article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
the study of ultracold atomic systems has received much attention in recent years , motivated largely by the prospect of realizing novel strongly correlated many - body physics .
a series of remarkable experimental breakthroughs have produced an extremely well - controlled physics playground @xcite .
the latest breakthrough is the ability to associate atoms via a feshbach resonance to form ultracold molecules in the ro - vibrational ground state @xcite , and the jila group has achieved a nearly degenerate gas of ground state polar molecules @xcite .
this is a true milestone in the program , since strong dipolar interactions between degenerate molecules in the presence of an external electric field allow for the design of exotic hamiltonians @xcite and are expected to give rise to exciting phenomena including roton softening @xcite , supersolidity @xcite , artificial photons @xcite , bilayer quantum phase transitions @xcite , and multi - layer self - assembled chains @xcite for bosonic molecules , and spontaneous interlayer superfluidity @xcite , itinerant ferroelectricity @xcite , fermi liquid anisotropy @xcite , fractional quantum hall effect @xcite , wigner crystallization @xcite , biaxial nematic phase @xcite , topological superfluidity @xcite , and a @xmath1 topological phase @xcite for fermionic molecules .
most , if not all , of these novel quantum phases will require temperatures on the order of @xmath2 or less ( with @xmath3 the fermi temperature ) , which will require further experimental advance .
efforts to overcome the current obstacles of collisional instability and insufficient cooling are already underway , though , and given the rate of progress in reaching the current state - of - the - art , the future is bright @xcite .
however , it is worthwhile to ask what interesting effects one might be able to observe in the immediate future with the temperature on the order of @xmath3 . from a condensed matter perspective ,
the very idea of having a @xmath4 interaction rather than the usual coulomb interaction is intriguing .
such a system has no parallel in solid state materials , and study of the previously unmotivated problem of many - body physics in a system interacting via a @xmath4 potential is still in its infancy .
of course , the actual form of the inter - molecular interaction potential is quite complicated , particularly at short range , and the scattering and stability of the molecules are sensitive to these details @xcite .
further investigation is required to allow full quantitative comparison between theory and experiment . for our purposes ,
though , it is sufficient to take a @xmath4 interaction with some undetermined short - range cutoff ; the low - energy many - body physics of a stable , dilute gas should not depend qualitatively on the short - range details .
it is then appropriate to consider what quantum many - body effects might be calculated for a gas with @xmath4 interactions that could be observed in cold polar molecule experiments . in this paper
, we calculate the compressibility ( or equivalently , the ordinary sound dispersion ) , zero sound dispersion , and effective mass at finite temperatures comparable to those achieved in current experiments .
these quantities should be readily accessible to experiment , and we present the first calculations to include thermal effects , as well as some trap effects .
as these calculations are carried out within a leading order perturbation approximation in the dipolar coupling constant , we are working within the standard weak - coupling theory in the sense of the landau fermi liquid theory .
this is the first theory one must do before one does anything else for the dipolar systems . for dipolar interactions , in contrast to the familiar coulomb case
, the weak - coupling regime corresponds to the _ low_-density limit .
this is fortuitous , since current experiments can not achieve high densities without significant loss rates .
for all practical purposes , the leading - order perturbative results are exact at typical experimental densities . however , lower densities correspond to higher @xmath0 , necessitating the calculations be done for finite temperature . as the density
is increased , the leading - order perturbative results should remain qualitatively correct as long there is no phase transition to break adiabaticity .
for very high densities one will enter the interesting strong - coupling regime and the weak - coupling theory will fail in a systematic way , but this regime is completely inaccessible at present in dipolar molecular systems . in two - dimensional semiconductor - based electron systems and in graphene
, studying compressibility experimentally @xcite and theoretically @xcite has been an important tool for studying quantum many - body effects in coulomb systems . in these two - dimensional condensed matter systems with the @xmath5 coulomb interaction , the hartree - fock approximation works remarkably well for understanding compressibility , even in the strongly interacting regime .
this is a general result of the frequency independence of the compressibility , and not specific to the coulomb interaction .
we allow for different trap geometries by considering uniform three - dimensional ( 3d ) , two - dimensional ( 2d ) , and one - dimensional ( 1d ) gases , as well as a nonuniform 3d gas in a periodic potential along @xmath6 .
we find that under certain conditions the compressibility varies nonmonotonically with temperature .
a closely related quantity with the same temperature dependence , @xmath7 ( with @xmath8 the fermi energy and @xmath9 the chemical potential ) , also varies nonmonotonically with density , even for temperatures on the order of @xmath3 .
the zero sound speed and effective mass also exhibit nontrivial dependencies .
however , in 3d , propagation of zero sound requires lower temperatures than currently feasible . in 2d and 1d , though , we find that zero sound propagates undamped even at experimentally realistic temperatures , assuming that the intermolecular potential behaves like @xmath4 at short enough distances .
in the low - dimensional situations , the geometry with dipoles aligned perpendicular to their separations is particularly relevant as this is the most stable configuration conducive to our approximation of a @xmath4 interaction , so we will focus on this case .
the layout of the paper is as follows : in section [ sec : vq ] we review the form of the dipolar interaction in momentum space . in section [ sec : results ] we present analytic low - temperature expansions for the self - energy , compressibility , and effective mass of a uniform system , as well as numerical results for arbitrary temperature .
we also present numerical results for the zero sound mode . in section [ sec : multilayer ]
we numerically obtain the compressibility and low - lying collective modes for the spatially inhomogeneous case of a 1d periodic external potential dividing the 3d cloud into multiple quasi-2d layers . in section [ sec : exp ] , we discuss experimental observation of quantum many - body effects in the compressibility , and we conclude in section [ sec : conclusions ] . in appendix
a , we briefly compare with the case of coulomb interactions .
we consider a gas of one - component fermionic molecules with number density @xmath10 , each possessing dipole moment @xmath11 , aligned by an external electric field @xmath12 so that the inter - molecular interaction potential at large separation @xmath13 is given by @xmath14 where @xmath15 is the angle between @xmath13 and @xmath12 .
we will parameterize the interaction using the dimensionless ratio @xmath16 , where @xmath17 is the average interparticle distance . for the different dimensionalities we have @xmath18 where @xmath19 is the noninteracting fermi wavevector . for fermionic krb ( @xmath20 debye , @xmath21 amu ) at densities around @xmath22 @xcite ( or @xmath23 or @xmath24 in lower dimensions ) , @xmath25 . strictly speaking , the fourier transform of @xmath26 has ultraviolet and infrared divergences
. however , recall that anyways this potential is not valid for arbitrarily short range and the system has some finite size . thus , as in ref .
@xcite , we introduce a short - range cutoff , @xmath27 , and a long - range cutoff , @xmath28 .
the fourier transform can then be performed to obtain @xmath29 \\
\overset{q\epsilon \rightarrow 0 , qr \rightarrow \infty}{\longrightarrow } \frac{8\pi}{3 } d^2 p_2 \!\left ( \cos \theta_{q } \right ) = 16 \pi^3 \lambda_{3d } p_2 \!\left ( \cos \theta_{q } \right ) \frac{e_f}{k_{f0}^3}\end{gathered}\ ] ] where @xmath30 is the second legendre polynomial , @xmath31 is the angle between @xmath32 and @xmath12 , and @xmath33 is the spherical bessel function of the first kind .
if there is strong confinement along @xmath6 , then assuming a fixed gaussian profile in that direction @xmath34 , and integrating eq .
[ eq : v3dq ] over this momentum profile , the effective 2d interaction is @xmath35 \frac{e_f}{k_{f0}^2}\end{gathered}\ ] ] where @xmath36 is the angle between @xmath12 and the @xmath6-axis , @xmath37 is the azimuthal angle between @xmath32 and the projection of @xmath12 onto the @xmath38 plane , and we have taken @xmath39 . alternatively , fixing @xmath13 to lie in the @xmath38 plane and taking the 2d fourier transform of eq .
[ eq : v3dr ] with a short - range cutoff , @xmath40 , as in ref .
@xcite yields the same result with a numerical factor of order unity in front of @xmath40 .
similarly , when there is strong confinement and a fixed gaussian profile @xmath41 in the radial direction , integrating out the radial degree of freedom in eq .
[ eq : v3dq ] yields an effective 1d interaction @xmath42 \frac{e_f}{k_{f0}^3 } \\
\overset{qw \rightarrow 0}{\longrightarrow } -\pi^3 \lambda_{1d } p_2 \!\left(\cos \theta_{e}\right ) \left [ \frac{1}{3 w^2 } + q^2\left(\gamma + 2\ln |q w| \right ) \right ] \frac{e_f}{k_{f0}^3}\end{gathered}\ ] ] where @xmath36 is again the angle between @xmath12 and the @xmath6-axis , @xmath43 is the incomplete gamma function , and @xmath44 is euler s constant .
( here again , performing a 1d fourier transform of a @xmath45 potential with a short - distance cutoff @xmath40 gives a similar dependence on @xmath46 and @xmath40 , although the numerical factors are different . )
unique to the quasi-1d case , the momentum dependent part of the effective interaction is not independent of the transverse width , even for very strong confinement .
to calculate the compressibility , @xmath47 @xcite , we first need to obtain the chemical potential in the presence of the dipolar interactions . in the uniform case ,
to first order , @xmath48 , where @xmath49 is the noninteracting chemical potential and @xmath50 is the isotropic part of the hartree - fock self - energy at the unperturbed fermi surface , @xmath51 with @xmath52 the noninteracting fermi distribution .
the zero - temperature self - energy has recently been calculated in ref .
below we extend this to finite temperature and obtain the inverse compressibility , effective mass , and zero sound dispersion .
[ [ subsec:3d ] ] 3d ~~ [ fig:3dmeffvst ] [ fig:3dmeffvsn ] [ fig:3dstability ] let us begin by considering the 3d case at finite temperature @xmath53 , with @xmath54 , @xmath55 \label{eq:3dsigma1 } \\ & = 2\pi\lambda_{3d } e_f \frac{k^3}{k_{f0}^3 } p_2 \!\left(\cos \theta_{\mathbf{k } } \right ) \int_0^{\infty } \frac{dx}{e^{\left(x^2 k^2/k_{f0}^2 - \tilde{\mu_0 } \right)/t } + 1 } \left [ -\frac{5}{2}x^2 + \frac{3}{2 } x^4 + \frac{3}{4 } x \left(x^2 - 1\right)^2 \ln \left| \frac{x-1}{x+1 } \right| \right ] \label{eq:3dsigma2 } \\
\sigma_{3d } \left ( \mathbf{k_{f0 } } \right ) & = 2\pi\lambda_{3d } e_f p_2 \!\left(\cos \theta_{\mathbf{k_{f0 } } } \right ) \biggl [ \frac { 5 \sqrt{\pi}}{8 } t^{3/2 } \text{li}_{\frac{3}{2 } } \left(-e^{\tilde{\mu_0}/t}\right ) - \frac { 9 \sqrt{\pi } } { 16 } t^{5/2 } \text{li}_{\frac{5}{2 } } \left(-e^{\tilde{\mu_0}/t}\right ) \notag \\ & \qquad + \frac{3}{4 } \int_0^{\sqrt{\tilde{\mu_0 } } } x dx \left(x^2 - 1\right)^2 \ln \left| \frac{x-1}{x+1 } \right| + \frac{3}{4 } \int_0^{\infty } \frac{dx \text{sgn } \left ( x^2 - \tilde{\mu_0 } \right ) } { e^{\left|x^2 - \tilde{\mu_0 } \right|/t } + 1 } x \left(x^2 - 1\right)^2 \ln \left| \frac{x-1}{x+1 } \right| \biggr ] \label{eq:3dsigma3 } \\ & = 2\pi\lambda_{3d } e_f p_2 \!\left(\cos \theta_{\mathbf{k_{f0 } } } \right ) \biggl\ { \frac{3 \tilde{\mu_0}^{5/2}}{10 } - \frac{5 \tilde{\mu_0}^{3/2}}{6 } + \frac{9 \pi^2 \tilde{\mu_0}^{1/2}}{48 } t^2 - \frac{5 \pi^2}{48 \tilde{\mu_0}^{1/2 } } t^2 - \frac{2}{15 } \notag \\ & \qquad + \frac{\pi^2}{16 } t^2 \left[- \left(1 - \tilde{\mu_0 } \right ) \left(1 + 2\zeta ' \left(-1\right ) + 2 \ln \frac{4\pi}{\left ( \sqrt{\tilde{\mu_0 } } + 1 \right)^2 } \right ) + \frac{\left(1 - \tilde{\mu_0 } \right)^2}{\tilde{\mu_0 } + \sqrt{\tilde{\mu_0 } } } \right ] + o\left(t^4\right ) + o\left(\left(1 - \tilde{\mu_0 } \right)^3
\right ) \biggr\ } \label{eq:3dsigma4 } \\ & = 4\pi\lambda_{3d } e_f p_2 \!\left(\cos \theta_{\mathbf{k_{f0 } } } \right ) \left [ -\frac{1}{3 } + \frac{\pi^2}{16 }
t^2 + o\left(t^4 \ln t\right ) \right ] \label{eq:3dsigma5}\end{aligned}\ ] ] @xmath56 \label{eq:3ddsigma2}\end{aligned}\ ] ] where @xmath57 is the angle between @xmath58 and @xmath12 , @xmath59 is the polylogarithm function , and @xmath60 is the derivative of the riemann zeta function . in this case , the self - energy has a purely @xmath61-wave angular dependence and does not affect the compressibility , which is given in units of the zero - temperature , noninteracting compressibility @xmath62 by @xmath63 the temperature - dependent effective masses along the radial and angular directions can be written to first - order as in the zero - temperature case @xcite : @xmath64 @xmath65 for general temperatures we must evaluate eq . and its derivative numerically . at low temperatures ,
though , eqs . and give analytic approximations on the fermi surface .
we have plotted the numerical inverse effective mass results at the fermi surface as functions of temperature and density in fig .
[ fig:3dmeff ] , as well as the low - temperature expansions .
note that the deviation of the inverse effective radial mass from the inverse bare mass is proportional to @xmath66 ( the integrand in eq .
is always negative ) , so the effective radial mass is smaller than the bare mass at the equator and larger ( or negative ) at the poles .
also , the effective angular mass is proportional to @xmath67 .
the average effective mass over the fermi sphere is thus unchanged by the interaction , but when the effective radial mass at the poles of the fermi surface is negative , the system is unstable . in the low - temperature limit
we have the stability condition @xmath68 , in excellent agreement with previous numerical results using a variational approach which gave @xmath69 @xcite . extending this result to finite temperature
, we obtain the critical line for krb in the @xmath70 plane plotted in fig .
[ fig:3dmeff](c ) .
however , the system actually becomes unstable at densities an order of magnitude lower due to formation of density waves perpendicular to the electric field @xcite .
we now turn our attention to the zero sound mode , which has recently been calculated in the zero - temperature limit @xcite .
we anticipate that zero sound will only propagate at low temperatures , so we assume that the landau fermi liquid description @xmath71 remains valid , with the temperature dependence entering through the single - particle energy , @xmath72 , and the fermi distribution function , @xmath73 .
the temperature dependence of the hartree - fock quasi - particle interaction , @xmath74 , is negligible since the interaction near the fermi surface is independent of the magnitude of the momenta in the limit @xmath75 @xcite .
note that , although we have not made the notation explicit , the interaction depends on the direction of momentum transfer . neglecting collisions , the linearized boltzmann equation @xcite is @xmath76 where @xmath77 is the deviation of the quasi - particle distribution from the equilibrium anisotropic distribution , @xmath78 . integrating out the radial degree of freedom and decomposing into spherical harmonics , @xmath79 , with the angle in the argument measured relative to the direction of momentum transfer , @xmath80 , we can rewrite the above as @xmath81 where @xmath82 @xmath83 and @xmath84 with @xmath85 the noninteracting fermi velocity and @xmath86 in the preceding we have used the hartree - fock single - particle energy in order to account for the anisotropy and obtain results consistent with refs . @xcite .
since the self - energy also appears in the fermi distribution , the integrand is evaluated in some thermally broadened range of momenta around the _ distorted _ fermi surface .
however , note that in order to consistently include only first - order corrections to the noninteracting single - particle energy , the self - energy terms themselves must be evaluated around the _ unperturbed _ fermi surface .
hence the added superscript denoting the subtle distinction , and for @xmath58 not too far below the fermi surface one can simply take @xmath87 , where the latter is given by eq . and the three vectors in the argument are collinear
. we will keep only the coupling between the @xmath88-wave and longitudinal @xmath89-wave modes as in ref .
@xcite , which yields good agreement with the results of ref .
the landau parameters for the dipolar interaction have previously been calculated @xcite , and in our notation and choice of basis the relevant nonzero parameters are @xmath90 the energy , @xmath91 , and damping , @xmath92 , of the collective modes for a given momentum , @xmath32 , are then given by the solutions of @xmath93 , where @xmath94 we numerically find solutions corresponding to underdamped zero sound propagation . in that case , the dispersion is given by @xmath95 and the damping , assumed to be small compared to @xmath91 , is given by @xmath96 the dispersion and damping for various temperatures are shown in figs .
[ fig:3dzerosound](a ) and [ fig:3dzerosound](b ) , respectively , with the density chosen such that @xmath97 for comparison with refs .
@xcite . for temperatures
up to about @xmath2 , the thermal effect is to increase the axial propagation speed while decreasing the range of angles for which zero sound propagates . as the temperature increases further , the axial propagation speed decreases .
note that at finite temperature , the mode can become overdamped long before it enters the single - particle excitation ( spe ) continuum , and eq
. ceases to have a solution . for temperatures above @xmath98
the damping is large even at @xmath99 , and above @xmath100 there is no solution to eq . for any @xmath31 .
clearly , the assumption of @xmath101 quickly breaks down with increasing temperature and propagation off the @xmath6-axis .
[ [ subsec:2d ] ] 2d ~~ in the remainder of this manuscript
our primary focus will be on dipoles aligned perpendicular to their motional degrees of freedom , so it will be unnecessary to distinguish between distorted and undistorted fermi surfaces , @xmath102 and @xmath103 , and we will drop the extra subscript . for the 2d case , with @xmath104 , @xmath36 the angle between @xmath12 and the normal to the plane , and @xmath105 the angle between the fermi wavevector and the projection of @xmath12 onto the plane , from the isotropic part of the interaction we obtain : @xmath106\right ) \biggr\ } \label{eq:2dsigmaiso3 } \\ & = 16\sqrt{\pi } \lambda_{2d } p_2 \!\left(\cos \theta_e\right ) e_f \left(\frac{8}{9}+\frac{\pi^2}{24 } t^2 + o\left[t^4 \ln t\right ] \right ) , \label{eq:2dsigmaiso4}\end{aligned}\ ] ] @xmath107 \label{eq:2ddsigmaisodk1 } \\ & = 16\sqrt{\pi } \lambda_{2d } p_2
\!\left(\cos \theta_e\right ) \frac{e_f}{k_{f } } \left(\frac{2}{3 } + \frac{\pi^2}{48}t^2 \left(1 + \ln \frac{\pi}{4 } + 12 \zeta'\left(-1\right)\right ) + \frac{\pi^2}{48}t^2 \ln t + o\left[t^4 \ln t\right ] \right ) \label{eq:2ddsigmaisodk2}\end{aligned}\ ] ] @xmath108 with @xmath109 and @xmath110 the complete elliptic integrals of the first and second kind @xcite .
although we will focus on the most stable scenario of dipoles aligned perpendicular to the plane , for completeness we also give the form of the self - energy arising from the anisotropic part of the interaction : @xmath111 \label{eq:2dsigmaani2 } \\ \sigma_{2d}^{\text{ani } } \left(\mathbf{k_{f}}\right ) & = -\frac{8\sqrt{\pi}}{3 } \lambda_{2d } e_f \sin^2
\theta_e \cos 2\phi_{\mathbf{k_{f } } } \biggl\ { \int_0^{\tilde{\mu_0 } } x dx \left(1+x\right ) \left[\left(2-x^2\right ) e\left(\frac{2\sqrt{x}}{1+x}\right ) + \left(1-x\right)^2 k\left(\frac{2\sqrt{x}}{1+x}\right ) \right ] \notag \\ & \qquad \qquad \qquad \qquad \qquad \qquad \qquad + \int_0^{\infty } x dx \frac{\text{sgn } \left(x^2 - \tilde{\mu_0 } \right)}{e^{\left|x^2 - \tilde{\mu_0 } \right|/t } + 1 } \left(1 + 2 \left(1-x\right ) + o\left[\left(1-x\right)^2 \ln |1-x| \right]\right ) \biggr\ } \label{eq:2dsigmaani3 } \\ & = -\frac{8\sqrt{\pi}}{3 } \lambda_{2d } e_f \sin^2 \theta_e \cos 2\phi_{\mathbf{k_{f } } } \left(\frac{8}{5}- \frac{\pi^2}{8 } t^2 + o\left[t^4 \ln t \right ] \right ) , \label{eq:2dsigmaani4}\end{aligned}\ ] ] in eqs .
and we have noted that at low temperature the integrand in the temperature - dependent term is only appreciable near @xmath112 and expanded the elliptic integrals about that point , permitting analytic integration .
we have shown the derivative of the self - energy with respect to density since the inverse compressibility is @xmath113 .
for arbitrary temperatures , the compressibility is obtained via numerical integration from eq . .
for low temperatures , recognizing that the fermi energy , @xmath8 , and reduced temperature , @xmath114 , and coupling strength , @xmath115 , all depend on density , we can obtain an analytic result by differentiating the self - energy given in eq . according to @xmath116
+ + when the orienting electric field is perpendicular to the plane ( @xmath117 ) , the compressibility is always positive ( assuming the gas is not so dense that the @xmath4 approximation to the potential breaks down ) . as shown in figs .
[ fig:2d](a ) and [ fig:2d](b ) , the compressibility is nonmonotonic in temperature for experimentally relevant densities .
the nonmonotonicity is due to the interaction - dependent term , @xmath118 , which decreases quadratically with @xmath114 before turning over and increasing with @xmath114 for fixed density . since @xmath119 and @xmath120 , both the temperature at which the peak occurs and the height of the peak are larger for higher density . in fig .
[ fig:2d](c ) , we show how the compressibility varies as the density is changed .
note that we do not use the natural unit of @xmath121 in this plot , since it is density - dependent . in the low - density ( high-@xmath114 ) limit ,
the compressibility behaves classically , as expected , and is drastically reduced in the high - density ( low-@xmath114 ) limit .
the nonmonotonicity in the compressibility shown in figs .
[ fig:2d](a ) and [ fig:2d](b ) occurs well below typical experimental temperatures of @xmath122 nk .
however , we may also consider the ratio of the finite - temperature interacting and zero - temperature noninteracting compressibilities , @xmath123 , which has the same temperature dependence as the compressibility . figures [ fig:2d](c ) and [ fig:2d](d ) show that this quantity also displays a nonmonotonic density dependence .
moreover , this nonmonotonic behavior is evident even at relatively high temperature , with @xmath124 .
[ for the homogeneous system considered here , @xmath125 24nk ( 240nk ) for @xmath126 @xmath127 ( @xmath128 @xmath127 ) . ] in this scenario , the peak is not due to nonmonotonicity in the interaction - dependent term , @xmath118 .
both @xmath118 and @xmath129 are monotonic in the density , but while @xmath118 increases with density , @xmath129 decreases .
the competition between the interacting and noninteracting density dependencies is the source of the nonmonotonicity .
figure [ fig:2d](e ) gives a sense of how @xmath130 changes as temperature and density are simultaneously changed in some manner .
although , it is hard to see on the logarithmic scale , the vertical portions of the lines actually bulge outwards slightly , consistent with the fixed density results plotted in figs .
[ fig:2d](a ) and [ fig:2d](b ) . clearly , nonmonotonic behavior is more pronounced if the measurements are performed at fixed temperature and variable density than the other way around .
we will discuss measurement in sec .
[ sec : exp ] . with the radial and angular effective masses , @xmath131 and @xmath132 , defined as in eqs . and
, we have plotted the radial mass at the fermi surface for the isotropic case in fig.[fig:2dmeff ] .
we see that at low temperatures or high densities the effective mass is diminished by the interaction while at high temperatures or low densities the effect of the interaction is small and the effective mass approaches its bare value .
the analytic low temperature results from eq . are shown as dotted lines .
approximations.,title="fig : " ] + approximations.,title="fig : " ] for dipoles aligned perpendicular to the plane , the 2d interaction is isotropic in contrast to the 3d case , and we need not include coupling to higher partial waves when calculating the zero sound . likewise , since the single - particle energy is also isotropic , to leading order we can simply use the bare single - particle energy .
the zero sound dispersion is then simply given by the solution of @xmath133 , where @xmath134 ^ 2 } \!\left[1 - \frac{\frac{v_0}{v_{f}}}{\sqrt{\frac{v_0 ^ 2}{v_{f}^2 } - x^2 - i0 } } \right]\end{gathered}\ ] ] note from eq .
that the thickness of the monolayer plays a critical role in determining the zero sound speed .
the dispersion is plotted in fig .
[ fig:2dzerosound ] as a function of temperature and density for a thickness of @xmath135 nm . in the limit @xmath136
, one can show that @xmath137 @xcite , and we numerically recover this limit to better than 2% .
the damping is negligible , and we have not shown it .
the mode propagates even at quite high temperatures due to the strong repulsive delta - function core of the effective 2d interaction .
in fact , it persists to arbitrarily high temperature for tight enough confinement .
this is an artifact of the ultraviolet divergence of the 2d fourier transform of the @xmath4 interaction , and in experiment this robustness will be limited by the range , @xmath138 , at which the actual intermolecular potential deviates from a @xmath4 behavior .
however , provided @xmath139 and @xmath140 , these results will remain valid and the collective mode will be robust at experimentally feasible temperature and density .
+ [ [ subsec:1d ] ] 1d ~~ + + for the 1d case , @xmath141 , and @xmath36 is the angle between @xmath12 and the trap axis . in this case ,
recall from eq . that the results will depend explicitly on the transverse width ( or short range interaction cutoff ) , @xmath40 . in the limit @xmath142 ,
the hartree - fock self - energy is @xmath143 \label{eq:1dsigma2 } \\ & = \frac{\pi^2}{2 } \lambda_{1d } e_f p_2
\!\left(\cos \theta_e \right ) \left [ \frac{8}{3}\left ( \gamma - \frac{2}{3 } + 2 \ln \left(2 k_f w\right ) \right ) + \left ( \gamma + 1 + 2 \ln \left(2 k_f w\right )
\right ) \frac{\pi^2}{6 } t^2 \right ] + o\left[t^4 \ln t\right ] \label{eq:1dsigma3}.\end{aligned}\ ] ] differentiating with respect to density yields @xmath144 + o\left[t^4 \ln t\right ] .\end{gathered}\ ] ] however , even if we take transverse size @xmath145 nm , corresponding to a very strong optical dipole trap , then at @xmath146 @xmath147 although the system is quasi-1d ( @xmath148 ) for @xmath149 nk , @xmath150 is actually of order unity .
this tends to diminish interaction effects since less repulsive off - axis interactions must be included in the effective 1d interaction .
( we did not hesitate to take @xmath151 in the 2d case because there the working assumption is that several independent quasi-2d layers will be formed by an optical lattice , which can be made considerably tighter than a dipole trap . in any event , the general treatment of the multilayer quasi-2d in sec .
[ sec : multilayer ] includes finite thickness effects . ) for arbitrary values of @xmath150 , then , a similar calculation without assuming @xmath152 in eq .
yields @xmath153 \biggr\ } + o\left(t^4 \ln t\right)\end{gathered}\ ] ] @xmath154 -\frac{\pi^2}{6 } t^2 \ln t + o\left(t^4 \ln t\right ) \biggr\}\end{gathered}\ ] ] @xmath155 \biggr\ } + o\left(t^4 \ln t\right)\end{gathered}\ ] ] where @xmath156 is a meijer g - function @xcite . the more general eq . was used for the low temperature approximation in fig .
[ fig:1d ] . for the second row of plots , with @xmath135
nm corresponding to a tight optical lattice forming an array of independent quasi-1d tubes , the simpler initial approach gives the same results except for at high densities .
figures [ fig:1d](a ) and [ fig:1d](d ) show a maximum in the compressibility at finite temperature as in the 2d case .
now , though , @xmath157 , so the temperature at which the peak occurs increases with density more rapidly than in 2d . for @xmath135 nm
, we have plotted results for higher densities and temperatures since these values can be reached while maintaining low - dimensionality .
in contrast to the 2d case , in a tightly confined system at reasonable density , the peak occurs at higher temperature as shown in fig .
[ fig:1d](d ) . however , it is important to note that in 1d this is no longer a signature of interaction effects since the compressibility exhibits a maximum even in a noninteracting system due to the nonmonotonic temperature dependence of @xmath158 .
in fact , the main effect of interactions is to add a constant shift to the inverse compressibility , making the peak relatively _ less _ dramatic . as in the 2d case
, we can also look for nonmonotonicity in the density dependence of @xmath159 . in the 2d case
, we found a maximum in @xmath130 in the presence of interactions .
here there is a weak maximum in the absence of interactions , due to the nonmonotonicity of @xmath49 . in the presence of interactions ,
this feature washes out , but the interactions may give rise to a more pronounced local maximum . unlike the 2d case , here the nonmonotonicity in density is predominantly due to nonmonotonic behavior of @xmath160 rather than competition between the interacting and noninteracting terms .
for dipoles perpendicular to the trap axis and @xmath161 nm , the density dependence at fixed temperature above a few nk is monotonic as in fig .
[ fig:1d](c ) .
however , for tighter radial confinement , the interaction effects become more pronounced and nonmonotonicity emerges at higher temperatures , as shown in fig .
[ fig:1d](f ) . at high densities , the effective 1d interaction at the fermi surface saturates due to the dependence on the finite transverse size . as a result
@xmath130 begins rising again .
actually , for @xmath162 nm ( @xmath163 nm ) , one - dimensionality is already beginning to break down for @xmath164 @xmath147 ( @xmath165 @xmath147 ) , since there @xmath166 .
the effective mass is plotted in fig .
[ fig:1dmeff ] as a function of density and temperature .
the effective mass is again diminished by the interaction and slowly increases towards its bare value at high temperatures . at low densities , as in 2d , the reduction of the effective mass by the interaction increases with density .
however , as the density increases further the effective mass eventually returns to its bare value since the effective 1d interaction saturates while the fermi velocity continues to increase . in order to clearly show the trend , we have included in fig .
[ fig:1dmeff](b ) extremely high densities .
the analytic low temperature results from eq . are shown as dotted lines .
nm ( a ) vs temperature at fixed density and ( b ) vs density at fixed temperature .
dotted lines show the low @xmath0 approximations.,title="fig : " ] + nm ( a ) vs temperature at fixed density and ( b ) vs density at fixed temperature .
dotted lines show the low @xmath0 approximations.,title="fig : " ] + as in the 2d case , we consider the zero sound dispersion given by @xmath167 , where @xmath168 ^ 2 } \\ \times \left[1 - \frac{\frac{v_0}{v_{f}}}{\frac{v_0}{v_{f } } - x + i0 } \right ] .\end{gathered}\ ] ] once again , the finite transverse size which appears in eq . is important in determining the zero sound speed .
the dispersion is plotted in fig .
[ fig:1dzerosound ] as a function of temperature and density for @xmath135 nm and dipoles aligned perpendicular to the axis of the tube . in the low - temperature limit , one can show that @xmath169 , and we numerically recover this limit to better than 1% .
one can not obtain this limit by going to high density instead of low temperature , since , unlike in 2d where @xmath170 approaches unity in the high density limit , here @xmath170 goes to zero like @xmath171 for @xmath172 .
thus , at high density the zero sound speed decreases to unity .
the damping strength , @xmath173 , is shown as a dashed line and it is negligible except near where the mode vanishes .
the collective mode is even more robust than in the 2d case since the 1d fourier transform of the interaction is even more divergent .
again , these results will remain physical until @xmath138 becomes comparable to either of the other length scales , @xmath174 or @xmath40 .
now we consider a 3d gas placed in a strong 1d optical lattice of depth @xmath175 and wavelength @xmath115 , resulting in a `` stack of pancakes '' configuration . in the limit where @xmath175 and @xmath115 become infinite , this is the same as the strictly 2d monolayer considered in sec .
[ subsec:2d ] . generally , however , we find that the nonmonotonic behavior discussed in sec .
[ subsec:2d ] persists in the presence of multilayer and finite thickness effects .
we will restrict our discussion to the case of dipoles aligned along the direction of the optical lattice .
because of the external periodic potential along @xmath6 , the effect of the interaction on the chemical potential to first order , @xmath176 , must now be written in real space : @xmath177 where the angular averaging over the fermi surface denoted by the brackets is necessary to separate the shift in chemical potential from the anisotropic deformation of the fermi surface , and @xmath178 .\end{gathered}\ ] ] the unperturbed single - particle wavefunction can be written in terms of the reciprocal lattice points , which are integer multiples of @xmath179 , in the bloch form @xmath180 the coefficients @xmath181 are easily obtained numerically , as are the corresponding single - particle energies needed to evaluate @xmath182 . in the above we have introduced a cutoff , @xmath183 , on the bloch momentum in order to perform the sum numerically . after some algebra , eq . can be recast as a sum over momenta , @xmath184 \bigg\rangle_{\omega_{\mathbf{k_f}}}\end{gathered}\ ] ] where @xmath185 is the 3d fourier transformed interaction defined in eq . and @xmath186 , @xmath187 .
we typically take @xmath188 , which we find to be sufficient for the lattice depths treated . to facilitate comparison with the strictly 2d case , we measure density in terms of the 2d density per layer , @xmath189 , and compressibility in units of @xmath190 , where @xmath191 . by first analytically performing the azimuthal integration in eq . and differentiating with respect to density , then numerically summing over reciprocal lattice points , averaging over the fermi surface , and performing the remaining two - dimensional integral , the compressibility can be efficiently obtained . in fig .
[ fig : multilayer ] we show the compressibility for krb in a deep optical lattice ( here the recoil energy @xmath192 ) . for strong lattices
, the compressibility behaves similarly versus temperature to the 2d results of fig .
[ fig:2d ] , with a maximum at nearly the same temperature . for weak lattices ,
the effect is less pronounced .
as density is varied , the ratio @xmath130 behaves as in the 2d case for low densities , with nonmonotonicity evident even for @xmath124 , but at high density @xmath193 becomes larger than the lattice bandgap and quasi - two - dimensionality breaks down . here
the compressibility exhibits discontinuous jumps as the chemical potential enters higher bands , reminiscent of the effect seen in ref .
@xcite . at very large densities ,
the recoil energy is negligible compared to the kinetic energy of the gas , and @xmath130 should increase with density as in the 3d case rather than decreasing with density as in the 2d case . goes to unity as @xmath10 goes to infinity .
however , since we have defined @xmath194 in this section in terms of the 2d energy scale @xmath195 rather than the 3d energy scale @xmath196 , the ratio plotted in fig . [
fig : multilayer](c ) should not go to unity , but should go like @xmath197 as @xmath10 goes to infinity .
] so , in addition to the interaction - induced local maximum analogous to the 2d case , there can also be a local minimum as the system crosses over to three - dimensional behavior .
this is similar to the finite confinement effect seen in the 1d case above . for the 3d multilayer system described above
we now calculate the low - lying collective modes at finite temperature , neglecting coupling between partial waves . in the limit of a large bandgap
, such couplings should be suppressed , and outside this limit the description should still be qualitatively valid . however , because of the spatial inhomogeneity , the 3d polarizability can no longer be written as a scalar function of the momentum transfer . in terms of the bloch wavefunctions defined in eq .
, the interaction and the zeroth order polarizability can be written in momentum space as @xmath198 @xmath199 where @xmath200 is the bare single - particle energy associated with @xmath201 , and @xmath202 the collective modes are given by the zeros of the determinant @xmath203 , where @xmath204 .
conservation of momentum in the @xmath38 plane and of quasi - momentum along @xmath6 reduces @xmath205 to a matrix function of the momentum transfer , @xmath206 , where @xmath207 is an integer denoting a reciprocal lattice vector @xmath208 and @xmath209 is in the first brillouin zone . some algebra results in a computationally feasible expression for the matrix elements , @xmath210 \\ \times \left[\sum_{m'n ' } u_{q+n+m'-n ' } \left(k_z - q_z+\frac{2\pi q}{\lambda}\right )
u_{m'}^{\ast } \left(k_z\right ) u_{n ' } \left(q_z+\frac{2\pi n}{\lambda}\right ) v_{3d } \left(q_{\perp } , q_z+\frac{2 \pi \left(n - n'\right)}{\lambda } \right ) \right].\end{gathered}\ ] ] the matrix elements can be evaluated numerically for a given frequency and momentum .
to proceed , one must obviously truncate the matrix at some finite reciprocal lattice vector , and we have used a @xmath211 matrix , as this seems to include the dominant terms for large potential depth , @xmath175 . as @xmath175 is decreased , the off - diagonal terms of the matrix vanish while the number of nonnegligible diagonal elements increases . in fig .
[ fig : multilayerzerosound ] we plot the lowest - lying underdamped collective modes for long - wavelength , in - plane momentum transfer , @xmath212 .
the zero sound speed is shown in units of the 2d fermi speed , @xmath213 .
in addition to the landau damping due to single particle excitations in the lowest band for @xmath214 , there are also regions of damping due to single - particle excitations in excited bands which occur for @xmath215 . the boundaries of these regions are shown in dotted lines .
the modes plotted are weakly thermally damped .
it is instructive to first consider the 20nk curve in fig .
[ fig : multilayerzerosound](a ) , where one can identify three distinct regions . at densities below @xmath216 ,
the thermal energy is comparable to the 2d energy scale , @xmath191 . for higher densities
, one might expect to see a density independent speed , as in sec .
[ subsubsec:2dzerosound ] and the zero temperature calculation of ref .
this appears to be the case for an intermediate region , with the value @xmath217 in qualitative agreement with the value @xmath218 obtained using the multilayer result of ref .
@xcite and inserting an estimate for the transverse size in the effective 2d interaction based on the harmonic approximation to the local wells of the optical lattice .
however , for even higher densities the speed abruptly drops towards the continuum .
this begins happening where @xmath193 becomes comparable to the bandgap , signalling a transition to 3d behavior . as we have seen in sec .
[ subsubsec:3dzerosound ] , the mode with momentum transfer perpendicular to the electric field must vanish in 3d . in this region , the behavior is roughly independent of @xmath219 , since the reduced temperature @xmath0 is effectively zero . for @xmath220nk ,
the mode does not propagate except at high densities , and we can not reach the low temperature limit without increasing the density into the 3d transition . for the weaker lattice in fig . [
fig : multilayerzerosound](b ) , even at @xmath221nk the intermediate plateau has vanished because higher bands are already becoming active as we increase the density out of the large @xmath0 regime . in fig .
[ fig : multilayerzerosound](b ) we see multiple modes entering the continuum , although only one is stable at low density .
also , at high density , faster zero sound modes appear above the first excited spe band . in fig .
[ fig : wzv100 ] we show the lowest underdamped collective excitation for momentum transfer along the @xmath6-axis in units of the frequency of the local harmonic well approximation to the optical lattice . in a harmonic trap
this is the kohn ( or sloshing ) mode @xcite , and it is independent of the interaction and the momentum . here the density dependence is due to anharmonicity of the potential .
the curve shown is independent of temperature and momentum . for @xmath222
, we do not find this mode .
-axis . ]
so , although one can observe a kohn - like collective mode along the direction of the optical lattice at currently attainable temperatures and densities , the observation of zero sound in a reasonably strong optical lattice with standard site separation on the order of a micron requires an order of magnitude improvement in the attainable temperature or density .
based on the 2d calculations of sec .
[ subsubsec:2dzerosound ] , it is reasonable to expect this requirement could be eased by using a more tightly spaced lattice .
+ we have shown that a quasi - low - dimensional experiment may be able to observe an interaction - induced maximum in the ratio of the finite - temperature interacting compressibility and the zero - temperature noninteracting compressibility , particularly if the measurement is performed with variable density .
now , the results we have shown are for homogeneous systems , while there is always a global confining potential in experiments . however ,
if we incorporate a weak harmonic trap through a simple local density approximation ( lda ) , @xmath223 , where one can think of a local compressibility , @xmath224 , then the trap can be beneficial in that precise _ in - situ _ measurement of the density profile automatically allows the simultaneous measurement of compressibility across a wide range of densities .
this technique has recently been used to measure local compressibility in an optical lattice @xcite .
the ratio @xmath225 is similarly accessible . in fig .
[ fig : trapped ] we have shown the ratio as a function of radial distance in the lda , and the associated density profile , for krb at reasonable temperature and peak density .
a gaussian profile with the same peak density and number of molecules as the calculated profile is shown in figs .
[ fig : trapped](b ) and [ fig : trapped](d ) for reference .
the 2d results shown are generated from the strictly 2d calculations of sec .
[ subsec:2d ] , rather than the more complicated multilayer case .
if the strictly 2d case is realized by applying intense counter - propagating beams of wavelength @xmath226 m along the axis of a cigar - shaped cloud that contains @xmath227 molecules and is @xmath228 m in length , with radius @xmath229 m @xcite , then a typical 2d plane will contain @xmath230 molecules . as shown in fig . [
fig : trapped](a ) , to reach suitably large densities with such a small number requires the in - plane trapping frequency to be strong enough that the spatial extent of the cloud is rather small .
this may preclude _ in - situ _ measurement of the profile , so for that specific method , it is preferable to apply the lattice perpendicular to the axis of the dipole trap in order to deal with larger clouds . rather than a cigar - shaped dipole trap , one might instead apply a pancake - shaped trap as in ref .
@xcite , but it would be rather difficult to reach the 2d regime without using an optical lattice . if the 1d case is realized by applying two orthogonal pairs of counter - propagating beams perpendicular to the axis of the initial cigar - shaped trap , then a typical quasi-1d tube may contain @xmath231 molecules .
this case is shown in figs .
[ fig : trapped](c ) and [ fig : trapped](d ) .
in addition to the interaction - induced maximum , there is a local minimum for sufficiently high densities , as discussed above . for those higher trap frequencies , the density decreases more rapidly and the features
are correspondingly narrowed .
however , there is the possible complication that in the presence of a trap , the hartree term is no longer just a constant . due to the long - range interactions , one may need to explicitly take the nonuniformity into account in a self - consistent manner with a spatially dependent hartree term , as was done , for example , in refs
. @xcite .
it is not clear how important this effect is , although one may estimate that it will be negligible if the interaction energy , @xmath232 , at distances on the order of the characteristic length scale of the trap , @xmath233 , is very small compared to the local fermi energy .
since we have already assumed the local fermi energy is much larger than @xmath234 by taking the lda , this implies @xmath235 is a sufficient condition for neglecting spatial variation of the hartree term . for krb , @xmath236 nm ,
so it is reasonable to treat the system the way we have as a first step .
further investigation is required , though , to fully resolve the issue .
in this work , our goal has been to propose well - defined experiments which can see quantum many - body effects in a stable , ultracold gas of dipolar molecules at finite temperature .
our work is admittedly a leading - order approximate calculation , however it should be generally valid qualitatively and , for the coupling strengths accessible to experiment , essentially exact .
our results indicate that the compressibility of a quasi - low - dimensional gas should exhibit a nonmonotonic temperature dependence for low temperatures .
( note that a similar maximum in the hartree - fock compressibility occurs in an electron gas with @xmath5 interactions , although there it is due to a @xmath237 term in the self - energy @xcite ; see appendix a. ) furthermore , the closely related quantity @xmath7 , which is the ratio of the finite - temperature interacting compressibility and the zero - temperature noninteracting compressibility , should also exhibit a quantum many - body effect , namely a nonmonotonic _ density _ dependence , which persists for @xmath238 .
we have also considered the finite temperature behavior of the effective mass and zero sound dispersion . in the 3d case
, the zero sound shows a nonmonotonic temperature dependence , in addition to angular anisotropy .
however , as expected , we have seen that the damping increases rapidly with temperature , so that the mode does not propagate at currently feasible temperatures and densities .
however , in quasi - low - dimensional scenarios , the mode is far more robust , propagating at relatively high temperatures for tight transverse confinement , assuming the length scale at which the interaction deviates from @xmath4 behavior is small compared to the transverse size .
we have performed a numerical calculation for a 3d dipolar gas in a 1d periodic potential , and in addition to finding similar behavior as in the 2d case in the presence of multilayer effects , we observe effects of excited bands at high density and the breakdown of two - dimensionality .
in particular , at moderate lattice depths we see discontinuous jumps in the compressibility and interesting collective mode structure .
finally , we have discussed a method for experimentally observing the characteristic nonmonotonicity of the compressibility in the presence of a global confining potential via high - resolution _ in - situ _ imaging .
below we briefly compare the results given above for a spinless 2d dipolar gas to results similarly derived for a 2d spinless coulomb gas with @xmath239 . in the latter case , the low - temperature behavior within hartree - fock is @xcite , @xmath240 where @xmath241 , with @xmath242 the bohr radius .
the @xmath237 term is not an artifact of the dimensionality , as a similar result has also been obtained for the 3d coulomb gas @xcite . here
again , the compressibility behaves nonmonotonically , due to the interaction - dependent term first decreasing , then increasing with temperature .
however , in contrast to the dipolar gas , this nonmonotonicity is evident already at leading order in @xmath114 .
t. sogo , l. he , t. miyakawa , s. yi , h. lu , and h. pu , new j. phys . * 11 * , 055017 ( 2009 ) .
the original work contains an error in the stability criterion , with the corrected critical value being cited in ref .
@xcite .
g. pupillo , a. micheli , h. p. buchler , p. zoller , in _ cold molecules : creation and applications _ , edited by r. v. krems , b. friedrich and w. c. stwalley ( taylor & francis , 2008 ) , preprint arxiv:0805.1896v1 . | the compressibility , zero sound dispersion , and effective mass of a gas of fermionic dipolar molecules is calculated at finite temperature for one- , two- , and three - dimensional uniform systems , and in a multilayer quasi - two - dimensional system .
the compressibility is nonmonotonic in the reduced temperature , @xmath0 , exhibiting a maximum at finite temperature
. this effect might be visible in a quasi - low - dimensional experiment , providing a clear signature of the onset of many - body quantum degeneracy effects .
the collective mode dispersion and effective mass show similar nontrivial temperature and density dependence . in a quasi - low - dimensional system
, the zero sound mode may propagate at experimentally attainable temperatures . |
the family of iron oxyarsenide @xmath5feaso@xmath6f@xmath7 ( @xmath5 = lanthanide element ) exhibits superconductivity with a maximum @xmath8 up to 56 k @xcite . additionally , the iron - arsenide compounds @xmath9fe@xmath0as@xmath0 ( @xmath9 = alkaline earth element ) , crystallizing in the thcr@xmath0si@xmath0-type structure , are known to become superconducting with @xmath8 s up to 38 k upon alkali metal substitution for the @xmath9 element @xcite , or partial transition metal substitution for fe @xcite .
in contrast to undoped bafe@xmath0as@xmath0 with a magnetic ground state , superconductivity with relatively low @xmath8 s was reported in the undoped alkali metal iron - arsenides kfe@xmath0as@xmath0 ( @xmath10 k ) and csfe@xmath0as@xmath0 ( @xmath1 k ) @xcite .
interestingly , rbfe@xmath0as@xmath0 is known to exist as well @xcite , although its physical properties have not been reported so far . here
we report on the superconductivity in undoped alkali metal iron arsenide rbfe@xmath0as@xmath0 .
as@xmath0 polycrystalline sample , measured in a magnetic field of 1 mt . superconductivity sets in at @xmath11 k. ] , here for 1 mt and 200 mt , measured in the zfc mode .
a relative shift of the onset of superconductivity of 0.15 k is observed . an additional magnetic moment in the normal state in the 200 mt measurement , originates from a major normal state magnetic contribution . ]
polycrystalline samples of rbfe@xmath0as@xmath0 were synthesized in two steps . first , rbas and fe@xmath0as were prepared from pure elements in evacuated and sealed silica tubes . then , appropriate amounts of rbas and fe@xmath0as were mixed , pressed into pellets and annealed at 650 @xmath12c for several days in evacuated and sealed silica ampoules .
powder x - ray diffraction analysis revealed , that the synthesized rbfe@xmath0as@xmath0 is single phase material with lattice parameters @xmath13 and @xmath14 .
magnetization data have been recorded using a quantum design mpms xl squid magnetometer , equipped with a reciprocating sample option .
a polycrystalline sample of rbfe@xmath0as@xmath0 was studied for its low temperature magnetic properties . in fig .
1 the magnetic moment in the field - cooled state ( fc ) and in the zero - field cooled state ( zfc ) in a magnetic field of 1 mt are shown .
the data are indicative of bulk superconductivity .
the distinct onset of diamagnetism due to superconductivity is observed at @xmath11 k. due to the limited temperature range of the equipment , the full development of the meissner state could not be recorded .
nevertheless , the observed zfc diamagnetic response mirrors bulk superconductivity and is consistent with the sample dimensions . the pronounced difference between the zfc and fc curves stemms from remarkable flux - pinning in the sample , suggesting rather high critical current density .
+ the upper critical field @xmath15 was estimated from magnetization measurements performed at various magnetic fields in the mixed state . in fig . 2 ,
two representative measurements of the magnetic moment versus temperature are displayed for @xmath16 mt and for @xmath17 mt . we defined the upper critical field @xmath15 as the magnetic field @xmath18 , where @xmath19 is located .
an obvious shift of the onset of superconductivity of 0.15 k is observed between the respective fields .
in addition to the diamagnetic signal due to superconductivity , a distinct paramagnetic response develops due to the normal state magnetic contribution , rendering an accurate determination of @xmath2 rather difficult . nevertheless ,
since a clear downward curvature is observed due to the onset of superconducting diamagnetism , the trend of @xmath2 can be followed down to 2 k. figure 3 shows a summary of the results up to a field of 0.8 t , anticipating a linear slope close to @xmath8 of @xmath3 t / k . assuming a simple whh temperature dependence @xcite , which is known not to be applicable for the fe pnictide superconductors with much higher transition temperatures
, one would extrapolate @xmath4 t , in comparision to the lower critical field @xmath20 mt , as we estimated from field dependent initial magnetization curves , and the thermodynamic critical field @xmath21 mt .
superconductivity is , obviously , of type ii .
+ the solid solution ( rb , ba)fe@xmath0as@xmath0 offers a particularly simple example where the interrelation between magnetic and superconducting ground states in the fe pnictides can be studied through the controlled shift of the fermi level .
bafe@xmath0as@xmath0 shows antiferromagnetic ordering competing with superconducting state .
appearently , doping of rbfe@xmath0as@xmath0 with ba leads to a natural picture of enhancing @xmath8 in the superconducting state , as the charge carrier concentration is varied .
the appearence of superconductivity in rbfe@xmath0as@xmath0 opens up the window for a new interpretation of the occurence of superconducting state in ( rb , ba)fe@xmath0as@xmath0 @xcite . for rbfe@xmath0as@xmath0 .
the estimate of @xmath4 t is made using the whh - approach . ]
superconductivity is observed in undoped rbfe@xmath0as@xmath0 with a @xmath11 k. in this sense , it is useful to consider rbfe@xmath0as@xmath0 as a superconductor , located at the opposite end to the nonsuperconducting compound bafe@xmath0as@xmath0 in the ( rb , ba)fe@xmath0as@xmath0 system
. therefore , superconductivity is enhanced by doping of an initially superconducting nonmagnetic parent compound .
the upper critical field at zero temperature of rbfe@xmath0as@xmath0 is estimated to be @xmath4 t.
this work was supported by the swiss national science foundation , by the nccr program manep , and partially by the polish ministry of science and higher education within the research project for the years 2007 - 2009 ( grant no .
n n202 4132 33 ) .
j. karpinski , n. d. zhigadlo , s. katrych , z. bukowski , p. moll , s. weyeneth , h. keller , r. puzniak , m. tortello , d. daghero , r. gonnelli , i. maggio - aprile , y. fasano , .
fischer , k. rogacki , and b. batlogg , physica c 469 ( 2009 ) 370 . | the iron arsenide rbfe@xmath0as@xmath0 with the thcr@xmath0si@xmath0-type structure is found to be a bulk superconductor with @xmath1 k. the onset of diamagnetism was used to estimate the upper critical field @xmath2 , resulting in @xmath3 t / k and an extrapolated @xmath4 t. as a new representative of iron pnictide superconductors , superconducting rbfe@xmath0as@xmath0 contrasts with bafe@xmath0as@xmath0 , where the fermi level is higher and a magnetic instability is observed .
thus , the solid solution series ( rb , ba)fe@xmath0as@xmath0 is a promising system to study the cross - over from superconductivity to magnetism .
rbfe@xmath0as@xmath0 , iron pnictides , upper critical field , transition temperature , superconductivity 74.70.dd , 74.25.op |
in wireless networks , radio signals can be overheard by unauthorized users due to the broadcast nature of wireless medium , which makes the wireless communication systems vulnerable to eavesdropping attack .
secret key encryption techniques have been widely used to prevent eavesdropping and ensure the confidentiality of signal transmissions .
however , the cryptographic techniques rely on secret keys and introduce additional complexities due to the dynamic distribution and management of secret keys . to this end , physical - layer security is emerging as an alternative paradigm to prevent the eavesdropper attack and assure the secure communication by exploiting the physical characteristics of wireless channels .
the physical - layer security work was pioneered by wyner [ 1 ] and further extended in [ 2 ] , where an information - theoretic framework has been established by developing achievable secrecy rates .
it has been proven in [ 2 ] that in the presence of an eavesdropper , a so - called _ secrecy capacity _ is shown as the difference between the channel capacity from source to destination ( called main link ) and that from source to eavesdropper ( called wiretap link ) . if the secrecy capacity is negative , the eavesdropper can intercept the transmission from source to destination and an intercept event occurs in this case . due to the wireless fading effect , the secrecy capacity is severely limited , which results in an increase in the intercept probability . to alleviate this problem , some existing work is proposed to improve the secrecy capacity by taking advantage of multiple antennas [ 3 ] and [ 4 ] .
however , it may be difficult to implement multiple antennas in some cases ( e.g. , handheld terminals , sensor nodes , etc . ) due to the limitation in physical size and power consumption . as an alternative ,
user cooperation is proposed as an effective means to combat wireless fading , which also has great potential to improve the secrecy capacity of wireless transmissions in the presence of eavesdropping attack . in [ 5 ] , the authors studied the secrecy capacity of wireless transmissions in the presence of an eavesdropper with a relay node , where the amplify - and - forward ( af ) , decode - and - forward ( df ) , and compress - and - forward ( cf ) relaying protocols are examined and compared with each other .
the cooperative jamming was proposed in [ 6 ] by allowing multiple users to cooperate with each other in preventing eavesdropping and analyzed in terms of the achievable secrecy rate . in [ 7 ] ,
the cooperation strategy was further examined to enhance the physical - layer security and a so - called noise - forwarding scheme was proposed , where the relay node attempts to send codewords independent of the source message to confuse the eavesdropper . in addition , in [ 8 ] and [ 9 ] , the authors explored the cooperative relays for physical - layer security improvement and developed the corresponding secrecy capacity performance , showing that the cooperative relays can significantly increase the secrecy capacity . in this paper , we consider a cooperative wireless network with multiple df relays in the presence of an eavesdropper and examine the best relay selection to improve wireless security against eavesdropping attack .
differing from the traditional max - min relay selection criterion in [ 10 ] where only the channel state information ( csi ) of two - hop relay links ( i.e. , source - relay and relay - destination ) are considered , we here have to take into account additional csi of the eavesdropper s links , in addition to the two - hop relay links csi .
the main contributions of this paper are summarized as follows .
first , we propose the best relay selection scheme in a cooperative wireless networks with multiple df relays in the presence of eavesdropping attack .
we also examine the direct transmission without relay and traditional max - min relay selection as benchmark schemes .
secondly , we derive closed - form expressions of intercept probability for the direct transmission , traditional max - min relay selection , and proposed best relay selection schemes in rayleigh fading channels . the remainder of this paper is organized as follows .
section ii presents the system model and describes the direct transmission , traditional max - min relay selection , and proposed best relay selection schemes . in section iii
, we derive closed - form intercept probability expressions of the direct transmission , traditional max - min relay selection , and proposed best relay selection schemes over rayleigh fading channels . in section
iv , we conduct numerical intercept probability evaluation to show the advantage of proposed best relay selection over traditional max - min relay selection . finally , we make some concluding remarks in section v.
consider a cooperative wireless network consisting of one source , one destination , and @xmath0 df relays in the presence of an eavesdropper as shown in fig .
1 , where all nodes are equipped with single antenna and the solid and dash lines represent the main and wiretap links , respectively .
the main and wiretap links both are modeled as rayleigh fading channels and the thermal noise received at any node is modeled as a complex gaussian random variable with zero mean and variance @xmath1 , i.e. , @xmath2 .
following [ 8 ] , we consider that @xmath0 relays are exploited to assist the transmission from source to destination and the direct links from source to destination and eavesdropper are not available , e.g. , the destination and eavesdropper both are out of the coverage area . for notational convenience ,
@xmath0 relays are denoted by @xmath3 .
differing from the existing work [ 8 ] in which all relays participate in forwarding the source messages to destination , we here consider the use of the best relay only to forward the message transmission from source to destination .
more specifically , the source node first broadcasts the message to cooperative relays among which only the best relay will be selected to forward its received signal to destination .
meanwhile , the eavesdropper monitors the transmission from the best relay to destination and attempts to interpret the source message .
following [ 8 ] , we assume that the eavesdropper knows everything about the signal transmission from source via relay to destination , including the encoding scheme at source , forwarding protocol at relays , and decoding method at destination , except that the source signal is confidential .
it needs to be pointed out that in order to effectively prevent the eavesdropper from interception , not only the csi of main links , but also the wiretap links csi should be considered in the best relay selection .
this differs from the traditional relay selection in [ 10 ] where only the two - hop relay links csi is considered in performing relay selection . similarly
to [ 8 ] , we here assume that the global csi of both main and wiretap links are available , which is a common assumption in the physical - layer security literature . notice that the wiretaps link s csi can be estimated and obtained by monitoring the eavesdropper s transmissions as discussed in [ 8 ] . in the following ,
we first describe the conventional direct transmission without relay and then present the traditional max - min relay selection and proposed best relay selection schemes . for comparison purpose
, this subsection describes the conventional direct transmission without relay .
consider that the source transmits a signal @xmath4 ( @xmath5 ) with power @xmath6 .
thus , the received signal at destination is expressed as @xmath7 where @xmath8 represents a fading coefficient of the channel from source to destination and @xmath9 represents additive white gaussian noise ( awgn ) at destination . meanwhile , due to the broadcast nature of wireless transmission , the eavesdropper also receives a copy of the source signal @xmath4 and the corresponding received signal is written as @xmath10 where @xmath11 represents a fading coefficient of the channel from source to eavesdropper and @xmath12 represents awgn at eavesdropper . assuming the optimal gaussian codebook used at source , the maximal achievable rate ( also known as channel capacity ) of the direct transmission from source to destination is obtained from eq .
( 1 ) as @xmath13 where @xmath14 is the noise variance .
similarly , from eq .
( 2 ) , the capacity of wiretap link from source to eavesdropper is easily given by @xmath15 it has been proven in [ 2 ] that the secrecy capacity is shown as the difference between the capacity of main link and that of wiretap link .
hence , the secrecy capacity of direct transmission is given by @xmath16 where @xmath17 and @xmath18 are given in eqs .
( 3 ) and ( 4 ) , respectively . as discussed in [ 2 ] , when the secrecy capacity is negative ( i.e. , the capacity of main link falls below the wiretap link s capacity ) , the eavesdropper can intercept the source signal and an intercept event occurs .
thus , the probability that the eavesdropper successfully intercepts source signal , called _ intercept probability _ , is a key metric in evaluating the performance of physical - layer security . in this paper
, we mainly focus on how to improve the intercept probability by exploiting the best relay selection to enhance the wireless security against eavesdropping .
the following subsection describes the relay selection for physical - layer security in the presence of eavesdropper attack .
considering the df protocol , the relay first decodes its received signal from source and then re - encodes and transmits its decoded outcome to the destination .
more specifically , the source node first broadcasts the signal @xmath4 to @xmath0 relays that attempt to decode their received signals .
then , only the best relay is selected to re - encode and transmit its decoded outcome to the destination .
notice that in the df relaying transmission , the source signal @xmath4 is transmitted twice from the source and relay , respectively . in order to make a fair comparison with the direct transmission
, the total amount of transmit power at source and relay shall be limited to @xmath6 . by using the equal - power allocation for simplicity ,
the transmit power at source and relay is given by @xmath19 .
in df relaying transmission , either source - relay or relay - destination links in failure will result in the two - hop df transmission in failure , implying that the capacity of df transmission is the minimum of the capacity from source to relay and that from relay to destination . hence ,
considering @xmath20 as the best relay , we can obtain the capacity of df relaying transmission from source via @xmath20 to destination as @xmath21 where @xmath22 and @xmath23 , respectively , represent the channel capacity from source to @xmath20 and that from @xmath20 to destination , which are given by @xmath24 and @xmath25 meanwhile , the eavesdropper can overhear the transmission from @xmath20 to destination .
hence , the channel capacity from @xmath20 to eavesdropper can be easily obtained as @xmath26 combining eqs .
( 6 ) and ( 9 ) , the secrecy capacity of df relaying transmission with @xmath20 is given by @xmath27 the following presents the traditional max - min relay selection and proposed best relay selection schemes .
let us first present the traditional max - min relay selection scheme for the comparison purpose . in the traditional relay selection scheme , the relay that maximizes the capacity of df relaying transmission @xmath28 is viewed as the best relay .
thus , the traditional relay selection criterion is obtained from eq .
( 6 ) as @xmath29 which is the traditional max - min relay selection criterion as given by eq .
( 1 ) in [ 10 ] . as shown in eq . ( 11 ) , only the main links csi @xmath30 and @xmath31 is considered in the max - min relay selection scheme without considering the eavesdropper s csi @xmath32 .
we now propose the best relay selection criterion considering the csi of both main and wiretap links , in which the relay that maximizes the secrecy capacity of df relaying transmission is selected as the best relay .
thus , the best relay selection criterion is obtained from eq . ( 10 ) as @xmath33 one can observe from eq .
( 12 ) that the proposed best relay selection scheme takes into account not only the main links csi @xmath30 and @xmath31 , but also the wiretap link s csi @xmath32 .
this differs from the traditional max - min relay selection criterion in eq . ( 11 ) where only the main links csi is considered .
notice that the transmit power @xmath6 in eq .
( 12 ) is a known parameter and the noise variance @xmath34 is shown as @xmath35 [ 12 ] , where @xmath36 is boltzmann constant ( i.e. , @xmath37 ) , @xmath38 is room temperature , and @xmath39 is system bandwidth . since the room temperature @xmath38 and system bandwidth @xmath39 both are predetermined , the noise variance @xmath40 can be easily obtained .
it is pointed out that using the proposed best relay selection criterion in eq .
( 12 ) , we can further develop a centralized or distributed relay selection algorithm . to be specific , for a centralized relay selection
, the source node needs to maintain a table that consists of @xmath0 relays and related csi ( i.e. , @xmath30 , @xmath31 and @xmath32 ) . in this way ,
the best relay can be easily determined by looking up the table using the proposed criterion in eq .
( 12 ) , which is referred to as centralized relay selection strategy . for a distributed relay selection ,
each relay maintains a timer and sets an initial value of the timer in inverse proportional to @xmath41 in eq .
( 12 ) , resulting in the best relay with the smallest initial value for its timer . as a consequence ,
the best relay exhausts its timer earliest compared with the other relays , and then broadcasts a control packet to notify the source node and other relays [ 11 ] .
in this section , we derive closed - form intercept probability expressions of the direct transmission , traditional max - min relay selection and proposed best relay selection schemes over rayleigh fading channels .
let us first analyze the intercept probability of direct transmission as a baseline for the comparison purpose .
as is known , an intercept event occurs when the secrecy capacity becomes negative .
thus , the intercept probability of direct transmission is obtained from eq .
( 5 ) as @xmath42 where the second equation is obtained by using eqs . ( 3 ) and ( 4 ) .
considering that @xmath43 and @xmath44 are independent exponentially distributed , we obtain a closed - form intercept probability expression of direct transmission as @xmath45 where @xmath46 , @xmath47 , and @xmath48 is the ratio of average channel gain from source to destination to that from source to eavesdropper , which is referred to as the main - to - eavesdropper ratio ( mer ) throughout this paper .
it is observed from eq . ( 14 ) that the intercept probability of direct transmission is independent of the transmit power @xmath6 , which implies that the wireless security performance can not be improved by increasing the transmit power .
this motivates us to exploit cooperative relays to decrease the intercept probability and improve the physical - layer security .
this subsection presents the intercept probability analysis of traditional max - min scheme in rayleigh fading channels . from eq . ( 11 ) , we obtain an intercept probability of the traditional max - min scheme as @xmath49 where @xmath50 denotes the channel capacity from the best relay to eavesdropper with df relaying protocol .
similarly , assuming that @xmath51 and @xmath52 are identically and independently distributed and using the law of total probability , the intercept probability of traditional max - min scheme is given by @xmath53 notice that @xmath54 , @xmath55 and @xmath56 follow exponential distributions with means @xmath57 , @xmath58 and @xmath59 , respectively . letting @xmath60 , we obtain eq .
( 17 ) at the top of the following page , @xmath61 } \frac{1}{{\sigma _ { me}^2}}\exp ( - \frac{x}{{\sigma _ { me}^2}})dx } } \\ & = \sum\limits_{m = 1}^m { \frac{1}{m}\int_0^\infty { \left ( { 1 + \sum\limits_{k = 1}^{{2^m } - 1 } { { { ( - 1)}^{|{{\mathcal{a}}_k}|}}\exp [ - \sum\limits_{i \in { { \mathcal{a}}_k } } { ( \frac{x}{{\sigma _ { si}^2 } } + \frac{x}{{\sigma _ { id}^2 } } ) } ] } } \right)\frac{1}{{\sigma _ { me}^2}}\exp ( - \frac{x}{{\sigma _ { me}^2}})dx } } \\ & = \sum\limits_{m = 1}^m { \frac{1}{m}\left ( { 1 + \sum\limits_{k = 1}^{{2^m } - 1 } { { { ( - 1)}^{|{{\mathcal{a}}_k}|}}{[1 + \sum\limits_{i \in { { \mathcal{a}}_k } } { ( \frac{{\sigma _ { me}^2}}{{\sigma _ { si}^2 } } + \frac{{\sigma _ { me}^2}}{{\sigma _ { id}^2 } } ) } ] ^{- 1 } } } } \right ) } \end{split}\ ] ] where the second equation is obtained by using the binomial expansion , @xmath62 represents the @xmath63-th non - empty sub - collection of @xmath0 relays , and @xmath64 represents the number of elements in set @xmath62 .
this subsection derives a closed - form intercept probability expression of the proposed best relay selection scheme . according to the definition of intercept event , an intercept probability of proposed scheme
is obtained from eq .
( 12 ) as @xmath65 where the second equation is obtained by using eq .
notice that random variables @xmath54 , @xmath55 and @xmath66 follow exponential distributions with means @xmath57 , @xmath58 and @xmath67 , respectively . denoting @xmath68 , we can easily obtain the cumulative density function ( cdf ) of @xmath69 as @xmath70 wherein @xmath71 . using eq .
( 19 ) , we have @xmath72 which completes the closed - form intercept probability analysis of oas scheme in rayleigh fading channels .
2 shows the numerical intercept probability results of direct transmission , traditional max - min relay selection , and proposed best - relay selection schemes by plotting eqs .
( 14 ) , ( 17 ) and ( 20 ) as a function of mer .
one can see from fig .
2 that for both cases of @xmath73 and @xmath74 , the intercept probability of proposed best relay selection scheme is always smaller than that of traditional max - min relay selection , showing the advantage of proposed best relay selection over traditional max - min scheme . fig .
3 depicts the intercept probability versus the number of relays @xmath0 of the traditional max - min and proposed best relay selection schemes .
it is observed from fig .
3 that the proposed best relay selection scheme strictly performs better than the traditional max - min scheme in terms of the intercept probability .
3 also shows that as the number of relays @xmath0 increases , the intercept probabilities of both the traditional and proposed relay selection schemes significantly decrease .
this means that increasing the number of relays can greatly improve the physical - layer security against eavesdropping attack .
in this paper , we investigated the physical - layer security in cooperative wireless networks with multiple df relays and proposed the best relay selection scheme to improve wireless security against eavesdropping attack .
we also examined the direct transmission and traditional max - min relay selection as benchmark schemes .
we derived closed - form intercept probability expressions of the direct transmission , traditional max - min relay selection , and proposed best relay selection schemes .
numerical intercept probability results showed that the proposed best relay selection scheme always performs better than the direct transmission and max - min relay selection schemes .
moreover , as the number of relays increases , the max - min relay selection and proposed best relay selection schemes both significantly improve , which shows the advantage of exploiting multiple relays for the physical - layer security improvement .
the work presented in this paper is partially supported by the auto21 network of centre of excellence , canada .
e. tekin and a. yener , the general gaussian multiple access and two - way wire - tap channels : achievable rates and cooperative jamming , " _ ieee trans .
inf . theory _ ,
54 , no . 6 , pp . 2735 - 2751 , jun .
2008 .
y. zou , j. zhu , b. zheng , and y .- d .
yao , an adaptive cooperation diversity scheme with best - relay selection in cognitive radio networks , " _ ieee trans .
signal process .
5438 - 5445 , oct . | due to the broadcast nature of wireless medium , wireless communication is extremely vulnerable to eavesdropping attack .
physical - layer security is emerging as a new paradigm to prevent the eavesdropper from interception by exploiting the physical characteristics of wireless channels , which has recently attracted a lot of research attentions . in this paper
, we consider the physical - layer security in cooperative wireless networks with multiple decode - and - forward ( df ) relays and investigate the best relay selection in the presence of eavesdropping attack . for the comparison purpose
, we also examine the conventional direct transmission without relay and traditional max - min relay selection .
we derive closed - form intercept probability expressions of the direct transmission , traditional max - min relay selection , and proposed best relay selection schemes in rayleigh fading channels .
numerical results show that the proposed best relay selection scheme strictly outperforms the traditional direct transmission and max - min relay selection schemes in terms of intercept probability .
in addition , as the number of relays increases , the intercept probabilities of both traditional max - min relay selection and proposed best relay selection schemes decrease significantly , showing the advantage of exploiting multiple relays against eavesdropping attack .
intercept probability , best relay selection , eavesdropping attack , physical - layer security , cooperative wireless networks . |
in last three years the observation of a narrow baryon state named @xmath4 predicted by diakonov , petrov and polyakov@xcite has been reported by a large number of experiments in the @xmath5 or @xmath6 decay channels .
several experiments , mostly at high energies , did not confirm the existence of @xmath4 .
the present situation and complete list of references to positive and null results can be found in the reviews@xcite .
here we present a new study of the reaction @xmath7 , @xmath8 , @xmath9 , with two independent samples of @xmath2 used@xcite .
a detailed description of svd-2 detector and its trigger system can be found elsewhere@xcite .
the components of the detector used in current analyses are : the high - precision microstrip vertex detector with active(si ) and passive(c , pb ) nuclear targets ; large aperture magnetic spectrometer ; multicell threshold cherenkov counter . for the analyses we use data obtained in the 70 gev proton beam of ihep accelerator at intensity @xmath10 protons / cycle .
the total statistics of @xmath11 inelastic events was collected .
the sensivity of this experiment for inelastic @xmath12-interactions taking in account the triggering efficiency was @xmath13 .
svd-2 has performed the searches of @xmath4-baryon in two independent samples of the data selected by the @xmath2 decay point . in the first one @xmath2 decayed within the vertex detector
were used . the first step in
the analysis was to find out events with a well defined secondary vertex at the distance of 0 to 35 mm by the beam direction from the beginning of the active target ( corresponding to the sensitive area of the vertex detector ) .
tracks from secondary vertex were explored to the magnetic spectrometer and their momenta were reconstructed .
the @xmath6 invariant mass spectrum ( fig.[fig1 ] ) shows a clear peak at the @xmath14 with the significance of about 6.2@xmath15 ( 205 signal over 1050 background events ) . to estimate the natural width of the observed peak additional quality cuts
were used : 1 ) the distance of the closest approach between @xmath16 tracks @xmath17 3 standard deviations and 2 ) the number of hits on the track in the magnetic spectrometer @xmath18 12 ( of 18 present ) . taking into account the experimental resolution of the svd-2 detector ( calculated using well - known resonances ) it was estimated that intrinsic width of the @xmath19-resonance is @xmath20 at 95% c.l . for analysis
ii , the `` distant '' @xmath2 ( decay region is @xmath21 , outside the vertex detector ) were used .
@xmath2-mesons were identified by their charged decay mode @xmath22 . to eliminate contamination from @xmath23
decays , candidates with ( @xmath24 ) mass hypothesis giving less than @xmath25 were rejected .
about @xmath26 @xmath2-mesons from the mass window @xmath27 were found in the selected events .
proton candidates were selected as positively charged tracks with a number of spectrometer hits more than 12 with a momentum @xmath28 .
pions of such energies should leave a hit in the cherenkov counter , therefore absence of hits in tcc was also required .
effective mass of the @xmath29 system is plotted in fig.[fig2 ] .
an enhancement is seen at the mass @xmath30 with a @xmath31 with statistical significance of 5.6@xmath15 .
it was verified that observed @xmath6-resonance can not be a reflection from other ( for example @xmath32 ) resonances . the mechanism for producing spurious peak around @xmath33 involving @xmath2 and @xmath23 decays overlap
was also checked .
the events were scanned using svd-2 event display and no `` ghost '' tracks were found .
it is impossible to determine the strangeness of this state in such inclusive reaction , however we did not observe any narrow structure in @xmath34 invariant mass spectrum in @xmath35 mass area ( fig.[fig4]a ) . when applying a different cut on @xmath23 momentum , @xmath36 , we observe a structure near @xmath37 ( fig.[fig4]b ) .
this peak may correspond to the @xmath38 , marked as one - star resonance in the pdg review@xcite .
our search for @xmath4-particle is an inclusive experiment with a significant background contribution .
we have made an attempt to apply a subtraction method to investigate @xmath39 creation region in terms of @xmath40 .
an effective mass distribution in a peak region was fitted with a sum of background ( b ) and gaussian functions .
background was taken as a product of a threshold function and a second degree polinomial , .
all the fit parameters were given reasonable seeds but no boundaries , to prove a fit stability .
we have chosen a peak region as a gaussian mean @xmath41 2 gaussian @xmath15 .
a number of effective background events under the peak , @xmath42 , was evaluated by integrating background function over the peak region .
@xmath40 distributions were plotted separately for a peak and an out - of - peak regions(``wings '' ) ; the latter ranged from the threshold to the 1.7 gev with a peak region cut out , and a result was scaled to a @xmath42 .
assuming that the background characteristics are uniform , we subtract `` wings '' distribution from the peak one .
choosing more narrow `` wings '' does not change the general shape of distribution ( a rise at @xmath43 ) but shows larger fluctuations .
these operations were performed over the data from both analyses .
acceptance corrections were taken from the simulations and were specific for each type of analysis .
the results are shown at figs.[fig3 ] and [ fig4 ] .
we plot also normalized curves of the predictions made in a baranov regge - based model@xcite . in this model ,
overall distribution comes from the sum of quark fragmentation(bell - shaped at @xmath43 ) and diquark one(seagull - shaped ) , taken with somewhat arbitrary weights . in @xcite the weights are taken as 1:10 , as coming from the analysis of non - exotic baryons creation .
our data may indicate some favoring to the quark fragmentation part of the model .
note that the analysis ii has a specific narrow acceptance in x due to the proton momentum restrictions .
we `` projected '' the @xmath40-result of analysis i to a second one to check the consistence of our observations .
for that , the _ inverse _ acceptance correction for analysis ii was applied to the result of analysis i. we found a plausible agreement of the distribution shapes ( fig.[fig6 ] ) and some difference in a total number of events .
the latter makes a contribution to the cross section error calculations . as positive as negative results on the theta particle search are very often presented as the ratio of theta to @xmath44 cross sections . in our case , we estimate it as 0.04 .
however we should note that these two particles may have quite different creation mechanisms .
for example , extracting @xmath40 for @xmath44 the same way as described above resulted in much more forward - oriented distribution ( not shown here ) , than for theta particle .
it weakens the reasons of making this comparison .
one may suppose that a best way to present such a ratio would be to accompany it with corresponding acceptances .
the re - evaluation of the theta creation cross section gives @xmath45 @xmath46 for @xmath47 .
it differs significantly from our previous estimates@xcite .
the main reason is an unusual form of @xmath40 distribution , assumed flat in our first publications .
taking into account our situation of using inclusive data , and undiscovered yet mechanism of theta particle creation ( that means difficulties in acceptance evaluation ) , we believe that the cross section value is subject to future investigations . in conclusion , svd-2 observes in two independent samples a signal in @xmath48 mass distribution with @xmath49,@xmath50 , significance of @xmath51 , @xmath45 @xmath46 for @xmath47 .
@xmath40 peaks at zero with , that agrees qualitatively to a regge - based model suggested by baranov@xcite . while in agreement with evidences of @xmath4 observation , there is no direct contradiction to null results in hadron - hadron fixed target collisions , mainly due to different acceptances at @xmath52 .
we thank ichep06 organizing committee for creating an excellent scientific atmosphere during the conference . s.p.baranov
kindly supplied us with the data used in his work@xcite , the discussions with him were also very useful . | the inclusive reaction @xmath0 was studied at ihep accelerator with @xmath1 proton beam using svd-2 detector .
two different samples of @xmath2 , statistically independent and belonging to different phase space regions were used in the analyses and a narrow baryon resonance with the mass @xmath3 was observed in both samples of the data . |
protostellar jets play an important role in star formation .
they are believed to be launched from the innermost parts of accretion disks around protostars , carrying away excess angular momentum from the disks , allowing the disk material to fall onto the protostars ( see the review by * ? ? ?
the launching mechanism of the jets is still uncertain , but can be constrained with the physical properties ( e.g. , morphology and kinematics ) of the jets . in particular , since most of the jets have knotty structures ( knots ) , studying the formation of these knots may help us to constrain how the jets are launched .
hh 211 is a well - studied jet consisting of a chain of well - defined knots , and thus it is a good candidate for studying the physical properties of the jet and the formation of the knots .
it is nearby , located in the ic 348 complex of perseus at a distance of @xmath2280 pc @xcite .
the jet is powered by a low - mass and low - luminosity ( @xmath33.6 l@xmath4 , corrected for the new distance ) class 0 protostar @xcite .
it has been detected in h@xmath5 @xcite , as well as sio and co .
it is highly collimated with a small - scale wiggle .
the wiggle appears to be reflection - symmetric and semi - periodical , and thus is likely due to an orbital motion of the jet source around a close binary companion @xcite .
similar wiggle structure was also observed later in other hh jets ( e.g. , hh 111 in the infrared in * ? ? ?
such an orbiting jet source model has also been proposed to produce the reflection - symmetric wiggle .
however , it is also possible that the wiggle structure is due to an instability in the jet .
therefore , in this paper , we compare the jet structures in sio ( j=8 - 7 ) in four epochs to ensure that the wiggle structure is indeed due to an orbital motion of the jet source , instead of an instability .
in addition , using more epochs of observations , we can refine the proper motion measurement of the knots , and thus the velocity and mass - loss rate of the jet derived earlier in @xcite .
the formation mechanism of the knots in the jet is still not well determined .
the knots seem to be a train of dense clumps produced by episodic ejections .
however , current simulations show that even if the mass - loss rate of the jet is constant , a small periodical variation in ejection velocity can also produce the knots . in these simulations ,
the knots are formed as the fast jet material catches up with the slow jet material .
this possibility is supported by previous observations that show a velocity variation ( e.g. , radial velocity ramps ) in the knots along the jet axis , not only in optical ( e.g. , * ? ? ?
* ; * ? ? ?
* ) but also in submillimeter wavelengths ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
notice that the jet can be well mapped in both wavelengths . in order to further support this possibility ,
we track some well - defined knots of hh 211 in sio ( j=8 - 7 ) in four epochs and check if the evolution of the velocity variation in the knots is consistent with that expected in the simulations .
the goal is to determine if the knots are formed because of episodic ejections of jet material or because of a small periodical variation in ejection velocity .
archive data of the submilimeter array ( sma ) were used for our study . in the archive , four epochs of sio ( j=8 - 7 ) observations of hh 211 were found , one in 2004 , one in 2008 , one in 2010 , and one in 2013 . the observations in 2010 and 2013 epochs were carried out in polarization mode .
the array configurations , resolutions , and sensitivities are listed in table 1 . these resolutions and sensitivities are sufficient for our study . in all epochs ,
continuum emission of the central source was detected at @xmath0 340 ghz , allowing us to center the jet for proper motion measurement .
the data in the 2004 , 2008 , and 2013 epochs have been reported in @xcite , @xcite , and @xcite , respectively .
here we only describe the data in the 2010 epoch . in this epoch , the hh 211 jet was observed on january 2010 in the extended configuration . like other epochs , the sio line and the 340 ghz continuum were observed using 345 ghz band receivers .
the receivers have two sidebands : a lower sideband with frequency ranging from 333.5 to 337.5 ghz , and an upper sideband with frequency ranging from 345.5 to 349.5 ghz .
each sideband has 48 chunks , and each chunk has 128 spectral channels with a velocity resolution of @xmath0 0.7 km s@xmath1 per channel . after combining these two sidebands without the line emission portion , we obtained a continuum with a bandwidth of @xmath0 7.5 ghz centered at @xmath0 340 ghz .
the projected baseline lengths range from 30 to 186 m. the primary beam has a size of @xmath0 35@xmath6 .
we used mir package to calibrate the visibility data .
the quasar 3c454.3 , quasars 3c84 and j0325 + 469 , and titan were used as passband , gain , and flux calibrator respectively .
the flux uncertainty is @xmath0 15 % .
we then used the miriad package to image the calibrated visibility data .
the calibrated visibility data were used to produce dirty maps and a dirty beam .
we cleaned the maps to obtain clean component maps .
the final images were obtained by restoring the clean component maps with a gaussian beam fitted to the main lobe of the dirty beam . with natural weighting ,
the synthesized beam has a size of @xmath7 at a position angle ( p.a . ) of @xmath0 63@xmath8 .
the rms noise level is @xmath0 0.1 jy beam@xmath1 for the sio channel map and @xmath0 1 mjy beam@xmath1 for the continuum map .
figure [ siorotate ] shows the sio maps of the hh 211 jet in four epochs .
the sio maps are rotated by clockwise to align the jet with the y - axis .
knots bk1 to bk6 indicate the blueshifted ( eastern ) knots , and rk1 to rk5 and rk7 indicate the redshifted ( western ) knots , as in @xcite .
all the maps are convolved with the same synthesized beam with a size of @xmath7 at a p.a . of @xmath0 70@xmath8 , and then plotted with the same contour levels .
notice that the noise level in the 2004 epoch ( @xmath0 5 @xmath7 10@xmath9 jy beam@xmath1 ) is about twice as those in the other three epochs ( @xmath0 2 @xmath7 10@xmath9 jy beam@xmath1 ) .
therefore , the contour levels of the map in the 2004 epoch start at @xmath0 0.1 jy beam@xmath1 , which corresponds to the third contour level in the other epochs
. the jet shows knotty and wiggle structures on either side of the source ( fig.[siorotate ] ) .
some knots , bk4 , bk5 , and rk4 , seem to become fainter in the later epochs .
the wiggle in the 2008 epoch was found to be reflection - symmetric about the central source and semi - periodical @xcite .
this kind of wiggle is likely due to an orbital motion of the jet source in a binary system and can be fitted reasonably well by such a model @xcite .
this wiggle of the jet is also seen in other epochs , with a position shift due to proper motion of the jet . as shown by the curvy lines in figure [ siorotate ] , the wiggles in other epochs
can be all fitted well with the orbiting jet source model after including a position shift due to the proper motion of the jet ( to be derived later ) .
therefore , the wiggle propagates along the jet axis and indeed must be due to an orbital motion of the jet source , not due to an instability in the jet .
since the knots are localized and their structures are similar in different epochs , the peak positions of the knots can be used to estimate the proper motion of the knots and thus the jet .
previously , @xcite used the 2004 and 2008 epochs to estimate the proper motion . here , we use two more epochs , 2010 and 2013 , providing a total of four epochs spanning over @xmath0 9 years , to refine the proper motion . in all epochs , the peaks of the continuum emission of the central source are used to center the jet .
six knots bk2 , bk3 , bk4 , bk6 , rk4 , and rk7 appear like single knots and thus are used for the proper motion measurement , as in @xcite . in figure [ siorotate ] ,
the triangles mark the peak positions of those knots , and the lines connect the peak positions for the same knots . fitting the peak positions of the same knots with a linear equation @xmath10 , [ where @xmath11 , @xmath12 , and @xmath13 stand for the peak position , the epoch of observation , and the slope ( i.e. , proper motion ) , respectively ] , we can obtain the proper motion , @xmath13 , of the knots averaged over the four epochs .
figure [ fitv ] shows a plot of the proper motion of the knots versus the distance of the knots from the central source .
the proper motion of the knots ranges from @xmath0 to @xmath0 per year .
since the knots are not spatially resolved , the errorbar in the proper motion here is estimated by dividing the possible error in the peak position of the knots by the time span ( @xmath0 9 years ) over the four epochs .
assuming that the possible error in the peak position is less than one third of the beam size along the jet axis , which is @xmath0 , the errorbar in the proper motion is @xmath0 per year .
the mean proper motion on the redshifted side ( @xmath0 per year as indicated as the dotted line in figure [ fitv ] ) is slightly smaller than that on the blueshifted side ( @xmath0 per year as indicated as the dashed line in figure [ fitv ] ) .
since the jet is almost in the plane of the sky , this difference may suggest the jet to have different velocities on each side , as seen in , e.g. , sz 102 @xcite
. however , this difference could also be due to an uncertainty in the center position of the jet , which is defined by the peak position of the continuum emission of the central source .
the mean proper motion of all the knots is @xmath0 @xmath14 per year , resulting in a transverse velocity of @xmath0 114 @xmath15 km s@xmath1 .
the proper motion here is @xmath0 30@xmath16 lower than that measured in @xcite using the first two epochs , which was @xmath0 @xmath14 per year .
this difference can be seen in figure [ siorotate ] , which shows that the slopes of the lines connecting the first two epochs are larger than those connecting the last two epochs for most of the knots .
this decrease of the slopes is unlikely to be real because it is unlikely for the knots at different distances to decelerate to roughly the same value at the same time ( figure [ fitv ] ) .
the decrease may be due to the poorest resolution in the 2004 epoch , so that the knots in that epoch are more unresolved than the knots in other epochs .
in addition , as discussed later , since sio traces the shocks in the knots and the shocks evolve with time , the detailed structures of the knots may change , and thus the peak position of the knots may not be from the same part of the knots , as the knots propagate
. we can now refine the velocity and the inclination angle for the jet , using the new mean proper motion measured here and the mean radial velocity of the jet measured in @xcite . since the mean radial velocity was found to be @xmath0 17 km s@xmath1 , the velocity of the jet is @xmath0 115 @xmath15 km s@xmath1 and is @xmath0 30@xmath16 lower than that measured before .
the inclination angle is @xmath0 9@xmath8 , slightly larger than that ( @xmath0 6@xmath8 ) measured before .
we can study the kinematics of the knots using the position - velocity ( pv ) diagram of the jet along the jet axis in sio emission , as shown in figure [ sioblue ] . here , we only present the pv diagram for the eastern side of the jet because the knots there have well - defined single - knot structures .
we use the 2013 epoch because the pv structures of knots bk2 and bk3 are clear .
these two knots are each associated with a linear pv structure consisting of two or more emission peaks ( marked by the triangles in the figure ) , with the fast material upstream to the slow material .
the solid lines connecting the peaks indicate the velocity gradients of the linear structures , and the gradient decreases from knot bk2 to bk3 .
knot bk6 could also be associated with such a linear pv structure , but a higher sensitivity observation is needed to confirm it . the innermost knot , bk1 , shows a much larger velocity range and a much longer structure than the other knots .
it may contain several unresolved sub - knots @xcite , showing a complicated pattern in the pv diagram . for knots bk2 and bk3 , the linear pv structures are also seen in earlier epochs in 2008 and 2010 , as shown in figure [ gradient ] ( presented with the original resolutions as listed in table 1 ) .
such a linear structure is also seen in the 2004 epoch in knot bk3 , but not in knot bk2 due to poor resolution . in the figure ,
the solid lines show the velocity gradients of the linear structures by connecting the emission peaks .
as can be seen , the gradients decrease with time from the 2004 to 2013 epochs as the knots propagate .
this decrease in the gradients is likely due to the time evolution of the shocks to be discussed later .
the dashed lines mark the peak positions of the knots in the sio maps shown in figure [ siorotate ] , and they are located at different positions on the pv structures at different epochs .
this would introduce an uncertainty in our proper motion measurement based on the emission peaks as discussed in the previous section .
the formation mechanism for a chain of knots in the jet is still undetermined . in hh 211 ,
the knots have been found to be located inside outflow cavities @xcite . here , their proper motions are found to be roughly the same ( even though there could be a slight difference on the opposite sides of the jet , as discussed in section 3.2 ) and independent of the distance from the central source ( figure [ fitv ] ) , indicative of no obvious deceleration of the knot motion along the jet axis .
this result supports the assumption that the knots are indeed located inside the cavities and thus are not decelerated due to interaction with the ambient medium .
this result also suggests that the underlying jet velocity is roughly constant and that the knots are formed due to an intrinsic property of the jet .
two possible mechanisms have been proposed for the formation of a chain of knots in the jet .
the first mechanism is that of episodic ejections of dense jet material . @xcite
have studied and ruled out this mechanism for the knots in the jets of planetary nebulae .
their simulations can also be applied to the knots of protostellar jets . in their simulations ,
the ejection velocity was assumed to be constant , and episodic ejections of dense jet material produce a chain of knots propagating away from the source at constant velocity , as seen in our observations .
however , if this is the case , the fast material would appear downstream of the slow material in the knots , inconsistent with our observations that show the fast material upstream of the slow material .
in addition , previous co maps also show a continuous structure of the jet extending out from the source to knot bk2 on the east and to knot rk2 on the west .
since co traces roughly the column density of the material , the jet is continuous near the source due to continuous ejection .
therefore , the knots here are unlikely to be produced by episodic ejections of jet material .
the second possible mechanism is a small periodical variation in ejection velocity .
extensive simulations have been performed to study this mechanism by , e.g. , @xcite , @xcite , , and @xcite . in those simulations , the ejection velocity of the jet
is assumed to have a small periodical and sinusoidal variation , producing a chain of knots propagating away from the central source at a roughly constant velocity , as seen in our observations . in this case , as the fast jet material catches up with the slow jet material , an internal shock region consisting of two shocks ( a backward shock at high velocity and a forward shock at low velocity ) and an internal working surface in between are formed , as shown in figure [ shock ] .
since sio is known to be a good shock tracer , the knots seen here in sio are expected to trace such shock regions . in the simulations ,
the velocity structure for a newly formed shock region is roughly linear , with the fast material upstream to the slow material .
therefore , knots bk2 and bk3 , which show such velocity structures , can be newly formed shock regions .
the velocity gradient of the pv structure of knot bk2 in 2008 epoch is steep , further supporting knot bk2 to be a newly formed shock region .
an internal shock region does not form immediately near the source because it takes time for the fast material to catch up with the slow material . in the simulations ,
the jet density is usually assumed to be constant ( although some with a small variation ) . in this case , the jet will appear to be continuous before the internal shock region fully forms .
as mentioned earlier , previous co maps containing both shocked and non - shocked material , already show such a continuous structure of the jet extending out from the source to knot bk2 , supporting this scenario . here in sio ,
knot bk1 is elongated and extended to knot bk2 ( figure [ siorotate ] ) , and thus may trace a region of the continuous structure where an internal shock is still forming .
the fact that knot bk1 has a larger velocity range than those of knots bk2 and bk3 also supports this possibility . to further support this possibility , we estimated the distance for an internal shock to fully form , using equation 3 in @xcite . assuming that the jet has a velocity of @xmath17 115 km s@xmath1 ( as derived earlier ) , and that the velocity variation has an amplitude of @xmath18 30 km s@xmath1 ( as inferred from the velocity range of knot bk1 in figure [ sioblue ] ) and a period of @xmath19 20 years ( since the interknot spacing is @xmath0 2 ) , then the distance for an internal shock to fully form is @xmath20 1.4 @xmath7 10@xmath21 cm ( @xmath0 ) .
this distance is in between knots bk1 and bk2 , supporting that knot bk1 traces a region where an internal shock is still forming and that knot bk2 traces a newly formed internal shock region . in the simulations ,
as the newly formed internal shocks propagate away from the central source , they expand , causing a decrease in the velocity gradient , as shown in figure [ shock ] of this paper and figure 8c of @xcite .
this expansion can also account for the observed decrease in the velocity gradient with time as the knots propagate , and the observed decrease in the velocity gradient from knot bk2 to bk3 .
this further supports that knots bk2 and bk3 are indeed the newly formed internal shocks . in knot
bk2 , the two peaks in the pv structures of knot bk2 can be identified as two shocks ( backward and forward ) . in knot bk3 , more than two peaks are seen in the pv structures . in this case , the one at the highest velocity and the one at the lowest velocity can trace the two shocks , and the ones in between may trace the unresolved internal working surface ( higher resolution observations are needed to confirm this ) . in summary ,
although episodic ejections of jet material can produce a chain of knots as seen in the observations , they seem to have difficulty in accounting for the velocity structure of the knots and the continuous structure of the jet near the source . on the other hand , a small variation in ejection velocity of the jet with a constant jet density
can account for the major features seen in our observations .
therefore , the mass - loss rate of the jet is likely to be roughly constant and the knots trace the internal shocks in the jet produced by a small velocity variation . with the new jet velocity
, we can refine the mass - loss rate of the jet and the accretion rate toward the central source derived before in @xcite .
the mass - loss rate of the two - sided jet is now @xmath22 m@xmath4 yr@xmath23 , about 30% lower than that derived before .
thus , the mechanical luminosity of the jet becomes @xmath24 using the mean jet velocity , @xmath25 , of @xmath0 115 km s@xmath1 .
this mechanical luminosity is about one third of the bolometric luminosity of the source . assuming that both the bolometric luminosity of the source and the mechanical luminosity of the jet come from the accretion toward the source
, we can derive the accretion rate of the source . if the central source is a single protostar with a total mass of @xmath26 60 m@xmath27 @xcite and a stellar radius of @xmath28 2 r@xmath4 @xcite ,
the accretion rate is @xmath29 , about 38% lower than that derived before .
therefore , the mass - loss rate is @xmath0 23% of the accretion rate , still consistent with those predicted in the jet launching models @xcite .
the central source could be a binary because of the reflection - symmetric wiggle of the jet @xcite . with the new jet velocity of @xmath0 115 km s@xmath1
, the mass of the jet source ( primary ) is found to have @xmath30 , and the mass of the secondary source ( companion ) is @xmath31 . since the bolometric luminosity scales with the mass of source , the jet source can have two thirds of the total bolometric luminosity . in this case , the accretion rate toward the jet source would be @xmath32 .
thus , the mass - loss rate is @xmath0 20% of the accretion rate , also consistent with those predicted in the jet launching models @xcite .
only one jet is seen from the binary system probably because the mass of the secondary source is small .
we have used 4 epochs ( 2004 , 2008 , 2010 , and 2013 ) of sma archive data to study the properties of the hh 211 jet in sio ( j=8 - 7 ) .
the jet consists of a chain of knots and shows a small - scale wiggle , as seen before .
the conclusions are the following : \1 .
the wiggle structures are similar in different epochs , and they can be fitted by an orbiting jet source model that includes a position shift due to proper motion of the jet .
this confirms that the wiggle structures propagate away from the central source and thus are due to an orbital motion of the jet source , not due to an instability .
six knots are used for proper motion measurements .
the proper motion of the knots are roughly the same ( @xmath0 114 km s@xmath1 ) and independent of the distance from the central source , indicative of no obvious deceleration of knot motion along the jet axis .
this supports the assumption that the knots are indeed located inside outflow cavities .
knots bk2 and bk3 have well - defined linear velocity structures with the fast jet material upstream to the slow jet material .
the velocity gradients decrease from knot bk2 to bk3 .
in addition , for each knot , the gradient decreases with time , as the knots propagate away from the central source .
these results are expected if these two knots trace internal shocks produced by a small periodical variation of ejection velocity of the jet .
the emission peaks seen in the velocity structures may trace different parts of the internal shocks .
k .- s.j . and c .- f.l .
acknowledge grants from the national science council of taiwan ( nsc 101 - 2119-m-001- 002-my3 , most 104 - 2119-m-001 - 015-my3 ) and the academia sinica ( career development award ) .
we also thank anthony moraghan for useful discussion . | hh 211 is a highly collimated jet with a chain of well - defined knots , powered by a nearby young class 0 protostar .
we have used 4 epochs ( 2004 , 2008 , 2010 , and 2013 ) of submillimeter array ( sma ) archive data to study the properties of the hh 211 jet in sio ( j=8 - 7 ) .
the jet shows similar reflection - symmetric wiggle structures in all epochs .
the wiggle structures can all be fitted by an orbiting jet source model that includes a position shift due to proper motion of the jet , indicating that the wiggle propagates along the jet axis .
thus , this suggests the wiggle is indeed due to an orbital motion of the jet source .
proper motions of the knots are measured by using the peak positions of the knots in four epochs , and they are roughly the same and independent of the distance from the central source . the mean proper motion of the knots is @xmath0 per year , resulting in a transverse velocity of @xmath0 114 km s@xmath1 , about 30% lower than that measured before
. knots bk2 and bk3 have a well - defined linear velocity structure , with the fast jet material upstream to the slow jet material .
the gradient of the velocity structure decreases from knot bk2 to bk3 .
in addition , for each knot , the gradient decreases with time , as the knot propagates away from the central source .
these results are both expected if the two knots trace internal shocks produced by a small periodical variation in ejection velocity of the jet . |
in the 1970s many - body theory flourished .
a topic that attracted much attention was the many - body effects on xps ( x - ray photoelectron spectroscopy ) spectra , from deep core levels in metals@xcite .
related effects were singularities that appeared near edges in absorption and emission spectra@xcite .
the cb ( conduction - band ) electrons can have many different effects on the xps peak from the core level . if the excitation could be considered adiabatic the screening of the core hole by the cb electrons would lead to a shift of the peak to higher kinetic energy ( lower binding energy of the core level ) .
now the excitation is not adiabatic , the excitation is swift .
the excitation frequency is large compared to the frequency components taking part in the screening of the core hole .
this has the effect that the cb electrons are shaken up and are left in a nonequilibrium state .
this is analogous to the following case : assume that a cork is floating in a glass of water .
if we remove the cork very slowly the water will be left very quiet and without any ripples .
if we on the other hand remove it very briskly the water will be left in an upset state with many surface waves excited . in a metal
the water waves correspond to plasmons , collective excitations of the cb electrons .
the shake - up effects will show up as plasmon replicas in the spectrum ; the main peak will be at the adiabatic position ; the first replica , a smaller copy of the main peak , will be at a shifted position to lower kinetic energy where the shift is equal to the plasma frequency ; the second replica is even smaller and shifted with two plasma frequencies ; this goes on until further peaks are absorbed by the background .
the main peak corresponds to the system left in quasi equilibrium(fully equilibrium would demand that the core hole were filled by one of the electrons ) , the second to a state where one plasmon is excited , the third to a state where two plasmons are excited and so on .
since all plasmons do not have identical energies , the plasmon curve shows dispersion , the replicas are not completely identical smaller versions of the main peak . in a metal also
single - particle excitations can take place .
these electron - hole pair excitations form a continuum , starting from zero frequency and upwards .
these excitations lead to a deformation , including a low - energy tail , of both the main peak and the plasmon replicas .
furthermore the finite life - time of the core hole causes a lorentzian broadening of all peaks and experimental uncertainties give a gaussian broadening .
the low - energy tailing is a characteristic of a metallic system , i.e. a system where the chemical potential is inside an energy band and not in a band - gap , and can e.g. be used by experimentalists to find out where in a complex sample the core hole is situated .
several parameters were introduced by doniach and sunji@xcite to characterize the xps line shapes and are still broadly used by experimentalists in the fitting procedure for xps spectra . in the present work we address pristine and doped graphene . in the pristine case
the chemical potential is neither inside an energy band nor in a band - gap .
the fermi surface is just two points in the brillouin zone .
this makes this system special .
as we will see there is still a low - energy tailing . in the doped case
the chemical potential is inside an energy band and we would expect to find , and find , a tailing .
however , the 2d ( two - dimensional ) character of the system means that the collective excitations are 2d plasmons with a completely different dispersion than in the ordinary 3d metallic systems .
the 2d plasmons give contributions that start already from zero frequency and upwards .
this means that they contribute to the tail and no distinct plasmon replicas are distinguishable .
making any quantitative interpretation of graphene core - hole spectra using doniach and sunji fitting of the spectra is not feasible .
it is more reasonable to calculate the spectra .
one purpose of this work is to provide the reader with the tools needed for such a calculation .
the material is organized in the following way : in sec . [
the ] we show how the core - hole spectra are derived for a 2d system .
the results and comparison with experimental spectra are presented in sec .
finally , we end with a brief summary and conclusion section , sec .
we use a model that is based on the one used by langreth@xcite for the core - hole problem in the 1970s and here modified and extended to fit our problem .
we have earlier@xcite with success used another modified version for the problem of exciton annihilation in quantum wells . in the excitation process
the photoelectron leaves the system and a core hole is left behind .
the shape of the xps spectrum depends on how fast the process is .
if it is very slow one may use the adiabatic approximation in which one assumes that the electrons in the system have time to relax around the core hole during the process .
when we derive the xps line shape we assume that the excitation process is very fast ; we use the sudden approximation in which the core - hole potential is turned on instantaneously .
the electrons have not time , during the process , to settle down and reach equilibrium in the potential from the core hole .
this results in shake - up effects in the form of single particle ( electron - hole pair ) excitations and collective ( plasmon ) excitations .
the electrons contributing to the shake - up effects are the electrons in the valence and conduction bands . from here on we refer to them as the electrons .
we use the assumption that the core hole does not recoil in the shake - up process and that there are no excitations within the core .
we approximate the core - hole potential with a pure coulomb potential .
the following model hamiltonian for the system is used : @xmath0 where @xmath1 , @xmath2 , @xmath3 , @xmath4 , and @xmath5 are the core - hole energy , the creation operator for the core hole , the annihilation operator for the core hole , the interaction potential for the interaction between the core hole and the electrons , and the hamiltonian for the electrons , respectively .
the operators @xmath4 and @xmath5 contain creation and annihilation operators for the electrons and no core - hole operators .
the core - hole number operator is @xmath6 .
it commutes with the hamiltonian , i.e. , @xmath7 = 0 $ ] . in the ground state of the system
there is no core hole and the electrons are in their ground state .
let us introduce two hamiltonians @xmath8 where the first is the hamiltonian before the core hole has been created and the second the hamiltonian after the creation .
let @xmath9 denote the ground state of @xmath10 .
it is also the ground state of @xmath11 .
then we have @xmath12 where @xmath13 is the ground state energy of the electron system . from now on we
drop the subscript @xmath14 on the core - hole operators and introduce the functions @xmath15 these functions are connected to the time ordered and retarded green s functions , according to @xmath16 and @xmath17\left| 0 \right\rangle \\
\quad \quad = - i\theta \left ( t \right)\left [ { { g^ > } ( t ) - { g^ < } ( t ) } \right ] .
\end{array } \label{equ6}\ ] ] the fourier transformed versions , @xmath18 have a direct physical meaning .
their sum is the spectral function , @xmath19 is the density of states for an occupied core - hole state , and @xmath20 for an unoccupied .
note that the creation and annihilation operators are for holes .
we want the density of states for a core electron .
thus we need to calculate @xmath20 .
we first determine the time dependent form .
we have @xmath21t/\hbar } } { c^\dag } \left| 0 \right\rangle \\ = { e^{ie_0^et/\hbar } } \left\langle 0 \right|c{e^ { - i\varepsilon nt/\hbar } } { e^ { - i\left [ { nv + { h_e } } \right]t/\hbar } } { c^\dag } \left| 0 \right\rangle \\ = { e^{i\left ( { e_0^e - \varepsilon } \right)t/\hbar } } \left\langle 0 \right|c{e^ { - i\left [ { v + { h_e } } \right]t/\hbar } } { c^\dag } \left| 0 \right\rangle \\ = { e^{i\left ( { e_0^e - \varepsilon } \right)t/\hbar } } { } _ e\left\langle 0 \right|{e^ { - i\left [ { v + { h_e } } \right]t/\hbar } } { \left| 0 \right\rangle _ e}{}_{ch}\left\langle 0 \right|1 - n{\left| 0 \right\rangle _ { ch}}\\ = { e^{i\left ( { e_0^e - \varepsilon } \right)t/\hbar } } \left\langle 0 \right|{e^ { - i\left [ { v + { h_e } } \right]t/\hbar } } \left| 0 \right\rangle , \end{array } \label{equ8}\ ] ] where we have let the operators operate to the left and right .
we have made use of the general relation @xmath22}}$ ] , for operators @xmath23 and @xmath24 , and that the core - hole number operator commutes with all terms of the hamiltonian .
it is now time to define our @xmath4 and the rest of the hamiltonian .
the excitation spectrum of the electrons is given by the dynamical structure factor , which is related to the dielectric function according to @xmath25 where @xmath26 is the 2d fourier transform of the coulomb potential .
the structure factor is nonzero in two regions ; one region is where electron - hole pairs are excited ; the other is on the plasmon dispersion curve .
all these excitations are bosons .
we assume that the bosons are independent .
let us for simplicity start by assuming that the bosons have a distinct dispersion curve , @xmath27 .
this is the case for plasmons and phonons .
later we will generalize this to include contributions from the electron - hole pair continuum .
we now get the following hamiltonian : @xmath28 where @xmath23 is the area of the 2d system and @xmath29 the coupling constant between the core hole and the boson excitation .
the operators @xmath30 and @xmath31 are boson creation and annihilation operators , respectively , obeying the commutation relation @xmath32 = { \delta _ { { \bf{q}},{\bf{q}}'}}$ ] and @xmath33 are the density operators , where @xmath34 denotes the position of the core hole . we treat the core hole as a classical particle , i.e. , the core - hole operators are c - numbers .
this is why the density operators do not contain any creation and/or annihilation operators .
let us here take the opportunity to derive the results using the adiabatic approximation .
since there is no kinetic energy terms for the core hole the hamiltonian can be diagonalized .
this is achieved by using the following unitary transformation : @xmath35 , \label{equ12}\ ] ] where @xmath36 letting @xmath37 we find that @xmath38 = { f^\dag } ( - { \bf{q } } ) = f({\bf{q}})$ ] and @xmath39 = { f^\dag } ( { \bf{q}})$ ] .
this gives @xmath40 the transformation of the hamiltonian gives @xmath41 } \\
- \frac{1}{{{a^{{1 \mathord{\left/ { \vphantom { 1 2 } } \right .
\kern-\nulldelimiterspace } 2}}}}}\sum\limits_{\bf{q } } { g({\bf{q}})\rho _ { ch}^\dag ( { \bf{q } } ) } \left ( { { c_{\bf{q } } } + f({\bf{q } } ) + c _ { - { \bf{q}}}^\dag + { f^\dag } ( - { \bf{q } } ) } \right)\\ = \sum\limits_{\bf{q } } { \hbar { \omega _ { \bf{q}}}\left ( { c_{\bf{q}}^\dag { c_{\bf{q } } } + { 1 \mathord{\left/ { \vphantom { 1 2 } } \right . \kern-\nulldelimiterspace } 2 } } \right ) } - \frac{1}{{{a^{{1 \mathord{\left/ { \vphantom { 1 2 } } \right . \kern-\nulldelimiterspace } 2}}}}}\sum\limits_{\bf{q } } { g({\bf{q}})\rho _ { ch}^\dag ( { \bf{q } } ) } \left ( { { c_{\bf{q } } } + c _ { - { \bf{q}}}^\dag } \right)\\ + \sum\limits_{\bf{q } } { \hbar { \omega _ { \bf{q}}}\left ( { c_{\bf{q}}^\dag f({\bf{q } } ) + { c_{\bf{q}}}{f^\dag } ( { \bf{q } } ) + { f^\dag } ( { \bf{q}})f({\bf{q } } ) } \right ) } \\ - \frac{2}{{{a^{{1 \mathord{\left/ { \vphantom { 1 2 } } \right .
\kern-\nulldelimiterspace } 2}}}}}\sum\limits_{\bf{q } } { g({\bf{q}})\rho _ { ch}^\dag ( { \bf{q } } ) } f({\bf{q } } ) .
\end{array } \label{equ16}\ ] ] substituting for the expression of @xmath42 we find @xmath43 thus @xmath44 where @xmath45 we see that the transformation of the hamiltonian in eq.([equ10 ] ) has the effect that the interaction term is changed into a constant energy term .
the interactions with the core hole produce an energy shift of the ground state .
this is the relaxation energy ; the gain in energy when the electrons relax around the core hole .
this is the shift of the xps peak ( towards higher kinetic energy ) one would get in the adiabatic approximation .
we now return to the sudden approximation and our green s function @xmath46{t \mathord{\left/ { \vphantom { t \hbar } } \right .
\kern-\nulldelimiterspace } \hbar } } } \left| 0 \right\rangle $ ] .
let us insert the identity on both sides of the exponential inside the ground state matrix element .
@xmath47{t \mathord{\left/ { \vphantom { t \hbar } } \right .
\kern-\nulldelimiterspace } \hbar } } } u\left| 0 \right\rangle . \end{array } \label{equ20}\ ] ] now , since there is no boson excited in the ground state we may write @xmath48{t \mathord{\left/ { \vphantom { t \hbar } } \right . \kern-\nulldelimiterspace } \hbar } } } \left| 0 \right\rangle = { e^{i\left ( { \sum\limits_{\bf{q } } { { \textstyle{1 \over 2}}\hbar { \omega _ { \bf{q } } } } } \right){t \mathord{\left/ { \vphantom { t \hbar } } \right .
\kern-\nulldelimiterspace } \hbar } } } \left| 0 \right\rangle \\
= { e^{ie_0^h{t \mathord{\left/ { \vphantom { t \hbar } } \right .
\kern-\nulldelimiterspace } \hbar } } } \left| 0 \right\rangle .
\end{array } \label{equ21}\ ] ] thus we have @xmath49{t \mathord{\left/ { \vphantom { t \hbar } } \right .
\kern-\nulldelimiterspace } \hbar } } } u{e^{i\left [ { \sum\limits_{\bf{q } } { \hbar { \omega _ { \bf{q}}}\left ( { c_{\bf{q}}^\dag { c_{\bf{q } } } + { 1 \mathord{\left/ { \vphantom { 1 2 } } \right . \kern-\nulldelimiterspace } 2 } } \right ) } } \right]{t \mathord{\left/ { \vphantom { t \hbar } } \right .
\kern-\nulldelimiterspace } \hbar } } } \left| 0 \right\rangle \\
= { e^ { - i\left ( { \varepsilon + \delta \varepsilon } \right){t \mathord{\left/ { \vphantom { t \hbar } } \right .
\kern-\nulldelimiterspace } \hbar } } } \left\langle 0 \right|{u^\dag } \left ( 0 \right)u\left ( t \right)\left| 0 \right\rangle , \end{array } \label{equ22}\ ] ] where @xmath50 and we see that @xmath51 is also a unitary transformation . now , from eqs.([equ12 ] ) and ( [ equ23 ] ) we find @xmath52\;\\ = \exp \left [ { \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } ( { c_{\bf{q}}}{e^{i{\omega _ { \bf{q}}}t } } - c _ { - { \bf{q}}}^\dag { e^ { - i{\omega _ { \bf{q}}}t } } ) } \right]\\ = \exp \left [ { \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } ( - c _ { - { \bf{q}}}^\dag { e^ { - i{\omega _ { \bf{q}}}t } } + { c_{\bf{q}}}{e^{i{\omega _ { \bf{q}}}t } } ) } \right ] . \end{array } \label{equ24}\ ] ] if we now once again make use of the relation @xmath22}}$ ] we find @xmath53\exp \left [ { \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } { c_{\bf{q}}}{e^{i{\omega _ { \bf{q}}}t } } } \right]\\ \quad \quad \times \exp \left [ { { \textstyle{1 \over 2}}\sum\limits_{{\bf{k}},{\bf{l } } } { { f^\dag } ( { \bf{l } } ) } { f^\dag } ( { \bf{k}}){e^ { - i{\omega _ { \bf{l}}}t}}{e^{i{\omega _ { \bf{k}}}t}}\underbrace { \left [ { c _ { - { \bf{l}}}^\dag , { c_{\bf{k } } } } \right ] } _ { - { \delta _ { { \bf{k } } , - { \bf{l } } } } } } \right]\\ = \exp \left [ { - \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } c _ { - { \bf{q}}}^\dag { e^ { - i{\omega _ { \bf{q}}}t } } } \right]\exp \left [ { \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } { c_{\bf{q}}}{e^{i{\omega _ { \bf{q}}}t } } } \right]\\ \quad \quad \times \exp \left [ { - { \textstyle{1 \over 2}}\sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } { f^\dag } ( - { \bf{q } } ) } \right]\\ = \exp \left [ { - \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } c _ { - { \bf{q}}}^\dag { e^ { - i{\omega _ { \bf{q}}}t } } } \right]\exp \left [ { \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } { c_{\bf{q}}}{e^{i{\omega _ { \bf{q}}}t } } } \right]\\ \quad \quad \times \exp \left [ { - { \textstyle{1 \over 2}}\sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } f({\bf{q } } ) } \right]\\ = \exp \left [ { - \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } c _ { - { \bf{q}}}^\dag { e^ { - i{\omega _ { \bf{q}}}t } } } \right]\exp \left [ { \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } { c_{\bf{q}}}{e^{i{\omega _ { \bf{q}}}t } } } \right]\\ \quad \quad \times \exp \left [ { - { \textstyle{1 \over 2}}\sum\limits_{\bf{q } } { { { \left| { f({\bf{q } } ) } \right|}^2 } } } \right ] .
\end{array } \label{equ25}\ ] ] we also need @xmath54\exp \left [ { - \sum\limits_{\bf{q } } { f({\bf{q } } ) } { c _ { - { \bf{q } } } } } \right]\\ \quad \quad \times \exp \left [ { - { \textstyle{1 \over 2}}\sum\limits_{\bf{q } } { { { \left| { f({\bf{q } } ) } \right|}^2 } } } \right ] .
\end{array } \label{equ26}\ ] ] substituting the results of eqs.([equ25 ] ) and ( [ equ26 ] ) in eq.([equ22 ] ) gives @xmath55\exp \left [ { - \sum\limits_{\bf{q } } { f({\bf{q } } ) } { c _ { - { \bf{q } } } } } \right]\\ \quad \quad \times \exp \left [ { - { \textstyle{1 \over 2}}\sum\limits_{\bf{q } } { { { \left| { f({\bf{q } } ) } \right|}^2 } } } \right]\exp \left [ { - \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } c _ { - { \bf{q}}}^\dag { e^ { - i{\omega _ { \bf{q}}}t } } } \right]\\ \quad \quad \times \exp \left [ { \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } { c_{\bf{q}}}{e^{i{\omega _ { \bf{q}}}t } } } \right]\exp \left [ { - { \textstyle{1 \over 2}}\sum\limits_{\bf{q } } { { { \left| { f({\bf{q } } ) } \right|}^2 } } } \right]\left| 0 \right\rangle . \end{array } \label{equ27}\ ] ] the factors that do not contain boson operators can be taken outside .
then we can let the left most factor operate to the left .
it produces a factor of unity .
the rightmost factor may operate to the right and it also produces a factor of unity .
we are left with @xmath56 here we make use of the relation @xmath57}}$ ] and find @xmath58\exp \left [ { - \sum\limits_{\bf{q } } { { { \left| { f({\bf{q } } ) } \right|}^2 } } } \right]\\ \times \left\langle 0 \right|\exp \left [ { - \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } c _ { - { \bf{q}}}^\dag { e^ { - i{\omega _ { \bf{q}}}t } } } \right]\\ \times \exp \left [ { - \sum\limits_{\bf{q } } { f({\bf{q } } ) } { c _ { - { \bf{q } } } } } \right]\\ \times \exp \left\ { { \left [ { - \sum\limits_{\bf{q } } { f({\bf{q } } ) } { c _ { - { \bf{q } } } } , - \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } c _ { - { \bf{q}}}^\dag { e^ { - i{\omega _ { \bf{q}}}t } } } \right ] } \right\}\left| 0 \right\rangle \\ = \exp \left [ { - i\left ( { \varepsilon + \delta \varepsilon } \right){t \mathord{\left/ { \vphantom { t \hbar } } \right . \kern-\nulldelimiterspace } \hbar } } \right]\exp \left [ { - \sum\limits_{\bf{q } } { { { \left| { f({\bf{q } } ) } \right|}^2 } } } \right]\\ \times \left\langle 0 \right|\exp \left [ { - \sum\limits_{\bf{q } } { f({\bf{q } } ) } { c _ { - { \bf{q } } } } } \right]\\ \times \exp \left\ { { \left [ { - \sum\limits_{\bf{q } } { f({\bf{q } } ) } { c _ { - { \bf{q } } } } , - \sum\limits_{\bf{q } } { { f^\dag } ( { \bf{q } } ) } c _ { - { \bf{q}}}^\dag { e^ { - i{\omega _ { \bf{q}}}t } } } \right ] } \right\}\left| 0 \right\rangle \\
= \exp \left [ { - i\left ( { \varepsilon + \delta \varepsilon } \right){t \mathord{\left/ { \vphantom { t \hbar } } \right . \kern-\nulldelimiterspace } \hbar } } \right]\exp \left [ { - \sum\limits_{\bf{q } } { { { \left| { f({\bf{q } } ) } \right|}^2 } } } \right]\\ \times \left\langle 0 \right|\exp \left [ { - \sum\limits_{\bf{q } } { f({\bf{q } } ) } { c _ { - { \bf{q } } } } } \right]\\ \times \exp \left\ { { \sum\limits_{{\bf{k}},{\bf{l } } } { f({\bf{k}}){f^\dag } ( { \bf{l}}){e^ { - i{\omega _ { \bf{l}}}t } } } \underbrace { \left [ { { c _ { - { \bf{k}}}},c _ { - { \bf{l}}}^\dag } \right]}_{{\delta _ { { \bf{k}},{\bf{l } } } } } } \right\}\left| 0 \right\rangle \\ = \exp \left [ { - i\left ( { \varepsilon + \delta \varepsilon } \right){t \mathord{\left/ { \vphantom { t \hbar } } \right . \kern-\nulldelimiterspace } \hbar } } \right]\exp \left [ { - \sum\limits_{\bf{q } } { { { \left| { f({\bf{q } } ) } \right|}^2 } } } \right]\\ \times \exp \left [ { \sum\limits_{\bf{q } } { { { \left| { f({\bf{q } } ) } \right|}^2}{e^ { - i{\omega _ { \bf{q}}}t } } } } \right]\\ \times \left\langle 0 \right|\exp \left [ { - \sum\limits_{\bf{q } } { f({\bf{q } } ) } { c _ { - { \bf{q } } } } } \right]\left| 0 \right\rangle \\ = { e^ { - i\left ( { \varepsilon + \delta \varepsilon } \right){t \mathord{\left/ { \vphantom { t \hbar } } \right .
\kern-\nulldelimiterspace } \hbar } } } { e^ { - \sum\limits_{\bf{q } } { { { \left| { f({\bf{q } } ) } \right|}^2 } } } } { e^{\sum\limits_{\bf{q } } { { { \left| { f({\bf{q } } ) } \right|}^2}{e^ { - i{\omega _ { \bf{q}}}t } } } } } .
\end{array } \label{equ29}\ ] ] thus we have @xmath59 we want the fourier transformed version .
it is @xmath60 where the so - called satellite generator is @xmath61 $ ] for doped graphene .
we see the typical 2d plasmon dispersion .
the plot is for the doping density @xmath62 but the result is very similar for any finite doping density . , width=226 ] in the unperturbed case , i.e. , when there is no interaction with the core hole we have @xmath63 so we can write @xmath64 where @xmath65 now we generalize this result to include all excitation processes in our system .
we first rewrite eq .
( [ equ35 ] ) @xmath66 where @xmath67 is the boson propagator .
now the boson propagator is replaced by @xmath68 times the dynamical structure factor of eq.([equ9 ] ) and the coupling constant by @xmath69 .
thus we get @xmath70 $ ] for doped graphene .
the plot is for the doping density @xmath62 but the result is very similar for any finite doping density.,width=207 ] the xps spectrum reflects the density of states of the core electron and becomes @xmath71}}{e^ { - a(t ) } } } \\
= \frac{1}{\pi } \int\limits_0^\infty { dt\cos \left [ { \left ( { \omega - d } \right)t + b\left ( t \right ) } \right]{e^ { - a(t ) } } } , \end{array } \label{equ38}\ ] ] where @xmath72,\\ b\left ( t \right ) = - \frac{{2{e^2}}}{{\pi \hbar } } \int\limits_0^\infty { \int\limits_0^\infty { dqd\omega } } \frac{1}{{{\omega ^2}}}{\mathop{\rm im}\nolimits } \frac { { - 1}}{{\varepsilon \left ( { q,\omega } \right)}}\sin \left ( { \omega t } \right),\\ d = \frac{{2{e^2}}}{{\pi \hbar } } \int\limits_0^\infty { \int\limits_0^\infty { dqd\omega } } \frac{1}{\omega } { \mathop{\rm im}\nolimits } \frac { { - 1}}{{\varepsilon \left ( { q,\omega } \right)}}. \end{array } \label{equ39}\ ] ] we give the energy relative to the core - level position when the interaction with the electrons are neglected . taking the finite life time of the core hole into account we arrive at the central theoretical result of this work , the xps spectrum .
the xps spectrum can be written as @xmath73dt } , \label{equ40}\ ] ] where @xmath74\left [ { 1 - \cos \left ( { wt } \right ) } \right]dq } dw } , \end{array } \label{equ41}\ ] ] @xmath75\sin \left ( { wt } \right)dq } dw } , \label{equ42}\ ] ] and @xmath76dq } dw}. \label{equ43}\ ] ] $ ] for pristine graphene .
the scaling of @xmath77 and @xmath78 is the same as in fig.[figu2].,width=207 ] all variables have been scaled according to @xmath79 and are now dimensionless .
the two last relations will be used below .
the quantity @xmath80 is the energy shift of the adiabatic peak .
we have taken the finite life - time of the core hole into account by introducing the first factor of the integrand of eq.([equ40 ] ) , where @xmath81 is the fwhm ( full width at half maximum ) of the lorentz broadened peak .
we have here assumed that the core - hole potential can be represented by a pure coulomb potential .
the results are valid for a general 2d system .
the particular system enters the problem through @xmath82 $ ] . for pristine graphene the dielectric function is@xcite @xmath83 where we have used the values@xcite @xmath84 m
/ s for the fermi velocity @xmath85 and @xmath86 for @xmath87 , respectively .
the background dielectric constant @xmath87 is the result of high frequency electronic excitations to higher lying empty energy bands and from lower lying filled energy bands .
when the graphene sheet is doped the dielectric function becomes much more complicated .
however it has been derived by several groups@xcite . the dielectric function in a general point in the complex frequency plane , @xmath77 , away from the real axis is@xcite @xmath88 } \right\};}\\ { f\left ( { q , w } \right ) = { \rm{asin}}\left ( { \frac{{1 - w}}{q } } \right ) + { \rm{asin}}\left ( { \frac{{1 + w}}{q } } \right)}\\ { \qquad \qquad - \frac{{w - 1}}{q}\sqrt { 1 - { { \left ( { \frac{{w - 1}}{q } } \right)}^2 } } + \frac{{w + 1}}{q}\sqrt { 1 - { { \left ( { \frac{{w + 1}}{q } } \right)}^2 } } , } \end{array } \label{equ46}\ ] ] where @xmath89 is the density of states at the fermi level and @xmath90 is the doping concentration .
in both eqs .
( [ equ45 ] ) and ( [ equ46 ] ) we let @xmath77 be real valued but add a small imaginary part to get the retarded forms of the dielectric functions . in figs.[figu1 ] and [ figu2 ]
we show the contour plot and surface plot , respectively , of @xmath91 $ ] for doped graphene .
the integrands in eqs.([equ41 ] ) , ( [ equ42 ] ) , and ( [ equ43 ] ) all have an extra factor @xmath92 for small @xmath77 which means that small energy transfers are important in the shake up structure of the spectra . in fig.[figu3 ] we give the corresponding surface plot for pristine graphene . here
the excitations are of electron - hole pair type .
we have taken the lorentzian broadening into account but how should we include the gaussian instrumental broadening ? here we use a trick .
if all shake up processes were involving only one discrete frequency , @xmath93 , the @xmath94- , @xmath95- , and @xmath96-functions would become @xmath97,\\ { b_g}\left ( t \right ) = - \eta \sin \left ( { { w_0}t } \right),\\ { d_g } = \eta { w_0},\\ \eta = \int { \frac{{{d^2}q}}{{{{\left ( { 2\pi } \right)}^2}}}\frac{{{{\left| { g\left ( { \bf{q } } \right ) } \right|}^2}}}{{{{\left ( { \hbar { \omega _ 0 } } \right)}^2 } } } } , \end{array } \label{equ47}\ ] ] -xps ) c 1s core - level spectrum of a suspended graphene sheet obtained in ref.@xcite using the photon energy 480 ev , red circles .
the solid curve is the theoretical result for an undoped free - standing graphene sheet .
a linear background was subtracted from the experimental data .
we used lorentzian and gaussian broadening of .4 ev and .666 ev , respectively , width=302 ] where @xmath98 is a strength constant and the spectrum would consist of a series of delta functions , @xmath99 but now for a gold supported graphene sheet .
the theoretical curve was obtained assuming a doping density of @xmath100 .
we used the same lorentzian and gaussian broadening as in fig.[figu5].,width=302 ] the amplitudes of the delta functions form a poisson distribution .
why is it a poisson distribution ?
the average number of bosons surrounding the core hole is @xmath98 .
the bosons do not interact and the core hole does not recoil when a boson is excited .
thus the probability that a boson is excited in a given instant does not depend on how many bosons are already excited .
then , according to probability theory the probability for having exactly @xmath90 bosons excited at a certain time is given by the poisson distribution .
when the strength parameter @xmath98 is large this turns into a gaussian distribution .
it is close to a gaussian already at @xmath101 .
the gaussian fwhm is given by @xmath102 .
the trick is now to choose a small enough value for @xmath103 so that @xmath98 becomes large enough .
then we add the resulting @xmath94- , @xmath95- , and @xmath96-functions to the original functions in eq.([equ40 ] ) .
thus we get the final spectrum with both lorentzian and gaussian broadened structures from @xmath104 } } } \right . } \\
\quad \quad \quad
\quad \times \cos \left\ { { \left [ { w - \left ( { d + { d_g } } \right ) } \right]t - \left [ { b\left ( t \right ) + { b_g}\left ( t \right ) } \right ] } \right\ } } \right ] . \end{array } \label{equ49}\ ] ] this is the relation we have used in finding the xps spectra presented in figs.[figu4]-[figu7 ] .
in fig.[figu4 ] we compare our results to an experimental xps spectrum@xcite represented by red circles .
the experiment was performed on a single quasi - free - standing epitaxial graphene layer on sic obtained by hydrogen intercalation@xcite .
this leads to a virtually undoped graphene sheet .
the photon energy used was 750 ev .
the leftmost peak is from graphene c1s and the rightmost from the c1s core level in the sic substrate .
the dashed curve is the result one would get if there were no interaction between the electrons and the core hole .
another way to express this is to say that the electron wave functions are frozen .
the energies are given relative to the position of this peak .
the peak is symmetric and has the full broadening but no structure from shake - up effects .
the thick solid curve is the adiabatic peak .
this is what the spectrum would look like if the excitation were adiabatic , i.e. , the excitation were so slow that the electrons were left in a quasiequilibrium state ; the core - hole potential was fully screened .
this peak is just the dashed peak shifted by the energy @xmath80 .
the thin solid curve is the full result from eq.([equ49 ] ) .
when we used the dielectric function from pristine graphene the curve emerged a little above the experimental tail .
when we added a very small doping level , @xmath105 , the agreement with experiment was perfect .
the reason for that a small amount of doping pushes down the tail a little is that there are some extra contributions at the top of the peak and the normalization then leads to a reduction away from the center of the peak . in figs.[figu5 ] and
[ figu6 ] we compare our results for pristine and doped free - standing graphene to the experimental results in ref.@xcite .
the experiments were performed using a photon energy of 480 ev on a suspended single graphene sheet and on a gold supported single graphene sheet . in the first sample
the graphene is more or less undoped ; in the second gold provides p - doping .
we subtracted a linear background from both experimental sets of data and used the same broadening parameters for both spectra .
the main peak in the doped sample is broader than in the undoped but this extra broadening comes from different shake - up effects in the two samples .
the theoretical spectrum was obtained assuming a doping density of @xmath100 .
this number @xmath106 should not be taken too seriously .
we varied the density in equidistant steps on a logarithmic scale and tried @xmath107 , @xmath108 and @xmath109 .
the one in the middle gave the best fit with experiments .
, from all spectra .
the symmetric peak is the result from the fully adiabatic approximation and the result when there is no interaction between the core hole and the electron - hole system .
the red dashed curve is the result from the undoped graphene sheet , the blue short - dashed from a sheet with doping concentration @xmath110 , and the green dotted curve from a sheet with doping concentration @xmath111 , respectively .
the curve for doping concentration @xmath112 has been omitted since it is very close to the curve for pristine graphene . ,
width=302 ] finally , in fig.[figu7 ] we show the theoretical results for different doping levels . in order to more clearly see the effect doping has on the peak shape
we have here removed all different shifts , @xmath80 , for the curves .
the solid black curve is the noninteracting curve or the adiabatic curve ( it is the same for all doping levels now when the shift has been removed ) .
the red dashed curve is the pristine result .
the blue short - dashed curve is for the doping density @xmath110 , and the green dotted curve from a sheet with doping concentration @xmath111 .
important doping effects show up first at doping concentrations exceeding @xmath112 .
the reason is that some electron - hole - pair excitation - channels are blocked with doping due to the pauli exclusion principle ; this is compensated by new electron - hole - pair excitation - channels and the new plasmon - excitation channel . in a recent work@xcite one came to the same conclusion as we regarding the need to make a full calculation of the core - hole spectra in graphene instead of using the doniach and sunji fitting of the peaks .
they treated the band - structure of graphene in a different way compared to our and relied on density functional theory .
they seem to have assumed a constant 2d plasmon frequency and thereby obtained sharp plasmon structures in the tail region of the peaks .
such structures are not observed in the experimental spectra .
to summarize , we have presented a detailed derivation of the core - level spectra of 2d systems and presented numerical results for pristine and doped free - standing graphene .
although , pristine graphene is not a metal its core - level spectrum shows a peak tailing , characteristic of metallic systems . the tailing increases with doping for doping concentrations exceeding @xmath112 .
the peak shape changes with further increase of the doping concentration .
this opens up for a complementary way to estimate the degree of doping of a sample .
we have compared our results to three different experimental spectra from two experimental groups .
the agreement is quite good which is very encouraging .
we have furthermore introduced a convenient way to introduce the effect of gaussian instrumental broadening in the formalism .
we thank professor leif i. johansson and professor chung - lin wu for providing us with their experimental data .
p. h. citrin , g. k. wertheim , and y. baer , phys .
lett . * 35 * , 885 ( 1975 ) .
p. h. citrin , g. k. wertheim , and y. baer , phys .
lett . * 41 * , 1425 ( 1978 ) .
s. doniach and m. sunji , j. phys .
c : solid state phys . * 3 * , 285 ( 1970 ) .
g. d. mahan , phys .
b * 11 * , 4814 ( 1975 ) .
david c. langreth , phys .
b * 1 * , 471 ( 1970 ) .
bo e. sernelius , and erik sderstrm , j. phys.:condens .
matter * 8 * 9071 ( 1996 ) .
j. gonz@xmath113les , f. guinea , and m. a. h. vozmediano , _ nucl .
phys . _ * 424 * , 595 ( 1994 ) .
b. wunsch , t .
stauber , f. sols , and f. guinea , _ new journal of physics _ * 8 * , 318 ( 2006 ) .
bo e. sernelius , _ epl _ * 95 * , 57003 ( 2011 ) .
e. h. hwang , and s. das sarma , phys .
b * 75 * , 205418 ( 2007 ) .
bo e. sernelius , phys .
b * 85 * , 195427 ( 2012 ) . the spectrum was not published before but how the sample was prepared is described in ref.@xcite .
j. hassan , c. virojanadara , a. meyer , i. g. ivanov , j. i. flege , s. watcharinyanon , j. falta , l. i. johansson , and e. janz ' en , mater .
forum * 717 - 720 * , 605 ( 2012 ) .
jhih - wei chen , chiang - lun wang , hung wei shiu , chi - yuan lin , chen - shiung chang , forest shih - sen chien , chia - hao chen , yi - chun chen , and chung - lin wu , applied physics express * 5 * , 085101 ( 2012 ) .
v. despoja , and m. sunji , phys .
b * 88 * , 245416 ( 2013 ) . | we calculate core - level spectra for pristine and doped free - standing graphene sheets .
instructions for how to perform the calculations are given in detail .
although pristine graphene is not metallic the core - level spectrum presents low - energy tailing which is characteristic of metallic systems .
the peak shapes vary with doping level in a characteristic way .
the spectra are compared to experiments and show good agreement .
we compare to two different pristine samples and to one doped sample .
the pristine samples are one with quasi - free - standing epitaxial graphene on sic obtained by hydrogen intercalation and one with a suspended graphene sheet .
the doped sample is a gold supported graphene sheet .
the gold substrate acts as an acceptor so the graphene sheet gets p - doped . |
by definition a wormhole ( wh ) is a bridge connecting two asymptotically flat regions .
usually such construction is a classical object and should satisfy to einstein equations ( see @xcite for more detailed introduction to this field ) .
the topology of the 4d wh is @xmath1 , where @xmath2 is the time , @xmath3 is the radial coordinate and @xmath4 is the cross section of the wh .
a space region near to the minimal cross section @xmath5 is called as a throat .
the basic problem for existing of the wh is concentrated on the throat : in this region a matter violates the so - called null energy condition . in this paper
we offer a model of the wh in which the throat is _ a set of quantum handles ( wormholes ) in the presence of the strong electric field_. these quantum handles can be considered as quantum wh s of a spacetime foam with separated mouthes ( see fig.[fig1 ] ) .
we remind that the hypothesized spacetime foam @xcite , @xcite is a set of quantum wormholes ( handles ) appearing in the spacetime on the planck scale level ( @xmath6 ) . for the macroscopic observer
these quantum fluctuations are smoothed and we have an ordinary smooth manifold with the metric submitting to einstein equations .
the presence of the spacetime foam with virtual wormholes ( vwh ) which have separated mouthes gives a physical effect : vwh s can entrap a part of the electric flux lines and in spite of the fact that at the infinity @xmath0 we will have a non - singular spacetime ( @xmath7 and @xmath8 are the charge and mass registered at the infinity ) .
the mechanism for creating the wh with quantum throat can be offered by the following way : 1 . the vwh s connect two regions with the strong electric field by such a way that the electric flux lines leak through vwh s from one region to another one @xcite , see fig.[fig2 ] .
each vwh is the solutions of the 5d einstein equations with @xmath9 .
3 . as usually the @xmath10 metric component can be considered as the 4d electric field which joins the electric flux lines of the above - mentioned two 4d regions .
4 . for this model
the whole spacetime is 5d one and outside of the throat the @xmath11 is non - dynamical variable and we have the 5d kaluza - klein theory in its initial interpretation ( @xmath12 and 5d gravity is equivalent to einstein - maxwell theory ) , inside of the throat the @xmath11 is dynamical variable and we have the ordinary kaluza - klein gravity with 4d metric + electromagnetic and scalar fields .
splitting off the 5d dimension takes place on the event horizon .
the matching conditions of the 4d and 5d fields on the event horizon is discussed in ref.@xcite . as
this composite wh is the classical solution of the 5d kaluza - klein equations a probability for the quantum creation of such vwh is not zero .
the model of the vwh is presented on the fig.([fig2 ] ) . on the fig.([fig3 ] ) are presented the two parts of the wh with the quantum throat . on the fig.[fig3]a
is presented the part of the wh with incoming force lines of the electric field and respectively on the fig.[fig3]b the part with outcoming force lines . for the 4d observer
each mouth is like to moving ( + /-
) electric charge .
but this movement is stochastic one and as a consequence we have @xmath13 @xmath14 where @xmath15 is the stochastical averaging . in ref.@xcite
is shown that such composite wh has a spin - like structure .
the cause of this is following .
the metric describing the 5d throat is @xmath16 where @xmath17 is the 5@xmath18 extra coordinate ; @xmath19 , @xmath20 ; @xmath21 are the @xmath22 polar coordinates ; here @xmath23 and @xmath24 are some constants .
we see that the signs of the @xmath25 and @xmath26 are not defined .
we remark that this 5d metric is located behind the event horizon therefore the 4d observer is not able to determine the signs of the @xmath25 and @xmath26 @xcite . moreover
this 5d metric ( [ ql1 ] ) can fluctuate between these two possibilities .
hence the external 4d observer is forced to describe such composite wh by means of a spinor .
this allows us to do the following assumption : the stochastical polarized spacetime foam approximately can be determinated by some effective field : a spinor field @xmath27 . in this case
a density @xmath28 of the vwh is @xmath29 where ( @xmath30 ) means the transposition and a density of an effective electric charge @xmath31 ( the mouthes of vwh entrapped the electric force lines ) is @xmath32 where @xmath7 is some charge .
as the hypothesized spacetime foam is a consequence of quantum gravity all our fields ( metric @xmath33 , electromagnetic filed @xmath34 and spinor field @xmath27 ) in this model should be quantized fields .
the interaction between these fields is very strong and we can not use the feynmann diagram technique . for the quantization of this model we will use the heisenberg quantization method which he applied for the non - linear spinor field @xcite .
the essence of this method consists in that the classical fields @xmath35 in field equations are excanched on the field operators @xmath36 . in this case
we have differential equations for operators .
certainly it is not clear what is a solution of such differential equations .
in fact heisenberg has shown that differential equations for the operator field is equivalent to some infinite set of differential equations for green functions ( for the small coupling constant this is dayson - schwinger equations system ) . following this way we write differential equations for the gravitational + electromagnetic fields in the presence of the spacetime foam @xmath37 as follows @xmath38 } - m \right ) \hat\psi & = & 0 , \label{ex2}\\ d_\nu \hat f^{\mu\nu } & = & 4\pi e \left ( \hat{\bar\psi}\hat\gamma^\mu \hat \psi \right ) \label{ex3}\end{aligned}\ ] ] where @xmath39 is the operator of ricci tensor ; @xmath40 is the operator of the sum of the energy - momentum tensor the spinor and maxwell fields ; @xmath41 is the operator of the gravitational field ; @xmath42 is the operator of the dirac matrices ; @xmath43 , @xmath44 $ ] means the antisymmetrization ; @xmath45 ( @xmath46 is the vier - bein index ) is the usual @xmath47-matrices @xmath48 where @xmath49 are the pauli matrices ; @xmath50 is the vier - bein operator ; greek indexes are the spacetime indexes ; latin indexes with the bar are vier - bein indexes ; @xmath51 is the operator of the maxwell tensor of the electromagnetic field ; @xmath52 is the operator of the potential ; @xmath8 and @xmath7 are some constants
. certainly this equation system is hopelessly complicated , and are impossible to find exact solutions .
we will consider equations for average values @xmath53 , @xmath54 , @xmath55 and so on .
in the first approximation we suppose that @xmath56 where @xmath57 can be @xmath39 , @xmath58 and so on ; @xmath59 can be @xmath41 , @xmath60 , @xmath52 and so on . in this case
we have the classical system of the einstein - dirac - maxwell equations , _
i.e. _ now the system ( [ ex1])-([ex3 ] ) is without ( @xmath61 ) .
fot our model we use the following ansatz : for the spherically symmetric metric @xmath62 for the electromagnetic potential @xmath63 for the spinor field @xmath64 the following is _ very important _ for us : the ansatz ( [ ex8 ] ) for the spinor field @xmath27 has the @xmath65 component of the energy - momentum tensor and the @xmath66 component of the current .
let we remind that @xmath27 determines the stochastical gas of the vwh s which can not have a preferred direction in the spacetime .
this means that after substitution expression ( [ ex6])-([ex8 ] ) into field equations they should be averaged by the spin direction of the ansatz ( [ ex8 ] ) in the expression ( [ ex5 ] ) is not only quantum but stochastical , too . ] .
after this averaging we have @xmath67 and @xmath68 and we have the following equations system describing our spherically symmetric spacetime @xmath69 where @xmath70 is some constant .
this equations system was investigated in @xcite and result is the following . a particle - like solution exists which has the following expansions near @xmath71 @xmath72 and the following asymptotical behaviour @xmath73 where @xmath74 is
the mass for the observer at infinity and @xmath75 is the charge of this solution .
our interpretation of this solution is presented on the fig.([fig4 ] ) .
the solution exists for @xmath76 and @xmath77 but for us is essential the first case with @xmath76 . in this case
the classical einstein - maxwell theory leads to the `` naked '' singularity .
the presence of the spacetime foam drastically changes this result : _ the appearance of the vwh s can prevent the formation of the `` naked '' singularuty in the reissner - nordstrm solution with @xmath0_.
thus our model is based on the following assumptions : * in basic the quantum ( virtual ) whs of the spacetime foam have the 5d throat . *
each quantum wh has a spin - like structure , * the spacetime foam effectively can be described with the help of a spinor field . as the consequence we have the result that the strong electric field separates the vwh s of the spacetime foam by such a way that they can prevent a singularity in the reissner - nordstrm solution ( with @xmath78 ) on account of the formation of vwhs . in the spirit of the einstein idea that the right - hand of gravitational equations should be zero we can note that this model of the wh with quantum throat is the vacuum model since the gauge fields can be considered as components of the metric in some multidimensional kaluza - klein gravity
i am grateful for financial support from the georg forster research fellowship from the alexander von humboldt foundation and h .- j .
schmidt for an invitation to potsdam university .
w. heisenberg , nachr .
gttingen * 8 * , 111(1953 ) ; w. heisenberg , zs .
naturforsch . , * 9a * , 292 ( 1954 ) ; w. heisenberg , f. kortel and h. mtter , zs . naturforsch . , * 10a * , 425 ( 1955 ) ; w. heisenberg , zs .
fr phys . , * 144 * , 1 ( 1956 ) ; p. askali and w. heisenberg , zs .
naturforsc . , * 12a * , 177 ( 1957 ) ; w. heisenberg , nucl . phys . , * 4 * , 532 ( 1957 ) ; w. heisenberg , rev .
* 29 * , 269 ( 1957 ) . | a wormhole with a quantum throat on the basis of an approximate model of the spacetime foam is presented .
an effective spinor field is introduced for the description of the spacetime foam .
the consequences of such model of the wormhole is preventing a `` naked '' singularity in the reissner - nordstrm solution with @xmath0 . |
galaxy formation is a complicated combination of many different physical processes , such as gravitational collapse ( eggen et al . 1962 ) , accretion of subgalactic clumps ( e.g. , searle & zinn 1978 ) , hierarchical merging of dark matter halos ( e.g. , white & rees 1978 ) , star formation and its feedback effects ( e.g. , larson 1974 ; dekel & silk 1986 ) , growth of bars and spiral arms ( e.g. , sellwood & carlberg 1984 ; athanassoula 2003 ) , clump dynamics ( e.g. , noguchi 1999 ) , feedback effects of active galactic nuclei ( e.g. , di matteo et al . 2005 ; springel et al . 2005 ) , hot and cold mode gaseous accretion ( e.g. , keres et al . 2005 ; dekel & birmboim 2006 ) , and environmental influences of groups and clusters ( e.g. , dressler 1980 ; bekki & couch 2011 ) .
it has been a key issue for galactic and extragalactic astronomy to understand the relative importance of each of these physical processes in the formation and evolution of galaxies with different masses and hubble types residing in different environments at a give cosmic time . in order to address this key issue ,
many theoretical studies have tried to construct as realistic galaxy formation models as possible by using sophisticated numerical simulations .
one of the missing elements in these previous numerical studies of galaxy formation and evolution is interstellar dust .
the efficient conversion from neutral to molecular hydrogen ( @xmath0 ) is possible on the surface of dust grains ( e.g. , gould & salpeter 1963 ; hollenbach & salpeter 1971 ) , and thermodynamical evolution of ism and interstellar chemistry are also controlled by dust ( e.g. herbst 2002 ) .
furthermore , fragmentation processes of metal - poor gas clouds can be significantly influenced by dust abundances so that dust can be very important for the formation of low - mass stars in the early universe ( e.g. , larson 2005 ; schneider & omukai 2010 ) . in spite of these importance of dust ,
dust - related physical processes have not been self - consistently included in almost all simulations of galaxy formation and evolution .
the observed physical processes of dust properties in galaxies have been investigated so far by one - zone chemical evolution models ( e.g. , dwek 1998 , d98 ; lisenfeld & ferrara 1998 ; hirashita 1999 ; edmunds 2001 ; inoue 2003 ; calura et al .
2008 ; asano et al . 2013 ; zhukovska & henning 2013 ; rowlands et al .
2014 ) .
bekki ( 2013a , 2015 ; b13a and b15 , respectively ) have first performed numerical simulations of galaxy formation and evolution which include self - consistently both ( i ) the formation , growth , and destruction processes of dust grains in asymptotic giant branch ( agb ) stars , supernovae ( sne ) , and ism and ( ii ) @xmath0 formation on dust grains .
thanks to the self - consistent modeling of dust physics , not only the cosmic evolution of dust but also the physical properties of galaxies ( e.g. , gas contents ) were compared with the latest observations ( e.g. , by herschel ) in b13a and b15 .
these simulations are different from previous ones ( ferrara et al .
1991 , f91 ; aguirre et al .
2001 ) in which dust grains are represented by test particles ( i.e. , moving in a gravitational potential of a galaxy without influencing its host galaxy at all ) so that the influences of dust on galaxy formation and evolution can not be investigated .
although our previous simulations made it possible for us to predict the cosmic evolution histories of dust abundances and distributions in galaxies , they have some problems in modeling dust .
one key problem is that gas and dust are assumed to move exactly the same way in our previous simulations ( b13a , b15 ) : the simulated spatial distributions and kinematics are the same between gas and dust in galaxies .
this could be a good approximation , given that the timescale of gas - dust frictional timescale is an order of @xmath1 yr for a typical range of ism properties ( e.g. , theis & orlova 2004 ) , which is much shorter than the dynamical timescales of most galaxies ( @xmath2 yr ) .
however , as shown in previous theoretical works on dust evolution in galaxies , only dust can be expelled from galactic disks through the effects of radiation pressure of stars on dust ( e.g. , f91 ) . also , the observed flat distribution of halo dust ( mnard et al .
2010 , m10 ) and the rather high dust - to - gas - ratio ( @xmath3 ) in m81 group ( e.g. , xilouris et al .
2006 ) strongly suggest that the spatial distributions of dust can be quite different from those of gas in some galaxy environments .
furthermore , chemical evolution of galaxies can be significantly influenced by dust removal from galaxies through radiation - driven winds ( bekki & tsujimoto 2014 ) , which implies that star formation processes in galaxies can be influenced by dust removal from galaxies owing to the processes dependent on dust and metal abundances .
therefore , we would need to model dust removal processes in galaxies more self - consistently so that we can discuss the roles of dust in galaxy formation and evolution and the observed wide variety of dust properties of galaxies in a more comprehensive way .
it would be a reasonable strategy for numerical studies of galaxy formation to treat gas and dust separately , because gas and dust respond differently to physical processes of ism .
the purpose of this paper is to describe a new method by which dust - related physics can be more self - consistently included in nbody / hydrodynamical simulations of galaxy formation and evolution .
dust and gas are represented by separate live particles in the present new simulation code so that dust - related physical processes such as radiation pressure on dust and dust - gas interaction can be better investigated . since this is the very first paper discussing the possible roles of dust in galaxy formation based on a new type of galaxy evolution simulations , we focus mainly on the methods to include dust - related physics in nbody / hydrodynamical simulations .
we briefly describe some key results of the new simulations , and we discuss extensively the possible roles of dust in galaxy formation and evolution in forthcoming papers .
the present model is not only a more sophisticated version of previous models on the formation of dust wind ( e.g. , chiao & wickamasinghe 1972 ; f91 ) but also an improved version of previous galaxy evolution models for dusty galaxies .
the important influences of radiation - driven dust wind on star - forming regions in galaxies have been recently discussed by a number of authors ( e.g. , thompson et al .
2005 ; hopkins et al 2012 ) .
these works , however , did not investigate the evolution of dust contents and the possibly separate distributions of gas and stars in a self - consistent manner . furthermore , previous models for spectral energy distributions ( sed ) of galaxies ( e.g. , bekki & shioya 2000 ; jonsson 2006 ) and for @xmath0 contents of galaxies ( e.g. , pelupessy et al .
2006 ; krumholz et al .
2009 ) assumed a fixed dust - to - metal ( @xmath4 ) ratio across the entire region of a simulated galaxy . such a fixed @xmath4 is neither observed ( e.g. , galametz et al .
2011 ) nor realistic in theoretical modeling of dust evolution ( e.g. , inoue 2003 ) .
therefore , the present model with variable @xmath4 in different local regions of a simulated galaxy can provide more detailed and accurate predictions on @xmath0 contents and seds of galaxies .
the plan of the paper is as follows .
first , we describe the details of the adopted new methods to implement dust - related physical processes in galaxy - scale chemodynamical simulations .
then , we present some key results of the new simulations on the roles of dust in the evolution of star formation rates , dust distributions , and gas dynamics within galaxies in 3 . in this section , we also discuss how the present results depend on the physical parameters of the adopted new method for dust - related physics . in 4 , we discuss ( i ) the advantages and disadvantages of the present new method over previous ones and ( ii ) further improvement of the method in our future works .
we summarize our conclusions in 5 .
although including dust - related physics in galaxy - scale nbody+hydrodynamical simulations is rather new in numerical studies of galaxy formation and evolution , there are already a number of more sophisticated simulations on the joint evolution of dust and gas in other areas of astronomy .
for example , laibe & price ( 2014 ) have already included the gas - dust interaction for multiple dust components in their sph simulations that can be used for the investigation of planet formation and evolution .
the methods and numerical techniques used in these simulations in other areas of astronomy would be useful in the present study .
however , we do not discuss these , because this is beyond the scope of this paper .
we accordingly focus on the modeling issues of galaxy - scale chemodynamical simulations with dust physics .
[ cols="<,^,^ , < " , ] the present new four - component chemodynamical code will enable us to investigate various aspects of dust - affected galaxy formation and evolution across the comic time .
the proposed live dust particle method will make it possible for us to discuss the different distributions of gas and dust in galaxies , the cosmic evolution of dust sizes and compositions , and the abundances of interstellar matters formed on dust grains .
below , we briefly discuss what we can investigate by using the new simulation code ( the figure in appendix c would be useful for readers of this paper to understand the following discussion better ) .
b15 failed to reproduce the observed flat profiles of dust halos of galaxies ( m10 ) , because the model adopted in b15 did not include dust - gas decoupling processes in ism ( e.g. , radiation pressure and dust - gas drag ) .
furthermore , the observed rather high dust - to - gas - ratio ( @xmath3 , which is about 6 times larger than the solar neighborhood ) in the outer halo of m81 by xilouris et al .
( 2006 ) was very difficult to be explained by b15 for the same reason .
as shown in this paper , the radial density profile of dust in a galaxy can be flattened by the effects of radiation pressure of stars ( thus the outer halo can have higher @xmath5 ) .
therefore , it is doubtlessly worthwhile for our future studies to investigate whether such an important effect of radiation pressure can be clearly achieved in cosmological hydrodynamical simulations of galaxy formation .
the present chemodynamical model includes the six key dust - related physical processes of galaxies , i.e. , dust formation , mixing , growth , destruction , consumption , and catalysis .
the formation of @xmath0 on dust grains is one of important interstellar chemistry processes on dust grains ( catalysis ) that can be discussed in the present study .
the present new live dust particle method combined with the proposed four - component galaxy model ( dm+stars+gas+dust ) can predict dust and gas 3d distributions separately and distributions and kinematics of different dust species ( e.g. , agb and sn dust ) separately too .
however , the live particle method is still less realistic in modeling some aspects of dusty ism , such as the size evolution of dust and the coagulation process of dust .
the bottleneck of the new method is that the total number of dust particles can endlessly increase if the formation of dust with different sizes and components in star - forming regions and evolved stars is considered in a fully self - consistent manner .
this means that the cpu / gpu time required for the calculation of dust physics at each time step of a simulation can become progressively longer as the simulation proceeds .
the total number of dust particles ejected from a single stellar particle ( @xmath6 ) is 3 in the present study so that the total number of particles ( @xmath7 ) can not dramatically increase .
however , if we investigate the time evolution of dust sizes , @xmath6 would possibly need to be more than @xmath8 to represent a statistically enough sample of dust grains with different sizes .
this means that we need to develop a fast algorism for calculating the dust - related physics of ism in a simulation in order to consider the dust size evolution in a self - consistent manner .
constructing the seds for arbitrary 3d geometries of gas , dust , and stars in galaxies based on theoretical and numerical models of galaxy formation has been one of key ares of theoretical studies of galaxies ( e.g. , bekki & shioya 2000 , 2001 ; jonsson 2006 ; popescu & tuffs 2013 ) .
these simulated seds and those based on some phenomenological models ( e.g. , dale et al .
2001 ) are very useful in interpreting the observed energy budgets ( e.g. , fraction of infrared light ) of dusty star - forming galaxies . in order to estimate seds from ssps ( e.g. , bruzual & charlot 2003 ) , the previous studies needed to assume ( i ) dust - to - metal - ratios ( to drive dust abundances from gas metallicities ) and ( ii ) dust size distribution ( for deriving the dust extinction curves ) .
if our future simulations with the new live dust particle method can predict dust size distributions properly , then they do not have to adopt the above assumption ( ii ) ( the above ( i ) is explicitly derived in the present code already ) .
thus , the self - consistent construction of seds based on dust size evolution will be possible in our future studies . although the gas - phase metallicity ( that does not include dust - phase metal ) of a gas particle should be used for estimating the radiation cooling rate of the gas in numerical simulations of galaxy formation and evolution , the total ism metallicity ( i.e. , gas - phase + dust - phase metal ) has been used in almost all simulations for cooling estimation , because dust was not included in the simulations .
although the present new model properly considered this dust - corrected cooling process , it did not include the possible time - dependent imf , which can cause significant changes in dust evolution , star formation histories , and chemical evolution of galaxies ( e.g. , bekki 2013b ; recchi & kroupa 2015 ) .
furthermore , the present study did not consider the radiative cooling rates that depend on both [ fe / h ] on [ mg / fe ] ( de rijcke et al . 2013 ) thus on the detailed chemical evolution of galaxies .
given that the dust - depletion levels are quite different between different elements ( e.g. , mg , fe , and ca ) , the present results might be changed to some extent if the cooling rates by de rijcke et al .
( 2013 ) are adopted .
thus our future studies will need to include both temporal and spatial variations of imfs and more sophisticated cooling models ( de rijcke et al .
2013 ) in order to model the thermal and dynamical evolution of ism regulated dust in a more self - consistent way .
the present study has introduced a new parameter for the mass fraction of dust destroyed by sne ( @xmath9 ) . in the simulation code , the dust particles that are within less than @xmath10 ( destruction radius ) from a sn are assumed to be destroyed and consequently lose some fraction ( @xmath9 ) of dust .
since a cluster of stars is formed from a gas particle , the previous results based on a model for dust destruction by a single sn can not be simply used in the present model . given
that @xmath9 can control chemical evolution of galaxies , the possible range of @xmath9 in star - forming gas clouds with multiple sne exploding at different epochs needs to be investigated in our future simulations .
the mixing radius ( @xmath11 ) of ejected metals from sne and agb stars and the dust destruction radius ( @xmath10 ) are set to be the same in the present study .
the appropriate values for @xmath11 and @xmath10 are determined from the adopted spatial resolution of a simulation ( i.e. , from @xmath12 ) . however , these two could be significantly different , because dust destruction processes might be more sensitive to physical properties of ism ( e.g. , magnetic fields and gas shock speed ; jones et al .
this means that we need to perform high - resolution sub - pc scale numerical simulations on the dynamical evolution of dust and metals ejected from sne and agb stars in galactic potentials .
the outputs from such simulations will be able to be included in larger galaxy - scale simulations so that we can investigate the dust - regulated galaxy formation and evolution for reasonable ranges of @xmath11 and @xmath10 .
although we have so far focused on the formation of @xmath0 on dust grains , the surface of grains in ism is crucial for the formation and evolution of many different interstellar species ( as ` catalysis ' ) , such as ice and organic matter ( e.g. , herbst & van dishoeck 2009 ) .
accordingly , the present new dust particle methods will enable us to predict the formation efficiency of many different species other than @xmath0 by considering physical conditions of gas and dust and chemistry on dust grains . as a dust particle moves in ism of a galaxy , the dust temperature , the collision rates of atoms and molecules , the strength of isrf , and other physical properties of dust and ism can change .
if these time evolution of the physical properties of the dust and its environments can be investigated by our future simulations , we will be able to predict the time evolution of interstellar species ( e.g. , organic matter ) .
given that recent infrared observations have already shown different abundances of @xmath13 and @xmath14 ices in different galaxies ( e.g. , shimonishi et al .
2010 ) , this prediction is something that we should do not in far future but in near future , though the modeling of @xmath13 and @xmath14 ices might be a formidable task .
we have developed a new four - component model of galaxy evolution in which a galaxy consists of dark matter , stars , gas , and dust .
the dust component is represented by live dust particles so that the present new model enables us to investigate separately the evolution of gas and dust.the dust particles are live in the following three senses .
firstly , dust particles can gravitationally interact not only with other components ( e.g. , dark matter ) but also with other dust particles . secondly
, dust can be destroyed by sne and reduce significantly its mass , and also it can grow by accretion of gas - phase metals in ism .
thirdly , dust grains can be the formation sites of water , organic compounds , and amino acid closely related to the origin of life , and accordingly we can study the evolution of these interstellar matter essential for life by using our new simulations in future .
first we have described the details of the model and discussed the limitations and advantages of the model in investigating galaxy formation and evolution .
we then have investigated the possible roles of dust in galaxy evolution by using the new code and obtained some preliminary results .
in particular , we have tried to grasp the possible roles of stellar radiation pressure on dust grains in the evolution of dust , gas - phase metals , and star formation histories ( sfhs ) of galaxies .
the preliminary results that we have obtained in this study are briefly summarized as follows : + ( 1 ) the radiation pressure of stars on dust in disk galaxies can influence the time evolution of 3d dust distributions .
dust particles in a disk galaxy can be levitated from the initially thin disk and some fractions of dust can escape from the disk to locate outside the virial radius of its dark matter halo .
this dynamical evolution of dust derived by stellar radiation strongly depends on the model parameters such as stellar ages ( @xmath15 ) , total dust extinction ( @xmath16 ) , dust composition ( e.g. , silicate or graphite ) , and frequency - averaged radiation pressure coefficient ( @xmath17 ) .
+ ( 2 ) the two basic roles of stellar radiation in the time evolution of 3d dust distributions are ( i ) thickening of the vertical structure of dust and ( ii ) flattening of the radial density profile .
the above four parameters ( e.g. , @xmath16 ) can determine the degrees of these thickening and flattening in the mass distributions of dust .
for example , the models with higher @xmath16 show a smaller degree of flattening in the dust distribution whereas those with younger disk ages ( i.e. , smaller @xmath15 ) show a larger degree of the flattening .
+ ( 3 ) stellar radiation pressure can reduce the mass density of dust in a galactic disk so that it can also reduce the @xmath0 formation efficiency on dust grains . as a result of this , the star formation rate ( sfr )
can be severely reduced with the reduction level depending largely on @xmath16 .
it should be noted , however , that this suppression of star formation by stellar radiation pressure is due to the adopted assumption that sfrs depend on the mass density of @xmath0 gas ( not the total gas density ) in the present study .
therefore , it is possible that the level of sf suppression by radiation pressure is overestimated in the present study .
+ ( 4 ) the inclusion of other dust - related physical processes , such as photo - electric heating and gas - dust heating , does not change the present results on the roles of stellar radiation pressure in galaxy evolution .
it is found that photo - electric heating can suppress galaxy - wide star formation in the models with and without radiation pressure .
furthermore , gaseous drag of dust can lead to the slight enhancement of galactic sfrs , because the drag can prevent dust from escaping from thin galactic disks .
we need to confirm these interesting effects of dust on galactic sfhs in our future studies by exploring a much wider parameter space for dust - related physical processes . + ( 5 ) one of the most important parameters for the evolution of dust and metal abundances is @xmath9 , which is defined as the mass fraction of interstellar dust that is destroyed by sne in a star - forming gas cloud .
although a possibly reasonable value of @xmath9 is @xmath18 in the present study , this parameter is yet to be constrained so well by observational and theoretical studies . given its importance , it is our future study to determine the possible range of @xmath9 and its dependence on the physical properties of star - forming gas clouds .
+ ( 6 ) the present results suggest that although dust does not influence galaxy evolution so dramatically as sn and agn feedback effects do , it can play significant roles in many aspects of galaxy formation and evolution . for example , the removal of metals through dust wind , suppression of @xmath0 formation due to dust removal , and photo - electric heating of gas by dust in ism can influence chemical evolution , star formation histories , and stellar and gas dynamics in galaxies , respectively . these roles of dust in galaxy formation and evolution will need to be investigated extensively in future theoretical studies of galaxies for better understanding the formation and evolution processes of galaxies . +
thus the present new live dust particle method has enabled us to reveal the possible influences of dust - related physical processes on galaxy evolution .
the dust properties of galaxies have been derived in the post - processing of simulation data for almost all previous theoretical simulations of galaxy formation and evolution ( e.g. , derivation of dust mass from metal mass by assuming dust - to - metal ratios ) .
the present study suggests that this post - processing approach might not be realistic because dust itself can cause significant changes in galaxy evolution processes .
the present new model is still somewhat idealized ( e.g. , non - inclusion of dust size distributions in ism ) and therefore needs to be further improved so that we can properly investigate the physical properties of dust ( e.g. , evolving dust compositions and sizes ) as well as their effects on galaxy formation and evolution , in particular , the evolution of galactic seds and molecular gas fraction . in our next paper
, we will discuss how galaxy formation processes can be influenced by dust in a @xmath19cdm cosmology by using a more sophisticated live dust particle model .
i ( kenji bekki ; kb ) am grateful to the referee for constructive and useful comments that improved this paper .
numerical simulations reported here were carried out on the three gpu clusters , pleiades , fornax , and gstar kindly made available by international center for radio astronomy research(icrar ) at the university of western australia , ivec , and the center for astrophysics and supercomputing in the swinburne university , respectively .
this research was supported by resources awarded under the astronomy australia ltd s astac scheme on swinburne with support from the australian government .
gstar is funded by swinburne and the australian government s education investment fund .
kb acknowledges the financial support of the australian research council throughout the course of this work .
aguirre , a. , hernquist , l. , schaye , j. , katz , n. , weinberg , d. h. , gardner , j. , 2001 , apj , 561 , 521 amblard , a. , et al . , 2010 , a&a , 518 , l9 andrievsky , s. m. , luck , r. e. , martin , p. , lpine , j. r. d. 2004 , a&a , 413 , 159 asano , r. s. , takeuchi , t. t. , hirashita , h. , nozawa , t. , 2013 , mnras , 432 , 637 athanassoula , e. , 2003 , mnras , 341 , 1179 bakes , e. l. o. , tielens , a. g. g. m. , 1994 , apj , 427 , 822 barsella , b. , ferrini , f. , greenberg , j. m. , aiello , s. , 1989 , a&a , 209 , 349 bekki , k. , 2013 , mnras , 432 , 2298 ( b13a ) bekki , k. , 2013 , mnras , 436 , 2254 ( b13b ) bekki , k. , 2014 , mnras , 444 , 1615 ( b14 ) bekki , k. , 2015 , accepted by apj ( b15 ) , arxiv1412.1239 bekki , k. , shioya , y. , 1998 , apj , 497 , 108 bekki , k. , shioya , y. , 2000 , apj , 542 , 201 bekki , k. , shioya , y. , 2001 , apjs , 134 , 241 bekki , k. , couch , w. j. , 2011 , mnras , 415 , 1783 bekki , k. , tsujimoto , t. , 2014 , mnras , 444 , 3879 bianchi , s. , ferrara , a. , 2005 , mnras , 358 , 379 bron , e. , le bourlot , j. , le petit , f. , 2014 , a&a , 569 , 100 bruzual , g. , charlot , s. , 2003 , mnras , 344 , 1000 calura , f. , pipino , a. , matteucci , f. , 2008 , a&a , 484 , 107 calzetti , d. , armus , l. , bohlin , r. c. , kinney , a. l. , koornneef , j. , storchi - bergmann , t. , 2000 , apj , 533 , 682 chiao , r. y. , & wickramasinghe , n. c. 1972 , mnras , 159 , 361 dale , d. a. , helou , g. , contursi , a. , silbermann , n. a. , kolhatkar , s. , 2001 , apj , 549 , 215 dekel , a. , silk , j. , 1986 , apj , 303 , 39 dekel , a. , birnboim , y. , 2006 , mnras , 368 , 2 de rijcke , s. , schroyen , j. , vandenbroucke , b. , jachowicz , n. , decroos , j. , cloet - osselaer , a. , koleva , m. , 2013 , mnras , 436 , 2254 di matteo , t. , springel , v. , hernquist , l. , 2005 , nature , 433 , 604 draine , b. t. , 1978 , apjs , 36 , 595 draine , b. t. , 2009 , physics of the interstellar and intergalactic medium dressler , a. , 1980 , apjs , 42 , 565 dunne , l. , eales , s. a. , 2001 , mnras , 327 , 697 dwek , e. , 1998 , apj , 501 , 643 ( d98 ) edmunds , m. , 2001 , mnras , 328 , 223 eggen , o. j. , lynden - bell , d. , sandage , a. r. , 1962 , apj , 136 , 748 ferrara , a. , ferrini , f. , barsella , b. , franco , j. , 1991 , apj , 381 , 137 ( f91 ) galametz , m. , madden , s. c. , galliano , f. , hony , s. , bendo , g. j. , sauvage , m. , 2011 , a&a , 532 , 56 galli , d. , palla , f. , 1998 , 335 , 403 gould , r. j. , salpeter , e. e. , 1963 , apj , 138 , 393 herbst , e. , 2002 , in highlights of astronomy , vol .
12 , edited by h. rickman .
san francisco , ca , astronomical society of the pacific , p55 herbst , e. , van dishoeck , e. f. , 2009 , ara&a , 47 , 427 hernquist , l. , & katz , n. , 1989 , 70 , 419 hirashita , h. , 1999 , apj , 522 , 220 hirashita , h. , 2012 , mnras , 422 , 1263 hollenbach , d. , salpeter , e. e. , 1971 , apj , 163 , 155 hopkins , p. f. , quataert , e. , murray , n. , mnras , 421 , 3522 inoue , a. k. , 2003 , pasj , 55 , 901 jones , a. p. , tielens , a. g. g. m. , hollenbach , d. j. , mckee , c. f. , 1994 , apj , 433 , 797 jonsson , p. , 2006 ,
mnras , 372 , 2 kennicutt , r. c. , jr . , 1998 , apj , 498 , 541 keres , d. , katz , n. , weinberg , d. h. , dave , r. , 2005 , mnras , 363 , 2 krumholz , m. r. , mckee , c. f. , tumlinson , j. , 2009 , apj , 693 , 216 laibe , g. , price , d. j. , 2014 , mnras , 444 , 1940 larson , r. b. , 1974 , mnras , 169 , 229 larson , r. b. , 2005 , mnras , 359 , 211 lisenfeld , u. , ferrara , a. , 1998 , apj , 498 , 145 [ mannucci et al .
[ ] mannucci , f. , della valle , m. , panagia , n. , 2006 , mnras , 370 , 773 mnard , b. , scranton , r. , fukugita , m. , richards , g. , 2010 , mnras , 405 , 1025 ( m10 ) mckee , c. f. , 1989 , in iau symp .
135 , interstellar dust , edited by louis j. allamandola and a. g. g. m. tielens , p431 navarro , j. f. , frenk , c. s. , white , s. d. m. , 1996 , apj , 462 , 563 ( nfw ) neto , a. f. , 2007 , mnras , 381 , 1450 noh , h. , vishniac , e. t. , cochran , w. d. , 1991 , apj , 383 , 372 noguchi , m. , 1999 , apj , 514 , 77 nozawa , t. , kozasa , t. , umeda , h. , maeda , k. , nomoto , k. , 2003 , apj , 598 , 785 pelupessy , f. i. , papadopoulos , p. p. , van der werf , p. , 2006 ,
apj , 645 , 1024 ( p06 ) popescu , c. c. , tuffs , r. j. , 2013 , mnras , 436 , 1302 predehl , p. , schmitt , j. h. m. m. , 1995 , a&a , 293 , 889 recchi , s. , kroupa , p. , 2015 ,
mnras , 446 , 4168 rowlands , k. , et al .
2014 , mnras , 441 , 1040 searle , l. , zinn , r. , 1978 , apj , 225 , 357 schmidt , m. , 1959 , apj , 129 , 243 schneider , r. , omukai , k. , 2010 , mnras , 402 , 429 sellwood , j. a. ; carlberg , r. g. , 1984 , apj , 282 , 61 shimonishi , t. , onaka , t. , kato , d. , sakon , i. , ita , y. , kawamura , a. , kaneda , h. , 2010 , interstellar matter and star formation : a multi - wavelength perspective , asi conference series , vol .
1 , pp 99 , edited by d. k. ojha springel , v. , di matteo , t. , hernquist , l. , 2005 , mnras , 361 , 776 sutherland , r. s. , dopita , m. a. , 1993 , apjs , 88 , 253 tasker , e. j. , 2011 , apj , 730 , 11 theis , c. , orlova , n. , 2004 , a&a , 418 , 959 thompson , t. a. , quataert , e. , murray , n. , 2005 , apj , 630 , 168 tielens , a. g. g. m. , 2005 , the physics and chemistry of the interstellar medium , cambridge university press .
tsujimoto , t. , nomoto , k. , yoshii , y. , hashimoto , m. , yanagida , s. , thielemann , f .- k .
, 1995 , mnras , 277 , 945 ( t95 ) van den hoek , l. b. ; groenewegen , m. a. t. , 1997 , a&as , 123 , 305 ( vg97 ) watson , w. d. , 1972 , apj , 176 , 103 white , s. d. m. , rees , m. j. , 1978 , mnras , 183 , 341 wolfire , m. g. , mckee , c. f. , hollenbach , d. , tielens , a. g. g. m. , 2003 , apj , 587 , 278 xilouris , e. , alton , p. , alikakos , j. , xilouris , k. , boumis , p. , & goudis , c. , 2006 , apj , 651 , l107 zhukovska , s. , henning , t. , 2013 , a&a , 555 , 99 zubko , v. , dwek , e. , arendt , r. g. , 2004 , apjs , 152 , 211
in the present live dust particle method , the formation of @xmath0 is possible only on the surface of dust grains in galactic ism .
accordingly , the time evolution of @xmath20 and sfhs of galaxies depend on the dynamical evolution of dust .
it is therefore possible that the present results on @xmath20 evolution and sfhs of galaxies depend on the total number of dust particles ( @xmath21 ) adopted in the present simulations . in order to investigate this important issue
, we have run four comparative mw - type disk models without stellar radiation pressure and different @xmath21 and investigated whether the results depend on @xmath21 .
a1 shows that the sfrs in mw - type disk galaxies do not depend strongly on @xmath21 as long as @xmath22 ( 30% of the total gas particle number ) .
the model with @xmath23 appears to underestimate the sfr in the initial bursty star formation in the central region of the disk .
it is confirmed that the final spatial distributions of dust does not depend strongly on @xmath21 either .
this weak dependence on @xmath21 is very encouraging , because we do not have to adopt an excessively large number of dust particles to simulate the dust evolution of galaxies .
thus we conclude that the adopted number of @xmath21 is enough to investigate the possible roles of dust in galaxy evolution .
dust growth via accretion of gas - phase metals onto already existing dust grains and and dust destruction by sne are both assumed to proceed locally depending on the physical conditions of ism in the present study . in order to implement these processes ,
we have introduced a new parameter @xmath24 that defines the sphere of influence within which the dust accretion and destruction rates are calculated for each dust particle based on the physical properties of gas around the dust particle .
since the present results can possibly depend strongly on this @xmath24 , we need to investigate the possible dependence .
b1 clearly shows that galactic sfhs are qualitatively very similar between the mw - type disk models with four different @xmath24 , though the sfrs at a given time step are slightly different between the four models .
this is again encouraging , because we can estimate a reasonable @xmath24 from the adopted gravitational softening length ( corresponding to the initial spatial resolution ) of a simulation .
however , this rather weak dependence on @xmath24 might be true only for the present galaxy evolution simulations for isolated disks .
thus , we need to confirm whether this is the case for other models such as galaxy formation models based on a @xmath19cdm cosmology .
c1 briefly summarizes the key elements of the new four - component chemodynamical model adopted in the present numerical study of galaxy evolution .
the six key physical processes included in the code are ( i ) mixing of metals and dust ejected from sne and agb stars , ( ii ) dust formation from condensation of metals in stellar winds of sne and agb stars , ( iii ) dust growth through accretion of gas - phase metals onto already existing dust grains , ( iv ) dust destruction by sne , ( v ) @xmath0 formation on dust grains from neutral hydrogen ( catalysis ) , and ( vi ) consumption of gas and dust by star formation in molecular clouds in galaxies .
the new model is quite different from previous standard three - component chemodynamical model in the sense that ( i ) dust - related physical processes are explicitly and self - consistently included and ( ii ) influences of dust evolution on chemical evolution of gas - phase metals is considered .
the improved predictability of the new code is discussed extensively in the main text . | interstellar dust plays decisive roles in the conversion of neutral to molecular hydrogen ( @xmath0 ) , the thermodynamical evolution of interstellar medium ( ism ) , and the modification of spectral energy distributions ( seds ) of galaxies .
these important roles of dust have not been self - consistently included in previous numerical simulations of galaxy formation and evolution .
we have therefore developed a new model by which one can investigate whether and how galaxy formation and evolution can be influenced by dust - related physical processes such as photo - electric heating , @xmath0 formation on dust , and stellar radiation pressure on dust in detail .
a novel point of the model is that different dust species in a galaxy are represented by ` live dust ' particles ( i.e. , not test particles ) .
therefore , dust particles in a galaxy not only interact gravitationally with all four components of the galaxy ( i.e. , dark matter , stars , gas , and dust ) but also are grown and destroyed through physical processes of ism .
first we describe a way to include dust - related physical processes in nbody+hydrodynamical simulations of galaxy evolution in detail .
then we show some preliminary results of dust - regulated galaxy evolution .
the preliminary results suggest that the evolution of dust distributions driven by radiation pressure of stars is very important for the evolution of star formation rates , chemical abundances , @xmath0 fractions , and gas distributions in galaxies .
[ firstpage ] ism : dust , extinction galaxies : ism galaxies : evolution infrared : galaxies stars : formation |
bayesian inference using markov chain monte carlo simulation methods is used extensively in statistical applications . in this approach ,
the parameters are generated from a proposal distribution , or several such proposal distributions , with the generated proposals accepted or rejected using the metropolis - hastings method ; see for example @xcite . in adaptive sampling
the parameters of the proposal distribution are tuned by using previous draws .
our article deals with diminishing adaptation schemes , which means that the difference between successive proposals converges to zero . in practice
, this usually means that the proposals themselves eventually do not change .
important theoretical and practical contributions to diminishing adaptation sampling were made by @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite and @xcite .
the adaptive random walk metropolis method was proposed by @xcite with further contributions by @xcite , @xcite and @xcite . @xcite
propose an adaptive independent metropolis - hastings method with a mixture of normals proposal which is estimated using a clustering algorithm .
although there is now a body of theory justifying the use of adaptive sampling , the construction of interesting adaptive samplers and their empirical performance on real examples has received less attention .
our article aims to fill this gap by introducing a @xmath0-copula based proposal density .
an antithetic version of this proposal is also studied and is shown to increase efficiency when the acceptance rate is above 70% . we also refine the adaptive metropolis - hastings proposal in @xcite by adding a heavy tailed component to allow the sampling scheme to traverse multiple modes more easily . as well as being of interest in its own right , in some of the examples we have also used this refined sampler to initialize the adaptive independent metropolis - hastings schemes . we study the performance of the above adaptive proposals , as well as the adaptive mixture of normals proposal of @xcite , for a number of models and priors using real data .
the models and priors produce challenging but realistic posterior target distributions .
@xcite is a longer version of our article that considers some alternative versions of our algorithms and includes more details and examples .
suppose that @xmath1 is the target density from which we wish to generate a sample of observations , but that it is computationally difficult to do so directly .
one way of generating the sample is to use the metropolis - hastings method , which is now described .
suppose that given some initial @xmath2 we have generated the @xmath3 iterates @xmath4 .
we generate @xmath5 from the proposal density @xmath6 which may also depend on some other value of @xmath7 which we call @xmath8 .
let @xmath9 be the proposed value of @xmath5 generated from @xmath10 .
then we take @xmath11 with probability @xmath12 and take @xmath13 otherwise .
if @xmath14 does not depend on @xmath15 , then under appropriate regularity conditions we can show that the sequence of iterates @xmath16 converges to draws from the target density @xmath1 .
see @xcite for details . in adaptive sampling
the parameters of @xmath14 are estimated from the iterates @xmath17 . under appropriate regularity conditions the sequence of iterates @xmath18 , converges to draws from the target distribution @xmath1 .
see @xcite , @xcite and @xcite .
we now describe the adaptive sampling schemes studied in the paper .
the adaptive random walk metropolis proposal of @xcite is @xmath19 where @xmath20 is the dimension of @xmath7 and @xmath21 is a multivariate @xmath20 dimensional normal density in @xmath7 with mean @xmath8 and covariance matrix @xmath22 . in , @xmath23 for @xmath24 , with @xmath25 representing the initial iterations , @xmath26 for @xmath27 with @xmath28 ; @xmath29 is a constant covariance matrix , which is taken as the identity matrix by @xcite but can be based on the laplace approximation or some other estimate .
the matrix @xmath30 is the sample covariance matrix of the first @xmath31 iterates .
the scalar @xmath32 is meant to achieve a high acceptance rate by moving the sampler locally , while the scalar @xmath33 is considered to be optimal @xcite for a random walk proposal when the target is multivariate normal .
we note that the acceptance probability for the adaptive random walk metropolis simplifies to @xmath34 we refine the two component random walk metropolis proposal in by adding a third component with @xmath35 and with @xmath36 .
we take @xmath37 if @xmath38 , @xmath39 for @xmath27 and @xmath40 . alternatively
, the third component can be a multivariate @xmath0 distribution with small degrees of freedom .
we refer to this proposal as the three component adaptive random walk .
the purpose of the heavier tailed third component is to allow the sampler to explore the state space more effectively by making it easier to leave local modes . to illustrate this issue we consider the performance of the two and three component adaptive random walk samplers when the target distribution is a two component and five dimensional multivariate mixture of normals .
each component in the target has equal probability , the first component has mean vector @xmath41 and the second component has mean vector @xmath42 .
both components have identity covariance matrices . for the three component
adaptive random walk we choose @xmath43 .
the starting value is @xmath44 for both adaptive random walk samplers .
figure [ fig : bimodal : normal ] compares the results and shows that the two component adaptive random walk fails to explore the posterior distribution even after 500 , 000 iterations , whereas the three component adaptive random walk can get out of the local modes .
+ the proposal density of the adaptive independent metropolis - hastings approach of @xcite is a mixture with four terms of the form @xmath45 with @xmath46 the parameter vector for the density @xmath47 .
the sampling scheme is run in two stages , which are described below . throughout each stage ,
the parameters in the first two terms are kept fixed .
the first term @xmath48 is an estimate of the target density and the second term @xmath49 is a heavy tailed version of @xmath48 .
the third term @xmath50 is an estimate of the target that is updated or adapted as the simulation progresses and the fourth term @xmath51 is a heavy tailed version of the third term . in the first stage
@xmath52 is a laplace approximation to the posterior if it is readily available and works well ; otherwise , @xmath52 is a gaussian density constructed from a preliminary run of 1000 iterates or more of the three component adaptive random walk . throughout , @xmath49 has the same component means and probabilities as @xmath48 , but its component covariance matrices are ten times those of @xmath48 .
the term @xmath50 is a mixture of normals and @xmath51 is also a mixture of normals obtained by taking its component probabilities and means equal to those of @xmath50 , and its component covariance matrices equal to 20 times those of @xmath50 .
the first stage begins by using @xmath53 and @xmath49 only with , for example , @xmath54 and @xmath55 , until there is a sufficiently large number of iterates to form @xmath50 .
after that we set @xmath56 and @xmath57 .
we begin with a single normal density for @xmath50 and as the simulation progresses we add more components up to a maximum of four according to a schedule that depends on the ratio of the number of accepted draws to the dimension of @xmath7 .
see appendix [ s : sim details ] . in the second stage ,
@xmath48 is set to the value of @xmath50 at the end of the first stage and @xmath58 and @xmath51 are constructed as described above .
the heavy - tailed densities @xmath49 and @xmath51 are included as a defensive strategy , as suggested by @xcite , to get out of local modes and to explore the sample space of the target distribution more effectively .
it is too computationally expensive to update @xmath50 ( and hence @xmath51 ) at every iteration so we update them according to a schedule that depends on the problem and the size of the parameter vector . see appendix [ s : sim details ] .
our article estimates the multivariate mixture of normals density for the third component using the method of @xcite who identify the marginals that are not symmetric and estimate their joint density by a mixture of normals using k - harmonic means clustering .
we also estimated the third component using stochastic approximation but _ without _ first identifying the marginals that are not symmetric .
we studied the performance of both approaches to fitting a mixture of normals and found that the clustering based approach was more robust in the sense that it does not require tuning for particular data sets to perform well , whereas for the more challenging target distributions it was necessary to tune the parameters of the stochastic approximation to obtain optimal results .
the results for the stochastic approximation approach are reported in @xcite .
we note that estimating a mixture of normals using the em algorithm also does not require tuning ; see @xcite , as well as @xcite for the online em
. however , the em algorithm is more sensitive to starting values and is more prone to converge to degenerate solutions , particularly in an mcmc context where small clusters of identical observations arise naturally ; see @xcite for a discussion .
the third proposal distribution is a mixture of a @xmath0 copula and a multivariate @xmath0 distribution .
we use the @xmath0 copula as the major component of the mixture because it provides a flexible and fast method for estimating a multivariate density . in our applications
this means that we assume that after appropriately transforming each of the parameters , the joint posterior density is multivariate @xmath0 . or more accurately , such a proposal density provides a more accurate estimate of the posterior density than using a multivariate normal or multivariate @xmath0 proposal .
the second component in the mixture is a multivariate @xmath0 distribution with low degrees of freedom whose purpose is to help the sampler move out of local modes and explore the parameter space more effectively .
@xcite and @xcite provide an introduction to copulas , with details of the @xmath0 copula given in @xcite .
let @xmath59 be a @xmath20-dimensional @xmath0 density with location @xmath60 , scale matrix @xmath61 and degrees of freedom @xmath62 and let @xmath63 be the corresponding cumulative distribution function .
let @xmath64 and @xmath65 be the probability density function and the cumulative distribution function of the @xmath15th marginal , with @xmath66 its parameter vector . then the density of the @xmath0 copula based distribution for @xmath67 is @xmath68 where @xmath69 is determined by @xmath70 , @xmath71 . for a given sample of observations , or in our case a sequence of draws , we fit the copula by first estimating each of the marginals as a mixture of normals .
for the current degrees of freedom , @xmath72 , of the @xmath0 copula , we now transform each observation @xmath73 to @xmath74 using @xmath75 this produces a sample of @xmath20 dimensional observations @xmath76 which we use to estimate the scale matrix @xmath77 as the sample correlation matrix of the @xmath76 s .
given the estimates of the marginal distributions and the scale matrix @xmath77 , we estimate the degrees of freedom @xmath62 by maximizing the profile likelihood function given by over a grid of values @xmath78 , with @xmath79 representing the gaussian copula .
we could use instead the grid of @xmath62 values in , but this is more expensive computationally and we have found our current procedure works well in practice . the second component of the mixture is a multivariate @xmath0 distribution with its degrees of freedom @xmath80 fixed and small , and with its location and scale parameters estimated from the @xmath7 iterates using the first and second sample moments .
the @xmath0 copula component of the mixture has a weight of 0.7 and the multivariate @xmath0 component a weight of 0.3 . to draw an observation from the @xmath0 copula , we first draw @xmath76 from the multivariate @xmath0 distribution with location 0 and scale matrix @xmath77 and degrees of freedom @xmath62 .
we then use a newton - raphson root finding routine to obtain @xmath16 for a given @xmath69 from for @xmath71 .
the details of the computation for the @xmath0 copula and the schedule for updating the proposals is given in appendix [ s : sim details ] .
using antithetic variables in simulations often increases sampling efficiency ; e.g. @xcite .
@xcite proposes using antithetic variables in markov chain monte carlo simulation when the target is symmetric .
we apply antithetic variables to the copula based approach which generalizes tierney s suggestion by allowing for nonsymmetric marginals . to the best of our knowledge ,
this has not been done before .
the antithetic approach is implemented as follows .
as above , we determine probabilistically whether to sample from the copula component or the multivariate @xmath0 component . if the copula component is chosen , then @xmath67 is generated as above and @xmath81 is also computed .
the values @xmath67 and @xmath82 are then transformed to @xmath76 and @xmath83 respectively and are accepted or rejected one at a time using the metropolis - hastings method . if we decide to sample from the multivariate @xmath0 component to obtain @xmath76 , we also compute @xmath84 , where @xmath60 is the mean of the multivariate @xmath0 , and accept or reject each of these values one at a time using the metropolis - hastings method .
we note that to satisfy the conditions for convergence in @xcite we would run the sampling scheme in two stages , with the first stage as above . in the second stage we would have a three component proposal , with the first two components the same as above .
the third component would be fixed throughout the second stage and would be the second component density at the end of the first stage .
the third component would have a small probability , e.g. 0.05 .
however , in our examples we have found it unnecessary to include such a third component as we achieve good performance without it .
this section studies the five algorithms discussed in section [ section : adap : samp ] .
the two component adaptive random walk metropolis ( rwm ) and the three component adaptive random walk metropolis ( rwm3c ) are described in section [ ss : arwm ] .
the adaptive independent metropolis - hastings with a mixture of normals proposal distribution fitted by clustering ( imh - mn - cl ) described in section [ ss : aimh ] .
the adaptive independent metropolis - hastings which is a mixture of a @xmath0 copula and a multivariate @xmath0 proposal , with the marginal distributions of the copula estimated by mixtures of normals that are fitted by clustering ( imh - tct - cl ) .
the fifth sampler is the antithetic variable version of imh - tct - cl , which we call imh - tct - cl - a .
these proposals are described in section [ ss : tct ] .
our study compares the performance of the algorithms in terms of the acceptance rate of the metropolis - hastings method , the inefficiency factors ( if ) of the parameters , and an overall measure of effectiveness which compares the times taken by all samplers to obtain the same level of accuracy .
we define the acceptance rate as the percentage of accepted values of each of the metropolis - hastings proposals .
we define the inefficiency of the sampling scheme for a given parameter as the variance of the parameter estimate divided by its variance when the sampling scheme generates independent iterates .
we estimate the inefficiency factor as @xmath85 where @xmath86 is the estimated autocorrelation at lag @xmath15 and the truncated kernel function @xmath87 if @xmath88 and 0 otherwise @xcite . as a rule of thumb ,
the maximum number of lags @xmath89 is given by the lowest index @xmath15 such that @xmath90 with @xmath91 being the sample size use to compute @xmath86 .
we define the equivalent sample size as @xmath92 , where @xmath93 , which can be interpreted as @xmath94 iterates of the dependent sampling scheme are equivalent to @xmath95 independent iterates .
the acceptance rate and the inefficiency factor do not take into account the time taken by a sampler . to obtain an overall measure of the effectiveness of a sampler ,
we define its equivalent computing time @xmath96 , where @xmath89 is the time per iteration of the sampler and @xmath97 .
we interpret @xmath98 as the time taken by the sampler to attain the same accuracy as that attained by @xmath94 independent draws of the same sampler . for two samplers @xmath99 and @xmath100 , @xmath101 is the ratio of times taken by them to achieve the same accuracy .
we note that the time per iteration for a given sampling algorithm depends to an important extent on how the algorithm is implemented , e.g. language used , whether operations are vectorized , which affects @xmath98 but not the acceptance rates nor the inefficiencies .
this section applies the adaptive sampling schemes to the binary logistic regression model @xmath102 using three different priors for the vector of coefficients @xmath103 .
the first is a non - informative multivariate normal prior , @xmath104 the second is a normal prior for the intercept @xmath105 , and a double exponential , or laplace prior , for all the other coefficients , @xmath106 the regression coefficients are assumed to be independent a priori .
we note that this is the prior implicit in the lasso @xcite .
the prior for @xmath107 is @xmath108 , where @xmath109 means an inverse gamma density with shape @xmath99 and scale @xmath100 .
the double exponential prior has a spike at zero and heavier tails than the normal prior . compared to their posterior distributions under a diffuse normal prior
, this prior shrinks the posterior distribution of the coefficients close to zero to values even closer to zero , while the coefficients far from zero are almost unmodified . in the adaptive sampling schemes we work with @xmath110 rather than @xmath107 as it is unconstrained .
the third prior distribution takes the prior for the intercept as @xmath111 and the prior for the coefficients @xmath112 as the two component mixture of normals , @xmath113 with the regression coefficients assumed a priori independent . @xcite
suggest using this prior for bayesian variable selection , with @xmath114 and @xmath115 small and large variances that are chosen by the user . in our article their values
are given for each of the examples below .
the prior for @xmath116 is uniform . in the adaptive sampling we work with the logit of @xmath116 because it is unconstrained .
this section models the probability of labor force participation by women , @xmath117 , in 1975 as a function of the covariates listed in table [ t : labor force part ] .
this data set is discussed by @xcite , p. 537 and has a sample size of 753 .
.variables used in labor force participation data regression [ cols= " < , < " , ] [ t : pap truncated ]
our article proposes a new copula based adaptive sampling scheme and a generalization of the two component adaptive random walk designed to explore the target space more efficiently than the proposal of @xcite .
we studied the performance of these sampling schemes as well as the adaptive independent metropolis - hastings sampling scheme proposed by @xcite which is based on a mixture of normals .
all the sampling schemes performed reliably on the examples studied in the article , but we found that the adaptive independent metropolis - hastings schemes had inefficiency factors that were often much lower and acceptance rates that were much higher than the adaptive random walk schemes .
the copula based adaptive scheme often had the smallest inefficiency factors and highest acceptance rates .
for acceptance rates over 70% the antithetic version of the copula based approach was the most efficient .
our results suggest that the copula based proposal provides an attractive approach to adaptive sampling , especially for higher dimensions .
however , the mixture of normals approach of @xcite also performed well and is useful for complicated and possibly multimodal posterior distributions .
the research of robert kohn , ralph silva and xiuyan mun was partially supported by an arc discovery grant dp0667069 .
we thank professor garry barret for the cps data and professor denzil fiebig for the pap smear data .
all the computations were done on intel core 2 quad 2.6 ghz processors , with 4 gb ram ( 800mhz ) on a gnu / linux platform using matlab 2007b . however , in the tct algorithm we computed the univariate cumulative distribution functions and inverse cumulative distribution functions of the @xmath0 , normal and mixture of normals distributions using matlab mex files based on the corresponding matlab code .
in addition , to speed up the computation , we tested each marginal for normality using the jarque - bera test at the 5% level .
if normality was not rejected then we fitted a normal density to the marginal .
otherwise , we estimated the marginal density by a mixture of normals . in stage 1 of the adaptive sampling schemes imh - mn - cl and imh - mn - sa that use a multivariate mixture of normals , the number of components ( @xmath118 ) used in the third term @xmath119 of the mixture
is determined by the dimension of the parameter vector ( @xmath120 ) and the number of accepted draws ( @xmath121 ) to that stage of the simulation .
in particular , @xmath122 if @xmath123 , @xmath124 if @xmath125 , @xmath126 if @xmath127 and @xmath128 if @xmath129 .
we now give details of the number of iterations , burn - in and updating schedules for all the adaptive independent metropolis - hastings schemes in the paper .
in addition , we update the proposal in stage 1 if in 100 successive iterations the acceptance rate is lower than 0.01 . *
normal prior : end of first stage = 5 000 ; burn - in = 75 000 ; number of iterations : 100 000 ; updates = [ 50 , 100 , 150 , 200 , 300 , 500 , 700 , 1000 , 2000 , 5000 , 10000 , 20000 , 30000 , 50000 , 75000 ] . * double exponential prior : end of first stage = 5 000 ; burn - in = 100 000 ; number of iterations : 150 000 ; updates = [ 50 , 100 , 150 , 200 , 300 , 500 , 700 , 1000 , 2000 , 5000 , 10000 , 20000 , 30000 , 50000 , 75000 , 100000 ] . * mixture of normals prior : end of first stage = 100 000 ; burn - in = 300 000 ; number of iterations : 400 000 ; updates = [ 100 , 150 , 200 , 300 , 500 , 700 , 1000 , 2000 , 3000 , 5000 , 7500 , 10000 , 15000 , 20000 , 30000 , 50000 , 75000 , 100000 , 125000 , 150000 , 175000 , 200000 , 225000 , 250000 , 300000 ] .
* normal and double exponential priors , quantiles 0.1 , 0.5 and 0.9 : end of first stage = 3 000 ; burn - in = 150 000 ; number of iterations : 200 000 ; updates = [ 100 , 150 , 200 , 300 , 500 , 700 , 1000 , 2000 , 3000 , 5000 , 7500 , 10000 , 15000 , 20000 , 30000 , 50000 , 75000 , 100000 , 150000 ] . * mixture of normals prior , quantiles 0.1 , 0.5 and 0.9 : end of first stage = 200 000 ; burn - in = 400 000 ; number of iterations : 500 000 ; updates = [ 100 , 150 , 200 , 300 , 500 , 700 , 1000 , 2000 , 3000 , 5000 , 7500 , 10000 , 15000 , 20000 , 30000 , 50000 , 75000 , 100000 , 125000 , 150000 , 175000 , 200000 , 225000 , 250000 , 275000 , 300000 , 325000 , 350000 , 375000 , 400000 ] .
probit random effects model , pap smear data .
for all three priors , end of first stage = 5 000 ; burn - in = 10 000 ; number of iterations : 20 000 ; updates = [ 20 , 50 , 100 , 150 , 200 , 300 , 400 , 500 , 600 , 700 , 800 , 900 , 1000 , 1100 , 1200 , 1300 , 1400 , 1500 , 2000 , 2500 , 3000 , 3500 , 4000 , 4500 , 5000 , 6000 , 7000 , 8000 , 9000 , 10000 , 12000 , 15000 ] . in this example
the importance sampling density is updated every @xmath130 iterations .
we now give the details of the sampling for both adaptive random walk metropolis algorithms .
for the hmda data the number of burn - in iterations was 300 000 and the total number of iterations was 500 000 for all three priors .
the corresponding numbers for the cps data with normal and double exponential priors are 500 000 and 1000 000 , and for the mixture of normals prior 1000 000 and 1500 000 .
the corresponding numbers for the pap smear data are 30 000 and 50 000 .
donald , s.g . ,
green , d.a . and
paarsch , h.j .
differences in wage distributions between canada and the united states : an application of a flexible estimator of distribution functions in the presence of covariates .
_ review of economic studies _ , 67(4 ) , 609 - 633 . | our article is concerned with adaptive sampling schemes for bayesian inference that update the proposal densities using previous iterates .
we introduce a copula based proposal density which is made more efficient by combining it with antithetic variable sampling .
we compare the copula based proposal to an adaptive proposal density based on a multivariate mixture of normals and an adaptive random walk metropolis proposal .
we also introduce a refinement of the random walk proposal which performs better for multimodal target distributions .
we compare the sampling schemes using challenging but realistic models and priors applied to real data examples .
the results show that for the examples studied , the adaptive independent metropolis - hastings proposals are much more efficient than the adaptive random walk proposals and that in general the copula based proposal has the best acceptance rates and lowest inefficiencies .
* keywords * : antithetic variables ; clustering ; metropolis - hastings ; mixture of normals ; random effects . |
selection , random genetic drift and mutations are the processes underlying darwinian evolution . for a long time
population geneticists have analyzed the dynamics in the simplest setting consisting of two genotypes evolving under these processes @xcite . in those studies
, a genotype represents an individual s genetic makeup , completely determining all relevant properties of the individual .
a key concept is the so - called fitness of a genotype which represents the selection pressure for the individuals .
the fitness defines the expected number of offspring an individual will produce . thus , selection acts on fitness differences preferring individuals with higher fitness over individuals with lower fitness .
usually it is assumed that individuals have fixed fitnesses defined by their genotype alone @xcite . yet
, experimental studies have revealed that many natural systems exhibit frequency - dependent selection @xcite , which means that an individual s fitness not only depends on its genotype , but also on its interactions with other individuals and hence on the frequency of the different genotypes in the population .
although such frequency - dependent selection had already been studied early by crow and kimura @xcite , only recently has it received more attention @xcite . in these theoretical and computational studies ,
individuals interactions are represented by interaction matrices from game theory .
this leads to a frequency dependence where the fitness depends directly on the interaction parameters in a linear way .
however , fitness may depend on many diverse factors such as cooperation ( i.e. individuals acting together to increase their fitness @xcite ) and resource competition , so that certain systems may exhibit frequency - dependent fitness that is nonlinear .
for example , in experiments certain hermaphrodites exhibit such nonlinear fitness - dependence @xcite . to the best of our knowledge
the impact of such nonlinear dependencies on coevolutionary dynamics has not been investigated theoretically . in this article
we show that nonlinear frequency dependence @xcite may induce new stable configurations of the evolutionary dynamics .
furthermore , we study the impact of asymmetric mutation probabilities on the dynamics @xcite , which was also neglected in most models until now @xcite . as in previous works on coevolutionary dynamics we base our work on the moran process in a non - spatial environment which is a well established model to study evolutionary dynamics and
was already used in many applications @xcite .
the moran process is a stochastic birth - death process which keeps the population size constant @xcite .
therefore , in a two - genotype model the system becomes effectively one - dimensional , so that the dynamics may be described by a one - dimensional markov chain with transition rates defined by the moran process .
we derive the stationary probability distribution of the system dynamics via its fokker - planck equation @xcite .
sharp maxima of the distribution reveal metastable points of the dynamics and a multitude of such maxima lead to stochastic switching dynamics between multiple stable points .
the article is structured as follows . in section
ii we introduce the model details and in section iii we derive the fokker - planck equation describing the probabilistic dynamics of the population .
using this equation we derive the stationary probability distribution that describes the long - time behavior of the system . in section
iv we analyze this probability distribution , which yields information about the impact of nonlinear frequency - dependent selection and of different mutation probabilities on the coevolutionary dynamics . in section
v we give a summary and discuss our results .
consider a population of @xmath0 individuals evolving in a non - spatial environment , meaning that the population is well - mixed so that each individual interacts with all other individuals at all times . in this population
the individuals may assume one out of two genotypes @xmath1 and @xmath2 .
the population sizes @xmath3 and @xmath4 ( @xmath5 ) evolve according to the time - continuous moran process described in the following , cf .
the number of individuals @xmath3 of genotype @xmath1 completely determines the state of the system as @xmath6 . at all times
the interactions of the individuals determine the actual ( frequency - dependent ) fitness , so that an individual s fitness of genotype @xmath1 or @xmath2 is defined by a fitness function @xmath7 or @xmath8 respectively .
the fitness functions @xmath7 and @xmath8 may be any functions with the only condition that @xmath9 for all @xmath10 $ ] as negative fitness is not defined . at rate
@xmath11 an individual of type @xmath1 produces an identical offspring which may mutate to genotype @xmath2 with probability @xmath12 .
this applies analogously to genotype @xmath2 .
then one individual of the population is chosen randomly to die , so that the population size @xmath0 stays constant and the variables @xmath13 change maximally by @xmath14 .
this is the so - called moran process which was originally introduced for a population of two genotypes with fixed fitnesses and no mutations occurring , cf .
however , this process is easily generalizable to more genotypes and frequency - dependent selection in the way described above , cf .
@xcite . note that in our model the rate of reproduction of e.g. type @xmath1 is directly determined by the term @xmath15 as e.g. in @xcite . in other models
the fitness is first normalized so that the rate of reproduction is given by @xmath16 @xcite , where @xmath17 \label{eq : meanfitness}\ ] ] is the population s mean fitness . while in both of these models the events
occur with the same probability , the times between the events differ by a common factor determined by the mean fitness ( [ eq : meanfitness ] ) .
thus , these models exhibit a quantitatively different time course , but the same event sequences . until now ,
usually linear functions have been considered for @xmath18 . however , in many applications nonlinear functions seem more appropriate @xcite .
for example , cooperation effects in game theory induce functions linearly increasing in @xmath3 , so that @xmath19 @xcite .
this would mean that fitness increases infinitely with the population size of genotype @xmath1 . yet , in any habitat there is only a limited amount of resources available for living .
therefore , fitness should decline when the population of one genotype becomes too large as all of the individuals will compete for the same resources .
we conclude that a fitness function including these two effects has to contain a nonlinear factor , e.g. @xmath20 where @xmath21 and @xmath22 .
here we choose a quadratic nonlinear factor as a simple example . naturally , all other nonlinear functions could be applicable as well .
note that linearly increasing fitness functions such as @xmath19 and @xmath23 lead to quadratic relations for the _ mean _ fitness @xmath24 ( see equation ( [ eq : meanfitness ] ) ) of the population which becomes @xmath25 the quadratic factor originates from the fact that the linear fitness function in ( [ eq : meanfitness ] ) is multiplied by the number of individuals leading to a quadratic dependence on @xmath26 , e.g. @xmath27 .
this factor enters the replicator equation in evolutionary game theory , thus leading to a quadratic fitness dependence in standard game theoretic problems @xcite .
however , this does not imply nonlinear interactions and thus fitness effects such as presented in equation ( [ eq : fitnessfunction ] ) .
therefore , the fitness functions that can be generated by evolutionary game theory are a special case of the functions that occur in the theory presented here
. fitness functions as for example presented in equation ( [ eq : fitnessfunction ] ) will as we show in this article lead to dynamics with more complex stability properties . usually , when analyzing the dynamics of such a model system as described above , it was assumed either that no mutations occur at all ( @xmath28 ) @xcite or that the mutation probabilities are equal ( @xmath29 ) @xcite . however , usually mutation probabilities can be highly diverse @xcite .
this has been considered in some examples @xcite , but until now the effect of different mutation probabilities in the introduced model system has not been studied systematically . here
, we explicitly study the effects caused by asymmetric mutation rates ( see e.g. figure [ fig : asymmetricexample ] ) .
in the following we assume the fitness of the individuals to be given by @xmath30 and @xmath31 with the interaction functions @xmath32 and @xmath33 . the model system is effectively one - dimensional with the variable @xmath34 completely describing the system state @xcite . at each event @xmath35 may change by at most @xmath36 depending only on the actual state of the system , so that the dynamics are markovian .
let the transition @xmath37 occur with the rate @xmath38 .
then this rate is determined by the fitnesses and mutation probabilities in the following way : * each individual in the population receives offspring with rate @xmath39 with @xmath40 .
this means that genotype @xmath1 receives one offspring with rate @xmath41 .
* such an offspring increases the population @xmath1 with probability @xmath42 , and population @xmath2 in case it mutates with probability @xmath12 . * one individuum is chosen uniformly to die belonging to genotype @xmath2 with probability @xmath43 .
* equivalently the number of individuals of genotype @xmath1 can increase if genotype @xmath2 receives one offspring , which mutates to genotype @xmath1 with probability @xmath44 , and one individuum of @xmath2 is chosen to die .
taken these processes together the transition rate for @xmath37 is@xmath45 decreasing @xmath35 to @xmath46 is equivalent to increasing the number of individuals of genotype @xmath2 , so that the transition rate@xmath47 is obtained analogously .
the master equation of the markov chain is then given by@xmath48 where we define @xmath49 .
as also @xmath50 , the system has reflective boundaries at @xmath51 and @xmath52 . if @xmath53 ( @xmath54 ) then @xmath55 ( @xmath56 ) is an absorbing state of the dynamics
it is possible to derive the exact solution to such an equation similarly to the derivations in @xcite which we do in appendix b. however , the obtained solutions are not very explicit and their implications are hard to grasp .
it is thus more useful to advance to the fokker - planck equation @xcite describing the system in the limit of large @xmath0 .
as previous works have shown , this approximation already leads to very good results for moderate population sizes of @xmath57 , cf .
we use the transformation @xmath58,\qquad s=\frac{t}{n},\qquad{\tilde{\mu}}_{ij}=\mu_{ij}\cdot n,\\[0.1 cm ] \rho(x , s)=p_{xn}(sn)n,\qquad\tilde{g}_{j}(x)=g_{j}(xn)\cdot n \label{eq : rescaledinteraction}\end{aligned}\ ] ] yielding the fokker - planck equation .
the scaling factors are derived in appendix a where the derivation of the fokker - planck equation in the scaling limit @xmath59 is shown in detail ( equation ( [ eq : appfpe ] ) ) .
this leads to@xmath60x(1-x)-\delta{\tilde{\mu}}\right)\rho(x , s)\big]+\frac{\partial^{2}}{\partial x^{2}}\big[x(1-x)\rho(x , s)\big]\label{eq : fpe}\ ] ] with the normalization condition @xmath61 for all @xmath62 . here
@xmath63 and @xmath64 are the rescaled interaction functions defined in ( [ eq : rescaledinteraction ] ) , @xmath65 is the rescaled mean mutation rate and @xmath66 is the rescaled mutation rate difference .
these functions have to be rescaled to obtain a non - degenerate limit , the so - called weak selection regime .
notice , that the drift term @xmath67x(1-x)-\delta{\tilde{\mu}}\ ] ] contains three different effects .
the fitness difference at each point @xmath68 causes a selective drift , the mean mutation rate causes a drift directed to @xmath69 and the mutation rate difference causes a one - directional drift towards one of the boundaries @xmath70 or @xmath71 .
the diffusion term@xmath72 reflects the undirected genetic drift .
if fitness differences are on a scale of @xmath73 , then mutation , selection and genetic drift all act on the same scale @xcite . then the dynamics are characterized by an interplay of these different effects . on the other hand , in the strong selection regime where fitness differences are on a scale @xmath74
genetic drift becomes negligible and the dynamics in the bulk of the system become deterministic for large @xmath0 @xcite .
however , as the factor @xmath75 vanishes for @xmath76 and @xmath77 the effects of mutations play an important role . clearly , without mutations @xmath70 and @xmath71 are absorbing states , but even in the strong selection regime they become non - absorbing for any positive mutation rate .
this is reflected in the form of the drift term @xmath78 , as only the factors reflecting mutations @xmath79 remain non - zero in the limits @xmath76 and @xmath77 . for a one - dimensional fokker - planck equation ( [ eq : fpe ] ) the stationary distribution is @xcite@xmath80 with the potential@xmath81 which is defined up to a constant irrelevant to our calculations as it is canceled by the normalization@xmath82 thus
, we obtain@xmath83dx\\ & = & \ln\left(x(1-x)\right)-{\tilde{\mu}}\ln(x(1-x))+\delta{\tilde{\mu}}\ln\left[\frac{x}{1-x}\right]-\int\left[{\tilde{g}_a}(x)-{\tilde{g}_b}(x)\right]dx\end{aligned}\ ] ] where we used@xmath84.\ ] ] the stationary solution becomes@xmath85dx}\label{eq : generalsolution}\ ] ] where the constant @xmath86 defined in ( [ eq : normalization ] ) has to be calculated numerically for given parameters .
this stationary distribution contains contributions from the above mentioned four different effects : 1 .
genetic drift is reflected by @xmath87 which diverges for @xmath88 at the boundaries .
the mean mutation rate @xmath89 causes the balancing term @xmath90 .
3 . the asymmetry in the mutation probabilities causes the term @xmath91 which diverges at one boundary and vanishes at the other .
all frequency - dependent selection effects are contained in the exponential factor @xmath92dx}$ ] which can take various shapes .
what are the possible shapes for the stationary distribution @xmath93 in the form given by equation ( [ eq : generalsolution ] ) ?
until now , the term @xmath91 representing the asymmetric mutation probabilities has not been considered to our knowledge and the interaction functions @xmath94 and @xmath95 in the selection term were considered to be at most linear in @xmath68 @xcite .
we should therefore be interested in the effects of nonlinear interaction functions and asymmetric mutation rates .
let us first analyze the dynamics for nonlinear interaction functions @xcite describing the effects of cooperation and limited resources as described by equation ( [ eq : fitnessfunction ] ) , so that@xmath96 already with @xmath97 such interaction functions induce dynamics stochastically switching between three metastable points .
an example is shown in figure [ fig : firstexample ] , where the theoretically calculated stationary distribution from equation ( [ eq : generalsolution ] ) is shown together with data from simulations with a population of @xmath98 individuals which is enough to obtain almost perfect fitting ( cf . also figure [ fig : nplot ] ) .
the fitness functions ( figure [ fig : firstexample]b ) both show at first an increase on increasing the number of individuals of genotype @xmath1 or @xmath2 from @xmath99 due to cooperation effects and then a strong decrease due to resource competion .
as the resulting fitness functions are asymmetric , also the stationary distribution is asymmetric .
there is a maximum at @xmath70 due to genetic drift .
selection drives the dynamics towards a metastable state at @xmath100 , because for @xmath101 genotype @xmath1 is fitter than @xmath2 thus increasing in frequency and for @xmath102 genotype @xmath1 is less fit than @xmath2 thus decreasing in frequency ( cf .
figure [ fig : firstexample]b ) .
the maximum of the stationary distribution is thus exactly at the point where @xmath103 . for @xmath104
again genotype @xmath1 is fitter than @xmath2 and thus the dynamics are driven towards @xmath71 by selection as well as genetic drift .
the mutational force induced by @xmath105 increases the height of the maximum at @xmath100 driving the system away from the maxima at @xmath70 and @xmath71 .
so , in this example genetic drift , mutation and selection all significantly influence the dynamics . )
( red , solid ) in perfect agreement with data from simulations with @xmath98 ( blue , @xmath106 ) .
( b ) shows the fitness functions of genotype @xmath1 ( blue , solid ) and @xmath2 ( red , dashed ) .
interaction parameters are @xmath107 , @xmath108 , @xmath109 and @xmath110 ( see equation ( [ eq : fofk ] ) and ( [ eq : interactions ] ) ) while the mutation rate is @xmath105 ( @xmath97).,width=566 ] let us now study the influence of the mutation rates in detail .
interestingly , for asymmetric mutation rates ( @xmath111 ) the factor @xmath91 always diverges in the interval @xmath112 $ ] .
for @xmath113 it diverges for @xmath76 , otherwise for @xmath77 .
this can cause the emergence of a maximum of the stationary distribution at @xmath70 ( or resp .
@xmath71 ) . figure [ fig : asymmetricexample ] shows an example where due to asymmetric mutation rates a maximum occurs at @xmath71 , when for @xmath97 there is an absolute minimum at @xmath71 as this stable state minimizes the population s mean fitness .
this means , that the system dynamics are mutation dominated near @xmath71 .
furthermore , the maximum located at @xmath70 for @xmath97 is shifted for @xmath113 to values @xmath114 causing a minimum of the stationary distribution at @xmath70 .
thus , in this example both selection and the asymmetry in mutation probabilities are the driving forces of the system s dynamics .
( red , solid ) and @xmath115 ( gray , dashed ) for a system with selective advantage for genotype @xmath2 as shown in ( b ) ( blue , solid : genotype @xmath1 ; red , dashed : genotype @xmath2 ) . solid and dashed lines in ( a ) show the theoretical curves from equation ( [ eq : generalsolution ] ) , crosses the data from simulations with @xmath98 ( @xmath106 : @xmath97 , @xmath116 : @xmath115 ) .
interaction parameters are @xmath117 , @xmath118 , @xmath119 and @xmath120 ( see equation ( [ eq : interactions ] ) ) while mean mutation rate is @xmath121.,width=566 ] for low mutation rates the population has a high tendency to become dominated by one of the two genotypes for long times which is illustrated in figure [ fig : mutplot]a .
thus , the dynamics stay close to the edges @xmath70 or @xmath71 waiting for one of the few mutations to occur . for low overall mutation rates differences in the mutation rates
have only weak effects .
high mutation rates cause the population to be drawn towards a mixture of both genotypes ( @xmath122 ) .
even more , for high mutation rates the differences in the mutation rates @xmath123 can have an important influence on the dynamics .
for instance figure [ fig : mutplot]b illustrates a shift of an existing stable point , which completely vanishes for high @xmath124 .
we conclude that asymmetric mutation rates can cause the emergence of new stable states ( as in figure [ fig : asymmetricexample ] ) , the disappearance of existing ones ( as in figure [ fig : mutplot ] ) , and shift existing stable states to new positions . ) for three different symmetric ( @xmath97 ) mutation rates : @xmath121 ( blue , solid ) , @xmath125 ( red , long dashed ) and @xmath126 ( green , short dashed ) .
this illustrates that for low mutation rates the dynamics stay close to the system s edges for long times .
here , differences in the mutation rates only slightly affect the shape of the stationary distribution . on the other hand ,
( b ) shows that for high mutation rates ( @xmath127 ) the dynamics are drawn stronger towards the middle . here
, differences in the mutation rates have a stronger impact , as the three curves with @xmath97 ( blue , solid ) , @xmath128 ( red , long dashed ) and @xmath129 ( green , short dashed ) demonstrate .
the increasing @xmath124 shifts the existing stable state towards the edge of the system until it vanishes .
system parameters were @xmath98 and a symmetric interaction function according to equation ( [ eq : interactions ] ) with one stable state at @xmath130 with @xmath131 , @xmath132.,width=566 ] as we saw above , the fit of the theoretically calculated stationary distribution to simulation data is almost perfect for weak selection , i.e. the maximal fitness difference satisfies @xmath133 , cf .
e.g. @xcite . to quantify the quality of the fit we measured the stationary distribution with the same parameters as above for different population sizes @xmath0
this is shown in figure [ fig : nplot ] demonstrating that for increasing @xmath0 the data gets ever closer to the theoretically obtained distribution .
for this we define the empirical distribution @xmath134 where @xmath135 is the discrete process defined in equation ( [ eq : meq ] ) .
@xmath136 is a time large enough for the process to reach stationarity and depends on the system size , as does the measurement time @xmath137 which is specified in figure [ fig : nplot ] . using this definition we quantify the quality of the fit using the distance measure @xmath138 which takes the mean distance of the measured empirical distribution @xmath139 from the theoretical distribution for every point except @xmath51 and @xmath52 where the density @xmath140 diverges . here , the the theoretical distribution is calculated by integrating the theoretical density over bins of the size @xmath141 around the points @xmath142 .
figure [ fig : nplot]b shows that the mean distance @xmath143 decays with increasing @xmath0 .
the decay of this quantity is dominated by the slow convergence at the domain boundaries @xmath51 and @xmath52 .
therefore , we additionally defined the maximum distance measure @xmath144}\left\{\left|\pi_k-\int_{k / n-1/2n}^{k / n+1/2n}\rho^\ast(x)dx\right|\right\}\label{eq : maxdistance}\ ] ] which gives the maximum distance between the measured probability distribution from the theoretical distribution for all points except @xmath51 and @xmath52 due to the same argument as above .
figure [ fig : nplot]b shows that the maximum distance @xmath145 decreases with increasing @xmath0 , however slower than the mean distance @xmath143 due to the divergence of @xmath146 at the boundaries .
altogether we conclude , that the theoretical curve fits the data very well already for @xmath147 .
however , near the domain boundaries special care has to be taken , as divergences of the theoretical curve lead to larger deviations . .
( a ) shows the theoretical curve from equation ( [ eq : generalsolution ] ) ( black , solid ) for the same system as in figure [ fig : firstexample ] .
data points show empirical distributions as defined in equation ( [ eq : empiricaldensity ] ) obtained in simulations for @xmath148 ( blue , @xmath106 ) , @xmath149 ( red , @xmath150 ) , @xmath151 ( green , @xmath152 ) and @xmath98 ( gray , @xmath116 ) .
( b ) shows the distances @xmath143 and @xmath145 of simulation data and theoretical curve for different @xmath0 for the mean distance measure @xmath143 defined in equation ( [ eq : meandistance ] ) ( blue , @xmath106 ) and the maximum distance measure @xmath145 defined in equation ( [ eq : maxdistance ] ) ( red , @xmath150 ) , demonstrating that the distance decays for increasing @xmath0 .
the solid line @xmath153 and the dashed line @xmath154 are added as a guide to the eye for the relation between measured distances and system size @xmath0 .
the measured empirical distributions for both ( a ) and ( b ) were obtained by simulating the system dynamics from an initial state drawn from @xmath140 for a mixing time @xmath155 and then recording the density for a time @xmath156.,width=566 ] furthermore , as the fokker - planck approximation only holds for weak selection we study the quality of the solution obtained throught the fokker - planck equation in dependence of the selection strength . for this
we introduce a scaling factor @xmath157 to the interaction functions , so that they become @xmath158 we find that the stationary solution of the fokker - planck equation well approximates the solution of the master equation @xmath159 for selection strengths up to @xmath157 of the order of 1 ( cf .
figure [ fig : selectionplot ] ) .
we derive the solution ( [ eq : mastersolutionintext ] ) of the master equation in appendix b. note that for the derivation of this solution no approximation is necessary and it hence fits data from simulations perfectly up to an error due to finite sampling in simulations . as figure [ fig : selectionplot ] illustrates , for strong selection @xmath160 the solution of the fokker - planck equation does not fit the exact solution of the master equation perfectly , but still catches the overall trend of the dynamics .
the error , quantified by the mean distance measure analogous to ( [ eq : meandistance ] ) , increases with @xmath157 as a power law both in comparison with the solution from the master equation and with data from simulations , cf .
figure [ fig : selectionplot]d .
all in all , for weak selection the fokker - planck approximation works well while for strong selection it does not perfectly predict the stationary distribution , but still reflects the overall trend of the dynamics .
if this approximation is not satisfying , the direct solution ( [ eq : mastersolutionintext ] ) of the master equation yields the exact distribution fitting the data for all selection strengths . )
obtained from the fokker - planck equation well fits the exact solution ( [ eq : mastersolutionintext ] ) from the master equation for weak selection , but not for strong selection .
( a ) shows the stationary solution from the fokker - planck equation ( red , solid ) for the system with interaction functions defined by equation ( [ eq : scaledinteractions ] ) for very weak selection @xmath161 together with the solution from the master equation ( blue , @xmath106 ) .
the distributions of the same system for weak selection @xmath162 and strong selection @xmath163 are shown in ( b ) and ( c ) , respectively .
all curves are plotted logarithmically to better allow a comparison of the deviations over all scales .
( c ) shows that the curve from the fokker - planck equation does not fit the exact master solution very well .
this is quantified in ( d ) showing the mean distance @xmath143 between fokker - planck and master solution ( blue , @xmath106 ) and between fokker - planck solution and an empirical density obtained from simulations ( red , @xmath150 ) .
the system parameters were @xmath98 , @xmath105 , @xmath131 and @xmath132 .
the measured empirical distributions were obtained by simulating the system dynamics from an initial state drawn from @xmath140 for a mixing time @xmath164 and then recording the density for a time @xmath165.,width=566 ] we have shown above that the dynamics of a two - genotype system can exhibit multiple stable points induced by nonlinear selection .
even more , there is no theoretical limit to the number of stable points in the system .
assume for example @xmath121 and @xmath97 , so that mutation exactly balances genetic drift .
then , the stationary distribution ( [ eq : generalsolution ] ) has a maximum at each point , where @xmath166 $ ] has a maximum .
theoretically @xmath94 and @xmath95 can be any function with an arbitrary amount of extreme points in @xmath112 $ ] , so that there is no limit to the stationary distribution s number of maxima , if selection dominates the dynamics .
however , for finite @xmath0 the number of possible maxima is naturally limited by @xmath167 .
figure [ fig : multiplestability ] shows an example , where we used periodic interaction functions@xmath168 although this is not a realistic interaction function in most applications , it demonstrates what is theoretically possible in the introduced system . )
( red , solid ) together with simulation data with @xmath98 ( blue , @xmath106 ) for the interaction functions given in ( [ eq : perfitfunc ] ) .
the computed stationary distribution diverges for @xmath76 and @xmath77 while the simulation data remains finite due to the finite number of individuals .
( b ) shows a sample path which exhibits switching between the different maxima of the distribution .
parameters are @xmath169 , @xmath170 , @xmath171 and @xmath97.,width=566 ]
in this article we analyzed a two - genotype system in a very general setting with ( possibly ) asymmetric mutation probabilities and nonlinear fitness functions @xcite in finite populations .
the underlying moran process is a well established model @xcite to gain an understanding of the interplay of selection , mutation and genetic drift in evolutionary dynamics .
however , the moran process is studied mostly with symmetric mutation probabilities and at most linear interaction functions .
we reasoned that neither need mutation probabilities be symmetric as experiments have shown , that mutation probabilities are often asymmetric @xcite nor can all interaction effects be described by linear interaction functions .
for example cooperation in game theory leads to an interaction function increasing linearly in the frequency of the cooperating genotype . yet , in many applications also cooperators in the end compete for the same type of resource which is limited .
therefore , due to limited resources a population being too large can not be sustained leading to a decrease in the fitness .
there is no linear function that can reflect both of these effects at the same time .
we derived the fokker - planck equation describing the dynamics of the number of individuals @xmath35 of genotype @xmath1 in the limit of large population sizes @xmath0 .
we quantified the quality of the fokker - planck approach for an example ( cf . figure 4 ) where the difference of simulation data and theoretical solution became almost not detectable for population sizes larger than @xmath147 .
actually , if the system exhibits absorbing states then the fokker - planck method does not work to study the corresponding quasi - stationary distributions .
instead , wkb methods are more appropriate to describe the system dynamics as for example in @xcite , where fixation resulting from large fluctuations was studied . in our model system
no such absorbing states exist , as long as the mutation rates are positive ( @xmath172 ) and therefore the fokker - planck equation is appropriate to describe the system dynamics .
we identified the individual effects of selection , mean mutation rate and mutation difference as well as genetic drift and derived the stationary probability distribution as determined by the fokker - planck equation . analyzing the distribution
, we found that asymmetries in the mutation probabilities may not only induce the shifting of existing stable points of the dynamics to new positions , but also lead to the emergence of new stable points .
thus , a genotype that has a selective disadvantage can anyway have a stable dynamical state where its individuals dominate the population due to a higher mutational stability ( see figure [ fig : asymmetricexample ] ) .
further we found , that dynamic fitness leads to multiple stable points of the dynamics induced by selection and also genetic drift .
we showed an example ( figure [ fig : firstexample ] ) where three stable points exist , two caused by selection and one by genetic drift .
we conclude that frequency - dependent fitness together with asymmetric mutation rates induces complex evolutionary dynamics , in particular if the interactions imply nonlinear fitness functions . theoretically , there is no limit for the number of stable points that the dynamics can exhibit ( see figure [ fig : multiplestability ] ) .
all in all , we interprete our results such that in real biological systems multiple metastable equilibria may exist , whenever species interact in a way complex enough to imply a fitness that nonlinearly depends on frequency . as a consequence , one species may exhibit a certain frequency for a long time before a sudden shift occurs and then a new frequency prevails .
such a change may thus occur even in the absence of changes of the environment ; it may be induced as well by a stochastic switching from one metastable state to another due to complex inter - species interactions .
the moran process is a standard tool to gain theoretical insights into experimental data @xcite . of course
, we do not propose here that any experimental setup may be exactly described by the moran process .
however , we think that it should be feasible to develop an experiment where two different mutants evolve with asymmetric mutation rates . to find an experimental setup where the two genotypes also exhibit nonlinear fitness could however prove more difficult .
rather , our study is a theoretical study indicating that nonlinear fitness may be the cause of multible stable states when observed in experimental data . for further studies on systems with more genotypes it should be useful to combine our considerations presented here with the work of traulsen et al .
@xcite , where an analysis of systems with more than two genotypes was carried out .
extending those results it may be possible to gain a better understanding of the effects of nonlinear interactions for many different genotypes . also , it may be interesting to study the effects of changing interactions , where the interactions change according to the system dynamics @xcite .
thus , our study might serve as a promising starting point to investigate how nonlinear frequency dependencies impact evolutionary dynamics in complex environments .
* acknowledgements * we thank steven strogatz for fruitful discussions during project initiation and stefan eule for helpful technical discussions .
stefan grosskinsky acknowledges support by epsrc , grant no .
ep / e501311/1 .
the master equation ( [ eq : meq ] ) may be transformed to a fokker - planck equation in the limit of large @xmath0 @xcite .
we introduce the transformation@xmath173 together with the rescaled functions@xmath174 we fix the scaling functions @xmath175 , @xmath176 and @xmath177 such that in the limit @xmath59 all terms in equation ( [ eq : meq ] ) remain finite so that mutation , selection and genetic drift all act on the same scale .
we further define@xmath178 and substituting all this into the master equation ( [ eq : meq ] ) , we obtain@xmath179\rho({x_{-}},s)\right.\nonumber \\ & & \left[(1-\mu_{ba})(1+g_{b}({x_{+}}))(1-{x_{+}}){x_{+}}+\mu_{ab}(1+g_{a}({x_{+}})){x_{+}}^{2}\right]\rho({x_{+}},s)\nonumber \\ & & -\left[(1-\mu_{ab})(1+g_{a}(x))x(1-x)+\mu_{ba}(1+g_{b}(x))(1-x)^{2}\right.\nonumber \\ & & + \left.\left.(1-\mu_{ba})(1+g_{b}(x))x(1-x)+\mu_{ab}(1+g_{a}(x))x^{2}\right]\rho(x , s)\right\ } \label{eq : app1}\end{aligned}\ ] ] we choose @xmath180 and @xmath181 so that in the limit @xmath59 the terms stay finite .
further we introduce the mean mutation rate @xmath65 and the mutation rate difference @xmath66 . to not overload the notation
we drop the time argument @xmath182 of @xmath183 in the following calculation .
this leads to @xmath184\right\ } \\ & & + n\left\ { { \tilde{g}_a}({x_{-}}){x_{-}}(1-{x_{-}})\rho({x_{-}})-{\tilde{g}_a}(x)x(1-x)\rho(x)+{\tilde{g}_b}({x_{+}}){x_{+}}(1-{x_{+}})\rho({x_{+}})-{\tilde{g}_b}(x)x(1-x)\rho(x)\right.\\ & & + \frac{{\tilde{\mu}}}{2}\left[(1 - 2{x_{-}})\rho({x_{-}})-(1 - 2{x_{+}})\rho({x_{+}})\right]\\ & & + \delta{\tilde{\mu}}\left[{x_{+}}\rho({x_{+}})-x\rho(x)-(1-{x_{-}})\rho({x_{-}})+(1-x)\rho(x)\right]\\ & & + \frac{{\tilde{\mu}}}{n}\left[\left({\tilde{g}_a}({x_{+}}){x_{+}}^{2}+{\tilde{g}_b}({x_{+}})({x_{+}}^{2}-{x_{+}})\right)\rho({x_{+}})-\left({\tilde{g}_a}(x)x^{2}+{\tilde{g}_b}(x)(x^{2}-x)\right)\rho(x)\right.\\ & & + \left.\left({\tilde{g}_a}({x_{-}})({x_{-}}^{2}-{x_{-}})+{\tilde{g}_b}({x_{-}})(1-{x_{-}})^{2}\right)\rho({x_{-}})-\left({\tilde{g}_a}(x)(x^{2}-x)+{\tilde{g}_b}(x)(1-x)^{2}\right)\rho(x)\right]\\ & & + \frac{\delta{\tilde{\mu}}}{n}\left[\left({\tilde{g}_a}({x_{+}}){x_{+}}^{2}-{\tilde{g}_b}({x_{+}})({x_{+}}^{2}-{x_{+}})\right)\rho({x_{+}})-\left({\tilde{g}_a}(x)x^{2}-{\tilde{g}_b}(x)(x^{2}-x)\right)\rho(x)\right.\\ & & + \left.\left.\left({\tilde{g}_a}({x_{-}})({x_{-}}^{2}-{x_{-}})-{\tilde{g}_b}({x_{-}})(1-{x_{-}})^{2}\right)\rho({x_{-}})-\left({\tilde{g}_a}(x)(x^{2}-x)-{\tilde{g}_b}(x)(1-x)^{2}\right)\rho(x)\right]\right\ } \end{aligned}\ ] ] in the limit @xmath59 the different terms with @xmath185 in front become second order derivatives with respect to @xmath68 , while the other terms become first order derivatives .
the terms which have a @xmath141 factor vanish in the limit @xmath59 and thus the above equation becomes @xmath186x(1-x)\rho(x)\right)-\delta{\tilde{\mu}}\rho(x)\right]+\frac{\partial^{2}}{\partial x^{2}}\left[x(1-x)\rho(x)\right]\label{eq : appfpe}\ ] ] which is the focker - planck - equation of the system .
we directly derive the stationary solution @xmath187 of the master equation ( [ eq : meq ] ) using the detailed balance equation @xmath188 which applies to any chain with only nearest neighbour transitions @xcite , cf .
also @xcite .
thus , using the rewritten balance equation @xmath189 iteratively , we obtain @xmath190 finally , we may use the normalization condition @xmath191 of the stationary distribution to eliminate the factor @xmath192 .
we then obtain the exact stationary solution of the master equation @xmath193 which can be evaluated numerically for any transition rates @xmath194 and @xmath195 . for more details on the exact solution of the master equation in the moran process
see for example the work by claussen and traulsen @xcite . | evolution is simultaneously driven by a number of processes such as mutation , competition and random sampling .
understanding which of these processes is dominating the collective evolutionary dynamics in dependence on system properties is a fundamental aim of theoretical research .
recent works quantitatively studied coevolutionary dynamics of competing species with a focus on linearly frequency - dependent interactions , derived from a game - theoretic viewpoint . however
, several aspects of evolutionary dynamics , e.g. limited resources , may induce effectively nonlinear frequency dependencies . here
we study the impact of nonlinear frequency dependence on evolutionary dynamics in a model class that covers linear frequency dependence as a special case .
we focus on the simplest non - trivial setting of two genotypes and analyze the co - action of nonlinear frequency dependence with asymmetric mutation rates .
we find that their co - action may induce novel metastable states as well as stochastic switching dynamics between them .
our results reveal how the different mechanisms of mutation , selection and genetic drift contribute to the dynamics and the emergence of metastable states , suggesting that multistability is a generic feature in systems with frequency - dependent fitness .
* keywords : * population dynamics ; dynamic fitness ; stochastic switching ; multistability |
the _ weak lefschetz property _ for a standard graded artinian algebra @xmath0 over a field @xmath1 is a natural property .
it says that there is a linear form @xmath2 such that the multiplication map @xmath3_i \rightarrow [ a]_{i+1}$ ] has maximal rank for all @xmath4 ( i.e. , it is injective or surjective ) .
its presence implies , for example , restrictions on the hilbert function and graded betti numbers of the algebra ( see @xcite ) .
recent studies have connected the weak lefschetz property to many other questions ( see , e.g. , @xcite ) .
thus , a great variety of tools from representation theory , topology , vector bundle theory , hyperplane arrangements , plane partitions , splines , differential geometry , among others has been used to decide the presence of the weak lefschetz property ( see , e.g. , @xcite ) .
an important aspect has also been the role of the characteristic of @xmath1 .
any artinian quotient of a polynomial ring in at most two variables has the weak lefschetz property regardless of the characteristic of @xmath1 ( see @xcite and ( * ? ? ?
* proposition 2.7 ) ) .
this is far from true for quotients of rings with three or more variables .
here we consider quotients @xmath5 , where @xmath6 $ ] and @xmath7 is a monomial ideal containing a power of @xmath8 , and @xmath9 . if @xmath7 has only three generators , then @xmath5 has the weak lefschetz property , provided the base field has characteristic zero ( see @xcite ) .
we focus on the case , where @xmath7 has four minimal generators , extending previous work in @xcite . to this end
we use a combinatorial approach developed in @xcite that involves lozenge tilings , perfect matchings , and families of non - intersecting lattice paths .
some of our results have already been used in @xcite . in section
[ sec : trireg ] , we recall the connection between monomial ideals in three variables and so - called triangular regions .
we use it to establish sufficient and necessary conditions for a balanced triangular subregion to be tileable ( see corollary [ cor : pp - tileable ] ) . in section
[ sec : alg ] , we show that the tileability of a triangular subregion @xmath10 is related to the semistability of the syzygy bundle of the ideal @xmath7 ( see theorem [ thm : tileable - semistable ] ) .
we further recall the relation between lozenge tilings of triangular regions and the weak lefschetz property .
all the results up to this point are true for arbitrary artinian monomial ideals of @xmath11 . in section [ sec : amaci ] we consider exclusively artinian monomial ideals with four minimal generators .
our results on the weak lefschetz property of @xmath5 are summarized in theorem [ thm : amaci - wlp ] . in particular , they provide further evidence for a conjecture in @xcite , which concerns the case where @xmath5 is a level algebra . furthermore , we determine the generic splitting type of the syzygy bundle of @xmath7 in all cases but one ( see propositions [ pro : st - nss ] and [ pro : split - type - semist ] ) . in the remaining case
we show that determining the generic splitting type is equivalent to deciding whether @xmath5 has the weak lefschetz property ( see theorem [ thm : equiv ] ) .
this result is independent of the characteristic .
besides introducing notation , we recall needed facts from the combinatorial approach to lefschetz properties developed in @xcite .
we also establish a new criterion for tileability by lozenges .
let @xmath6 $ ] be a standard graded polynomial ring over a field @xmath1 , i.e. , @xmath12 . unless specified otherwise , @xmath1 is always an arbitrary field .
all @xmath11-modules in this paper are assumed to be finitely generated and graded .
let @xmath13_j$ ] be a graded quotient of @xmath11 .
the _ hilbert function _ of @xmath0 is the function @xmath14 given by @xmath15_j$ ] .
the _ socle _ of @xmath0 , denoted @xmath16 , is the annihilator of @xmath17 , the homogeneous maximal ideal of @xmath11 , that is , @xmath18 .
let @xmath7 be a monomial ideal of @xmath11 . as @xmath5 is standard graded , the monomials of @xmath11 of degree @xmath19 that are _ not _ in @xmath7 form a @xmath1-basis of @xmath20_d$ ] .
let @xmath21 be an integer .
consider an equilateral triangle of side length @xmath22 that is composed of @xmath23 downward - pointing ( @xmath24 ) and @xmath25 upward - pointing ( @xmath26 ) equilateral unit triangles .
we label the downward- and upward - pointing unit triangles by the monomials in @xmath27_{d-2}$ ] and @xmath27_{d-1}$ ] , respectively , as follows : place @xmath28 at the top , @xmath29 at the bottom - left , and @xmath30 at the bottom - right , and continue labeling such that , for each pair of an upward- and a downward - pointing triangle that share an edge , the label of the upward - pointing triangle is obtained from the label of the downward - pointing triangle by multiplying with a variable .
the resulting labeled triangular region is the _ triangular region ( of @xmath11 ) in degree @xmath22 _ and is denoted @xmath31 .
see figure [ fig : triregion - r](i ) for an illustration . and with respect to @xmath5.,title="fig : " ] + _ ( i ) @xmath32 _ and with respect to @xmath5.,title="fig : " ] + _ ( ii ) @xmath33 _ throughout this manuscript we order the monomials of @xmath11 with the _ graded reverse - lexicographic order _ , that is , @xmath34 if either @xmath35 or @xmath36 and the _ last _ non - zero entry in @xmath37 is _
negative_. for example , in degree @xmath38 , @xmath39 thus in @xmath32 , see figure [ fig : triregion - r](iii ) , the upward - pointing triangles are ordered starting at the top and moving down - left in lines parallel to the upper - left edge .
we generalize this construction to quotients by monomial ideals .
let @xmath7 be a monomial ideal of @xmath11 .
the _ triangular region ( of @xmath5 ) in degree @xmath22 _ , denoted by @xmath40 , is the part of @xmath31 that is obtained after removing the triangles labeled by monomials in @xmath7 .
note that the labels of the downward- and upward - pointing triangles in @xmath40 form @xmath1-bases of @xmath20_{d-2}$ ] and @xmath20_{d-1}$ ] , respectively .
it is more convenient to illustrate such regions with the removed triangles darkly shaded instead of being removed .
see figure [ fig : triregion - r](ii ) for an example .
notice that the regions missing from @xmath31 in @xmath40 can be viewed as a union of ( possibly overlapping ) upward - pointing triangles of various side lengths that include the upward- and downward - pointing triangles inside them .
each of these upward - pointing triangles corresponds to a minimal generator of @xmath7 that has , necessarily , degree at most @xmath41 .
we can alternatively construct @xmath40 from @xmath31 by removing , for each minimal generator @xmath42 of @xmath7 of degree at most @xmath41 , the _ puncture associated to @xmath43 _ which is an upward - pointing equilateral triangle of side length @xmath44 located @xmath45 triangles from the bottom , @xmath46 triangles from the upper - right edge , and @xmath47 triangles from the upper - left edge .
see figure [ fig : triregion - punctures ] for an example .
we call @xmath44 the _ side length of the puncture associated to @xmath43 _ , regardless of possible overlaps with other punctures in @xmath10 .
as constructed by removing punctures.,title="fig : " ] + _ ( i ) @xmath48 _ as constructed by removing punctures.,title="fig : " ] + _ ( ii ) @xmath49 _ we say that two punctures _ overlap _ if they share at least an edge .
two punctures are said to be _ touching _ if they share precisely a vertex .
a _ lozenge _ is a union of two unit equilateral triangles glued together along a shared edge , i.e. , a rhombus with unit side lengths and angles of @xmath50 and @xmath51 .
lozenges are also called calissons and diamonds in the literature .
see figure [ fig : triregion - intro ] . together with one of its @xmath52 tilings . ] together with one of its @xmath52 tilings . ]
fix a positive integer @xmath22 and consider the triangular region @xmath31 as a union of unit triangles .
thus a _ subregion _
@xmath53 is a subset of such triangles .
we retain their labels . as
above , we say that a subregion @xmath54 is _
@xmath24-heavy _ , _ @xmath26-heavy _ , or _ balanced _
if there are more downward pointing than upward pointing triangles or less , or if their numbers are the same , respectively .
a subregion is _ tileable _ if either it is empty or there exists a tiling of the region by lozenges such that every triangle is part of exactly one lozenge .
since a lozenge in @xmath31 is the union of a downward - pointing and an upward - pointing triangle , and every triangle is part of exactly one lozenge , a tileable subregion is necessarily balanced .
let @xmath53 be any subregion . given a monomial @xmath43 with degree less than @xmath22 , the _ monomial subregion _ of @xmath54 associated to @xmath43 is the part of @xmath54 contained in the triangle @xmath45 units from the bottom edge , @xmath46 units from the upper - right edge , and @xmath47 units from the upper - left edge .
in other words , this monomial subregion consists of the triangles that are in @xmath54 and the puncture associated to the monomial @xmath43 .
see figure [ fig : triregion - subregion ] for an example .
( see figure [ fig : triregion - intro ] ) associated to @xmath55 . ]
replacing a tileable monomial subregion by a puncture of the same size does not alter tileability .
* lemma 2.1)[lem : replace - tileable ] let @xmath53 be any subregion . if a monomial subregion @xmath56 of @xmath54 is tileable , then @xmath54 is tileable if and only if @xmath57 is tileable .
moreover , each tiling of @xmath54 is obtained by combining a tiling of @xmath57 and a tiling of @xmath56 .
let @xmath58 be a monomial subregion , and let @xmath59 be any subregions such that @xmath60 .
if @xmath61 and @xmath62 are both tileable , then @xmath54 is tileable if and only if @xmath63 is , by lemma [ lem : replace - tileable ] .
in other words , replacing a tileable monomial subregion of a triangular region by a tileable monomial subregion of the same size does not affect tileability .
* theorem 2.2)[thm : tileable ] let @xmath64 be a balanced triangular region , where @xmath65 is any monomial ideal
. then @xmath54 is tileable if and only if @xmath54 has no @xmath24-heavy monomial subregions .
let @xmath7 be a monomial ideal of @xmath11 whose punctures in @xmath31 ( corresponding to the minimal generators of @xmath7 having degree less than @xmath22 ) have side lengths that sum to @xmath66
. then we define the _ over - puncturing coefficient _ of @xmath7 in degree @xmath22 to be @xmath67 . if @xmath68 , @xmath69 , or @xmath70 , then we call @xmath7 _ under - punctured _ , _ perfectly - punctured _ , or _ over - punctured _ in degree @xmath22 , respectively
. let now @xmath64 be a triangular region with punctures whose side lengths sum to @xmath66 .
then we define similarly the _ over - puncturing coefficient _ of @xmath54 to be @xmath71 . if @xmath72 , @xmath73 , or @xmath74 , then we call @xmath54 _ under - punctured _ , _ perfectly - punctured _ , or _ over - punctured _ , respectively
. observe that different monomial ideals can determine the same triangular region of @xmath31 .
consider , for example , @xmath75 and @xmath76 .
then @xmath77 , and @xmath78 but @xmath79 .
however , given a triangular region @xmath80 , there is a unique largest ideal @xmath81 that is generated by monomials whose degrees are bounded above by @xmath41 and that satisfies @xmath82 .
we call @xmath83 the _ monomial ideal of the triangular region @xmath54_. note that @xmath84 , and equality is true if and only if the ideals @xmath7 and @xmath83 are the same in all degrees less than @xmath22 . [ rem : puncture coeff ] if a monomial subregion @xmath54 of @xmath31 has no overlapping punctures , then @xmath85 is equal to the number of downwards - pointing unit triangles in @xmath54 minus the number of upward - pointing unit triangles in @xmath54 .
perfectly - punctured regions admit a numerical tileability criterion .
[ cor : pp - tileable ] let @xmath64 be a triangular region . then any two of the following conditions imply the third : 1 .
@xmath54 is perfectly - punctured ; 2 .
@xmath54 has no over - punctured monomial subregions ; and 3 .
@xmath54 is tileable .
suppose @xmath54 is tileable .
then @xmath54 has no @xmath24-heavy monomial subregions by theorem [ thm : tileable ] .
thus , every monomial subregion of @xmath54 is not over - punctured if and only if no punctures of @xmath54 overlap . hence ( ii ) implies ( i ) by remark [ rem : puncture coeff ] because @xmath54 is balanced .
for the converse it is enough to show : if some punctures of @xmath54 overlap , then @xmath54 is over - punctured . indeed ,
if no punctures overlap , then @xmath54 is perfectly punctured because @xmath54 is balanced .
so assume two punctures of @xmath54 overlap .
then the smallest monomial subregion @xmath56 of @xmath54 containing these two punctures does not overlap with any other puncture of @xmath54 and is uniquely tileable .
hence @xmath57 is tileable by lemma [ lem : replace - tileable ] , and thus @xmath86 , as desired . if @xmath54 is non - tileable , then @xmath54 has a @xmath24-heavy monomial subregion .
since every @xmath24-heavy monomial subregion is also over - punctured , it follows that @xmath54 has an over - punctured monomial subregion .
any subregion @xmath53 can be associated to a bipartite planar graph @xmath87 that is an induced subgraph of a honeycomb graph ( see @xcite ) .
we are interested in the bi - adjacency matrix @xmath88 of @xmath87 .
this is a zero - one matrix whose determinant enumerates signed lozenge tilings ( see ( * ? ? ?
* theorem 3.5 ) ) .
if @xmath64 for some monomial ideal @xmath7 , then @xmath88 admits an alternative description .
indeed , consider the multiplication map @xmath89_{d-2 } \rightarrow [ r / i]_{d-1}$ ] .
let @xmath90 be the matrix to this linear map with respect to the monomial bases of @xmath20_{d-2}$ ] and @xmath20_{d-1}$ ] in reverse - lexicographic order .
then the transpose of @xmath90 is the bi - adjacency matrix @xmath91 ( see ( * ? ? ?
* proposition 4.5 ) ) .
here we need only a special case of these results .
[ prop : pm - det ] assume @xmath92 is a non - empty balanced subregion . if @xmath93 is not zero , then @xmath54 is tileable .
balancedness of @xmath54 is equivalent to @xmath94_{d-2 } = \dim_k [ r / i]_{d-1}$ ] .
it follows that @xmath88 is a square matrix by ( * ? ? ?
* proposition 4.5 ) .
now ( * ? ? ?
* theorem 3.5 ) gives the assertion .
we conclude this section with a criterion that guarantees non - vanishing of @xmath95 . to this end
we recursively define a puncture of @xmath53 to be a _ non - floating _
puncture if it touches the boundary of @xmath96 or if it overlaps or touches a non - floating puncture of @xmath54 .
otherwise we call a puncture a _ floating _ puncture .
for example , the region @xmath54 in figure [ fig : triregion - intro ] has three non - floating punctures ( in the corners ) and three floating punctures , two of them are overlapping and have side length two .
* corollary 4.7 ) [ prop : same - sign ] let @xmath54 be a tileable triangular region , and suppose all floating punctures of @xmath54 have an even side length . then @xmath97 .
in this section , we use the connection to triangular regions to reinterpret some algebraic properties . throughout this subsection
, we assume the characteristic of @xmath1 is zero .
let @xmath7 be an artinian ideal of @xmath98 $ ] that is minimally generated by forms @xmath99 .
the _ syzygy module _ of @xmath7 is the graded module @xmath100 that fits into the exact sequence @xmath101 its sheafification @xmath102 is a vector bundle on @xmath103 , called the _ syzygy bundle _ of @xmath7 .
it has rank @xmath104 .
semistability is an important property of a vector bundle .
let @xmath105 be a vector bundle on projective space .
the _ slope _ of @xmath105 is defined as @xmath106 .
furthermore , @xmath105 is said to be _
semistable _ if the inequality @xmath107 holds for every coherent subsheaf @xmath108 .
if the inequality is always strict , then @xmath105 is said to be _
stable_. brenner established a beautiful characterization of the semistability of syzygy bundles to monomial ideals .
since we only consider monomial ideals in this work , the following may be taken as the definition of ( semi)stability herein .
* proposition 2.2 & corollary 6.4 ) [ thm : stable - syz ] let @xmath7 be an artinian ideal in @xmath109 $ ] that is minimally generated by monomials @xmath110 , where @xmath1 is a field of characteristic zero
. then @xmath7 has a semistable syzygy bundle if and only if , for every proper subset @xmath81 of @xmath111 with at least two elements , the inequality @xmath112 holds , where @xmath113 is the degree of the greatest common divisor of the @xmath114 with @xmath115 .
further , @xmath7 has a stable syzygy bundle if and only if the above inequality is always strict .
we use brenner s criterion to rephrase ( semi)stability in the case of a monomial ideal of @xmath116 $ ] in terms of the over - puncturing coefficients of ideals .
note , in particular , that @xmath117 .
[ cor : t - stable ] let @xmath7 be an artinian ideal in @xmath6 $ ] that is minimally generated by monomials @xmath110 of degree at most @xmath22 . for every proper subset @xmath81 of @xmath111 with at least two elements , let @xmath118 be the monomial ideal that is generated by @xmath119 , where @xmath120 has degree @xmath113 .
then @xmath7 has a semistable syzygy bundle if and only if , for every proper subset @xmath81 of @xmath111 with at least two elements , the inequality @xmath121 holds .
furthermore , @xmath7 has a stable syzygy bundle if and only if the above inequality is always strict . since @xmath122
, this follows immediately from theorem [ thm : stable - syz ] . in order to apply this result
we slightly extend the concept of a triangular region @xmath10 .
label the vertices in @xmath31 by monomials of degree @xmath22 such that the label of each unit triangle is the greatest common divisor of its vertex labels .
then a minimal monomial generator of @xmath7 with degree @xmath22 corresponds to a vertex of @xmath31 that is removed in @xmath10 .
we consider this removed vertex as a puncture of side length zero .
observe that this is in line with our general definition of the side length of a puncture .
using corollary [ cor : pp - tileable ] , we see that semistability is strongly related to tileability of a region .
[ thm : tileable - semistable ] let @xmath7 be an artinian ideal in @xmath6 $ ] generated by monomials whose degrees are bounded above by @xmath22 , and let @xmath64 .
if @xmath54 is non - empty , then any two of the following conditions imply the third : 1 .
@xmath7 is perfectly - punctured ; 2 .
@xmath54 is tileable ; and 3 .
@xmath102 is semistable .
assume @xmath7 is perfectly punctured , that is , @xmath123 .
we will show that @xmath54 is tileable if and only if @xmath102 is semistable . if @xmath54 is tileable , then @xmath124 which implies @xmath125 .
hence no punctures of @xmath54 overlap .
this further implies @xmath126 for any subset @xmath0 of the generators of @xmath7 .
thus @xmath127 , since no punctures overlap and every subregion is not over - punctured by corollary [ cor : pp - tileable ] .
hence , @xmath128 is semistable by corollary [ cor : t - stable ] . if @xmath128 is semistable
, then @xmath129 holds for any subset @xmath0 of the generators of @xmath7 .
this implies , in particular , that no punctures of @xmath54 overlap .
hence @xmath130 and so @xmath131 .
furthermore , since no pair of punctures overlap , having no over - punctured regions is the same as having no @xmath24-heavy regions ( see remark [ rem : puncture coeff ] ) .
thus , by corollary [ cor : pp - tileable ] , @xmath54 is tileable .
now assume @xmath7 is not perfectly - punctured , but @xmath54 is tileable .
we have to show that @xmath102 is not semistable . arguing as in the proof of corollary [ cor : pp - tileable ] , we conclude that @xmath54 is over - punctured and must have overlapping punctures .
consider two such overlapping punctures of @xmath54 .
then the smallest monomial subregion @xmath56 containing these two punctures does not overlap with any other puncture of @xmath54 with positive side length .
hence @xmath132 is tileable and @xmath133 .
if @xmath63 is still over - punctured , then we repeat the above replacement procedure until we get a perfectly - punctured monomial subregion of @xmath54 . abusing notation slightly ,
denote this region by @xmath63 .
let @xmath81 be the largest monomial ideal containing @xmath7 and with generators whose degrees are bounded above by @xmath22 such that @xmath134 .
observe that @xmath135 .
notice that a single replacement step above amounts to replacing the triangular region to an ideal @xmath136 by the region to the ideal @xmath137 , where @xmath138 is a greatest common divisor of the minimal generators of @xmath136 that correspond to two overlapping punctures .
these generators have degrees less than @xmath22 .
assume now that @xmath63 is empty .
then @xmath7 has two relatively prime minimal generators , say @xmath139 , whose corresponding punctures overlap and are not both contained in a proper monomial subregion of @xmath31 . since @xmath7 is artinian it has @xmath140 minimal generators .
moreover , all minimal generators of @xmath7 other than @xmath141 and @xmath142 have degree @xmath22 .
it follows that @xmath143 . since @xmath144 , corollary [ cor : t - stable ] shows that @xmath145 is not semistable .
it remains to consider the case where @xmath63 is not empty , i.e. , @xmath81 is a proper ideal of @xmath11
. let @xmath110 and @xmath146 be the minimal monomial generators of @xmath7 and @xmath81 , respectively .
partition the generating set of @xmath7 into @xmath147 .
notice @xmath148 .
in particular , @xmath149 as @xmath81 is a proper artinian ideal .
set @xmath150 .
observe @xmath151 as the subregion of @xmath10 associated to @xmath152 is tileable , hence not under - punctured .
moreover , @xmath153 as @xmath154 , we conclude that @xmath155 and , in particular , @xmath156 for each @xmath157 .
assume @xmath158 for all @xmath157
. then @xmath159 . but
this implies @xmath160 , which is absurd .
hence , there is some @xmath161 such that @xmath162 . since @xmath163 it follows that @xmath164 . indeed , this is immediate if @xmath165 .
if @xmath166 , then it is also true because @xmath167 .
now corollary [ cor : t - stable ] gives that @xmath102 is not semistable .
we get the following criterion when focusing solely on the triangular region . recall that @xmath83 denotes the monomial ideal of a triangular region @xmath54 as introduced above remark [ rem : puncture coeff ] .
[ cor : semistability - by - region ] let @xmath7 be an artinian ideal in @xmath6 $ ] generated by monomials whose degrees are bounded above by @xmath22 , and let @xmath64 .
assume @xmath54 is non - empty and tileable . 1 . if @xmath168 , then @xmath102 is not semistable .
@xmath169 is semistable if and only if @xmath54 is perfectly - punctured .
note that @xmath170 implies @xmath171 .
since @xmath54 is balanced , we get @xmath172 .
hence theorem [ thm : tileable - semistable ] gives our assertions . for stability , we obtain the following result . [
pro : pp - stable ] let @xmath7 be an artinian ideal in @xmath6 $ ] generated by monomials whose degrees are bounded above by @xmath22 .
if @xmath64 is non - empty , tileable , and perfectly - punctured , then @xmath169 is stable if and only if every proper monomial subregion of @xmath54 is under - punctured .
we may assume @xmath173 .
as @xmath54 is perfectly - punctured , we have that @xmath174 . in particular , no punctures of @xmath54 overlap .
using corollary [ cor : t - stable ] , we see that @xmath102 is stable if and only if @xmath175 for all proper subsets @xmath81 of the set of minimal generators of @xmath7 .
this is equivalent to every proper monomial subregion of @xmath54 being under - punctured ( see remark [ rem : puncture coeff ] ) . by the preceding theorem and proposition
, we have an understanding of semistability and stability for perfectly - punctured triangular regions .
however , when a region is over - punctured and non - tileable more information is needed to infer semistability .
[ exa : stability ] there are monomial ideals with stable syzygy bundles whose corresponding triangular regions are over - punctured and non - tileable .
see figure [ fig : nss - examples](i ) for a specific example .
+ _ ( i ) @xmath176 _ + _ ( ii ) @xmath177 _ + _ ( iii ) @xmath178 _ moreover , the ideal @xmath179 has a semistable , but non - stable syzygy bundle ( the monomial subregion associated to @xmath180 breaks stability ) , and the ideal @xmath181 has a non - semistable syzygy bundle ( the monomial subregion associated to @xmath182 breaks semistability ) . both of their triangular regions , see figures [ fig : nss - examples](ii ) and ( iii ) , respectively , are over - punctured and non - tileable .
we recall some results that help decide the presence of the weak lefschetz property .
in fact , one needs only check near a `` peak '' of the hilbert function .
* proposition 2.3 ) [ pro : wlp ] let @xmath183 be an artinian standard graded @xmath1-algebra , and let @xmath184 be a general linear form .
suppose @xmath0 has no non - zero socle elements of degree less than @xmath185 for some integer @xmath186
. then @xmath0 has the weak lefschetz property , provided one of the following conditions is satisfied : * @xmath187_{d-2 } \rightarrow [ a]_{d-1}$ ] is injective and @xmath187_{d-1 } \rightarrow [ a]_{d}$ ] is surjective .
* @xmath187_{d-2 } \rightarrow [ a]_{d-1}$ ] is bijective .
moreover , for monomial algebras , it is enough to decide whether the sum of the variables is a lefschetz element .
* proposition 2.2 ) [ pro : mono ] let @xmath188 be a monomial artinian @xmath1-algebra , where @xmath1 is an infinite field . for any integer @xmath22 ,
the following conditions are equivalent : 1 .
the multiplication map @xmath189_{d-1 } \to [ a]_d$ ] has maximal rank , where @xmath190 is a general linear form . 2 .
the multiplication map @xmath191_{d-1 } \to [ a]_d$ ] has maximal rank .
as pointed out above proposition [ prop : pm - det ] , for a monomial ideal @xmath192 $ ] , the bi - adjacency matrix @xmath91 can be described using multiplication by @xmath193 .
we thus get the following criterion for the presence of the weak lefschetz property , where we consider the entries of @xmath91 as elements of the base field @xmath1 .
* corollary 4.7)[cor : wlp - biadj ] let @xmath7 be an artinian monomial ideal in @xmath6 $ ]
. then @xmath5 has the weak lefschetz property if and only if , for each positive integer @xmath22 , the matrix @xmath194 has maximal rank .
this can be used to infer the weak lefschetz property in sufficiently large characteristic from its presence in characteristic zero .
* proposition 7.9)[pro : char-0-to - p ] let @xmath5 be any artinian monomial algebra such that @xmath5 has the weak lefschetz property in characteristic zero .
if @xmath7 contains the powers @xmath195 , then @xmath5 has the weak lefschetz property in positive characteristic whenever @xmath196 .
this section presents an in - depth discussion of artinian monomial ideals of @xmath11 with exactly four minimal generators .
they are called artinian monomial almost complete intersections .
these ideals have been discussed , for example , in @xcite and ( * ? ? ?
* section 6 ) .
in particular , we will answer some of the questions posed in @xcite .
besides addressing the weak lefschetz property , we discuss the splitting types of the syzygy bundles of these ideals .
particular attention is paid if the characteristic is positive .
some of our results are used in @xcite for studying ideals with the rees property . each artinian ideal of @xmath116 $ ] with exactly four monomial minimal generators is of the form @xmath197 where @xmath198 , @xmath199 and @xmath200 , such that at most one of @xmath201 , @xmath202 , and @xmath203 is zero .
if one of @xmath201 , @xmath202 , and @xmath203 is zero , then @xmath204 has type two . in this case
, the presence of the weak lefschetz property has already been described in @xcite .
thus , throughout this section we assume that the integers @xmath205 , and @xmath203 are all positive ; this forces @xmath204 to have cohen - macaulay type three .
more precisely : ( * ? ? ?
* corollary 7.3 ) [ pro : amaci - semistable ] let @xmath206 be defined as above , and suppose @xmath1 is a field of characteristic zero .
. then @xmath7 has a semistable syzygy bundle if and only if the following three conditions are satisfied : [ exa : amaci - nss ] let @xmath216 , and thus @xmath217 .
then the syzygy bundle of @xmath7 is not semistable as @xmath218 .
however , the triangular region @xmath219 is balanced and @xmath220 . hence , @xmath7 does not have the weak lefschetz property if and only if the characteristic of @xmath1 is @xmath221 .
[ exa : amaci - nss-2 ] consider the ideal @xmath222 with five minimal generators
. then theorem [ cor : t - stable ] gives that the syzygy bundle of @xmath81 is not semistable .
notice that @xmath223 is balanced .
however , @xmath224 , and so @xmath225 never has the weak lefschetz property , regardless of the characteristic of @xmath1 .
again , the conclusion of this result may fail in positive characteristic .
indeed , for the ideal @xmath227 in example [ exa : amaci - nss ] we get @xmath228 , but it does not have the weak lefschetz property in characteristic @xmath221 . the following result addresses the weak lefschetz property in the cases that are left out by propositions [ pro : amaci - nss - wlp ] and [ pro : amaci - not-3 ] .
its first part extends ( * ? ? ?
* lemma 7.1 ) from level to arbitrary monomial almost complete intersections .
observe that balanced triangular regions correspond to an equality of the hilbert function in two consecutive degrees , dubbed `` twin - peaks '' in @xcite .
note that condition ( i ) in proposition [ pro : amaci - semistable ] says that @xmath40 has punctures of nonnegative side lengths @xmath231 , and @xmath232 .
furthermore , conditions ( ii ) and ( iii ) therein are equivalent to the fact that the degree of the least common multiple of any two of the minimal generators of @xmath7 is at least @xmath22 , i.e. , the punctures of @xmath40 do not overlap . using the assumption that @xmath22 is an integer
, it follows that @xmath40 is perfectly - punctured , and thus balanced .
since the punctures of @xmath40 do not overlap , the punctures of @xmath233 are not overlapping nor touching .
thus we conclude that the degrees of the socle generators of @xmath5 are at least @xmath185 .
hence , corollary [ cor : wlp - biadj ] and proposition [ pro : wlp ] together give that @xmath5 has the weak lefschetz property if and only if @xmath230 is not zero in @xmath1 .
in the situation of proposition [ pro : amaci - balanced ] , the fact that @xmath5 has the weak lefschetz property implies that @xmath40 is tileable by proposition [ prop : pm - det ] .
tileability remains true even if @xmath5 fails to have the weak lefschetz property . by propositions [ pro : amaci - nss - wlp ] and
[ pro : amaci - not-3 ] , we know that the syzygy bundle of @xmath7 is semistable and @xmath229 is an integer . hence by proposition [ pro : amaci - balanced ] , @xmath40 is perfectly - punctured .
now we conclude by theorem [ thm : tileable - semistable ] .
[ rem : axes - central ] the central puncture is _ axes - central _ if it is ( approximately ) equidistant from a corner puncture and the opposite wall , for each of the three punctures . more specifically , suppose @xmath234 , @xmath235 , @xmath236 , and @xmath237 .
there are two cases to consider : 1 .
if @xmath0 , @xmath238 , and @xmath239 have the same parity , then the region is of the form @xmath240 2 .
if @xmath0 and @xmath238 differ in parity from @xmath239 , then the region is of the form @xmath241 the explicit signed enumerations for these regions can be found in ( * ? ? ?
* theorems 1 , 2 , 4 , & 5 ) .
however , the desired consequence for our use is that the signed enumeration is nonzero if and only if not all of @xmath0 , @xmath238 , and @xmath239 are odd .
moreover , if it is nonzero , then the largest prime divisor of the enumeration is bounded above by @xmath242 .
[ thm : amaci - wlp ] let @xmath243 be an artinian ideal with four minimal generators such that @xmath201 , @xmath202 , and @xmath203 are all positive
. assume the base field @xmath1 has characteristic zero , and consider the following conditions : * if one of the conditions ( i ) - ( iv ) is not satisfied , then @xmath5 has the weak lefschetz property . *
assume all the conditions ( i ) - ( iv ) are satisfied .
then : * * the multiplication map @xmath244_{j-2 } \to [ r / i]_{j-1}$ ] has maximal rank whenever @xmath245 .
* * the algebra @xmath5 has the weak lefschetz property if one of the following conditions is satisfied : * * * condition ( ii ) is an equality . * * * @xmath246 is divisible by 6 . * * * @xmath247 . * * * the region @xmath40 has an axes - central puncture ( see remark [ rem : axes - central ] ) and one of @xmath231 , and @xmath248 is not odd . * * * @xmath249 , @xmath250 , and @xmath47 or @xmath203 is even .
* * the algebra @xmath5 fails to have the weak lefschetz property if one of the following conditions is satisfied : * * * the region @xmath40 has an axes - central puncture ( see remark [ rem : axes - central ] ) and all of @xmath231 , and @xmath248 are odd ; or * * * @xmath249 , @xmath250 , and both @xmath47 and @xmath203 are odd .
the assumptions in ( b ) guarantee that the punctures of @xmath80 do not overlap and the degrees of the socle generators of @xmath5 are at least @xmath185 . then condition ( i )
implies that the puncture to the generator @xmath252 touches another puncture , whereas condition ( ii ) says that this puncture has an even side length . in either case , @xmath5 has the weak lefschetz property by proposition [ prop : same - sign ] .
the condition @xmath247 is equivalent to @xmath253 . after taking into account all lozenges forced by the puncture to @xmath254 ,
the remaining subregion of @xmath10 is a hexagon , and so @xmath255 ( see , e.g. , proposition [ prop : same - sign ] ) . 1 .
theorem [ thm : amaci - wlp ] can be extended to fields of sufficiently positive characteristic by using proposition [ pro : char-0-to - p ] .
this lower bound on the characteristic can be improved whenever one knows the determinant of @xmath194 .
question 8.2(2c ) in @xcite asked if there exist non - level almost complete intersections which never have the weak lefschetz property .
the almost complete intersection @xmath256 is not level and never has the weak lefschetz property , regardless of field characteristic , as @xmath257 . in the previous subsection
, we considered one way of centralizing the inner puncture of a triangular region associated to a monomial almost complete intersection .
we called such punctures `` axes - central . '' in this section , we consider another method of centralizing the inner puncture of such a triangular region .
it turns out this method of centralization is equivalent to the algebra being level .
consider the ideal @xmath206 as above .
let @xmath22 be an integer and assume that @xmath64 has one floating puncture .
we say the inner puncture of @xmath54 is a _ gravity - central puncture _ if the vertices of the puncture are each the same distance from the puncture opposite to it ( see figure [ fig : gravity - central ] ) .
level almost complete intersections were studied extensively in ( * ? ? ?
* sections 6 and 7 ) .
in particular , migliore , mir - roig , and the second author proposed a conjectured characterization for the presence of the weak lefschetz property for such algebras .
we recall it here , though we present it in a different , but equivalent , form to better elucidate the reasoning behind it .
* conjecture 6.8 ) [ conj : level - wlp ] let @xmath259 be an ideal of @xmath6 $ ] , where @xmath1 has characteristic zero , @xmath260 , @xmath261 , and @xmath262 is divisible by three . if @xmath263 is not @xmath264 or @xmath265
, then @xmath5 fails to have the weak lefschetz property if and only if @xmath266 is even , @xmath262 is odd , and @xmath250 or @xmath267 .
furthermore , @xmath5 fails to have the weak lefschetz property in the two exceptional cases .
the necessity part of this conjecture was proven in ( * ? ? ?
* corollary 7.4 ) ) by showing that @xmath5 does not have the weak lefschetz property if @xmath266 is even , @xmath262 is odd , and @xmath250 or @xmath267 .
this result is covered by theorem [ thm : amaci - wlp](b)(3)(v ) because the region is mirror symmetric .
it remained open to establish the presence of the weak lefschetz property .
theorem [ thm : amaci - wlp ] does this in many new cases .
we apply theorem [ thm : amaci - wlp ] with @xmath269 .
then the side length of the inner puncture of @xmath40 is @xmath270 .
hence ( i ) follows from theorem [ thm : amaci - wlp](b)(ii ) .
claim ( ii ) is a consequence of theorem [ thm : amaci - wlp](b)(iv ) as the given condition implies the inner puncture is axes - central .
notice that @xmath274 is simultaneously axis- and gravity - central precisely if either @xmath275 and @xmath268 , or @xmath276 and @xmath277 . in the former case
, the weak lefschetz property in characteristic zero is completely characterized below , strengthening ( * ? ? ?
* corollary 7.6 ) .
assume now @xmath282 .
then @xmath5 fails the weak lefschetz property if @xmath201 and @xmath45 are odd by ( * ? ? ?
* corollary 7.6 ) ( or theorem [ thm : amaci - wlp](b)(3)(v ) ) . otherwise , @xmath5 has this property by proposition [ pro : level - wlp ] . for @xmath282 ,
the triangular region @xmath283 was considered by krattenthaler in @xcite .
he described a bijection between cyclically symmetric lozenge tilings of the region and descending plane partitions with specific conditions .
the generic splitting type of a vector bundle on projective space is an important invariant .
however , its computation is often challenging . in this section
we consider the splitting type of the syzygy bundles of monomial almost complete intersections in @xmath11 .
these are rank three bundles on the projective plane .
for the remainder of this section we assume @xmath1 is an infinite field .
let @xmath206 as above .
recall from section [ sub : syz ] that the syzygy module @xmath100 of @xmath7 is defined by the exact sequence @xmath284 and the syzygy bundle @xmath102 on @xmath285 of @xmath7 is the sheafification of @xmath100 .
its restriction to any line @xmath286 of @xmath285 splits as @xmath287 .
the triple @xmath288 depends on the choice of the line @xmath286 , but is the same for all general lines .
this latter triple is called the _ generic splitting type _ of @xmath102 .
since @xmath7 is a monomial ideal , proposition [ pro : mono ] implies that the generic splitting type @xmath288 can be determined if we restrict to the line defined by @xmath289 . for computing the generic splitting type of @xmath102
, we use the observation that @xmath290 , where @xmath291 $ ] , and @xmath292 . define an @xmath293-module @xmath294 by the exact sequence @xmath295 using the ,
possibly non - minimal , set of generators @xmath296 of @xmath81 .
then @xmath297 , where @xmath288 is the generic splitting type of the vector bundle @xmath102 .
the castelnuovo - mumford regularity of the ideal @xmath81 is @xmath298 . 1 . using , for example , the sequence ( [ eqn : syz - j ] ) , one gets @xmath299 .
if any of the generators of @xmath81 is extraneous , then the degree of that generator is one of @xmath300 , @xmath301 , or @xmath302 .
3 . as the regularity of @xmath81
is determined by the betti numbers of @xmath303 , we obtain that @xmath304 if the sequence ( [ eqn : syz - j ] ) is a minimal free resolution of @xmath81 .
[ lem : reg-2amaci ] let @xmath291 $ ] , where @xmath1 is a field of characteristic zero .
consider the ideal @xmath305 of @xmath293 , and assume that the given generating set is minimal .
then @xmath306 is @xmath307 second , the algebra @xmath310 has the strong lefschetz property in characteristic zero ( see , e.g. , ( * ? ? ?
* proposition 4.4 ) ) .
thus , the hilbert function of @xmath311 is @xmath312_j } = \max\{0 , \dim_k{[s/(x^a , y^b)]_j } - \dim_k{[s/(x^a , y^b)]_{j-\gamma}}\}.\ ] ] by analyzing when the difference becomes non - positive , we get that @xmath313 third , notice that @xmath314 hence , multiplication by @xmath315 induces the short exact sequence @xmath316(-\alpha-\beta ) \stackrel{\times x^\alpha y^\beta}{\longrightarrow } s/{\mathfrak{a}}\rightarrow s/(x^a , y^b , x^\alpha y^\beta ) \rightarrow 0.\ ] ] it implies @xmath317 recall that proposition [ pro : amaci - semistable ] gives a characterization of the semistability of the syzygy bundle @xmath318 , using only the parameters @xmath45 , @xmath46 , @xmath47 , @xmath201 , @xmath202 , and @xmath203 .
we determine the splitting type of @xmath318 for the nonsemistable and the semistable cases separately .
we first consider the case when the syzygy bundle is not semistable , and therein we distinguish four cases .
it turns out that in three cases , at least one of the generators of the ideal @xmath81 is extraneous .
[ pro : st - nss ] consider the ideal @xmath243 with four minimal generators .
assume that the base field @xmath1 has characteristic zero and , without loss of generality , that @xmath319 .
set @xmath320 , and denote by @xmath288 the generic splitting type of @xmath102 .
assume that @xmath102 is not semistable .
then : using @xmath319 , ( * ? ? ?
* theorem 6.3 ) implies that the maximal slope of a subsheaf of @xmath102 is @xmath332 . since @xmath102 is not semistable , we have @xmath333 ( see proposition [ pro : amaci - semistable ] ) .
moreover , the generic splitting type of @xmath102 is determined by the minimal free resolution of @xmath292 as a module over @xmath334 $ ] .
we combine both approaches to determine the generic splitting type . since @xmath335 , all polynomials in @xmath293
whose degree is at least @xmath336 are contained in @xmath337 .
hence , @xmath338 if @xmath321 , and the claim in case ( i ) follows by remark [ rem : splitting - type ] . in case ( ii )
, it follows that @xmath341 and @xmath342 , and thus @xmath343 . using equation ,
we conclude that @xmath344 observe now that @xmath345 is equivalent to @xmath346 .
this implies @xmath347 , and thus @xmath348 . using remark [ rem : splitting - type ] again , we get the generic splitting type of @xmath102 as claimed in ( ii ) .
consider now case ( iii ) .
then @xmath349 , which gives @xmath350 .
the second assumption in this case also implies @xmath351 , which is equivalent to @xmath352 and also to @xmath353 .
similarly , we have that @xmath354 , which is equivalent to @xmath355 and also to @xmath356 .
it follows that @xmath357 hence lemma [ lem : reg-2amaci ] yields @xmath358 this shows that @xmath359 . setting @xmath360 , remark [ rem : splitting - type ] provides the generic splitting type in case ( iii )
. finally consider case ( iv ) .
then @xmath361 , and @xmath362 is equal to the degree of the least common multiple of two of the minimal generators of @xmath7 .
in fact , @xmath363 is the slope of the syzygy bundle @xmath364 of the ideal generated by these two generators .
thus , the harder - narasimhan filtration ( see ( * ? ? ? * definition 1.3.2 ) ) gives an exact sequence @xmath365 where @xmath366 is a semistable torsion - free sheaf on @xmath285 of rank two and first chern class @xmath367 .
its bidual @xmath368 is a stable vector bundle .
thus , by the theorem of grauert and mlich ( see @xcite or ( * ? ? ?
* corollary 1 of theorem 2.1.4 ) ) , its generic splitting type is @xmath369 .
now the claim follows by restricting the above sequence to a general line of @xmath285 .
we have seen that the ideal @xmath292 has at most three minimal generators in the cases ( i ) - ( iii ) of the above proposition . in the fourth case ,
the associated ideal @xmath370 may be minimally generated by four polynomials .
[ exa : st - nss-4mingen ] consider the ideal @xmath371 then the corresponding ideal @xmath81 is minimally generated by @xmath372 , and @xmath373 .
the syzygy bundle of @xmath102 is not semistable , and its generic splitting type is @xmath374 by proposition [ pro : st - nss](iv ) .
order the entries of the generic splitting type @xmath375 of the semistable syzygy bundle @xmath102 such that @xmath376 . in this case
, the splitting type determines the presence of the weak lefschetz property if the characteristic of @xmath1 is zero ( see ( * ? ? ? * theorem 2.2 ) ) .
the following result is slightly more precise .
[ pro : split - type - semist ] let @xmath1 be a field of characteristic zero , and assume the ideal @xmath206 has a semistable syzygy bundle . set @xmath377 . then the generic splitting type of @xmath102 is @xmath378 the grauert - mlich theorem @xcite gives that @xmath379 and @xmath380 are both nonnegative and at most 1 .
moreover , @xmath381 , and @xmath382 satisfy @xmath383 ( see remark [ rem : splitting - type](i ) ) .
this gives the result if @xmath384 .
it remains to consider the case when @xmath385 .
then @xmath386 and @xmath387 are the only possible generic splitting types . by proposition
[ pro : amaci - semistable](i ) , the minimal generators of the ideal @xmath292 have degrees that are less than @xmath22 .
hence @xmath388 if and only if the splitting type of @xmath102 is @xmath389 .
since @xmath94_{d-2 } = \dim_k [ r / i]_{d-1}$ ] , using proposition [ pro : amaci - balanced ] , we conclude that @xmath390 if and only if @xmath5 does not have the weak lefschetz property .
we have just seen that the ideal @xmath292 has regularity @xmath22 if @xmath5 fails the weak lefschetz property .
this implies that the multiplication map @xmath391_{j-2 } \to [ r / i]_{j-1}$ ] is surjective whenever @xmath392 .
moreover , since the minimal generators of @xmath81 have degrees that are less than @xmath22 , we have the exact sequence @xmath393 in the above proof of theorem [ thm : amaci - wlp ] we saw that the four punctures of @xmath10 do not overlap and that @xmath40 is balanced .
hence @xmath394 has 3 more downward - pointing than upward - pointing triangles , that is , @xmath395_{d-2 } = \dim_k [ r / i]_{d-3 } + 3.\ ] ] it follows that the multiplication map in the exact sequence @xmath396_{d-3 } \longrightarrow [ r / i]_{d-2 } \longrightarrow s / j \longrightarrow 0\ ] ] is injective because @xmath397_{d-2 } = 3 $ ] . hence @xmath391_{j-2 } \to [ r / i]_{j-1}$ ] is injective whenever @xmath398 .
the second author would like to thank the authors of @xcite ; it was during a conversation in the preparation of that paper that he learned about the use of the grauert - mlich theorem for an alternative way of deducing the injectivity of the map @xmath20_{d-3 } \longrightarrow [ r / i]_{d-2}$ ] in the above argument if the characteristic of @xmath1 is zero .
[ exa : syzygy ] consider the ideal @xmath399 .
it never has the weak lefschetz property , by theorem [ thm : amaci - wlp](vii ) .
the bundle @xmath400 has generic splitting type @xmath401 .
notice that the similar ideal @xmath402 has the weak lefschetz property in characteristic zero as @xmath403 .
the generic splitting type of @xmath404 is @xmath405 .
we summarize part of our results for the case where @xmath7 is associated to a tileable triangular region .
in particular , if @xmath1 is an infinite field of arbitrary characteristic , then the splitting type can be used to determine the presence of the weak lefschetz property .
[ thm : equiv ] let @xmath406 $ ] , where @xmath1 is an infinite field of arbitrary characteristic .
assume @xmath7 satisfies conditions ( i)(iv ) in theorem [ thm : amaci - wlp ] and @xmath320 is an integer .
then the following conditions are equivalent : 1 .
the algebra @xmath5 has the weak lefschetz property .
the determinant of @xmath194 ( i.e. , the enumeration of signed perfect matchings of the bipartite graph @xmath407 ) is not zero in @xmath1 .
the generic splitting type of @xmath102 is @xmath408 .
regardless of the characteristic of @xmath1 , the arguments for proposition [ pro : amaci - balanced ] show that @xmath40 is balanced .
moreover , the degrees of the socle generators of @xmath5 are at least @xmath185 as shown in theorem [ thm : amaci - wlp](b)(1 ) .
hence , proposition [ pro : wlp ] gives that @xmath5 has the weak lefschetz property if and only if the multiplication map @xmath409_{d-2 } \to [ r / i]_{d-1}\ ] ] is bijective .
now , corollary [ cor : wlp - biadj ] yields the equivalence of conditions ( i ) and ( ii ) . as
above , let @xmath288 be the generic splitting type of @xmath102 , where @xmath376 , and let @xmath370 be the ideal such that @xmath410 .
the above multiplication map is bijective if and only if @xmath411 .
since @xmath412 and @xmath413 , it follows that @xmath411 if and only if @xmath414 .
hence , conditions ( i ) and ( iii ) are equivalent . | deciding the presence of the weak lefschetz property often is a challenging problem .
continuing studies in @xcite , in this work an in - depth study is carried out in the case of artinian monomial ideals with four generators in three variables .
we use a connection to lozenge tilings to describe semistability of the syzygy bundle of such an ideal , to determine its generic splitting type , and to decide the presence of the weak lefschetz property .
we provide results in both characteristic zero and positive characteristic . corresponding author . ] |
life is intimately related to movement on many different time and length scales , from molecular movements to the motility of cells and organisms .
one type of movement which is ubiquitous on the molecular and cellular scale , although not specific to the organic world , is brownian motion or passive diffusion : biomolecules , vesicles , organelles , and other subcellular particles constantly undergo random movements due to thermal fluctuations.@xcite within cells , these random movements depend strongly on the size of the diffusing particles , because the effective viscosity of the cytoplasm increases with increasing particle size.@xcite while proteins typically diffuse through cytoplasm with diffusion coefficients in the range of @xmath0/s to tens of @xmath0/s and therefore explore the volume of a cell within a few minutes to several tens of minutes ( for a typical cell size of a few tens of microns ) , a 100 nm sized organelle typically has a diffusion coefficient of @xmath1/s within the cell,@xcite and would need @xmath2 days to diffuse over the length of the cell . for fast and efficient transport of large cargoes , cells therefore use active transport based on the movements of molecular motors along cytoskeletal filaments.@xcite these molecular motors convert the chemical free energy released from the hydrolysis of atp ( adenosinetriphosphate ) into directed motion and into mechanical work .
they move in a directed stepwise fashion along the linear tracks provided by the cytoskeletal filaments .
there are three large families of cytoskeletal motors , kinesins and dyneins which move along microtubules , and myosins which move along actin filaments .
the filaments have polar structures and encode the direction of motion for the motors . a specific motor steps predominantly in one direction , the forward direction of that motor .
backward steps are usually rare as long as the motor movement is not opposed by a large force .
motor velocities are typically of the order of 1 @xmath3 m/s , which allows a motor - driven cargo to move over typical intracellular distances in a few seconds to a few minutes . on the other hand ,
the force generated by a motor molecule is of the order of a few pn , which is comparable or larger than estimates for the viscous force experienced by typical ( @xmath4 100 nm sized ) motor - driven cargoes in the cytoplasm .
a large part of our present knowledge about the functioning of molecular motors is based on _ in vitro _
experiments which have provided detailed information about the molecular mechanisms of the motors and which have allowed for systematic measurements of their transport properties.@xcite in order to obtain such detailed information , the overwhelming majority of these experiments has addressed the behavior of single motor molecules . within cells ,
however , transport is often accomplished by the cooperation of several motors rather than by a single motor as observed by electron microscopy@xcite and by force measurements@xcite and the analysis of cargo particle trajectories _ in vivo_.@xcite in order to understand the cargo transport in cells , it is therefore necessary to go beyond the single molecule level and to address how several motors act together in a team , in particular in cases where the cooperation of different types of motors is required such as bidirectional cargo transport .
the latter situation , i.e. the presence of different types of motors bound to one cargo particle , is rather common and has been observed for kinesins and dyneins , kinesins and myosins as well as for different members of the kinesin family and even for members of all three motor families.@xcite in this article , we review our recent theoretical analysis@xcite of the cooperation of several motors pulling one cargo .
we emphasize the ability of transport driven by several motors to deal with high viscosities and present an extended discussion of the case where a strong viscous force opposes the movement of the cargo particle .
we also discuss how diffusion can be enhanced by motor - driven active transport and conclude with some remarks on the regulation of active transport .
motors along a cytoskeletal filament . the number of motors which actually pull the cargo changes in a stochastic fashion due to the binding and unbinding of motors to and from the filament . ] to study the cooperation of several molecular motors theoretically , we have recently introduced a model which describes the stochastic binding and unbinding of motors and filaments as well as the movements of the cargo particle to which these motors are attached.@xcite the state of the cargo particle is described by the number @xmath5 of motors bound to the filament . as shown in fig .
[ f1 ] , this number changes stochastically between 0 and @xmath6 , the total number of motors bound to the cargo , since motors bind to and unbind from the filament.@xcite the model is therefore defined by a set of rates @xmath7 and @xmath8 which describe the unbinding and binding of a motor , respectively , and which depend on the number @xmath5 of bound motors , and by a set of velocities @xmath9 with which the cargo particle moves when pulled by @xmath5 motors . in the simplest case , the motors bind to and unbind from the filament in a fashion independent of each other .
in that case , the binding and unbinding rates are given by @xmath10 with the single motor unbinding and binding rates @xmath11 and @xmath12 , respectively .
for non - interacting motors , the cargo velocity is independent of the number of pulling motors and given by the single motor velocity , @xmath13 , as shown both by microtubule gliding assays and by bead assays for kinesin motors.@xcite for this case we have obtained a number of analytical results.@xcite in particular , the model indicates a strong increase of the average run length , i.e. , the distance a cargo particle moves along a filament before it unbinds from it . for motors which bind strongly to the filament , so that @xmath14 , the average run length is given by @xmath15 and essentially increases exponentially with increasing number of motors . using the single molecule parameters for conventional kinesin ( kinesin 1 ) , we have estimated that run lengths in the centimeter range are obtained if cargoes are pulled by 78 motors.@xcite as these long run lengths exceed the length of a microtubule ( typically a few tens of microns ) , they can however only be realized if microtubules are aligned in a parallel and isopolar fashion and if cargoes can step from one microtubule to another as observed _ in vitro _ using aligned microtubules.@xcite the increase of cargo run lengths with increasing number of motors has been observed in several _ in vitro _ experiments,@xcite
however it has been difficult to determine the number of motors pulling the cargo .
one method to determine the motor number is to use a combination of dynamic light scattering measurements and comparison of measured run length distributions with theoretical predictions.@xcite if the cargo is pulled against an opposing force @xmath16 , this force is shared among the bound motors , so that each bound motor experiences the force @xmath17 . under the influence of an external force ,
the single motor velocity decreases approximately linearly , @xmath18 , and the unbinding rate increases exponentially , @xmath19 as obtained from optical tweezers experiments.@xcite the two force scales are the stall force @xmath20 and the detachment force @xmath21 .
for a cargo pulled by several motors , the velocities and unbinding rates in the different binding states are then given by @xmath22 since the velocity now depends on the number of bound motors , the velocity of the cargo changes every time a motor unbinds or an additional motor binds to the filament .
the trajectory of the cargo therefore consists of linear segments with constant velocity , and the distribution of the instantaneous velocities has several peaks which become more and more distinct if the force @xmath16 is increased .
in addition , the sharing of the force induces a coupling between the motors which leads to cascades of unbinding events , since the unbinding of one motor increases the force and , thus , the unbinding rate for the remaining bound motors .
such unbinding cascades occur also in many other biophysical systems which have a similar unbinding dynamics , in particular they have been studied extensively for the forced unbinding of clusters of adhesion molecules.@xcite for the motors , the most important consequence of this type of force - induced coupling of the motors is that an increase in force not only slows down the motors , but also decreases the number of bound motors .
therefore , the force - velocity relation given by the average velocity as a function of the load force is a nonlinear relation for cargoes pulled by several motors , although it is approximately linear for a single motor.@xcite rather than being imposed by an optical laser trap or other force fields that can be directly controlled _ in vitro _ , an opposing force can also arise from other motors which pull the cargo into the opposite direction .
the presence of two types of motors which move into opposite directions bound to the same cargo is commonly found in cells and is required for bidirectional transport in essentially unidirectional systems of filaments as they are typical for the microtubule cytoskeleton . in general , the two types of motors interact both mechanically by pulling on each other _ and _ via biochemical signals or regulatory molecules .
if there are only mechanical interactions our model predicts a tug - of - war - like instability : if the motors pull on each other sufficiently strongly , one species will win , and the cargo performs fast directed motion rather than being stalled by the pulling of motors in both directions . since the number of motors pulling the cargo is typically small , the direction of motion will however be reversed from time to time with a reversal frequency which decreases as the motor numbers are increased .
one universal force that is always experienced by molecular motors is the viscous drag caused by the medium through which the cargo is pulled . in water or aqueous solutions , however , the viscous drag of the cargo is usually negligible since it corresponds to a force of only a small fraction of the motor stall force .
for example , a bead with diameter 1 @xmath3 m which moves at 1 @xmath3 m/s through water experiences a viscous force of 0.02 pn which is tiny compared to a motor stall force of a few pn .
therefore , _ in vitro _ experiments are hardly affected by the viscosity of the solution , and changes in motor number do not lead to a change of the cargo velocity unless the viscosity is increased to @xmath23 times that of water.@xcite in highly viscous environments , this is different : if the viscous drag force is of the same order of magnitude as the single motor stall force , the velocity can be increased if the number of motors which share this force is increased . the latter effect has been observed in microtubule gliding assays with high solution viscosity where for low motor density on the surface the velocity decreases as a function of the microtubule length , while for high motor density the velocity is independent of the microtubule length.@xcite for a cargo pulled by @xmath5 motors , inserting the stokes friction force @xmath24 ( with the friction coefficient @xmath25 ) into the linear force velocity relation leads to@xcite @xmath26 this equation shows that the velocity increases with increasing number of motors if @xmath27 is not negligibly small compared to one . in particular , in the limit of high viscosity or large @xmath28 , for which the last approximation in eq .
( [ eq : vn_gamma ] ) is valid , the velocity is proportional to the number @xmath5 of pulling motors .. as a consequence of this , an increase of @xmath5 to motor numbers large compared to @xmath29 will increase the consumption of atp without substantially increasing the cargo velocity . ] in a highly viscous environment , the cargo s velocity distribution therefore exhibits maxima at integer multiples of a minimal velocity .
similar velocity distributions have recently been observed for vesicles and melanosomes in the cytoplasm , see refs . .
to first order in @xmath30 the motors experience the force @xmath31 which implies that the force per bound motor is independent of the number of bound motors and that the motors behave as independent motors for large viscous force , however with an increased effective single motor unbinding rate @xmath32 .
we have emphasized that large particles experience a strong viscous drag in the cytoplasm and that therefore brownian motion is too slow to drive transport of large particles in the cell .
while this observation suggests that active transport is necessary within cells , it does not imply that the active transport must necessarily be directed transport .
alternatively , active transport could also be used to generate effectively diffusive motion , which is faster than passive brownian motion , e.g. if a cargo particle performs a sequence of active molecular motor - driven runs in random direction .
we call this effectively diffusive motion , which depends on chemical energy , active diffusion.@xcite in cells , active diffusion can be achieved either by ( i ) switching the direction of motion by switching between different types of motors which walk along a unipolar array of filaments or by ( ii ) a single type of motor and isotropic ( e.g. , bidirectional or random ) arrangements of filaments .
the first case is typical for microtubule - based transport : microtubules are often arranged in a directed fashion , either in radial systems emanating from a central microtubule organizing center with their plus ends pointing outwards or in unidirectional systems where microtubules are aligned in a parallel and isopolar fashion such as in axons .
bidirectional movements along these unidirectional microtubule arrangements have been observed for a large variety of intracellular cargoes.@xcite these movements are driven by a combination of plus end and minus end directed motors . on the other hand , actin - based movements are often of the second type , since the actin cytoskeleton usually forms an isotropic random mesh , on which , e.g. , myosin v - driven cargoes perform random walks.@xcite active diffusion has also been observed for random arrays of microtubules in cell extracts.@xcite let us mention that these two types of active diffusion are highly simplified .
more complex scenarios include bidirectional , but biased movements along microtubules and the switching of cargoes between microtubules and actin filaments.@xcite in vitro , one can use various techniques such as chemically or topographically structured surfaces,@xcite motor filament self - organization,@xcite and filament crosslinking on micropillars@xcite to create well - defined patterns of filaments for active diffusion which may be useful to enhance diffusion in bio - nanotechnological transport systems.@xcite the maximal effective diffusion coefficient which can be achieved by molecular motor - driven active diffusion is given by @xmath33 where @xmath34 is the length of essentially unidirectional runs , given by either the average run length before unbinding from filaments or the mesh size of the filament pattern , and @xmath35 is the probability that the cargo is bound to a filament.@xcite in order to obtain large effective diffusion one therefore has to make sure that the cargo particle is bound to filaments most of the time , and that it has an average run length which is comparable or larger than the pattern mesh size .
one possibility to satisfy both conditions is to use a sufficiently large number of motors .
a larger number of motors also decreases the probability of switching direction at an intersection,@xcite so that unidirectional runs exceeding the mesh size can be achieved .
since the motor velocity is rather insensitive to the viscosity for small viscosities , active diffusion is much less affected by the viscous drag of the solution than passive brownian motion,@xcite in particular if a cargo is pulled by several motors . for a cargo of size 100 nm , brownian motion in water
is characterized by a diffusion coefficient in the order of a few @xmath3m@xmath36/s .
an increase in viscosity by a factor of 10 leads to a decrease of the diffusion coefficient by that factor .
the active diffusion coefficient , on the other hand can be estimated to be of similar size or slightly smaller ( using @xmath37 m/s , @xmath3810
@xmath3 m and @xmath39 ) , however the latter value is essentially unaffected by an increase of the viscosity by up to a factor 100 compared to the viscosity of water , since the viscous force which arises from the movement of such a bead is only a fraction of @xmath40 of the motor stall force .
the effect is increased if a cargo is pulled by several motors , since the viscous force @xmath28 starts to affect the motor movement only if it is of the order @xmath41 .
therefore only viscosities which lead to viscous forces that exceed @xmath42 have a considerable effect on active diffusion .
in the preceding sections we have discussed how cells achieve cargo transport over large distances through a viscous environment by the cooperation of a small number of molecular motors . in principle , cells could also use a single motor which generates a larger force and binds more strongly to the corresponding filament rather than a team of motors , but motor cooperation appears to be preferred .
the use of several motors has the advantage that the transport parameters can be easily controlled by controlling the number of pulling motors , e.g. by activating motors , by increasing their binding to filaments or by recruiting additional motors to the cargo .
the use of multiple weak bonds rather than a single strong bond in order to enable simple ways of control appears to be a general principle in cellular biology which applies to various cellular processes as diverse as the build - up of strong , but at the same time highly dynamic clusters of adhesion molecules@xcite and the binding of transcription factors to dna where programmable specificity of binding is achieved by sequence - dependent binding of the transcription factor to a short stretch of dna.@xcite if it is true that the main purpose of motor cooperation is the controllability of motor - driven movements , one may ask whether this function imposes constraints on the properties of the motors . for example , in order to both up- and down - regulate the cargo velocity or run length , it is clearly desirable to be able to both increase and decrease the number of pulling motors and , thus , to have an average number , @xmath43 , of pulling motors that is not too close to either 1 or @xmath6 , the total number of motors , but rather is of the order of @xmath44 . itself may be fixed by the tradeoff between increased cargo velocity and efficient atp usage , such that @xmath45 . ] the average number of pulling motors can be estimated@xcite by @xmath46 $ ] , so that the requirement @xmath47 implies @xmath48 or , for cargoes subject to a strong viscous force , @xmath49 .
the latter condition represents a relation between the single motor parameters possibly imposed by a functional constraint related to motor cooperation .
the parameters of conventional kinesin approximately satisfy this relation , which could however be coincidental
. it would therefore be interesting to see whether this relation is also satisfied by the parameters of other motors .
if this is not the case , these differences might provide hints towards functional differences between different types of motors .
cargo transport in cells is often carried out by small teams of molecular motors rather than by single motor molecules .
we have described how one can analyze the transport by several motors using a phenomenological model based on the understanding of single motors that has been obtained from single molecule experiments during the last decade .
our theoretical approach is sufficiently versatile to be extended to more complex situations where additional molecular species are present such as the cooperation of two types of motors or the interaction of motors with regulatory proteins . in particular
, it will be interesting to use this model to study control mechanisms of motor - driven transport .
sk acknowledges support by deutsche forschungsgemeinschaft ( kl 818/1 - 1 ) . | molecular motors power directed transport of cargoes within cells . even if a single motor is sufficient to transport a cargo , motors often cooperate in small teams .
we discuss the cooperative cargo transport by several motors theoretically and explore some of its properties .
in particular we emphasize how motor teams can drag cargoes through a viscous environment . |
ethanol and methanol strongly self - associate due to hydrogen bonding @xcite . these primary polar liquids with c - oh group form very similar hydrogen bond acceptors and hydrogen bond donors and they exhibit extended network @xcite .
on mixing , these two liquids does not form azeotropes .
also , methanol does not form azeotrope with water , whereas ethanol forms an azeotrope with it .
although the binary system of ethanol and methanol is assumed to be ideal , their behavior at low concentrations is far from ideal @xcite . +
vast experimental information is available concerning the dielectric properties , excess volume parameters , ftir spectra and refractive index of binary system of ethanol - water @xcite , methanol - water @xcite , ethanol - higher alcohols @xcite , methanol - higher alcohols @xcite , ethanol - pyrindine @xcite , methanol - pyridine @xcite , etc .
all these systems show a deviation from ideal behaviour for most physical and chemical properties at @xmath1 . on the other hand ,
comparatively very few reports are available for the ethanol - methanol binary system .
al . , found the non - ideal behavior of ethanol - methanol at low concentrations while investigating the activity measurements of ethanol - methanol - acetone system @xcite .
the current study is undertaken to understand the anomalous behavior of intermolecular interactions of ethanol - methanol system .
dielectric , ftir , surface tension , density , refractive index and molecular dynamics studies for this system is undertaken over the complete concentration range , with emphasis on concentration regions where anomalous behavior in properties is seen .
the excess values of molar volume and refractive index , molar refraction and total partial pressure were determined using the experimental values .
complex impedance cole - cole plots were used to study the relaxation mechanism in the ethanol - methanol system . +
molecular dynamics simulations were done to verify the behavior of ethanol - methanol binary system .
the structure of molecules and the number of hydrogen - bonded molecules determined from the simulations are used to interpret the the experimental results of surface tension , density , refractive index and dielectric values relating hydrogen bond structure .
it is understood from the present study that the methanol molecules are responsible for increase in hydrogen bond density and they play the role of a mediator in connecting ethanol molecules .
methanol and ethanol were purchased from sigma aldrich and merck emplura respectively .
the purities of methanol and ethanol were @xmath2 .
the precision of the binary system was @xmath3 and they were measured using a weighing balance .
density of the binary systems were measured by using a @xmath4 specific gravity bottle , calibrated with distilled water .
surface tension of the binary systems over the whole concentration range were measured using rame - hart contact angle goniometer .
pendant drop method was used in calculating the surface tension .
the polar and dispersive parts of surface tension were calculated using the contact angle datas .
the mid infra - red spectra of the binary systems were measured by fourier transform infra red spectrometer of perkin elmer .
broadband dielectric studies were carried out using vector network analyzer with dielectric assessment kit of rhode and scharwz .
the refractive index measurements with na light were carried out for the binary system using abbe refractometer which was calibrated using doubly distilled water .
all the experiments were performed at room temperature .
the well - known structures of methanol and ethanol were first constructed using the avogadro package @xcite .
the saved input files were optimized using the gamess @xcite package , using an ab - initio hartree - fock @xcite with a @xmath5 basis set @xcite .
the simulations were carried out using gromacs @xcite .
visualizations of the simulations were obtained using the vmd software @xcite .
+ methanol and ethanol mixtures from @xmath6 to @xmath7 proportion of methanol were produced .
further details of the solution concentrations are given in table .
all the samples were subjected to energy minimization procedures , using the steepest descent algorithm @xcite , with an energy tolerance of 500 kj / mol .
any outlying molecules were manually replaced , assuring that the concentration remain unchanged .
+ following the energy minimization , the system was subjected to two equilibration processes .
first is an nvt equilibration in which the thermostat algorithm is used to set the temperature right .
this thermally equilibrated system is then subject to an npt equilibration wherein the system is compressed to obtain the correct density .
this is done using a barostat @xcite and a thermostat algorithm @xcite .
we can now conclude that the system is very much comparable to a laboratory case .
methanol is a polar molecule capable of h - bonding both with itself and with other oxygen or nitrogen containing molecules , like water , thus making it completely soluble in water .
however , with a single methyl group , methanol has only weak london dispersion forces with itself and with other molecules @xcite .
+ with an increase in the alkyl chain , ethanol has an increase in the dispersion forces which is comparably greater than the stronger dipole - dipole and hydrogen bonding intermolecular forces in alcohols @xcite . as a result , methanol is insoluble in alkanes , such as hexane and heptane , whereas ethanol and propanol are completely miscible with these alkanes .
+ on the addition of the two liquids , there will be an expansion or contraction in volume of the mixture @xcite .
the volume expansion occurs when the geometry of molecular structure is such that it does not fit one component of molecules into the other . the steric hindrance and the loss of dipolar association can also lead to expansion in the volume of the mixture .
volume contraction occurs when there is an increase in chemical interaction between the constituent molecules and if the geometry of molecular structure favors the fitting of component molecules with each other , volume of the mixture decreases .
the contraction shows that the molecules of one component are accommodated into the interstitials of the other component@xcite .
+ on addition of methanol to ethanol , the hydrogen bonds in ethanol aggregates are disrupted by the methanol molecules .
subsequently when the hydrogen bond strength between ethanol molecules overcomes the dispersive interaction between ethanol and methanol , methanol molecules are encaged by ethanol . on further addition of methanol ,
bridged structures of ethanol - methanol are formed .
+ the experimental values of surface tension , density and refractive index plotted against the concentration of methanol are shown in figure [ 1 ] . in the figure , region a , b and c indicates @xmath8 , @xmath9 to @xmath10 and @xmath11 of methanol in ethanol respectively . usually with the increase in hydrogen bond density ,
the density of the system will increase .
the surface tension is a measure of strength of intermolecular forces and in region a , the surface tension shows a slight increase with an addition of methanol . in region
a , density @xmath12 is seen to drop dramatically with an addition of small amount of methanol to ethanol .
the decrease in density indicates the decrease in magnitude of volume contraction on mixing of two liquids .
it is also clear that , with the addition of methanol , the dipolar association of ethanol molecules start to break . + on further addition of methanol , the system shows a complex variation in density and surface tension at the region b. the observed increase / decrease in density at this region indicates the increase / decrease in magnitude of contraction on mixing of these liquids . in particular , a dramatic increase in contraction is observed at @xmath13 of methanol concentration .
this indicates that at @xmath13 of methanol , the interaction between ethanol and methanol molecules is large and this leads to accommodation of methanol molecules within ethanol aggregates . at @xmath14 ,
the density drops to 0.70 g / ml indicating the decrease in magnitude of contraction that results from the break up of interaction between molecules by the rupture of hydrogen bonded chains and loosening of dipole interactions . as a whole , in the region
b , dipole - dipole interactions and hydrogen bonding that exists in pure ethanol and pure methanol decrease and the intermolecular interactions between ethanol and methanol increase .
this is also confirmed from the surface tension measurements .
+ above @xmath15 of methanol ( region c ) , the density and surface tension is constant indicating the equilibrium attained by the system .
it is also clear that with the increase in methanol concentration , the hydrogen bond density increases .
+ there is an non - ideal decrease in refractive index of the binary system from 1.361 to 1.327 with decrease in ethanol concentration . from the figure 1 , a decrease in refractive index
is observed at region a and c that indicates that the polarization due to electrons is reduced with an addition of methanol to ethanol .
+ in the region b ( concentration range of @xmath9 to @xmath15 ) , an interesting behavior of refractive index is observed .
it is observed that the refractive index remains unchanged at @xmath16 to @xmath17 , @xmath18 to @xmath13 and @xmath14 to @xmath19 of methanol in ethanol .
this is an indication of clathrate formation ie . ,
the formation of open structures containing cages .
the cages are generated due to the hydrogen bonding between the molecules of solute being added . here
the clathrate formation is mainly attributed to the hydrogen bonding between methanol and ethanol molecules .
this also shows that electronic polarization is stable at these concentrations .
+ on examining the static dielectric permittivity of the binary system , there is an increase in permittivity with respect to the concentration of methanol as shown in figure [ 3 ] . in region
a , the binary system does nt show any variation in permittivity . on
further increasing methanol concentration , the static permittivity at region b shows a gradual increase in permittivity till @xmath18 , a fall at @xmath13 followed by a sharp rise at @xmath14 of methanol .
there is no uniformity in the dielectric values unlike in refractive index .
this change clearly indicates the formation of temporarily induced dipoles and the increase in corresponding dispersion interaction between the ethanol - methanol molecules at region b. this also confirms the breaking of hydrogen bond network of like molecules and formation of hydrogen bonded ethanol - methanol bridged structures which is also reflected in the ftir spectra . at region
c , the dielectric permittivity is observed to increase with methanol concentration indicating the formation of dipoles .
+ the dielectric loss is seen to drop at region b ( @xmath9 to @xmath15 of methanol ) , which confirms the polarization due to temporarily induced dipoles .
the dielectric loss for all concentrations are shown in figure [ 3 ] .
the dielectric permittivity and dielectric loss of the binary system for the frequency range @xmath20 to @xmath21 are shown in figure [ 4 ] .
+ the further analysis of measured complex permittivity is performed using the curve fitting technique based on the cole - cole model .
this is done to find the relaxation time .
the cole - cole plots of all the concentrations are performed and are shown in figure [ 5 ] and figure [ 6 ] . at high frequencies above @xmath22 , the cole - cole plot of all concentration of methanol is located on the same line .
this implies that at all mole fractions of methanol in ethanol , the relaxation mechanism is same in this frequency region .
permittivity being on the same line at this frequency region indicates that all cole - cole plots must have have the same limiting high frequency , @xmath23 . in this region , the relaxation laws are the same and the same dipole moment is responsible for the observations between @xmath22 to @xmath24 .
both ethanol and methanol contain oh group and the dielectric signal observed in this high frequency region is attributed to the dipole moment of the oh groups .
+ at low frequencies , the cole - cole plots have the same shape of varying magnitudes .
the relaxation at this low frequency region is associated with the instantaneous dipoles formed and its re - orientation among the like molecules .
deviation of some points from cole - cole plots is attributed to the presence of interacting species in various concentrations .
+ the hydrogen bonds in methanol are stronger than in ethanol whereas in ethanol , dispersion forces dominate over hydrogen bonds .
so with the addition of methanol to ethanol , weakening of dispersion forces take place .
the breaking up of hydrogen bond network of ethanol and formation of ethanol - methanol hydrogen bonds are seen as alternate red and blue shifts in the oh stretching peak of the infra red spectra .
the red shift of @xmath25 in oh stretching frequency at @xmath26 of methanol manifests the strength of formation of hydrogen bonds . the ir spectrum with respect to the molar fraction of methanol is shown in figure [ 2 ] .
the hydrogen bonds start to form between unlike molecules and get stabilized after @xmath15 of methanol concentration in ethanol . beyond @xmath15 , as expected , the number of hydrogen bonds between ethanol and methanol molecules increase and is shown with a red shift in the transmittance spectra .
+ on deconvultion of the oh peak , three peaks of frequency @xmath27 , @xmath28 and @xmath29 corresponding to multimers , open chain tetramers and cyclic multimers are found . the deconvoluted spectrum for the concentration of @xmath14 of methanol is shown in figure [ 6 ] .
+ mole fraction of methanol.,width=480 ] the deconvulted oh peak positions are shown in figure [ 7 ] . from the figure
it is observed that the multimer peak shows alternate red and blue shifts .
this confirms the instability of intermolecular interactions with the increase in concentration of methanol .
this also confirms the formation of long chain multimers composed of ethanol - methanol and ethanol molecules .
the position of the peak corresponding to the cyclic multimers seem to be at the same frequency across the concentration range of @xmath9 to @xmath15 of methanol in ethanol(region b ) .
this attributes to clathrate formation as seen from the refractive index measurements .
the comparison of the area under the three deconvoluted peaks is shown in the figure [ 8 ] .
it is observed from the figure that , the formation of cyclic multimers are more compared to tetramers and pentamers .
+ to further confirm the role of dispersion forces in the binary system , the excess values of properties are calculated and found that this binary system deviates from its ideal behavior [ figure [ 9 ] ] . from the excess molar volume plot , it is clear that volume increases for the concentration of @xmath9 to @xmath15 .
this confirms that the strength of intermolecular forces decreases .
this is attributed to the formation of multimer structures through dispersion forces between the ethanol molecules , along with a hydrogen bond formed between ethanol and methanol .
+ this dominant role of dispersive forces is further confirmed by measuring the polar and dispersive parts of surface tension using the contact angle values .
figure [ 10 ] shows the variations of polar and dispersive parts of surface tension with the mole fraction of methanol .
it is observed from the results that the strength of intermolecular forces decreases with increase in mole fraction of methanol .
it is also found that , the dispersive forces of ethanol decreases and hydrogen bond density increases with an addition of methanol to ethanol .
it is attributed that methanol molecules are responsible for the hydrogen bonding since ethanol clusters are not strictly hydrogen bonded .
+ molecular dynamics simulations were performed to study the nature of interaction between ethanol and methanol molecules of different concentrations , to compare with the experimental results .
the simulations were equilibrated such that they resemble the ambient laboratory conditions . + ethanol aggregates ( multimers ) , ethanol - methanol joint structures , ethanol - methanol - ethanol bridges , ethanol monomers and methanol monomers
were commonly observed structures at the various concentrations ( figure [ 12 ] ) .
+ clathrates of various sizes typically dimers to pentamers are observed ( figure [ 12 ] ) . as the concentration of methanol increases ,
the number and size of these aggregates reduces and is shown in figure [ 11 ] .
methanol molecules , with their stronger hydrogen bonding nature insert themselves among the ethanol molecules , sterically hindering ethanol from forming aggregates .
this is seen experimentally in figure [ 1 ] as the regions of constant refractive index .
the methanol molecules also act as a bridge between ethanol molecules forming ethanol - methanol - ethanol structures . at concentrations close to @xmath1 ,
ethanol - methanol joint aggregates are also formed in good numbers . at higher concentrations of methanol
, ethanol molecules mainly have monomeric forms , while ethanol - methanol joint aggregates became less in number as well .
+ concentration of methanol in ethanol as shown here , there are fewer number of aggregated structures or clathrates.,width=480 ] concentration of methanol in ethanol b ) the ethanol closed cage structure ( clathrate ) obtained at @xmath30 concentration of methanol in ethanol c ) pentameric open cage structure of ethanol found commonly at several lower concentrations @xmath31 d ) a firmly hydrogen bond bridged methanol molecule , holding onto ethanol molecules e ) a very common trimeric form of ethanol , width=480 ] methanol molecules , however , do not form aggregates among themselves . a graph plotted between hydrogen bonds per volume and percentage concentration of methanol ( figure
[ 13 ] ) shows that there is a deviation from the expected trend at around @xmath32 concentration .
the graph shows that the hydrogen bond density increases with the increase in concentration of methanol . as per the refractive index and surface tension studies ,
this is the concentration range where complex properties are observed .
the formation of these aggregates are proposed to be the reason for these properties .
+ table 1 provides the output of molecular dynamics simulations . from the graph
it is seen that in the concentration region between @xmath9 and @xmath32 , the slope of the graph is much lesser than that at higher concentration regions .
in fact in this concentration region the hydrogen bond density increases only by @xmath33 while at higher concentrations , there are larger changes over a @xmath9 change in concentration .
it is clearly understood from this data that there is a direct correlation between the number of molecules of methanol and the h - bond density .
the decreased hydrogen bond density at @xmath34 concentration and similar hydrogen bond density at @xmath26 and @xmath18 are also noted from the table .
so , the hydrogen bond density , as per the graph , is a function of the number of methanol molecules present , as the ethanol aggregates are not strictly hydrogen - bonded .
this is in complete agreement with the experimental results .
+ .variation of hydrogen bond density with the concentration of methanol [ cols="<,^,^,^,^,^,^,^ " , ]
the surface tension , density , refractive index and dielectric permittivity have been reported for the ethanol - methanol binary system .
the results show distinct type of interactions leading to deviation in properties at @xmath9 to @xmath15 of methanol .
complex variation in density and surface tension values is observed at @xmath9 to @xmath15 which is attributed to decrease in dipole -dipole interactions and hydrogen bonding that exists in pure ethanol and pure methanol and the increase in intermolecular interaction between ethanol and methanol .
the ftir results confirm the presence of multimers and tetramers rather than dimers and monomers .
refractive index measurements show the formation of clathrates at @xmath16 to @xmath17 and @xmath14 to @xmath19 of methanol .
the variation in dielectric values at these concentrations clearly indicates the formation of temporarily induced dipoles .
excess parameters confirmed the domination of dispersion forces at low concentrations in the ethanol - methanol binary system .
molecular dynamics simulations were performed to compare with the experimental results .
it was found from the simulations that the ethanol aggregates are not strictly hydrogen - bond constructed .
it can be concluded that methanol involves extensive h - bonding and with increase in methanol concentration , the hydrogen bond density also increases , which agree with the experimental results .
the authors thank naval research board of india for the contact angle goniometer .
caibin xiao , hugo bianchi , and peter r tremaine,``excess molar volumes and densities of methanol - water at temperatures between 212 k and 462 k and pressures of 7 mpa and 13.5 mpa'',j .
thermodynamics , 1997 , 29 , 261 - 286 .
jose j , cano - gomez , gustavo a , iglesias - silva , edgar o , castrejon - gonzalez , mariana ramos - estrada , and kenneth r. hall , `` density and visocsity of binary liquid mixtures of ethanol and 1-hexanol and ethanol+1-heptanol from 293.15 to 328.15 k at 0.1 mpa '' , j. chem .
eng . data .
2015 , 60 , 1945 - 1955 ezekiel d. dikio1 , simphiwe m. nelana , david a. isabirye , eno e. ebenso , `` density , dynamic viscosity and derived properties of binary mixtures of methanol , ethanol , n - propanol , and n - butanol with pyridine at t @xmath35 ( 293.15 , 303.15 , 313.15 and 323.15 ) k '' , int .
j. electrochem .
, 7 ( 2012 ) 11101 - 11122 marcus d hanwell , donald e curtis , david c lonie , tim vandermeersch , eva zurek and geoffrey r hutchison ; `` avogadro : an advanced semantic chemical editor , visualization , and analysis platform '' journal of cheminformatics 2012 , 4:17 m.w.schmidt , k.k.baldridge , j.a.boatz , s.t.elbert , m.s.gordon , j.h.jensen , s.koseki , n.matsunaga , k.a.nguyen , s.su , t.l.windus , m.dupuis , j.a.montgomery , general atomic and molecular electronic structure system , j. comput .
, 14 , 1347 - 1363(1993 ) m. m. francl , w. j. pietro , w. j. hehre , j.s .
binkley , m.s .
gordon , d.j .
de frees , j. a. polpe , self - consistent molecular orbital methods .
xxiii , a polarization - type basis set for second - row elements , j. chem .
77 , 3654 - 3665 ( 1982 ) bekker , h. , berendsen , h. j. c. , dijkstra , e. j. , achterop , s. , van drunen , r. , van der spoel , d. , sijbers , a. , keegstra , h. , reitsma , b. , renardus , m. k. r. gromacs : a parallel computer for molecular dynamics simulations . in physics computing 92 ( singapore , 1993 ) .
de groot , r. a. , nadrchal , j. , eds . .
world scientific .
pronk , s. , pall , s. , schulz , r. , larsson , p. , bjelkmar , p. , apostolov , r. , shirts , m. r. , smith , j. c. , kasson , p. m. , van der spoel , d. , hess , b. , lindahl , e. gromacs 4.5 : a high - throughput and highly parallel open source molecular simulation toolkit .
bioinformatics 29(7 ) : 845 - 854 , 2013 . pall , s. , abraham , m. j. , kutzner , c. , hess , b. , lindahl , e. tackling exascale software challenges in molecular dynamics simulati ons with gromacs . in : solving software challenges for exascale . vol .
markidis , s. , laure , e. eds .
vol.8759 . .
springer international publishing switzerland london 2015 3 - 27 .
abraham , m. j. , t. , schulz , r. , r. , pall , s. , smith , j. c. , hess , b. , lindahl , e. gromacs : high performance molecular simulations through multi - level parallelism from laptops to supercomputers .
softwarex 1 - 2 , 19 - 25 , 2015 .
saravanakumar k and kubendran t.r .
, density and viscosities for the binary mixtures of 1 , 4-dioxane and benzene or chlorobenzene at 303.15 , 308.15 , 313.15 k and a pressure of 0.1mpa , research journal of chemical sciences , vol .
2(4 ) , 50 - 56 , april ( 2012 ) | at low concentrations of methanol in ethanol - methanol binary system , the molecular interactions are seen to be uniquely complex . it is observed that the ethanol aggregates are not strictly hydrogen - bonded complexes ; dispersion forces also play a dominant role in the self - association of ethanol molecules . on the addition of small amount of methanol to ethanol
, the dipolar association of ethanol is destroyed .
the repulsive forces between the two moieties dominate the behavior of the binary system at lower concentration of methanol . at higher concentration of methanol @xmath0 , the strength and extent ( number ) of formation of hydrogen bonds between ethanol and methanol increases .
the geometry of molecular structure at high concentration favors the fitting of component molecules with each other .
intermolecular interactions in the ethanol - methanol binary system over the entire concentration range were investigated in detail using broadband dielectric spectroscopy , ftir , surface tension and refractive index studies .
molecular dynamics simulations show that the hydrogen bond density is a direct function of the number of methanol molecules present , as the ethanol aggregates are not strictly hydrogen - bond constructed which is in agreement with the experimental results . |
a point charge with a constant acceleration @xmath0 in its instantaneous rest system follows the relativistic trajectory@xcite @xmath1 the equation relating @xmath2 and @xmath3 is a hyperbola , leading to the common designation of this motion as ` hyperbolic motion ' . the trajectory in eq .
( [ zt ] ) corresponds to a charge that comes to rest at @xmath4 at time @xmath5 after traveling an infinite distance from the infinite past where its speed @xmath61 .
the electric field produced by the accelerating charge has been calculated in ref .
the field lines show an unusual , non - physical behavior at @xmath7 ( for observation time t=0 ) where they abruptly stop .
this is because the retarded position of the charge is outside the past light cone for any @xmath8 .
this problem is recognized in @xcite , and they attempt to correct it by introducing truncated hyperbolic motion " , in which the accelerated part of the motion is preceded by motion with a constant velocity . however , their results in figures 3 and 4 of their paper show an unphysical discontinuous behavior for the electric field lines as the charge passes through the transition from constant velocity to accelerated motion . in this paper , we modify the calculation of the electric field lines , resulting in continuous electric field lines that satisfy the appropriate physical constraints . in section 2 of this paper ,
we summarize the calculation in @xcite of the lienard - wiechert electric field for hyperbolic motion , and agree with their conclusion that the abrupt end of electric field lines at @xmath7 is unphysical . in section 3
, we derive the electric field for the constant velocity phase of the motion and for the accelerated phase from the same standard lienard - wiechert formula , only using different retarded time formulas for the two different situations .
consequently , our electric field lines are continuous as the retarded time increases through the transition time between constant velocity and accelerated motion , and our figures 2 and 3 do not show the discontinuous behavior of the corresponding figures 3 and 4 in @xcite . as the transition time approaches negative infinity the constant velocity part of the electric field develops a delta function contribution that has been proposed by others as necessary to preserve gauss s law for the electric field .
the lienard - wiechert electric field of a unit point charge is given by @xmath9 } { ( r-{\bf r{\mbox{\boldmath$\cdot$}}v})^3}\right\}_{\rm ret},\hspace{.3 in } \label{lw}\ ] ] where @xmath10 is the vector from the unit point charge to the point of observation .
all variables ( @xmath11 ) on the right - hand side of eq .
( [ lw ] ) are evaluated at the retarded time , @xmath12 . for the trajectory in eq .
( [ zt ] ) , the electric field is derived in eqs .
( 2)-(11 ) of @xcite .
we summarize their derivation below . the variables in eq .
( 2 ) are given ( in cylindrical coordinates ) by @xmath13{\bf\hat z}= { \mbox{\boldmath$\rho$}}+\left(z-\sqrt{z_0 ^ 2+t_r^2}\right){\bf{\hat z}}\\ { \bf v}&=&\frac{t_r{\bf{\hat z}}}{\sqrt{z_0
^ 2+t_r^2}}\label{v}\\ { \bf a}&=&\frac{z_0 ^ 2{\bf{\hat z}}}{(z_0 ^ 2+t_r^2)^{\frac{3}{2}}}.\end{aligned}\ ] ] for simplicity , we evaluate the electric field at time @xmath14 . for @xmath14 ,
the retarded time is negative and satisfies the relation @xmath15 with the solution @xmath16 putting eqs .
( 3)-(7 ) into eq . ( [ lw ] )
gives the electric field ( after some algebra ) @xmath17}{[(\rho^2+z^2+z_0 ^ 2)^2-(2 zz_0)^2]^{\frac{3}{2}}}\theta(z ) .
\label{eh}\ ] ] the electric field must vanish for negative z because it can be seen from eq .
( [ tr ] ) that there is no solution for @xmath18 for negative @xmath19 .
physically , this results from the fact that a charge with the trajectory given by eq . ( [ zt ] ) is always outside the past light cone of any point with negative @xmath19 ( at @xmath20 ) , and consequently can not have any effect for negative z. the field lines for @xmath21 can be found from either eq .
( [ lw ] ) or eq .
( [ eh ] ) .
they , of course , give the same answer because eq .
( [ eh ] ) is derived from eq .
( [ lw ] ) .
the electric field lines are plotted in fig . 2 of @xcite , and in fig . 1 of this paper . )
, with the retarded time given by eq .
( [ tra ] ) .
the scale is set with @xmath22.,height=384 ]
a remarkable feature of these curves is that the field lines stop abruptly at @xmath7 because there is no solution for the retarded time for negative @xmath19 .
it is recognized in @xcite that the field given by eq .
( [ eh ] ) and represented in our fig .
1 can not be a physical electric field , because an electric field can not just end in empty space without terminating on electric charge .
this would violate maxwell s equation @xmath23 , which leads to gauss s law and the requirement that the normal component of e be continuous across any surface without a surface charge .
we attribute this failure of the lienard - wiechert electric field to the fact that the limit @xmath24 ( even if in the distant past ) , required for complete hyperbolic motion , is unphysical in itself .
following @xcite , we modify the motion of the point charge so that it is originally moving at a constant velocity @xmath25 until a time @xmath26 , after which it moves with a constant acceleration @xmath27 in its instantaneous rest system . reference @xcite calls this motion truncated hyperbolic motion " .
we continue to use eq .
( [ lw ] ) for constant velocity by just setting @xmath28 equal @xmath29 .
the only other change we have to make is to use the retarded time appropriate to the constant velocity motion . for constant velocity motion up to the time @xmath26
at which the acceleration starts , the velocity is given by @xmath30 which follows by just substituting @xmath26 into eq .
( [ v ] ) .
the constant velocity trajectory of the charge is given by @xmath31 where we have introduced the notation @xmath32 for the position of the charge at the time the acceleration starts .
then , the retarded time for the constant velocity motion of the charge satisfies the relation @xmath33 with the solution @xmath34}{z_0 ^ 2},\quad t_r < t . \label{trv}\ ] ] note that the restriction @xmath35 no longer applies .
the electric field for the constant velocity part of the truncated hyperbolic motion is given by substituting eqs .
( [ vv ] ) , ( [ zqv ] ) , and ( [ trv ] ) into eq .
( [ lw ] ) with * a * taken as @xmath29 . the result for @xmath36
is given in [ 2 ] as@xcite @xmath37^{3/2}},\quad t\le t , \label{ev}\ ] ] where @xmath38 is the vector to the point of observation from the point @xmath39 on the z - axis which the unit point charge would have reached had it followed the constant velocity trajectory until the observation time , @xmath14 .
using eq .
( [ bcr ] ) , we can write eq .
( [ ev ] ) as @xmath40}{[\rho^2z_0 ^ 2+(zz - z_0 ^ 2)^2]^{3/2}},\quad t\le t. \label{er}\ ] ] we have derived the electric field lines for all the figures of this paper by the following procedure : we start close to the point @xmath41 at @xmath42 intervals from the z axis , where we calculate initial values @xmath43 and @xmath44 . then we determine each next point in the field lines by @xmath45 where @xmath46 is a small increment ( we chose @xmath47 . ) .
the electric field values were calculated from eq .
( [ lw ] ) , using eq .
( [ tra ] ) for @xmath48 and eq .
( [ trv ] ) for @xmath49 .
we also used eq .
( [ eh ] ) for @xmath48 and eq .
( [ er ] ) for @xmath49 as a check on our calculations .
the field lines of @xmath50 for several values of t , the time at which the acceleration starts , are shown in figures 2 , 3 , and 4 .
these figures can be considered corrected versions of figures 3 , 4 , and 5 of @xcite . ) with @xmath51 .
the acceleration is zero for the lines outside the sphere ( shown as a circle on the figure ) .
the lines inside the sphere use eq .
( [ tra ] ) for the retarded time , and those outside the sphere use eq .
( [ trv]).,height=384 ] in figure 2 , and subsequent figures , the transition from constant velocity motion to accelerated motion takes place on a sphere defined by the equation @xmath52 the field lines outside the sphere are straight lines directed radially outward from the point @xmath53 on the z - axis .
the acceleration electric field lines are curved lines starting at the point @xmath41 on the z - axis , and ending at the transition sphere . at the transition sphere they are in the @xmath54 direction ,
directed radially outward from z , the retarded position of the charge .
this can be seen from the cross product of @xmath55 , which vanishes just inside the sphere .
figure 2 shows that the field lines are continuous through the transition from constant velocity motion to accelerated motion .
as the field lines pass through the spherical surface between the accelerated and the constant velocity motion , the normal component of @xmath50 is continuous , and gauss s law is satisfied at the surface .
this can be seen from eq .
( [ lw ] ) , where the discontinuity in the acceleration does not affect the radial component ( @xmath56 ) of the retarded field .
this is in contrast to fig . 3 of @xcite , where the field lines inside and outside the sphere are disconnected , and the continuity of the normal component of @xmath50 and gauss s law are violated .
an electric field does not usually have a kink in its field lines .
this is because maxwell s equation @xmath57 usually requires that the transverse component of e be continuous as a field line crosses a surface .
this follows from stokes law which relates the discontinuity in the transverse component of @xmath50 to the integral @xmath58 , where @xmath59 and @xmath60 are just inside and just outside the transition sphere .
the integral is zero if @xmath61 is finite , but in the present case @xmath61 is infinite because the acceleration is discontinuous . if we write the electric field @xmath50 in terms of its constant velocity and acceleration parts as @xmath62 , then @xmath63 has a delta function part , given by @xmath64 , so the integral equals @xmath65 , which is just the discontinuity in the transverse component of @xmath50 .
the break in the direction of the field lines in fig .
3 is seen to be more extreme than that in fig . 2 , but the normal component of @xmath50 is still continuous , and the discontinuity in @xmath50 transverse is still compensated by the discontinuity in @xmath66 .
in contrast to our fig . 3 , the corresponding fig .
4 in @xcite is particularly unusual .
that figure shows no relation between the constant velocity motion and the ensuing accelerated motion .
a connecting field " was introduced to somehow close the gap between these disconnected field lines .
but the connecting field has no physical basis for this case .
figure 4 has a very early transition time , @xmath67 , and shows field lines that look like fig . 1 for positive @xmath68 , but seem to be approaching a delta function near @xmath7 . the sequence of figures 2 , 3 , and 4 is seen to be progressing from an almost normal electric field to a combination of the field of fig . 1 with the addition of a delta function along the z - axis
we can tell that the field lines are approaching a delta function along the z - axis for large negative transition times by looking at eq .
( [ er ] ) for the constant velocity part of the electric field as @xmath69 . in this limit ,
the rho component , @xmath70 , becomes infinite for @xmath7 , and gets arbitrarily small for @xmath71 . at the same time , for @xmath69 , the integral of @xmath70 from @xmath72 to the spherical surface at @xmath73 is given by @xmath74^{3/2}}\nonumber\\ & = & \frac{2\rho}{\rho^2+z_0 ^ 2}. \label{inte}\end{aligned}\ ] ] the integration in eq .
( [ inte ] ) has to be over the entire range of the constant velocity electric field , from just at the transition sphere to @xmath72 , in order to get the correct factor for the delta function .
we conclude from this result that the electric field emitted by a unit charge in truncated hyperbolic motion with the transition time @xmath75 is given by the acceleration field in eq .
( [ eh ] ) plus the delta function we have calculated above .
the result is @xmath17}{[(\rho^2+z^2+z_0 ^ 2)^2 -(2zz_0)^2]^{\frac{3}{2}}}\theta(z ) + \frac{2{\mbox{\boldmath$\rho$}}}{\rho^2+z_0 ^ 2}\delta(z ) .
\label{edelta}\ ] ] this is just the electric field that has been conjectured in @xcite , and derived by others@xcite using differing methods .
+ * acknowledgement * + i would like to thank david j. griffiths and daniel j. cross for useful correspondence .
9 we use gaussian units with c = 1 .
joel franklin and david griffiths , the fields of a charged particle in hyperbolic motion " , am .
j. phys .
* 82 * , 755 - 763 ( 2014 ) see , for instance , the derivation in example 10.4 in _ introduction to electrodynamics _ , 3rd edn , d. j. griffiths ( san francisco : addison wesley , 1999 ) h. bondi and t. gold , the field of a uniformly accelerated charge , with special reference to the problem of gravitational acceleration , " proc . r. soc .
london ser . * a 229 * , 416 - 424 ( 1955 ) d. g. boulware , radiation from a uniformly accelerated charge , " ann .
phys . * 124 * , 169 - 188 ( 1980 ) daniel j. cross , completing the lienard - wiechert potentials : the origin of the delta function fields for a charged particle in hyperbolic motion " , arxiv:1409.1569 ( 2014 ) | we find the electric field of a point charge in ` truncated hyperbolic motion ' , in which the charge moves at a constant velocity followed by motion with a constant acceleration in its instantaneous rest frame .
the same lienard - wiechert formula holds for the acceleration phase and the constant velocity phase of the charge s motion .
the only modification is that the formula giving the retarded time is different for the two motions , and the acceleration is zero for the constant velocity motion .
the electric field lines are continuous as the retarded time increases through the transition time between constant velocity and accelerated motion . as the transition time approaches negative infinity the electric field develops a delta function contribution that has been introduced by others as necessary to preserve gauss s law for the electric field . |
research in the last decade showed that to understand the assembly and formation of massive galaxies it is critical to study submillimeter galaxies ( smgs ; see * ? ? ? * for a review ) . since their first detection ten years ago @xcite , more than three hundred dust - enshrouded high - z objects ( z @xmath72 )
have been detected by scuba and by mambo in the submillimeter and millimeter regime ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . due to the large beam size in the ( sub)mm ( scuba : 15 , mambo : 11 ) , it is impossible to do a proper identification in other wavebands ( such as the optical and near - infrared ) based on bolometer data only .
thus , typically interferometric observations at radio wavelength with the vla were used to identify the location of the smgs , yielding a radio identification rate of @xmath8% ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , depending on both the depths of the radio and the bolometer maps .
the main drawback of this technique is , however , that the radio flux is redshift dependent , whereas the ( sub)mm flux density at z @xmath9 is not ( due to the inverse k correction ; e.g. , * ? ? ?
* ) : one would thus expect the highest redshift ( z @xmath10 ) sources ( e.g. , * ? ? ? * ) to be very faint at radio wavelengths if they were to follow the radio - fir correlation ( e.g. , * ? ? ?
we note that the radio - identified smg with the highest redshift lies at z @xmath11 @xcite .
based on this sample , co observations of radio identified smgs ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) lead to the main conclusion that smgs are massive galaxies in formation and @xcite argue that the contribution to the star - formation rate density ( sfrd ) from this population is missed by uv - selected surveys .
if one only uses radio - identified sources ( e.g. , * ? ?
* ; * ? ? ?
one thus has to be aware of the fact that the radio selection introduces a bias towards the low end of the true redshift distribution . to summarize , our knowledge about the nature and properties of the smg population is mainly based on radio - identified smgs a biased sample .
one of the open questions in the investigations of smgs is if there exists a significant very high - redshift tail ( z @xmath12 ; e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the existence of such a tail would not be in contradiction to constraints from measurements of the cosmic infrared background @xcite .
interferometric observations in the ( sub)mm hold the promise to locate the positions of bright smgs to sub - arcsec accuracy
. one of the brightest scuba sources in the goods north field is goods 850 - 5 ( wang et al . 2004 ; this source is labeled gn 10 in pope et al .
2005 ) with a submm flux @xmath13 mjy @xcite mjy ] . until very recently this bright smg
did not have a known counterpart which motivated observations with the iram plateau de bure interferometer ( pdbi ; * ? ? ?
* ) to obtain an accurate position for this source . in the meanwhile ,
@xcite reported the location of goods 850 - 5 through sma observations . based on their detection in the submm ( and subsequent identification in ir observations ) and a 2 @xmath14 detection in the radio ( @xmath15jy ) , @xcite argue that the redshift of goods 850 - 5 is z @xmath12 and could even lie between z @xmath16 - with a much lower probability at z@xmath17 . here
we report the results based on our pdbi observations .
we adopt the cosmological parameters @xmath18 and @xmath19 @xcite .
furthermore , we assume a salpeter initial mass function from 0.1 to 100 @xmath20 .
we use the ab magnitude scale @xcite .
the final map resulting from our observations between december 2005 and august 2006 with the plateau de bure millimeter interferometer in c and d configuration are based on 6.8 hours of on - source observing time .
the phase center ( see table [ tab : position ] ) was set at the nominal scuba position published in @xcite .
the source was observed in the 1 mm and 3 mm band simultaneously .
the 1 mm receiver was tuned at 240.000 ghz ( corresponding to 1.25 mm ) to obtain an accurate position .
the 3 mm receiver was tuned to 98.3188 ghz , i.e. , the frequency of the co(2 - 1 ) transition redshifted to z @xmath21 .
this is the redshift of the optical source goods850 - 5a @xcite which is situated close ( @xmath22 ) to the scuba bolometer position .
the total bandwith of 580 mhz covers a redshift space of @xmath23 at that redshift ( i.e. within the redshift uncertainty of the source , cf : z @xmath24 in cohen et al .
2000 and pope et al .
2006 ; z @xmath25 in wang et al . 2004 ; z @xmath26 in wirth et al . 2004 ) .
the data were calibrated through observations of standard bandpass ( 3c273 , j0418@xmath0380 ) , phase / amplitude ( 1044@xmath0719 , 1150@xmath0497 ) and flux calibrators ( crl 618 , 3c273 , 1044@xmath0719 , 1150@xmath0497 ) and reduced with the gildas software packages clic and map .
the fwhm of the beam is @xmath27 at 240 ghz and @xmath28 at 98.3 ghz .
we detected the dust continuum emission of goods 850 - 5 at 5 @xmath14 significance at @xmath29 mjy .
the derived position is ra(j2000)@xmath3012:36:33.45 , dec(j2000)@xmath30 62:14:08.95 ( hereafter : mm [email protected] ) with an estimated position uncertainty of @xmath310.3 .
the position and flux density of mm [email protected] are derived from point source fits of the visibilities in the uv plane and are consistent with the values derived in the image plane .
the difference between the scuba @xcite and our millimeter interferometric position is only 0.02 in ra and 0.48 in dec , and our position is consistent with the sma position from @xcite , see summary in table [ tab : position ] . in the 3 mm band
no line was detected down to a rms of 0.8 mjy beam@xmath32 ( channel width : 40 mhz , 120 km s@xmath32 ) on the position of the optical source at z @xmath21 ( and also at the position of goods 850 - 5 ) .
this corresponds to an upper limit ( 3 @xmath14 ) of the integrated flux intensity of 1.2 jy km s@xmath32 , adopting a typical co width for a smg of 500 km s@xmath32 ( e.g. , * ? ? ?
* ) . however , we note that a number of co detections of smgs have even lower integrated flux intensities than this 3 @xmath14 upper limit ( e.g. , * ? ? ? * ) .
we have used new , deep vla radio continuum observations ( morrison et al . 2008 ) to constrain the radio flux density of goods 850 - 5 .
these observations reveal a highly significant radio source ( @xmath33jy ) ; i.e. a source that is significantly brighter than the upper limit reported in @xcite .
our derived flux density is however in agreement with the 3 @xmath14 detection reported in @xcite .
the radio emission is unresolved with a fwhm @xmath34 . in our analysis
, we use the exquisite multi - wavelength dataset available for the goods north field ( e.g. , alexander et al .
2003 ; giavalisco et al .
2004 ; dickinson et al .
, in preparation ) and have corrected these data products for the offset of 0.38 in declination between vla and the goods imaging products ( see also * ? ? ?
the pdbi / vla source is undetected in the deep acs images down to @xmath35 mag
. however goods 850 - 5 is detected at low significance in spitzer irac measurements at the few @xmath4jy level @xcite as summarized in table [ tab : flux ] .
the source is also detected at 24 @xmath4 m ( table [ tab : flux ] ) but undetected both at mips 70 @xmath4 m and 160 @xmath4 m at a 3 @xmath14 level ( @xmath36 mjy and @xmath37 mjy ) . however , due to the large beam size ( 18 at 70 @xmath4 m and 40 at 160 @xmath4 m ) our source would be confused with other objects within the beam .
we conclude that the irac counterpart presented in @xcite is the proper counterpart of mm [email protected] m fluxes by @xcite are consistent within 1.5 @xmath14 with the measurements reported by @xcite ] . due to the lack of a spectroscopic identification
, we apply several photometric redshift estimators focusing on different parts of the sed in order to put constraints on the possible redshift
range of the source . applying the radio/(sub)millimeter index @xcite on the pdbi flux and the radio flux
we estimate a redshift z@xmath38 .
furthermore , we use the infrared galaxy sed templates from @xcite to fit the sed of our source ( using the mips 24 @xmath4 m , scuba , pdbi and vla flux ) and this method proposes a redshift of z @xmath39 , within the errors consistent with the radio/(sub)mm spectral index estimates and infrared luminosity @xmath40 . we obtain the same result by only using the mips 24 @xmath4 m and the vla flux .
the use of the 850/1.2 mm ratio as a proper redshift indicator is very limited ( see fig . 13 in * ? ? ?
* ) . however , the scuba to pdbi flux ratio of 2.6 limits the possible redshift range from z @xmath41 ( see plot in * ? ? ?
as @xcite we have also applied the widely used photometric redshift code _ hyperz _
@xcite on the optical / near - infrared data from acs and irac .
our redshift estimates give a redshift z @xmath6 , i.e. consistent with the range of values proposed by @xcite .
however , we note that with this method only the stellar component of the source is fitted and the applied templates do not reflect the characteristic properties of infrared galaxies . assuming a redshift z @xmath5 , taking into account the optical / near - ir measurements from acs / spitzer and applying the reddening law by @xcite we estimate the extinction to be a@xmath42 . for comparison ,
@xcite and @xcite derived for smg samples mean reddening of a@xmath43 and 3.0 , based on optical / near - ir colors .
the source is undetected both in the soft ( 0.5 - 2 kev ) and hard ( 2 - 8 kev ) bands in the 2 ms chandra observations of goods north ( e.g. , * ? ? ?
however , the lack of x - ray detection can not exclude the presence of an heavily obscured agn in goods 850 - 5 .
using irac , @xcite proposed a diagnostic tool in order to distinguish between agn and starburst as dominating energy source . applying this tool on mm
[email protected] would propose an agn at redshift z @xmath44 .
however , we note that @xcite and @xcite pointed out that starbursts at z @xmath5 contaminate the irac selection for z @xmath45 agns . using either the template fitting from @xcite ( resulting in an infrared luminosity @xmath40 and taking into account the conversion between infrared luminosity @xmath46 and sfr @xcite ) , we derive a star formation rate sfr @xmath47 . by simply using the pdbi flux of 5.0 mjy at 1.25 mm we obtain a star formation rate sfr @xmath48 @xcite .
this is consistent with the sfr derived from the radio flux for a source at z @xmath5 of sfr @xmath49 ( see discussion in ivison et al .
2002 ) .
we conclude that the multi - wavelength data suggests that mm j123633@xmath06314 is a starburst at z @xmath504 . a total of 21 smgs in goods north are securely identified @xcite .
in addition to mm [email protected] there are four smgs that have even fainter vla 20 cm fluxes ( @xmath51jy ) than our source .
similar to the counterpart of mm
[email protected] none of them has an acs counterpart and their irac fluxes are faint ( @xmath52jy ) .
one of these smgs is the well - studied source hdf850.1 @xcite which is one of the excellent candidates to lie at z @xmath5 or higher . to summarize ,
about 20% of securely identified smgs could lie at high redshift z @xmath5 or higher .
@xcite and @xcite also reported the discovery of reliable candidates of redshift z @xmath5 or higher in other deep fields .
we have used the iram plateau de bure interferometer to obtain an accurate position and flux at 1.3 mm wavelengths of the smg mm [email protected] alias goods 850 - 5 ( @xmath1 mjy ) .
this source is coincident with a source detected in deep vla radio continuum observations at high significance ( @xmath33jy ) .
the source has also been detected by recent observations obtained at the sma @xcite and by spitzer @xcite in the irac and 24 @xmath4 m bands .
the photometric redshift indicators using the mir- and fir sed of mm [email protected] all point towards a redshift of z @xmath5 . based on our millimeter and radio detection
, we consider a much higher redshift ( z @xmath6 , as proposed by wang et al .
2007 ) unlikely .
however , only future facilities with wide mm - receivers planned e.g. the upgraded pdbi or the lmt will unambiguously solve the question of the true redshift of goods 850 - 5 ( alma will unfortunately not be able to reach goods north ) . in any case , goods 850 - 5 represent one of the few solid candidates of a high - z ( z @xmath10 ) smg that contribute to the cosmic sfrd at early cosmic times . we would like to thank the iram staff for help provided during the observations and especially arancha castro - carrizo and roberto neri for helpful support during the data reduction at iram grenoble .
furthermore , we are grateful to dave frayer for 70 and 160 @xmath4 m measurements on the pdbi position and emanuele daddi for instructive conversations and help in applying the @xcite sed templates on goods 850 - 5 .
finally , we appreciate the comments of the anonymous referee which clarified a number of points and improved our manuscript .
acs @xmath53 , irac 8.0 @xmath4 m , mips 24 @xmath4 m and radio images of the field of mm [email protected] .
pdbi contours of 1.25 continuum emission of mm [email protected] overlaid on the acs @xmath53 image and start at 3@xmath14 with steps of 1 mjy .
3.0 circles are drawn on the pdbi position in the irac 8.0 @xmath4 m , mips 24 @xmath4 m and radio images . ] | we have obtained a position ( at sub - arcsecond accuracy ) of the submillimeter bright source goods 850 - 5 ( also known as gn10 ) in the goods north field using the iram plateau de bure interferometer at 1.25 mm wavelengths ( mm [email protected] , flux density : @xmath1 mjy ) .
this source has no optical counterpart in deep acs imaging down to a limiting magnitude of i@xmath2 mag and its position is coincident with the position found in recent sub - millimeter mapping obtained at the sma @xcite . using deep vla imaging at 20 cm , we find a radio source ( @xmath3jy ) at the same position that is significantly brighter than reported in wang et al .
the source is detected by spitzer in irac as well as at 24 @xmath4 m . we apply different photometric redshift estimators using measurements of the dusty , mid / far - infrared part of the sed and derive a redshift z @xmath5 .
given our detection in the millimeter and radio we consider a significantly higher redshift ( e.g. , z @xmath6 * ? ? ?
* ) unlikely .
mm [email protected] alias goods 850 - 5 nevertheless constitutes a bright representative of the high - redshift tail of the submillimeter galaxy population that may contribute a significant fraction to the ( sub)millimeter background . |
to meet its science requirements , the large synoptic survey telescope ( lsst ) @xcite places stringent requirements on the accuracy of its photometric calibration .
the uniformity of photometric zeropoints across the sky is required to be 10 mmag in the grizy bands , and 20 mmag in u. additionally , color accuracy is required to be 5 mmag for colors that do not involve u , and 10 mmag for those that do . a plan for achieving these requirements
is presented in @xcite .
achieving this plan , and verifying the accuracy of its results , relies on external data in the form of photometric standards .
the goal of this paper is to explore the possibility that gaia @xcite can contribute to the external data needed for lsst photometric calibration .
the paper is organized as follows .
section [ sec : lsstphotocal ] presents an overview of the lsst photometric calibration process , and the role of external data .
section [ sec : gaiadata ] is a brief summary of the gaia data products that are relevant to our purpose .
section [ sec : directgaiastandards ] presents the transformation of gaia magnitudes into lsst magnitudes , and simulated results for both ms stars and wd .
section [ sec : syntheticg ] presents the synthetic g band approach and simulated results for ms stars .
section [ sec : conclusion ] summarizes our results , and implications for future work .
the overall strategy for lsst photometric calibration is based on @xcite .
they proposed _ directly _ measuring the system throughput as a function of wavelength , focal plane position , and time .
further , the _ normalization _ of the throughput in each observation ( the gray - scale zeropoint ) and the _ shape _ of the throughput curve ( the wavelength dependent terms ) , are explicitly separated and measured with separate procedures for both the telescope system response and the atmospheric transmission .
calibration systems based on this approach are already in use at panstarrs and des .
the goal of the calibration process is to produce standard magnitudes , defined by @xmath4 where @xmath5 is the normalized standard bandpass for band b , set a priori for the survey , @xmath6 is the object flux at the top of the atmosphere , and @xmath7 = 3631 jy . to calculate @xmath8
the calibration system must determine , for each observation , the actual realized bandpass from detected electrons to the top of the atmosphere , @xmath9 , and a gray zeropoint @xmath10 .
with these quantities in hand , the standard magnitude is calculated by @xmath11 where @xmath12 is the instrumental magnitude ( log of counts ) and @xmath13 is the _ shape _ of the source spectrum , which we will refer to as the sed .
several hardware systems are required to implement the approach , which we briefly describe : * a dome screen projector designed to provide uniform ( @xmath14 variation ) illumination across the field of view , while minimizing stray light .
this projector system will have the capability to not only illuminate the screen with broadband white light , but also narrow - band light to measure the system response at individual wavelengths .
the narrow - band light will be generated by a tunable laser , capable of producing light from @xmath15 nm and tunable in 1 nm increments .
the brightness of the screen is measured with a nist - calibrated photodiode , so that the relative intensity at different wavelengths can be precisely determined . * a 1.2-m auxiliary telescope with an @xmath16 spectrograph , located adjacent to the lsst itself .
this auxiliary telescope will obtain spectra of a chosen set of atmospheric probe stars across the sky to determine an atmospheric absorption model .
* water vapor monitoring system , consisting of a gps system and a microwave radiometer copointed with the lsst telescope , and monitoring a similar field of view .
this supplements the auxiliary telescope spectra , which are unable to track the sometimes rapid variations of water vapor in time and space .
lsst uses a form of self calibration , as pioneered by @xcite , to set the zero point for each exposure . the inputs to the self calibration process are the sed - corrected magnitudes of the calibration stars in band @xmath17 , @xmath18 , for the entire survey .
these are calculated from the instrumental magnitudes @xmath12 , the object seds @xmath19 , the standard bandpass for the photometric system , @xmath5 , and the system bandpass at the moment of observation , @xmath20 as @xmath21 self calibration utilizes the facts that the survey observes each calibration star many times over the course of the survey , and that the survey fields overlap , to construct a least squares system for the zero points of each observation .
the resulting set of zero points for each band has an arbitrary offset , which makes it impossible to determine object colors without additional information .
this additional information can come from a variety of sources : * photometric standards in individual lsst bands . * photometric standards in a bandpass which can be synthesized from the lsst bands .
* locus of a well understood stellar population in color - color space .
the need for external information goes beyond simply determining the band zero point offsets , which , in principle , could be obtained from only a single standard . in practice , results from self calibration will contain some level of spatially dependent error , which tends to be greatest for the low - order spatial modes . assessing the magnitude of this error , and correcting it
if needed , can only be performed if one or more of the above sources of information is available as a function of position on the sky .
this is , of course , the more demanding application , and it is the one we focus on in this paper . as we show below , gaia data can contribute both to band zero point determination and to verification / correction of self calibration errors through the first two of the approaches listed above .
we do not discuss the third approach here , but refer the reader to @xcite as an example of its use .
gaia s principal photometric data product @xcite is calibrated magnitudes through a broadband g filter .
additionally , it produces low resolution red ( rp ) and blue ( bp ) spectra , and higher resolution ( rvs ) spectra within a narrow band intended mostly to supply radial velocities .
it is tempting to use the rp and bp spectra directly to estimate the stellar sed @xmath22 and then use that to calculate the color terms needed to transform from the gaia g band to the 6 lsst bands .
the biggest problem with this approach is that rp and bp spectra , due to their low resolution and relatively broad psf , can not be directly inverted to give the incident stellar sed with sufficient accuracy .
preliminary simulations showed that systematic errors of order 0.1 mag are incurred , which rules out the direct approach for our purposes .
we are better off taking a different approach , and instead using apsis @xcite , gaia s astrophysical parameters inference system .
we realized that apsis , which of course utilizes bp / rp as a major input , would lead to smaller errors for our case .
there are two reasons for this .
first , apsis takes account of data beyond bp / rp , in particular the parallax to determine absolute luminosity .
second , apsis has built in the knowledge that these are not arbitrary spectra , but the spectra of stars .
this prior knowledge significantly reduces the errors in the color terms @xmath23 ( defined in section [ sec : transforming ] ) from our initial naive concept .
apsis utilizes all available gaia data for each star to infer the stellar parameters @xmath0 , @xmath1 , @xmath24 $ ] , and @xmath25 .
@xmath25 is a stellar extinction parameter used by apsis , which is within a few percent of the more commonly used extinction at 550 nm , @xmath2 , for values of @xmath26 near 3 . for our purposes
the difference between @xmath25 and @xmath2 is inconsequential , and in the remainder of the paper we use the more common @xmath2 . for sufficiently bright stars ( @xmath27 ) , this includes spectra from the rvs instrument , which substantially improves the accuracy of the resulting stellar parameters .
the usable range of magnitudes for lsst standards is roughly @xmath28 , with the bright end set by detector saturation , and the faint end set by signal - to - noise for a 15 sec exposure .
therefore , we must make use of stellar parameters determined without rvs .
table [ table : apsis_accuracy ] shows the predicted accuracy of these parameters .
as mentioned in section [ sec : lsstphotocal ] , the errors remaining after self calibration can be assessed and/or eliminated if we have an accurate set of photometric standards in the six lsst bands with a suitable distribution across the sky .
work is in progress @xcite to obtain a small set of suitable white dwarfs observed with hst whose photometry can be accurately transformed to the lsst bands , but their number and spatial distribution may not be adequate to the task addressed here .
the minimum number of standards required depends on the spatial power spectrum of the self calibration errors , and is difficult to determine in the absence of actual lsst data .
the low order spatial error modes ( @xmath29 ) from self - calibration can be adequately measured over the roughly @xmath30 steradians observed by lsst with roughly ten standards .
clearly , it would be desirable to have more , if only to verify that the power spectrum is negligible at higher @xmath31 values .
as an upper limit , 2000 standards would provide one standard in each lsst field of roughly 10 square degrees .
every lsst observation would then include a standard , and spatial error modes to @xmath32 would be tightly constrained .
the number of standards employed in practice will be guided by the error characteristics of the self calibration process , and possibly constrained by the availability of ground - based spectroscopy resources ( see section [ sec : usingwd ] ) each observation of a standard constrains the self calibration equations by a direct measurement of the zero point at that place and time : @xmath33 in this paper we are concerned with the uncertainties in determining @xmath34 that result solely from uncertainties in @xmath35 . in evaluating these ,
we make two assumptions : * @xmath36 is known exactly .
this is of course not the case , since it depends on a variety of measurements with finite uncertainty , but this allows us to isolate the uncertainty which is due solely to the precision of the standard .
* @xmath36 is sufficiently close to @xmath37 that the difference can be ignored in this context .
this is not obviously the case , since @xmath37 is in principle arbitrary , but in the case of lsst we will pick @xmath37 to be statistically as close as possible to @xmath36 .
we therefore approximate the calibration errors that result from an imperfectly characterized standard as @xmath38 where @xmath39 is the difference between the actual and assumed sed of the standard . to make use of gaia data for lsst photometric standards , we must transform gaia g magnitudes into lsst magnitudes in each of the six lsst filter bands .
as figure [ fig : bandpasses ] shows , the gaia g band is very broad compared with the lsst bands , and has an overlap with each of them .
this suggests that an accurate transformation might be possible given an achievable level of knowledge of the stellar sed .
the transformation process is as follows : @xmath40 where @xmath17 denotes one of the six lsst bands , and @xmath41 the gaia g band .
@xmath42 we assume that the stellar flux , @xmath6 , can be adequately parameterized by the apsis stellar parameters @xmath43 , which in turn means that the color @xmath23 can also be so parameterized : @xmath44 our main concern in this paper is with the uncertainty in the resulting @xmath45 magnitudes when the stellar parameters are determined by apsis .
the uncertainty is given by @xmath46 the expected uncertainty in gaia photometry , @xmath47 , is presented in @xcite , and reproduced in figure [ fig : gaiaphotometricerrors ] .
note that the uncertainties in the figure are for single visit photometry .
the expected uncertainty in the end of mission photometry is lower by roughly @xmath48 , yielding uncertainties of roughly 0.4 mmag at g=16 and 0.8 mmag at g=18 .
it is clear that @xmath47 makes a negligible contribution to the total uncertainty of 5 mmag required for lsst , as discussed in section [ sec : lsstphotocal ] .
therefore we require the uncertainty @xmath49 to be 5 mmag or less .
the remainder of the paper is largely occupied with the assessment of @xmath49 . given the enormous number of stars in both the lsst and gaia catalogs in the relevant magnitude range , we can be highly selective in the stellar types we select as standards .
the most useful stellar type for our purpose is not immediately clear , since it depends on the details both of the performance of gaia apsis , and on the shape of the lsst filter bandpasses .
we first examine the suitability of main sequence ( ms ) stars . in order to estimate the precision to which lsst magnitudes
can be found using gaia s apsis , we simulate the observed magnitudes of the stars in the phoenix spectral library @xcite for the gaia g band and the lsst u , g , r , i , z and y bands .
the phoenix spectra do not include interstellar reddening due to dust , so we redden them with the prescription of @xcite using @xmath50 .
we then used the published uncertainties in the stellar parameters determined by apsis ( in @xmath0 , @xmath1 , @xmath51 and @xmath2 ) to find the standard deviation of the colors @xmath23 for objects whose parameters are within the apsis uncertainties . in this fashion , the standard
deviation and other statistics concerning the predicted precision of gaia measurements of stars can be found .
we neglect the covariance between the various stellar parameters , which surely results in an overestimate of the dispersion in the resulting colors .
this assumption is necessary due to our lack of knowledge of the covariance terms , which have not yet been published .
the overestimate is likely in the range of 2 to 4 .
it is necessary to interpolate between the phoenix data points because the phoenix spectra used as input are widely spaced in stellar parameter space , so that typically only a few spectra are within the range of uncertainty of the apsis stellar parameters .
the interpolation program linearly interpolates in @xmath52 and @xmath2 between the magnitudes calculated from the available spectra .
the program selects the interpolated points drawn from a 4-dimensional normal distribution with variances determined by the expected apsis errors , and then calculates the lsst magnitudes for each point .
it then finds the standard deviation of the @xmath23 colors of the indistinguishable points , in order to find the predicted error from gaia data .
.accuracy of astrophysical parameter estimation ( internal rms errors ) with the apsis algorithm aeneas using bp / rp spectra and parallaxes , but without rvs data @xcite .
[ table : apsis_accuracy ] [ cols="^,>,>,^,^,^ " , ] [ table : wd_sd ]
in this section we turn to the second approach mentioned in section [ sec : lsstphotocal ] , that the six lsst bands can be combined into an approximate gaia g band , and then compared directly with gaia g band photometry , a possibility suggested by @xcite . looking again at figure [ fig : bandpasses ]
, it seems plausible that the gaia g bandpass can be reasonably approximated by linear combination of the lsst bandpasses @xmath53 which leads to @xmath54 where @xmath55 is the flux of an object with sed @xmath56 integrated over the bandpass @xmath57 .
if the approximation were an equality , then we could directly compare , for all objects in the gaia catalog within the lsst magnitude range , this synthetic g band flux to the actual measured g band flux .
any discrepancies beyond the expected random scatter would indicate a systematic problem with the lsst photometric calibration process .
the spatial dependence of these discrepancies , and their dependence on stellar parameters , particularly @xmath0 , would give clues to what aspects of the calibration process were at fault .
additionally , it might also be possible to correct the lsst fluxes to minimize the discrepancies , while maintaining consistency with our knowledge of stellar populations in color - color space , thereby improving the lsst calibration . in practice
, there will be a significant difference between @xmath58 and @xmath59 , so that @xmath60 leading to a systematic flux error @xmath61 figure [ fig : gaiagsynthresid ] shows @xmath62 as a function of @xmath63 , expressed in magnitudes , evaluated over the entire phoenix sed set restricted to stars with @xmath64 .
this restriction corresponds roughly to @xmath65 .
cooler stars show much larger residuals .
as shown in the figure , if the @xmath66 magnitudes are corrected with a third order polynomial in @xmath63 , then the remaining scatter is @xmath67 mmag , negligible in the lsst context .
note that the same approach could be used for wd , with a smaller scatter , but also a much smaller sample of stars .
it is clear from this that @xmath66 magnitude residuals will be diagnostic of any systematic errors in lsst calibration that exceed the 5 mmag level required by the srd .
if a significant level of systematic error is found , the @xmath66 data can be employed to diagnose the causes , though not as directly as with the direct standards discussed in section [ sec : directgaiastandards ] .
could we use the @xmath66 data to directly constrain , and thereby improve , the self calibration process , as we can with the direct standards ?
the main limitation is that , given an error @xmath62 for some particular star , there is not a unique way to decompose it into the errors in the individual lsst bands . in the six dimensional lsst photometry space ,
we can move it anywhere on the five dimensional hyperplane given by @xmath68 to remove this degeneracy , we can additionally require that the star must remain on the color - color track for its stellar population . to the extent that this track is scatter free , the corrections to the individual lsst magnitudes can be uniquely assigned . again , wd may be more valuable than ms stars for this purpose .
a quantitative assessment of the resulting accuracy is beyond the scope of this paper .
we investigated two different approaches to using gaia data to provide accurate photometric standards for wide - field surveys such as lsst .
in the first , gaia g - band photometry is combined with stellar parameters derived from the gaia apsis pipeline ( for ms stars ) or from ground based spectroscopy ( for wd ) to determine the transformation from g - band to the six lsst bands .
the accuracy achieved with ms stars falls well short of lsst requirements , though with potential for improvement through the use of ground based spectroscopy .
direct standards derived from wd have the potential to meet lsst requirements , particularly if they are restricted to high galactic latitudes .
additional work will be required to realize this potential in practice .
possible systematic errors in wd sed models must be carefully assessed , and the accuracy of interstellar reddening derived from ground - based spectroscopy or from gaia rp / bp spectroscopy needs to be more carefully quantified .
both of these tasks will become possible in the near future using multi band hst photometry in conjunction with ground - based spectroscopy .
it is worth noting that the ground - based observations required to realize the potential of gaia wd for photometric calibration may require substantial investment of telescope resources if a large number of standards is desired .
a spectrum of a single wd in our magnitude range requires roughly two hours of exposure time on a 6 meter class telescope to reach the necessary signal - to - noise .
similar considerations apply to the use of ms stars . in the second approach ,
the six lsst bands are linearly combined into a synthetic g - band which approximates the gaia g - band . although the two g - bands have significant wavelength dependent differences , the resulting two sets of magnitudes have very small differences for most stars with @xmath69 .
this allows a useful check on the spatial dependence of lsst calibration accuracy , and within currently unknown limits , the correction of such systematic errors .
it seems worthwhile to actively pursue both approaches during the next years , as gaia science data is released , and lsst nears operation .
we wish to thank drs .
abhijit saha ( noao ) , edward olszewski ( u. arizona ) , and eljko ivezi ( u. washington ) for many useful discussions which have improved the paper . | we consider the possibility that the gaia mission can supply data which will improve the photometric calibration of lsst . after outlining the lsst calibration process and the information that will be available from gaia , we explore two options for using gaia data .
the first is to use gaia g - band photometry of selected stars , in conjunction with knowledge of the stellar parameters @xmath0 , @xmath1 , and @xmath2 , and in some cases @xmath3 , to create photometric standards in the lsst u , g , r , i , z , and y bands .
we consider both main sequence ( ms ) stars and da white dwarfs ( wd ) .
the accuracies of the resulting standard magnitudes from ms stars are found to be insufficient to satisfy lsst requirements using gaia data alone , but with the potential to do so when supplemented with ground based spectroscopy .
the accuracies of the wd derived standards are generally adequate , but also require ground based spectroscopy .
the second option is combine the lsst bandpasses into a synthetic gaia g band , which is a close approximation to the real gaia g band .
this allows synthetic gaia g photometry to be directly compared with actual gaia g photometry at a level of accuracy which is useful for both verifying and improving lsst photometric calibration . |
the hubbard type of models in the theory of strongly correlated electron systems are notably successful @xcite . in particular
, it is remarkable the way they reproduce the properties of mott insulators , such as metal - transitions oxides and copper - oxygen layered htc compounds @xcite .
however , the efforts for developing approaches having more basic foundations had not ceased , due to the expectation that they could open the way for obtaining more exact and specific results@xcite . in this sense ,
methods that are grouped into the so called band theory picture are also known as first principle calculations in the literature .
they are electronic structure calculations that begin with the interactions among electrons or atoms in vacuum .
the study of the band structure they predict , in principle should offer a road toward the effective and precise determination of the physical properties of each material@xcite .
some of them are : the configurations interaction scheme ( ci ) ; the local density approximations method ( lda)@xcite , the local spin density approximations procedure ( lsda ) and hartree - fock method ( hf ) . however , the above mentioned potentialities of those first principles approaches had been failing in describing many of so called strongly correlated electron system@xcite .
for example , the lsda , a sophisticated generalization of the lda procedure , was devised to describe local spin structures@xcite . however , although the method had offered satisfactory descriptions of the physical results in few materials , this success had not been universal and it also wrongly predicted the properties of some compounds , by example , the here considered la@xmath5cuo@xmath6 .
the motivation of the present work arose from a primary suspicion that perhaps the self - consistent hartree fock ( hf ) method , could had been underestimated in its possibilities for helping in the above described searches@xcite . in this sense
, it can be firstly remarked that it is widespread the criteria that for obtaining behaviors such as the mott insulator character , it becomes necessary the presence of short range correlations among electrons with spin quantized in different directions . by example , paraphrasing one type of mott s argument for specific systems : `` ... two electrons with spin resting on contrary directions are forbidden to occupy the same wannier orbital ... '' . on another hand , the orthodox hf approaches does not have in consideration the correlations among electrons of different spins .
therefore two electrons with opposite spins do not disturb one each other and consequently both of them can occupy the same wannier orbital .
clearly , the usual hf approach , seems not to be viable for investigating a system in which the mott s argument is appropriate .
however , the physical sources of the validity of cited mott s statement in some systems are not completely clear . by example : which is the physical origin of these short range correlations assumed in it ? .
even , the proper concept of correlations , roughly described as : everything missing in the single - particle hf state for to be the real many body ground state , makes clear how clouded their origins are . as the result of the study presented here ,
we believe that many of the so called correlation effects , can be effectively described even in the framework of the hf scheme , after removing certain symmetry restrictions which obstacle the finding of the best hf solutions .
such constraints are usually imposed on the space formed by the single particle orbitals , which are employed to construct the determinant like states among which the hf one shows minimal energy . by example , if after solving the hf problem , it occurs that the resulting self - consistent potential breaks the symmetry of the original crystalline lattice , it could create a gap and thus produce a mott kind of insulating solution .
this effect was originally discovered by slater in ref . .
this symmetry breaking effect has been also more recently underlined and deepened in ref . .
however , the removal of this kind of symmetry restrictions alone had not been able to describe the insulator properties of a large class of materials@xcite .
one of the central results of the present investigation , as it will be described just in what follows , is the identification of another important kind of symmetry restrictions that seemingly had been overlooked for long time .
it can be cited here that a fully unrestricted formulation of the hf problem was early done by dirac in ref . .
this work will consider the hartree - fock self - consistent problem as applied to a simple one band model of the la@xmath5cuo@xmath6@xcite , but following an unusual way . in order to leave freedom to obtain paramagnetic , ferromagnetic and antiferromagnetic solutions in the same context
, we look for single particle orbitals being non separable in their spacial and spinor dependence , i.e. they will have the structure @xmath7 .
in other words , in those states there is no an absolute common quantization direction for the electron spin .
thus , in each position the spin is quantized in a specific direction , and the equations of motion to be used will reflect this fact .
note , that to proceed in this way is not other thing that to apply the dirac s unrestricted formulation of the hf procedure@xcite .
we think that the restriction to @xmath8 and @xmath3 types of orbitals , usually employed in hf electronic band and quantum chemistry calculations , prohibits from the start the prediction of possible spontaneously symmetry breaking effects @xcite .
such a particular structure , excessively reduces the space of functions to be examined and consequently annihilates some possibilities to obtain exotic solutions ( like that ones that are present in the strong correlation effects ) .
we believe that in the context of the band theory , or more precisely , under the hf approach , it could be possible to reproduce the main characteristics of a wide class of _ mott insulator _ kind of materials .
the results of the simple model investigated here , as it will be seen , support this possibility for important compounds such as la@xmath0cuo@xmath6 , which under certain doping levels turn out in high temperature superconductors@xcite .
the present work heads in that direction , with eyes in also showing the potentialities of implementing the above considerations in self - consistent hf calculations , not only for describing electronic bands and the associated ground states , but also for studying atomic and molecular structures .
the exposition proceeds as follows . in the section [ chap:2 ]
we describe the details of the hf self - consistent method to be employed in next sections .
specifically , it will be discussed the imposition of restrictions on the space of single particle states in which the solutions will be searched and the possible physical consequences they could carry on . in section [ chap:3 ] the effective model we are going to employ will be exposed . the symmetry restrictions to be assumed and its corresponding tight - binding bloch basis will be defined .
section [ chap:4 ] is devoted to derive the hf equations associated to interacting electrons which free hamiltonian is given by the specially constructed effective tight binding model .
the tight binding bloch basis associated to the model is chosen in this section to have the maximal symmetry given by the group of translations leaving invariant the copper - oxygen plane in la@xmath0cuo@xmath6 .
considering the previously mentioned paramagnetic solution in a generic form , the free parameters of the effective hamiltonian are adjusted to reproduce the form of the only half filled band in the matheiss calculation of the band structure of la@xmath0cuo@xmath6 @xcite .
after defined the free hamiltonian of the model , the paramagnetic and antiferromagnetic solutions are obtained . the bloch basis shown in section [ chap:3 ]
was employed .
both solutions are also compared in this section and the properties of all the obtained states are analyzed in corresponding subsections .
appendix [ aelem ] presents notations , algebraic developments , constants and definitions cited along the writing . in a final seccin the main conclusions of the work are reviewed and various further tasks for its extension are described .
in the language of quantum mechanic the state of a system of n particles is described by a function depending on each one particle s spinor and spacial coordinates @xmath9 , where @xmath10 represents the corresponding set of quantum numbers @xcite .
the hf approximation consists on supposing that the above mentioned state can be expressed as a linear combination of @xmath11-products of orthonormalized orbitals @xmath12 with @xmath13 .
each one of these orbitals is interpreted as a single particle state , because it defines amplitude and probability distributions depending on a single particle coordinates . as usual , in what follows the word coordinates will mean the spacial as well as the spinor ones .
if the particles are fermions , the previously mentioned linear combination is called slater determinant @xcite .
let @xmath14 be the n - electrons system hamiltonian , including kinetic plus interaction with the environment hamiltonian @xmath15 , besides coulomb interaction among pairs of electrons @xmath16 .
the hf equations of motion for this system , leading the dynamic of the single particle states in a self - consistent way , are @xmath17\ \phi_{\eta}(x , s)&\\ \nonumber-\sum_{\eta_1}[\ \sum_{s^{\prime}}\int d^2x^{\prime } \phi^*_{\eta_1}(x^{\prime},s^{\prime } ) v(x , x^{\prime } ) \phi_{\eta}(x^{\prime},s^{\prime})\ ] \phi_{\eta_1}(x , s)&=\varepsilon_{\eta}\ \phi_{\eta}(x , s),\ \ \ \ \ \\end{aligned}\ ] ] where @xmath18 is a label in the basis formed by the solutions .
that is , each hf electron state is influenced by the presence of the other ones .
the self - consistent hamiltonian has two components , the coulomb like type of mean potential which their fellows create : the direct potential and the contribution reflecting the fact that two electrons can not occupy the same state : the exchange potential @xcite . the hf energy of the @xmath11 electrons system and the interaction energy of an electron in the @xmath19 state with the remaining ones , are given by @xmath20 the bracket notation definition is given in appendix [ aelem ] within the subsection [ sub : brac ] .
it can be noted that the system of equations ( [ hf ] ) is rotational invariant because it is written without imposing a spatially absolute direction for the spin quantization of the single electron orbitals .
this rotational invariant formulation of the self - consistent hf procedure was firstly introduced by dirac in ref . .
to solve ( [ hf ] ) is a complicated task because it is a system of coupled integro - differential equations .
the iterative method is one of the most frequently employed for solving this kind of systems and it is usually complemented by the imposition of symmetry restrictions that simplify the space of states to be investigated . however , the use of such constraints could avoid the obtaining of special solutions non obeying the added symmetry conditions .
although in some cases they could retain the minimal energy one , the method can hide the existence of interesting excited states and inclusive could wrongly predict the excitation features in some cases , as it will be seen in what follows .
a very common symmetry restriction usually employed in band theory and quantum chemistry calculations is to consider that single particle solutions of ( [ hf ] ) have spin quantized in a given direction in every point of the space @xcite .
that is @xmath21 where @xmath22 represent the pauli spinors with spin up and down in a certain direction respectively .
if the spacial functions @xmath23 and @xmath24 are the same , the hf calculation is called a restricted one , if they are different , the procedure is called unrestricted @xcite .
it is interesting to investigate the consequences of considering the possibly existence of non separable single particle states being solutions of the hf problem .
a positive answer to this question can open a natural context for obtaining solutions exhibiting magnetic properties and to allow their comparison with paramagnetic ones .
another important kind of restrictions posed on the first principles band theory evaluations is the a priori impositions of crystal symmetries . to beforehand impose a symmetry on the assumed to be searched solution , although it could be well rooted in the studied system , has the risk of hiding a possible spontaneous breaking of that invariance .
this could occurs due to the reduction of the space orbitals in which we are searching . in that case
it can turns out that the obtained solutions will not be an absolute extremal of the energy functional , but a conditional one , due to the fixed symmetry constraint .
for instance , let us consider the functional space formed by the allowed orbitals and the maximal subset of orbitals u which is invariant under a certain group of transformations * t*. consider also the maximal subset u@xmath25 being invariant under the group of transformations * * t**@xmath26 which is a subgroup of * t*. then , the set u@xmath26 obtained from imposing less symmetry restrictions a priori , should contain the set u. therefore , after to find the extremes of the same functional in u and u@xmath26 , it could be possible to obtain different results . in this case , the solution in u@xmath26 , in general , shall be the most stable of both .
however , also it could be the case that looking for a solution in u@xmath25 , an extremal function also pertaining to u arises as a solution .
in such a situation , such a configuration could be found from the beginning by finding the extreme of the functional in u , that is , by imposing more symmetry restrictions . in terms of the hf scheme
, this could mean that the states corresponding to both solutions have identical occupied single - particle states , but they curiously might show different sets of excited ones .
therefore , depending on the particular features of the material , removing a priori imposed symmetry restrictions on the set of allowed orbitals of the hf procedure , can predict new properties for the excited single particle states of the system .
such one could be for instance the gap appearance .
this effect can have physical relevance after noting that at finite temperatures the more stable state will be preferred by the system and then the state showing a gap should be expected to be selected at non zero temperature .
in this section the basis of the effective band model used to describe the dynamic of the less bounded la@xmath0cuo@xmath6 electrons will be presented .
the main considerations for defining the model and the determination of its characteristic parameters are given .
in subsection [ cuo ] the simplified electronic model for the copper - oxygen planes , as well the main definitions in its structure are introduced .
subsections [ sec : traslacionessr ] and [ tb ] are devoted to define the symmetry transformation group defining the tight binding bloch basis .
it is known that at low temperature la@xmath0cuo@xmath6 is an antiferromagnetic - insulator @xcite . however , in evident contradiction with the experiments the linear augmented plane waves ( lapw ) technics @xcite predicts a metal and paramagnetic zero temperature properties for this material .
nevertheless , such band calculation results show that the conduction electrons are strongly coupled to the bravais lattice centers of the copper oxygen planes .
clearly this tight - binding behavior is determined by the interaction of the electrons with its surrounding effective environment .
this defines the initial hypothesis of our model .
the less bounded electron in the la@xmath0cuo@xmath6 molecule is the cu@xmath27 s not paired one . that is , at difference from o@xmath28 ions , cu ones do not have their last shell ( 3d ) closed .
those copper 3d electrons fill the last band of la@xmath0cuo@xmath6 solid and in what follows they shall be referred as : the electron gas .
it seems appropriate to consider those electrons as strongly linked to cuo@xmath0 cells and moreover , given the above mentioned arguments , with special preference for the cu centers @xcite .
thus , our bravais lattice is going to be the squared net coincident with the array of copper sites ( see the figure [ f : bravais ] ) .
the presence of electrons pertaining to the various fully filled bands in the material plus the nuclear charges , plays a double role in the model .
firstly , it will act as an effective polarizable environment , which screens the field created by electron charges constituting the electron gas in the half filled band . consequently
, we will introduce a dielectric constant @xmath29 which will screen the coulomb interaction . secondly , as suggested by its spacial distribution and magnitude , the mean field created by the environment will be assumed to act as a periodic potential w@xmath30 being responsible for tight - binding confining of the electron to the cu centers .
it is also primordial in the model to take into consideration the interaction f@xmath31 among the electron gas and the jellium neutralizing its charges .
this background will be modeled here as a gaussian distribution of positive charges @xmath32 surrounding each lattice point and with characteristic radius _ b _ . in resume
, the free hamiltonian of the model takes the form @xmath33 where @xmath34 is the i - th electron s squared momentum operator ; _ m _ is the electron mass ; @xmath35 is the vacuum permitivity and @xmath36 moves on bravais lattice .
the versors @xmath37 and @xmath38 are resting on the direction defined by the lattice s nearest neighbours , see figure [ f : bravais ] a ) .
it is known that the distance between cu nearest neighbours is p@xmath39 3.8 @xcite .
we also consider the interaction among pairs of electrons in the form @xmath40 which , as remarked above , includes a dielectric constant associated to the presence of the effective environment .
we are seeking here for hf solutions with orbitals having a non separable spin and orbit structures .
thus , it was considered that the spin can show a different projection for the different wannier wavepackets to be superposed in defining those orbitals .
the spin for each of them will be either or @xmath2 or @xmath3 type , according they are linked either to one or the other of the two sublattices shown in figure [ f : bravais ] a ) .
thus , the single particle eigenstates will be chosen to be invariant , only under the reduced group of translations which transform each of those sublattices on itself .
it is important employ a procedure leaving independently identified the characteristics of the electron states in each one of the sublattices . that will allow to analyze solutions with dissimilar qualities in a same framework , that is : antiferromagnetic * afm * , paramagnetic * pm * and ferromagnetic * fm * ones
for this purpose , let us define the points of the two sublattices with indices r = 1 , 2 , as follows @xmath41 where @xmath42 and @xmath43 form the basis versors on each one of them .
the solutions we are looking for will be eigenfunctions of the operators @xmath44 belonging to the reduced discrete translation group which transforms a given sublattice on itself : @xmath45 if the bravais lattice were infinite , the brillouin s zone ( b.z . )
associated to @xmath44 would be the shadowed one on figure [ f : zb ] a ) .
note , that the continent square in this figure represents the b.z .
associated to the group of translations which leaves invariant the absolute lattice ( the lattice formed by the cu atoms in the cuo planes ) . however , given the impossibility of considering an infinite lattice for numerically solving the hf problem , then is also not allowed to consider its associated b.z . as continuous .
therefore we will impose periodic boundary conditions on the @xmath46 in the absolute lattice s boundaries x@xmath47= -l p y l p , x@xmath0= -l p y l p ( see figure [ f : zb ] b ) ) .
this condition determines the allowed set of * k * @xmath48 therefore , after recalling the discussion given in the introduction , note that we are now demanding less crystal symmetry on the single particle states which we are looking for , since a lower number of constraints are being imposed on the space of single particle states in which the solutions are searched .
label the eigenfunctions of the group of translations @xmath44 in this infinite point lattice .
the grey square indicates the corresponding brillouin zone ( b.z . ) of the sublattices .
the length of its side is @xmath49 .
b ) the net of points shows the discrete character of the brillouin zone when the absolute point lattice is finite with periodic conditions fixed in its boundaries .
the unit of the scale means @xmath50.,title="fig:",width=153,height=153 ] label the eigenfunctions of the group of translations @xmath44 in this infinite point lattice .
the grey square indicates the corresponding brillouin zone ( b.z . ) of the sublattices .
the length of its side is @xmath49 .
b ) the net of points shows the discrete character of the brillouin zone when the absolute point lattice is finite with periodic conditions fixed in its boundaries .
the unit of the scale means @xmath50.,title="fig:",width=153,height=153 ] the tight binding bloch basis which we are going to use is the following @xmath51 where @xmath11 is the number of electrons in the electron gas , @xmath52 is the spin z projection operator , where z is the orthogonal direction to the copper oxygen cuo@xmath0 planes ; @xmath53 -1 , 1 , are the eigenvalue of the previously mentioned operator and _ r _ = 1 , 2 , is the label which indicates each one of the sublattices . as we are going to work on a half filling condition , then @xmath11 coincides with the number of cells in the crystal with fixed periodic boundary conditions @xmath54 .
note , that due to the tiny overlapping among nearest neighbors approximation , the exact orthogonal character is only weakly lost between elements corresponding to different sublattices and having the same spin quantization .
that occurs because nearest neighbors belong to different arrays
. however the orthogonality between different elements corresponding to the same sublattice , as well as unity norm for every elements , is rigourously maintained .
this follows because they are constructed as bloch states in their corresponding sublattices .
the wannier orbitals @xmath55 represent the probability amplitude of encountering one electron in the vicinity of the site @xmath56 , that is on the given cell cuo@xmath0 .
let us now describe some simplifications that will be adopted in order to solve the hf problem .
the central aim of this work is not to make an exact study of the problem , instead we seek for approximate solutions qualitatively well reflecting the physical properties of la@xmath0cuo@xmath6 and other compounds . following this principle
we take for the wannier orbitals the explicit form given in ( [ base de bloch ] ) . physically
, this means to consider the effective potential created by the environment on each electron of the half filled band , is a quadratic function having a minimum on the copper sites and strongly confining the electrons to it .
this last consideration is similar to the one made on the _ t - j _ one band model .
in this section the main results of this work and their discussion are presented . in subsection [ secc:41 ] the equivalent matrix problem , resulting from projecting the hf system of equations ( [ hf ] ) on the tight binding bloch basis ( [ base de bloch ] ) defined in the previous section is presented .
subsection [ smst ] shows how after imposing the maximal translational symmetry and an @xmath2 and @xmath3 spin nature on the orbitals , the model is able to reproduce the dispersion relation of la@xmath0cuo@xmath6 metal - paramagnetic half filled band of the precise band calculations given in ref . .
the solutions presented in subsections [ solantiferro ] and [ solpara ] , illustrate the consequences of releasing symmetry restrictions on the space in which hf solutions are searched .
the first of them corresponds to a mott s insulator - antiferromagnetic ground state .
that is , the state corresponds to an insulator even tough it has one electron per cell , precisely in the same way as la@xmath0cuo@xmath6 behaves .
the second solution obtained corresponds to a paramagnetic state showing a pseudogap and exactly the same hf energy and set of occupied single particle states , that the previously obtained metal - paramagnetic state@xcite .
it is important to mention that all the band diagrams shown in this section are plotted on the same energy scale and the zero energy point coincides with the fermi level of the antiferromagnetic ground state ( iaf ) presented in subsection [ solantiferro ] .
let the searched single particle states represented in the explicitly nonseparable form @xmath57 where _ l _ is the additional quantum number needed for indexing the stationary state on question , which we are going to define precisely further ahead .
after subsituting ( [ e : representaciontb ] ) , ( [ freehamiltonian ] ) and ( [ f : coulombefectivo ] ) , in ( [ hf ] ) ; followed by projecting the obtained result on the basis @xmath58 and an extensive algebraic work , it is possible to arrive at the following self - consistent matrix problem for the coefficients appearing in the expansion ( [ e : representaciontb ] ) : @xmath59\textbf{.}\textrm{b}^{\textbf{k},l}= \widetilde{\varepsilon}_{l}(\textbf{k } ) \ \textrm{i}_{\textbf{k}}\textbf{.}\textrm{b}^{\textbf{k},l},\end{aligned}\ ] ] where each of the quantities @xmath60 represents a vector having four components given by the four possible pairs @xmath61 .
the constants @xmath62 are dimensionless . in them , _
_ represents the vacuum charge of the electron ; @xmath63 is the reduced planck constant ; _ a _ is the characteristic radius of the wannier orbitals @xmath64 and _ p _ is the nearest neighbors separation .
it is clear now that we can define _
l _ = 1 , 2 , 3 , 4 , as a label indicating each of the four solutions to be obtained for every value of quasi - momentum * k*. also , all the implicit parameters in the following 4@xmath654 matrices are dimensionless @xmath66 the set of quantities ( [ eq18 ] ) constitute the matrix representations of the periodic potential created by the mean field w@xmath30 , the direct and exchange terms in ( [ hf ] ) , the interaction potential with the neutralizing
jellium of charges f@xmath31 defined in ( [ freehamiltonian ] ) and the overlapping matrix among nearest neighbors , respectively .
each one of the four pairs , @xmath67 and @xmath61 defines a row and a column respectively , of the matrix in question .
the explicit forms of the matrix elements are given in appendix [ aelem ] .
in this new representation the normalization condition for the hf single particle states and the hf energy of the system take the forms @xmath68,\end{aligned}\ ] ] where @xmath69 is the heaviside function .
the system ( [ ecumatricial ] ) is non linear on the variables @xmath70 , which are the four components of each vector @xmath71 .
they can be interpreted as a measure of the probability amplitude of finding an electron in the state ( * k * , l ) , in the sublattice @xmath72 , with spin z - projection @xmath73 .
thinking in numerically solving the equations by the method of iterations , it is convenient to pre - multiply them by @xmath74 for each * k*. note that for each * k * four eigenvalues ( _ l _ = 1 , 2 , 3 , 4 . ) will be obtained , or equivalently , four bands on the z.b .
[ e : zb ] ) . from eq .
( [ ecumatricial ] ) it can be observed that in the representation ( [ base de bloch ] ) , the hf potentials and in general the total hamiltonian of the system , resulted as block diagonal with respect the sets of states indexed by the same * k*. this fact is a consequence of the commutation of each one of them with every element of the reduced group of discrete translations .
that is , the group of translations which leaves invariant a sublattice . in this subsection
we will search for hf solutions having their orbitals on the space of bloch functions being eigenfunction of the maximal group of translations leaving invariant the absolute lattice .
in other words , we demand the maximum possible symmetry under translations .
it will follows that our model is capable of acceptably reproduce the profile of the conduction band dispersion calculated in ref . for the la@xmath0cuo@xmath6 .
we will adjust the free parameters of the model in order to reproduce in the best way the band dispersion reported in ref . .
the parameters are : the dielectric constant of the effective environment @xmath29 ; the characteristic radius of the gaussian wannier orbitals @xmath75 ; the jumping probability between nearest sites for an electron @xmath76 ( it is fixed by the effective environment ) and the radius in which the gaussian orbitals associated to the neutralizing jellium of charges decays @xmath77 ( see section [ cuo ] ) .
in what follows the wavy hats will mean dimensionless , see appendix [ aelem]([e : dimensiones ] ) .
let us define the bloch basis for the space of orbitals in which the solution will be searched as @xmath78 where @xmath79 are the quasimomenta of the single particle bloch states which are eigenfunctions of the maximal group of translations
. also is important to define @xmath80 , and * r * , which are the amount of cells in the absolute lattice and the corresponding parametrization ( [ redabsoluta ] ) , respectively
. the functions @xmath81 are the gaussian orbitals defined in section [ tb ] .
the searched hf orbitals will have the form @xmath82 as expressed on the above mentioned basis .
this time , the equivalent matrix problem for the vector
@xmath83 is of second order for each value of ( @xmath84,l ) .
that is , its solutions will be two component vectors .
consequently , _
l _ will take the values 1 or 2 now .
thus , in analogy to ( [ ecumatricial ] ) the new set of equations to be solved results in the form @xmath85\textbf{.}\bar{\textrm{b}}^{\textbf{q},l}= \widetilde{\varepsilon}_{l}(\textbf{q } ) \ \bar{\textrm{i}}_{\textbf{q}}\textbf{.}\bar{\textrm{b}}^{\textbf{q},l}.\end{aligned}\ ] ] let @xmath86 and @xmath87 anyone of the 2x2 matrices in ( [ ecumatricial2 ] ) and its 4x4 equivalent on ( [ ecumatricial ] ) respectively , the relationship between these matrix elements is the following one @xmath88 the direct and exchange potentials respective matrices 2x2 and 4x4 relationship becomes slightly more complicated .
besides also satisfying the above mentioned relation , each vector components in their `` 4x4 '' definitions must be removed from the sublattice label dependence and multiplied by @xmath89 ( those new quantities are the vector components of the 2x2 problem ) .
for instance , making reference to the definitions given in appendix [ aelem ] @xmath90.\end{aligned}\ ] ] .
figure b ) shows the doubly degenerated bands associated to the same paramagnetic and metallic state .
note the close correspondence between these results and the ones obtained by matheiss in ref . .
the zero energy level in all the band diagrams is the fermi energy of the isolator and antiferromagnetic solution presented in subsection [ solantiferro ] .
the domain of the plot is the b.z . of the sublattice shown in fig .
2 a ) . , title="fig:",width=144,height=144 ] .
figure b ) shows the doubly degenerated bands associated to the same paramagnetic and metallic state .
note the close correspondence between these results and the ones obtained by matheiss in ref . .
the zero energy level in all the band diagrams is the fermi energy of the isolator and antiferromagnetic solution presented in subsection [ solantiferro ] .
the domain of the plot is the b.z . of the sublattice shown in fig .
2 a ) . , title="fig:",width=144,height=144 ] analogously to the ones showed in the previous section , in this representation , the normalization condition for the single particle states and the hf energy of the system take the forms @xmath91.\end{aligned}\ ] ] the method employed for solving all the self - consistent matrix problems considered in this work was an iterative one which started from a particular and estimated as convenient state configuration . in the case
being examined in this section , for beginning the iterations , we used a paramagnetic state .
the figure [ f : bm ] shows the paramagnetic , metallic and doubly degenerate band obtained from the iterative process on ( [ ecumatricial2 ] ) .
a half filling condition has been assumed , that is , a state with one electron per cell is being considered .
specifically , for the case of @xmath92 electrons , the occupied states inside the b.z
. are showed in figure [ f : bm ] a ) by the points inside the shadowed region .
the chosen parameters were : @xmath29=10 , which is a common value for semiconductors , @xmath75=0.25 , @xmath77=0.05 and @xmath76=-0.03 ( see appendix [ aelem ] ) , by following the criterium of fixing a bandwidth of 3.8 ev @xcite .
the here obtained band topologically coincides with the conduction band presented in ref . .
in both of them the fermi level on the @xmath93-@xmath94 direction is a square which vertices touch the middle of the b.z .
( of the cuo lattices ) sides and also the maximal and minimal energies lay on coinciding points . therefore , in this subsection the basis of the effective model employed here and its main set of parameters have been defined . as we have stated before ,
the solution of the system of equations eq .
( [ ecumatricial ] ) was performed by the method of successive iterations .
the results which presented from now on were found by employing the parameter values @xmath29 , @xmath75 , @xmath76 and @xmath77 , which were determined in the previous section .
it is important to be noted that @xmath75 , @xmath77 @xmath951 and also that @xmath76 must be of the order of the overlapping among nearest neighbors factor .
it was necessary to start the iteration process from a particular state having an antiferromagnetic character from the beginning , in order to achieve convergence toward the solution presented in this subsection . in the figure
[ f : bandas ] two sets of results following for a half filling band are shown .
they correspond to two lattices of 20x20 and 30x30 cells .
the bands are depicted on the same scale of energies .
the difference between them is of the order of 10@xmath96 dimensionless units of energy @xmath97= 8.3 ev .
evidently , they are bands corresponding to insulating states.the close similarity of both results indicates that the thermodynamical limit has being satisfactorily achieved for the considered sizes of the periodic system .
the hf energy of this hf solution was the lowest among of all the ones found .
in coincidence with the experimental evidence , they are states with local magnetic moment resting on the direction of the sublattice x@xmath98 ( see figure [ f : magnetizacion ] ) . in the next section
we will show the difference between this hf energy of this state and the ones corresponding to the other determined hf solutions and shall comment on this respect .
= 1.32 ev .
b ) a sample of 30x30 cells , e@xmath99 = 1.32 ev .
the parameter values chosen were @xmath75 = 0.25 , @xmath77 = 0.05 , @xmath76 = -0.03 and @xmath29 = 10 .
the zero energy level is fixed in the fermi level of the 20@xmath6520 system .
note that the difference in energy between the two bands is not appreciable in the employed energy scale .
the domains of both plots is the b.z . of the sublattice shown in fig .
2 a ) , title="fig:",width=153,height=153 ] = 1.32 ev .
b ) a sample of 30x30 cells , e@xmath99 = 1.32 ev .
the parameter values chosen were @xmath75 = 0.25 , @xmath77 = 0.05 , @xmath76 = -0.03 and @xmath29 = 10 .
the zero energy level is fixed in the fermi level of the 20@xmath6520 system .
note that the difference in energy between the two bands is not appreciable in the employed energy scale .
the domains of both plots is the b.z . of the sublattice shown in fig .
2 a ) , title="fig:",width=153,height=153 ] one important quantity which has been experimentally measured is the magnetization .
it is therefore motivating to inspect the prediction of the obtained hf state for this magnitude .
its definition is given by the expression @xmath100 where @xmath101 and @xmath102= @xmath103 , @xmath104=@xmath105 and @xmath106=@xmath107 are the pauli matrices .
in figure [ f : magnetizacion ] a ) the only non vanishing component of ( [ e : m ] ) in this solution is plotted .
an interesting result is that it has been experimentally observed that la@xmath0cuo@xmath6 has a magnetic moment of 0.68 @xmath4 per cu site on the cuo plane @xcite .
the value obtained from evaluating the above formula for our hf state turns out to be 0.67 @xmath4 .
therefore , the considered here discussion satisfactorily predicts the whole antiferromagnetic structure of la@xmath0cuo@xmath6 .
[ f : magnetizacion ] of the more stable hf state here determined rests in the direction 1 - 2 .
a ) this figure shows the projection @xmath108 of the dimensionless magnetization , in the 1 - 2 direction .
the magnetization unit is @xmath109 .
b ) the picture shows a scheme of the mean magnetic moment per site in the lattice . for the shown solution
its modular value is 0.67@xmath110.,title="fig:",width=144,height=144 ] of the more stable hf state here determined rests in the direction 1 - 2 .
a ) this figure shows the projection @xmath108 of the dimensionless magnetization , in the 1 - 2 direction .
the magnetization unit is @xmath109 .
b ) the picture shows a scheme of the mean magnetic moment per site in the lattice . for the shown solution its modular value is 0.67@xmath110.,title="fig:",width=144,height=144 ] it is also helpful to remark that the corresponding single particle states carry a more intensive antiferromagnetism as more closer they are from the fermi surface .
therefore , this property offers a clear explanation of the gradual loss observed in the antiferromagnetic order under the doping with holes @xcite . in figure 6 the dependence of the angle @xmath111 between the magnetic moments per cell on each of the sublattice 1 and 2 , shown by each one of the single particle bloch states ,
is plotted .
these components are defined as the integrals of the magnetic moment density over all the unit cells of the _ absolute _ lattice centered in the sublattice points .
note that the states laying just on the fermi surface are perfectly antiferromagnetic ones , and that the more away from the boundaries the orbitals are , the less antiferromagnetic they become .
then , this hf solution indicates that after the orbitals are allowed to spatial dependent spin orientations , the electrons prefer to reorient their spin when traveling between contiguous lattice cells .
this effect that can be interpreted as clean `` correlation '' effect when considered within a restricted hf picture .. ) is plotted against their bloch states quasimomenta .
note that the states on the boundary have a perfect antiferromagnetism and that they become less antiferromagnetic as their quasimomenta move away from the boundary .
the region of the plot is the b.z . of the sublattice shown in fig .
2 a ) ] [ f : angulo ] it also follows that the size of the zone in which the antiferromagnetism is strongest , inversely depends on the dielectric constant .
thus , the less is the coulomb interaction among the electrons in the half filled electron band , smaller becomes the antiferromagnetic character of the single particle states and the region in which the antiferromagnetism accumulate .
only the single particle states staying exactly on the fermi surface remain having a perfect antiferromagnetism .
in addition , the magnitude of the gap also decreases with the increasing of the dielectric constant .
it can be observed that to remove the translational symmetry restrictions does not mean renouncing to obtain a paramagnetic state as solution . as a matter of fact ,
this solution effectively exists and its properties are quite interesting . in section [ sec1:1 ]
we had already commented about la@xmath0cuo@xmath6 and its antiferromagnetic _
mott s insulator _ properties .
however , at an intermediate level of doping , this material presents very special properties . after the breaking of the antiferromagnetic order and in a certain temperature t and doping ranges , the material transits to phases showing a so called pseudogap . in the la@xmath0cuo@xmath6 and other htc materials the presence of this property has been observed in some regions of their paramagnetic phase * pm * and in the superconductor phase * sc * @xcite .
[ f : bandas1 ] in figure [ f : bandas1 ] the band spectra corresponding to the hf paramagnetic ground state obtained from eq .
( [ ecumatricial ] ) is shown .
note the existence of a pseudogap which reaches a maximum value of 0.1 ev @xmath112 10@xmath113 k ( equivalent to 0.012 dimensionless unit of energy @xmath97 = 8.3 ev ) .
the parameters given in the subsection [ smst ] were employed for this evaluation .
the pseudogap amplitude rises for smaller dielectric constants @xmath29 , that is , with the reduction of the screening .
it is an interesting result that the hf energy of this ground state is exactly coincident with the one corresponding to the paramagnetic and metallic state presented in subsection [ smst ] .
moreover , the lowest energy band in both solutions , also exactly coincide , with an upper bound error of @xmath114 in dimensionless energy units ( that means @xmath115 in ev ) .
thus , the occupied single particle states in both solutions are identical and in consequence the momentum dependence of the filled energy bands also coincide . henceforth , the difference between the two solutions refers only to the non occupied states .
those states can not be considered in the band model presented in subsection [ smst ] .
for instance , consider the expressions for two states , one occupied an another empty , which are associated to the same four quasimomentum value @xmath116 where @xmath117 represents the gaussian orbitals @xmath118 centered on @xmath56 . comparing with the expression ( [ estadopmatheiss ] ) , it may be noted that the occupied single particle state ( [ ocupado ] ) can be expressed on the basis ( [ basematheiss ] ) , and in fact , coincides with the state @xmath119 obtained from solving ( [ ecumatricial2 ] ) .
however the excited state ( [ exitado ] ) was not allowed in the space of orbitals employed in solving ( [ ecumatricial2 ] ) , and thus its attainment is a neat consequence of the removing of the usually imposed symmetry under the maximal group of translations .
more precisely , these excited states showing a pseudogap , appeared thanks to the allowed independence between the bloch functions defined in both sublattices , introduced in this work .
this result shows an interesting potential use that this kind of constraints removing procedure could introduce in the determination of new excitation properties of the hf solutions in various applications . in the following table
the hf energies per particle are shown for the paramagnetic - metallic ( pm ) , paramagnetic with pseudogap ( ppg ) and insulator - antiferromagnetic ( iaf ) ground states ; with zero point energy assumed on the last one : .energy differences between the various hf states . [ cols=">,>,>,>",options="header " , ] it is an interesting outcome that the energy difference pm ( ppg)-iaf and the nel temperature of this kind of materials are both of the order of 10@xmath120 k. thus , the results suggest the possibility of having further success in applying the approach started in this work to the description of some regions of the phase diagram of the la@xmath0 cuo@xmath6 . on the question about the relative stability between the pm and ppg states , it can be estimated that due to the presence of a pseudogap , the ppg ground state should become more stable than the pm at non vanishing temperatures .
this should be the case because for creating excitations on the ppg ground state , temperatures of hundreds of kelvin degrees are needed , while excitations in the pm ground state can occur in every range of temperatures .
[ f : bandas2 ] in figure [ f : bandas2 ] the ppg ( pm ) and iaf occupied bands are depicted in a common frame .
the main difference in their energies corresponds to the single particle states being closer to fermi surface .
as we had noted before , the same behavior has the antiferromagnetic character of the single particle states of the iaf solution .
therefore , it can be hoped that from the solution of ( [ ecumatricial2 ] ) under doping with holes , both solutions move toward a common ground state without absolute magnetic order , in correspondence with the pattern shown by the phase diagrams of these materials @xcite .
that is , at a critical doping level value a breaking of the antiferromagnetic order could occur and the system presumably transits to a phase showing a pseudogap .
the results of this work support the potential of the hf self - consistent method for the description of some secular properties exhibited by materials such as copper - oxygen compounds , specifically la@xmath0cuo@xmath6 .
those are properties usually associated to strong correlation effects , that could be explained by _
first principle _ calculations , after modified to incorporate procedures to search for spontaneous symmetry breaking solutions and general ways of exploring the spin structure of the hf obitals .
note that single particle orbitals are not only employed in the hf method .
they are also essential ingredients of more general schemes as the various types of density functional methods .
thus , the kind of reasoning employed in this work seems to be easily easily implemented in such discussions . in the present work we explore those paths in a simplified manner . in order to avoid the intrinsic complexities of la@xmath0cuo@xmath6 material
, it was necessary to utilize a simple model which was sufficiently flexible for reproducing the dispersion profile of the only half filled band of la@xmath0cuo@xmath6 reported in ref . .
after imposing the maximal symmetry under translations a paramagnetic and metallic ground state ( pm ) was obtained which dispersion properties topologically also coincide with the results given in ref . .
then , the free parameters of effective model were fixed from the requirement of reproducing the bandwidth of the half filled band obtained in ref . .
employing those parameters and from removing some symmetry restrictions , solutions were obtained which show interesting properties . in agreement with the experiment , the insulating - antiferromagnetic ( iaf ) solution turns to be the most stable among all hf states found .
some of its properties are enumerated below : 1 .
the gap magnitude diminish with the increasing of the screening constant @xmath29 .
the antiferromagnetic structure the hf orbitals increases when the states approach the fermi level .
in addition , the size of the outlying region in which antiferromagnetism persists depends on the screening created by the effective environment @xmath29 .
that is , by increasing screening , the size of the antiferromagnetic region reduces .
thus , the idea arises that after doping with holes ( that is , solving for hf solutions not at half filling condition as it is done here ) the antiferromagnetic zone , which is precisely concentrated near the fermi level could be annihilated , producing in this way a phase transition to a non magnetically ordered ground state .
this possibility indicates a way to describe the normal state properties of the htc superconductors through a simple hf study .
the magnetic moments per cell which are evaluated rest on the right direction respect to the lattice and its modular value of 0.67 @xmath4 satisfactorily coincides with the experimentally measured result 0.68 @xmath4 .
this outcome support the adjust made of the parameters defining the effective model employed . the other hf solution presented in this work corresponds to a paramagnetic state showing a pseudogap ( ppg ) .
few properties of this state are described in what follows : 1 .
the pseudogap magnitude diminishes with the increasing of the screening constant @xmath29 .
2 . just as we hoped from the beginning of the work , at zero temperature the states pm and ppg resulted identical .
the difference between them is only given by the excitations of the system .
thus , the removal of some symmetry restrictions , defines in this case , new properties for unoccupied single particle states .
it seems feasible that in other materials , it could be possible to obtain a gap instead of a pseudogap even in the absence of magnetic order .
such an outcome could show the ability of a properly formulated hf description in describing all the kinds of mott insulators , being or not magnetically ordered .
the amplitude of the obtained pseudogap reaches a value of the order of 10@xmath120k .
thus , it is possible that the system at finite temperature should prefer the ppg state over the pm one . that is an interesting signal , knowing that la@xmath121m@xmath122cuo@xmath6 presents pseudogap in the * pm * phase .
4 . in the same way as it happens for the antiferromagnetic character of the iaf ground state , the difference between the one particle energies of iaf and the ppg ( pm ) states , is greater for states being closer to the fermi surface .
that is , it happens for the electrons with more energy and consequently the first ones in disappear under doping with holes .
therefore , as it was mentioned before , these results suggest the possibility to describe a transition iaf@xmath123ppg under doping with holes , within the here studied effective model . one methodological conclusion of this work seems to be worth to be mention .
it corresponds to the fact that the results fully clarifies that the solutions of a general hf problem non necessarily should turn to be a set of single particle orbitals , all having an @xmath2 or @xmath124 spin structure at all points of the space .
this is directly shown by the particular example of the iaf solution in which the hf single particle states have a neatly non separable character in their spin and orbital dependence .
this possibility , although being simple , and perhaps even expected to be correct after to direct the attention to it , looks to be relevant for the description of magnetic and strongly correlated systems .
the above enumerated conclusions motivated new objectives to be considered in extending the work .
our general interests are twofold : in one side to achieve a more complete and precise investigation about the here detected unexplored potentialities of the _ first principle _ calculations . in another direction , to look for the possibility of describing the htc superconductivity in the framework of the here studied simple model .
some of the more specific issues of future searches in connection those general objectives are the following ones : * to generalize the discussion done in this work to introduce the doping with holes as a new parameter .
then , it will be possible to investigate the effects of the doping on the here determined ground states .
of particular relevance in this sense appears the iaf state . * to compute the zero temperature electron green function of the system , and use it to evaluate the effective polarization of the la@xmath0cuo@xmath6 in the obtained states . * with the polarization results at hand , attempt to solve the bethe - salpeter equation for two holes in the iaf ground state ( and also in the ppg one ) , to find whether it is possible or not to define the existence of preformed cooper pairs in the model .
the possibility for its existence was suggested by the results of ref . ,
in which it was argued that a strong 2d - screening of the coulomb interaction is created by a half filled band of tight binding electron . * to continue the application of the ideas advanced here in help clarifying the widely known debate between the mott and slater pictures about the electronic structure of solids . * to apply a rotational invariant hf calculation for obtaining states of molecular or atomic systems having incomplete shells .
we guess , that this application could help to reduce the value of the correlation energies in those systems .
this idea is suggested by the results of this work which show that the hf energy can be optimized in many of the currently considered physical systems , due to the difficulties in determining the best among all possible self - consistent solutions .
we express our gratitude to a. gonzlez , c. rodrguez , a. delgado , y. vazquez - ponce and n. g. cabo - bizet by helpful conversations and comments .
in addition , the relevant support received from the proyecto nacional de ciencias bsics ( pncb , citma , cuba ) and from the network n-35 of the office of external activities ( oea ) of the asictp ( italy ) are deeply acknowledged .
from those definitions , the used reference conventions can be inferred .
that is , the distances are expressed in units of the cu nearest neighbors separation _
= 3.6 ; and the energies and potential interactions , in units of the quantity @xmath97 , that for the employed parameters is equivalent to 8.3 ev .
we had already defined along the paper the wannier orbitals @xmath117 , as normalized gaussian functions , centered in @xmath56 and with characteristic parameter _
a_. in order to simplify notation when the jellium in ( [ freehamiltonian ] ) is going to be used , is useful to define : @xmath127 let us use ( [ wannierfondo ] ) for defining the jellium s potential @xmath128 in obtaining the matrix elements of the jellium potential ( [ fondoapendice ] ) , the following notation is useful : where labels _ r _ and _ t _ move on the sublattices independently . in the context of the infinite lattice problem is easy to see the following property @xmath130 which can be proved by making a pair of changes of variables on the second order integral in the right hand side of ( [ notacionint ] ) . for finite systems
the prove is a little more complicated .
firstly , the integrals extend over the region occupied by the lattice , and for this region the integration does not remain invariant under the translations that must be done .
it becomes necessary to periodically extend the coulomb interaction beyond the boundaries .
that is , modified it to the form @xmath131 where @xmath132 and @xmath133 = ... , -1 , 0 , 1 , ... , and @xmath134 is the length of the side of the squared region which the system occupies .
however when the system is sufficiently large , the error done by no extending it periodically is not important ; and it vanishes in thermodynamic limit .
if we define @xmath139 then it follows @xmath140 let us consider the definitions of * * r**@xmath141 and * * r**@xmath142 given in ( [ e : subred ] ) , and recall that the only wannier orbitals which have non vanishing overlapping are those centered on the same site or the ones centered in nearest neighbors ( the closest neighbors belong to different sublattice ) . then for fixed * * r**@xmath143 and * * r**@xmath144 , the only no vanishing among all the quantities in the left hand side of ( [ propdir ] ) are @xmath145 and @xmath146 in which _ i_=1, ... ,4 .
the four quantities * * p**@xmath147 are defined as @xmath148 that is , they move over the neighbors which are closest to the site on the origin .
the employed procedure reduces the number of integrals appearing in ( [ integral1 ] ) , ( [ integral2 ] ) , inclusive those corresponding to the right hand side in ( [ integralfondo ] ) and anyone of those appearing when the matrix elements of the periodic potential w@xmath149 , or the projection of the tight binding bloch basis between any two elements , are searched . by example
@xmath150 which respectively fulfill the following properties @xmath151 given the periodicity of w@xmath149 in the absolute sublattice .
@xmath155,\\\nonumber & & \\
i_{\textbf{k},(t,\alpha_z),(r,\sigma_z)}&=&\delta _ { \alpha_z,\sigma_z } \left[i_{00}\ \delta_{t , r}+2i_{01}(\cos k_1p+\cos k_2p)\ \delta_{t , r+1}\right ] , \\\nonumber & & \\
f_{\textbf{k},(t,\alpha_z),(r,\sigma_z ) } & = & \delta _ { \alpha_z,\sigma_z }
\left[f_{00}\ \delta_{t , r}+2f_{01}(\cos k_1p+\cos k_2p)\ \delta_{t , r+1}\right],\\\nonumber\end{aligned}\ ] ] where @xmath156 = 0 , represents a change in the zero point energy ; and @xmath76 is a free parameter describing our lack of knowledge about the periodic potential .
the other appearing parameters are defined as @xmath157 where _ n _ is the number of electrons in the electron gas and , as we had already defined , the symbol @xmath158 means dimensionless . for simplicity , in expression ( [ e : matrizdirecta ] )
the einstein summation convention for the indices @xmath161 is employed .
further , the matrix elements of the exchange potential are @xmath162,\end{aligned}\ ] ] where @xmath163 \times\langle\widetilde{\textbf{r}}^{(t^{\prime},t^{\prime\prime})},\textbf{0}|\widetilde{v}|\textbf{0},\widetilde{\textbf{r}}^{(t^{\prime},t^{\prime\prime})}\rangle , \\\nonumber&&\\ s^{(\textbf{k},\textbf{k}^\prime , t^{\prime},t^{\prime\prime})}_1&=&\frac{2}{n}\sum_{i}\sum_{\widetilde{\textbf{r}}^{(t^{\prime},t^{\prime\prime})}}\cos[\textbf{k}\cdot\textbf{p}_i+(\textbf{k}-\textbf{k}^\prime)\cdot\ \textbf{r}^{(t^{\prime},t^{\prime\prime } ) } ] \times\langle\widetilde{\textbf{p}}_i+\widetilde{\textbf{r}}^{(t^{\prime},t^{\prime\prime})},\textbf{0}|\widetilde{v}|\textbf{0},\widetilde{\textbf{r}}^{(t^{\prime},t^{\prime\prime})}\rangle , \\\nonumber&&\\ s^{(\textbf{k},\textbf{k}^\prime , t^{\prime},t^{\prime\prime})}_3&=&\frac{2}{n}\sum_{i , j}\sum_{\widetilde{\textbf{r}}^{(t^{\prime},t^{\prime\prime})}}\cos[\textbf{k}\cdot(\textbf{p}_i+\textbf{p}_j)+(\textbf{k}-\textbf{k}^\prime)\cdot\ \textbf{r}^{(t^{\prime},t^{\prime\prime } ) } ] \times\langle\widetilde{\textbf{p}}_i+\widetilde{\textbf{r}}^{(t^{\prime},t^{\prime\prime})},\widetilde{\textbf{p}}_j|\widetilde{v}|\textbf{0},\widetilde{\textbf{r}}^{(t^{\prime},t^{\prime\prime})}\rangle.\end{aligned}\ ] ] any one of the fourth fold integrals presented in previous sections , can be partially integrated in quadratures , in such a way they final calculation is reduced to numerically evaluate first order integrals . thus @xmath164}{2\sqrt{2\pi\widetilde{a}^2}}\times\int^{2\pi}_{0}d\phi\ \exp\{-\frac{[(\widetilde{r}_{x_1}+\frac{1}{2})\sin\phi-(\widetilde{r}_{x_2}-\frac{1}{2})\cos\phi]^2}{2\widetilde{a}^2}\}\\\nonumber & \times&\text { erfc}\{-\frac{[(\widetilde{r}_{x_2}-\frac{1}{2})\sin\phi-(\widetilde{r}_{x_1}+\frac{1}{2})\cos\phi]^2}{2\widetilde{a}^2}\},\end{aligned}\ ] ] where erfc is the complement error function .
similarly @xmath165}{\sqrt{\frac{4(1+\zeta^2 ) \pi\widetilde{a}^2}{\zeta^2}}}\times\int^{2\pi}_{0}d\phi\ \exp\{-\frac{[(\widetilde{r}_{x_1}+\frac{1}{2})\sin\phi-\widetilde{r}_{x_2}\cos\phi]^2}{\frac{(1+\zeta^2)\widetilde{a}^2}{\zeta^2 } } \}\\\nonumber & \times&\text { erfc}\{-\frac{[\widetilde{r}_{x_2}\sin\phi-(\widetilde{r}_{x_1}+\frac{1}{2})\cos\phi]^2}{\frac{(1+\zeta^2)\widetilde{a}^2}{\zeta^2 } } \},\end{aligned}\ ] ] | it is argued that a hartree - fock ( hf ) solution for coulomb interacting electrons in a simple model of the cu - o planes in la@xmath0cuo@xmath1 , is able to predict some of the most interesting properties of this material , such as its insulator character and the antiferromagnetic order .
moreover , the natural appearance of pseudogaps in their paramagnetic and superconductor phases are also suggested by the discussion .
these results follow after the elimination of some symmetry restrictions which are usually imposed on the single particle hf orbitals
. one of them is the simplification of their spinor dependence to be of the so called @xmath2 or @xmath3 types .
this constraint , seems to strongly reduce the largest space of orbitals corresponding to the rotational invariant hf formulation originally introduced by dirac .
we also remove , the demand on the hf orbitals of having a bloch structure in the bravais lattice of the crystal .
this procedure allows for the consideration of hf solutions having the same symmetry than the known ( symmetry breaking ) antiferromagnetic structure of the material .
it turns out that the most stable hf solution of the problem corresponds to an antiferromagnetic and insulating state ( iaf ) , which magnetization rests on the experimentally determined direction .
encouragingly , the evaluated magnetic moment per cell is 0.67 @xmath4 , a result that satisfactorily reproduces the experimentally measured value : 0.68 @xmath4 .
another hf solution having a slightly higher energy arises and corresponds to a paramagnetic state showing a pseudogap ( ppg ) .
it follows after only imposing the bloch stucture on the single particle states .
finally a third paramagnetic but metallic solution ( no gap ) is also obtained by including both of the mentioned restrictions .
the interesting result follows that this state only differs from the ppg pseudogap state in the form of excited empty orbitals .
that is , the occupied single particle states and the hf energy at t = 0 for both solutions , are identical . in general , the discussion helps to clarify the role of the antiferromagnetic correlations in the structure of the physics of the htsc materials .
in addition , these initial results , indicate a promising way for start conciliating the mott and slater pictures in the physics of the strongly correlated electron systems . |
one of the principal problems in solar physics is understanding how the sun s corona is heated to very high temperatures .
recent work on coronal loops indicates that they have physical properties that are difficult to reconcile with theoretical models .
coronal loops with temperatures near 1mk are observed to persist longer than a characteristic cooling time , suggesting steady or quasi - steady heating ( e.g. , * ? ? ?
* ; * ? ? ?
steady heating models , however , can not reproduce the high electron densities observed in these loops @xcite .
multi - thread , impulsive heating models have been proposed as a possible heating scenario ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
such models are motivated by our understanding of the energy release during magnetic reconnection in flares ( e.g. , * ? ? ?
* ) . in these models
impulsive heating leads to high densities and multiple , sub - resolution `` threads '' lead to long lifetimes relative to the cooling time for an individual loop .
these models are severely constrained by the relatively narrow distributions of temperatures that are often observed in loops with apex temperatures near 1mk ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
a narrow distribution of temperatures suggests that the loop can contain only a few independent threads .
one difficulty with fully testing coronal heating scenarios such as these with hydrodynamic models has been the spareness of data .
previous work on loop evolution has generally focused on measurements imaging instruments ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , which have limited diagnostic capabilities .
current solar observatories , however , allow for coronal loops to be observed in unprecedented detail . the euv imaging spectrometer ( eis ) on the _ hinode _ mission provides high spatial and spectral resolution observations over a very wide range of coronal temperatures .
eis plasma diagnostics yield important constraints on the physical properties of coronal loops .
the x - ray telescope ( xrt ) on _ hinode _ complements these observations with high spatial and temporal resolution observations of the high temperature corona .
the multiple viewpoints of the twin _ stereo _ spacecraft allow for loop geometry , a critical parameter in the modeling , to be measured using the euv imagers ( euvi ) .
the _ transition region and coronal explorer _ ( _ trace _ ) currently provides the highest spatial resolution images of the solar corona . in this paper
we use _ stereo _ , _ hinode _ , and _ trace _ observations of an evolving loop in a post - flare loop arcade to make quantitative comparisons between a multi - thread , impulsive heating model and measured densities , temperatures , intensities and loop lifetimes .
an important component of this work is the development of methods for integrating the different observations into hydrodynamic simulations of the loop .
we find that it is possible to reproduce the extended loop lifetime , the high electron density , and the narrow differential emission measure ( dem ) with a multi - thread model provided the time scale for the energy release is sufficiently short .
the model , however , does not reproduce the evolution of the high temperature emission observed with xrt .
one goal of investigating the heating on individual loops is to motivate the modeling of entire active regions or even the full sun ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
it is possible , however , that there is not a single coronal heating mechanism that can be applied to all coronal loops .
for example , it may be that steady heating is the dominant heating scenario on some fraction of coronal loops ( e.g. , * ? ? ? * ; * ? ? ? * ) . even if impulsive heating of the kind discussed here is only a minor contributor to the heating of the solar corona , this study provides important insights into the energy release during magnetic reconnection , a fundamental process in astrophysical and laboratory plasmas .
in this section we provide an overview of the instruments and observations used in this study .
a summary of the observations is shown in figure [ fig : summary ] .
the loop considered here is a post - flare loop from a very small event ( goes class b2.5 ) that peaked around 19:00 ut on may 2 , 2007 . the eis instrument on _ hinode _ produces stigmatic spectra in two wavelength ranges ( 171212 and 245291 ) with a spectral resolution of 0.0223 .
there are 1 and 2 slits as well as 40 and 266 slots available .
the slit - slot mechanism is 1024 long but a maximum of 512 pixels on the ccd can be read out at one time .
solar images can be made using one of the slots or by stepping one of the slits over a region of the sun .
telemetry constraints generally limit the spatial and spectral coverage of an observation .
see @xcite and @xcite for more details on the eis instrument . for these observations
the 1 slit was stepped over the active region and 15s exposures were taken at each position .
an area of @xmath0 was imaged in about 71 minutes .
a total of 20 spectral windows were read out of the ccd and included in the telemetry stream .
the raw data were processed using ` eis_prep ` to remove dark current , warm pixels , and other instrumental effects using standard software . during the processing
the observed count rates are converted to physical units .
intensities from the processed data are computed by fitting the observed line profiles with gaussians .
the eis rasters are co - aligned to account for any spatial offsets ( see @xcite for a discussion ) .
spacecraft jitter during the raster has not been accounted for .
hinode _ housekeeping logs suggest relatively small displacements ( less than one pixel ) for the narrow field of view of interest here . for larger structures
spacecraft jitter can be important .
eis rasters in a number of different emission lines are shown in figure [ fig : eis ] , and show post - flare loops at various temperatures in the lower part of the active region .
these rasters also indicate a brief data gap due to orbital eclipse .
one limitation of these eis data is the lack of temporal information .
better information on the temporal evolution of these loops is provided by the imaging instruments , such as the euvi @xcite on the _ solar terrestrial relations observatory _ ( _ stereo _ ) mission .
the euvi is a normal incidence , multilayer telescope which can observe the sun in 4 wavelength bands centered at 284 , 195 , 171 , and 304 .
euvi observes the full sun and therefore has reduced spatial resolution ( 1.6 pixels ) relative to the other observations that we consider here .
there are two _ stereo _ spacecraft with identical instrument packages .
the twin _ stereo _ spacecraft drift away from the earth at about 23@xmath1 per year . on may 2 , 2007
the separation between the spacecraft was small , about 6@xmath1 .
the euvi images taken around the time of the eis raster are indicated in figure [ fig : summary ] . because of telemetry constraints the image cadence is limited . for these observations
171 images were taken at a relatively high cadence ( @xmath2s ) while the images at the other wavelengths were taken at lower cadences ( @xmath31200s ) .
the raw data are processed using ` euvi_prep ` to produce calibrated , co - aligned images .
images of the active region and flare from _ stereo b _ euvi are shown in figure [ fig : euvi ] .
the xrt on _ hinode _ is a high cadence , high spatial resolution ( approximately 1 pixels ) grazing incidence telescope that images the sun in the soft x - ray and extreme ultraviolet wavelength ranges .
temperature discrimination is achieved through the use of focal plane filters . because xrt can observe the sun at short wavelengths
, xrt images can observe high temperature solar plasma very efficiently .
the thinner xrt filters allow longer wavelength euv emission to be images and extend the xrt response to lower temperatures .
further details on xrt are given in @xcite .
as indicated in figure [ fig : summary ] , the principal xrt images taken during this time period were in three filters , ti - poly , al - thick , and be - thick , at a variable cadence .
unfortunately , the exposure times on the be - thick images are too short for the images to be used for analyzing active region loops .
the standard processing routine ` xrt_prep ` is used to remove the ccd bias , dark current , and calibrate the images .
images are also `` dejittered '' using _ hinode _ spacecraft housekeeping data so that the images are co - aligned with respect to each other .
_ hinode _ tracks solar rotation so there is no need to account for it in the images .
example ti - poly images are shown in figure [ fig : xrt ] .
these images have a field of view of @xmath4 .
the _ trace _ instrument is a high resolution normal incidence telescope .
the primary and secondary mirrors are divided into quadrants and a rotating shutter is used to select which quadrant is illuminated .
three of the quadrants are coated with multilayers for imaging at euv wavelengths .
the multilayer coatings have peak sensitivities at approximately 171 , 195 , and 284 .
the fourth quadrant is coated with aluminum and magnesium fluoride for imaging very broad wavelength ranges near 1216 , 1550 , 1600 , and 1700 .
images in all of the wavelengths are projected onto a single detector , a @xmath5 ccd .
each ccd pixel represents a solar area approximately 0.5 on a side .
the instrument is described in detail by @xcite .
the initial in - flight performance is reviewed by @xcite and @xcite . during this period
_ trace _ observed mainly in the 171 channel at a cadence of about 60s with occasional 1600 and white light context images .
all of the images have a @xmath4 field of view .
as shown in figure [ fig : summary ] , there are periodic data gaps in the trace data due to orbital eclipses .
all of the _ trace _ images are processed using a standard application of ` trace_prep ` .
additionally , the images are despiked and co - aligned with respect to each other to account for solar rotation and drifts in the pointing .
a simple cross - correlation method is used for this purpose . the _ trace _
171 images are very similar in appearance to the euvi 171 images .
the primary goal of this study is to compare multi - thread , hydrodynamic simulations with the emission observed in an evolving coronal loop . as we will discuss in more detail later in the paper ,
hydrodynamic simulations involve solving the equations for the conservation of mass , momentum , and energy in the loop given some input heating rate .
relating the heating rate to physical observables is a critical element of the modeling .
our strategy is to measure the electron density with eis and use a family of hydrodynamic simulations to infer the required heating rate for this density .
previous numerical simulations suggest that for a fixed loop length there is a power law relationship between the peak electron density and the input energy @xcite .
since we want the loop length to be fixed , the other critical element of this modeling is an accurate measurement of the loop geometry , including the inclination .
observations from the twin euvi instruments allow the loop geometry to be measured , and we use the _ stereo _ software package developed for this purpose @xcite .
once the density and loop geometry have been determined we can perform hydrodynamic simulations and synthesize the expected emission during the entire evolution of the loop .
the simulation results can then be compared with light curves determined from _ trace _ and xrt and in this section we discuss how these light curves are calculated .
the distribution of temperatures in the loops is an important constraint on the modelings and in this section we also discuss the calculation of the differential emission measure distribution with the eis spectra . our analysis requires the identification of loops observed simultaneously with both eis and _ trace_. to facilitate this we wrote routines to co - align eis rasters with _ trace _ images and to display 24 bit color images using an eis raster for one color channel and a trace image for another color channel .
an animation of these images allowed us to quickly identify times when the eis slit was co - spatial with a loop observed with _
_ trace _ s small field of view and frequent data gaps due to orbital eclipse make finding good data sets more difficult than anticipated . to optimize the co - alignment between eis and _ trace
_ we had to allow for a roll angle between the images .
this is in addition to the usual spatial shifts between the pointing information contained in the data headers . since the _ trace _ data was taken in the /x 171 channel we used the eis 184.536 raster for co - alignment .
the eis intensities for the loop of interest are summarized in figure [ fig : eis_ints ] .
once the region of interest was identified we manually selected spatial positions along the loop in the eis 195.119 raster .
these points are used as spline knots to define the loop coordinate system ( @xmath6 ) , with @xmath7 along the loop and @xmath8 perpendicular to it ( see @xcite figure 3 ) . since these loop coordinates are not necessarily aligned to the ccd we have interpolated to determine the intensities along the selected segment .
the loop segments displayed in figure [ fig : eis_ints ] are interpolated to 0.2 per pixel . using the loop
coordinate system it is a simple matter to compute the intensity averaged along the loop segment .
the coordinate system derived from 195.119 is used for all of the eis rasters . to further isolate the intensity of the loop we identify two background points and fit a single gaussian with a linear background to the selected region .
background subtraction is essential to separating the intensity in the loop from the contribution of the ambient corona , but there is no unique method for computing it .
analysis of euvi data , which has the advantage of providing two different lines of sight for a single loop , suggests that the background subtracted intensities can be computed consistently , although there can be considerable uncertainties for individual measurements @xcite .
since we are interested in emission that is co - spatial we extract the same region from all of the eis rasters .
the resulting intensities are shown in figure [ fig : eis_ints ] . to test how co - spatial
the emission is at various temperatures , we have calculated a simple correlation coefficient between the background subtracted intensity in each line and 195.119 , which represents a middle ground between the highest and lowest temperature emission that is observed . for this loop
well correlated , co - spatial emission is observed for and below . for the emission at higher temperatures ( xvi )
the correlation is poor .
it is clear from the images shown in figures [ fig : eis ] and [ fig : eis_ints ] that the combination of high spatial resolution and good temperature discrimination allows eis to probe the interrelationship of emission at different temperatures .
the loop intensities suggest that the emission at different temperatures is generally not co - spatial and that the dem in the loop of interest should be relatively narrow .
this is shown more clearly in the `` multicolor '' image of this region presented in figure [ fig : eis_color ] .
this image , which is a 24 bit image formed from rasters in 3 different emission lines , would be white in regions where the dem is broad and the emission is strong in all three lines .
there are some composite regions that are cyan ( green @xmath9 blue ) , but the post - flare loop arcade is generally dominated by emission in the primary colors suggesting relatively narrow distributions of temperature in each loop .
to investigate the temperature structure of this loop more quantitatively we compute the differential emission measure using the background subtracted loop intensities for the loop segment .
the intensities are related to the differential emission measure by the usual expression @xmath10 where @xmath11 is the plasma emissivity and @xmath12 is the differential emission measure distribution .
we consider a gaussian dem model @xmath13,\ ] ] which allows for a dispersion in the temperature distribution .
since the density is an important parameter in determining the emissivities of many of these lines we leave it is a free parameter . to determine the best - fit parameters ( @xmath14 , @xmath15 , @xmath16 , @xmath17 ) we use a levenberg - marquardt technique implemented in the ` mpfit ` package
the chianti 5.2 atomic physics database ( e.g. , @xcite ) is used to calculate the emissivities .
the abundances of @xcite and the low density ionization fractions of @xcite are assumed .
there are several subtleties to computing the emission measure parameters .
one is the statistical uncertainty associated with each intensity . because the intensities are averaged over a significant area the statistical uncertainties are generally small .
the systematic errors introduced by the background subtraction and the atomic data , however , are large , but difficult to estimate .
we simply assume that the relative error is 20% of the measured intensity .
another question is how to deal with the emission from lines such as 262.984 , that do not show evidence for the loop and have no measured intensity .
for these lines the background subtracted intensity is zero .
observations of these lines provide important constraints on the high temperature component of the dem and must be included . to account for possible errors in computing
the background subtracted intensities for these lines we assume that the uncertainty in the intensity is 20% of the background instead of 20% of the loop intensity . since the background can be large this represents a substantial enhancement of the uncertainty .
the resulting dem for this loop segment is shown in figure [ fig : eis_ints ] .
the observed and computed intensities are given in table [ table : ints ] .
from this analysis we obtain an electron density of @xmath18 . for this work we use the 203.826/202.044 lines to provide the bulk of the density sensitivity . recently @xcite have noted systematic discrepancies between the various density sensitive line ratios from and . in light of this
we compared the densities inferred from 203.826/202.044 and 186.880/195.119 with those from 258.375/261.058 in a series of other active region and quiet sun observations .
the lines , which were not included in this study , are relatively weak and not sensitive over as large a range as the and lines . however , the atomic data for is potentially more reliable than the atomic data available for the complex fe ions .
we find that the densities derived from the and ratios are in excellent agreement and emphasize the ratio here .
the densities inferred from 186.880/195.119 are as much as a factor of 3 higher .
details of this analysis will be presented in a future paper . the dispersion in the temperature is found to be @xmath19 , which is comparable to other active region loop observations with eis @xcite .
these lines observed in the quiet corona above the limb indicate much narrower temperature distributions ( @xmath20 , @xcite ) .
here we find a dispersion in temperature that is several times greater , indicating that this loop is not strictly isothermal .
the relatively intense , co - spatial emission observed from both and provides the best direct evidence for a distribution of temperatures in this loop .
these lines have peak temperatures of formation that are about 1mk apart .
the application of a delta function emission measure to equation [ eq : ints ] confirms that a single temperature model can not adequately reproduce the intensities observed in these emission lines .
in searching for coronal loops observed with both eis and _ trace _ we co - aligned the eis rasters with the _ trace _ images for this period .
thus to compute the _ trace _ intensities we use the same coordinate system and apply it to all of the _ trace _ images taken during the time of interest . since the co - alignment between eis and _ trace _ is not perfect ,
the spline knots selected in the eis raster are modified slightly to better align with the loop .
background subtracted intensities are computed for each of the available _ trace _ images taken during this time using the same procedure that was used on the eis data . for each image the region defined by the spline knots
was extracted , straightened , and averaged along the loop coordinate to produce the intensity as a function of the perpendicular coordinate .
the intensity at each time refers to the background subtracted intensity integrated over the loop .
the _ trace _ light curve is shown in figure [ fig : trace_ints ] . for most of this period the background subtracted loop intensities are in the noise and the evolution of this loop can be seen clearly .
we have manually selected the region around the peak in the light curve and fit it with a single gaussian .
this fitting yields a gaussian width of about 294s . to extract the three dimensional geometry of this loop we use the software package developed by markus aschwanden and the _ stereo _ team for this purpose .
an application of this software is discussed in detail in @xcite and @xcite .
the initial processing co - registers images from _ stereo a _ and _ b _ to account for differences in spacecraft roll angle and spatial resolution .
the next step is to outline the loop in the _ stereo a _ image .
the selected coordinates are projected onto the corresponding _
b _ image . since geometry of the loop has not yet been determined this projection yields a range of possible coordinates in the _ b _ image and the user selects the position of the loop within this range .
once the coordinates of the loop have been selected in both images the geometry of the loop is determined from simple trigonometric relationships . due to the gap in the _ a _ data , an image pair at 21:16 ut , after the peak emission observed in this loop ,
is used , so there is some ambiguity in the loop identification .
the selection of nearby structures in the loop arcade generally yield very similar results .
the projection of the extracted loop geometry onto the euvi _ b _ image available at the peak of the loop emission ( 20:59 ut ) also outlines the observed loop very well .
thus it does not appear that the calculation of the loop geometry is significantly impacted by the data gap .
the selected loop and the projection of this loop in various planes is shown in figure [ fig : euvi_stereo ] .
images taken with the xrt provide information on the evolution of the loop at high temperatures .
we have computed an xrt light curve similar to that computed for _
trace_. to use the loop coordinates derived from eis we first co - align the xrt images with the eis 262.984 raster , which shares some common features .
we then compute the background subtracted intensities by selecting two background points and doing a linear fit .
the emission seen earlier in the event is much broader than what is observed trace and so we select a wider area to compute the intensities .
the resulting xrt light curve is shown in figure [ fig : xrt_ints ] . as is indicated by the light curve ,
the xrt images clearly show strong emission in the region that is eventually occupied by the loops observed with eis and _ trace_. it is also clear , however , that xrt does not show any individual loops that are as narrow as those that are seen at cooler temperatures .
consistent with this , the loop cross section measured with xrt is systematically wider than what is measured with eis and _ trace_. to illustrate these differences we have constructed multicolor images from xrt and _ trace_. these images , which are presented in figure [ fig : xrt_trace ] , use different color channels to display the xrt and _ trace _ data in the same picture . to illustrate the differences in morphology as the plasma cools
we have offset the times of the selected images by one hour .
we show examples of the loop cross sections in figure [ fig : xrt_trace_ints ] .
this difference between high temperature and low temperature emission in flares and active regions is well documented ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , and may be further evidence for fine scale structure in the solar corona .
we will discuss this point in more detail later in the paper .
one of the primary paradoxes of coronal loops with temperatures near 1mk is the disparity between the rapid cooling suggested by the high electron densities and the relatively long observed lifetimes .
eis density diagnostics allow us to make more rigorous comparisons between these time scales .
densities inferred from observed intensities requires an accurate determination of the differential emission measure , the absolute instrumental calibration , and the loop geometry .
this measurement also represents a lower bound on the density . the density inferred from density sensitive
line ratios circumvents many of these problems .
if we assume that the loop is cooling only through radiation the energy balance is simply @xmath21 where @xmath22 is the radiative loss function for an optically thin plasma . in the limit of no flows
the energy is @xmath23 .
the plasma pressure given by @xmath24 , with @xmath25 is the boltzmann constant .
the radiative cooling time is defined as @xmath26 and is given by @xmath27 using the temperature and density derived from the eis dem analysis ( @xmath28 and @xmath29 ) and a radiative loss rate of @xmath30erg @xmath31 s@xmath32 @xcite we obtain a radiative cooling time of 341s .
the radiative cooling time is not directly comparable to the loop lifetime that we have measured with _
trace_. if we make the additional assumptions that @xmath33 and @xmath34 we can relate the radiative cooling time ( @xmath35 ) to the timescale for changes in the temperature ( @xmath36 ) @xmath37 numerical simulations suggest that @xmath38 @xcite . finally , to compare with the observed loop lifetime
we must incorporate the temperature changes into the trace temperature response curve .
this yields @xmath39\ ] ] where @xmath40 is the peak temperature of the trace response and @xmath41 is the gaussian width of the trace response ( see @xcite equation 9 ) .
this yields @xmath42 with @xmath43mk and @xmath44mk @xcite we obtain @xmath45s , which is smaller than the observed gaussian width of 294s .
this mismatch between the predicted and observed loop lifetime is one of the key motivations for the multi - thread modeling of coronal loops . by assuming that the observed emission comes from a series of loops that are heated at different times it is clear that we can create a composite loop with the required lifetime .
the challenge is to also match the relatively narrow dem and the lifetime of the loop as it is observed with xrt . to simulate the evolution of this loop we consider numerical solutions to the full hydrodynamic loop equations using the nrl solar flux tube model ( solftm ) .
we adopt many of the same parameters and assumptions that were used in previous simulations with this code and we refer the reader to the earlier papers for additional details on the numerical model ( e.g. , @xcite ) .
for example , we assume that the loop is symmetric and only simulate the evolution over half of the loop length .
we also assume that each loop has a constant cross section . for this work
we consider a heating function for each thread that is a simple spatially uniform heating rate @xmath46 where @xmath47 is a triangular envelope that peaks at time @xmath48 and has width @xmath49 , which we set to be 100s .
the parameter @xmath50 is a small background heating rate that provides the initial loop equilibrium . as we mentioned earlier ,
our strategy for inferring the heating rates from the observations is to use the density determined from eis and the results from systematic hydrodynamic simulations .
this will determine the peak heating rate for the ensemble of threads .
the heating rate for the other threads will be determined by assuming a gaussian envelope for the heating function . to determine the relationship between the input heating rate and the densities and temperatures observed during the cooling phase of the loop evolution ,
we have performed 21 simulations with @xmath51 varying between @xmath52 and @xmath53erg @xmath54 s@xmath32 . using the parameters derived from the euvi observations ,
the loop length is fixed at 135 mm and the loop inclination is fixed at 68.5@xmath1 . for each simulation
we average over the loop apex to determine a representative density and temperature .
the simulation results are summarized in figure [ fig : apex ] .
we find that the apex density at 1.3mk , which is the peak temperature in the dem , essentially scales as @xmath55 .
this implies that for a fixed loop length the observed intensity is linearly proportional to the input energy .
the relationship shown in figure [ fig : apex ] indicates that a heating rate of 0.8erg @xmath54 s@xmath32 is required to reproduce an apex density of @xmath56 .
the density - heating rate relationship is valid for a single loop . for a multi - thread simulation
we assume the heating rate for each thread is related to this peak heating rate by @xmath57,\ ] ] where @xmath58 determines the duration of the heating envelope .
the parameter @xmath59 is chosen so that all of the times are positive .
the heating events are spaced so that as the heating in the previous loop ends the heating in the next loop begins .
this heating scenario is illustrated in figure [ fig : eflare ] for @xmath60 , 200 , and 300s .
once the individual hydrodynamic simulations are run , we average over the loop apex at each time step to compute a representative temperature and density . these densities and temperatures
are then used as inputs to the _ trace _ temperature response to calculate the expected count rates in the _ trace _ 171 , 195 , and 284 channels as function of time .
the simulation times are shifted so that the peak in the 171emission corresponds to the observed peak .
since we are not interested in resolving the differences in absolute calibration among the various instruments , we also introduce a scaling factor so that the peak simulated emission matches what is observed .
the simulation results are shown in figure [ fig : trace_sim ] and indicate that @xmath61s simulation , which yields a simulated loop lifetime of @xmath62s , best matches the observations .
we also use the simulated densities and temperatures as a function of time to compute the expected intensities in many of the emission lines that can be observed with eis .
light curves from selected emission lines are shown in figure [ fig : eis_sim ] . since the absolute time for the simulation
has been established through the comparisons with _ trace _ , we select the simulated eis intensities that correspond to the time of the eis observations and use them as inputs to the same differential emission measure code that was used to produce figure [ fig : eis_ints ] . the resulting simulated dem is shown in figure [ fig : eis_sim ] .
the simulated intensities are given in table [ table : ints ] .
the agreement between the observed and simulated differential emission measure is relatively good .
the simulation captures the salient features of the observations , a relatively high density and a narrow dem .
for these simulated intensities we do obtain a somewhat lower electron density ( @xmath63 for the simulation and 9.7 for the observation ) and peak temperature ( @xmath64 for the simulation and 6.11 for the observation ) .
the dispersion in the dem also does not match the observation exactly ( @xmath65 and 5.48 ) . given the approximate nature of the simulations we consider these discrepancies to be small .
the difference in the density comes about because we have inferred the heating rate from a family of single - loop hydrodynamic simulations but the emission is actually a composite from several threads .
it is likely that iterating on this solution would yield better agreement with the observations , but this is unlikely to yield addition physical insights . finally , we have simulated the expected xrt emission for the open / ti - poly filter combination using the standard xrt software routine ` xrt_t_resp ` .
the simulated and observed light curves are shown in figure [ fig : xrt_sim ] .
this comparison presents the greatest challenge to the modeling .
the modeled composite intensities , which have been scaled to match the observations at 20:00 ut , clearly do not extend back in time enough to cover the entire evolution of the emission in this region .
the peak observed emission in this loop occurs at approximately 19:30 ut , before the simulation has even begun .
given the diffuse nature of the xrt emission and the difficulty of isolating individual loops at high temperatures perhaps the simplest explanation may be that the xrt light curve includes the contributions of many loops in addition to the loop we isolated using the eis and _ trace _ data .
the identification of individual loops at very high temperatures with xrt is likely to be hampered by the slow evolution of plasma during the conductive cooling phase .
this is evident by the slow evolution of the threads in the simulation . in figures [ fig : eis_sim ] and [ fig : xrt_sim ] , for example , we see that the threads last for approximately 1 hour at high temperatures . at the lower temperatures , when radiative losses are much higher , the cooling is dominated by radiation and the evolution is much faster . in these simulations
the threads last only for about 10 minutes in the _ trace _
171 bandpass .
this difference suggests that the differentiated loops seen at lower temperatures , such as those illustrated in figure [ fig : eis_color ] , would appear as a single structure in xrt .
these differences are also related to the broad temperature response of xrt .
the relatively narrow line emission imaged with eis and _ trace _ emphasizes small differences in temperature .
alternatively , the inability of the model to reproduce the observed xrt emission may reflect inadequacies with the hydrodynamic simulations during the conductive phase of the cooling .
it may be that the heating is not as impulsive as we have assumed .
many previous studies have suggested a gradual decay in the heating ( e.g. , * ? ? ?
these differences in the assumed heating may be related to the changes in the topology of the magnetic field during the evolution of the event .
observations of post - flare loop arcades have shown that newly reconnected field lines relax from cusp - shaped to approximately semi - circular during the early phases of the cooling @xcite .
the comparisons between the hot and cool emission shown in figure [ fig : xrt_trace ] clearly suggests that field line shrinkage is occurring in this event .
the cool post - flare loops observed with _ trace _ are generally observed at the lowest heights of the arcade and do not overlap with the high temperature xrt emission seen at the top of the arcade .
the implications of field line shrinkage on hydrodynamic simulations has not been investigated . in general , the conversion of magnetic energy into thermal energy through the process of magnetic reconnection is not well understood ( see , for example , @xcite ) .
more detailed analysis of mhd simulations is needed to better understand the evolution of coronal loops after reconnection ( e.g. , * ? ? ?
in this paper we have made use of the unprecedented opportunity to observe evolving coronal loops in detail . we have used _ stereo _ , eis , _ trace _ , and xrt data to constrain a multithread model of coronal heating and compare with observations .
these comparisons indicate that it is possible to reproduce the high densities , long lifetimes , and relatively narrow emission measure distributions inferred from the data so long as the heating envelope of the heating is sufficiently narrow .
the most challenging comparisons are with xrt , where the model fails to reproduce the extended lifetime of the emission at high temperatures .
it is not clear if this is due to our inability to isolate narrow loops at high temperatures or to problems with the assumed envelope on the heating .
recent analysis of the eis spectral range has identified , xv , xvi , and xvii emission lines that can be used in the analysis of high temperature plasma @xcite .
these lines will provide additional information on plasma evolution during the conductive phase and have been incorporated into the latest eis observing sequences .
new active region observations should be available during the rise of the next solar cycle .
ultimately our goal is to apply the multithread modeling described here to non - flaring active region loops .
it is encouraging that the simulated eis differential emission measure curve derived here is similar to those derived for lower density active region loops @xcite .
it remains to be seen , however , that this models can also reproduce the loop evolution observed in _ trace _ and xrt .
the launch of the atmospheric imaging assembly ( aia ) on the _ solar dynamics observatory _
( _ sdo _ ) , which will combine full disk imaging , _
trace_-like spatial resolution , 10s cadences , and multiple filters , will greatly expand the number of useful active region observations that combine eis plasma diagnostics and loop evolution .
the authors would like to thank jim klimchuk for helpful discussions on the time scales for radiative cooling , and markus aschwanden for assistance with the stereo loop geometry code .
amy winebarger contributed significant improvements to the interface to the hydrodynamic code .
hinode is a japanese mission developed and launched by isas / jaxa , with naoj as domestic partner and nasa and stfc ( uk ) as international partners .
it is operated by these agencies in co - operation with esa and nsc ( norway ) . , s. , antiochos , s. k. , & klimchuk , j. a. 2002 , in esa special publication , vol .
505 , solmag 2002 .
proceedings of the magnetic coupling of the solar atmosphere euroconference , ed .
h. sawaya - lacoste , 207 , 200 , and 300s .
the simulation times have been adjusted so that the peak emission in 171 matches what is observed .
the magnitude of the simulated emission has also been scaled to match the observations .
the lifetime of the loop is calculated from a gaussian fit to the composite light curve .
the @xmath66s simulation most closely matches to observed loop lifetime . ]
lrrrr 275.352 & 37.2 & 37.4 & 66.7 & 61.2 + 184.536 & 190.3 & 194.4 & 339.5 & 342.6 + 188.216 & 436.9 & 357.9 & 439.7 & 496.7 + 195.119 & 718.6 & 853.2 & 718.6 & 773.7 + 202.044 & 595.5 & 545.1 & 301.2 & 277.1 + 203.826 & 207.9 & 200.3 & 120.7 & 120.9 + 274.203 & 0.0 & 106.1 & 33.9 & 28.7 + 284.160 & 0.0 & 296.1 & 36.9 & 31.2 + 262.984 & 0.0 & 2.71 & 0.1 & 0.1 + [ table : ints ] | the high densities , long lifetimes , and narrow emission measure distributions observed in coronal loops with apex temperatures near 1mk are difficult to reconcile with physical models of the solar atmosphere .
it has been proposed that the observed loops are actually composed of sub - resolution `` threads '' that have been heated impulsively and are cooling .
we apply this heating scenario to nearly simultaneous observations of an evolving post - flare loop arcade observed with the euvi/_stereo _ , xrt/_hinode _ , and _ trace _ imagers and the eis spectrometer on _
hinode_. we find that it is possible to reproduce the extended loop lifetime , high electron density , and the narrow differential emission measure with a multi - thread hydrodynamic model provided that the time scale for the energy release is sufficiently short .
the model , however , does not reproduce the evolution of the very high temperature emission observed with xrt . in xrt
the emission appears diffuse and it may be that this discrepancy is simply due to the difficulty of isolating individual loops at these temperatures .
this discrepancy may also reflect fundamental problems with our understanding of post - reconnection dynamics during the conductive cooling phase of loop evolution . |
every physically relevant computational model must be mapped into physical space - time and _ vice versa _
@xcite . in this line of thought , von neumann s self - reproducing cellular automata @xcite have been envisioned by zuse @xcite and other researchers @xcite as `` calculating space ; '' i.e. , as a locally connected grid of finite automata @xcite capable of universal algorithmic tasks , in which intrinsic @xcite observers are embedded @xcite .
this model is conceptually discreet and noncontinuous and resolves the eleatic `` arrow '' antinomy @xcite against motion in discrete space by introducing the concept of information about the state of motion in between time steps .
alas , there is no direct physical evidence supporting the assumption of a tessellation of configuration space or time . given enough energy , and without the possible bound at the planck length of about @xmath0 m
, physical configuration space seems to be potentially infinitely divisible .
indeed , infinite divisibility of space - time has been utilized for proposals of a kind of `` zeno oracle '' @xcite , a progressively accelerated turing machine @xcite capable of hypercomputation @xcite .
such accelerated turing machines have also been discussed in the relativistic context @xcite .
in general , a physical model capable of hypercomputation by some sort of `` zeno squeezing '' has to cope with two seemingly contradictory features : on the one hand , its infinite capacities could be seen as an obstacle of evolution and therefore require a careful analysis of the principal possibility of motion in finite space and time _ via _ an infinity of cycles or stages . on the other hand ,
the same infinite capacities could be perceived as an advantage , which might yield algorithms beyond the turing bound of universal computation , thus extending the church - turing thesis .
the models presented in this article unify the connectional clarity of von neumann s cellular automaton model with the requirement of infinite divisibility of cell space . informally speaking ,
the scale - invariant cellular automata presented `` contain '' a multitude of `` spatially '' and `` temporally '' ever decreasing copies of themselves , thereby using different time scales at different layers of cells .
the cells at different levels are also capable to communicate , i.e. , exchange information , with these copies , resulting in ever smaller and faster cycling cells .
the second model is based on petri nets which can enlarge themselves .
the advantage over existing models of accelerated turing machines which are just turing machines with a geometrically progression of accelerated time cycles resides in the fact that the underlying computational medium is embedded into its environment in a uniform and homogeneous way . in these new models , the entire universe , and not just specially localized parts therein , is uniformly capable of the same computational capacities .
this uniformity of the computational environment could be perceived as one further step towards the formalization of continuous physical systems @xcite in algorithmic terms . in this respects ,
the models seem to be closely related to classical continuum models , which are at least in principle capable of unlimited divisibility and information flows at arbitrary small space and time dimensions . at present however , for all practical purposes , there are finite bounds on divisibility and information flow . to obtain a taste of some of the issues encountered in formalizing this approach , note that an infinite sequence of ever smaller and faster cycling cells leads to the following situation . informally speaking ,
let a _ self - similar cellular automaton _
be a variant of a one - dimensional elementary cellular automaton , such that each cell is updated twice as often as its left neighbor .
the cells of a self - similar cellular automaton can be enumerated as @xmath1 .
starting at time 0 and choosing an appropriate time unit , cell @xmath2 is updated at times @xmath3 .
remarkably , this definition leads to indeterminism . to see this ,
let @xmath4 be the state of cell @xmath5 at time @xmath6 .
now , the state @xmath7 depends on @xmath8 , which itself depends on @xmath9 and so on , leading to an infinite regress . in general , in analogy
to thomson s paradox @xcite , this results in an undefined or at least nonunique and thus indeterministic behavior of the automaton .
this fact relates to the following variant of zeno s paradox of a runner , according to which the runner can not even get started @xcite .
he must first run to the half way point , but before that he must run half way to the half way point and so on indefinitely . whereas zeno s runner can find rescue in the limit of convergent real sequences
, there is no such relieve for the discrete systems considered .
later on , two restrictions on self - similar automata ( build from scale - invariant cellular automata ) are presented , which are sufficient conditions for deterministic behavior , at least for finite computations .
furthermore , a similar model based on a variant of petri nets will be introduced , that avoids indeterminism and halts in the infinite limit , thereby coming close to the spirit of zeno s paradox .
the article is organized as follows .
section [ sec - tm ] defines the turing machine model used in the remainder of the article , and introduces two hypercomputing models : the accelerated and the right - accelerated turing machine . in section [ chap : sica ] self - similar as well as scale - invariant cellular automata are presented .
section [ chap : hypercomputer ] is devoted to the construction of a hypercomputer based on self - similar cellular automata .
there is a strong resemblance between this construction and the right - accelerated turing machine , as defined in section [ sec - tm ] .
a new computing model , the self - similar petri net is introduced in section [ chap : petri ] .
this model features a step - to - step equivalence to self - similar cellular automata for finite computations , but halts in the infinite case .
the same construction as in section [ chap : hypercomputer ] is used to demonstrate that self - similar petri nets are capable of hypercomputation .
the final section contains some concluding remarks and gives some directions for future research .
the turing machine is , beside other formal systems that are computationally equivalent , the most powerful model of classical computing @xcite .
we use the following model of a turing machine @xcite . formally , a _
turing machine _ is a tuple @xmath10 , where @xmath11 is the finite set of states , @xmath12 is the finite set of tape symbols , @xmath13 is the set of input symbols , @xmath14 is the start state , @xmath15 is the blank , and @xmath16 is the set of final states . the next move function or transition function @xmath17 is a mapping from @xmath18 to @xmath19 , which may be undefined for some arguments .
the turing machine @xmath20 works on a tape divided into cells that has a leftmost cell but is infinite to the right .
let @xmath21 .
one step ( or move ) of @xmath20 in state @xmath22 and the head of @xmath20 positioned over input symbol @xmath23 consists of the following actions : scanning input symbol @xmath23 , replacing symbol @xmath23 by @xmath24 , entering state @xmath25 and moving the head one cell either to the left ( @xmath26 ) or to the right ( @xmath27 ) . in the beginning , @xmath20 starts in state @xmath28 with a tape that is initialized with an input word @xmath29 , starting at the leftmost cell , all other cells blank , and the head of @xmath20 positioned over the first symbol of @xmath30 .
we need sometimes the function @xmath17 split up into three separate functions : @xmath31 .
the configuration of a turing machine @xmath20 is denoted by a string of the form @xmath32 , where @xmath33 and @xmath34 . here
@xmath22 is the current state of @xmath20 , @xmath35 is the tape content to the left , and @xmath36 the tape content to the right of the head including the symbol that is scanned next .
leading and trailing blanks will be omitted , except the head has moved to the left or to the right of the non - blank content .
let @xmath32 and @xmath37 be two configurations of @xmath20 .
the relation @xmath38 states that @xmath20 with configuration @xmath32 changes in one step to the configuration @xmath37 .
the relation @xmath39 denotes the reflexive and transitive closure of @xmath40 .
the original model of a turing machine as introduced by alan turing contained no statement about the time in which a step of the turing machine has to be performed .
in classical computation , a `` yes / no''-problem is therefore decidable if , for each problem instance , the answer is obtained in a finite number of steps .
choosing an appropriate time scheduling , the turing machine can perform infinitely many steps in finite time , which transcends classical computing , thereby leading to the following two hypercomputing models .
the concept of an accelerated turing machine was independently proposed by bertrand russell , ralph blake , hermann weyl and others ( see refs .
@xcite ) .
an accelerated turing machine is a turing machine which performs the @xmath41-th step of a calculation in @xmath42 units of time .
the first step is performed in time 1 , and each subsequent step in half of the time before . since @xmath43
, the accelerated turing machine can perform infinitely many steps in finite time .
the accelerated turing machine is a hypercomputer , since it can , for example , solve the halting problem , see e.g. , ref . @xcite .
if the output operations are not carefully chosen , the state of a cell becomes indeterminate , leading to a variation of thomson s lamp paradox .
the open question of the physical dynamics in the limit reduces the physical plausibility of the model .
the following model of a hypercomputing turing machine has a different time scheduling , thereby avoiding some of the paradoxes that might arise from the previous one .
let the cells of the tape be numbered from the left to the right @xmath44 .
a right - accelerated turing machine is a turing machine that takes @xmath42 units of time to perform a step that moves the head from cell @xmath45 to one of its neighbor cells .
[ th - right - acc - tm ] there exists a right - accelerated turing machine that is a hypercomputer .
let @xmath46 be a universal turing machine .
we construct a turing machine @xmath47 that alternates between simulating one step of @xmath46 and shifting over the tape content one cell to the right .
we give a sketch of the construction , ref .
@xcite contains a detailed description of the used techniques .
the tape of @xmath47 contains one additional track that is used to mark the cell that is read next by the simulated @xmath46 .
the finite control of @xmath47 is able to store simultaneously the state of the head of @xmath46 as well as a tape symbol of @xmath46 .
we assume that the input of @xmath46 is surrounded by two special tape symbols , say @xmath48 . at the start of a cycle , the head of @xmath47 is initially positioned over the left delimiter @xmath48 .
@xmath47 scans the tape to the right , till it encounters a flag in the additional track that marks the head position of @xmath46 . accessing the stored state of @xmath46
, @xmath47 simulates one step of @xmath46 thereby marking either the left or the right neighbor cell as the cell that has to be visited next in the simulation of @xmath46 .
if necessary , a blank is inserted left to the right delimiter @xmath48 , thereby extending the simulated tape of @xmath46 . afterwards
the head of @xmath47 moves to the right delimiter @xmath48 to start the shift over that is performed from the right to the left .
@xmath47 repeatedly stores the symbols read in its finite control and prints them to the cell to the right . after the shift over ,
the head of @xmath46 is positioned over the left delimiter @xmath48 which finishes one cycle .
we now give an upper bound of the cycle time .
let @xmath41 be the number of cells , from the first @xmath48 to the second one .
without loss of generality we assume that @xmath49 contains the left @xmath48 .
@xmath47 scans from the left to the right and simulates one step of @xmath46 which might require to go an additional step to the left . if cell @xmath50 is to be read next , the head of @xmath46 can not move to the right , otherwise it would fall off the tape of @xmath46 . therefore the worst case occurs if the cell @xmath51 is marked as cell that @xmath46 has to be read next . in this case
we obtain @xmath52 .
the head of @xmath47 is now either over cell @xmath53 , or over cell @xmath45 if a insertion was performed .
the shift over visits each cell @xmath54 three times , and @xmath49 two times .
therefore the following upper bound of the time of the shift over holds : @xmath55 .
we conclude that if the cycle started initally in cell @xmath45 it took less than time @xmath56 .
if @xmath46 halts on its input , @xmath47 finishes the simulation in a time less than @xmath57 .
@xmath47 therefore solves the halting problem of turing machines .
we remark that if @xmath46 does not halt , the head of @xmath47 vanishes in infinity , leaving a blank tape behind .
a right - accelerated turing machine is , in contrast to the accelerated one , in control over the acceleration .
this can be used to transfer the result of a computation back to slower cells .
the construction of an infinite machine , as proposed by davies @xcite , comes close to the model of a right - accelerated turing machine , and his reasoning shows that a right - accelerated turing machine could be build within a continuous newtonian universe .
_ cellular automata _ are dynamical systems in which space and time are discreet . the states of cells in a regular lattice are updated synchronously according to a local deterministic interaction rule .
the rule gives the new state of each cell as a function of the old states of some `` nearby '' neighbor cells .
each cell obeys the same rule , and has a finite ( usually small ) number of states . for a more comprehensive introduction to cellular automata
, we refer to refs .
@xcite .
a _ scale - invariant cellular automaton _ operates like an ordinary _ cellular automaton _ on a cellular space , consisting of a regular arrangement of cells , whereby each cell can hold a value from a finite set of states . whereas the cellular space of a cellular automaton consists of a regular one- or higher dimensional lattice , a scale - invariant cellular automaton operates on a cellular space of recursively nested lattices which can be embedded in some euclidean space as well .
the time behavior of a scale - invariant cellular automaton differs from the time behavior of a cellular automaton : cells in the same lattice synchronously change their state @xcite , but as cells are getting smaller in deeper nested lattices , the time steps between state changes in the same lattice are assumed to _ decrease _ and approach zero in the limit .
thereby , a finite speed of signal propagation between adjacent cells is always maintained .
the scale - invariant cellular automaton model gains its computing capabilities by introducing a local rule that allows for interaction between adjacent lattices @xcite .
we will introduce the scale - invariant cellular automaton model for the one - dimensional case , the extension to higher dimensions @xcite is straightforward .
a scale - invariant cellular automaton , like a cellular automaton , is defined by a cellular space , a topology that defines the neighborhood of a cell , a finite set of states a cell can be in , a time model that determines when a cell is updated , and a local rule that maps states of neighborhood cells to a state .
we first define the cellular space of a scale - invariant cellular automaton . to this end , we make use of standard interval arithmetic . for a scalar @xmath58 and a ( half - open ) interval @xmath59
set : @xmath60 and @xmath61 .
we denote the unit interval @xmath62 by @xmath63 . the cellular space @xmath64 , the set of all cells of the scale - invariant cellular automaton , is the set @xmath65 .
the neighborhood of a cell @xmath66 is determined by the following operators @xmath67 .
for a cell @xmath68 in @xmath64 let @xmath69 be the left neighbor , @xmath70 the right neighbor , @xmath71 the parent , @xmath72 the left child , and @xmath73 the right child of @xmath66 .
the predicate @xmath74 is true if and only if the cell @xmath66 is the left child of its parent .
the cellular space @xmath64 is the union of all lattices @xmath75 , where @xmath76 is an integer .
this topology is depicted in fig .
[ fig:1-dim - interaction ] . for notational
convenience , we introduce a further operator , this time from @xmath64 to @xmath77 , that maps a cell to its both child cells : @xmath78 .
we remark that according to the last definition for each cell either @xmath74 or @xmath79 is true .
later on , we will consider scale - invariant cellular automata where not each cell has a parent cell . if @xmath68 is such a cell , we set by convention @xmath80 if @xmath81 , otherwise @xmath82 .
all cells in lattice @xmath83 are updated synchronously at time instances @xmath84 where @xmath5 is an integer .
the time interval between two cell updates in lattice @xmath83 is again a half - open interval @xmath85 and the cycle time , that is the time between two updates of the cell , is therefore @xmath86 .
a simple consequence of this time model is that child cells cycle twice as fast and the parent cell cycle half as fast as the cell itself .
the time scale @xmath87 is the set of all possible time intervals , which is in the one - dimensional case equal to the set @xmath64 : @xmath88 .
the temporal dependencies of a cell are expressed by the following time operators @xmath89 . for a time
inverval @xmath90 let @xmath91 , @xmath92 , @xmath93 , and @xmath94 .
the predicate @xmath95 is true if and only if the state change of a cell at the beginning of @xmath6 occurs simultaneously with the state change of its parent cell .
the usage of time intervals instead of time instances , has the advantage that a time interval uniquely identifies the lattice where the update occurs .
[ fig : timeops ] depicts the temporal dependencies of a cell : to the left it shows a coupled state change , to the right an uncoupled one .
we remark that we denoted space and time operators by the same symbols , even if their mapping is different . in applying these operators
, we take in the remainder of this paper care , that the context of the operator is always clearly defined . at any time
, each cell is in one state from a finite state set @xmath96 .
the cell state in a given time interval is described by the state function @xmath97 , which maps cells and time intervals to the state set .
the space - time scale @xmath98 of the scale - invariant cellular automaton describes the set of allowed pairs of cells and time intervals : @xmath99 .
then , the state function @xmath100 can be expressed as a mapping @xmath101 .
the local rule describes the evolution of the state function .
for a cell @xmath66 and a time interval @xmath6 , where @xmath102 is in @xmath98 , the evolution of the state is given by the local rule @xmath103 of the scale - invariant cellular automaton @xmath104 in accordance with the definition , the expanded form of a expression of the kind @xmath105 is @xmath106 .
the local rule @xmath103 is a mapping from @xmath107 to @xmath96 . beside the dependencies on the states of the neighbor cells ,
the new state of the cell further depends on whether the cell is the left or the right child of its parent cell and whether the state change is coupled or uncoupled to the state change of its parent cell .
formally , a scale - invariant cellular automaton @xmath108 is denoted by the tuple @xmath109 .
there are some simplifications of the local rule possible , if one allows for a larger state set .
for instance , the values of the predicates @xmath110 and @xmath111 could be stored as substate in the initial configuration .
if the local rule accordingly updates the value of @xmath111 , the dependencies on the boolean predicates could be dropped from the local rule .
as noted in the introduction the application of the local rule in its general form might lead to indeterministic behavior .
the next subsection introduces two restrictions of the general model that avoid indeterminism at least for finite computations .
a special case of the local rule is a rule of the form @xmath112 , which is the constituting rule of a one - dimensional 3-neighborhood cellular automaton . in this case , the scale - invariant cellular automaton splits up in a sequence of infinitely many nonconnected cellular automata .
this shows that the scale - invariant cellular automaton model is truly an extension of the cellular automaton model and allows us to view a scale - invariant cellular automaton as an infinite sequence of interconnected cellular automata .
we now examine the signal speed that is required to communicate state changes between neighbor cells . to this end , we select the middle point of a cell as the source and the target of a signal that propagates the state change of a cell to one of its neighbor cells .
a simple consideration shows that the most restricting cases are the paths from the space time points @xmath113 , @xmath114 , @xmath115 to @xmath116 if not @xmath95 .
the simple calculation delivers the results @xmath117 , and @xmath118 , respectively , hence a signal speed of 1 is sufficient to deliver the updates in the given timeframe .
a more general examination takes also the processing time of a cell into account .
if a cell in @xmath83 takes time @xmath119 to process their inputs and if we assume a finite signal speed of @xmath120 , the cycle time of a cell in @xmath83 must be at least @xmath121 . in sum , as long as the processing time is proportional to the diameter of a cell , we can always find a scaling factor @xmath122 , such that the scale - invariant cellular automaton has cycle times that conform to the time scale @xmath87 .
the construction of a hypercomputer in section [ chap : hypercomputer ] uses a simplified version of a scale - invariant cellular automaton , which we call a self - similar cellular automaton . a _ self - similar cellular automaton _ has the cellular space @xmath123 , the time scale @xmath124 , and the finite state set @xmath96 .
the space - time scale of a self - similar cellular automaton is the set @xmath125 .
the self - similar cellular automaton has the following local rule : for all @xmath126 @xmath127 the local rule @xmath103 is a mapping from @xmath128 to @xmath96 .
formally , a self - similar cellular automaton @xmath108 is denoted by a tuple @xmath109 . by restricting the local rule of a scale - invariant cellular automaton
, a self - similar cellular automaton can also be constructed from a scale - invariant cellular automaton .
consider a scale - invariant cellular automaton whose local rule does not depend on the cell neighbors @xmath129 , @xmath130 , and @xmath131 .
then , the resulting scale - invariant cellular automaton contains the self - similar cellular automaton as subautomaton .
we introduce the following notation for self - similar cellular automata .
we index a cell @xmath132 by the integer @xmath133 , that is a cell with index @xmath76 has a cyle time of @xmath134 .
we call the cell @xmath135 the upper neighbor and the cell @xmath136 the lower neighbor of cell @xmath76 . time instances can be conveniently expressed as a binary number .
if not stated otherwise , we use the cycle time of cell 0 as time unit .
we noted already in the introduction that the evolution of a scale - invariant cellular automaton might lead to indeterministic behavior .
we offer two solutions , one based on a special quiescent state , the other one based on a dynamically growing lattice .
a state @xmath22 in the state set @xmath96 is called a quiescent state with regard to the short - circuit evaluation , if @xmath137 , where the question mark sign `` @xmath138 '' either represents an arbitrary state or a boolean value , depending on its position .
whenever a cell is in state @xmath22 , the cell does not access its lower neighbor .
the cell remains as long in the quiescent state as long as the upper neighbor is in the quiescent state , too .
this modus of operandi corresponds to the short - circuit evaluation of logical expressions in programming languages like c or java .
if the self - similar cellular automaton starts in an initial configuration of the form @xmath139 at cell @xmath140 , the infinite regress is interrupted , since cell @xmath141 evolves to @xmath22 without being dependent on cell @xmath142 .
let @xmath22 be a state in the state set @xmath96 , called the quiescent state .
a dynamically growing self - similar cellular automaton initially starts with the finite set of cells @xmath143 and the following boundary condition .
whenever cell @xmath140 or the cell with the highest index @xmath76 is evolved , the state of the missing neighbor cell is assumed to be @xmath22 .
the self - similar cellular automaton dynamically appends cells to the lower end when needed : whenever the cell with the highest index @xmath76 enters a state that is different from the quiescent state , a new cell @xmath136 is appended , initialized with state @xmath22 , and connected to the cell @xmath76 . to be more specific :
if @xmath76 is the highest index , and cell @xmath76 evolves at time @xmath144 to state @xmath145 , a new cell @xmath136 in state @xmath22 is appended .
the cell performs its first transition at time @xmath146 , assuming state @xmath22 for its missing lower neighbor cell .
we note that the same technique could also be applied to append upper cells to the self - similar cellular automaton , although in the remainder of this paper we only deal with self - similar cellular automata which are growing to the bottom .
both enhancements ensure a deterministic evaluation either for a configuration where only a finite number of cells is in a nonquiescent state or for a finite number of cells . a configuration of a self - similar cellular automaton @xmath108 is called finite if only a finite number of cells is different from the quiescent state .
let @xmath147 be a finite configuration and @xmath148 the next configuration in the evolution that is different to @xmath147 .
@xmath148 is again finite .
we denote this relationship by @xmath149 .
the relation @xmath150 is again the reflexive and transitive closure of @xmath151 .
a self - similar cellular automaton as a scale - invariant cellular automaton can not halt by definition and runs forever without stopping .
the closest analogue to the turing machine halting occurs , when the configuration stays constant during evolution .
such a configuration that does not change anymore is called final .
in this section , we shall construct an accelerated turing machine based on a self - similar cellular automaton . a self - similar cellular automaton which simulates the turing machine @xmath47 specified in the proof of theorem [ th - right - acc - tm ] in a step - by - step manner
is a hypercomputer , since the resulting turing machine is a right - accelerated one .
we give an alternative construction , where the shift over to the right is directly embedded in the local rule of the self - similar cellular automaton .
the self - similar cellular automaton will simultaneously simulate the turing machine and shift the tape content down to faster cycling cells .
the advantages of this construction are the smaller state set as well as a resulting faster simulation .
let @xmath10 be an arbitrary turing machine .
we construct a self - similar cellular automaton @xmath152 that simulates @xmath20 as follows .
first , we simplify the local rule by dropping the dependency on @xmath153 , obtaining @xmath154 the state set @xmath96 of @xmath155 is given by @xmath156 we write @xmath157 for an element @xmath158 in @xmath159 , @xmath160 for an element @xmath161 in @xmath18 , and @xmath162 for an element @xmath163 in @xmath164 . to simulate @xmath20 on the input @xmath165 in @xmath166 , @xmath167
, @xmath155 is initialized with the sequence @xmath168 starting at cell 0 , all other cells shall be in the quiescent state @xmath169 .
if @xmath170 , @xmath155 is initialized with the sequence @xmath171 , and if @xmath172 , the empty word , @xmath155 is initialized with the sequence @xmath173 .
we denote the initial configuration by @xmath174 , or by @xmath175 if we want to emphasize the dependency on the input word @xmath30 .
the computation is started at time 0 , i.e. the first state change of cell @xmath76 occurs at time @xmath134 . the elements @xmath176 and @xmath177 act as head of the turing machine including the input symbol of the turing machine that is scanned next . to accelerate the turing machine , we have to shift down the tape content to faster cycling cells of the self - similar cellular automaton , thereby taking care that the symbols that represent the non - blank content of the turing machine tape are kept together .
we achieve this by sending a pulse , which is just a symbol from a subset of the state set , from the left delimiter @xmath178 to the right delimiter @xmath179 and back .
each zigzag of the pulse moves the tape content one cell downwards and triggers at least one move of the turing machine .
furthermore a blank is inserted to the right of the simulated head if necessary .
the pulse that goes down is represented by exactly one element of the form @xmath180 , or @xmath181 , the upgoing pulse is represented by the element @xmath182 . the specification of the values for the local rule @xmath103 for all possible arguments is tedious , therefore we use the following approach . a coupled transition of two neighbor cells can perform a simultaneous state change of the two cells . if the state changes of these two neighbor cells is independent of their other neighbors , we can specify the state changes as a transformation of a state pair into another one .
let @xmath183 be elements in @xmath96 .
we call a mapping of the form @xmath184 a block transformation . the block transformation
@xmath184 defines a function mapping of the form @xmath185 and @xmath186 for all @xmath187 in @xmath96 .
furthermore , we will also allow block transformations that might be ambiguous for certain configurations .
consider the block transformations @xmath184 and @xmath188 that might lead to an ambiguity for a configuration that contains @xmath189 . instead of resolving these ambiguities in a formal way
, we will restrict our consideration to configurations that are unambiguous .
the evolution of the self - similar cellular automaton @xmath155 is governed by the following block transformations : 1 .
_ pulse moves downwards . _
set @xmath190 @xmath191 @xmath192 if @xmath193 set @xmath194 @xmath195 @xmath196 if @xmath197 set @xmath198 @xmath199 @xmath200 set @xmath201 @xmath202 @xmath203 2 .
_ pulse moves upwards_. set @xmath204 @xmath205 @xmath206 if to a certain cell no block transformation is applicable the cell shall remain in its previous state .
furthermore , we assume a short - circuit evaluation with regard to the quiescent state : @xmath207 , whereby the lower neighbor cell is not accessed . [ cols="^,^,^,^,^,^ " , ] we illustrate the working of @xmath155 by a simple example .
let @xmath208 be the formal language consisting of strings with @xmath41 0 s , followed by @xmath41 1 s : @xmath209 .
a turing machine that accepts this language is given by @xmath210 @xcite with the transition function depicted in fig .
[ fig : example - delta ] .
note that @xmath208 is a context - free language , but @xmath20 will serve for demonstration purposes .
the computation of @xmath20 on input @xmath211 is given below : @xmath212 fig .
[ fig : example - hyper - sca-2 ] depicts the computation of @xmath155 on the turing machine input 01 .
the first column of the table specifies the time in binary base .
@xmath155 performs 4 complete pulse zigzags and enters a final configuration in the fifth one after the turing machine simulation has reached the final state @xmath213 .
[ fig : evolution ] depicts the space - time diagram of the computation .
it shows the position of the left and right delimiter ( gray ) and the position of the pulse ( black ) .
we split the proof that @xmath155 is a hypercomputer into several steps .
we first show that the block transformations are well - defined and the pulse is preserved during evolution . afterwards
we will prove that @xmath155 simulates @xmath20 correctly and we will show that @xmath155 represents an accelerating turing machine .
let @xmath214 be the set of elements that represent the downgoing pulse , @xmath215 be the singleton that contains the upgoing pulse , @xmath216 , and @xmath217 the remaining elements .
the following lemma states that the block transformations are unambiguous for the set of configurations we consider and that the pulse is preserved during evolution .
if the finite configuration @xmath147 contains exactly one element of @xmath218 then the application of the block transformations [ tr : start - state ] [ tr : up - lhd ] is unambiguous and at most one block transformation is applicable .
if a configuration @xmath148 with @xmath219 exists , then @xmath148 contains exactly one element of @xmath218 as well .
note that the domains of all block transformations are pairwise disjoint .
this ensures that for all pairs @xmath220 in @xmath221 at most one block transformation is applicable . block transformations [ tr : start - state ] [ tr : new - blank ] are all subsets or elements of @xmath222 , block transformation [ tr : reflection - right ] is element of @xmath223 , block transformations [ tr : up ] and [ tr : up - state ] are subsets of @xmath224 , and finally block transformation [ tr : up - lhd ] is element of @xmath225 .
since the domain is either a subset of @xmath226 or @xmath227 the block transformations are unambiguous if @xmath147 contains at most one element of @xmath218
. a configuration @xmath148 with @xmath219 must be the result of the application of exactly one block transformation . since each block transformation preserves the pulse
, @xmath148 contains one pulse if and only if @xmath147 contains one .
we introduce a mapping @xmath228 that aims to decode a self - similar cellular automaton configuration into a turing machine configuration .
let @xmath147 be a finite configuration .
then @xmath229 is the string in @xmath230 that is formed of @xmath147 as following : 1 . all elements in @xmath231 are omitted .
all elements of the form @xmath157 are replaced by @xmath23 and all elements of the form @xmath232 or @xmath233 are replaced by the two symbols @xmath22 and @xmath23 .
all other elements of the form @xmath23 are added as they are .
4 . leading or trailing blanks of the resulting string are omitted .
the following lemma states that @xmath155 correctly simulates @xmath20 .
let @xmath50 , @xmath51 be configurations of @xmath20 . if @xmath234 , then there exist two finite configurations @xmath235 , @xmath236 of @xmath155 such that @xmath237 , @xmath238 , and @xmath239 . especially if the initial configuration @xmath174 of @xmath155 satisfies @xmath240
, then there exists a finite configuration @xmath236 of @xmath155 , such that @xmath238 and @xmath241 .
if @xmath50 has the form @xmath242 we consider without loss of generality @xmath243 .
therefore let @xmath244 .
if @xmath245 or @xmath246 and @xmath247 we choose @xmath248 .
if @xmath246 and @xmath249 we insert an additional blank : @xmath250 . in any case @xmath251 holds .
we show the correctness of the simulation by calculating a complete zigzag of the pulse for the start configuration : @xmath252 .
the number of the block transformation that is applied , is written above the derivation symbol .
we split the zigzag up into three phases . 1
. pulse moves down from the left delimiter to the left neighbor cell of the simulated head .
+ for @xmath253 we obtain @xmath254 + if @xmath255 the pulse piggybacked by the left delimiter @xmath256 is already in the left neighbor cell of the head and this phase is omitted .
2 . downgoing pulse passes the head .
+ if in the beginning of the zigzag the head was to the right of the left delimiter then @xmath257 if @xmath258 no further block transformation is applicable and the configuration is final .
the case @xmath259 will be handled later on .
we now continue the derivation [ der : start ] . if @xmath260 then @xmath261 + if @xmath262 then @xmath263 we distinguish two cases : @xmath245 and @xmath246 .
if @xmath245 then @xmath264 if the next steps of @xmath20 are moving the head again to the right , block transformation [ tr : right-2 ] will repeatedly applied , till the head changes its direction or till the head is left of the right delimiter @xmath179 . if the turing machine @xmath20 changes its direction before the right delimiter is reached , we obtain @xmath265 or if the direction change happens just before the right delimiter then @xmath266 if @xmath267 or if the right - moving head hits the right delimiter the derivation has the following form @xmath268 which inserts a blank to the right of the simulated head .
downgoing pulse is reflected and moves up . + we proceed from configurations of the form @xmath269 .
then @xmath270 which finishes the zigzag .
note that the continuation of derivations [ der : right ] and [ der : new - blank ] is handled by the later part of derivation [ der : up ] .
we also remark that the zigzag has shifted the whole configuration one cell downwards .
all block transformations except transformations [ tr : right-2 ] and [ tr : left-1 ] keep the @xmath228-value of the configuration unchanged .
block transformations [ tr : right-2 ] and [ tr : left-1 ] correctly simulate one step in the calculation of the turing machine @xmath20 : if @xmath271 , @xmath272 , and @xmath273 then @xmath274 .
let @xmath275 be the resulting configuration of the zigzag .
we conclude that @xmath276 holds .
we have chosen @xmath235 in such a way that at least one step of @xmath20 is performed , if @xmath20 does not halt , either by block transformation [ tr : right-2 ] or [ tr : left-1 ] . if @xmath20 does not halt the configuration after the zigzag is again of the form @xmath252 . the case @xmath246 and
@xmath277 is excluded by derivation [ der : new - blank ] , which inserts a blank to the right of the head , if @xmath277 .
this means that @xmath275 has the same form as @xmath235 and that any subsequent zigzag will perform at least one step of @xmath20 as well if @xmath20 does not halt .
in summary , we conclude that @xmath155 reaches after a finite number of zigzags a configuration @xmath236 such that @xmath238 . on the other hand , if @xmath20 halts , @xmath155 enters a final configuration since derivations [ der : tm - step - left ] or [ pulse - passed ] are not applicable anymore and the pulse can not cross the simulated head .
since we have chosen @xmath174 to be of the same form as @xmath235 in the beginning of the proof , the addendum of the lemma regarding the initial configuration is true .
next , the time behavior of the self - similar cellular automaton @xmath155 will be investigated .
let @xmath278 be a finite configuration of @xmath155 that starts in cell @xmath76 .
if @xmath20 does not halt , the zigzag of the pulse takes 3 cycles of cell @xmath76 and @xmath155 is afterwards in a finite configuration @xmath279 that starts in cell @xmath136 . without loss of generality , we assume that the finite configuration starts in cell 0 .
we follow the zigzag of the pulse , thereby tracking all times , compare with fig .
[ fig : example - hyper - sca-2 ] and fig .
[ fig : evolution ] .
the pulse reaches at time 1 cell 1 , and at time @xmath280 cell 2 .
in general , the downgoing pulse reaches cell @xmath281 in time @xmath282 . at time
@xmath283 the cell @xmath141 changes to @xmath181 which marks the reversal of direction of the pulse .
the next configuration change ( @xmath284 ) occurs at @xmath285 .
the pulse @xmath182 reaches cell @xmath286 in time @xmath287 and in general cell @xmath281 in time @xmath288 . the final configuration change of the zigzag ( @xmath289 ) that marks also the beginning of a new pulse zigzag occurs synchronously in cell 0 and cell 1 at time 3 .
we remark that the overall time of the pulse zigzag remains unchanged if the simulated head inserts a blank between the two delimiters .
[ th - rca ] if @xmath20 halts on @xmath30 and @xmath155 is initialized with @xmath175 then @xmath155 enters a final configuration in a time less than 6 cycles of cell 0 , containing the result of the calculation between the left and right delimiter .
if @xmath20 does not halt , @xmath155 enters after 6 cycles of cell 0 the final configuration that consists of an infinite string of the quiescent element : @xmath290 .
@xmath155 needs 3 cycles of cell 0 to perform the first zigzag of the pulse .
after the 3 cycles the configuration is shifted one cell downwards , starting now in cell 1 .
the next zigzag takes 3 cycles of cell 1 which are 3/2 cyles of cell 0 , and so on .
each zigzags performs at least one step of the turing machine @xmath20 , if @xmath20 does not halt .
we conclude that if @xmath20 halts , @xmath108 enters a final configuration in a time less than @xmath291 cycles of cell 0 .
if @xmath20 does not halt , the zigzag disappears in infinity after 6 cycles of cell 0 leaving a trail of @xmath169 s behind .
if @xmath20 is a universal turing machine , we immediately obtain the following result , which proves that @xmath155 is a hypercomputer for certain turing machines @xmath20 .
let @xmath46 be a universal turing machine .
then @xmath292 solves the halting problem for turing machines .
initialize @xmath292 with an encoded turing machine @xmath20 and an input word @xmath30 .
then @xmath155 enters a final configuration with the result of @xmath20 on @xmath30 in less than 6 cycles of cell 0 if and only if @xmath20 halts . in the current form of turing machine simulation
the operator has to scan a potentially unlimited number of cells to determine whether @xmath20 has halted or not , which limits its practical value .
if @xmath20 has halted , we would like to propagate at least this fact back to the upper cells .
the following obvious strategy fails in a subtle way . add a rule to @xmath155 that whenever @xmath176 has no next move , replaces it by the new symbol @xmath293 .
add the rule @xmath294 to @xmath155 that propagates @xmath293 upwards to cell @xmath140 .
the propagation upwards is only possible if we change also the block transformation [ tr : up - lhd ] to @xmath295 , thereby introducing a new symbol @xmath296 that is not subject of the short - circuit evaluation .
the last point , even if necessary , causes the strategy to fail , since if @xmath155 does not halt , @xmath155 is after 6 cycles in the configuration @xmath297 that leads to indeterministic behavior of @xmath155 .
this is in so far problematic , since we can not be sure whether a state @xmath293 in cell @xmath140 is really the outcome of a halting turing machine or the result of indeterministic behavior . instead of enhancing the self - similar cellular automaton model
, we will introduce in the next section a computing model that is computational equivalent for finite computations , but avoids indeterminism for infinite computations .
the evolution of a cellular automaton as well as the evolution of a self - similar cellular automaton depends on an extrinsic clock representing a global time that triggers the state changes .
since a self - similar cellular automaton can not halt , a self - similar cellular automaton is forced to perform a state change , even if no state with a causal relationship to the previous one exists , leading to indeterministic behavior , as described in the introduction . in this section ,
we present a model based on petri nets , the self - similar petri nets , with a close resemblance to self - similar cellular automata . even though petri nets in general are not deterministic , there exist subclasses that are . as will be shown below
, self - similar petri nets are deterministic .
they are also capable of hypercomputing , but compared to self - similar cellular automata , their behavior differ in the limit . whereas a self - similar cellular automaton features indeterministic behavior , the self - similar petri net halts .
petri introduced petri nets in the 1960s to study asynchronous computing systems .
they are now widely used to describe and study information processing systems that are characterized as being concurrent , asynchronous , distributed , parallel , nondeterministic , and/or stochastic .
it is interesting to note that very early , and clearly ahead of its time , petri investigated the connections between physical and computational processes , see e.g. , ref .
@xcite . in
what follows , we give a brief introduction to petri nets to define the terminology . for a more comprehensive treatment
we refer to the literature ; e.g. , to ref . @xcite .
a petri net is a directed , weighted , bipartite graph consisting of two kinds of nodes , called places and transitions .
the weight @xmath298 is the weight of the arc from place @xmath25 to transition @xmath6 , @xmath299 is the weight of the arc from transition @xmath6 to place @xmath25 .
a marking assigns to place @xmath25 a nonnegative integer @xmath76 , we say that @xmath25 is marked with @xmath76 tokens .
if a place @xmath25 is connected with a transition @xmath6 by an arc that goes from @xmath25 to @xmath6 , @xmath25 is an input place of @xmath6 , if the arc goes from @xmath6 to @xmath25 , @xmath25 is an output place .
a petri net is changed according to the following transition ( firing ) rule : 1 .
a transition @xmath6 may fire if each input place @xmath25 of @xmath6 is marked with at least @xmath298 tokens , and 2 . a firing of an enabled transition @xmath6 removes @xmath298 tokens from each input place @xmath25 of @xmath6 , and adds @xmath299 tokens to each output place @xmath25 of @xmath6 .
formally , a petri net @xmath300 is a tuple @xmath301 where @xmath218 is the set of places , @xmath302 is the set of transitions , @xmath303 is the set of arcs , @xmath304 is the weight function , and @xmath305 is the initial marking . in graphical representation , places are drawn as circles and transitions as boxes . if a place is input place of more than one transition , the petri net becomes in general indeterministic , since a token in this place might enable more than one transition , but only one can actually fire and consume the token .
the subclass of petri nets given in the following definition avoids these conflicts and is therefore deterministic . in a standard petri net ,
tokens are indistinguishable . if the petri net model is extended so that the tokens can hold values , the petri net is called a colored petri net . a marked graph is a petri net such that each place has exactly one input transition and exactly one output transition .
a colored petri net is a petri net where each token has a value .
it is well - known that cellular automata can be modeled as colored petri nets . to do this , each cell of the cellular automaton
is replaced by a transition and a place for each neighbor .
the neighbor transitions send their states as token values to their output places , which are the input places of the transition under consideration .
the transition consumes the tokens , calculates the new state , and send its state back to its neighbors .
a similar construction can be done for self - similar cellular automata , leading to the class of self - similar petri nets . underlying graph of a self - similar petri net . ]
a _ self - similar petri net _ is a colored petri net with some extensions .
a self - similar petri net has the underlying graph partitioned into cells that is depicted in fig .
[ petri ] .
we denote the transition of cell @xmath41 by @xmath306 , the place to the left of the transition by @xmath307 , the place to the right of the transition by @xmath308 and the central place , in the figure the place above the transition , by @xmath309 .
let @xmath96 be a finite set , the state set , @xmath310 be the quiescent state , and @xmath103 be a ( partial ) function @xmath311 .
the set @xmath312 is the value set of the tokens .
tokens are added to a place and consumed from the place according to a first - in first - out order .
initially , the self - similar petri net starts with a finite number of cells @xmath313 , and is allowed to grow to the right .
the notation @xmath314 defines the following action : create a token with value @xmath315 and add it to place @xmath25 .
the firing rule for a transition in cell @xmath41 of a self - similar petri net extends the firing rule of a standard petri net in the following way : 1 .
if the transition @xmath306 is enabled , the transition removes token @xmath316 from place @xmath307 , token @xmath317 from @xmath309 and tokens @xmath318 from @xmath308 .
the value of token @xmath319 shall be of the form @xmath320 in @xmath321 , the other token values @xmath322 and @xmath323 shall be in @xmath96 .
if the tokens do not conform , the behavior of the transition is undefined .
2 . the transition calculates @xmath324 .
( left boundary cell ) _ if @xmath325 then @xmath326 , @xmath327 , @xmath328 , @xmath329 .
( inner cell ) _ if @xmath330 and @xmath41 is not the highest index , then : @xmath331 , @xmath332 , @xmath333 , @xmath334 .
( right boundary cell ) _
if @xmath41 is the highest index then : 1 . _
( quiescent state ) _ [ firing - rule - quiescent ] if @xmath335 then @xmath336 , @xmath337 , @xmath338 , @xmath338 2 . _
( new cell allocation ) _ if @xmath145 then a new cell @xmath339 is created and connected to cell @xmath41 . furthermore : @xmath331 , @xmath332 , @xmath338 , @xmath333 , @xmath334 , @xmath340 , @xmath341 , @xmath341 .
formally , we denote the self - similar petri net by a tuple @xmath342 .
a self - similar petri net is a marked graph and therefore deterministic .
the initial markup is chosen in such a way that initially only the rightmost transition is enabled .
( initial markup ) let @xmath343 be an input word in @xmath344 and let @xmath300 be a self - similar petri net with @xmath41 cells , whereby @xmath345 . the initial markup of the petri net is as follows : * @xmath346 , ( @xmath347 , @xmath348 ) for @xmath349 , ( @xmath350 , @xmath351 ) for @xmath352 * @xmath353 for @xmath354 , @xmath355 for @xmath356 , * @xmath357 for @xmath358 , @xmath359 for @xmath360 , and @xmath338 .
note that the place @xmath308 is initialized with two tokens .
we identify the state of a cell with the value of its @xmath361-token
. if @xmath361 is empty , because the transition is in the process of firing , the state shall be the value of the last consumed token of @xmath361 . token flow in a self - similar petri net .
] fig .
[ token - flow ] depicts the token flow of a self - similar petri net consisting of @xmath362 cells under the assumption that the self - similar petri net does not grow .
tokens that are created and consumed by the same cell are not shown .
the numbers indicate whether the firing is uncoupled ( 0 ) or coupled ( 1 ) .
the only transition that is enabled in the begin is @xmath363 , since @xmath364 was initialized with 2 tokens .
the firing of @xmath363 bootstraps the self - similar petri net by adding a second token to @xmath365 , thereby enabling @xmath366 , and so on , until all transitions have fired , and the token flow enters periodic behavior .
we now compare self - similar petri nets with self - similar cellular automata .
we call a computation finite , if it involves either only a finite number of state updates of a self - similar cellular automaton , or a finite number of transition firings of a self - similar petri net , respectively .
[ lemma : comp - equivalence ] for finite computations , a dynamically growing self - similar cellular automaton @xmath109 and a self - similar petri net @xmath342 are computationally equivalent on a step - by - step basis if the start with the same number of cells and the same initial configuration .
let @xmath300 be a self - similar petri net which has @xmath41 cells initially .
for the sake of the proof consider an enhanced self - similar petri net @xmath367 that is able to timestamp its token .
a token @xmath368 of @xmath367 does not hold only a value , but also a time interval .
we refer to the time interval of @xmath368 by @xmath369 and to the value of @xmath368 by @xmath370 .
we remark that the timestamps serve only to compare the computations of a self - similar cellular automaton and a self - similar petri net and do not imply any time behavior of the self - similar petri net .
the firing rule of @xmath367 works as for @xmath300 , but has an additional pre- and postprocessing step : * _ ( preprocessing ) _ let @xmath317 , @xmath316 , @xmath371 , and @xmath372 be the consumed token , where the alphabetical subscript denotes the input place and the numerical subscript the order in which the tokens were consumed . calculate @xmath373 , where @xmath374 is the inverse time operator of @xmath375 .
if @xmath376 or @xmath377 or @xmath378 the firing fails and the transition becomes permanently disabled . * _ ( postprocessing ) _ for each created token @xmath368 , set @xmath379 .
the initial marking must set the @xmath6-field , otherwise the first transitions will fail . for the initial tokens in cell @xmath76 ,
set @xmath380 for both tokens in place @xmath381 , @xmath382 , and @xmath383 .
set @xmath384 for the second token in @xmath308 .
the firings of cell @xmath76 add tokens with timestamps @xmath385 to the output place @xmath386 .
if transition @xmath387 does not fail , the state function for the arguments @xmath388 and @xmath389 is well - defined : @xmath390 if cell @xmath76 has produced or was initialized in place @xmath391 with a token @xmath368 with @xmath379 and @xmath392 .
let @xmath97 be the state function of the scale - invariant cellular automaton @xmath108 .
due to the initialization , the two state functions are defined for the first @xmath41 cells and first time intervals @xmath393 .
assume that the values of @xmath100 and @xmath394 differ for some argument or that their domains are different .
consider the first time interval @xmath395 where the difference occurs : @xmath396 , or exactly one of @xmath397 or @xmath398 is undefined .
if there is more than one time interval choose an arbitrary one of these . since @xmath395 was the first time interval where the state functions differ , we know that @xmath399 , @xmath400 , @xmath401 , and @xmath402 .
we handle the case that the values of the state functions are different or that @xmath394 is undefined for @xmath403 whereas @xmath100 is .
the other case ( @xmath394 defined , but not @xmath100 ) can be handled analogously .
if @xmath404 , we conclude that tokens with timestamps @xmath405 , @xmath406 , @xmath407 , @xmath408 were sent to cell @xmath76 , and no other tokens were sent afterwards to cell @xmath76 , since the timestamps are created in chronological order .
hence , the precondition of the firing rule is satisfied and we conclude that @xmath409 , which contradicts our assumption .
the allocation of new cells introduces some technicalities , but the overall strategy of going back in time and concluding that the conditions for a state change or cell allocation were the same in both models works here also .
we complete the proof , by the simple observation that @xmath300 and @xmath367 perform the same computation .
the proof can be simplified using the following more abstract argumentation .
a comparison of fig .
[ token - flow ] with fig .
[ fig : timeops ] shows that each computation step has in both models the same causal dependencies .
since both computers use the same rule to calculate the value of a cell , respectively the value of a token , we conclude that the causal nets @xcite of both computations are the same for a finite computation , and therefore both computers yield the same output , in case the computation is finite .
a large number of different approaches to introducing time concepts to petri nets have been proposed since the first extensions in the mid 1970s .
we do not delve into the depths of the different models , but instead , define a very simple time schedule for the class of self - similar petri nets .
a timed self - similar petri net is a self - similar petri net that fires as soon as the transition is enabled and where a firing of an enabled transition @xmath387 takes the time @xmath134 . in the beginning of the firing ,
the tokens are removed from the input places , and at the end of the firing the produced tokes of the firing are simultaneously entered into the output places .
this time model can be satisfied if the cells of the timed self - similar petri net are arranged as the cells of a self - similar cellular automaton . under the assumption of a constant token speed , a firing time that is proportional to the cell length , and an appropriate unit of time we yield again cycle times of @xmath134 .
we now come back to the simulation of turing machines and construct a hypercomputing timed self - similar petri net , analogous to the hypercomputing self - similar cellular automaton in section [ chap : hypercomputer ] .
let @xmath10 be an arbitrary turing machine .
let @xmath96 be the state set that we used in the simulation of a turing machine by a self - similar cellular automaton , and let @xmath103 the local rule that is defined by the block transformations [ tr : start - state ] - [ tr : up - lhd ] , without the short - circuit evaluation . by lemma [ lemma : comp - equivalence ]
we know that the timed self - similar petri net @xmath410 simulates @xmath20 correctly for a finite number of turing machine steps .
hence , if @xmath20 halts on input @xmath30 , @xmath411 enters a final configuration in less than 6 cyles of cell 0 .
we examine now the case that @xmath20 does not halt . a pivotal difference between a self - similar cellular automaton and a self - similar petri net
is the ability of the latter one to halt on a computation .
this happens if all transitions of the self - similar petri net are disabled .
[ lemma : apn - halting ] let @xmath10 be an arbitrary turing machine and @xmath30 an input word in @xmath166 .
if @xmath20 does not halt on @xmath30 , the timed self - similar petri net @xmath411 halts on @xmath175 after 6 cycles of cell 0 . as long as the number of cells is finite , the boundary condition [ firing - rule - quiescent ] of the firing rule adds by each firing two tokens to the @xmath391-place of the rightmost cell that successively enable all other transitions as well .
this holds no longer for the infinite case .
let @xmath20 be a turing machine , and @xmath30 an input word , such that @xmath20 does not halt on @xmath30 .
we consider again the travel of the pulse zigzags down to infinity for the timed self - similar petri net @xmath411 with initial configuration @xmath175 , thereby tracking the marking of the @xmath391-places for times after the zigzag has passed by .
the first states of cell @xmath140 are @xmath256 , @xmath178 , @xmath178 , and @xmath169 , including the initial one .
the state @xmath169 is the result of the firing at time 3 , exhausting thereby the tokens in place @xmath412 . at time 3
the left delimiter ( @xmath256 ) of the pulse zigzag is now in cell 1 .
cell 1 runs from time 3 on through the same state sequence @xmath256 , @xmath178 , @xmath178 , and @xmath169 , thereby adding in summary 4 tokens to @xmath412 . after creating the token with value @xmath169 ,
@xmath413 is empty as well .
we conclude that after the zigzag has passed by a cell , the lower cell sends in summary 4 tokens to the upper cell , till the zigzag has left the lower cell as well .
for each cell @xmath76 these four tokens in @xmath414 enable two firings of cell @xmath76 thereby adding two tokens to @xmath415 .
these two tokens of @xmath415 enable again one firing of cell @xmath416 thereby adding one token to @xmath417 .
we conclude that each cell fires 3 times after the zigzag has passed by and that the final marking of each @xmath391 is one .
hence , no @xmath391 has the necessary two tokens that enable the transition , therefore all transitions are disabled and @xmath411 halts at time 6 . since @xmath411 halts for nonhalting turing machines , there are no longer any obstacles that prevent the construction of the proposed propagation of the halting state back to upper cells .
we replace block transformation [ tr : start - state ] with the following two and add one new .
if @xmath193 set @xmath418 if @xmath197 or @xmath419 is not defined set @xmath420 if @xmath419 is not defined set @xmath421 the following definition propagates the state @xmath293 up to cell @xmath140 : @xmath422 we denote the resulting timed self - similar petri net by @xmath423 .
the following theorem makes use of the apparently paradoxical fact , that @xmath423 halts if and only if the simulated turing machine does not halt .
let @xmath46 be a universal turing machine .
then @xmath424 solves the halting problem for turing machines .
consider a turing machine @xmath20 and an input word @xmath30 .
initialize @xmath424 with @xmath425 where @xmath426 is the encoding of @xmath20 and @xmath30 . if @xmath20 does not halt on @xmath30 , @xmath424 halts at time 6 by lemma [ lemma : apn - halting ] .
if @xmath20 halts on @xmath30 , then one cell of @xmath424 enters the state @xmath293 by block transformation [ tr : h ] or [ tr : down - to - head2 ] according to theorem [ th - rca ] and lemma [ lemma : comp - equivalence ] and taking the changes in @xmath103 into account . the mapping [ eq : h - up ] propagates @xmath293 up to cell 0 .
an easy calculation shows that cell 0 is in state @xmath293 , in time 7 or less .
we have proven that @xmath424 is indeed a hypercomputer without the deficiencies of the scale - invariant cellular automaton - based hypercomputer .
we end this section with two remarks .
the timed self - similar petri net @xmath411 sends a flag back to the upper cells , if the simulated turing machine halts . strictly speaking ,
this is not necessary , if the operator is able to recognize whether the timed self - similar petri net has halted or not . on the other hand ,
a similar construction is essential , if the operator is interested in the final tape content of the simulated turing machine . transferring the whole tape content of the simulated turing machine upwards , could be achieved by implementing a second pulse that performs an upwards - moving zigzag .
the construction is even simpler as the described one , since the tape content of the turing machine becomes static as soon as the turing machine halts .
the halting problem of turing machines is not the only problem that can be solved by self - similar cellular automata , scale - invariant cellular automata , or timed self - similar petri nets , but is unsolvable for turing machines .
a discussion of other problems unsolvable by turing machines and of techniques to solve them within infinite computing machines , can be found in davies @xcite .
we have presented two new computing models that implement the potential infinite divisibility of physical configuration space . these models are purely information theoretic and do not take into account kinetic and other effects . with these provisos , it is possible , at least in principle , to use the potential infinite divisibility of space - time to perform hypercomputation , thereby extending the algorithmic domain to hitherto unsolvable decision problems .
both models are composed of elementary computation primitives .
the two models are closely related but are very different ontologically .
a cellular automaton depends on an _ extrinsic _ time requiring an _ external _ clock and a rigid synchronization of its computing cells , whereas a petri net implements a causal relationship leading to an _ intrinsic _ concept of time .
scale - invariant cellular automata as well as self - similar petri nets are built in the same way from their primitive building blocks .
each unit is recursively coupled with a sized - down copy of itself , potentially leading to an infinite sequence of ever decreasing units .
their close resemblance leads to a step - by - step equivalence of finite computations , yet their ontological difference yields different behaviors for the for the case that the computation involves an infinite number of units : a scale - invariant cellular automaton exhibits indeterministic behavior , whereas a self - similar petri net halts .
two supertasks which operate identically in the finite case but differ in their limit is a puzzling observation which might question our present understanding of supertasks .
this may be considered an analogy to a theorem @xcite in recursive analysis about the existence of recursive monotone bounded sequences of rational numbers whose limit is not a computable number .
one striking feature of both models is their scale - invariance .
the computational behavior of these models is therefore the first example for what might be called scale - invariant or self - similar computing , which might be characterized by the property that any computational space - time pattern can be arbitrary squeezed to finer and finer regions of space and time .
although the basic definitions have been given , and elementary properties of these new models have been explored , a great number of questions remain open for future research .
the construction of a hypercomputer was a first demonstration of the extraordinary computational capabilities of these models .
further investigations are necessary to determine their limits , and to relate them with the emerging field of hypercomputation @xcite
. another line of research would be the investigation of their phenomenological properties , analogous to the statistical mechanics of cellular automata @xcite .
m. margenstern and k. morita , `` a polynomial solution for 3-sat in the space of cellular automata in the hyperbolic plane , '' journal of universal computer science * 5 * , 563573 ( 1999 ) .
j. durand - lose , `` abstract geometrical computation for black hole computation , '' in _ machines , computations , and universality , 4th international conference , mcu 2004 , saint petersburg , russia , september 21 - 24 , 2004 , revised selected papers _ , m. margenstern , ed .
( springer , 2005 ) , pp .
i. nmeti and g. dvid , `` relativistic computers and the turing barrier , '' applied mathematics and computation * 178 * , 118142 ( 2006 ) , special issue on hypercomputation .
o. bournez and m. l. campagnolo , `` a survey on continuous time computations , '' in _ new computational paradigms . changing conceptions of what is computable _ , s. cooper , b. lwe , and a. sorbi , eds .
( springer verlag , new york , 2008 ) , pp .
383423 .
h. gutowitz , `` cellular automata : theory and experiment , '' physica * d45 * , 3483 ( 1990 ) , previous ca conference proceedings in _ international journal of theoretical physics _
* 21 * , 1982 ; as well as in _ physica _ , * d10 * , 1984 and in * complex systems * * 2 * , 1988 .
e. specker , `` nicht konstruktiv beweisbare stze der analysis , '' the journal of smbolic logic * 14 * , 145158 ( 1949 ) , reprinted in @xcite ; english translation : _ theorems of analysis which can not be proven constructively_. | two novel computing models based on an infinite tessellation of space - time are introduced .
they consist of recursively coupled primitive building blocks .
the first model is a scale - invariant generalization of cellular automata , whereas the second one utilizes self - similar petri nets .
both models are capable of hypercomputations and can , for instance , `` solve '' the halting problem for turing machines .
these two models are closely related , as they exhibit a step - by - step equivalence for finite computations . on the other hand , they differ greatly for computations that involve an infinite number of building blocks : the first one shows indeterministic behavior whereas the second one halts .
both models are capable of challenging our understanding of computability , causality , and space - time . |
in contrast to conventional superconductivity and antiferromagnetism , metallic ferromagnetism is in general an intermediate or strong coupling phenomenon . since there
do not exist systematic investigation schemes to solve such types of problems the stability of metallic ferromagnetism is still not sufficiently understood .
this is true even within the simplest electronic correlation model , the one - band hubbard model,@xcite in spite of significant progress made recently .
the hubbard interaction is very unspecific , i.e. , does not depend on the lattice structure or dimension .
hence the dispersion , and thereby the shape of the density of states ( dos ) , is of considerable importance for the stability of ferromagnetism .
this was recognized already by gutzwiller,@xcite hubbard,@xcite and kanamori@xcite in their initial work on the hubbard model .
however , the approximations used in the early days of many - body theory were usually not reliable enough to provide definite conclusions .
an exception are the exact results by nagaoka@xcite on the stability of ferromagnetism at @xmath4 in the case of one electron above or below half filling .
they show an important lattice sensitivity but , unfortunately , are not applicable in the thermodynamic limit . over the years
the stability of metallic ferromagnetism has turned out to be a particularly difficult many - body problem whose explanation requires subtle _ nonperturbative _ techniques .
there has been an upsurge of interest in this topic most recently @xcite .
these investigations confirm that ferromagnetism is favored in systems with ( i ) frustrated lattices ( which suppress antiferromagnetism ) and ( ii ) high spectral weight near the band edge closest to the fermi energy ( which improve the kinetic energy of the polarized electrons ) . taken together , these properties imply a strongly asymmetric dos of the electrons .
ferromagnetism on bipartite lattices having a symmetric dos may still be possible , but seems to require very large values of @xmath0.@xcite with the exception of refs .
all previous calculations refer to the ground state .
it is therefore of interest to obtain an answer to the question : _ how does the distribution of spectral weight in the dos influence the stability regime of ferromagnetism at finite temperatures ? _ it should be noted that a strongly peaked , asymmetric dos is a considerably more complex condition for ferromagnetism than the stoner criterion .
the latter merely asserts that , at @xmath5 , the critical interaction for the instability is determined by the inverse of the dos precisely _ at _ the fermi energy @xmath6 , @xmath7 , thus neglecting antiferromagnetism and the structure of the dos away from @xmath6 .
stoner ( i.e. , hartree - fock@xcite ) theory is a purely static mean - field theory which ignores correlation effects , e.g. , the correlation - induced redistribution of momentum states and the dynamic renormalizations of the band shape and width .
so the question remains : _ how essential are genuine correlation effects for the stability of itinerant ferromagnetism at finite temperatures ? _
a third question concerns the suitability of the hubbard model itself as a model for ferromagnetism .
indeed there is no compelling _ a priori _ reason why the hubbard model should be a good model for ferromagnetism at all .
not only does it neglect band degeneracy , a feature observed in all ferromagnetic transition metals ( fe , co , ni ) , it also ignores the ( weak ) direct heisenberg exchange interaction which is equivalent to a ferromagnetic spin - spin interaction and hence favors ferromagnetism in the most obvious way.@xcite the proposition by hirsch and coworkers@xcite that this interaction plays a key role in metallic ferromagnetism was disputed by campbell et al.@xcite so the controversial question is : _ how important is the direct heisenberg exchange interaction for the stability of itinerant ferromagnetism in the one - band hubbard model at finite temperatures ? _ in this paper quantitative answers to the three questions formulated above are given within the dynamical mean field theory ( dmft ) .
the dmft , a nonperturbative approach , becomes exact in the limit of large coordination numbers.@xcite when applied to @xmath8 , where the coordination number is @xmath9 , the dmft has proven to yield accurate and reliable results , especially in the context of long - range magnetic order.@xcite it treats local correlations exactly while spatial fluctuations are neglected . in this situation
the momentum integral entering in the local propagator will be replaced by an energy integral involving only the dos of the noninteracting electrons .
the latter may be viewed as an input parameter . in our investigation
the question concerning the importance of the distribution of spectral weight within the band for the stability of ferromagnetism will therefore be studied using a model dos of the noninteracting electrons whose shape can be changed continuously from symmetric to strongly asymmetric by varying an asymmetry parameter .
the paper is structured as follows : in sec . ii we present the model under investigation , the dynamical mean - field equations , and the analytical and numerical steps needed to construct magnetic phase diagrams .
the model dos is introduced in sec .
the results of our investigation and quantitative answers to the questions posed above are presented in sec .
iv . a discussion where these results are put into perspective ( sec .
v ) closes the presentation .
the minimal model allowing one to treat an asymmetric dos , electronic on - site correlations , and the nearest - neighbor heisenberg exchange interaction is given by @xmath10 here @xmath11 with the vector of pauli matrices @xmath12 .
we note that there are three other nearest - neighbor contributions of the coulomb interaction which might also effect the stability of the ferromagnetic phase ( see appendix ) . within the dmft the coupling constants in ( [ exthubm ] ) and ( [ hub ] ) have to be scaled with the lattice coordination number @xmath13 as@xcite @xmath14 , where we consider nearest - neighbor hopping @xmath15 only . by analogy to classical spin models@xcite the hartree - fock approximation yields the exact result for the @xmath1-term in high dimensions . in the following
we investigate the influence of the direct exchange term on the properties of the hubbard model in @xmath16 .
since the hubbard model is su(2 ) spin - symmetric we can , without loss of generality , assume a magnetization parallel to the @xmath17-axis .
the hartree@xcite decoupling then takes the form @xmath18 . in terms of the magnetization @xmath19 and its expectation value @xmath20 , where @xmath21 and @xmath22 is the number of lattice sites , the hamiltonian ( [ exthubm ] ) can be written as@xcite @xmath23 apparently the influence of the exchange term in the limit @xmath24 is that of a ( weiss ) magnetic field which vanishes in the paramagnetic phase ( @xmath25 )
therefore , in this phase , all one - particle properties of the system are those of the pure hubbard model . however , two - particle functions , especially the ferromagnetic susceptibility , are modified ( see sec . [ iiib ] ) . in @xmath26 ,
the hubbard model ( [ hub ] ) is equivalent to an anderson impurity model complemented by a self - consistency condition.@xcite written in terms of matsubara frequencies @xmath27 , self energy @xmath28 , the dos of the noninteracting electrons @xmath29 , and thermal average @xmath30 the resulting coupled equations for the green function in the homogeneous phase read : @xmath31 the solution of the @xmath32integrated dyson equation ( [ sk2 ] ) is straightforward and can be performed analytically for the dos used in this paper ( see sec .
iv ) . by contrast
the solution of eq .
( [ sk1 ] ) is highly nontrivial ( for details of the notation see ref . ) .
it is achieved using the auxiliary - field quantum monte carlo ( qmc ) algorithm by hirsch and fye,@xcite where a discretization of imaginary time , @xmath33 , is introduced . here
, @xmath34 denotes the number of independent matsubara frequencies .
physical quantities are obtained in the @xmath35 limit . the second order phase transition from a para- to
a ferromagnetic phase occurs at the zero of the inverse susceptibility @xmath36 , calculated@xcite in the paramagnetic phase .
it is sufficient to perform all simulations for the pure hubbard model since the influence of @xmath37 on the susceptibility is given by the following random - phase approximation ( rpa)-like expression : @xmath38 where @xmath39 is the susceptibility of the pure hubbard model .
this type of relation holds for pairs of two - particle interactions ( here @xmath0 and @xmath37 ) in arbitrary dimensions , whenever one interaction ( here @xmath37 ) is treated in hartree - fock approximation and the other one ( here @xmath0 ) exactly.@xcite in general , since the calculation of the susceptibility involves the derivative of the self - energy @xmath40 with respect to some field @xmath41,@xcite this follows from the fact that the self - energy of the full hamiltonian can be expressed as @xmath42=\sigma_{u,0}[g_{u , f^ * } ] + \sigma_{0,f^*}^{(1)}[g_{u , f^*}].\ ] ] here @xmath43 $ ] refers to the diagrammatic skeleton expansion of @xmath44 , where all lines are fully dressed propagators @xmath45 .
since the hartree - fock term @xmath46 only contributes in the symmetry - broken phase , all @xmath37-renormalizations of @xmath47 vanish in the symmetric phase . evaluating @xmath48 in the symmetric phase , the first term in ( [ sigma ] ) leads to the same contributions as without @xmath37-interaction , while the second term introduces the rpa - like term proportional to @xmath37 . for the @xmath15-@xmath49 model in dmft the analogue of ( [ rpa ] ) was derived by pruschke et al.@xcite since @xmath37 only enters the calculations via eq .
( [ sigma ] ) we are left with four physical parameters of the pure hubbard model : hubbard interaction @xmath0 , electron density @xmath2 , temperature @xmath3 and an asymmetry parameter @xmath50 for the kinetic energy ( see sec .
iv ) . for each set of these five parameters eqs . ( [ sk2 ] ) and ( [ sk1 ] )
are iterated with typically @xmath51 monte carlo sweeps until convergence is reached , i.e. , the difference between two consecutive values of ( @xmath52@xmath53 ( @xmath45@xmath54 @xmath44@xmath55 is smaller than @xmath56 ( measured by the norm @xmath57 ; the energy scale is defined in sec .
subsequently eight measurements of the susceptibility are performed with a reduced number of @xmath58 monte carlo sweeps .
thus the result for each parameter set consists of an averaged susceptibility @xmath59 and its statistical error @xmath60 .
we neglect the propagation of the error in ( @xmath52@xmath61 since it is always an order of magnitude smaller than @xmath60 .
the extrapolation to @xmath62 is performed by a quadratic least squares fit of @xmath59 , using at least six different values of @xmath63 for @xmath64 $ ] .
further details regarding the technical treatment can be found in refs . . for mean - field theories like the dmft a linear behavior of the inverse susceptibility , i.e. , a curie - weiss law ,
is expected and observed in the vicinity of the transition .
thus the curie temperature @xmath65 can be obtained as the zero of a linear fit of @xmath66 drawn from values of @xmath63 for four to six different temperatures ( see , e.g. , figs .
[ fvst ] and [ fvsta ] , where @xmath67 ( [ rpa ] ) is plotted ) . the error of @xmath65 is obtained from the errors @xmath60 by error propagation and therefore denotes only statistical , not systematic errors ( e.g. , due to the extrapolation schemes used ) .
however , we checked the accuracy of our results by varying the procedure , e.g. , extrapolating @xmath68 instead of @xmath59 .
due to the vanishing of spatial fluctuations within the dmft the topology of the underlying lattice enters the self - consistency equation ( [ sk2 ] ) only via the noninteracting dos , at least for homogeneous phases .
the choice of a particular model spectral function thus represents a special ( not unique ) set of hopping elements @xmath69 in the hamiltonian ( [ exthubm ] ) which characterize the structure of the underlying lattice .
contributions to the kinetic energy by , e.g. , next - nearest - neighbor hopping can lead to an asymmetrically shaped dos , which is apparently favorable for the stability of ferromagnetism . to investigate this stabilizing effect quantitatively
, we propose a model dos with a shape - controlling parameter .
this parameter allows us to change smoothly from a noninteracting dos with ( i ) a symmetric shape ( mimicking nearest - neighbor hopping on a bipartite lattice ) to ( ii ) an asymmetrically peaked dos ( similar to a cubic lattice with next - nearest - neighbor hopping ) to ( iii ) a dos with a square - root divergence at the band edge ( e.g. , a fcc lattice with next - nearest - neighbor hopping @xmath70 ) .
the shape of the model dos thus qualitatively captures key features of real lattices . the spectral function which we use throughout the paper is given by @xmath71 with @xmath72 and bandwidth @xmath73 .
the well - known semielliptic dos of the bethe lattice with infinite number of nearest neighbors is recovered for @xmath74 . by increasing the parameter
@xmath50 spectral weight is shifted towards the lower band edge ( fig . [ dos ] ) .
for @xmath75 the dos diverges at the lower band edge like an inverse square - root . the particular choice of the model dos ( [ ndos ] ) has the advantage that the numerical effort of solving the self - consistency equation ( [ sk2 ] ) is rather small since the hilbert transform can be calculated analytically .
in the following we set the variance to unity , @xmath76 ^ 2=1 $ ] , thereby fixing the energy scale .
this leads to @xmath77 for all values of @xmath50 .
for @xmath74 it is equivalent to choosing @xmath78 on the bethe lattice . while for the study of ferromagnetism within the dmft the lattice structure only enters via the dos it is possible to construct ( infinitely many ) corresponding dispersion relations @xmath79(@xmath32 ) or , equivalently , sets of hopping elements @xmath80 .
a realization in @xmath81 that is symmetric , @xmath82 , and monotonous , @xmath83 , for @xmath84 is given by # 1#2 @xmath85 where an inversion yields @xmath86 .
generalizations to other dimensions are possible with , e.g. , @xmath79(@xmath32@xmath87@xmath32@xmath88@xcite or @xmath79(@xmath32@xmath89@xcite although in principle one could thus choose a lattice corresponding to the dos ( [ ndos ] ) , this will not be done here , since we only study homogeneous phases .
antiferromagnetism or incommensurate phases are not expected to be important far away from half filling . only for the case
@xmath74 ( bethe lattice ) at @xmath90 the stability of an antiferromagnetic phase is investigated .
on a bipartite lattice with perfect nesting , the ground state of the pure hubbard model at half filling is antiferromagnetic for all @xmath91 , at least in dimensions @xmath92 .
in this situation a ferromagnetic state is strongly disfavored also in the general model , eq .
( [ exthubm ] ) . at large @xmath0 , however , when the model reduces to an effective heisenberg model ( which , in high dimensions , is exactly described by weiss mean - field theory ) , already a small value of the direct exchange interaction , @xmath93 , is sufficient to stabilize a ferromagnetic ground state.@xcite indeed , the heisenberg model well describes the @xmath37@xmath0 ground state phase diagram at half filling down to @xmath94 ; this is evident from fig .
[ fsymm ] , where a comparison with our qmc results is shown for a symmetric dos [ eq .
( [ ndos ] ) with @xmath74 ] . at small @xmath0 the phase boundary between a paramagnetic and a ferromagnetic state
is correctly reproduced by hartree - fock theory ( but only for @xmath96 ) .
this is not surprising since in @xmath26 the @xmath1-term is treated exactly within this approximation .
also included in fig .
[ fsymm ] is the line below which a fully saturated ferromagnetic state becomes unstable against single spin flips as first computed by hirsch@xcite for cubic lattices . for the bethe lattice this line is given exactly by the hartree - fock result @xmath97 for @xmath98 and @xmath99 for @xmath100 .
this can be seen from eqs . 5 and 7 in ref . and the known analytic expression for the hilbert transform of the semielliptic dos .
the remarkable agreement between the qmc results and this curve for @xmath101 suggests that ( at zero temperature ) the region of partial polarization is very narrow already at intermediate interaction strength @xmath0 .
the @xmath37@xmath3 phase diagram for a strongly peaked dos ( @xmath102 ) at filling @xmath103 is shown in fig .
the qmc results for the ferromagnetic phase boundary can be extrapolated linearly to zero temperature leading to a ground state phase diagram .
clearly the values of @xmath37 necessary to stabilize ferromagnetism are significantly reduced in comparison to the bipartite case . in particular , for @xmath104 and @xmath105 the extrapolation lines cross the ordinate at positive temperatures .
thus , an asymmetric dos # 1#2 _ model _ ( @xmath106 ) above a critical interaction strength @xmath107 with @xmath108 .
[ fvst ] and [ tvsnf ] prove that already a small direct exchange coupling @xmath37 can significantly enhance ferromagnetic tendencies and thus give the final `` kick '' towards ferromagnetism for systems that are close to an instability .
this influence is stronger at larger densities when the local magnetic moments are enhanced ( fig .
[ tvsnf ] ) .
the lower critical densities are very small , but larger than those predicted by hartree - fock theory , since hartree - fock always overestimates the size of the ferromagnetic regime .
# 1#2 # 1#2 quite generally , the value of the exchange interaction @xmath1 in metals can be expected to be rather small .
note , however , that for three - dimensional lattices the _ scaled _ quantity @xmath109 is an order of magnitude larger than the exchange coupling @xmath1 itself .
hubbard s crude estimate@xcite of @xmath110 would therefore imply that , e.g. , at @xmath104 a ( scaled ) exchange interaction as large as @xmath111 is not completely unrealistic . the dependence of the phase boundary on the asymmetry parameter @xmath50 is systematically studied in fig . [ fvsta ] at @xmath112 for a relatively small electron density @xmath113 . for a symmetric or slightly asymmetric dos ( @xmath74 , @xmath114 )
the system only becomes ferromagnetic for @xmath115 even at @xmath5 . for @xmath116 , when the shape of the dos is roughly triangular , the critical @xmath37 is considerably reduced .
but only when a marked peak develops ( i.e. , for @xmath117 ) does the critical @xmath37 drop to zero ; ferromagnetism is then stable even in the pure hubbard model . from now on we restrict our studies to this case ( @xmath106 ) . the @xmath3 vs. @xmath2 phase diagram is shown in fig . [ tvsna ] for @xmath112 and three different shapes of the dos ranging from strongly asymmetric ( @xmath118 ) to divergent at the lower band edge ( @xmath75 ) .
evidently the ferromagnetic phase is largest at @xmath75 .
we want to stress however , that the divergence does _ not _ change the physics qualitatively ( except at @xmath119 ) .
a moderately strong peak near the band edge is all that is needed to stabilize ferromagnetism . #
1#2 # 1#2 for symmetric densities of states at @xmath120 ( bethe lattice and hypercubic lattice ) we find _ suppression _ of the ferromagnetic ( as well as the antiferromagnetic ) susceptibility away from half filling.@xcite this does not exclude the possibility for ferromagnetism on bipartite lattices at much larger values of @xmath0 .
indeed , very recently ferromagnetism was found within the noncrossing approximation for the @xmath15@xmath49 model on a hypercubic lattice in the limit @xmath121 for @xmath122 away from half filling.@xcite apparently , at least for moderate @xmath0 , the bipartite lattice with only nearest - neighbor - hopping is not a natural `` environment '' for ferromagnetism the asymmetry of the noninteracting dos is crucial . in fig .
[ tvsnu ] the @xmath3@xmath2 phase diagram is shown for different values of the on - site interaction @xmath0 .
evidently the ferromagnetic phase becomes more favorable for increasing @xmath0 : both the maximal curie temperature and the ( upper ) critical density rise .
this effect is seen to be qualitatively similar to an increase of the exchange interaction @xmath37 or the asymmetry of the dos @xmath50 ( figs .
[ tvsnf ] and [ tvsna ] , respectively ) .
our qmc results are compared with hartree - fock theory in figs .
[ tvsnhf][tvsuhf ] .
we note that , applied to the hubbard model , the dmft includes hartree - fock theory as its static limit and is thus superior in any dimension .
[ tvsnhf ] shows the vast overestimation of the ferromagnetic phase within hartree - fock . the maximal curie temperature obtained in this approximation is more than an order of magnitude too large . at such high temperatures details of the dos
are averaged out and consequently the density dependence ( e.g. , the position of the maximum ) is completely artificial . in fig .
[ uvsnhf ] the @xmath0 vs. @xmath2 ground state phase diagram is shown and compared to the stoner criterion.@xcite at low @xmath2 the stoner curve clearly approaches the qmc curve .
since the dos vanishes smoothly at the lower band edge for @xmath123 both curves diverge for @xmath124 .
figure [ tvsuhf ] focuses on the limit of large @xmath0 .
the weak coupling hartree - fock theory fails again : it predicts an unbounded linear increase of @xmath65 with @xmath0 , @xmath125 , whereas qmc shows that @xmath65 has a finite limit for @xmath126 .
it is expected that such a finite limit exists for all densities .
a saturation is also suggested by the curves in fig .
[ tvsnu ] .
it arises from the suppression of double occupancies by correlations .
in contrast to the hartree - fock prediction the interaction energy goes to zero for @xmath126 , thus only the bandwidth remains as an energy scale . in the special case @xmath102 , @xmath127 one finds@xcite @xmath128 . #
1#2 # 1#2 # 1#2 # 1#2 one might argue that comparison of the dmft results should not be made with hartree - fock itself but with hartree - fock plus quantum corrections , since the latter are known to reduce many of the deficiencies of hartree - fock .
such corrections have been discussed by van dongen@xcite and freericks and jarrell.@xcite the latter authors showed how quantum fluctuations modify the stoner criterion by subtracting the particle - particle susceptibility . evaluating these corrections in the case of fig .
[ tvsuhf ] we find that the ferromagnetic phase is completely suppressed ( as previously observed in ref . for a symmetric dos ) . at @xmath102
this holds for all densities @xmath129 .
thus the `` corrections '' to hartree - fock are seen to _ underestimate _ the ferromagnetic region by far .
after more than three decades of research it has become clear at last@xcite that the hubbard model can describe itinerant ferromagnetism even on regular lattices and at moderate @xmath0-values for a wide range of electronic densities @xmath2 .
since ferromagnetism is an intermediate to strong coupling problem the question concerning its `` mechanism '' has , in principle , no straightforward answer .
this is in contrast to _ weak _ coupling phenomena , e.g. , conventional superconductivity , which can be explained within perturbation theory .
nevertheless a good starting point for an understanding of the origin of itinerant ferromagnetism can be obtained in the strong coupling limit . at @xmath4
doubly occupied sites are excluded and the hubbard model reduces to a ( complicated ) kinetic energy . to avoid doubly occupied sites in a paramagnetic phase
the dos is then necessarily strongly renormalized compared with the noninteracting case , whereas for the saturated ferromagnetic phase the interacting dos is the same as the noninteracting one except for a shift between the spin - up and -down bands . in this situation details of the structure of the noninteracting dos become relevant in selecting the state with the lowest energy .
this physical picture is , in principle , similar to that underlying the nagaoka mechanism ; however , the latter only addresses the kinetic energy of a single hole and it was so far not possible to generalize it to thermodynamically relevant densities .
our investigations within the dmft explicitly show that a moderately strong peak at the band edge closest to the fermi energy is sufficient to stabilize ferromagnetism .
furthermore a strong asymmetry of the dos implies a nonbipartite lattice which frustrates the competing antiferromagnetism near half filling .
the mechanism described above is completely different from the mere band shift of the hartree - fock theory .
this weak coupling approach does not take into account the dynamical renormalization of the dos in the paramagnetic phase and thus predicts ferromagnetism for any dos , even at relatively small values of @xmath0 and for high temperatures .
the comparison with dmft clearly shows that hartree - fock ( i ) overestimates transition temperatures by more than an order of magnitude , ( ii ) renders the dependence of @xmath65 on @xmath0 qualitatively incorrect , and ( iii ) predicts ferromagnetism for the symmetric dos , where ( at least for @xmath120 ) this is not found .
these shortcomings of hartree - fock theory are due to the neglect of dynamical fluctuations which are at the heart of the correlation problem .
the heisenberg exchange interaction , not considered in the pure hubbard model , provides another mechanism that may order the fluctuating local moments arising by the suppression of double occupancies .
we found that for a symmetric dos rather large values of @xmath1 are needed to stabilize ferromagnetism .
however , for an asymmetric dos with a peak near the band edge already small values of the exchange interaction may provide the final `` kick '' towards ferromagnetism . in any case it reduces the critical on - site interaction and increases the critical temperatures of the ferromagnetic phase boundary . while the dmft correctly describes the dynamic fluctuations of the interacting many - body system , it neglects spatial fluctuations and short - range order .
hence one should suspect that this approach overestimates the transition temperatures @xmath65 .
within dmft ulmke@xcite estimated @xmath65 for a three - dimensional fcc lattice to be of the order of 500 to 800 k which is in the range of realistic transition temperatures .
we may expect spatial fluctuations to reduce these temperatures . on the other hand , band degeneracy ,
not considered in our model so far , is expected to increase @xmath65 .
indeed , band degeneracy and hund s rule couplings which are clearly present in realistic systems can be rigorously shown to improve the stability of ferromagnetism at least for special parameter values.@xcite the incorporation of band degeneracy , for which the dmft also provides a suitable framework , is the most important feature that has to be included in future investigations of the hubbard model.@xcite the additional nearest - neighbor interactions discussed in the appendix may provide yet another mechanism for ferromagnetism and will be studied in the future .
we are very grateful to m. kollar and m. ulmke for numerous helpful discussions .
one of us ( n.b . ) thanks the fulbright commission for support and the department of physics at the university of illinois for hospitality
. computations were performed on the cray t90 of the hlrz jlich .
in wannier representation the coulomb interaction gives rise to the purely local interaction @xmath0 as well as to four nearest - neighbor interactions . @xcite besides the heisenberg exchange interaction these are the density - density interaction , the pair - hopping term , and the off - diagonal `` bond - charge site - charge''@xcite interaction .
the latter effectively describes a density - dependent hopping @xcite and leads to a narrowing of the band .
hence this term is expected to stabilize saturated ferromagnetism . since the quantum dynamics of this term make a systematic investigation difficult even within the dmft
its detailed study has to be postponed to the future .
the pair - hopping term also weakly enhances ferromagnetic tendencies.@xcite among all nearest - neighbor interactions the density - density term @xmath130 is largest and is thus investigated explicitly in the following . in the case of d - electrons hubbard roughly estimated this term to @xmath131=2 - 3 ev , an order of magnitude smaller than the hubbard interaction @xmath0.@xcite however , since there are @xmath13 neighbors contributing , the total energy of the nearest - neighbor density - density interactions may in some materials even surpass that of the hubbard interaction .
this raises the question of the importance of the @xmath131-term , in particular its influence on the ferromagnetic phases investigated in the present paper .
it was already pointed out by mller - hartmann@xcite that in the limit @xmath132 and with the proper scaling @xmath133 the nearest - neighbor density - density interaction reduces to its hartree contribution which may then be viewed as a simple , site - dependent shift of the chemical potential . in the absence of broken translational symmetry
the chemical potential must compensate this shift to keep the electron density fixed .
then there is no effect at all .
on bipartite lattices translational symmetry can be broken by a charge density wave ( cdw ) with different electron densities on a and b sublattices , i.e. , with order parameter @xmath134 . to study this possible
ordering we analyze the instability towards a cdw in the following .
similar to the exchange term @xmath1 the hartree contribution of the interaction @xmath131 leads , even in the presence of other interactions , to an rpa - like pole in the cdw susceptibility [ cf .
( [ rpa])]:@xcite @xmath135 thus a second order phase transition to a cdw occurs at @xmath136 .
since next - nearest - neighbor hopping frustrates cdw order , the maximal instability towards a cdw is expected for the symmetric dos with @xmath74 in eq .
( [ ndos ] ) .
half filling is optimal in this case . for these parameters we determined the phase diagram fig .
[ vvsu ] employing the qmc technique ( for details concerning the calculation of the cdw susceptibility see ref . ) . within dmft
a cdw ordering occurs for @xmath137 ( at not too high temperatures ) . compared to the hartree - fock approximation
the cdw phase boundary of the full model is only slightly moved towards larger values of @xmath138 . a similar deviation from hartree - fock
was found by means of qmc simulations in @xmath81 by hirsch,@xcite in @xmath139 by zhang and callaway,@xcite and within perturbation theory for both all these studies demonstrate that the cdw is stable for @xmath140 . while this relation may in principle hold for some transition metals , ferromagnets apparently do not show spatial charge ordering .
therefore the adequate correlated electron model for a ferromagnet appears to have parameters in the range @xmath141 .
then the nearest - neighbor density - density interaction @xmath138 has no influence on the phase diagram , especially on the border of the ferromagnetic phase , at least in @xmath26 . even in @xmath8
the hartree diagram gives the main contribution of the interaction @xmath138 since spatial fluctuations , leading to genuine correlations , are suppressed as @xmath142 .
moreover in @xmath81 and at half - filling the effect of @xmath138 on the ferromagnetic phase boundary is still small.@xcite therefore , over an extended range of parameters the nearest - neighbor term @xmath138 has almost no influence and thus its importance is seen to be much smaller than its value suggests .
e. mller - hartmann , in _ proceedings of the v. symposium `` physics of metals '' _ , edited by e. talik and j. szade , page 22 , poland , 1991 , ustro ; e. mller - hartmann , t. hanisch , and r. hirsch , physica b * 186 - 188 * , 834 ( 1993 ) ; t. hanisch and e. mller - hartmann , ann .
physik * 2 * , 381 ( 1993 ) ; t. hanisch , b. kleine , a. ritzl , and e. mller - hartmann , , 303 ( 1995 ) ; p. wurth , g. s. uhrig , and e. mller - hartmann , , 148 ( 1996 ) ; t. hanisch , g. s. uhrig , and e. mller - hartmann , phys .
b * 56 * , 13960 ( 1997 ) . m. jarrell and t. pruschke , z. phys .
b * 90 * , 187 ( 1993 ) ; m. j. rozenberg , g. kotliar , h. kajter , g. a. thomas , d. h. rapkine , j. m. honig and p. metcalf , phys . rev .
lett . * 75 * , 105 ( 1995 ) ; m. j. rozenberg , i. h. inoue , h. makino , f. iga and y. nishihara , phys . rev .
lett . * 76 * , 4781 ( 1996 ) ; m. ulmke , r. t. scalettar , a. nazarenko and e. dagotto , phys . rev .
b * 54 * , 16523 ( 1996 ) .
the fock contribution of the @xmath1 term vanishes as @xmath142 since it is given by two green functions connecting neighboring sites . in @xmath8
the fock term renormalizes the band - width : @xmath143\}$ ] ( see eq . ( 29 ) in ref .
, which readily yields this equation for the paramagnetic phase ) .
the renormalization is maximal at half - filling and reduces the critical value of @xmath37 for the onset of ferromagnetism , e.g. , for the half - filled cubic lattice from @xmath144 to @xmath145 within the hartree - fock approximation ( @xmath146 ) .
this effect of order @xmath142 is neglected in the dmft .
the definition of generalized ( e.g. , staggered ) susceptibilities for bipartite ( @xmath147- ) lattices requires the introduction of an index @xmath148 in eq .
( [ hfnn ] ) that indicates the sublattice .
the corresponding generalizations are straightforward ( see , e.g. , ref . ) . | a systematic investigation of the microscopic conditions stabilizing itinerant ferromagnetism of correlated electrons in a single - band model is presented .
quantitative results are obtained by quantum monte carlo simulations for a model with hubbard interaction @xmath0 and direct heisenberg exchange interaction @xmath1 within the dynamical mean - field theory .
special emphasis is placed on the investigation of ( i ) the distribution of spectral weight in the density of states , ( ii ) the importance of genuine correlations , and ( iii ) the significance of the direct exchange , for the stability of itinerant ferromagnetism at finite temperatures .
we find that already a moderately strong peak in the density of states near the band edge suffices to stabilize ferromagnetism at intermediate @xmath0-values in a broad range of electron densities @xmath2 .
correlation effects prove to be essential : slater hartree - fock results for the transition temperature are both qualitatively and quantitatively incorrect .
the nearest - neighbor heisenberg exchange does not , in general , play a decisive role .
detailed results for the magnetic phase diagram as a function of @xmath0 , @xmath1 , @xmath2 , temperature @xmath3 , and the asymmetry of the density of states are presented and discussed . 2 |
in recent years , advances in quantum control have emerged through the introduction of appropriate and powerful tools coming from mathematical control theory @xcite and by the use of sophisticated experimental techniques to shape the corresponding control fields @xcite .
all these efforts have lead nowadays to an unexpected and satisfactory agreement between theory and experiment . on the theoretical side ,
one major tool to design the control field is optimal control theory ( oct ) @xcite . over the past few years
, numerical iterative methods have been developed in quantum control to solve the optimization problems .
basically , they can be divided into two families , the gradient ascent algorithms @xcite and the krotov @xcite or the monotonic @xcite ones .
the design of optimal control fields by standard iterative algorithms @xcite can require the computation of several hundreds numerical propagations of the dynamical quantum system .
while the efficiency of this procedure has been established for low dimensional quantum systems , this approach can numerically be prohibitive for large dimensions . in this latter case
, it is possible to use more easily accessible numerical methods , such as the local control approach @xcite . roughly speaking
, the optimization procedure is built from a lyapunov function @xmath0 over the state space , which is minimum ( or maximum ) for the target state .
a control field that ensures the monotonic decrease ( or increase ) of @xmath0 is constructed with the help of the first derivative @xmath1 of @xmath0 .
note that this approach has largely been explored in quantum control @xcite .
thanks to the progresses in numerical optimization techniques , it is now possible to design high quality control along with some experimental imperfections and constraints .
recent studies have shown how to extend the standard optimization procedures in order to take into account some experimental requirements such as the spectral constraints @xcite , the non - linear interaction between the system and the laser field @xcite , the robustness against experimental errors @xcite . in view of experimental applications in quantum control , it is also desirable to design pulse sequences with a zero global time - integrated area .
several works have pointed out the inherent experimental difficulties associated with the use of non - zero area fields @xcite , in particular for laser fields in the thz regime .
since the dc component of the field is not a solution of maxwell s equation , such pulses are distorted when they propagate in free space as well as through focusing optics .
the standard optimization procedures do not take into account this basic requirement , designing thus non - physical control fields . in this framework ,
a question which naturally arises is whether one can adapt the known optimization algorithms to this additional constraint .
this paper aims at taking a step toward the answer of this open question by proposing new formulations of optimal and local control algorithms .
the zero - area requirement for the laser fields is mathematically fulfilled by the introduction of a lagrange multiplier and the derivation of the corresponding optimal equations .
the goal of this paper is to explore the efficiency of the generalized optimal and local control algorithms on two key control problems of molecular dynamics : orientation and photodissociation .
the remainder of this paper is organized as follows .
the new formulations of optimization control algorithms are presented in sec .
section [ sec3 ] is devoted to the application of the optimal control algorithm to the enhancement of molecular orientation of co by thz laser fields at zero temperature .
local control is used in sec .
[ sec4 ] to manipulate efficiently the photodissociation of heh@xmath2 .
conclusion and prospective views are given in sec .
we consider a quantum system whose dynamics is governed by the following hamiltonian : @xmath3 where @xmath4 is the control field .
the state @xmath5 of the system satisfies the differential equation : @xmath6 with @xmath7 the initial ( @xmath8 ) state .
the units used throughout the paper are atomic units .
as mentioned in the introduction , we add the physical constraint of zero - area on the control field : @xmath9 where @xmath10 is the control ( or total pulse ) duration .
the goal of the control problem is to maximize the following standard cost functional at time @xmath11 : @xmath12 -\lambda \int_0^{t_f } e(t)^2dt , \label{eq : joc}\ ] ] where @xmath13 is the target state .
other cost functionals could of course be chosen .
more specifically , a monotonic optimal control algorithm can be formulated to satisfy the zero - area constraint .
for such a purpose let us consider the following cost functional @xmath14-\mu \left[\int_0^{t_f}{e(t)dt}\right]^2-\lambda \int_0^{t_f } { [ e(t)-e_{ref}(t)]^2/s(t)dt}\ , , \label{eq : newjoc}\ ] ] where @xmath15 is a reference pulse and @xmath16 an envelope shape given by @xmath17 @xcite .
the function @xmath16 ensures that the field is smoothly switched on and off at the beginning and at the end of the control .
the positive parameters @xmath18 and @xmath19 , expressed in a.u . , weight the different parts of @xmath20 , which penalize the area of the field ( @xmath18- term ) and its energy ( @xmath19- term ) . in this algorithm , we determine the field @xmath21 at step @xmath22 from the field @xmath23 at step @xmath24 , such that the variation @xmath25 . at step @xmath24 , the reference field @xmath15 is taken as @xmath26 and we denote by @xmath27 its time - integrated area .
note that the choice @xmath28 leads to a smooth transition of the control field between two iterations of the algorithm @xcite .
the variation @xmath29 can be expressed as follows : @xmath30-\lambda ( e_{k+1}-e_k)/s(t)-2\mu a_k ( e_{k+1}-e_k)]dt,\ ] ] @xmath31 is obtained from backward propagation of the target @xmath32 taken as an intial state for eq.(2 ) and we assume that @xmath33 .
one deduces that the choice : @xmath34}{\lambda}-(e_{k+1}-e_k)-2\frac{\mu}{\lambda}a_k s(t)\ ] ] ensures a monotonic increase of @xmath35 .
this leads to the following correction of the control field : @xmath36}{2\lambda}-\frac{\mu}{\lambda}s(t)a_k .
\label{eq : newfield}\ ] ] the monotonic algorithm can thus be schematized as follows : 1 .
guess an initial control field , which at step @xmath24 is taken as @xmath37 .
2 . starting from @xmath38 , propagate @xmath31 backward with @xmath26 3 .
evaluate the correction of the control field @xmath39 as obtained from eq .
( [ eq : newfield ] ) , while propagating forward the state @xmath40 of the system from @xmath7 with @xmath21 .
4 . with the new control ,
go to step 2 by incrementing the index @xmath24 by 1 .
the same formalism can be used to extend the local control approach to the zero - area constraint .
we refer the reader to @xcite for a complete introduction of this method . using the same notations as in the previous section , we consider as lyapunov function @xmath41 the cost functional : @xmath42 where @xmath43 is any operator such that @xmath44 $ ] .
the cost @xmath41 is maximum at a given time @xmath10 if @xmath45 is maximum and @xmath46 .
the computation of the time derivative of @xmath41 gives : @xmath47}{i}|\psi(t ) \rangle
-2\mu a(t ) e(t).\ ] ] one can choose : @xmath48}{i}|\psi(t ) \rangle -2\mu
a(t)\big ) , \label{eq : fieldlc}\ ] ] to ensure the monotonic increase of @xmath41 from the initial state of the system @xmath7 .
the parameter @xmath49 is a small positive parameter which is adjusted to limit the amplitude of the control field . in order to illustrate the efficiency of the algorithm with the zero - area constraint , we introduce two measures : @xmath50 where @xmath51 and @xmath52 are the optimized field with and without the zero - area constraint , respectively
in this section , we present the theoretical model for the control of molecular orientation by means of thz laser pulses @xcite . we consider the linear co molecule described in a rigid rotor approximation and driven by a linearly polarized electric field @xmath53 along the @xmath54- axis of the laboratory frame .
the molecule is assumed to be in its ground vibronic state .
the hamiltonian of the system is given by : @xmath55 the first term is the field - free rotational hamiltonian , where @xmath56 is the rotational constant of the molecule and @xmath57 the angular momentum operator .
the second term represents the interaction of the system with the laser field , the parameter @xmath58 being the permanent molecular dipole moment .
the spatial position of the diatomic molecule is given in the laboratory frame by the spherical coordinates @xmath59 . due to the symmetry of revolution of the problem with respect to the @xmath54- axis
, the hamiltonian @xmath60 depends only on @xmath61 , the angle between the molecular axis and the polarization vector .
numerical values of the molecular parameters are taken as @xmath62 and @xmath63 . in the case of zero rotational temperature ( @xmath64 ) which is assumed hereafter ,
the system is described at time @xmath65 by the wave function @xmath66 and its dynamics by the time - dependent schrdinger equation ( eq . ( [ eq : dyn ] ) ) .
the dynamics is solved numerically using a third - order split operator algorithm @xcite .
we finally recall that the degree of orientation of the molecular system is given by the expectation value @xmath67 .
in this paragraph , we test the new formulation of the optimal control algorithm presented in sec .
[ secopti ] on the specific example of co orientation dynamics .
the initial state is the ground state denoted @xmath68 in the spherical harmonics basis set @xmath69 . due to symmetry
, the projection @xmath70 of the angular momentum on the @xmath71axis is a good quantum number .
this property leads to the fact that only the states @xmath72 , @xmath73 , will be populated during the dynamics .
in the numerical computations , we consider a finite hilbert space of size @xmath74 .
note that this size is sufficient for an accurate description of the dynamics considering the intensity range of the laser field used in this work . to define the cost functional , rather than maximizing the expectation value @xmath75 , we focus on a target state @xmath38 maximizing this value in a finite - dimensional hilbert space @xmath76 spanned by the states @xmath77 with @xmath78
. the details of the construction of @xmath38 can be found in refs .
@xcite . here
, the parameter @xmath79 is taken to be @xmath80 . the guess field is a gaussian pulse of 288 fs full width at half maximum , centered at @xmath81 .
the optimization time , @xmath10 , is taken to be the rotational period @xmath82 of the molecule .
the parameter @xmath19 defined in the cost functional is fixed to 100 .
both , with and without zero - area constraint , the optimized field achieves a fidelity higher than 99@xmath83 .
the results are shown in fig.[fig : jt_area ] .
note that , since the ratio @xmath84 has the dimension of time , we found convenient to plot the evolution of the measures with respect to @xmath85 ( panel ( b ) ) .
the upper right panel , fig.[fig : jt_area](a ) , shows the deviation from unity of the fidelity ( or final objective ) @xmath86 as a function of the number of iterations . from unity as a function of the number of iterations for @xmath87 corresponding to the optimization without zero - area constraint ( dashed dotted blue ) and for @xmath88 ( solid black @xmath89 , dashed black @xmath90 ) .
( b ) plot of @xmath91 ( solid black ) and @xmath92 ( solid green ) as a function of the parameter @xmath18 times the final time @xmath10 .
( c ) the guess field ( dashed red ) and the optimized fields without ( dotted dashed blue ) and with ( solid black ) zero - area constraint .
( d ) comparison of the time evolution of the expectation value of @xmath93 with the guess field ( dashed red ) and with the fields without ( dotted dashed blue ) and with ( solid black ) zero - area constraint . ]
as could be expected , optimization with zero - area constraint requires a large number of iterations to reach a high fidelity . for @xmath87 ,
i.e. without zero - area constraint , the final cost reaches a value of 0.99 after 30 iterations , while in the case for which @xmath94 , this value is reached after 40 iterations .
the variations of the @xmath91 and @xmath92 with respect to the parameter @xmath85 are shown in fig.[fig : jt_area](b ) . for @xmath95 ,
the area of the optimized field obtained with the zero - area constraint is two orders of magnitude smaller than the one obtained without the zero - area constraint .
note that the values of @xmath91 and @xmath92 displayed in fig.[fig : jt_area](b ) are computed with optimized fields which achieve a fidelity higher than 99@xmath83 .
figure [ fig : jt_area](c ) compares the guess ( reference ) field and the optimized fields with and without the zero - area constraint . the optimized field with the zero - area constraint shown in fig . [ fig : jt_area](c ) is obtained with @xmath96 .
the optimized fields are similar to the guess field at the beginning of the control and become slightly different for @xmath97 .
figure [ fig : jt_area](d ) corresponds to the time evolution of @xmath98 during two rotational periods ( one with the field and one without ) .
it compares the dynamics of @xmath99 induced by the three fields shown in the lower left panel .
clearly , the dynamics with the guess field is very different from the one obtained with the optimized fields , which have similar features .
we consider the photodissociation of heh@xmath2 through the singlet @xmath100 excited states coupled by numerous nonadiabatic interactions .
a local control strategy has recently been applied to find a field able to enhance the yield in two specific dissociation channels either leading to he@xmath101 , or to he@xmath102 @xcite .
the diatomic system is described by its reduced mass @xmath70 and the internuclear distance @xmath103 . due to the short duration of the pulses (
as compared with the rotational period ) a frozen rotation approximated is valid .
in addition , the molecule is assumed to be pre - aligned along the @xmath54-direction of the laboratory frame .
the initial state @xmath104 is the vibrationless ground electronic adiabatic state .
dynamics is performed in the diabatic representation .
the total hamiltonian is written as : @xmath105 where @xmath106 is the field free hamiltonian , @xmath107 is the number of diabatic electronic states @xmath108 under consideration and @xmath109 .
all the adiabatic potential energy curves and the nonadiabatic couplings @xmath110 have been computed in ref .
@xmath111 are the diabatized dipole transition matrix elements .
we shall consider only parallel transitions among the singlet @xmath100 states induced by the @xmath112 dipole operator by assuming the internuclear axis pointing along the @xmath54-direction .
the adiabatic - to - diabatic transformation matrix @xmath113 has been derived by integrating @xmath114 from the asymptotic region where both representations coincide .
the local control of a nonadiabatic dissociation requires a careful choice of the operator @xmath43 referred to in eq .
( [ eq : jlc ] ) since it has to commute with the field free hamiltonian @xcite . in the nonadiabatic case ,
the projectors on either adiabatic or diabatic states are thus not relevant as they do not commute with this hamiltonian due to kinetic or potential couplings respectively .
this crucial problem can be overcome by using projectors on eigenstates of @xmath106 , i.e. on scattering states correlating with the controlled exit channels . in this example
, the operator @xmath43 takes the form : @xmath115 where @xmath116 represents the two channels leading to the target he@xmath117 fragments , the objective being @xmath118 .
the local control field now reads @xmath119 involving two adjustable parameters @xmath120 and @xmath18 . ,
dashed red : @xmath18 = 0.05 , dotted blue : @xmath18 = 0.10 .
the parameter @xmath18 is expressed in atomic units .
( b ) objectives of the local control measured by the total he@xmath117 yield ( same color code than in ( a ) ) .
( c ) @xmath121 ( solid black ) and @xmath122 ( solid green ) as a function of @xmath18 .
( d ) fields after filtration of the low frequencies of the spectrum removing the static stark component .
( e ) objectives with the filtered fields .
( f ) @xmath121 and @xmath122 ( same color code than in ( c ) ) as a function of @xmath18 for the filtered fields . ]
the ingoing scattering states @xmath123 are estimated using a time - dependent approach based on mller - operators @xcite as derived in ref .
this control strategy remains local in time but can preemptively account for nonadiabatic transitions that occur later in the dynamics .
the photodissociation cross section @xcite shows that there is no spectral range where the he@xmath117 dissociation channels dominate .
the local control finds a very complicated electric field which begins by an oscillatory pattern followed after 10 fs by a strong complex positive shape whose area is obviously not zero ( see fig . [
fig : heh_area](a ) ) .
we therefore choose this example to check the efficiency of the zero - area constraint algorithm .
we use @xmath124 4.2 and different values of @xmath18 to assess the effect of the zero - area constraint on the control field .
figure [ fig : heh_area](a ) shows the dramatic evolution of the pulse profile for two values of @xmath18 ( 0.05 and 0.10 ) .
the main change is the suppression of the erratic positive structure which can be interpreted as a stark field .
however , as shown in fig .
[ fig : heh_area](c ) the variation of @xmath122 is not monotonic and the best correction is actually obtained for a given value of @xmath18 around @xmath125 , leading to almost zero value for @xmath122 .
close to the optimal @xmath18 , one observes a clear improvement of the zero - area constraint , but at the price of a decrease of the objective ( from @xmath126 without constraint to @xmath127 with @xmath128 ) as shown in fig .
[ fig : heh_area](b ) .
however , a brute force strategy by filtration of near - zero frequencies already provides a large correction to the non vanishing area while also decreasing the yield . the field with @xmath129 and after filtration of the low frequencies is shown in black line in fig .
[ fig : heh_area](d ) and the corresponding yield is given in fig .
[ fig : heh_area](e ) .
the objective is divided by a factor 2 .
we also filter the fields generated by the constrained control with @xmath18 equal to 0.05 and 0.10 which improves even further the zero - area constraint .
the measures @xmath122 evaluated with the filtered fields ( fig . [ fig : heh_area](f ) ) are smaller than the unfiltered ones by about two orders of magnitude , while the objectives are reduced by about a factor 1.5 , as seen by comparing fig .
[ fig : heh_area](e ) and fig .
[ fig : heh_area](b ) .
we have presented new formulations of optimization algorithms in order to design control fields with zero area .
the procedure is built from the introduction of a lagrange multiplier aiming at penalizing the area of the field .
the value of the corresponding parameter is chosen to express the relative weight between the area and the other terms of the cost functional , i.e. the projection onto the target state and the energy of the electric field .
the efficiency of the algorithms is exemplified on two key control problems of molecular dynamics : namely , the molecular orientation of co and the photodissociation of heh@xmath2 . from a theoretical view point , the zero - area constraint is crucial since all electromagnetic fields which are physically realistic must fulfill this requirement .
surprisingly , this physical requirement has been considered , up to date , by only very few works in quantum control . as
such , this work can be viewed as a major advance of the state - of - the - art of optimization algorithms that could open the way to promizing future prospects .
s. v. acknowledges financial support from the fonds de la recherche scientifique ( fnrs ) of belgium . financial support from the conseil rgional de bourgogne and the quaint coordination action ( ec fet - open )
is gratefully acknowledged by d. s. and m. n .. | we propose a new formulation of optimal and local control algorithms which enforces the constraint of time - integrated zero - area on the control field .
the fulfillment of this requirement , crucial in many physical applications , is mathematically implemented by the introduction of a lagrange multiplier aiming at penalizing the pulse area .
this method allows to design a control field with an area as small as possible , while bringing the dynamical system close to the target state .
we test the efficiency of this approach on two control purposes in molecular dynamics , namely , orientation and photodissociation . |
the phase - field method has become the method of choice for simulating microstructure formation during solidification .
it owes its popularity mainly to its algorithmic simplicity : the cumbersome problem of tracking moving solid - liquid interfaces or grain boundaries is avoided by describing the geometry in terms of one or several phase fields .
the phase fields obey simple partial differential equations that can be easily coded by standard numerical methods .
the foundations of the phase - field method and its application to solidification have been the subject of several recent review articles @xcite , and it seems of little use to repeat similar information here . instead , in this paper
several topics are discussed where robust phase - field modelling tools are not yet available because some fundamental questions remain open . in sec .
[ sec2 ] , the thin - interface limit of two - sided phase - field models is examined , and it is shown that the currently available approaches can not in general eliminate all effects linked to the finite interface thickness . in sec .
[ sec3 ] , orientation - field models for polycrystalline solidification are discussed , and it is shown that the standard equation of motion usually written down for the orientation field is not appropriate for the evolution of coherent crystalline matter .
finally , in sec . [ sec4 ] , the inclusion of microscopic fluctuations in the phase - field equations is reviewed , and it is shown that the standard approach can not be used in a straightforward way to investigate the process of nucleation . the common point of these topics is that they pose challenges or limitations for straightforward computations .
indeed , a characteristic feature of the phase - field method is that its equations can often be written down following simple rules or intuition , but that their detailed properties ( which have to be known if quantitative simulations are desired ) become only apparent through a mathematical analysis that can be quite involved .
therefore , it is not always easy to perceive the limits of applicability of the method .
it is hoped that the present contribution will be helpful to point out some pitfalls and to stimulate further discussions that will facilitate the solution of these issues .
the precision and performance of phase - field models have been greatly enhanced in the last decade by a detailed control of their properties .
phase - field models are rooted in the mean - field description of spatially diffuse interfaces by order parameters . however , to be useful for simulating microstructure formation in solidification , phase - field models need to bridge the scale gap between the thickness of the physical solid - liquid interfaces and the typical scale of the microstructures .
this is achieved by increasing the interface width in the model , sometimes by several orders of magnitude .
obviously , this procedure magnifies any physical effect that is due to the diffuseness of the interface .
therefore , to guarantee precise simulations , all these effects have to be controlled and , if possible , eliminated . the privileged tool to achieve
this is the so - called _ thin - interface limit _ :
the equations of the phase - field model are analysed under the assumption that the interface thickness is much smaller than any other physical length scale present in the problem , but otherwise arbitrary .
the procedure of matched asymptotic expansions then yields the effective boundary conditions valid at the macroscale , which contain all effects of the finite interface thickness up to the order to which the expansions are carried out .
this procedure was pioneered by karma and rappel , who analysed the symmetric model of solidification ( equal diffusion constants in the solid and the liquid ) and obtained a thin - interface correction to the expression of the kinetic coefficient @xcite .
the use of this result has made it possible to carry out quantitative simulations of free dendritic growth of a pure substance , both at high and low undercoolings @xcite .
it turned out , however , that the generalisation of this method to a model with arbitrary diffusivities is far from trivial @xcite , since several new thin - interface effects appear , which can not all be eliminated simultaneously .
a solution to this problem was found later for the case of the one - sided model ( zero diffusivity in the solid ) with the introduction of the so - called antitrapping current @xcite , and it was shown that quantitative simulations of alloy solidification are possible with this model @xcite , including multi - phase @xcite and multi - component alloys @xcite .
recently , several extensions of the antitrapping current were put forward to generalise the approach to the case of finite diffusivity in the solid @xcite , and simulations were presented which show that the approach works well for the instability of a steady - state planar interface @xcite and for free dendritic growth @xcite . however , as will be shown below , this is only a partial solution to the problem of developing a general quantitative model , since there is a second , independent thin - interface effect that can not be removed by an antitrapping current , namely , the kapitza resistance . for the sake of concreteness , consider the standard phase - field model for the solidification of a pure substance as discussed in refs .
the evolution equation for the phase field reads @xmath0 where @xmath1 is the phase field , with @xmath2 and @xmath3 corresponding to solid and liquid , respectively , @xmath4 is the relaxation time of the phase field , @xmath5 is the interface thickness , and @xmath6 is a dimensionless coupling constant . the field @xmath7 is a dimensionless temperature defined by @xmath8 , where @xmath9 , @xmath10 and @xmath11 are the melting temperature , latent heat , and specific heat , respectively .
it is assumed for simplicity that @xmath11 is the same in both phases .
the temperature is governed by a diffusion equation with a source term , @xmath12 + \frac 12 \partial_t h(\phi ) .
\label{diffusion}\ ] ] here , @xmath13 , which satisfies @xmath14 , is a function that describes the release or consumption of latent heat during the phase transition , and @xmath15 interpolates between the thermal diffusivities of the liquid and the solid , @xmath16 and @xmath17 , @xmath18 where the interpolation function @xmath19 satisfies @xmath20 and @xmath21 . for simplicity ,
crystalline anisotropy has not been included in the above model because it is not necessary for the present discussion .
furthermore , the equations have been stated in the language of a two - sided thermal model , but with some modifications ( as detailed in refs .
@xcite ) , they also apply to the isothermal solidification of a binary alloy . in this case , @xmath7 is a dimensionless chemical potential ( conjugate to the concentration of one of the alloy components ) , and @xmath15 is the chemical diffusivity . in the following
, two simple one - dimensional solutions of these equations will be analysed .
the first is a steady - state planar front that propagates with constant velocity @xmath22 in the positive @xmath23 direction into a liquid of undercooling @xmath24 ( @xmath25 for @xmath26 ) , and leaves behind a constant temperature .
this solution only exists if the liquid is undercooled beyond the hypercooling limit , that is , @xmath27 .
the sharp - interface solution to this problem is readily obtained and reads @xmath28 for an interface located at @xmath29 ( in the frame moving with the interface ) . here ,
@xmath30 and @xmath31 are the limit values of the temperature when the interface is approached from the solid and the liquid side , respectively . in the standard formulation of the free boundary problem of solidification
, it is assumed that the temperature is the same on the two sides of the interface , @xmath32 . then , the use of the two boundary conditions @xmath33 , where @xmath34 is the linear kinetic coefficient , and @xmath35 ( the stefan boundary condition ) determines the solution , @xmath36 ( a simple consequence of heat conservation ) , and @xmath37 .
the phase - field equations can be analysed and related to this sharp - interface solution by the method of matched asymptotic expansions in the limit where the interface thickness @xmath5 is much smaller than the diffusion length @xmath38 .
this calculation has been presented in detail in refs .
@xcite and will not be repeated here .
the essential outcome is that , in general , the two asymptotes of the bulk phases do _ not _ correspond to the same temperature . the difference is given , to the lowest order , by @xmath39 , \label{tjump1}\ ] ] where @xmath40 is the equilibrium profile of the phase field .
the physical interpretation of this temperature jump is _ trapping _ : when the diffusivity decreases upon solidification , the heat generated at the rear of the interface gets trapped . in the alloy version of the model ,
this is nothing but the well - known solute trapping effect . indeed , in sharp - interface models of alloy solidification the chemical potential exhibits a jump at the interface when solute trapping occurs . in the phase - field model ,
the temperature profile through the interface is determined by the interplay between the rejection of latent heat and the diffusion away from the interface ; therefore , it is natural that the heat source function @xmath13 and the diffusivity function @xmath15 appear in eq .
( [ tjump1 ] ) .
whereas , thus , this discontinuity is physically correct , it generates problems for simulations . to see this , is is sufficient to rewrite eq .
( [ tjump1 ] ) in order to make the relevant scales apparent . since the only length scale in eq .
( [ pf ] ) is the interface thickness @xmath5 , the equilibrium solution @xmath41 is a function only of the reduced variable @xmath42 . using this together with the interpolation of @xmath15 given by eq .
( [ ddef ] ) , eq .
( [ tjump1 ] ) becomes @xmath43 @xmath44\ ;
d\eta\qquad { \rm and } \label{fdef}\ ] ] @xmath45 the temperature jump is thus proportional to the velocity , the interface thickness , and the difference of the two integrals ; the latter depends only on the choice of the interpolation functions . if @xmath5 is the physical interface thickness ( a few angstroms ) , this effect is negligibly small , but if the interface thickness is increased by a large factor to make simulations feasible , this leads to potentially large errors in the simulations . as discussed in detail in refs .
@xcite , it is not possible to eliminate this macroscopic discontinuity simply by the choice of appropriate interpolation functions , due to other constraints not discussed here .
the solution put forward in ref .
@xcite and further developed in ref .
@xcite is the introduction of an antitrapping current : eq .
( [ diffusion ] ) is replaced by @xmath46 where the antitrapping current @xmath47 is given by @xmath48 where @xmath49 is a shorthand for the time derivative @xmath50 , @xmath51 is the unit normal vector to the interface , and @xmath52 is a new interpolation function .
this term induces a current which is directed from the solid to the liquid , and proportional to the interface velocity ( through the factor @xmath49 ) .
it thus `` pushes '' heat from the solid to the liquid side of the interface when the interface moves , and can be used to adjust the temperature jump at the interface . for the one - sided model ( @xmath53 ) with the standard choices @xmath54 and @xmath55 , it was shown that a constant @xmath56 leads to a vanishing jump in @xmath7 , because it modifies the function @xmath57 in eq .
( [ fdef ] ) such that @xmath58 .
thus , continuity of the temperature between the two sides of the interface ( local equilibrium ) is restored for arbitrary @xmath5 and @xmath22 , as long as the asymptotic analysis remains valid .
recently , several authors have put forward generalisations of this approach @xcite for arbitrary ratio of the diffusivities . for the case
analysed above ( that is , the current far inside the solid vanishes ) , they reduce to the simple prescription that the same expression for the antitrapping current can be used , but with an additional prefactor that can be written as @xmath59 , @xmath60 indeed , the asymptotic analysis shows @xcite that in this way the temperature jump can be eliminated .
however , this is not the only thin - interface effect that can arise in the two - sided case . to see this , consider now a different situation , namely an immobile interface in a temperature gradient .
such an interface can be easily obtained in experiments by maintaining a pure substance between two walls which are held below and above the melting temperature , respectively . when the interface is stationary , @xmath61 by definition , and eq .
( [ diffusion ] ) implies that the system is crossed by a constant heat current flowing from the liquid into the solid , @xmath62 with @xmath63 a positive constant . as before ,
the centre of the interface is located at @xmath29 , and the solid is located in the domain @xmath64 .
this situation can be analysed without performing a perturbation expansion , since it is sufficient to integrate eq .
( [ current ] ) to obtain a solution for @xmath7 , @xmath65 where @xmath66 is the temperature at @xmath29 .
the sharp - interface solution for this case is simply given by @xmath67 matching the asymptotes of the phase - field and sharp - interface expressions , it is straightforward to show that there is again a temperature jump given by @xmath68,\ ] ] this time proportional to the _
current_. if the phase - field profile is replaced by its equilibrium shape , this can be rewritten as @xmath69 with @xmath70 this temperature jump corresponds to a surface thermal resistance , also called kapitza resistance , first found for an interface between liquid helium and metal @xcite .
indeed , in a sharp - interface picture it is generally necessary to assign a surface resistance to an interface for a complete description of heat transfer , because transport through an interface can be decomposed into three elementary steps : transport in one bulk phase , crossing of the interface , and transport in the other phase .
the surface resistance describes the kinetics associated with the crossing of the interface ( its inverse is sometimes referred to as the interfacial transfer coefficient ) .
it is characterised either by the value of the resistance , @xmath71 , or by a length that is obtained by dividing this resistance by the conductivity of the liquid phase . here , this characteristic length is simply @xmath72 , which is of the order of the interface thickness . since
this quantity is actually an interface excess of the inverse diffusivity ( in complete analogy to the interface excesses for equilibrium quantities obtained by the well - known gibbs construction ) , it can also be negative this does not violate the laws of thermodynamics because the _ local _ transport coefficients are strictly positive .
if the surface resistance is finite , the temperature in the sharp - interface model is _ not _ continuous at the interface , but exhibits a jump that is proportional to the current crossing the interface . in the alloy version of the model
, this corresponds to a jump in chemical potential that is proportional to the solute flux @xcite .
such discontinuities have been thoroughly investigated @xcite , and can be measured in experiments @xcite and detected in molecular dynamics simulations @xcite for solid - liquid interfaces .
thus , like the trapping effect , the surface resistance is a natural effect that is proportional to the interface thickness .
if the interface thickness is to be upscaled , it should therefore also be eliminated .
however , is is immediately clear that this effect can not be eliminated by any antitrapping current proportional to @xmath49 as given by eq .
( [ antinew ] ) : since the interface does not move , @xmath73 and the antitrapping current vanishes , independently of the current @xmath63 that crosses the interface .
the authors of both refs .
@xcite have recognised the importance of the current @xmath63 .
they have developed generalised expressions for the antitrapping current with coefficients that depend on the value of @xmath63 .
as long as the interface velocity remains non - zero , the formal asymptotic analysis shows that it is still possible to eliminate the temperature jump .
however , for a fixed current @xmath63 , the expressions of the coefficients diverge when @xmath22 tends to zero , such that the asymptotic analysis is not valid in this limit .
thus , it seems unlikely that this approach can be used as a robust method for simulations . in summary , there exist two independent thin - interface effects , one proportional to @xmath22 , and one proportional to @xmath63 . on a very fundamental level , this is just the consequence of the fact that the interface motion is driven by a diffusion equation , which has two independent boundary conditions .
the corresponding physical quantities are the currents on the two sides of the interface , or one current and the velocity . a general solution to eliminate both thin - interface effects ( which are linearly independent ) does not seem to exist at this moment , but the above considerations can at least be used to obtain simple criteria when the prescription of eq .
( [ antinew ] ) can be used .
indeed , eqs .
( [ tjump1a ] ) and ( [ tjump2 ] ) show that if @xmath74 ( note that , since @xmath7 is dimensionless , @xmath63 has the dimension of a velocity ) , the kapitza effect is much smaller than the trapping effect , and can thus be neglected .
this is generally the case for equiaxed dendritic growth , in which the gradients outside the growing dendrite , which determine the growth speed , are much larger than the gradients inside the solid .
indeed , it was shown in ref . @xcite that eq .
( [ antinew ] ) works well in this case .
however , problems might arise in the case of alloy solidification in a temperature gradient or for multicomponent alloys with widely different solute diffusivities , since in this case large currents of heat or certain solutes may cross an interface whose velocity is controlled by a different diffusion field .
such cases have to be critically examined before simulation results can be trusted .
the size and shape of the crystalline grains formed upon solidification is one of the most important factors that determine materials properties .
therefore , phase - field models that are to be helpful for materials design must be capable of dealing with the evolution of polycrystals , both during solidification of individual columnar or equiaxed grains from the melt and during the subsequent evolution of the grain structure after impingement .
this can be achieved using the multi - phase - field approach @xcite , in which each grain is represented by a different phase field , even if they are of the same thermodynamic phase .
the properties of each individual grain boundary or interface can then be specified separately @xcite , and it has been demonstrated that good quantitative control of the grain boundary properties can be achieved @xcite .
the problem of handling several hundreds or even thousands of phase fields simultaneously can be solved by recognising that only a small number of fields are important at any given point of space ( see for example @xcite ) .
an alternative approach is the orientation - field method .
its starting point is the remark that it would be desirable , both for efficiency and simplicity , to formulate a model that works only with a small number of field variables .
indeed , the orientation of a crystal can be described by one scalar quantity ( an angle ) in two dimensions , and three scalars in three dimension ( for instance , the euler angles ) .
orientation - field models for pure substances in two dimensions that work with a single phase field , an orientation field ( the local angle of the crystalline structure with respect to a fixed coordinate system ) , and the temperature field were put forward in refs .
@xcite , and generalised for alloy solidification @xcite and to three dimensions @xcite .
while these models are elegant and simple in their formulation and therefore hugely appealing , it is pointed out here that the evolution equation of the angle field , which takes the form of a simple relaxation equation , does not correctly describe the microscopic evolution of the orientation field since it does not take into account the connectivity of matter and the resulting geometrical conservation laws . for simplicity , anisotropy and crystallographic effects
will again be neglected , and it is sufficient to consider a two - dimensional system .
the dimensionless free energy of the orientation - field model is @xcite @xmath75 d\vec r , \label{modeli}\ ] ] where now @xmath76 and @xmath2 in the liquid and the solid , respectively , @xmath77 and @xmath78 are positive constants , @xmath79 and @xmath80 are monotonous functions that satisfy @xmath81 and @xmath82 , and @xmath83 is the local free energy density , with @xmath7 the same dimensionless temperature field as previously ; the standard choice is @xmath84 .
recently , an alternative model was developed @xcite , @xmath85 d\vec r , \label{modelii}\ ] ] where @xmath86 is a constant . in the following
, these models will be called model i and model ii .
they both have some features that distinguish them from standard phase - field models .
model i contains a term proportional to @xmath87 , which has a singular derivative at @xmath88 .
model ii has only a regular square gradient term in @xmath89 , but it is multiplied by a singular function of the phase field @xmath1 , which diverges in the limit @xmath90 ( the solid ) .
these singular features are needed to create stable grain boundary solutions , that is , localised spatial regions where the phase field departs from its solid value and the angle field exhibits rapid variations .
both models have a variational structure for the dynamics of the phase field and the angle field , that is @xmath91 @xmath92 which means that both @xmath1 and @xmath93 evolve such as to follow the gradient of the free energy , with @xmath94 and @xmath95 being the corresponding mobilities ( which may be functions of the fields ) . in the following
, it will be shown that eq .
( [ thetaeq ] ) is incorrect for coherent crystalline matter . to illustrate the problems with this equation of motion ,
it is again useful to analyse a simple one - dimensional situation , which is a tricrystal .
a slab of crystalline orientation @xmath96 is sandwiched between two crystals of identical orientation @xmath97 , as shown in the left side of fig .
the two crystals on the sides of the system are assumed to be clamped to a substrate , that is , @xmath97 for all times . in both models
, this initial condition evolves with time : the orientation of the central slab remains homogeneous , but changes with time to approach the orientation of the outer crystals .
the final state is a uniform solid of orientation @xmath97 : the central slab has disappeared .
of course , this process can take place since it corresponds to a minimisation of the free energy : the two grain boundaries with their positive grain boundary energy are eliminated .
however , the pathway of this dynamics is not appropriate for the evolution of a coherent crystal .
in fact , eq .
( [ thetaeq ] ) corresponds to the dynamics of matter which has orientational , but no positional order , such as a liquid crystal .
indeed , if in model i the term proportional to @xmath87 is omitted or in model ii the singular coupling function is replaced by a regular one , the resulting model can be mapped to the standard landau - de gennes model for nematic liquid crystals in two dimensions @xcite .
the free energies in eqs .
( [ modeli ] ) and ( [ modelii ] ) have been designed to stabilise grain boundaries , which do not exist in a nematic liquid crystal .
the energetics of the models are thus quite different from liquid crystals . in contrast
, the type of the dynamics has stayed the same . to understand where is the difference in dynamics between liquid crystals and crystals , consider the elongated molecules of a nematic liquid crystal characterised by a director field of a certain orientation @xmath96 . since the molecules have no bonds ,
it is possible to change the local orientation while keeping the centres of mass fixed , by just making each molecule rotate around its centre of mass ( of course , in a dense liquid crystal , this exact procedure is not possible because of steric exclusion , but the director can still be changed with only short - range displacements of the centres of the molecules ) .
the system is thus free to _ locally _ change orientation in order to lower its free energy , and thus follows eq .
( [ thetaeq ] ) .
this is obviously not the case in crystalline matter : it is not possible to rotate a unit cell without displacing the surrounding neighbours , because bonds ( or , more generally , the positional ordering of elements ) define a connectivity .
it is easy to grasp that the evolution depicted in fig.[fig1 ] is impossible if the connectivity of the central slab is preserved .
thus , a consistent evolution equation for @xmath93 has to take into account this connectivity , or , in other words , the evolution of the positions .
this is , in general , a complicated undertaking .
two elementary situations where it easy to obtain an equation are ( i ) rigid body rotation , in which case the ( advected ) time derivative of the local angle is given by the curl of the local velocity field , or ( ii ) purely elastic deformations of the solid , in which case the orientation is not an independent quantity but can be deduced from the elastic displacement field . here
, a third possibility will be briefly discussed , namely , plastic deformation .
this corresponds precisely to a change in the connectivity of matter .
if the matter in question can be considered reasonably crystalline ( as opposed to , for example , an amorphous material ) , its geometry can formally always be described by a density of dislocations , which are singularities of the displacement field if a perfect crystal is taken as the reference state .
if , furthermore , grain boundaries remain coherent ( that is , no grain boundary sliding takes place ) , the evolution of the local orientation can be linked to the motion of dislocations .
a complete description is far outside of the scope of this article ; the interested reader is referred to ref .
@xcite for a detailed introduction to the continuum theory of defects . here
, only two simple examples will be qualitatively treated for illustration .
consider again the tricrystal configuration . in the sketch shown in fig .
[ fig2 ] , only one set of crystal planes is shown for clarity , and the central slab has a small misorientation with respect to the outer crystals .
in this situation , the two low - angle grain boundaries consist of individual edge dislocations . the inner crystal can now rotate by an elementary process : take one of the edge dislocations of the left grain boundary ( marked by a circle ) and make it glide towards the other grain boundary .
this process involves only local reconnection events .
when the dislocation arrives at the right grain boundary , it can annihilate with a dislocation of the opposite sign . as a result
, one dislocation has disappeared from each grain boundary .
of course , this process can repeat itself until no dislocation is left , and the grain boundaries have disappeared .
it should be stressed that this pathway for rearrangement exhibits large energy barriers , since the elastic energy of a single dislocation is much higher in the centre of the slab than at its original position within the grain boundary .
therefore , if only thermal fluctuations are driving this process ( no external strains ) , it will be extremely slow . on a more quantitative level
, the misorientation through a grain boundary is linked to the density of dislocations by simple geometrical arguments .
therefore , it is natural that the misorientation is lowered when the dislocation density in the grain boundaries decreases .
furthermore , it is obvious that the rotation rate of the central slab is proportional to the current of dislocations crossing the crystal .
thus , a consistent equation of motion for the orientation should be based on the evolution of the dislocation density .
however , the development of such an equation is a difficult task , because the motion of dislocations is determined by their complicated elastic interactions , as well as by external strain and interactions with other defects . despite intense activity on the phase - field modelling of defects , elasticity , and plasticity ( see @xcite for a recent overview ) , such an equation seems at present out of reach .
let us now come back to the outcome of the simulations for the tricrystal configuration .
the functional derivative of the gradient term in eq .
( [ modeli ] ) of model i generates a non - local diffusion equation for the angle field , which has to be regularised as described in ref .
@xcite . for a constant mobility
, the nonlocal interaction between the grain boundaries leads to a rotation rate that is almost independent of the distance between the grain boundaries . in model ii , the rotation rate of the central crystal decreases exponentially with the distance between the grain boundaries @xcite . in both cases ,
the central slab eventually disappears .
while , quantitatively , neither of these evolutions is likely to be accurate , qualitatively the result is the same as the one achieved by dislocation motion .
to see that there can be qualitative differences between the two dynamics , consider now a circular grain of orientation @xmath98 inserted in an infinite monocrystal of orientation @xmath97 .
suppose that the misorientation ( which is equal to @xmath98 ) is small , such that the grain boundary is made of individual dislocations separated by a typical distance @xmath99 which is much larger than the lattice spacing .
furthermore , suppose that the grain radius @xmath100 is large , @xmath101 , such that on the scale of the grain the boundary can still be described as a continuous line . for simplicity ,
disregard any anisotropy in the grain boundary energy or mobility .
then , the grain will shrink by standard motion by curvature , and the dislocations will simply move towards the centre of the grain .
note that the motion of the dislocations might not be strictly radial due to their coupling to the crystal structure ; however , this does not change the present discussion , as long as no annihilation of dislocations takes place . indeed , in this case , the total number of dislocations is conserved , and the dislocation density is simply proportional to @xmath102 , which _ increases _ with time as the grain shrinks .
this means that the misorientation also increases with time , and if the outer crystal is fixed , the circular inner grain has to perform a rigid body rotation away from the orientation of the outer crystal .
this seems surprising at first , since for low - angle grain boundaries the grain boundary energy is an increasing function of the misorientation .
however , this process is perfectly possible if it leads to a decrease of the total energy of the grain boundary , which is given by @xmath103 , with @xmath104 the misorientation - dependent grain boundary energy .
its time derivative is @xmath105,\ ] ] where @xmath106 is the derivative of @xmath107 with respect to the misorientation .
the evolution can thus take place if the first term , which is always negative since @xmath108 , is large enough to outweigh the second one , which is positive . in that case
, the geometrical constraints thus predict an increase of @xmath98 with time .
the orientation - field models make exactly the opposite prediction : since the angle field evolves _ locally _ such as to lower the energy , the misorientation of the inner grain should _ decrease _ with time .
recently , this situation was investigated by numerical simulations @xcite using the phase - field crystal model @xcite , which gives a faithful microscopic picture of dislocations .
an increase of the misorientation with time was observed , consistent with the geometrical constraints .
a previous study that had compared phase - field and molecular dynamics simulations @xcite and had reached different conclusions was limited to high misorientations , such that the above hypotheses were not satisfied . in conclusion , the simple relaxation equation for the angle field , eq .
( [ thetaeq ] ) is not consistent with the coherent crystalline structure of matter , and can sometimes lead to predictions that are even qualitatively wrong . for practical purposes , the quantitative importance of the committed errors might be small when the evolution of a large - scale grain structure is considered , but this has to be confirmed for each case at hand .
it is worth mentioning that orientation - field models have been used to investigate the interplay between the positional and orientational degrees of freedom during the solidification of spherulites @xcite or in the presence of foreign - phase particles @xcite .
these studies were performed with a vanishing orientational mobility @xmath95 in the solid , and are thus not affected by the problem pointed out here . indeed , in the interfacial region where the structure of the solid in not yet fully established , the concept of a rotational mobility is valid .
many phase - field simulations include fluctuations , which are often introduced in a purely qualitative way to trigger instabilities or to create some disorder in the geometry of the microstructures .
the role of fluctuations has been investigated more quantitatively in connection with the formation of sidebranches in free dendritic growth @xcite .
the standard approach is to include fluctuations as langevin terms in the field equations , with coefficients deduced from the fluctuation - dissipation theorem . before proceeding further ,
this procedure will be summarised .
after inclusion of noise , eqs .
( [ pf ] ) and ( [ diffusion ] ) for the solidification of a pure substance become ( see ref .
@xcite for details ) @xmath109 @xmath110 where @xmath111 is assumed ( symmetric model ) , and lengths and times have been scaled by the interface thickness @xmath5 and the phase - field relaxation time @xmath4 , respectively . here , @xmath112 and @xmath113 are random fluctuations of the phase field and random microscopic heat currents , respectively .
they are assumed to be @xmath114-correlated in space and time , @xmath115 @xmath116 with dimensionless amplitudes @xmath117 and @xmath118 given by @xmath119 @xmath120 where @xmath99 is the spatial dimension , and the quantity @xmath121 is determined by materials properties only , @xmath122 where @xmath123 , @xmath9 , @xmath11 , @xmath10 , and @xmath124 are boltzmann s constant , the melting temperature , the specific heat , the latent heat , and the capillary length , respectively .
the latter is given by @xmath125 , where @xmath107 is the surface free energy .
with the help of this expression for the capillary length , @xmath121 can be rewritten as @xmath126 , which makes its physical meaning more transparent : it is the ratio of the thermal energy and a capillary energy scale , and can thus be seen as a non - dimensional temperature . in a finite - difference discretization of timestep @xmath127 and grid spacing @xmath128 , the noise terms are implemented by drawing , at each grid point @xmath129 and for each time step @xmath130 , independent gaussian random variables of correlation @xmath131 where @xmath132 and @xmath133 are now kronecker symbols , and similarly for @xmath134 .
this procedure was shown to yield the correct interface fluctuations at equilibrium in numerical simulations @xcite .
an obvious question then arises , namely , can this method also be used to explore nucleation ?
phase - field methods have been used recently to investigate homogeneous and heterogeneous nucleation , both in single - phase and multi - phase systems ( see , for example , @xcite ) . in particular , it was found that for high undercoolings , diffuse - interface models yield better agreement with experiments than classical nucleation theory , since the size of the nuclei is not much larger than the thickness of the diffuse interfaces ; therefore , the free energy barriers calculated in phase - field models can differ significantly from classical nucleation theory .
is it sufficient , then , to add thermal noise as described above to obtain quantitative simulations of nucleation processes ?
the answer to this question is negative .
the reason is that , for strong noise , field equations like the phase - field model are renormalized by the fluctuations .
this is a well - known fact in statistical field theory , but its implications do not yet seem to have been fully appreciated in the phase - field community .
therefore , it is useful to briefly sketch a few calculations that can be found in textbooks ( see , for example , @xcite ) .
they are , therefore , neither new nor complete ; however , they will prepare the ground for understanding the conclusions on the phase - field method at the end of this section . instead of the full phase - field model ,
consider a single equation for a scalar field @xmath1 that reads @xmath135 where @xmath136 is a non - conserved noise that is @xmath114-correlated , @xmath137 with @xmath138 a suitably non - dimensionalized temperature ( such as @xmath121 , see the discussion after eq . ( [ fexpt ] ) ) , and the deterministic part of the equation derives from the functional @xmath139d\vec r , \label{hamil}\ ] ] where @xmath140 is a local potential of the field @xmath1 ( lengths , times , and energies are dimensionless ) .
it is important to stress that @xmath141 is _ not _ a free energy functional , but the hamiltonian of the field theory .
( [ langevin ] ) generates an evolution in which each microscopic field configuration appears with probability @xmath142 in the limit of infinite evolution time .
here , @xmath143 is the partition function , @xmath144 and @xmath145 denotes a functional integration over the field @xmath1 .
the free energy is then obtained by the standard formula @xmath146 .
the free energy can be calculated exactly for the case of a quadratic potential , @xmath147 , where @xmath148 is a constant . to carry out the calculations ,
it is useful to consider a discrete version of the model . for simplicity , consider as the domain of integration @xmath22 a @xmath99-dimensional torus of size @xmath149 with periodic boundary conditions .
when this system is discretized with the usual finite difference formulae using @xmath150 grid points in each direction and hence a grid spacing @xmath151 , the integral in eq .
( [ hamil ] ) becomes a sum over a finite number of variables . in one dimension , @xmath152 , \label{hamilex}\ ] ] with the convention that @xmath153 .
for the discretized system , the functional integration in eq .
( [ funcint ] ) is replaced by a simple integration over the field variables at each grid point , @xmath154 since the hamiltonian of eq .
( [ hamilex ] ) is a quadratic form in the @xmath155 s , this is a @xmath150-dimensional gaussian integral which can be evaluated using standard formulae .
the most convenient way is to use a discrete fourier transform to find the eigenvalues of the quadratic form .
the final result for the free energy is ( up to a constant that can be dropped ) @xmath156 for dimensions @xmath157 , the same calculation can be repeated without difficulties , and the result is @xmath158 where the sum is now over an independent index @xmath159 for each dimension ( @xmath160 ) , and is normally taken over the first brillouin zone , @xmath161 . for an arbitrary potential @xmath140 , an exact calculation is generally impossible .
statistical field theory has developed sophisticated approximation methods , in particular perturbation expansions .
formally , every potential can be written as a perturbation of a quadratic potential . the perturbation expansion ( where the expansion parameter is the temperature , which sets the fluctuation strength ) is cumbersome and usually visualised in terms of diagrams @xcite .
fortunately , the first order result can be understood in a relatively simple manner if we are interested in homogeneous systems .
more precisely , consider the spatial average of the field , @xmath162 which is a fluctuating quantity .
the probability distribution of @xmath163 can be written as @xmath164 where @xmath165 is the free energy density . to first order in the perturbation expansion , @xmath166 where the correction to the original ( `` bare '' ) potential @xmath167 is identical to the exact result for the quadratic potential , with the constant @xmath168 replaced by the second derivative of the bare potential , taken at @xmath163 .
this results from a quadratic approximation ( second - order taylor expansion ) of the bare potential around @xmath163 .
the result @xmath165 is a renormalized potential for @xmath163 . ) and from numerical simulations , for @xmath169 , @xmath170 , @xmath171 .
only the part close to one of the potential wells is shown .
the zero of @xmath172 was chosen at the minimum of the renormalized potential .
the bin size for the histograms was @xmath173 . ]
these calculations can be readily verified numerically . as an example
, the standard double - well potential was used , @xmath174 ( usually called @xmath175-potential in the field - theory literature ) , and simulated in a two - dimensional system of size @xmath176 with a grid spacing of @xmath170 and @xmath169 , using the standard discretization method described above with a timestep @xmath177 , and an initial condition @xmath178 . in time intervals of @xmath179 , @xmath163 was calculated , and in total @xmath180 points were sampled . then , the free energy can be obtained by making a histogram of the values of @xmath163 , and taking the logarithm of the counts ( the normalisation contributes only a constant to @xmath172 and can be disregarded ) .
the comparison between the simulation and the prediction of eq .
( [ theory ] ) in fig .
[ fig3 ] shows excellent agreement .
it can be seen that the minimum of the free energy density is shifted with respect to its `` bare '' value @xmath181 .
this can be understood intuitively by the following reasoning .
the system starts in the well of the `` bare '' potential , at @xmath181 .
the random fluctuations push the system in both directions with equal probability , but since the potential is asymmetric , the restoring force is larger for fluctuations towards @xmath182 than towards @xmath183 ; therefore , smaller values are more likely to occur . in the example chosen here , the shift is small ( the minimum is close to @xmath184 ) , but for increasing temperature , the correction becomes larger and larger ( for an example of such simulations , see @xcite ) , and eventually a phase transition occurs ( the double well disappears ) ; in this regime , of course the first - order perturbation result is inaccurate .
the correction also depends on the discretization .
this is physically sound : a finer discretization introduces more degrees of freedom per unit volume in the discretized system , and hence allows for more fluctuation modes that contribute to the free energy . with a slight change of perspective ,
this can also be seen as the natural result of a coarse - graining procedure .
indeed , if the free energy is calculated from a given microscopic model by coarse - graining ( averaging ) over cells with a certain size @xmath128 larger than the size of the microscopic elements , both the free energy density and the amplitude of the fluctuations that remain after the averaging ( which thus have a wavelength larger than @xmath128 ) depend on the choice of @xmath128 , as was recently demonstrated explicitly for a simple lattice gas model @xcite . however , a problem arises in the continuum picture : it is easy to verify that , when the grid spacing @xmath128 tends to zero , the sum in eq .
( [ theory ] ) diverges for @xmath185 .
this is a classical example of an ultraviolet divergence .
thus , eq . ( [ langevin ] ) has no continuum limit , and if it is written down in continuum language , it is implicitly understood that an ultraviolet cutoff must be specified .
a reasonable physical value for a cutoff in condensed - matter systems is the size of an atom .
let us now discuss the implications of these facts for phase - field modelling .
even though the above calculation have not been carried out for the full model ( @xmath1 and @xmath7 ) , it is clear that renormalization occurs . if a phase - field model is seen as a simulation tool for a problem that is defined in terms of macroscopic parameters , the relevant quantities that need to be adjusted in the model are the renormalized ones .
for instance , thermophysical properties are usually interpolated assuming that the phase field takes fixed values in the bulk phases ( @xmath186 ) . if , on average , this is no longer the case , such as in the example of fig .
[ fig3 ] , these interpolations become incorrect .
an obvious idea to cure this problem is to choose the `` bare '' potential such that the renormalized potential has the desired properties . for the @xmath175-potential , which is renormalizable , one may choose @xmath187 and determine the constants @xmath188 and @xmath189 by the two conditions @xmath190 and @xmath191 using eq .
( [ theory ] ) .
for the example shown above , the values @xmath192 and @xmath193 indeed restore the correct bulk properties .
however , in a quantitative phase - field model , the macroscopic properties not only of the bulk phases , but also of the interfaces need to be controlled .
it is far from obvious that the above procedure , designed for homogeneous systems , will work .
this is even more so for the critical nucleus needed to evaluate the nucleation barrier .
it is instructive to examine some orders of magnitude . in nickel ,
the value of @xmath121 is @xmath194 @xcite , of order unity ; it can be expected that this value is of similar order of magnitude for other substances with microscopically rough interfaces .
an inspection of eqs .
( [ fnoisedef][fudef ] ) reveals that _ if _ phase - field simulations are carried out with the `` natural '' interface thickness , which is of the order of the capillary length @xmath124 , the fluctuations are of order unity ( recall that @xmath117 and @xmath118 are equivalent to @xmath138 in the numerical example ) , and renormalization can not be neglected .
this is a natural consequence of the fact that real solid - liquid interfaces do indeed exhibit very strong fluctuations , as evidenced from molecular dynamics simulations @xcite ; therefore , a mean - field approximation ( such as the phase - field model without noise ) is not accurate .
in contrast , if ( as in refs .
@xcite ) a much larger interface thickness is used , the fluctuation strength is greatly reduced , and the difference between `` bare '' and renormalized free energy is small .
note , however , that even in this limit a sufficient refinement of the grid would create noticeable fluctuation corrections .
we are thus faced with the conclusion ( opposite to the usual point of view in phase - field modelling ) that the use of the simple prescription of ref .
@xcite is more precise for larger interface thickness and coarser grids .
it is noted in passing that the concept of the sharp - interface limit , central for the asymptotic analysis in the deterministic case , has to be reexamined because a new length scale ( the microscopic cutoff for the fluctuations ) has been introduced . in conclusion
, it is clear that the use of the phase - field method with fluctuations is subject to caution , at least on small length scales . to gain a better understanding , the fluctuation effects on the couplings of the phase - field variables need to be investigated
furthermore , a good control of the discretization effects needs to be achieved ; the introduction of a simple cutoff will most likely be insufficient , since the renormalized free energy of eq .
( [ theory ] ) also depends on the grid structure .
while a large body of results on these topics can certainly be found in the field - theory literature , the development of quantitative models for specific materials remains a challenging task .
in this paper , some open questions concerning various aspects of phase - field modelling of solidification have been discussed , and potential future directions of research have been outlined .
the selection of topics is necessarily incomplete , both concerning the problems and the potential solutions .
for instance , the rapid development of the phase - field crystal approach @xcite and related methods currently opens up interesting new perspectives for the modelling of polycrystals , which are not discussed further here .
the common point of the topics treated here is that they illustrate the dual nature of the phase - field method . on the one hand
, it is a genuine representation of condensed - matter systems and their evolution in terms of order parameters on a mesoscopic scale . on the other hand , with the help of mathematical analysis ,
it can be turned into an efficient simulation tool for the solution of free boundary problems . as in the past , the development of more efficient and robust models for materials modelling will most likely benefit from the pursuit and confrontation of _ both _ of these two complementary viewpoints .
therefore , the further development of the phase - field method remains an exciting research topic at the frontiers of physics , mathematics , and materials science .
i thank jean - marc debierre , tristan ducousso , alphonse finel , lszl grnsy , herv henry , alain karma , yann le bouar , jesper mellenthin , tams pusztai , and james warren for stimulating discussions on these and many other topics .
t. ducousso , _ tude de la solidification dirige par la mthode du champ de phase : comparaison thorie exprience pour un alliage binaire dilu _ , ph.d .
thesis , universit paul czanne , marseille , france , 2009 . | three different topics in phase - field modelling of solidification are discussed , with particular emphasis on the limitations of the currently available modelling approaches .
first , thin - interface limits of two - sided phase - field models are examined , and it is shown that the antitrapping current is in general not sufficient to remove all thin - interface effects .
second , orientation - field models for polycrystalline solidification are analysed , and it is shown that the standard relaxational equation of motion for the orientation field is incorrect in coherent polycrystalline matter .
third , it is pointed out that the standard procedure to incorporate fluctuations into the phase - field approach can not be used in a straightforward way for a quantitative description of nucleation . |
the fundamental atmospheric ( effective temperature , surface gravity , and metallicity ) and physical ( mass and age ) parameters of stars provide the major observational foundation for chemo - dynamical studies of the milky way and other galaxies in the local group . with the dawn of large spectroscopic surveys to study individual stars , such as segue @xcite , rave @xcite , gaia - eso @xcite , and hermes @xcite , these parameters are used to infer the characteristics of different populations of stars that comprise the milky way .
stellar parameters determined by spectroscopic methods are of a key importance . the only way to accurately measure metallicity is through spectroscopy , which thus underlies photometric calibrations ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , while high - resolution spectroscopy is also used to correct the low - resolution results ( e.g. , * ? ? ?
the atmospheric parameters can all be estimated from a spectrum in a consistent and efficient way .
this also avoids the problem of reddening inherent in photometry since spectroscopic parameters are not sensitive to reddening .
the spectroscopic parameters can then be used alone or in combination with photometric information to fit individual stars to theoretical isochrones or evolutionary tracks to determine the stellar mass , age , and distance of a star .
a common method for deriving the spectroscopic atmospheric parameters is to use the information from fe and fe absorption lines under the assumption of hydrostatic equilibrium ( he ) and local thermodynamic equilibrium ( lte ) .
many previous studies have used some variation of this technique ( e.g. , ionisation or excitation equilibrium ) to determine the stellar atmospheric parameters and abundances , and henceforth distances and kinematics , of fgk stars in the milky way .
for example , some have used this procedure to estimate the effective temperature , surface gravity , and metallicity of a star ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) , while others use photometric estimates of effective temperature in combination with the ionisation equilibrium of the abundance of iron in lte to estimate surface gravity and metallicity ( e.g. , * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ; * ? ? ? * ) .
however , both observational ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) and theoretical evidence ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) suggest that systematic biases are present within such analyses due to the breakdown of the assumption of lte .
more recently , @xcite and @xcite quantified the effects of non - local thermodynamic equilibrium ( nlte ) on the determination of surface gravity and metallicity , revealing very substantial systematic biases in the estimates at low metallicity and/or surface gravity .
it is therefore extremely important to develop sophisticated methods , which reconcile these effects in order to derive accurate spectroscopic parameters .
this is the first in a series of papers , in which we develop new , robust methods to determine the fundamental parameters of fgk stars and then apply these techniques to large stellar samples to study the chemical and dynamical properties of the different stellar populations of the milky way . in this work ,
we utilise the sample of stars selected from the rave survey originally published in ( * ? ? ?
* hereafter r11 ) to formulate the methodology to derive very accurate atmospheric parameters .
we consider several temperature scales and show that the balmer line method is the most reliable among the different methods presently available .
further , we have developed the necessary tools to apply on - the - fly nlte corrections to fe lines , utilising the grid described in @xcite .
we verify our method using a sample of standard stars with interferometric estimates of effective temperature and/or _ hipparcos _ parallaxes .
we then perform a comprehensive comparison to standard 1d , lte techniques for the spectral analysis of stars , finding significant systematic biases .
nlte effects in iron are most prominent in low - metallicity stars @xcite .
we therefore chose the metal - poor sample from r11 for our study .
these stars were originally selected for high - resolution observations based on data obtained by the rave survey in order to study the metal - poor thick disk of the milky way .
spectral data for these stars were obtained using high - resolution echelle spectrographs at several facilities around the world .
full details of the observations and data reduction of the spectra can be found in r11 .
briefly , all spectrographs delivered a resolving power greater than 30,000 and covered the full optical wavelength range .
further , nearly all spectra had signal - to - noise ratios greater than @xmath4 per pixel .
the equivalent widths ( ews ) of both fe and fe lines , taken from the line lists of @xcite and @xcite , were measured using the ares code @xcite .
however , during measurement quality checks , we found that the continuum was poorly estimated for some lines .
we therefore determined ews for these affected lines using hand measurements .
we computed the stellar parameters for each star using two different methods . in the first method , which is commonly used in the literature
, we derived an effective temperature , @xmath5 , surface gravity , @xmath6 , metallicity , @xmath7 , and microturbulence , @xmath8 , from the ionisation and excitation equilibrium of fe in lte .
this is hereafter denoted as the lte - fe method .
we used an iterative procedure that utilised the ` moog ` analysis program @xcite and 1d , plane - parallel ` atlas - odf ` model atmospheres from kurucz computed under the assumption of lte and he . in our procedure , the stellar effective temperature was set by minimising the magnitude of the slope of the relationship between the abundance of iron from fe lines and the excitation potential of each line .
similarly , the microturbulent velocity was found by minimising the slope between the abundance of iron from fe lines and the reduced ew of each line .
the surface gravity was then estimated by minimising the difference between the abundance of iron measured from fe and fe lines .
iterations continued until all of the criteria above were satisfied .
finally , @xmath7 was chosen to equal the abundance of iron from the analysis .
our results for this method are described in section [ sec - lte ] .
the second method , denoted as the nlte - opt method , consists of two parts .
first , we determined the optimal effective temperature estimate , @xmath9 , for each star ( see section [ sec - temp ] for more details ) .
then , we utilised ` moog ` to compute a new surface gravity , @xmath10 , metallicity , @xmath11 , and microturbulence , @xmath12 .
this was done using the same iterative techniques as the lte - fe method , that is the ionisation balance of the abundance of iron from fe and fe lines .
there are , however , three important differences .
first , the stellar effective temperature was held fixed to the optimal value , @xmath9 .
second , we restricted the analysis to fe lines with excitation potentials above 2 ev , since these lines are less sensitive to 3d effects as compared to the low - excitation lines ( see the discussion in * ? ? ?
third , the abundance of iron from each fe line was adjusted according to the nlte correction for that line at the stellar parameters of the current iteration in the procedure .
the nlte corrections were determined using the nlte grid computed in @xcite and applied on - the - fly via a wrapper program to ` moog ` .
note that the nlte calculations presented in @xcite were analogously calibrated using the ionisation equilibria of a handful of well - known stars .
our extended sample , including more stars with direct measurements of surface gravity and effective temperature , provide support for the realism of this calibration .
the grid extends down to @xmath13 .
we imposed a routine which linearly extrapolated the nlte corrections to below this value .
the results of extrapolations were checked against nlte grids presented in @xcite and no significant differences were found .
further , @xcite found very small nlte corrections for fe lines .
we therefore do not apply any correction to the fe lines .
iterations continued until the difference between the average abundance of iron from the fe lines and the nlte - adjusted fe lines were in agreement ( within @xmath14 dex ) and the slope of the relationship between the reduced ew of the fe lines and their nlte - adjusted iron abundance was minimised .
sections [ sec - temp ] and [ sec - nlte ] describe our final stellar parameter estimates for this method .
the initial lte - fe stellar parameters for our sample stars are listed in table [ tab - par ] .
residuals in the minimizations of this technique gave typical internal errors of 0.1 dex in both @xmath6 and @xmath7 and @xmath15 k in @xmath5 .
as we show in the following sections , these small internal errors can be quite misleading as they are not representative of the actual accuracy of stellar parameter estimates .
often , especially in metal - poor stars , estimates of @xmath16 , @xmath17 , and @xmath18 , that result from this method are far too low when compared to other , more accurate data ( cf . , r11 ) .
@rrrrrrrrrrrrrrrrr@ & & & & & & & & + + star & @xmath16 & @xmath19 & @xmath17 & @xmath18 & @xmath20 & @xmath21 & @xmath22 & @xmath23 & @xmath24 & @xmath25 & @xmath26 & @xmath16 & @xmath19 & @xmath17 & @xmath18 & @xmath20 + & ( k ) & ( k ) & ( @xmath27 ) & ( @xmath27 ) & & & ( k ) & ( k ) & ( @xmath28 k ) & ( k ) & ( k ) & ( k ) & ( k ) & ( @xmath27 ) & ( @xmath27 ) & + c0023306 - 163143 & 5128 & 58 & 2.40 & -2.63 & 1.3 & & & & 5528 & 5400 & 100 & 5443 & 101 & 3.20 & -2.29 & 0.9 + c0315358 - 094743 & 4628 & 40 & 1.51 & -1.40 & 1.5 & & & & 4722 & 4800 & 100 & 4774 & 89 & 2.06 & -1.31 & 1.6 + c0408404 - 462531 & 4466 & 40 & 0.25 & -2.25 & 2.2 & & & & 4600 & & & 4600 & 90 & 1.03 & -2.10 & 2.1 + c0549576 - 334007 & 5151 & 50 & 2.53 & -1.94 & 1.3 & & & & 5379 & 5400 & 100 & 5393 & 82 & 3.16 & -1.70 & 1.1 + c1141088 - 453528 & 4439 & 40 & 0.39 & -2.42 & 2.1 & & & & 4592 & 4500 & 200 & 4562 & 123 & 1.10 & -2.28 & 1.9 + this table is published in its entirety in the electronic edition of the mnras . a portion is shown here for guidance regarding its form and content .
it was found in @xcite and @xcite that taking into account nlte in the solution of excitation equilibrium does not lead to a significant improvement of the stellar effective temperature .
this was also supported by our test calculations for a sub - sample of stars .
fe lines formed in lte or nlte are still affected by convective surface inhomogeneities and overall different mean temperature / density stratifications , which are most prominent in strong low - excitation fe lines @xcite . using 1d hydrostatic models with either lte or nlte radiative transfer thus leads to effective temperature estimates that are too low when the excitation balance of fe lines is used ( see below ) .
it is therefore important that the stellar effective temperature be estimated by other means .
we used three different methods to compute the effective temperature .
the first estimate , @xmath25 , was derived from the wings of the balmer lines , which is among the most reliable methods available for the effective temperature determination of fgk stars ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the only restriction of this method is that for stars cooler than @xmath29 k , the wings of h lines become too weak to allow reliable determination of @xmath16 .
profile fits of h@xmath30 and h@xmath31 lines were performed by careful visual inspection of different portions of the observed spectrum in the near and far wings of the balmer lines which were free of contaminant stellar lines .
figures [ fig - bal3 ] and [ fig - bal5 ] show two example fits to h@xmath32 .
note that the balmer lines were self - contained within a single order in each spectrum .
therefore , we did not use neighbouring orders for the continuum normalisation .
theoretical profiles were computed using the siu code with ` mafags - odf ` model atmospheres @xcite .
same as ` atlas - odf ` ( section [ sec - par ] ) , the mafags models were computed with kurucz opacity distribution functions , thus the differences between the model atmosphere stratifications are expected to be minimal in our range of stellar parameters . for self - broadening of h lines , we used the @xcite theory .
as shown by @xcite this method successfully reproduces the balmer line spectrum of the sun within @xmath33 k , and provides accurate stellar parameters that agree very well with _ hipparcos _
astrometry @xcite .
the errors are obtained directly from profile fitting , and they are largely internal , @xmath34 to 100 k. in the spectrum of the metal - poor ( @xmath35 ) dwarf , j142911.4 - 053131 .
the solid , red curve shows the best fit to the data ( at 5700 k ) , while the blue , dashed curves represent @xmath36 k around the fit.,width=317 ] in the spectrum of the metal - poor ( @xmath37 ) giant , j230222.8 - 683323 .
the solid , red curve shows the best fit to the data ( at 5200 k ) , while the blue , dashed curves represent @xmath36 k around the fit.,width=317 ] a key advantage of the balmer lines is that they are insensitive to interstellar reddening , which affects photometric techniques ( see below ) .
however , the balmer line effective temperature scale could be affected by systematic biases , caused by the physical limitation of the models .
the influence of deviations from lte in the h line formation in application to cool metal - poor stars was studied by @xcite .
comparing our results to the nlte estimates by @xcite for the stars in common , we obtain : @xmath38(our - m08 ) @xmath39 k ( hd 122563 metal - poor giant ) , @xmath38(our - m08 ) @xmath40 k ( hd 84937 , metal - poor turn - off ) . the difference is clearly within the @xmath16 uncertainties . on the other side
, it should be kept in mind that the atomic data for nlte calculations for hydrogen are of insufficient quality and , at present , do not allow accurate quantitative assessment of nlte effects in h , as elaborately discussed by @xcite .
likewise , the influence of granulation is difficult to assess .
@xcite presented 3d effective temperature corrections for balmer lines for a few points on the hrd , for which 3d radiative - hydrodynamics simulations of stellar convection are available . for the sun
consistent with the ` mafags - odf ` model atmospheres adopted here .
] , they find @xmath41 k , and for a typical metal - poor subdwarf with [ fe / h ] @xmath42 , @xmath43 of the order @xmath44 to @xmath45 k ( average over h@xmath30 , h@xmath31 , and h@xmath46 ) .
however , in the absence of consistent 3d nlte calculations , it is not possible to tell whether 3d and nlte effects will amplify or cancel for fgk stars .
thus , we do not apply any theoretical corrections to our balmer effective temperatures . currently , the only way to understand whether our balmer @xmath16 scale is affected by systematics is by comparing with independent methods , in particular interferometry .
we , therefore , computed the balmer @xmath16 for several nearby stars with direct and indirect interferometric angular diameter measurements .
the results are listed in table [ tab - inter ] , while we plot the difference between our balmer estimate and that from interferometry in figure [ fig - int ] .
both @xmath16 scales show an agreement of @xmath47 k for stars with @xmath48 , while the balmer estimate is @xmath49 k warmer than @xmath50 at the lowest metallicities .
these differences are well within the combined errors in the interferometric and balmer measurements .
this suggests that deviations from 1d he _ and _ lte are either minimal , or affect both interferometric and balmer @xmath16 in exactly same way .
also note , for the stars in common with @xcite , our estimates are fully consistent .
.effective temperatures determined from direct interferometric measurements of angular diameters . [ cols= " > , > , > , > , > , > , > , < " , ]
in figure [ fig - lte ] , we compare our final nlte - opt stellar parameters to those derived using the lte - fe method .
the differences in the estimates of effective temperature , surface gravity , metallicity , and microturbulence all display clear trends with decreasing metallicity .
the microturbulent velocity is underestimated by @xmath51 until @xmath52 , where @xmath8 becomes larger than @xmath12 .
the differences between @xmath16 range from @xmath53 to @xmath0 k for metal - poor giants , and @xmath54 to @xmath55 k for dwarfs .
the differences for @xmath56 and [ fe / h ] reach a factor of @xmath57 in surface gravity ( @xmath58 dex ) and a factor of @xmath59 in metallicity ( @xmath60 [ fe / h ] @xmath61 dex ) at [ fe / h ] @xmath62 . .
the difference in effective temperature , surface gravity , and metallicity show a large systematic increase with decreasing metallicity .
the dual trends seen in @xmath63 , @xmath64 , and @xmath65 are a result of the r11 effective temperature calibration , in which the authors found that stars with effective temperatures less than 4500 k only required a small correction to @xmath5 .
therefore , these stars stand out in the plots.,width=317 ] figure [ fig - hrd ] illustrates how the different lte - fe and nlte - opt results can change the position of each star in the @xmath17 vs. @xmath66 plane .
in addition , we have included several evolutionary tracks , computed using the garstec code @xcite , for comparison . generally , the nlte - opt estimates of surface gravity and effective temperature trace the morphology of the theoretical tracks much more accurately . several features are most notable .
the nlte - opt parameters lead to far less stars that lie on or above the tip of the red giant branch , and more stars occupy the middle or lower portion of the rgb . also , stars at the turn - off and subgiant branch are now more consistent with stellar evolution calculations . figures [ fig - lte ] and [ fig - hrd ] further prompted us to determine the relative importance of the effective temperature scale versus the nlte corrections for gravities and metallicities in the nlte - opt method .
we singled out the effect of the nlte corrections by deriving additional , lte - opt surface gravity and metallicity estimates using lte iron abundances combined with our @xmath9 estimate .
note , as with the nlte - opt method , fe lines which have an excitation potential below 2 ev were excluded .
the comparison between these lte - opt estimates and the final nlte - opt estimates are shown in figure [ fig - copt ] . as evident from this figure ,
solving for ionisation equilibrium in nlte also leads to _ systematic _ changes in the @xmath56 and [ fe / h ] , such that lte gravities are under - estimated by @xmath67 to @xmath68 dex , whereas the error in metallicity is about @xmath69 to @xmath70 dex .
these effects are consistent with that seen in @xcite .
we thus conclude that reliable effective temperatures are necessary to avoid substantial biases in a spectroscopic determination of @xmath56 and [ fe / h ] , such as displayed in figure [ fig - lte ] .
we have shown here that , at present , excitation balance of fe lines with 1d hydrostatic model atmospheres in lte does not provide the correct effective temperature scale , supporting the results by bergemann et al .
( 2012 ) . on the contrary ,
balmer lines provide such a scale .
furthermore , nlte effects on ionisation balance are necessary to eliminate the discrepancy between fe and fe lines , an effect that is present , regardless of the adopted @xmath16 . only in this way
is it possible to determine accurate surface gravity and metallicity from fe lines .
in this work , we explore several available methods to determine effective temperature , surface gravity , and metallicity for late - type stars .
the methods include excitation and ionization balance of fe lines in lte and nlte , semi - empirically calibrated photometry ( r11 ) , and the infra - red flux method ( irfm ) . applying these methods to the large set of high - resolution spectra of metal - poor fgk stars selected from the rave survey
, we then devise a new efficient strategy which provides robust estimates of their atmospheric parameters .
the principal components of our method are ( i ) balmer lines to determine effective temperatures , ( ii ) nlte ionization balance of fe to determine @xmath17 and @xmath18 , and ( iii ) restriction of the fe lines to that with the lower level excitation potential greater than 2 ev to minimize the influence of 3d effects @xcite .
a comparison of the new nlte - opt stellar parameters to that obtained from the widely - used method of 1d lte excitation - ionization of fe , lte - fe , reveals significant _ systematic biases _ in the latter .
the difference between the nlte - opt and lte - fe parameters systematically increase with decreasing metallicity , and can be quite large for the metal - poor stars : from 200 to 400 k in @xmath16 , 0.5 to 1.5 dex in @xmath17 , and 0.1 to 0.5 dex in @xmath18 .
these systematic trends are largely influenced by the difference in the estimate of the stellar effective temperature , and thus , a reliable effective temperature scale , such as the balmer scale , is of critical importance in any spectral parameter analysis .
however , a disparity between the abundance of iron from fe and fe lines still remains .
it is therefore necessary to include the nlte effects in fe lines to eliminate this discrepancy .
the implications of the very large differences between the nlte - opt and lte - fe estimates of atmospheric parameters extend beyond that of just the characterisation of stars by their surface parameters and abundance analyses .
spectroscopically derived parameters are often used to derive other fundamental stellar parameters such as mass , age and distance through comparison to stellar evolution models .
the placement of a star along a given model will be largely influenced by the method used to determine the stellar parameters .
for example , distance scales will change , which could affect the abundance gradients measured in the milky way ( e.g. , r11 ) , as well as the controversial identification of different components in the mw halo @xcite .
we explore this in greater detail in the next paper of this series @xcite .
we acknowledge valuable discussions with martin asplund , and are indebted to ulrike heiter for kindly providing interferometric temperatures for several _ gaia _ calibration stars .
we also acknowledge the staff members of siding spring observatory , la silla observatory , apache point observatory , and las campanas observatory for their assistance in making the observations for this project possible .
greg ruchti acknowledges support through grants from esf eurogenesis and max planck society for the firststars collaboration .
aldo serenelli is partially supported by the european union international reintegration grant pirg - ga-2009 - 247732 , the micinn grant aya2011 - 24704 , by the esf eurocores program eurogenesis ( micinn grant eui2009 - 04170 ) , by sgr grants of the generalitat de catalunya and by the eu - feder funds .
ali , a. w. , & griem , h. r. 1965 , physical review , 140 , 1044 allende prieto c. , beers t. c. , wilhelm r. , newberg h. j. , rockosi c. m. , yanny b. , lee y. s. , 2006 , apj , 636 , 804 alonso a. , arribas s. , martinez - roger c. , 1995 , a&a , 297 , 197 an d. , et al . , 2009 , apj , 707 , l64 rnadttir a. s. , feltzing s. , lundstrm i. , 2010 , a&a , 521 , a40 asplund m. , 2005 , ara&a , 43 , 481 bai g. s. , zhao g. , chen y. q. , shi j. r. , klochkova v. g. , panchuk v. e. , qiu h. m. , zhang h. w. , 2004 , a&a , 425 , 671 barden s. c. , et al . , 2008 ,
spie , 7014,149 barklem p. s. , 2007 , a&a , 466 , 327 barklem p. s. , piskunov n. , omara b. j. , 2000a , a&a , 355 , l5 barklem p. s. , piskunov n. , omara b. j. , 2000b , a&a , 363 , 1091 barklem p. s. , stempels h. c. , allende prieto c. , kochukhov o. p. , piskunov n. , omara b. j. , 2002 , a&a , 385 , 951 barklem p. s. , stempels h. c. , kochukhov o. , piskunov n. , omara b. j. , 2003 , in 12th cambridge workshop on cool stars , stellar systems , and the sun , the future of cool - star astrophysics , ed .
a. brown , g.m .
harper , and t.r .
ayres , ( university of colorado ) , 1103 beers t. c. , et al . , 2012 , apj , 746 , 34 bergemann m. , kudritzki r .-
, plez b. , davies b. , lind k. , gazak z. , 2012a , apj , 751 , 156 bergemann m. , lind k. , collet r. , magic z. , asplund m. , 2012b , arxiv , arxiv:1207.2455 bruntt h. , et al . , 2012 , mnras , 423 , 122 carollo d. , et al . , 2010 , apj , 712 , 692 casagrande l. , ramrez i. , melndez j. , bessell m. , asplund m. , 2010 , a&a , 512 , a54 casagrande l. , schnrich r. , asplund m. , cassisi s. , ramrez i. , melndez j. , bensby t. , feltzing s. , 2011 , a&a , 530 , a138 cayrel r. , vant veer - menneret c. , allard n. f. , stehl c. , 2011 , a&a , 531 , a83 chiavassa a. , freytag b. , masseron t. , plez b. , 2011 , a&a , 535 , a22 cowley c. r. , castelli f. , 2002 , a&a , 387 , 595 creevey o. l. , et al . , 2012 , a&a , 545 , a17 drimmel r. , cabrera - lavers a. , lpez - corredoira m. , 2003 , a&a , 409 , 205 franois p. , et al . , 2003 , a&a , 403 , 1105 fuhrmann k. , 1998 , a&a , 338 , 161 fuhrmann k. , axer m. , gehren t. , 1993 , a&a , 271 , 451 fulbright j. p. , 2000 , aj , 120 , 1841 gehren t. , shi j. r. , zhang h. w. , zhao g. , korn a. j. , 2006 , a&a , 451 , 1065 gilmore g. , et al . , 2012 , msngr , 147 , 25 gonzlez hernndez j. i. , bonifacio p. , 2009
, a&a , 497 , 497 grupp f. , 2004a , a&a , 420 , 289 grupp f. , 2004b , a&a , 426 , 309 holmberg j. , nordstrm b. , andersen j. , 2007 , a&a , 475 , 519 huber d. , et al . , 2012 , apj , arxiv:1210.0012 ivans i. i. , kraft r. p. , sneden c. , smith g. h. , rich r. m. , shetrone m. , 2001 , aj , 122 , 1438 johnson j. a. , 2002 , apjs , 139 , 219 kervella p. , thvenin f. , di folco e. , sgransan d. , 2004 , a&a , 426 , 297 lai d. k. , bolte m. , johnson j. a. , lucatello s. , heger a. , woosley s. e. , 2008 , apj , 681 , 1524 lind k. , bergemann m. , asplund m. , 2012 , arxiv , arxiv:1207.2454 ludwig h .-
, behara n. t. , steffen m. , bonifacio p. , 2009
, a&a , 502 , l1 mashonkina l. , et al . , 2008 , a&a , 478 , 529 mashonkina l. , gehren t. , shi j .-
r . , korn a. j. , grupp f. , 2011 , a&a , 528 , a87 mcwilliam a. , preston g. w. , sneden c. , searle l. , 1995 , aj , 109 , 275 mozurkewich d. , et al . , 2003 , aj , 126 , 2502 prochaska j. x. , naumov s. o. , carney b. w. , mcwilliam a. , wolfe a. m. , 2000 , aj , 120 , 2513 ruchti g. r. , et al .
, 2010 , apj , 721 , l92 ruchti g. r. , et al . , 2011 , apj , 737 , 9 schnrich r. , asplund m. , casagrande l. , 2011 , mnras , 415 , 3807 serenelli a. , bergemann m. , ruchti g. r. , 2012 , mnras , in prep . shchukina n. g. , trujillo bueno j. , asplund m. , 2005 , apj , 618 , 939 sneden c. , 1973 , apj , 184 , 839 sousa s. g. , santos n. c. , israelian g. , mayor m. , monteiro m. j. p. f. g. , 2007 , a&a , 469 , 783 steinmetz m. , et al . , 2006 , aj , 132 , 1645 thvenin f. , idiart t. p. , 1999 ,
apj , 521 , 753 torres g. , 2010 , aj , 140 , 1158 weiss a. , schlattl h. , 2008 , ap&ss , 316 , 99 yanny b. , et al . , 2009 , aj , 137 , 4377 | we present a comprehensive analysis of different techniques available for the spectroscopic analysis of fgk stars , and provide a recommended methodology which efficiently estimates accurate stellar atmospheric parameters for large samples of stars .
our analysis includes a simultaneous equivalent width analysis of fe and fe spectral lines , and for the first time , utilises on - the - fly nlte corrections of individual fe lines .
we further investigate several temperature scales , finding that estimates from balmer line measurements provide the most accurate effective temperatures at all metallicites .
we apply our analysis to a large sample of both dwarf and giant stars selected from the rave survey .
we then show that the difference between parameters determined by our method and that by standard 1d lte excitation - ionisation balance of fe reveals substantial systematic biases : up to @xmath0 k in effective temperature , @xmath1 dex in surface gravity , and @xmath2 dex in metallicity for stars with @xmath3 .
this has large implications for the study of the stellar populations in the milky way .
[ firstpage ] stars : abundances stars : late - type stars : population ii |
entanglement in disordered spin chains has received much attention recently @xcite .
the main reason is that , as opposed to on - site disordered systems @xcite , long - distance correlations are not destroyed in this case , but only modified in subtle ways .
thus , for the 1d heisenberg and xx models with uncorrelated random couplings , the von neumann entropy of blocks of size @xmath0 is known to violate the area law and grow as @xmath1 , similarly to the conformal case @xcite .
the prefactor , nonetheless , is different : it is still proportional to the the central charge of the associated conformal field theory ( cft ) , but multiplied by an extra @xmath2 factor .
moreover , the rnyi entropies do not satisfy the predictions of cft @xcite , because these models are not conformal invariant .
a very relevant tool of analysis is the strong disorder renormalization group ( sdrg ) devised by dasgupta and ma @xcite , which shows that the ground state of heisenberg or xx chains with strong disorder can be written as a product of _ random singlets _ ,
in which all spins are paired up making su(2 ) singlet bonds .
furthermore , the renormalization procedure prevents the bonds from crossing , i.e. , the bond structure will always be _
planar_. the paired spins are often neighbours , but not always .
as it was shown @xcite , the probability distribution for the singlet bond lengths , @xmath3 falls as a power - law , @xmath4 , with @xmath5 .
entanglement of a block can be obtained just by counting the number of singlets which must be cut in order to isolate the block , and multiplying by the entanglement entropy of one bond , which is @xmath2 . under the sdrg flow ,
the variance of the couplings increases and its correlation length decreases , thus approaching the so - called _ infinite randomness fixed point _ ( irfp ) @xcite .
is this fixed point unique ?
not necessarily .
if the couplings present a diverging correlation length , we might have other fixed points of the sdrg .
for example , if the couplings decay exponentially from the center , they give rise to the _ rainbow phase _ , in which singlets extend concentrically @xcite . in that case , all couplings are correlated .
but we may also devise ensembles of couplings which present long - range correlations , but are still random
. a glimpse of some further fixed points can be found by observing the statistical mechanics of the secondary structure of rna @xcite . a simple yet relevant model is constituted by a closed 1d chain with an even number of rna bases , which we call sites , which are randomly coupled in pairs with indices of different parity @xcite .
each pair constitutes an rna bond , and the only constraint is that no bonds can cross .
therefore , the ensemble of secondary structures of rna can be described in terms of planar bond structures , just like ground states of disordered spin - chains .
wiese and coworkers @xcite studied the probability distribution for the bond lengths , and found @xmath6 , with @xmath7 .
furthermore , the studies of rna folding included a very interesting second observable .
the planar bond structure can be mapped to the height function of a discretized interface @xcite .
we can define the expected roughness of windows of size @xmath0 , @xmath8 , as the deviation of the height function over blocks of size @xmath0 , which can be shown to scale in rna folding structures like @xmath9 , with @xmath10 .
interestingly , @xmath11 .
as we will show , the interface roughness is very similar to the entanglement entropy of blocks of size @xmath0 , and they are characterized by similar exponents . in the irfp phase for random singlets , notice that the entropy is characterized by a zero exponent , due to the logarithmic growth , and @xmath5 .
therefore , it is also true that @xmath11 .
we may then ask , what is the validity of this scaling relation ? does the rna folding case correspond to some choice of the ensemble of coupling constants for a spin - chain ?
can we obtain other fixed points which interpolate between the irfp and the rna folding cases ?
we may keep in mind that the couplings in some spin chain models ( e.g. , the xx model ) can be mapped into modulations of the space metric @xcite .
thus , we are obtaining , in a certain regime , the relation between the statistical structure of the space metric and the statistical properties of entanglement of the vacuum , i.e. , the ground state of the theory .
this article is organized as follows .
section [ sec : model ] introduces our model and the renormalization procedure used throughout the text .
moreover , it discusses the consequences of the planarity of the pairing structures which characterize the states . in section [ sec : results ] we establish our strategy to sample highly correlated values of the couplings , and show numerically the behavior of the entropy and other observables . in section [ sec : rna ] we focus on the relation between the rna folding problem and our disordered spin chains , and determine an inverse algorithm to compute a parent hamiltonian for any planar state , exemplifying it with the rna folding states .
how generic are planar states is the question addressed in section [ sec : generic ] , showing that they are non - generic through the study of their entanglement entropy .
the article ends in section [ sec : conclusions ] discussing our conclusions and ideas for further work .
let us consider for simplicity a spin-1/2 xx chain with @xmath12 ( even ) sites and periodic boundary conditions , whose hamiltonian is @xmath13 where the @xmath14 are the coupling constants , which we will assume to be positive and strongly inhomogeneous .
more precisely , we assume that neighboring couplings are very different .
notice that we do not impose them to be random . in order to obtain the ground state ( gs ) , we can employ the _ strong disorder renormalization group _ ( sdrg ) method of dasgupta and ma @xcite . at each renormalization step , we pick the maximal coupling , @xmath14 , decimate the two associated spins , @xmath15 and @xmath16 , and establish a _ singlet bond _ between them .
the neighboring sites are then joined by an effective coupling given by second order perturbation theory : @xmath17 among the next neighbours of the link , @xmath18 and @xmath19 .
it is convenient to use a set of auxiliary variables , that we will call _ log - couplings _ : @xmath20 .
the main reason is that , for them , the dasgupta - ma renormalization rule becomes _ additive _ :
@xmath21 once the sdrg procedure is finished we can read our gs as a product state of _ singlet _ valence bonds .
@xmath22 where @xmath23 denotes a set of @xmath24 pairing bonds among the @xmath12 spins .
many properties of these ground states have been studied in the last thirty years @xcite .
one of the most salient of those properties is the fact that the pairing @xmath23 which results from the renormalization procedure _ must be planar _ ,
i.e. , it can be drawn without any two bonds crossing .
states of the form which fulfill this requirement will be called from now on _
planar states_. in more formal terms , let ( even ) @xmath12 be the number of nodes , and a bond is defined as an ordered pair @xmath25 , where @xmath26 are the nodes joined . in principle , @xmath27 .
we define the covered nodes by the bond as @xmath28 .
notice that , if @xmath29 and @xmath30 are consecutive , @xmath31 . given two bonds , @xmath32 and @xmath33
, we say that @xmath34 if @xmath35 .
if neither @xmath36 or @xmath37 , we say that the bonds _ cross_. a planar bond structure is defined as a set of @xmath24 bonds which _ do not cross_. thus , the bonds form a _
nested graph_. an important remark is that the two nodes joined by bond must have different parity .
[ fig : illust ] for an illustration .
let us assume that the nodes are indexed counterclockwise .
we can now define for each node @xmath15 a value @xmath38 to be either @xmath39 or @xmath40 depending on whether it is the source or the sink of a bond , as shown in fig .
[ fig : illust ] . of course
, the sum of all the @xmath38 around the full system should be zero : @xmath41 .
the @xmath38 can be considered as _
slopes _ of a height function , @xmath42 of course , this definition is not translation invariant , since we start counting at node 1 . in order to avoid that ,
let @xmath43 denote the absolute minimum of this height function .
then , we can define the _ absolute height function _
, @xmath44 .
its meaning is the following : it denotes the number of bonds passing above the link in the circle joining nodes @xmath15 and @xmath16 . by construction
, this absolute height function has , at least , one zero .
there is a close analogy between planar pairings and the _ dyck language _ @xcite .
a dyck word is a string of symbols from the alphabet @xmath45 such that the number of @xmath46 counted from the left is always greater or equal to the number of @xmath47 .
equivalently , they are the set of _ properly balanced _ parenthesis .
this means that their height function @xmath48 , as defined in eq .
, is positive for all @xmath15 .
the difference between our planar pairings and the dyck language resides entirely in the periodic boundary conditions .
if , in our _ circular _ planar structures , we break at the absolute minimum of the height function , the analogy with dyck words becomes complete .
how many different planar states are there for a system of fixed size @xmath49 ?
let us denote this value by @xmath50 . disregarding the ordering of the sites in each bond , which merely contributes a general @xmath51 factor
, we can provide a recursive relation .
site 1 must be linked to an even site , @xmath52 .
then it creates two _ regions _ , one of size @xmath53 and the other @xmath54 .
thus , we get @xmath55 along with @xmath56 , which is known in the literature as _ segner s recurrence
_ @xcite , which gives rise to the catalan numbers : @xmath57 given a planar state of the form , we can easily compute the entanglement entropy of any block @xmath58 : using 2 as the base for the logarithms , it coincides with the number of bonds which must be cut in order to separate it from the rest of the system @xcite .
@xmath59 \label{eq : define_entropy}\ ] ] where @xmath60 stands for the _ exclusive or _ ( xor )
symbol , which means that _ either _ @xmath61 or @xmath62 , but not both .
we can prove the following theorem which relates the height function and entanglement .
let @xmath63 $ ] denote the block @xmath64 .
then , @xmath65 } ) = h_{i-1 } + h_j - 2\min_{k\in \{i-1 \cdots j\ } } h_k .
\label{eq : entropy_height}\ ] ] the meaning of that equation is the following .
@xmath66 represents the bonds that _ enter _ the block from its left end , and @xmath67 the bonds which exit from its right . for an example , see fig .
[ fig : illust ] .
the block marked with the dashed box is @xmath68}$ ] .
the number of bonds entering from the left is @xmath69 , and the number of bonds leaving from the right is @xmath70 .
but not all those bonds contribute to the entropy .
some of them just _ fly over _ the block , and we can separate the block without touching them .
let @xmath71 be the number of those flying bonds , in our example @xmath72 , the bond from site 1 to site 8 .
the links entering from the left , @xmath66 are either overflying ( @xmath71 ) or not ( @xmath73 ) : @xmath74 .
similarly , on the right we have @xmath75 , and the block entropy is given by @xmath76
. we will proceed to prove that @xmath71 is given by the minimum of the height function inside the block .
since the bonds which contribute to the entropy , @xmath73 and @xmath77 , do not _ fly _ over the block , they must either _ end _ inside it ( @xmath73 ) or _ start _ inside it @xmath77 . since the bonds can not cut , the @xmath73 bonds from the left must have ended _ before _ any of the @xmath77 start . at that very moment , only the _ flying _ bonds will remain . in fig .
[ fig : illust ] , this moment takes place between sites 5 and 6 , @xmath78 , @xmath79 and the block entropy is @xmath80 .
thus , the minimum value of the height function is , exactly , @xmath71 .
we have @xmath81 , as required .
notice that we can rewrite expression as @xmath82})=(h_{i-1}-h_{min } ) + ( h_j - h_{min})$ ] , thus showing a connection between the block entropy and the _ average variation _ of the height within the block , i.e. the roughness of the interface .
the main difference is that the entanglement entropy gives special relevance to the boundaries .
the statistical properties of the ground states of hamiltonians of the form when the couplings @xmath83 are picked randomly and _ uncorrelated _ have been determined in a series of papers @xcite .
the sdrg procedure converges to the so - called _ infinite randomness fixed point _ ( irfp ) . along the rg ,
the variance of the effective couplings grow , and their correlation length decreases .
it has been shown that the average entanglement entropy of a block of size @xmath0 follows the expression @xcite : @xmath84 where @xmath85 is a scaling function and @xmath86 is the central charge of the associated cft , i.e. , the one which corresponds to the homogeneous ( conformal ) case , with all the @xmath14 equal . in our case , @xmath87 .
surprisingly , expression is very similar to the conformal expression @xcite : @xmath88 the scaling function @xmath85 is , in fact , rather similar to @xmath89 @xcite .
another relevant observable which helps characterize the irfp is the _ bond - length _ probability , i.e. , given a singlet bond @xmath90 , determine the probability distribution for its length @xmath91 , @xmath92 .
this value is directly related to the two - point correlation function @xcite . in the uncorrelated case ,
it is known to behave , for @xmath93 , as a power - law : @xmath94 , where @xmath5 @xcite .
our aim is to characterize the ground states of hamiltonian when the couplings @xmath14 are random , but present non - trivial correlations .
if these correlations are short ranged , they will be washed away by the renormalization procedure , and return to the irfp .
thus , we will consider the case of long - range correlations .
let us establish a procedure to obtain samples from sets of log - couplings @xmath95 which present long - range correlations , by employing a suitable fourier expansion : @xmath96 where @xmath97 are a set of allowed momenta , @xmath98 , with @xmath99 .
we do not include moment zero , since it would amount to a global constant which would be irrelevant for the sdrg .
the values @xmath100 and @xmath101 are chosen as independent random variables .
the phase @xmath101 is taken to be uniformly distributed in @xmath102 and @xmath103 where the @xmath104 are independent gaussian variates with zero average and variance one , and @xmath105 is a fixed spectral exponent . if @xmath106 , all momenta in expression get the same weight , and we obtain again an _ uncorrelated _ set of @xmath107 .
as we increase @xmath105 , the larger momenta get less and less weight , and we are left with only the lowest momenta .
this implies that the set of @xmath107 have stronger correlations .
[ fig : corrj ] ( a ) shows typical samples for increasing values of @xmath105 .
the ensemble of log - couplings presents zero correlations in momentum space , but strong correlations in real space for increasing @xmath105 .
the correlation function is translation invariant by construction , and given by @xmath108 for @xmath109 , fig .
[ fig : corrj ] ( b ) shows the correlation as a function of the distance , normalized to have a maximal value of one . for @xmath106 ,
the correlation is identically zero for all @xmath110 .
for @xmath111 it approaches a cosine function . the value @xmath112 , which will have special relevance in the rest of the text , appears marked .
we should remark that , although eq . makes perfect sense for all finite values of @xmath12 , the expression diverges in the thermodynamic limit for @xmath113 .
[ fig : diagrams ] shows some sample planar pairings for different values of @xmath105 , in the range from @xmath106 ( no correlations ) to @xmath114 ( large correlations ) , along with their corresponding height diagrams .
the average entanglement entropy as a function of the block size @xmath0 , for a fixed value of @xmath109 , @xmath115 realizations and several values of @xmath105 is shown in fig .
[ fig : entropy ] ( a ) .
the upper part of the panel is devoted to @xmath116 , while the lower one shows more detail for @xmath117 .
notice that , for @xmath117 , the function @xmath118 is nearly independent of @xmath105 .
we propose a finite - size fit of the form : @xmath119 where the scaling function @xmath85 is determined via a fourier series expansion in the same line as @xcite : @xmath120 the best fit values of @xmath121 are small and nearly independent of the spectral exponent @xmath105 .
we have found @xmath122 and @xmath123 , both slowly decaying as @xmath105 increases .
the inset of the top panel of fig .
[ fig : entropy ] shows these scaling functions for different values of @xmath105 .
the values of the exponent @xmath124 present more relevance .
fig [ fig : entropy ] ( b ) show these exponents , found by three different strategies : ( i ) _ finite - size _ , using a full fit to expression for @xmath109 , ( ii ) _ local exponent _ , fitting the entropy for small blocks to a form @xmath125 also for @xmath109 , ( iii ) _ global exponent _ , fitting @xmath126 to a form @xmath127 for different values of @xmath12 , up to @xmath128 .
the three expressions differ slightly for larger @xmath105 , although they keep a general trend : for @xmath117 , @xmath124 is very close to zero , while for @xmath111 we see @xmath129 .
this signals a _
volumetric _ growth of the entropy @xmath130 .
the discrepancies between the values of @xmath124 measured by the different strategies , as seen in fig .
[ fig : entropy ] ( b ) , may be of numerical origin . interestingly , for @xmath117 ,
the @xmath118 curves are nearly identical , and the best finite - size fit to the whole function is _ not _ given by the power - law expression , but expression , i.e. a logarithmic behavior .
even for @xmath131 , the best fit is logarithmic , but with a slightly larger prefactor .
it is difficult to determine whether there is a smooth crossover between @xmath106 and @xmath111 or a sharp transition at @xmath131 , below which the entropy grows logarithmically , i.e. : if the irfp extends to the region @xmath132 .
another interesting observable is provided by the study of the height function which characterizes the state , given by eq . .
as we will show , the profiles are fractals , of similar nature to the ones appearing in the study of rough interfaces @xcite .
let us define the roughness , or width @xmath133 , of the interface for a given length scale @xmath0 as the average deviation of the heights in windows of that size .
then , the family - vicsek ansatz assumes that @xmath134 .
[ fig : roughness ] ( a ) shows the roughness as a function of the window size @xmath0 , taking @xmath115 realizations for each value of @xmath105 .
the top frame shows a log - log plot , while in the bottom one only the @xmath135-axis is logarithmic .
the difference is notorious : for @xmath136 , the roughness follows a clear power law , with exponent @xmath137 which grows up to one ( shown as a straight line ) .
for @xmath117 , instead , the behavior is better fit by a logarithmic function @xmath138 .
this provides further support to the conjecture that the behavior for @xmath117 corresponds to the irfp .
panel ( b ) of fig .
[ fig : roughness ] depicts the probability distribution for the bond - length .
a power - law is established , i.e. , @xmath139 , and @xmath140 is shown to depend on @xmath105 .
for @xmath117 , the curves appear to be parallel , i.e. , show the same exponent , and only differing in their prefactor @xcite . fig .
[ fig : roughness ] ( c ) shows the values of the three exponents , entropy ( @xmath124 ) , roughness ( @xmath137 ) and bond - length distribution ( @xmath140 ) as a function of the correlation parameter @xmath105 .
notice that @xmath124 is very similar to @xmath137 , as suggested by relation which links the block entropy to the height fluctuations .
both exponents grow with @xmath105 , starting near zero for uncorrelated spin chains and saturating at a value close to 1 .
the bond - length exponent @xmath140 behaves in the opposite way , starting at @xmath5 for uncorrelated spin chains and decreasing towards zero .
the region for @xmath117 is peculiar : while the bond - length @xmath140 exponent is still @xmath141 , the other two exponents are very close to zero , since the true behavior is expected to be logarithmic . the limit @xmath111 is also rather special .
a look at the last panel of [ fig : diagrams ] shows that _ rainbow - like _ structures become more and more prominent .
the limit in which only the lowest momentum modulation survives gives rise to a perfect rainbow state , which presents volumetric entanglement @xcite , i.e. , @xmath130 .
this explains the limit @xmath129 for the entropy exponent for large @xmath105 .
similarly , the height function becomes a nearly perfect wedge , which explains the @xmath142 behavior . in that extreme ,
the bond - length distribution is completely flat , since all bond - lengths show up once for each realization , thus @xmath143 .
as it was briefly discussed in the introduction , planar pairings also appear naturally in the study of the secondary structure of folded rna strands @xcite . the model developed by wiese and coworkers @xcite works in the following way : ( 1 )
a pair of sites with different parity are chosen randomly and paired ; ( 2 ) further pairs are chosen in the same way , always under the constraint that no previous bonds can be crossed . in their seminal work @xcite
, the authors studied the roughness of the equivalent height function and the bond - length distribution , showing that they both follow a power - law behavior , @xmath144 and @xmath145 .
then they proved that @xmath146 .
if we assume the scaling equivalence of the roughness and the entropy , this result is also fulfilled in uncorrelated random spin chains , where we have @xmath147 ( because of the logarithmic behavior of the entropy ) and @xmath5 @xcite . on the other hand , this relation does not hold for correlated spin chains .
we may ask what is the range of validity of the relation @xmath148 ( or @xmath146 ) . extending the results of @xcite
we can provide a proof of that statement in the case of uncorrelated bonds . indeed , let us consider a block of size @xmath0 and
let us number the sites from @xmath149 to @xmath0 .
the bond at site @xmath15 will be cut by the block if it goes left and its length is larger or equal than @xmath15 , or if it goes right and its length is larger than @xmath150 .
so , we have an estimate for the average entropy : @xmath151 this equation implies a double integration . if @xmath6 , it leads to @xmath152 , as we desired . as it follows from fig .
[ fig : roughness ] ( c ) , this is not true for the planar state ensembles generated with correlated couplings .
in fact , in the _ rainbow _ limit , we have @xmath153 , which suggests a strong correlation between the bonds .
how strong is the connection between the rna folding and disordered spin chains ?
can we obtain an ensemble of couplings @xmath83 such that the ground states of hamiltonian correspond to the planar states obtained in rna folding ?
this question leads us to the study of the more general _ inverse sdrg problem_. if we regard the sdrg as a mapping between sets of couplings and planar pairings , we might be able to reverse the algorithm , and obtain the set of couplings which give rise to a certain planar pairing . in other terms , a parent 1d hamiltonian for a given planar state . in this section
we will show that ( 1 ) every planar state has a ( non - unique ) parent 1d hamiltonian and ( 2 ) an explicit algorithm to obtain the _ optimal _ set of couplings , in a sense to be determined later .
the aim is to obtain the logarithmic couplings , @xmath95 , given the set of bonds , @xmath154 .
our proposed algorithm works as follows ( see fig . [
fig : illust_inverse ] for an illustration ) : * sort the bonds in order of increasing length . *
consider the bonds of length one , fix their internal log - couplings to @xmath155 .
in the first row of fig .
[ fig : illust_inverse ] , we put a zero under links @xmath156 and @xmath157 .
* flank these zeroes with log - couplings of value @xmath149 at both sides .
see the second row of fig .
[ fig : illust_inverse ] , where the arrows in the new values point to the zero which they flank .
* now consider the bonds of length three .
find the effective log - coupling which would appear as their _ renormalization value _ ( which must be 2 ) .
flank them with log - couplings of value @xmath158 at both sides , as in the third row of fig .
[ fig : illust_inverse ] .
* consider the rest of the bonds in order of increasing lengths .
for each of them , find their _ renormalization value _ and flank them with log - couplings of value one unit higher .
* log - couplings may never decrease along the procedure .
if two values collide , take the larger
. this procedure yields couplings which , by construction , always give rise to the desired bond structure .
moreover , because the value of each bond is computed using the sdrg itself , we ensure a certain _ optimality condition _ : among the sets of couplings yielding the desired state , our choice will always require the minimal span of coupling values .
for example , this hamiltonian will yield the largest possible gap .
[ fig : rnainverse ] ( a ) shows the couplings which give rise to a give instance of the rna folding problem with @xmath159
. we have run @xmath115 simulations of the rna folding algorithm and obtained the optimal couplings for different system sizes up to @xmath109 .
[ fig : rnainverse ] ( b ) shows the ( translation invariant ) correlation function for the log - couplings in the @xmath160 , @xmath161 and @xmath162 cases .
the values present long range correlations , but not a clear power - law behavior .
moreover , the couplings field @xmath95 is _ not _ gaussian . in the inset of fig .
[ fig : rnainverse ] ( b ) we show the histogram , in logarithmic scale , for @xmath107 .
the marginal probability distribution is not gaussian .
instead , it is a power - law , with an empirical exponent close to @xmath163 .
since we have determined that all planar states have a 1d parent hamiltonian , we may still ask how dense are planar states within the hilbert space . in other terms , how generic they are .
we can define an ensemble of planar states for @xmath12 sites under the condition that all possible planar pairings have the same probability . in order to sample that ensemble
, we just apply a correction to the rna folding sampling strategy . in the rna folding algorithm , the pair of sites @xmath90 which will constitute the next bond is chosen with equal probabilities among those which do not cut any previous bond .
but , following that procedure , not all planar pairings are sampled with the same probability .
this can be corrected if the probabilities for each pair @xmath90 are not equal , but proportional to the number of planar pairings which are consistent with the presence of that bond .
let us consider a certain _ empty patch _ of length @xmath164 in a planar pairing which is under construction , i.e. , a set of contiguous spins which have not been paired yet . as we know , there are @xmath165 possible ways to create a planar pairing on that empty patch .
the spin with index 1 must be paired with some spin inside the patch , let us refer to its index as @xmath97 .
then , after bond @xmath166 is established , the number of different possible planar pairings will be @xmath167 .
thus the probability with which bond @xmath166 should be taken is just @xmath168 , which is known to be less than one by construction , as we see in eq . .
repeating this procedure , we can sample the planar pairing ensemble with equal probabilities .
we have found numerically the average block entropy as a function of the block size @xmath0 for this ensemble of states , and found that it grows as @xmath169 , with @xmath170 .
the precise value is not very relevant , but it allows us to conclude that planar states are highly non - generic quantum states , because for generic states we should obtain @xmath171 , i.e. , a volumetric growth of the entropy .
in this article we have applied the sdrg to study the ground state properties of a strongly disordered random spin chain with long - range correlations between its couplings .
the states can be described as valence bond states with planar bond structures , and they can have arbitrarily large entanglement entropy .
concretely , we have chosen the couplings such that their logarithm is expressed as a fourier series with random coefficients , falling as a power - law of the momentum @xmath172 .
for @xmath117 the behavior is very similar to the infinite randomness fixed point ( irfp ) found for uncorrelated coupling constants .
nonetheless , for @xmath136 , the block entropy behaves as a power - law of the block size , @xmath173 , with @xmath124 a function of the exponent @xmath105 which seems to interpolate smoothly between @xmath174 and @xmath175 as @xmath111 .
the bond length probability , which is related to the correlator , is also characterized by a power - law , @xmath6 , with @xmath5 for @xmath132 and falling to @xmath176 for @xmath111 .
this extreme , @xmath111 , corresponds to the case where only the lowest momentum @xmath177 contributes to the correlation between the couplings , and the state becomes a _ rainbow state_. as we have shown , the planar states can be mapped to a 1d interface , whose roughness behaves approximately like the entanglement entropy , as it is suggested by expression .
remarkably , the system described constitutes a family of local 1d hamiltonians whose ground states violate the area law to any desired degree .
we have also considered the inverse renormalization problem : given a ( planar ) valence bond state , to obtain its ( 1d ) parent hamiltonian . in this way we were able to study the ensemble of random spin chains whose ground states would correspond to the planar structures which show up in other physical situations , such as the rna folding problem .
these _ engineered _ random spin chains present a behavior of the entanglement entropy and the correlators which do not correspond to any value of @xmath105 .
this suggests that the phase diagram of random spin chains with large correlations between the couplings is richer than expected .
inhomogeneous spin chains can be mapped , in some cases , to models which represent the motion of fermionic matter on a curved spacetime @xcite , where the metric is given by the coupling constants .
thus , our study shows that the statistical properties of the metric show up as statistical properties of the entanglement of the vacuum , i.e. , the ground state of the corresponding hamiltonian .
moreover , we can also find , using the inverse renormalization algorithm , the optimal spatial geometry which gives rise to a certain vacuum entanglement .
these results may shed light on the relation between entanglement and space - time @xcite .
we would like to thank j. cuesta for insights into the statistical mechanics of rna folding .
this work was funded by grants fis-2012 - 33642 and fis-2012 - 38866-c05 - 1 , from the spanish government , quitemad+ s2013/ice-2801 from the madrid regional government and sev-2012 - 0249 of the `` centro de excelencia severo ochoa '' programme .
instead of using the minimal length of the bond , we can also employ their _ renormalized length _
, i.e. , the number of links which were decimated in its formation during the sdrg procedure . as opposed to the minimal length , the renormalized length can be larger than @xmath178 .
this alternative definition does not affect the computation of entropies , but it changes the heights and the bond - lengths .
but the universal features described in this chapter were not changed , and thus we do not show specific results . | average block entanglement in the 1d xx - model with uncorrelated random couplings is known to grow as the logarithm of the block size , in similarity to conformal systems . in this work
we study random spin chains whose couplings present long range correlations , generated as gaussian fields with a power - law spectral function .
ground states are always planar valence bond states , and their statistical ensembles are characterized in terms of their block entropy and their bond - length distribution , which follow power - laws .
we conjecture the existence of a critical value for the spectral exponent , below which the system behavior is identical to the case of uncorrelated couplings . above that critical value
, the entanglement entropy violates the area law and grows as a power law of the block size , with an exponent which increases from zero to one .
similar planar bond structures are also found in statistical models of rna folding and kinetic roughening , and we trace an analogy between them and quantum valence bond states . using an inverse renormalization procedure
we determine the optimal spin - chain couplings which give rise to a given planar bond structure , and study the statistical properties of the couplings whose bond structures mimic those found in rna folding . |
entanglement is one of the strangest features of quantum mechanics .
few would disagree that it distinguishes quantum mechanics unequivocally from classical physics .
while it has played an essential role in the understanding of quantum mechanics , paradoxically it has been mostly absent from the theory of quantum phase transitions . as in the case of classical critical phenomena , the theory of a quantum phase transition is naturally based on the description of the scaling behavior of local observables .
indeed , much of the theory of quantum criticality is based on this extension of the framework of classical criticality allowing for the natural dynamics specified by the quantum mechanical hamiltonian of the system.@xcite in this standard description the main effect of quantum mechanics is captured by the dynamic critical exponent associated @xmath4 that governs the relative scaling of space and time and is determined by the dynamics of the quantum system .
it is a major theoretical challenge is to understand the role of observables that have no classical analogue in quantum criticality .
quantum entanglement has been proposed as a candidate for such a measure.@xcite while there are other measures of entanglement , the entanglement entropy has proven to be the most tractable analytically and simple to define .
one begins with a pure state of a bipartite quantum system @xmath5 . by restricting oneself to observing either @xmath6 or @xmath7 subregion ,
the subsystem is described by a mixed state with a non - trivial reduced density matrix .
suppose that @xmath6 is the observed region so that the degrees of freedom in @xmath7 are traced out .
the reduced density matrix is @xmath8 .
non - local quantum correlations are then encoded in the von neumann entropy , @xmath9 . for the case of a total system @xmath10 in a system in a pure state ,
the entanglement entropy is symmetric @xmath11 so that @xmath12 should only depend on common properties of the two regions .
given the non - local character of the entanglement entropy , its behavior in extended systems and quantum fields theories has a strong geometric flavor .
an important early was the realization that in systems away from quantum criticality in @xmath13 space dimensions , systems dominated by short distance physics , the von neumann entanglement entropy scales as the _ area _ of the observed region , @xmath14 , where @xmath15 is the linear size of the region ( say @xmath6 ) and @xmath16 is a non - universal constant.@xcite the scaling behavior of the entanglement entropy has been studied extensively in 1d quantum critical systems and it is by now reasonably well understood .
such systems are effectively relativistic ( @xmath17 ) @xmath18 dimensional conformal field theories ( cft ) .
it has been shown that in 1d the entanglement entropy of a subsystem @xmath6 of linear size @xmath15 of an otherwise infinite system ( _ i.e. _ of linear size @xmath19 ) obeys a logarithmic scaling law,@xcite @xmath20 , where @xmath21 is the central charge of the cft , and @xmath22 is the short distance cutoff .
the growing popularity of entanglement entropy study is due to the fact that many universal properties of quantum systems like the central charge , excitation spectra or boundary entropy can be extracted from the entanglement entropy without the need to specify an observable : all that is needed is a consistent partition of the system .
in addition , the _ dynamical _ entanglement entropy , entanglement generated in a quantum quench , has also been studied recently for a number of one - dimensional quantum critical systems.@xcite even in strongly disordered quantum systems , which do not have a local order parameter such as random spin chains at infinite disorder fixed points@xcite as well as at the anderson transition and quantum hall plateau transition,@xcite the scaling behavior of the entanglement entropy has been shown to be a faithful measure of quantum criticality .
however , much less is known about the scaling of entanglement in spatial dimensions @xmath23 , and how it relates to the scaling properties of local observable . while the leading scaling behavior of the entanglement entropy is the area law ,
its prefactor is not universal , except in @xmath24 where the area law becomes the universal logarithmic scaling .
for @xmath23 the only universal contributions may only arise from sub - leading terms ( relative to the area law ) .
the study of these universal corrections requires more detailed and subtle calculations .
in general the situation is not as clear .
an exception is the special case of _ topological phases _ in @xmath25 , where there is a finite , i.e. @xmath26 , universal correction to the area law which is given in terms of the topological invariants of the effective topological field theory of the phase .
@xcite recent work has showed that several quantum critical systems in @xmath25 do not have logarithmic sub - leading corrections and , instead , have a _ universal _
finite term.@xcite in this paper we reexamine the scaling behavior of entanglement in a special class of quantum ( multi)critical points in @xmath25 with dynamic scaling exponent @xmath27 .
these conformal quantum critical points have the special property that the amplitudes of field configurations @xmath28 in their wave functions have local scale - invariant weights.@xcite simple examples of such systems are 2d quantum dimer models@xcite and their generalizations.@xcite the norm of these ground state wavefunctions is thus equivalent to a partition function of a suitable two dimensional classical statistical model at criticality . labeling a configuration of the classical statistical model by the field @xmath29 , this can be expressed , @xmath30 with the explicit form of the wavefunction , one can compute the entanglement entropy by constructing the replicated reduced density matrix , @xmath31 . for these models
it was shown that the entanglement entropy has a universal sub - leading correction , @xcite @xmath32 where @xmath33 for the case of the quantum dimer models at their quantum critical ( rokhsar - kivelson ) point . here
@xmath3 is the compactification radius ( see below ) of the coarse - grained height model , dual to the qdm , the quantum lifshitz model.@xcite the same scaling behavior has been shown to hold , within the @xmath34-expansion , in relativistic @xmath35 quantum field theory , the prototype of a quantum phase transition.@xcite in ref.@xcite it was shown that the computation of the `` spectral moment '' of the reduced density matrix @xmath36 for the case of local scale - invariant wave functions is the same as the computation of a ratio in which the numerator is given by a classical partition function of the form of eq . for @xmath1 copies of the system which are required to agree on the boundary of region @xmath6 , while in the denominator no such constraint is imposed .
posed in this way , the computation of @xmath37 , and the von neumann entanglement entropy @xmath38 , is a problem in a `` replicated '' classical 2d critical systems with a `` conformal defect '' along some curve,@xcite i.e a problem in boundary ( euclidean ) conformal field theory ( bcft ) on a somewhat unusual manifold .
the scaling behavior of @xmath37 with sub - system size is then not as surprising as it is strongly reminiscent of the finite - size scaling of the free energy in large finite critical systems.@xcite on the other hand , it is known that the structure of the cft strongly determines its possible allowed bcfts .
thus , the structure of the cft that underlies these scale invariant wave functions must similarly play a key role in the behavior of the entropies @xmath37 .
this natural relation of the entanglement entropy with boundary ( or surface ) critical behavior on a suitably defined manifold appears naturally in the path integral formulation.@xcite in this work we use bcft methods to reexamine this problem for the case of the quantum lifshitz wave function studied before in refs.@xcite paying close attention to the role of the compactification radius @xmath3 by constructing explicitly the boundary states of the associated bcft.@xcite here we show that in the limit of a large aspect ratio , @xmath39 ( where @xmath22 is the short - distance cutoff and @xmath40 is the linear size of the full system ) the entanglement entropy has the universal finite term @xmath33 , in agreement with the previous result of ref.@xcite . on the other hand , numerical estimates of the sub - leading corrections of the scaling of the entanglement entropy in the 2d qdm , obtained using the wave function of the lattice model and using extrapolation methods
, suggested that there may be additional universal contributions to @xmath0.@xcite these authors attribute the apparent disparity to the boundary conditions used in refs.@xcite .
this motivated us to reexamine the bcft of this problem and to give an explicit form of the boundary states .
in bcft , boundary conditions are in one to one correspondence with the primaries of the bulk cft.@xcite for the case of multiple copies at hand here the bulk cft possesses additional symmetries .
the imposition of boundary conditions along a specified curve , a `` conformal defect , '' reduces these symmetries .
a similar situation occurs when considering defect lines in the ising model.@xcite a better understanding of systems of @xmath1-coupled critical systems is also crucial for reasons outside the entanglement entropy . from a broader viewpoint ,
a familiar condensed matter context are defects in lattice models . as mentioned earlier , one such example are defect lines in the ising model.@xcite more generally , conformal defects are very hard to classify . even for two copies of a free bosonic theory , central charge @xmath41 ,
the complete classification remains elusive.@xcite theoretically , the classification of such defects is an important issue .
these represent fixed points of a bcft .
a familiar example is the @xmath42-channel kondo problem.@xcite more complicated examples of @xmath1 intersecting theories also appear in a more applied context .
the intersection of multiple quantum wires at the same point is one example.@xcite as mentioned earlier , systems of @xmath1-coupled degrees of freedom also make their appearance in the theory of quantum brownian motion@xcite and in the dissipative hofstadter model.@xcite in this work , we add the entanglement entropy in conformal quantum critical points to this list .
we compute the quantity @xmath31 in terms of the original degrees of freedom and show that there is no additional factor as suggested by ref.@xcite . while our result agrees with our previous work , we note that this happens only in the asymptotic limit of a long cylinder and that for finite sized systems , there is generically a non - trivial @xmath1-dependence .
to arrive at these conclusions , we identify a new boundary state describing the boundary condition that @xmath1 copies of a system are stitched together at an interface , and we argue that additional sub - leading corrections to eq.([eq : sqcp ] ) are a result of differing boundary conditions used .
we show that the correct boundary condition for the original degrees of freedom is one where the extra factor vanishes .
this turns out to correspond to a subclass of conformal defects , at the common boundary .
in addition , we show that there is a geometrical interpretation to @xmath31 as the ratio of classical partition functions defined on different tori , and that the universal sub - leading term is the `` thin torus limit '' of this ratio . with an explicit calculation
, we find analytically that in the @xmath43 the entanglement spectrum is given by the characters of the underlying conformal field theory describing the ground state wavefunction .
this provides a case where it is possible to verify analytically the conjecture of li and haldane.@xcite finally , we demonstrate that in a string theory language , it is possible to think of the common boundary condition as a brane with a background gauge field and that @xmath31 is the expectation value of an appropriately defined vertex operator ( in the `` target space '' not in the cft of the world - sheet ) .
the paper is organized as follows : in the next section , we review the construction of @xmath31 for conformal quantum critical models .
we then specialize to the quantum dimer model in section [ sec : qdm ] with the bulk of the calculation in sec [ sec : calc ] .
a brief review of boundary conformal field theory in [ sec : bcft ] is given as it plays a central role in the main calculation . using the explicit expression for @xmath31
we show in section [ sec : es ] that the entanglement spectrum has a level degeneracy indicative of the underlying conformal field theory describing the ground state wavefunction . in section [ sec : disc ]
we relate the boundary condition to a brane with a specific magnetic and electric fields and show that @xmath31 can be computed as the expectation value of a vertex operator .
* note added to the text : * while this paper was being refereed , we received a preprint by oshikawa @xcite and became aware of a number of inconsistencies in our treatment of the boundary state .
in particular , it was pointed out that our glueing condition violated a permutation symmetry in the problem .
these corrections and its consequences are noted in section [ sec : note ] .
we are interested in the von neumann entropy for systems at conformal quantum critical points . these are systems where the norm of the ground state wavefunction is equivalent to a partition function of a two dimensional classical statistical model at criticality ,
_ i.e. _ eqn .
( [ eqn : gs ] ) . with the explicit form of the wavefunction
, one can compute the entanglement entropy .
this is the von neumann entropy of the reduced density matrix @xmath44 since the ground state wave function is a local function of the field @xmath45 , a general matrix element of the reduced density matrix is a trace of the density matrix of the pure state @xmath46 $ ] over the degrees of freedom of the `` unobserved '' region @xmath7 , denoted by @xmath47 . hence the matrix elements of @xmath48 take the form @xmath49 \,\
, e^{\displaystyle{-\left(\frac{1}{2 } s^{a}(\phi^{a}_{i } ) + \frac{1}{2 } s^{a}(\phi^a_{i+1 } ) + s^b(\phi^b_{i } ) \right ) } } \ , , \nonumber \\\end{aligned}\ ] ] where the degrees of freedom satisfy the _ boundary condition _ at the common boundary @xmath50 : @xmath51 this problem can be thought of as a problem in boundary conformal field theory by letting @xmath52 parameterize the boundary interaction between copy @xmath53 and @xmath54 , at the strong coupling limit @xmath55 the copies are required to have the same configuration on the boundary and at @xmath56 the copies do not interact at the boundary .
formally , the trace over @xmath1 copies of the reduced density matrix can be written as the ratio of the partition functions in these two limits , @xmath57 the theoretical challenge is to compute this ratio .
various formal mathematical devices have been devised , but until now a direct approach has been lacking .
it is desirable to understand the boundary condition in terms of the original degrees of freedom since those are most directly related to a physical dimer covering of the lattice .
here we consider the simple case where the action in ( [ eqn : gs ] ) is the gaussian free field theory , described by a bosonic field with the property that it is identified on a circle of radius @xmath3 , @xmath58 , @xmath59 one can think of this ground state ( [ eqn : gs ] ) as a superposition lattice configurations in a statistical model such as the dimers on a square lattice @xcite and its generalizations @xcite or as coverings in the eight - vertex model @xcite with special choice of the baxter weight @xcite .
one can think of this action as describing a @xmath18 dimensional system defined on some manifold , a cylinder here . to connect with our pervious work , we let the circumference be of length @xmath15 and the length @xmath40 .
periodic boundary conditions are imposed in the @xmath60-direction .
the field @xmath61 is a linear combination of the holomorphic @xmath62 and anti - holomorphic @xmath63 parts , @xmath64 .
holomorphic and anti - holormophic mode expansions can be written as @xmath65 where @xmath66 and the zero modes are given by @xmath67 the constants @xmath68 are canonically conjugate to these zero modes , @xmath69=i$ ] .
the primaries of the boson field theory are labelled by the value of the zero modes , @xmath70 . where @xmath40 is the length of the cylinder and @xmath15 is the length of the circumference .
[ fig : copies],scaledwidth=40.0% ] at the ends of the cylinder , dirichlet boundary conditions can be chosen @xmath71 .
on the oscillator modes , one finds that @xmath72 here , @xmath73 where @xmath74 .
for the zero modes , dirichlet boundary conditions set the winding modes @xmath75 . at the boundary @xmath76 continuity of the fields implies that @xmath77 .
this restricts the modes to obey @xmath78 which amounts to simple continuity of the field at the boundary in agreement with ( [ eq : bcphigamma ] ) .
now , we choose to observe the lower half of the cylinder ( @xmath79 ) , region @xmath7 , and compute the reduced density matrix for the remaining region @xmath6 by the replica trick .
the fields at the boundary should be continuous with each other as described in ( [ eq : bcphigamma ] ) and follow an explicit relationship among the modes given by ( [ eqn : continuity ] ) .
computing the reduced density matrix is hence a computation of a ratio of partition functions ( [ eqn : rhoa ] ) .
this is most simply done in the boundary conformal field theory framework .
this formalism is briefly reviewed in [ sec : bcft ] .
as it will play a central role , those unacquainted should proceed there .
more extensive reviews of the subject matter can be found in other canonical texts.@xcite to make use of boundary conformal field theory , we fold the system at the boundary at @xmath76 ( see figure [ fig : copies ] ) so that there are @xmath80 cylinders of half the total length @xmath40 . at @xmath81 , dirichlet boundary conditions were imposed on the ends which relates the holomorphic and anti - holomorphic modes .
the boundary condition ( [ eqn : dbc ] ) can be regarded as an eigenvalue expression for the annihilation operators , and the ishibashi states can be written as the coherent state @xmath82 where and @xmath83 and the state @xmath84 labels states in the fock space where the winding mode @xmath75 , as required by dirichlet boundary conditions ( [ eqn : dbc ] ) . with two copies , the matrix @xmath85 is given by , @xmath86 .
the boundary state is a linear combination of these ishibashi states , given by ( [ eqn : bstate ] ) and ( [ eqn : bosonbstates ] ) @xmath87 @xmath88 is the @xmath89-factor associated with the dirichlet boundary condition for the free boson and can be computed explicitly as mentioned in ( [ eqn : gfactor ] ) .
we now construct the non - trivial boundary state that is at the boundary @xmath76 .
below , we specialize to the case of @xmath90 copies , but the result is easily generalized to arbitrary @xmath1 . at @xmath56 , the two copies are decoupled , and the boundary condition becomes @xmath91 and the momentum modes are restricted to obey , @xmath92 the equation ( [ eqn : uncoupled ] ) can be regarded as eigenvalue equations for the destruction operators and the ishibashi state for each can be written succinctly as @xmath93 where @xmath94 a similar construction can be done for the @xmath95 case . in a previous version of the paper , @xmath96 was given incorrectly and was pointed out by oshikawa @xcite to violate a permutation symmetry of the problem . instead
all possible boundary conditions must be summed ( see section [ sec : note ] ) .
the resulting matrix is given by equation ( [ eqn : matrix ] ) .
the state @xmath97 denote states with eigenvalues obeying the matrix relationship @xmath98 where the form of @xmath99 depends on the value of @xmath52 .
it is given by ( [ eqn : pfree ] ) for @xmath100 and ( [ eqn : matrix ] ) for @xmath95 .
hence , the boundary state can be written as a linear combination of ishibashi states again , @xmath101 where the sum is over modes on the correct compactification lattice .
this subtlety was neglected in a previous version of the paper , but was pointed out by oshikawa @xcite . here , the @xmath89-factor will play an important role .
we call the @xmath89-factor associated with the @xmath102 limit @xmath103 and the @xmath89-factor associated with the @xmath104 limit @xmath105 .
we leave the specific evaluation of these normalization factors till later . the partition function
is then easily evaluated as the overlap of the two boundary states .
@xmath106 where the quantity @xmath107 is the piece only involving the zero modes and @xmath108 involves the oscillator modes .
the respective matrices describing the glueing conditions inserted , one readily finds , @xmath109 the zero modes require some care . in a previous version of this paper , the winding and momentum modes were assumed to be independent , but the glueing matrix forces a enforces a condition .
instead one must sum over a lattice @xmath110 as noted by oshikawa @xcite .
this is dual lattice which satisfies the glueing condition on the zero modes , @xmath98 where @xmath99 is given by ( [ eqn : matrix ] )
. hence for @xmath102 , one finds , @xmath111 by a similar set of arguments , for @xmath112 one finds , @xmath113 the results are easily generalized to arbitrary @xmath1 . writing in terms of @xmath114-functions ( see [ sec : appendixtheta ] ) , the uv and ir limit partition functions are given by , @xmath115 as a non - trivial check , for @xmath116 we note that @xmath117 is the modular parameter for a cylinder of total length @xmath15 and observe that the term in paranthesis is the @xmath118 character ; it is the partition function for @xmath1-decoupled bosons on a cylinder of length @xmath15 as it should be . in light of this observation ,
@xmath105 should be equal to 1 .
the factors @xmath103 and @xmath105 are normalization factors that have yet to be fixed .
these can be fixed by cardy s conditions and one finds that , @xmath119 crucially , the ratio @xmath120 , and hence we arrive at the main result of this paper , @xmath121 where @xmath122 and @xmath123 are given by ( [ eqn : zinfty ] ) and ( [ eqn : zzero ] ) respectively and we have factored out the constant factors @xmath124 explicitly . here , we have included the regulator dependent contribution in the function @xmath125 that comes from the short distance cutoff which we have hitherto suppressed.@xcite we comment on its specific form below . more importantly , we note that the each of terms @xmath126 are characters of a @xmath118 , @xmath127 conformal field theory.@xcite thus far , we have been concerned only with the regulated part of @xmath31 and have neglected the divergent contributions from the short distance cutoff .
these can be recovered by a careful @xmath128-function regularization.@xcite in general , it was shown that in the limit of interest the free energy for a system on a _ smooth _ open manifold scales as,@xcite @xmath129 here @xmath6 is the area of the cylinder , @xmath130 the perimeter of the boundary and @xmath131 is the euler character of the manifold ( zero for cylinders ) .
the coefficients @xmath132 and @xmath133 depend on the short distance cutoff .
the order one contributions are what have been computed explicitly thus far .
we note in both limits , the total area is simply @xmath134 where @xmath6 is the area of a single cylinder so that in the ratio ( [ eqn : rhoa ] ) this contribution cancels .
the difference in the two limits lies in the perimeter of the boundary . in the decoupled limit
, there are @xmath80 unshared boundaries of length @xmath15 and @xmath1 shared ( where the two halves are joined ) boundaries so that the total perimeter of the boundary is @xmath135 . meanwhile ,
in the strongly coupled regime all the manifolds coincide ( smoothly ) on a single interface of length @xmath15 and there are still the same @xmath80 unshared boundaries .
the total length of the perimeter in this case is @xmath136 .
the important point is that there is also a divergent non - universal contribution to @xmath31 . to leading order , @xmath137 where the prefactor @xmath133 is non - universal and depends on the short distance regulator.@xcite it is important to note that this analysis neglects the effects cusps and corners that occur if the boundary is not smooth , giving rise to conical singularities , but these have been shown to give @xmath15 dependent contributions that scale as a power law that may be non - trivial functions of @xmath1 and to additional universal ( logarithmic ) corrections to the entanglement entropy.@xcite it is important to observe that as @xmath138 , @xmath139 as it should be .
now , in the present long cylinder limit the @xmath114-functions and @xmath140-function are equal to the identity to leading order in @xmath141 .
our main result is that the universal sub - leading term to the entanglement entropy is , @xmath142 where , once again , @xmath3 is the compactication radius of the boson .
this agrees with numerical result for the quantum dimer model.@xcite
while the result for @xmath0 is the same as our previous result,@xcite the interpretation is much different .
the full analytic expression for @xmath31 , for instance , differs and it is only in the asymptotic limit that the two results are identical . in the very long cylinder limit ,
@xmath143 the characters asymptotically approach the identity , and one finds that @xmath31 identically reduces to the one obtained previously.@xcite however , for large but finite values of the aspect ratio @xmath144 , @xmath31 has a non - trivial @xmath1-dependence .
naturally , an immediate question is what information is contained in @xmath31 in this picture ?
we show below that the entanglement spectrum is indeed described by the underlying conformal field theory describing the ground state wavefunction .
more importantly , several immediate questions arise .
first , the form of @xmath31 is relatively simple and one wonders if there is a deeper reason for this .
secondly , we have constructed boundary states @xmath145 and @xmath146 which lack a straightforward classification as free or fixed .
they represent instead a coupling between different copies .
an understanding of these states is clearly desirable .
lastly , our result for @xmath0 hinged on @xmath120 and a natural question is whether other values are possible . in
what follows , we address each of these issues . with an explicit expression for
@xmath31 , one in fact can construct all the moments of the rnyi entropy , @xmath147 one can examine the degeneracy of states by examining the higher moments , @xmath43 , of the rnyi entropy .
it has been postulated , but not shown analytically , that the higher moments of the rnyi entropy should be given by the characters of the underlying conformal field theory describing the ground state wavefunction.@xcite the characters of a conformal field theory count the number of independent states occurring at a given energy level . here
, we find that in the @xmath43 limit the contribution from the strong @xmath102 coupled sector tends to the identity so that the rnyi entropy in this limit is given by , @xmath148 where the @xmath149 indicate subleading constant contributions .
here , @xmath150 is the euler function that is related to the @xmath140-function through ramanujan s identity .
@xmath151 @xmath152 is related to the partitions of integers , @xmath153 , and @xmath154 where @xmath155 , @xmath156 in the long cylinder limit , @xmath157 is exponentially small and @xmath158 .
one finds that , @xmath159 asymptotically the multiplicities are given by the the partition of integers which exactly describes the number of states at a given energy in a @xmath127 free bosonic conformal field theory .
therefore , in a finite size computation of the entanglement spectrum the degeneracies of the eigenvalues of the reduced density matrix are ( asymptotically ) given by the integers @xmath153 .
this relation was conjectured ( for quantum hall wave functions ) by li and haldane.@xcite one issue still untouched by the preceding discussion is where additional universal corrections to @xmath0 might come from .
some insight can be gained by realizing that the problem of @xmath80 free bosonic field theories interacting only at a common boundary has been studied in the context of quantum brownian motion in a magnetic field,@xcite and open strings in a background gauge field .
@xcite the connection can be seen more concretely by considering @xmath31 in the path integral formulation @xmath160 } } { \int d\varphi \ , e^{-s_0[\varphi ] } } , \label{eq : newrep}\ ] ] where @xmath161=\lim_{\lambda \to \infty } s_\lambda[\varphi]$ ] and @xmath162=\lim_{\lambda \to 0}s_\lambda[\varphi]$ ] describes the bosonic action of @xmath1 scalar fields satisfying boundary conditions specified by @xmath52 . once again , the numerator describes @xmath1 fields that are forced to coincide at the boundary ( hence @xmath163 ) and in the denominator the @xmath1 fields are decoupled from each other ( hence @xmath164 ) . we will now write @xmath165 $ ] in a form that we find more useful as follows .
the first step in understanding this problem in terms of quantum brownian motion in a magnetic field or , equivalently , open strings in a background gauge field , is to fold the system accross the boundary , thus doubling the number of fields .
let @xmath166 , with @xmath167 denote a @xmath80 component scalar field whose upper @xmath1 components label the ( folded ) fields from the @xmath6 regions and its remaining ( lower ) @xmath1 components are those of the @xmath7 region , @xmath168 the action for the @xmath169 field is ( for so far unspecified boundary conditions ) @xmath170 = \frac{1}{2 } \int d^{2}x \ , \sum_{i=1}^{2n } \left(\partial_{\mu}\phi^{i } \right)^{2 } .
\label{eqn : gauged}\ ] ] here we choose our coordinates so that @xmath171 is the direction parallel to the length of the cylinders and @xmath172 runs along the circumference .
now we make a @xmath173-dual " transformation on the @xmath7 field , @xmath174 .
this corresponds to a symmetry of the action ( on the @xmath7 fields ) with respect to the interchange of their winding and charge modes,@xcite @xmath175 here @xmath176 and @xmath177 label the winding and charge numbers of the zero modes of the fields @xmath178 . under @xmath173-duality
the compactification radius @xmath3 transforms as @xmath179 , while neumann boundary conditions transform into dirichlet boundary conditions , and viceversa .
the fields only interact with each other at the common boundary via the boundary conditions . to this end
we introduce a field @xmath180 localized at the boundary , @xmath181 = \frac{1}{2}\int d^{2}x \ , \ , \sum_{i=1}^{2n } \left ( \partial_{\mu } \phi^{i } \right)^{2 } - \oint dt \ , a_{i } \partial_{t } \phi^{i}.\ ] ] where @xmath182 is an antisymmetric matrix we define below .
formally , this action describes the `` dissipative hofstadter model '' @xcite ( with vanishing potential ) . upon varying the action ,
the fields @xmath183 are found to obey the usual wave equation with the boundary condition , @xmath184 if @xmath182 is a constant matrix , independent of @xmath183 , and anti - symmetric , then clearly @xmath185 .
one can think of @xmath182 as the 2-form field strength tensor associated with a gauge field @xmath186 .
now , letting @xmath182 be the @xmath187 matrix , @xmath188 it is readily seen that eq.([eqn : bcf ] ) yields the desired boundary conditions at the common boundary for the scalar fields by a suitable choice of the @xmath189 matrix @xmath190 .
this construction can be used to represent both the numerator and denominator of eq . by the choices ( in this basis ) @xmath191 for @xmath163 and @xmath164 respectively .
previously , a matrix was given for @xmath192 that violated the permutation symmetry present in the boundary condition .
summing over possible glueing conditions yields the matrix above .
the matrix @xmath192 is then identical to the matrix @xmath193 ( see equation ( [ eqn : matrix ] ) ) of oshikawa @xcite up to an overall unimportant sign . in analogy with electromagnetism
, there is a electric and magnetic field at the boundary @xmath194 that mediates the interaction between the @xmath80-dimensional systems . in terms of quantum
brownian motion,@xcite one can regard the problem as a particle at @xmath194 on a @xmath80-dimensional space moving in a electric and magnetic field . as an open string
, one sees that the boundary can be thought of as a brane carrying a magnetic and electric field.@xcite the analogy also gives a possible interpretation of the result obtained by ref .
@xcite . in the context of quantum brownian motion in @xmath1-dimensions
, it has been shown that there is a plethora of possible boundary states where @xmath124 is non - trivial .
these boundary states correspond to a different electromagnetic field and hence a different coupling of the replicas in our picture .
we have thus constructed a gauge field ( and associated 2-form ) that describe the boundary interaction .
one can go further by rewriting @xmath31 in the suggestive form @xmath195 where @xmath196 corresponds to the choice @xmath197 and @xmath198 to @xmath199 ( both defined above ) .
the universal sub - leading correction to the entanglement entropy is the asymptotic behavior of this correlation function . in string theory , such objects
are generically called vertex operators.@xcite this operator can be understood as counting a topological charge @xmath200 where the field @xmath183 is a map , @xmath201 .
integrating by parts yields the correct boundary field .
importantly , there exists a basis of the field @xmath202 where @xmath182 is an @xmath1-block diagonal matrix of anti - symmetric @xmath203 tensors .
@xmath183 can then be written as a tensor product of maps @xmath204 and @xmath205 can be thought of as a product of holonomies characterized by the homotopy group @xmath206 .
an interesting observation is that the gauge field @xmath186 is fixed by the free part of the action of eq.([eqn : gauged ] ) so that a perturbation that brings the system into a topological phase , e.g. @xmath207 has no effect on the boundary condition and hence @xmath186 remains the same .
we expect that this correlation function , and hence the entanglement entropy , should take different values in the different phases .
while this paper was being put into production , we received a preprint by oshikawa @xcite and we became aware of a number of inconsistencies in our previous work .
in particular , the matrix used to described the coupled boundary condition did not possess the full permutation symmetry of the problem and the winding and momentum quantum numbers of the zero modes were not treated correctly . as noted by oshikawa , @xcite there is a permutation symmetry in the problem and each copy and definition of region a and b are interchangeable .
one should then sum over all possible combinations of matrices @xmath208 , where @xmath208 is a shift matrix .
the glueing condition is then , @xmath209 the resulting matrix @xmath193 is in fact the matrix @xmath193 given by oshikawa @xcite up to an overall unimportant sign .
explicitly , @xmath210 using the sum of all possible glueing conditions because of the permutation symmetry , the partition function ( [ eqn : zinfty ] ) is given by the expression found by oshikawa @xcite .
@xmath211 here , @xmath110 is a lattice that satisfies the boundary condition on the zero modes , @xmath98 the result for @xmath116 is unchanged and given by equation ( [ eqn : zzero ] ) . as pointed out by oshikawa @xcite ,
the correction normalization is , @xmath212 the universal subleading term to the entanglement entropy is dependent on this ratio of @xmath89-factors so that the result ( [ eqn : result3 ] ) is now @xmath213 several results in the paper are modified because of this . in particular , the quantity @xmath31 ( i.e. equation ( [ eqn : result1 ] ) ) is now given by @xmath214 then in the limit @xmath215 one finds that the distribution of eigenvalues of the rnyi entropy for large moments @xmath1 , is given by @xmath216 where @xmath150 is the partition of integers and @xmath149 include subleading constant terms .
eigenvalues are distributed according the level counting for the free @xmath127 boson theory that describes the ground state wavefunction , as mentioned in the body of the paper .
before concluding with broader and more speculative issues , we summarize our results for the quantum dimer model .
we found that in the limit @xmath217 , @xmath31 is given by the expression of eq.([eqn : corr ] ) .
this was done in terms of the original degrees of freedom and it is hoped that this clarifies what boundary condition must be used at the common interface . in the limit @xmath218 of @xmath219 , it was found that the universal finite part of the entanglement entropy is @xmath220 this result coincides with the numerical result . @xcite
we found that for finite sized systems , there is a non - trivial @xmath1-dependence , not reflected in the construction of fradkin and moore.@xcite the source of this difference laid in the subtle details of defect lines in critical systems . in this work
, we showed that the boundary condition is not described by the same notion of dirichlet boundary conditions as in the original system .
we circumvented the difficulties here by working directly with the original degrees of freedom . by considering the original degrees of freedom
, we further demonstrated that in the limit @xmath43 , @xmath31 has a distribution of eigenvalues characterized by the correct underlying conformal field theory of the ground state wavefunction , confirming , at least for this case , a conjecture put forth by li and haldane.@xcite attempting to understand where the universal corrections to entanglement entropy come from , we related the problem formally to work done on quantum brownian motion and branes with a background electromagnetic field , and showed that @xmath31 can be understood as an expectation value of a vertex operator . in this work we focused on the quantum dimer model , _
i.e. _ models where the norm of the ground state wavefunction is related to the free gaussian field theory , but the methods can be readily extended to different conformal quantum critical models .
it would be interesting to see if a similar structure exists for more complicated systems , such as non trivial topological theories with non - abelian excitations.@xcite for some simple cases , an exact solution is possible.@xcite we thank paul fendley , duncan haldane , vincent pasquier , and jean - mare stphan for discussions .
we especially thank masaki oshikawa for sending us his pre - print and discussing his results with us .
this work was supported in part by the national science foundation grant dmr 0758462 at the university of illinois .
to exploit conformal invariance , it is useful to think of the system as being on a cylinder with circumference @xmath221 and length @xmath15 with boundary conditions @xmath222 on the field on the left and right ends of the system respectively .
quantum mechanically , this corresponds to evolving a one dimensional system defined on the line @xmath171 in time @xmath221 .
the partition function is given by the usual expression , @xmath223 if the hamiltonian possess conformal invariance , then one knows that time and space can be interchanged , @xmath224 or equivalently , the system is invariant under the modular transformation @xmath12 .
one now has a cylinder which is wrapped around in the spatial direction and extending upward in time .
the corresponding hamiltonian in this picture can be regarded as propagating the system for the time interval @xmath15 from the initial and final state @xmath225 @xmath226 the states here belong to the hilbert space of states quantized on the circle , _
i.e. _ they can be decomposed into linear combinations of states in the representation of the virasoro algebra which are labelled by @xmath227 , the highest weights .
@xcite because conformal invariance is so restrictive in two dimensions , one can say more about the boundary states @xmath228 .
one typically imposes the condition that @xmath229 where @xmath230 are the holomorphic and anti - holomorphic components of the stress energy tensor , @xmath231 . in the @xmath232 basis , this means that the diagonal components of the stress energy tensor vanish at the boundary @xmath233 . if the boundary is in the time direction , this means no momentum flows out of the system .
the stress energy generates the conformal symmetry so that the boundary states must satisfy the condition , @xmath234 \vert a \rangle = 0 \label{eqn : cfbc } .\ ] ] fourier transforming , this can be written in terms of virasoro generators , @xmath235 \vert a \rangle = 0 .\ ] ] this implies that the boundary state @xmath236 must made out of states with the holomorphic and anti - holomorphic sectors stitched together in a specific way , _
i.e. _ @xmath237 here , @xmath238 labels the descendant level in the representation @xmath239 that belongs to the subset of representations that appear simultaneously in the holomorphic and anti - holomorphic sectors of the virasoro algebra .
the state @xmath240 are known as the ishibashi states.@xcite the ishibashi states turn out to form a basis for the possible boundary states , and one can write an arbitrary state @xmath236 as a linear combination of the ishibashi states , @xmath241 hence , the characterization of a conformal boundary condition is reduced to finding the matrix elements @xmath242 .
now , using the expression ( [ eqn : bstate ] ) into ( [ eqn : cstring ] ) one finds that @xmath243 the overlap can be identified with the character of the representation @xmath53 , @xmath244 . now noting that ( [ eqn : ostring ] ) can be written as a sum of characters , and using the fact that the two quantities are in fact equivalent by conformal invariance leads one to the relationship @xmath245 where @xmath246 are the multiplicities that indicate the number of times a representation @xmath53 appears in the hilbert space with boundary conditions @xmath6 and @xmath7 .
one then notes that the characters transform among themselves by the modular @xmath12-matrix so that @xmath247 if the characters are linearly independent , then this leads one to cardy s equation which relates the multiplicities @xmath246 that characterize the spectrum of the theory for fixed boundary conditions @xmath222 and the matrix elements @xmath242 that characterize the boundary states , @xmath248 the key problem in boundary conformal field theory is finding a set of boundary states where the multiplicities are non - negative integers.@xcite for the free boson , a solution to this requirement is @xmath249 where @xmath250 and @xmath68 are defined in ( [ eqn : zeromode ] ) . note that linear combinations of boundary states @xmath236 also satisfy the above constraints .
an additional choice that is imposed is that @xmath251 , that is to say that the identity representation appears exactly once in the spectrum of the theory with @xmath252 boundary conditions .
operatively , this fixes the normalization of the boundary states so that in the long cylinder limit @xmath253 contains the identity exactly once .
this gives the @xmath89-factor of the boundary state,@xcite @xmath254 it has been conjectured that relevant boundary perturbations drive the system to fixed points given by lower values of the @xmath89-factor.@xcite in this sense , the @xmath89-factor is also a characteristic of the boundary condition .
the @xmath255-functions are defined as @xmath256 where @xmath73 and @xmath257 the @xmath258 function is then defined as @xmath259 the action of the modular transformation @xmath260 on @xmath261-functions can be found by making use of the possion resummation formula , @xmath262
10 ian affleck . .
in j. jacobsen , s. ouvry , v. pasquier , d. serban , and l.f .
cugliandolo , editors , _ les houches 2008 , session lxxxix : exact methods in low - dimensional statistical physics and quantum computing _ , oxford , uk , 2010 .
oxford university press . | we examine the scaling behavior of the entanglement entropy for the 2d quantum dimer model ( qdm ) at criticality and derive the universal finite sub - leading correction @xmath0 .
we compute the value of @xmath0 without approximation working directly with the wave function of a generalized 2d qdm at the rokhsar - kivelson qcp in the continuum limit . using the replica approach
, we construct the conformal boundary state corresponding to the cyclic identification of @xmath1-copies along the boundary of the observed region .
we find that the universal finite term is @xmath2 where @xmath3 is the compactification radius of the bose field theory quantum lifshitz model , the effective field theory of the 2d qdm at quantum criticality .
we also demonstrated that the entanglement spectrum of the critical wave function on a large but finite region is described by the characters of the underlying conformal field theory .
it is shown that this is formally related to the problems of quantum brownian motion on @xmath1-dimensional lattices or equivalently a system of strings interacting with a brane containing a background electromagnetic field and can be written as an expectation value of a vertex operator . |
the theory of hyperplane arrangements has deep connections with many areas of mathematics , such as combinatorics , representation theory , algebraic geometry , algebraic topology , singularity theory , and the theory of hypergeometric functions .
one beautiful application of the use of hyperplane arrangements in algebraic geometry is the proof of brieskorn s conjecture by deligne @xcite in 1972 .
real hyperplane arrangements serve as the motivation for the axioms that define oriented matroids .
see @xcite for detailed reviews of the theory of hyperplane arrangements .
real hyperplane arrangements have been used recently in the physics literature to analyze the fiber structure and the network of flop transitions between different resolutions of elliptically fibered calabi - yau threefolds @xcite .
compactifications of m - theory on calabi - yau threefolds form an elegant bridge between the study of supersymmetric gauge theories in five dimensional spacetime and certain types of highly structured hyperplane arrangements defined by a lie algebra @xmath2 and a representation @xmath3 of @xmath2 .
these hyperplane arrangements have the peculiarity of being defined not in the full affine space as they are restricted to the dual fundamental weyl chamber of the lie algebra @xmath2 .
[ def : incgeogeneral ] let @xmath2 be a reductive lie algebra over @xmath4 , let @xmath5 be a split , real form of a cartan subalgebra of @xmath6 and let @xmath7 be a representation of @xmath2 .
we denote by @xmath8 the real hyperplane arrangement consisting of the kernels of the weights of @xmath7 restricted to a dual fundamental weyl chamber in @xmath5 . because all cartan subalgebras are conjugate and all fundamental weyl chambers canonically related by the weyl group action , the incidence geometry @xmath8 is independent of the choice of @xmath5 and of a dual fundamental weyl chamber in @xmath5 .
in the rest of the paper , we call the dual fundamental weyl chamber " the _ weyl chamber_. an arrangement of hyperplanes stratifies the ambient space into open polyhedral cones called _ faces_. a non - empty intersection of a finite number of hyperplanes of the arrangement is called a _ flat_. a @xmath9-face ( resp .
@xmath9-flat ) is a face ( resp .
a flat ) that generates a @xmath9-dimensional linear space .
faces of maximum dimension are called _
a classical combinatorial problem asks how many @xmath9-flats and @xmath9-faces a given hyperplane arrangement contains .
zaslasky @xcite enumerated chambers and bounded chambers of an affine hyperplane arrangement using the mbius function of its semi - lattice of flats .
the purpose of this paper is to count the numbers of @xmath9-faces and @xmath9-flats of hyperplane arrangements @xmath10 where @xmath11 is the vector representation of @xmath0 ( the first fundamental representation ) and @xmath12 is its second exterior power ( the second fundamental representation ) .
since our arrangements are restricted to the weyl chamber , we can not apply zaslasky s result directly .
however , the highly symmetric nature of the hyperplane arrangements under consideration allows us to study them via their extreme rays .
we also analyze the geometries @xmath13 when @xmath2 is not simply laced .
the case of @xmath14 is discussed in @xcite .
the counting problem solved in this paper is formulated purely in terms of representation theory .
however , its motivation comes from the geometry of string theory . in this subsection , we review the motivation .
the reader uninterested in the interface of string theory and algebraic geometry can safely ignore this subsection .
hyperplane arrangements of the form @xmath15 arise naturally in the study _ coulomb branch _ of a gauge theory in five - dimensional supersymmetric theory with eight supercharges @xcite .
such theories contain different types of particles organized into representations of the supersymmetric algebra called _
supermultiplets_. we will consider only vector multiplets and hypermultiplets . when a gauge theory is defined in a five dimensional supersymmetric theory ,
the vector multiplets transform in the adjoint representation while the hypermultiplets transform in some representation @xmath3 of @xmath2 .
the hypermultiplets can become massive through the higgs mechanism along hyperplanes passing through the origin of the weyl chamber of @xmath2 .
these hyperplanes partition the weyl chamber into different connected regions called _ coulomb phases _ of the gauge theory , which are characterized by a cubic prepotential depending on real coordinates @xmath16 of the weight space .
intriligator - morrision - seiberg computed the prepotential of a five dimensional gauge theory for a theory in which hypermultiplets transform in the representation @xmath17 of a lie algebra @xmath2 .
the result is the real function @xcite : @xmath18 depending on the real vector @xmath19 in the weight space .
we only wrote the terms of @xmath20 that are relevant for our purposes .
the variable @xmath21 runs through the simple roots of @xmath2 , and @xmath22 through the weights of the representation @xmath7 under which the hypermultiplets transform .
the cubic absolute values generate singularities , which correspond to hyperplanes along which new massless particles are generated .
a well - defined phase of the coulomb branch of the theory is a connected region in which the quantities @xmath23 and @xmath24 each take fixed signs .
we fix the sign of @xmath23 by requiring @xmath19 to be in the fundamental weyl chamber .
the condition that @xmath25 for all @xmath26 restricts us inside a specific chamber of the arrangement @xmath27 .
the counting of flats is connected to the enumeration of mixed _ coulomb - higgs branches _ of the gauge theory .
a five dimensional gauge theory with 8 supercharges can be geometrically engineered by a compactification of m - theory on a calabi - yau threefold @xcite .
when the calabi - yau threefold is elliptically fibered , its singular fibers determine the lie algebra @xmath2 and the matter representation @xmath3 of the gauge theory .
a smooth elliptic fibration with a rational section is a resolution of a singular weierstrass model , and flop transitions join pairs of crepant resolutions of the weierstrass model .
it is conjectured that the network of ( partial ) crepant resolutions of the weierstrass model is isomorphic to the adjacency graph of the faces of the hyperplane arrangement @xmath15 .
more precisely , it is conjectured that each crepant resolution corresponds to a unique chamber of @xmath15 @xcite , and that two crepant resolutions are connected by a flop if and only if the intersection of the corresponding chambers is a face of codimension one .
their common face is then conjectured to correspond to a unique partial resolution of the weierstrass model .
see @xcite for examples of explicit matchings between ( partial ) resolutions and faces of a hyperplane arrangements @xmath15 .
there are subtleties in presence of non - trivial @xmath28-factorial terminal singularities , see for example @xcite .
a proof of this conjecture will likely rely heavily on ideas from mori s program along the lines of @xcite .
the intriligator - morrision - seiberg prepotential @xmath29 can also be obtained geometrically @xcite . in the case of @xmath30 ,
the gauge group is @xmath1 , and therefore the relevant elliptic fibrations each have a mordell - well group of rank one and a singular fiber of type i@xmath31 over a divisor of the discriminant locus .
models for elliptic fibrations with a mordell - weil group of rank one are discussed in @xcite and @xcite .
elliptic fibrations for geometric engineering of @xmath1 gauge theories have been extensively studied recently in the physics literature ( see for example @xcite and references therein ) .
we are now ready to discuss the enumerations performed in this paper .
we denote the number of @xmath9-faces of @xmath32 by @xmath33 and collect these numbers in the following formal power series ( generating function ) : @xmath34 the following theorem enumerates faces of @xmath30 .
it is proved in sec : genfuncfaces .
[ thm : facecount ] the generating function counting faces of @xmath35 by dimension is given by the following rational function : @xmath36 thm : facecount implies a simple recurrence relation for @xmath33 .
[ cor : facerecurse ] the sequence @xmath33 obeys the recurrence relation : @xmath37 for @xmath38 . we can also obtain formulas for the generating polynomials @xmath39 g(s , t ) = \sum_{k=0}^n g(n , k)t^k.\ ] ] [ cor : facegenpoly ] the generating polynomials @xmath40 are given by @xmath41 cor : facegenpoly yields the following values of @xmath40 : .3 cm from the above formulae , we can read off the value of @xmath33 for @xmath42 as the coefficient of @xmath43 in @xmath40 : @xmath44 we can also extract formulae for @xmath33 for @xmath9 small and @xmath45 small from thm : facecount .
[ cor : facesmalllarge ] we have @xmath46 and @xmath47 cor : facesmalllarge yields the following formulae for @xmath33 and @xmath48 for @xmath49 : @xmath50 thm : facecount and corollaries [ cor : facerecurse ] , [ cor : facegenpoly ] , and [ cor : facesmalllarge ] are proved in sec : genfuncfaces .
we denote the number of @xmath9-flats of @xmath32 by @xmath51 and collect these numbers in the following formal power series ( generating function ) : @xmath52 the following theorem enumerates flats of @xmath30 .
[ thm : flatcount ] the generating function counting flats of @xmath35 by dimension is given by the following rational function : @xmath53 thm : flatcount is proven sec : genfuncflats .
it implies a simple recurrence relation for @xmath51 , the analogue of cor : facerecurse for flats .
[ cor : flatrecurse ] the sequence @xmath51 obeys the recurrence relation @xmath54 for @xmath55 . the taylor expansion of thm : flatcount to order 10 in @xmath56 is : from the above formula , we can read off the value of @xmath51 for @xmath42 as the coefficient of @xmath57 in @xmath58 : @xmath59 the form of the series @xmath58 does not allow simple explicit formulae for @xmath60 or for @xmath51 as in corollaries [ cor : facegenpoly ] and [ cor : facesmalllarge ] , respectively .
thm : flatcount is proved in sec : genfuncflats .
( 1,1 ) & ( 1,2 ) & ( 1,3 ) & & ( 1,n ) + & ( 2,2 ) & ( 2,3 ) & & ( 2,n ) + & & ( 3,3 ) & & ( 3,n ) + & & & & + & & & & ( n , n ) a face is determined by assigning signs to @xmath61 for all values of @xmath62 such that @xmath63 .
these weights can be organized in a right - justified young tableau as in figure [ fig : signtableau ] .
it is important to realize that not all sign patterns are allowed .
since we are in the weyl chamber , we have @xmath64 using the identities @xmath65 one can prove the following simple sign rules called _ sign flows _ in the physics literature @xcite : the signs in a tableau corresponding to a chamber satisfy the following conditions .
* all the boxes above or on the left of a box with positive entries are also positive .
* all the boxes below or on the right of a box with negative entries are also negative .
a tableau satisfying these two rules corresponds to a unique chamber of @xmath66 .
however , a different notation for chambers will be more convenient for dealing with extreme rays
. a different notation for chambers will be more convenient for dealing with extreme rays .
the sign rules implies that the positive entries of each row are next to each other and start on the left border of the table .
it is therefore efficient to denote a given chamber by the numbers of positive entries on each row .
the sign rules is then automatically satisfied if these numbers form a decreasing sequence @xmath67 , where we denote by @xmath68 the number of positive entries on the @xmath69th row of the tableau .
we do nt count the rows that do not have any positive entries .
an entry of the tableau located on the @xmath69th row and the @xmath70th column ( with @xmath71 ) is positive if and only if @xmath72 and @xmath73 .
it is negative otherwise .
this justifies the following definition .
[ def : subsettochamber ] for @xmath74 $ ] , define a face @xmath75 of @xmath35 as the subset of @xmath5 on which @xmath76 we can equivalently write a subset @xmath77 $ ] as a _
characteristic vector _
@xmath78 , where @xmath79 and @xmath80 .
a characteristic vector @xmath79 defines a subset @xmath81 $ ] such that @xmath82 ( resp .
@xmath83 ) if and only if @xmath84 ( resp .
@xmath85 ) .
the notation of def : subsettochamber allows us to state the following classification of chambers and extreme rays . [ thm : chamberstructure ] the chambers and extreme rays of @xmath30 satisfy the following properties . a. [ assert : chamberbiject ] the map @xmath86 defines a bijection from @xmath87}$ ] to the set of chambers of @xmath35 . in particular
, @xmath35 has @xmath88 chambers .
b. [ assert : chamberraysprelim ] the extreme rays of @xmath75 are generated by the vectors @xmath89 where @xmath90 where @xmath91 where @xmath91 counts the elements of @xmath92 that are greater or equal to @xmath93 , and the vectors @xmath94 non - negatively span @xmath75 .
in particular , the geometry @xmath30 is simplicial .
the enumeration of chambers of @xmath30 announced in thm : chamberstructureassert : chamberbiject recovers the count of phases of @xmath1 with fundamental and antisymmetric matter from the physics literature @xcite .
theorem [ thm : chamberstructure ] is proven in sec : extremerayschambers .
suppose that @xmath95 with @xmath96 .
a direct calculation shows that the box diagram associated to @xmath75 has @xmath97 plusses in the @xmath69th row for @xmath98 and all minus signs in the @xmath69th row for @xmath99 .
see tables [ tab : example.gl2 ] and [ tab : example.gl3 ] for examples .
c | l|c| l sign tableau & subset of @xmath100 $ ] & characteristic vector & interior points + ' '' '' & @xmath101 & @xmath102 & @xmath103 + ' '' '' & @xmath104 & @xmath105 & @xmath106 + ' '' '' & @xmath107 & @xmath108 & @xmath109 + ' '' '' & @xmath110 & @xmath111 & @xmath112 l | c| c | l sign tableau & subset of @xmath100 $ ] & characteristic vector & interior points of the chamber + ' '' '' - & - & - + & - & - + & & - & @xmath110 & @xmath113 & @xmath114 + ' '' '' + & - & - + & - & - + & & - & @xmath107 & @xmath115 & @xmath116 + ' '' '' + & + & - + & - & - + & & - & @xmath104 & @xmath117 & @xmath118 + ' '' ''
+ & + & + + & - & - + & & - & @xmath119 & @xmath120 & @xmath121 + ' '' '' + & + & + + & + & - + & & - & @xmath122 & @xmath123 & @xmath124 + ' '' '' + & + & - + & + & - + & & - & @xmath101 & @xmath125 & @xmath126 + ' '' '' + & + & + + & + & + + & & - & @xmath127 & @xmath128 & @xmath129 + ' '' '' + & + & + + & + & + + & & + & @xmath130 & @xmath131 & @xmath132 in light of thm : chamberstructure , we can introduce additional structure on the set of extreme rays of @xmath30 .
we will equip the set of extreme rays with a partial order and use the combinatorics of the resulting partially ordered set ( poset ) to study faces and flats in @xmath30 .
the following definitions describe the posets that will be relevant to us .
[ def : quarterplaneposet ] the _ discrete quarter plane poset _ is the set @xmath133 endowed with the cartesian order induced by the usual order of the set @xmath134 of non - negative integers : @xmath135 this is a graded poset with grading function @xmath136 .
we call @xmath137 the _ level _ of @xmath138 .
[ def : posetforextrays ] we denote by @xmath139 the subset of the discrete quarter plane that consists of points at level less or equal to @xmath140 .
@xmath139 is a poset with the order induced by the cartesian order defined above .
we denote by @xmath141 the poset @xmath142 where @xmath143 is greater than all the elements of @xmath139 .
we denote by @xmath144 the poset @xmath139 with the origin removed : @xmath145 we are now ready to relate @xmath144 to the set of extreme rays of @xmath30 .
[ def : vfn ] define a function @xmath146 by @xmath147 the following theorem follows directly from thm : chamberstructure .
[ thm : raystruture ] the extreme rays of @xmath35 satisfy the following properties .
a. the function @xmath148 defines a bijection from @xmath144 to the set of extreme rays of @xmath35 .
b. the set of extreme rays that lie in a chamber @xmath75 are the extreme rays @xmath149 where @xmath150 .
we will characterize faces and flats by the extreme rays they contain .
for that reason we define a function which returns the extreme rays lying in a given subset of the weyl chamber .
[ def : raysoperator ] define a function @xmath151 from @xmath152 to the power set of the set of extreme rays of @xmath30 as follows .
for @xmath153 let @xmath154 be the set of extreme rays of @xmath30 that lie in @xmath155 . the following theorem , proven in sec : extremeraysfaces relates faces in @xmath30 to the combinatorics of @xmath144 .
[ thm : characterization.faces ] for all @xmath9 , the function @xmath156 induces a bijection from the set of @xmath9-faces of @xmath30 to the set of @xmath9-chains in @xmath144 . here
, we say that @xmath157 is a @xmath9-chain if @xmath92 is a chain and @xmath158 . to state an analogue of thm : characterization.faces for flats , we will need to define structures called _ ensembles _ that will play the role of chains .
[ def : ensemble ] an _ ensemble _ is the restriction to @xmath144 of a union @xmath159 \right)\cap { \mathbb{e}^\ast}_n\ ] ] of intervals @xmath160 $ ] of @xmath141 satisfying the following four conditions : 1 .
@xmath161 2 .
@xmath162 for @xmath163 ; 3 .
@xmath164 for @xmath165 ; and 4 .
@xmath166 or @xmath167 .
we say that @xmath168 is a _
@xmath9-ensemble _ if @xmath169 , so that @xmath9 counts the number of distinct levels of elements of @xmath168 . figure [ fig : ensembleoneeg ] .
shows one example of an ensemble .
2 -ensemble in @xmath170 .
* figure ( a ) represents the @xmath171-ensemble @xmath92 which is the restriction to @xmath172 of the union of intervals @xmath173 \cup [ ( 3,3),(4,4 ) ] \cup [ ( 5,5),\infty]$ ] .
theorem [ thm : characterization.flats ] guarantees that @xmath174 is the set of extreme rays of a unique 10-flat in @xmath175 .
figure ( b ) illustrates a maximal chain of @xmath92 , which corresponds to a @xmath171-face of @xmath175 . ]
-ensemble in @xmath170 .
* figure ( a ) represents the @xmath171-ensemble @xmath92 which is the restriction to @xmath172 of the union of intervals @xmath173 \cup [ ( 3,3),(4,4 ) ] \cup [ ( 5,5),\infty]$ ] .
theorem [ thm : characterization.flats ] guarantees that @xmath174 is the set of extreme rays of a unique 10-flat in @xmath175 .
figure ( b ) illustrates a maximal chain of @xmath92 , which corresponds to a @xmath171-face of @xmath175 . ]
[ thm : characterization.flats ] for all @xmath9 , the function @xmath156 induces a bijection from the set of @xmath9-flats of @xmath30 to the set of @xmath9-ensembles of @xmath144 . theorem [ thm : characterization.flats ] is proven in section [ sec : extremeraysflats ] on page .
denote by @xmath11 the vector representation of @xmath176 and of @xmath177 . recall that the representation @xmath178 of @xmath177 factors as @xmath179 , where @xmath180 is irreducible .
let @xmath181 ( resp .
@xmath182 ) denote the unique 26-dimensional ( resp .
7-dimensional ) irreducible representation of @xmath183 ( resp .
@xmath184 ) .
we are now ready to discuss the geometry @xmath185 when @xmath2 is non - simply laced simple lie algebra .
[ thm : nonsimplylaced ] suppose that @xmath186 and suppose that @xmath2 has rank @xmath140 .
the geometry @xmath185 is a cone over a simplex @xmath187 .
in particular , there is 1 chamber and the number of @xmath9-faces ( resp .
@xmath9-flats ) is @xmath188 .
there is only one other geometry @xmath13 when @xmath2 is simple and non - simply laced and @xmath7 is a direct sum of minuscule and quasi - minuscule representation .
this is the geometry @xmath189 it is mentioned in ( * ? ? ?
* section 6.3 ) that @xmath190 and @xmath191 have no chambers .
this is in contradiction with the conclusion of thm : nonsimplylaced .
a direct calculation shows that all the weights of @xmath7 are scalar multiples of the roots of @xmath2 in the cases discussed in the theorem .
therefore , the hyperplanes @xmath192 for @xmath22 a weight of @xmath7 coincide with vanishing loci of roots of @xmath2 .
the theorem then follows from the fact that the weyl chamber is simplicial . .
* the geometry contains 4 chambers , 5 half - lines and the origin .
there is one 2-flat ( the half - plane below the line @xmath193 ) , three 1-flats ( the interior walls @xmath194 , @xmath195 , @xmath196 ) , and one 0-flat ( the origin ) . ] in sec : prelims , we recall basic facts regarding the lie algebra @xmath0 and from the theory of hyperplane arrangements . in sec : extrayschmfaces , we prove theorems [ thm : chamberstructure ] , [ thm : raystruture ] , and [ thm : characterization.faces ] . in sec : facecount , we prove thm : facecount , and in sec : flatcount , we prove theorems [ thm : flatcount ] and [ thm : characterization.flats ] .
m.e . is grateful to shu - heng shao and shing - tung yau for helpful discussions .
m.e . and a.n .
thank william massey and all the organizers of the 20th conference for african american researchers in mathematical sciences ( caarms 20 ) where this started .
is supported in part by the national science foundation ( nsf ) grant dms-1406925
elliptic fibrations and string theory " .
is supported by the harvard college research program .
in sec : prelimliealg , we fix notation for the weyl chamber of @xmath0 . in sec : hyparrange , we recall some facts from the theory of hyperplane arrangements . a detailed list of symbols is presented in tab : notations .
let @xmath197 denote the cartan subalgebra of @xmath0 consisting of the diagonal matrices .
we identify @xmath5 with @xmath198 via the isomorphism @xmath199 the positive roots of @xmath0 are @xmath200 .
we define the open weyl chamber of @xmath0 as the set of vectors @xmath201 forming a non - increasing sequence @xmath202 and denote by @xmath203 the closure of @xmath204 .
define @xmath205 .
the weights of the vector representation are @xmath206 for @xmath207 and the weights of the antisymmetric representation are @xmath61 for @xmath208 . when @xmath209 , the hyperplane @xmath210 is the vanishing locus of the weight @xmath206 of the vector representation of @xmath0 . when @xmath211 , the hyperplane @xmath212 is the vanishing locus of the weight @xmath61 of the antisymmetric representation of @xmath0 . by definition ,
each chamber is a convex rational polyhedral cone ( finite intersection of rational half spaces ) . in the following
we will refer to a convex rational polyhedral cone simply as `` cone '' .
face _ is either a chamber or the intersection of a chamber with a supporting hyperplane .
there are finitely many faces , and each face is a lower dimensional cone .
the chambers cover @xmath155 and form a _
fan_. a fan is a finite collection of cones such that for each cone all its faces are in the fan and the intersection of two cones is a face of each @xcite . by classical theorems of weyl and minkowski , a cone
is finitely generated if and only if it is a finite intersection of closed linear half spaces ( see ( * ? ? ?
* theorem 1.3 ) ) .
it follows that each chamber of @xmath32 admits a system of generators .
a cone of dimension @xmath140 is said to be _ simplical _ if it is non - negatively spanned by @xmath140 linearly independent vectors .
a face of dimension one which is a half - line is called an _
extreme ray_. a _ facet _ is a face of co - dimension one .
the roots of the lie algebra @xmath0 define a hyperplane arrangement known as the _ braid arrangement _
@xcite , which has @xmath213 chambers , namely the weyl chambers of @xmath214 .
the braid arrangement is also called the coxeter arrangement of type @xmath214 .
the weights of the fundamental representation define an arrangement of hyperplanes known as the _ boolean arrangement _ @xcite .
the weights of the antisymmetric representation @xmath215 define an arrangement of hyperplanes known as the _ threshold arrangement _ @xcite .
altogether , the roots of @xmath0 and the weights of the first two fundamental representations define the coxeter arrangement of type @xmath216 @xcite .
in sec : extremerayschambers , we prove theorems [ thm : chamberstructure ] and [ thm : raystruture ] , and in sec : extremeraysfaces , we prove thm : characterization.faces . given a vector @xmath217 , we denote by @xmath218 ( resp .
@xmath219 ) the vector obtained from @xmath220 by rearranging its components in non - increasing ( resp .
non - decreasing ) order .
the idea of the proof is to derive for each @xmath78 an explicit linear isomorphism @xmath221 from @xmath198 to @xmath5 such @xmath222 .
the elementary basis vectors of @xmath198 then map under @xmath221 to the set of extreme rays of @xmath223 .
define @xmath224 the absolute values of the coordinates @xmath225 of a point @xmath226 are pairwise distinct and non - zero , because @xmath227 and @xmath228 for all @xmath208 .
after reordering , we can write the set @xmath229 as a strictly increasing vector @xmath230 so that @xmath231 for all @xmath69 because @xmath232 there exists a unique sequence @xmath233 of signs such that @xmath234 for all @xmath235 .
we see immediately that @xmath236 defines a continuous function from @xmath237 to @xmath238 .
a direct calculation shows that @xmath239 and that @xmath240 .
we now prove part assert : chamberbiject of the theorem .
suppose that @xmath241 .
consider the vector @xmath242 note that the absolute values of the components of @xmath243 are @xmath244 .
we see immediately that @xmath245 for all @xmath78 , so that @xmath246 . because @xmath247 and @xmath246 , the face @xmath223 contains a point of @xmath204 not lying in any hyperplane of @xmath30 .
it follows that @xmath223 is a chamber of @xmath30 for all @xmath248 .
it remains to prove that @xmath249 defines a bijection from @xmath238 to the set of chambers of @xmath30 .
consider a chamber @xmath250 of @xmath30 and let @xmath251 be an interior point .
note that @xmath240 .
we see immediately that @xmath252 because @xmath253 is a chamber of @xmath30 .
suppose that @xmath254 .
it follows that @xmath243 and @xmath255 lie in the same connected component of @xmath237 . because @xmath236 is continuous
, we obtain that @xmath256 so that @xmath249 is injective .
part assert : chamberbiject follows .
we now prove part assert : chamberraysprelim .
consider a chamber @xmath223 of @xmath30 .
let @xmath257 denote the subset of @xmath100 $ ] corresponding to @xmath258 .
note that @xmath259 with @xmath260 .
hence , @xmath223 is non - negatively spanned by @xmath261 .
it follows that @xmath94 are the extreme rays of @xmath223 , as desired .
recall the definition of a chain in a poset .
a _ * chain * _ in a poset is a subset of the poset in which any two elements are comparable .
the _ length _ of a chain is its size , and a _
@xmath9-chain _ is a chain of length @xmath9 .
it follows from thm : chamberstructure that @xmath262 induces a bijection from the set of chambers of @xmath30 to the set of @xmath140-chains in @xmath144 .
to prove thm : characterization.faces , we will exploit the simpliciality of the chambers of @xmath30 , which was also proved in thm : chamberstructure .
suppose that @xmath250 is a @xmath9-face of @xmath30 .
consider a chamber @xmath249 containing @xmath250 . because @xmath249 is simplicial ( by thm : chamberstructure )
, @xmath250 must be the non - negative span of @xmath9 extreme rays of @xmath249 .
in particular , we obtain that @xmath263 is a subset of @xmath264 of size @xmath9 . because @xmath265 is an @xmath140-chain in @xmath144 and @xmath266
is a subset of @xmath265 of size @xmath9 , it follows that @xmath266 is a @xmath9-chain in @xmath144 .
therefore , @xmath262 is a function from the set of @xmath9-faces of @xmath30 to the set of @xmath9-chains of @xmath144 .
the fact that every face is non - negatively spanned by the extreme rays that it contains ensures that @xmath262 induces an injection from the set of @xmath9-faces of @xmath30 to the set of @xmath9-chains in @xmath144 .
it remains to prove that @xmath262 is surjective .
suppose that @xmath267 is a @xmath9-chain .
let @xmath250 denote the non - negative span of @xmath268 .
consider an @xmath140-chain @xmath269 containing @xmath270 which clearly exists , and the corresponding chamber @xmath249 ( which satisfies the property that @xmath271 ) .
it follows from the simpliciality of @xmath249 ( thm : chamberstructure ) that @xmath250 is a sub - face of @xmath249 of dimension @xmath9 and that @xmath272 .
therefore , every @xmath9-chain in @xmath144 can be expressed in the form @xmath273 for some @xmath9-face @xmath250 of @xmath30 .
the theorem follows .
in sec : prelimonposet , we describe some basic properties of the posets @xmath139 and @xmath144 . in sec : genfuncfaces , we prove thm : facecount . given two points @xmath274 and @xmath275 in @xmath144 , such that @xmath276 , consider the rectangle with vertical and horizontal edges whose ne - sw diagonal is the segment @xmath277 .
the lattice points of this rectangle are exactly the points of the interval @xmath278 $ ] in @xmath144 .
see figures [ fig : posetsn ] on page for examples of intervals in @xmath144 .
the discrete quarter - plane @xmath133 is equipped with a grading given by the level function .
the origin is the only point of level zero .
denote by @xmath279 the set of points of level @xmath69 of @xmath133 .
the level sets @xmath279 are all anti - chains and form a partition of @xmath133 .
see fig : levelset for an example .
note that @xmath280 for all @xmath281 , the interval @xmath282 in @xmath139 ( or in @xmath144 ) is isomorphic to @xmath144 as a graded poset .
the isomorphism is induced by the translation by @xmath274 in @xmath133 : the map @xmath283 defines an isomorphism from @xmath284 to @xmath282 .
3 and @xmath144 .
* ] and @xmath144 . * ] and @xmath144 . * ] in order to compute the generating function @xmath285 we first derive a recurrence relation for @xmath33 .
[ prop : facerecursion ] the integers @xmath33 satisfy the recurrence relation @xmath286 for @xmath287 , where @xmath288 for @xmath289 and for @xmath290 $ ] .
we prove prop : facerecursion by tracking the minimum of each @xmath9-chain and interpreting the @xmath291 largest elements of the chain as @xmath292-chain in @xmath293 with @xmath294 .
the proposition is clear for @xmath295 .
assume for the remainder of the proof that @xmath296 .
denote by @xmath297 the set of @xmath9-chains in @xmath144 with minimum @xmath274 .
suppose that @xmath298 .
recall that translation by @xmath299 in @xmath133 defines an isomorphism from @xmath282 to @xmath300 .
we see immediately that the process of removing @xmath274 from a @xmath9-chain and translating by @xmath299 ( i.e. , the function defined by @xmath301 ) defines a bijection from @xmath297 to the set of @xmath292-chains in @xmath302 .
it follows that @xmath303 .
note in particular that @xmath304 for @xmath305 .
recall that @xmath144 contains @xmath306 elements of level @xmath307 for @xmath308 .
grouping by level , we obtain that @xmath309 as desired .
we need to introduce the correction term @xmath310 in order to deal with the case of @xmath295 .
after some algebra , we can derive the expression for @xmath311 announced in thm : facecount .
translating prop : facerecursion in to the language generating functions yields the linear equation @xmath312 where we used the identity @xmath313 the theorem follows by solving the linear equation for @xmath311 . corollaries [ cor : facerecurse ] , [ cor : facegenpoly ] and [ cor : facesmalllarge ] follow from thm : facecount by elementary algebraic manipulations . in particular , cor : facerecurse can be derived by noting that the coefficient of @xmath57 in @xmath314 vanishes for @xmath315 due to thm : facecount .
cor : facegenpoly can be derived by computing a partial fraction decomposition for the expression for @xmath311 given in thm : facecount . on the other hand , cor :
facesmalllarge is proved by regarding @xmath311 as the sum of geometric series with ratio @xmath316 .
in sec : extremeraysflats , we prove thm : characterization.flats , and in sec : genfuncflats , we prove thm : flatcount .
thm : characterization.flats will be derived from the following two lemmas .
the first asserts that @xmath317 is a flat whenever @xmath318 is an ensemble , and the second asserts that every ensemble is of the form @xmath317 for some flat @xmath318 .
to prove lem : flatyieldsensemble , we exploit the the fact that the set of extreme rays of a flat is the intersection of the sets of extreme rays of the hyperplanes containing it .
recall that the set of extreme rays of @xmath319 @xmath320 corresponds in the poset @xmath144 to the union of intervals @xmath321\cup [ \varphi(i , j)+(1,1),\infty],\ ] ] where @xmath322 notice that @xmath323 with equality if and only if @xmath324 when @xmath209 , the second interval is empty and @xmath325 is a weight of the fundamental representation @xmath11 .
the first interval is empty when @xmath326 and completely contained in the @xmath327-axis ( reps .
@xmath220-axis ) when @xmath328 ( resp .
@xmath329 ) .
pedagogical examples are presented in fig : flats.hyperplane . * .
the set of extreme rays of a hyperplane @xmath330 ( @xmath71 ) corresponding to a union of intervals @xmath331\cup [ \varphi(i , j)+(1,1),\infty]$ ] in the poset @xmath139 with @xmath332 . for extreme values ( @xmath328 , @xmath333 , @xmath329 , @xmath334 , @xmath335 ) one of the two intervals
can be contained in one of the axis , shrink to a point , or disappear altogether .
the points @xmath336 and @xmath337 are the only points of @xmath144 not correspond to any extreme ray of any hyperplane @xmath330 .
[ fig : flats.hyperplane ] ] * .
the set of extreme rays of a hyperplane @xmath330 ( @xmath71 ) corresponding to a union of intervals @xmath331\cup [ \varphi(i , j)+(1,1),\infty]$ ] in the poset @xmath139 with @xmath332 . for extreme values ( @xmath328 , @xmath333 , @xmath329 , @xmath334 , @xmath335 ) one of the two intervals
can be contained in one of the axis , shrink to a point , or disappear altogether .
the points @xmath336 and @xmath337 are the only points of @xmath144 not correspond to any extreme ray of any hyperplane @xmath330 .
[ fig : flats.hyperplane ] ] * .
the set of extreme rays of a hyperplane @xmath330 ( @xmath71 ) corresponding to a union of intervals @xmath331\cup [ \varphi(i , j)+(1,1),\infty]$ ] in the poset @xmath139 with @xmath332 . for extreme values ( @xmath328 , @xmath333 , @xmath329 , @xmath334 , @xmath335 ) one of the two intervals
can be contained in one of the axis , shrink to a point , or disappear altogether .
the points @xmath336 and @xmath337 are the only points of @xmath144 not correspond to any extreme ray of any hyperplane @xmath330 .
[ fig : flats.hyperplane ] ] * .
the set of extreme rays of a hyperplane @xmath330 ( @xmath71 ) corresponding to a union of intervals @xmath331\cup [ \varphi(i , j)+(1,1),\infty]$ ] in the poset @xmath139 with @xmath332 . for extreme values ( @xmath328 , @xmath333 , @xmath329 , @xmath334 , @xmath335 ) one of the two intervals
can be contained in one of the axis , shrink to a point , or disappear altogether .
the points @xmath336 and @xmath337 are the only points of @xmath144 not correspond to any extreme ray of any hyperplane @xmath330 .
[ fig : flats.hyperplane ] ] * .
the set of extreme rays of a hyperplane @xmath330 ( @xmath71 ) corresponding to a union of intervals @xmath331\cup [ \varphi(i , j)+(1,1),\infty]$ ] in the poset @xmath139 with @xmath332 . for extreme values ( @xmath328 , @xmath333 , @xmath329 , @xmath334 , @xmath335 ) one of the two intervals
can be contained in one of the axis , shrink to a point , or disappear altogether .
the points @xmath336 and @xmath337 are the only points of @xmath144 not correspond to any extreme ray of any hyperplane @xmath330 .
[ fig : flats.hyperplane ] ] * .
the set of extreme rays of a hyperplane @xmath330 ( @xmath71 ) corresponding to a union of intervals @xmath331\cup [ \varphi(i , j)+(1,1),\infty]$ ] in the poset @xmath139 with @xmath332 . for extreme values
( @xmath328 , @xmath333 , @xmath329 , @xmath334 , @xmath335 ) one of the two intervals can be contained in one of the axis , shrink to a point , or disappear altogether .
the points @xmath336 and @xmath337 are the only points of @xmath144 not correspond to any extreme ray of any hyperplane @xmath330 .
[ fig : flats.hyperplane ] ] * .
the set of extreme rays of a hyperplane @xmath330 ( @xmath71 ) corresponding to a union of intervals @xmath331\cup [ \varphi(i , j)+(1,1),\infty]$ ] in the poset @xmath139 with @xmath332 .
for extreme values ( @xmath328 , @xmath333 , @xmath329 , @xmath334 , @xmath335 ) one of the two intervals can be contained in one of the axis , shrink to a point , or disappear altogether .
the points @xmath336 and @xmath337 are the only points of @xmath144 not correspond to any extreme ray of any hyperplane @xmath330 .
[ fig : flats.hyperplane ] ] * .
the set of extreme rays of a hyperplane @xmath330 ( @xmath71 ) corresponding to a union of intervals @xmath331\cup [ \varphi(i , j)+(1,1),\infty]$ ] in the poset @xmath139 with @xmath332 . for extreme values ( @xmath328 , @xmath333 , @xmath329 , @xmath334 , @xmath335 ) one of the two intervals
can be contained in one of the axis , shrink to a point , or disappear altogether .
the points @xmath336 and @xmath337 are the only points of @xmath144 not correspond to any extreme ray of any hyperplane @xmath330 . [
fig : flats.hyperplane ] ] * .
the set of extreme rays of a hyperplane @xmath330 ( @xmath71 ) corresponding to a union of intervals @xmath331\cup [ \varphi(i , j)+(1,1),\infty]$ ] in the poset @xmath139 with @xmath332 . for extreme values ( @xmath328 , @xmath333 , @xmath329 , @xmath334 , @xmath335 ) one of the two intervals
can be contained in one of the axis , shrink to a point , or disappear altogether .
the points @xmath336 and @xmath337 are the only points of @xmath144 not correspond to any extreme ray of any hyperplane @xmath330 .
[ fig : flats.hyperplane ] ] * .
the set of extreme rays of a hyperplane @xmath330 ( @xmath71 ) corresponding to a union of intervals @xmath331\cup [ \varphi(i , j)+(1,1),\infty]$ ] in the poset @xmath139 with @xmath332 . for extreme values
( @xmath328 , @xmath333 , @xmath329 , @xmath334 , @xmath335 ) one of the two intervals can be contained in one of the axis , shrink to a point , or disappear altogether .
the points @xmath336 and @xmath337 are the only points of @xmath144 not correspond to any extreme ray of any hyperplane @xmath330 .
[ fig : flats.hyperplane ] ] consider a @xmath9-flat @xmath318 and denote by @xmath168 the set of extreme rays of @xmath318 .
we will prove that @xmath168 is an ensemble .
suppose that @xmath338 note that @xmath339 using the structure of extreme rays of hyperplanes @xmath319 and the identity @xmath340\cap[c , d]=[a\vee c , b\wedge d],\ ] ] we see immediately that @xmath341 can be minimally expressed in the form @xmath342
\cup [ a_m , b_m].\ ] ] we claim that this expression exhibits @xmath168 as an ensemble . condition ( 1 ) is satisfied because @xmath343 condition ( 2 ) follows from the minimality of the expression for @xmath168 .
note that for all @xmath344 there exists @xmath345 such that @xmath346 and @xmath347 . in particular
, we have @xmath348 for all @xmath349 , which is condition ( 3 ) .
note that either @xmath350 for some @xmath345 for @xmath351 .
condition ( 4 ) follows from the fact that @xmath323 for all @xmath352 .
therefore , @xmath168 is an ensemble .
it remains to prove that @xmath168 is a @xmath9-ensemble .
note that because @xmath168 is a union of intervals , the number of distinct levels of elements of @xmath168 is the same as the length of any maximal chain in @xmath168 .
suppose that @xmath168 is an @xmath353-ensemble and that @xmath354 is an @xmath353-chain .
thm : characterization.faces ensures that there exists an @xmath353-face @xmath250 whose set of extreme rays is @xmath355 . because @xmath250 is a
non - negatively spanned by @xmath355 and @xmath318 is non - negatively spanned by @xmath268 , we obtain that @xmath356 .
it follows that @xmath357 .
consider a @xmath9-face @xmath356 , which exists because @xmath318 is a @xmath9-flat and every flat is the union of the ( finitely many ) faces it contains .
we see immediately that @xmath358 .
thm : characterization.faces guarantees that @xmath266 is a @xmath9-chain in @xmath144 , from which it follows that @xmath359 .
hence , we have @xmath360 , so that @xmath168 is a @xmath9-ensemble .
the lemma follows .
[ lem : getspecialflats ] given two points of @xmath361 and @xmath362 of @xmath141 such that @xmath363 , there exists a flat @xmath318 of @xmath30 such that @xmath364\cup [ a , \infty]\right ) \cap { \mathbb{e}^\ast}_n = \operatorname{rays}(l).\ ] ] \cup [ a,\infty]\cap { \mathbb{e}^\ast}_n$ ] * with @xmath365 in @xmath141 , @xmath366 , and @xmath367 .
the three figures illustrate how to construct a flat @xmath318 whose set of extreme rays @xmath317 corresponds to the ensemble @xmath368 \cup [ a,\infty]\cap { \mathbb{e}^\ast}_n$ ] . in every case , @xmath318 is the intersection of two hyperplanes .
there are three cases to consider : ( case 1 ) @xmath369 ( case 2.a ) @xmath369 and @xmath370 , ( case 2.b ) @xmath371 and @xmath372 . ]
\cup [ a,\infty]\cap { \mathbb{e}^\ast}_n$ ] * with @xmath365 in @xmath141 , @xmath366 , and @xmath367 .
the three figures illustrate how to construct a flat @xmath318 whose set of extreme rays @xmath317 corresponds to the ensemble @xmath368 \cup [ a,\infty]\cap { \mathbb{e}^\ast}_n$ ] . in every case , @xmath318 is the intersection of two hyperplanes .
there are three cases to consider : ( case 1 ) @xmath369 ( case 2.a ) @xmath369 and @xmath370 , ( case 2.b ) @xmath371 and @xmath372 . ]
\cup [ a,\infty]\cap { \mathbb{e}^\ast}_n$ ] * with @xmath365 in @xmath141 , @xmath366 , and @xmath367 .
the three figures illustrate how to construct a flat @xmath318 whose set of extreme rays @xmath317 corresponds to the ensemble @xmath368 \cup [ a,\infty]\cap { \mathbb{e}^\ast}_n$ ] . in every case , @xmath318 is the intersection of two hyperplanes .
there are three cases to consider : ( case 1 ) @xmath369 ( case 2.a ) @xmath369 and @xmath370 , ( case 2.b ) @xmath371 and @xmath372 . ]
* case 1 : @xmath369 .
consider the flat @xmath375 of @xmath30 .
a direct calculation shows that @xmath364 \cup [ a , \infty]\right ) \cap { \mathbb{e}^\ast}_n = \operatorname{rays}\left(l\right).\ ] ] * case 2 : @xmath376 .
write @xmath377 .
consider the flat @xmath378 of @xmath30 .
a direct calculation shows that @xmath364 \cup [ a , \infty]\right ) \cap { \mathbb{e}^\ast}_n = \operatorname{rays}\left(l\right).\ ] ] consider an ensemble given by @xmath379\cup [ a_1 , b_1]\cup \cdots
\cup [ a_k , b_k]\right).\ ] ] define the intervals @xmath380\cup [ a_{i+1 } , \infty ) \text { for } 0 \le i \le k-1\\ d_k&= ( 0 , b_k].\end{aligned}\ ] ] lem : getspecialflats guarantees that each @xmath381 is the set of extreme rays of a flat @xmath382 . consider the flat @xmath383 a direct calculation shows that @xmath384 as desired .
lem : flatyieldsensemble ensures that @xmath151 induces a function from the set of @xmath9-flats of @xmath30 to the set of @xmath9-ensembles in @xmath144 . because every flat is non - negatively spanned by its extreme rays , the induced function is injective .
it remains to prove that the induced function is surjective .
let @xmath168 be a @xmath9-ensemble .
lem : ensembleyieldsflat guarantees the existence of a flat @xmath250 whose set of extreme rays is @xmath168 . the fact that @xmath250 is a @xmath9-flat follows from lem : flatyieldsensemble and the fact that @xmath168 is a @xmath9-ensemble .
therefore , the induced function is surjective , which implies the theorem .
a subset @xmath267 is a @xmath9-ensemble if and only if @xmath387 is a @xmath9-pseudo - ensemble ( necessarily with starting point 0 ) . in particular
, the function defined by @xmath388 induces a bijection from the set of @xmath9-ensembles in @xmath144 to the set of @xmath9-pseudo - ensembles in @xmath139 with starting point @xmath389 . for all @xmath390 the set
@xmath391 is the unique @xmath392-pseudo - ensemble in @xmath139 .
denote by @xmath393 the number of @xmath9-pseudo - ensembles in @xmath139 for @xmath287 .
define @xmath394 for @xmath289 and note that @xmath394 for @xmath395 and for @xmath396 .
the following two lemmas illustrate the role of @xmath393 in counting ensembles : thm : flatcount will follow from the recurrence relations presented in the lemmas by routine algebraic manipulations involving generating functions . the idea of lem : rhonkrecursion is to divide into cases based on the starting point of a pseudo - ensemble . after translating the starting point to @xmath389 and removing @xmath389
, we obtain an ensemble .
the lemma is clear for @xmath399 .
assume for the remainder of the proof that @xmath400 .
denote by @xmath401 the set of @xmath9-pseudo - ensembles in @xmath139 with starting point @xmath274 .
recall that translation by @xmath299 induces an isomorphism from @xmath282 to @xmath302 .
we see immediately that the composite of translation by @xmath299 and the removal of @xmath389 ( i.e. , the function @xmath402 ) induces a bijection from @xmath401 to the set of @xmath9-ensembles in @xmath302 .
hence , we have @xmath403 recall that @xmath144 contains @xmath306 elements of level @xmath307 for @xmath308 . grouping by level , we obtain that @xmath404 as desired .
we need to introduce the correction term @xmath405 in order to deal with the case of @xmath399 . the proof of lem : hnkrecursion is similar .
we instead divide into cases based on the level of @xmath406 , the maximum of the first interval in the definition of an ensemble .
translating @xmath407 to @xmath408 and intersecting with @xmath139 , we obtain a pseudo - ensemble .
the lemma is clear for @xmath409 and for @xmath410 .
assume for the remainder of the proof that @xmath411 .
denote by @xmath412 the set of @xmath9-ensembles in @xmath144 whose first interval is @xmath413 $ ] . the hypothesis that @xmath414 ensures that @xmath415 , while the definition of an ensemble guarantees that @xmath415 for @xmath416 .
recall that translation by @xmath299 induces an isomorphism from @xmath282 to @xmath302 .
recall that translation by @xmath417 induces an isomorphism from @xmath418 to @xmath419 .
we see immediately that the composite of translation by @xmath417 and intersection with @xmath133 ( i.e. , the function @xmath420 ) induces a bijection from @xmath297 to the set of @xmath421-pseudo - ensembles in @xmath422 .
hence , we have @xmath423 the proof of cor : flatrecurse is similar to that of cor : facerecurse .
in particular , cor : flatrecurse can be derived by noting that the coefficient of @xmath57 in @xmath428 vanishes for @xmath55 due to thm : flatcount .
p. orlik and h. terao , http://dx.doi.org/10.1007/978-3-662-02772-1 [ _ arrangements of hyperplanes _ ] , vol .
300 of _ grundlehren der mathematischen wissenschaften [ fundamental principles of mathematical sciences]_. springer - verlag , berlin , 1992 . http://dx.doi.org/10.1007/978-3-662-02772-1 .
g. m. ziegler , http://dx.doi.org/10.1007/978-1-4613-8431-1[_lectures on polytopes _ ] , vol .
152 of _ graduate texts in mathematics_. springer - verlag , new york , 1995 .
http://dx.doi.org/10.1007/978-1-4613-8431-1 .
m. esole , s .- h .
shao , and s .- t .
yau , `` singularities and gauge theory phases , '' http://arxiv.org/abs/1402.6331[arxiv:1402.6331 [ hep - th ] ]
. m. esole , s .- h .
shao , and s .- t .
yau , `` singularities and gauge theory phases ii , '' http://arxiv.org/abs/1407.1867[arxiv:1407.1867 [ hep - th ] ] .
h. hayashi , c. lawrie , d. r. morrison , and s. schafer - nameki , `` box graphs and singular fibers , '' http://dx.doi.org/10.1007/jhep05(2014)048 [ _ jhep _ * 1405 * ( 2014 ) 048 ] , http://arxiv.org/abs/1402.2653[arxiv:1402.2653 [ hep - th ] ] . t. zaslavsky , `` facing up to arrangements : face - count formulas for partitions of space by hyperplanes , '' _ mem . amer . math .
soc . _ * 1 * no . issue 1 , 154 , ( 1975 ) vii+102 .
m. esole , s. g. jackson , r. jagadeesan and a. g. nol , `` incidence geometry in a weyl chamber ii : @xmath429 , '' arxiv:1601.05070 [ math.rt ] . s. ferrara , r. minasian , and a. sagnotti , `` low - energy analysis of m and f theories on calabi - yau threefolds , '' http://dx.doi.org/10.1016/0550-3213(96)00268-4[_nucl .
* b474 * ( 1996 ) 323342 ] , http://arxiv.org/abs/hep-th/9604097[arxiv:hep-th/9604097 [ hep - th ] ] .
k. a. intriligator , d. r. morrison , and n. seiberg , `` five - dimensional supersymmetric gauge theories and degenerations of calabi - yau spaces , '' http://dx.doi.org/10.1016/s0550-3213(97)00279-4[_nucl.phys._ * b497 * ( 1997 ) 56100 ] , http://arxiv.org/abs/hep-th/9702198[arxiv:hep-th/9702198 [ hep - th ] ] . m. esole and s .- t .
yau , `` small resolutions of su(5)-models in f - theory , '' http://dx.doi.org/10.4310/atmp.2013.v17.n6.a1[_adv .
* 17 * ( 2013 ) 11951253 ] , http://arxiv.org/abs/1107.0733[arxiv:1107.0733 [ hep - th ] ] .
k. matsuki , `` weyl groups and birational transformations among minimal models , '' http://dx.doi.org/10.1090/memo/0557[_mem .
soc . _ * 116 * no .
557 , ( 1995 ) vi+133 ] .
http://dx.doi.org/10.1090/memo/0557 . d .- e .
diaconescu and r. entin , `` calabi - yau spaces and five - dimensional field theories with exceptional gauge symmetry , '' http://dx.doi.org/10.1016/s0550-3213(98)00689-0[_nucl .
phys . _ * b538 * ( 1999 ) 451484 ] , http://arxiv.org/abs/hep-th/9807170[arxiv:hep-th/9807170 [ hep - th ] ] . m. esole , s. jackson , r. jagadeesan , and a. nol .
incidence geometry in a weyl chamber ii : @xmath429 http://arxiv.org/abs/1601.05070[arxiv:1601.05070 [ math - rt ] ] .
m. esole and s .- h .
shao , `` m - theory on elliptic calabi - yau threefolds and 6d anomalies , '' http://arxiv.org/abs/1504.01387[arxiv:1504.01387 [ hep - th ] ] .
d. r. morrison and d. s. park , `` f - theory and the mordell - weil group of elliptically - fibered calabi - yau threefolds , '' http://dx.doi.org/10.1007/jhep10(2012)128[_jhep_ * 10 * ( 2012 ) 128 ] , http://arxiv.org/abs/1208.2695[arxiv:1208.2695 [ hep - th ] ] . m. esole , m. j. kang , and s .- t .
yau , `` a new model for elliptic fibrations with a rank one mordell - weil group : i. singular fibers and semi - stable degenerations , '' http://arxiv.org/abs/1410.0003[arxiv:1410.0003 [ hep - th ] ] .
v. braun , t. w. grimm , and j. keitel , `` geometric engineering in toric f - theory and guts with u(1 ) gauge factors , '' http://dx.doi.org/10.1007/jhep12(2013)069[_jhep_ * 1312 * ( 2013 ) 069 ] , http://arxiv.org/abs/1306.0577[arxiv:1306.0577 [ hep - th ] ] .
m. cvetic , d. klevers , and h. piragua , `` f - theory compactifications with multiple u(1)-factors : constructing elliptic fibrations with rational sections , '' http://dx.doi.org/10.1007/jhep06(2013)067[_jhep_ * 1306 * ( 2013 ) 067 ] , http://arxiv.org/abs/1303.6970[arxiv:1303.6970 [ hep - th ] ] . m. kuntzler and s. schafer - nameki , `` tate trees for elliptic fibrations with rank one mordell - weil group , '' http://arxiv.org/abs/1406.5174[arxiv:1406.5174 [ hep - th ] ] .
c. lawrie , s. schafer - nameki , and j .-
wong , `` f - theory and all things rational : surveying u(1 ) symmetries with rational sections , '' http://arxiv.org/abs/1504.05593[arxiv:1504.05593 [ hep - th ] ] .
c. mayrhofer , e. palti , and t. weigand , `` u(1 ) symmetries in f - theory guts with multiple sections , '' http://dx.doi.org/10.1007/jhep03(2013)098[_jhep_ * 1303 * ( 2013 ) 098 ] , http://arxiv.org/abs/1211.6742[arxiv:1211.6742 [ hep - th ] ] .
d. r. morrison and w. taylor , `` sections , multisections , and u(1 ) fields in f - theory , '' http://arxiv.org/abs/1404.1527[arxiv:1404.1527 [ hep - th ] ] .
w. fulton , _ introduction to toric varieties _ ,
131 of _ annals of mathematics studies_. princeton university press , princeton , nj , 1993 .
the william h. roever lectures in geometry . | we study the central hyperplane arrangement whose hyperplanes are the vanishing loci of the weights of the first and the second fundamental representations of @xmath0 restricted to the dual fundamental weyl chamber .
we obtain generating functions that count flats and faces of a given dimension .
this counting is interpreted in physics as the enumeration of the phases of the coulomb and mixed coulomb - higgs branches of a five dimensional gauge theory with 8 supercharges in presence of hypermultiplets transforming in the fundamental and antisymmetric representation of a @xmath1 gauge group as described by the intriligator - morrison - seiberg superpotential . |
since haider and liu postulated a possible existence of @xmath1-mesic nuclei @xcite , many experimental groups performed measurements dedicated to search for the new kind of nuclear matter in which the @xmath1 meson is bound within a nucleus via the strong interaction . however , till now , none of the experiments have brought the clear evidence for the bound state existence .
the status of the search was recently described in the following reviews @xcite .
some of the experiments set the upper limits for the bound state production in several processes .
group estimated the upper limit of total cross section for @xmath5 process to the value of 270 nb and for @xmath5 @xmath6-@xmath7 @xmath8 to the value 70 nb .
cosy - gem measurement of @xmath9 @xmath10 brought the upper limit of the total cross section for @xmath11-@xmath12 production equal to 0.46 @xmath13 0.16(stat ) @xmath13 0.06(syst ) nb @xcite .
the wasa measurement in 2008 results in the upper limit of of the total cross section for the @xmath14-@xmath12 creation in @xmath2 @xmath15 reaction , which varies from 20 nb to 27 nb for the range of the bound state width from 5 mev to 35 mev @xcite .
the measurement carried out two years later permitted to lower the upper bound for the cross section of @xmath16-@xmath17 @xmath4 process down to the value of few nanobarns .
additionally , the upper limit of the preliminary total cross section was determined for the first time for the @xmath14-@xmath12 production in @xmath2 @xmath18 reaction @xcite .
this paper presents the preliminary results obtained for the aforementioned processes .
in november 2010 , wasa - at - cosy collaboration carried out the experiment dedicated for the search for @xmath0-@xmath1 bound states in @xmath2 @xmath3 and @xmath2 @xmath4 reactions . the ramped beam technique was used to vary the momentum continuously from 2.127 gev / c to 2.422 gev / c , which corresponds to a range of excess energies @xmath19 from -70 to 30 mev @xcite .
the detailed description of the wasa experimental setup is presented in @xcite .
analysis for the @xmath2 @xmath3 and @xmath2 @xmath4 reactions were carried out independently .
next , the set of the cross - check tests was performed to assure the consistency at the pid level .
the @xmath20 ions and nucleon - pion pairs were identified in the forward and central detector , respectively .
the deposited energy patterns in thick scintillator layers of the forward hodoscope was used to identify the @xmath20 ions ( the ) .
the neutral pion @xmath21 was reconstructed based on the invariant mass of two gamma quanta while the neutron was identified via the missing mass technique @xcite .
the proton and @xmath22 identification was based on the measurement of the energy loss in the thin plastic scintillator barrel combined with the energy deposited in the electromagnetic calorimeter @xcite .
the events which may correspond to the bound states production were selected using criteria based on monte carlo simulations for the @xmath1-mesic nuclei production and decay .
we apply the cuts in the momentum of @xmath20 in the cm frame , nucleon cm kinetic energy , pion cm kinetic energy and the opening angle between nucleon - pion pair in the cm .
the region rich in signal corresponds to the momenta of the @xmath20 in the range . for this region
the excitation function was obtained by normalizing the events selected in individual excess energy intervals by the corresponding integrated luminosities ( the detailed description of the luminosity determination one can find in ref .
@xcite ) and corrected for acceptance and efficiency .
the excitation function does not reveal the resonance - like structure , which could be the signature of the @xmath1-mesic nuclei existence @xcite , however the interpretation of the results is still in progress .
so far , the upper limit of the total cross section for the @xmath16-@xmath17 @xmath3 and @xmath16-@xmath17 @xmath4 processes was determined on the 90% confidence level .
preliminary , the upper limits were obtained by the fit of the sum of the polynomial and breit - wigner functions to the experimentally determined excitation functions .
it varies from 21 to 36 nb for the first channel and from 5 to 9 nb for the second channel for the bound state width ranging from 5 to 50 mev ( see fig . [ result_sigma_upp ] ) .
-@xmath17 @xmath3 ( left panel ) and @xmath16-@xmath17 @xmath4 ( right panel ) reaction as a function of the width of the bound state .
the binding energy was set to 30 mev .
the green areas denote the systematic uncertainties @xcite .
[ result_sigma_upp],title="fig:",width=226,height=151 ] -@xmath17 @xmath3 ( left panel ) and @xmath16-@xmath17 @xmath4 ( right panel ) reaction as a function of the width of the bound state .
the binding energy was set to 30 mev .
the green areas denote the systematic uncertainties @xcite .
[ result_sigma_upp],title="fig:",width=226,height=151 ] a possible broad state in the case of @xmath16-@xmath17 @xmath3 reaction can not be excluded by the current data set @xcite .
the kinematic region , where we expect the evidence of the signal from the bound state corresponding to @xmath20 momenta in the cm system in range , can not be fully described only by the combination of the considered background processes ( see left panel of fig .
[ fit_plots ] ) .
in contrast , as shown in the right panel of fig . [ fit_plots ] the experimental excitation function is very well fitted by the background contributions for the region where the signal is not expected .
@xmath3 ( green squares ) and @xmath23 @xmath24 @xmath3 ( magenta squares ) . a sum of both background contributions is shown as blue triangles . left and right panels show results for the regions rich in signal and poor in signal , respectively .
the figure is adopted from @xcite .
[ fit_plots],title="fig:",width=226,height=151 ] @xmath3 ( green squares ) and @xmath23 @xmath24 @xmath3 ( magenta squares ) .
a sum of both background contributions is shown as blue triangles . left and right panels show results for the regions rich in signal and poor in signal , respectively .
the figure is adopted from @xcite .
[ fit_plots],title="fig:",width=226,height=151 ]
the excitation functions were determined for @xmath16-@xmath17 @xmath4 and @xmath16-@xmath17 @xmath3 processes , however none of them reveal any direct narrow structure which could be signature of the bound state with width less than 50 mev .
the interpretation of the resuts is still in progress .
so far preliminary upper limit of the total cross section for the @xmath1-mesic @xmath0 formation and decay was estimated . in case of @xmath16-@xmath17 @xmath4 reaction we obtained the preliminary upper limit of the total cross section in order of few nb which is about four times lower in comparison with the result obtained from 2008 data @xcite . comparing to theoretically estimated value @xcite ,
the obtained upper limit value does not exclude the existence of the bound state .
the excitation function for the reaction @xmath16-@xmath17 @xmath3 was obtained for the first time in the experiment .
the obtained upper limit is here by factor of five larger than predicted value therefore , we can conclude , that the current measurement does not exclude the existence of bound state also in this process @xcite .
moreover , the excitation function obtained for this reaction is a subject of interpretation of few theoretical groups with respect to very wide or bound state @xcite . in may 2014
, we extended the search for to @xmath20-@xmath1 sector @xcite .
we chose processes corresponding to the three mechanisms : ( i ) absorption of the @xmath1 meson by one of the nucleons , which subsequently decays into @xmath25-@xmath26 pair e.g. : @xmath27 ( @xmath20-@xmath7 @xmath28 , ( ii ) decay of the @xmath1 -meson while it is still `` orbiting '' around a nucleus e.g. : @xmath27 ( @xmath20-@xmath7 @xmath29 or @xmath27 ( @xmath20-@xmath7 @xmath30 reactions and ( iii ) @xmath1 meson absorption by few nucleons e.g. : @xmath27 ( @xmath20-@xmath7 @xmath31 or @xmath27 ( @xmath20-@xmath7 @xmath32 .
almost two weeks of measurement with an average luminosity of about 6@xmath33 @xmath34 s@xmath35 allowed to collect a world largest data sample for @xmath20-@xmath1 .
the data analysis is in progress .
the search for @xmath1 and @xmath36 - mesic bound states is carried out also by other international collaborations , e.g. at j - parc @xcite and at gsi @xcite . in parallel ,
several theoretical studies are ongoing @xcite .
we acknowledge support by the foundation for polish science - mpd program , co - financed by the european union within the european regional development fund , by the polish national science center through grants no .
dec-2013/11/n / st2/04152 , 2011/01/b / st2/00431 , 2011/03/b / st2/ 01847 and by the ffe grants of the forschungszentrum jlich .
k. yoshiki _ et .
arxiv:1503.03566 , proceedings of the 20th international conference on particles and nuclei ( panic 14 ) , doi:10.3204/desy - proc-2014 - 04/106 _ , 24 - 29 august 2014 , hamburg , germany , page 286 , 2015 . | we performed a search for @xmath0-@xmath1 bound state in @xmath2 @xmath3 and @xmath2 @xmath4 reactions with the wasa - at - cosy facility using a ramped beam technique .
the measurement was carried out with high statistics and high acceptance .
the signature of @xmath1-mesic nuclei was searched for by the measurement of the excitation functions in the vicinity of the @xmath1 production threshold for each of the considered channels .
we did not observe the narrow structure which could be interpreted as a bound state .
the preliminary upper limits of the total cross sections for the bound state production and decay varies from 21 nb to 36 nb for the @xmath2 @xmath3 channel , and from 5 nb to 9 nb for the @xmath2 @xmath4 channel for the bound state width ranging from 5 to 50 mev . |
the effects of weak gravitational lensing by the large - scale structure have been detected in several samples of high redshift qsos , intermediate redshift galaxies , and batse grbs . in the case of point sources , qsos and grbs ,
weak lensing manifests itself as angular ( anti-)correlations between these sources and foreground inhomogeneously distributed mass @xcite , while in the case of galaxies weak lensing is detected through its coherent shear effect ( see @xcite for a recent review ) . in principle
, there is another , more direct way of detecting weak lensing , which uses fluxes of standard candles .
if the observed magnitudes of standard candles are corrected for cosmological distances then the effect of lensing can be seen : brighter sources will lie behind regions of mass density excess , while fainter ones will have mass deficits in their foregrounds .
the best example of cosmological standard candle , supernovae type ia ( snia ) have been extensively observed with the purpose of determining the global geometry of the universe @xcite .
nuisance effects like evolution , variations in individual sn , and gray dust extinction have been studied theoretically and observationally , and have either been corrected for or shown to be small .
weak lensing , another nuisance effect has been addressed theoretically by several authors @xcite and found to be unimportant given the current uncertainties .
for example , @xcite used ray tracing through cosmological simulations and found that the lensing induced dispersions on truly standard candles are @xmath3 and @xmath4 mag at redshift @xmath5 and @xmath6 , respectively , in a cobe - normalized cold dark matter universe with @xmath7 , @xmath8 , @xmath9km / s / mpc and @xmath10 .
these are small variations compared to the current errors which are @xmath11 mag .
even though weak lensing effects are estimated to be small for @xmath12 , they are predicted to be non - negligible for higher redshift sources , so it is not surprising that the highest redshift snia , sn1997ff at @xmath13 has been examined by several authors @xcite for the effects of weak lensing due to galaxies along the line of sight .
present day high-@xmath0 snia samples are dominated by lower redshift sne , and so have not been examined for the effects of lensing .
the main goal of this work is to determine if the observed fluxes of the cosmologically distant snia have suffered significantly from lensing induced ( de- ) amplifications .
the largest homogeneous compilation of snia has been recently published by @xcite : table 15 of that paper contains 74 sne at @xmath14 .
the authors use four different light curve fitting methods ( mlcs , @xmath15 , modified dm15 , and bayesian adapted template match ) to estimate distances to sne .
the final quoted distance is the median of the estimates of the four individual methods , and the uncertainty is the median of the error of the contributing methods .
the analysis presented in @xcite yields values of the global cosmological parameters ; if a flat model is assumed , then @xmath16 and @xmath17 .
we use these values in all the analysis of the present paper . as tracers of foreground mass density we use apm galaxies @xcite .
apm provides near full coverage of the sky in the northern and southern hemispheres , at @xmath18 . in our analysis
we use only the central @xmath19 of apm plates . since the plate centres are separated by @xmath20 ,
there exist small portions of the sky that are not covered by any plate . as
a result of these cuts , only 55 of the 74 sne lie on the usable parts of apm plates .
the median redshift of the 55 sne is @xmath21 .
is not in our sample : it fell in the cracks between the apm plates . ]
since most of the sne have rather low redshifts , care must be taken to ensure that galaxies are foreground to the sne . furthermore , because sne span a large range of nearby redshifts , from @xmath22 to @xmath23 , the optimal lens redshift @xmath24 will depend on @xmath25 much more compared to a typical case where sources ( say , qsos ) are at @xmath26 and so the redshift of optimal lenses is roughly independent of @xmath25 . in our analysis
we adjust @xmath24 for each sn source by selecting the appropriate limiting apparent magnitude , mag@xmath27 for apm galaxies on red plates .
@xcite gives an empirical expression for the median redshift @xmath28 of a galaxy sample with a given faint magnitude flux cutoff .
this median redshift can be equated with the optimal lens redshift @xmath24 , and hence the magnitude limit of the foreground galaxies can be determined for every sn separately . however , there is a small catch . for @xmath29 optimal @xmath30 .
the galaxy redshift distribution whose median redshift @xmath31 has a considerable tail extending beyond @xmath32 . to avoid the problem of source / lens redshift overlap we use @xmath33 , where factor of 2 was chosen arbitrarily .
we explore the dependence of the results on this factor in section [ robust ] .
around every sn we draw a circle of radius @xmath34 , and count the number of galaxies , @xmath35 , in the appropriate magnitude range .
this number is compared to the average number density in control circles , @xmath36 .
fractional galaxy excess is @xmath37 .
control circles are confined to the same apm plate as the sn , and to the same distance from the plate centre as the sn ( to reduce the effects of vignetting ) ; however , scattering the control circles randomly on the plate does not change the results significantly . for each
sn we also calculate @xmath38 , where @xmath39 is the number of control circles , out of total @xmath40 , that have less galaxies in them than the circle around the sn .
in other words , @xmath38 is the rank of the sn circle among its control ` peers ' . if sne are randomly distributed with respect to the foreground galaxies , then average @xmath41 .
if sne have an excess ( deficit ) of galaxies in front of them then their @xmath42 will be greater ( less ) than 0.5 .
analogous to the medians being more stable than averages , @xmath38 rank statistic is more stable than @xmath43 . , and @xmath17 , and extinction , k - corrected distances from @xcite .
@xmath44 is a measure of the number density of foreground galaxies in circles of radius @xmath34 around sne ( see section [ anal ] for details ) .
there are a total of 55 sources ( with magnitude errors ) , but only 50 ( filled points ) are used in the analysis .
the 50 points are grouped into three bins , whose horizontal size is shown as thick horizontal `` error - bars '' .
the corresponding vertical error - bars show the deviation of the mean of the points in each bin ( the rms is 4 times larger ) , and the intersection of the thick lines are the averages of the mag@xmath45 of sne in each bin .
the thin slanting line is the best - fit to 50 sne , and has a slope @xmath46 .
the significance of the correlation is @xmath2.,width=317 ] figure [ figone ] shows absolute magnitudes , @xmath47 , and @xmath38 ranks of 55 sne found on apm plates .
the effect of flux dimming due to cosmological distances has been taken out , i.e. all the sne have been ` brought ' to the same redshift ; the constant magnitude offset on the vertical axis is irrelevant for this work .
there are two sne whose magnitudes make them @xmath48 outliers , sn1997o , and sn1997bd represented by empty circles with dotted line error - bars .
we exclude these from our analysis . as quoted in @xcite .
sn1997o and sn1994h were excluded from the primary fit ( fit c ) by the analysis of @xcite .
we do not exclude sn1994h from our analysis , but if we did it would improve the trend , as its coordinates on fig .
[ figone ] are ( 0.427 ; -0.606 ) . ] .
distance errors , @xmath49 were derived by @xcite , and listed in column 9 of their table 15 . the solid histogram is for 55 sn ; the dashed line represents 31 sn from the high-@xmath0 supernova search team .
the 3 outlier sne with @xmath50 are not used in the present analysis.,width=317 ] the distance ( or , magnitude ) errors , as estimated by @xcite for the 55 sne are shown in fig .
[ figthree ] , as the solid line histogram .
the dashed line represents the subsample of 31 sne from high-@xmath0 supernova search team @xcite , whose errors appear to be generally smaller than those of the supernova cosmology project @xcite .
we use sne from both the groups , but exclude three outliers in fig .
[ figthree ] , whose @xmath51 . in fig .
[ figone ] these are represented by empty circles with dashed line error - bars .
thus , we exclude a total of 5 sne from our analysis , leaving us with 50 , shown as the solid points in fig .
[ figone ] .
these 50 sne exhibit a relation between their @xmath47 and @xmath38 , in the sense that brighter sne have an excess of galaxies in their foregrounds . for illustration purpose
only , we bin the 50 sne into three bins ; fig .
[ figone ] shows the extent of the bins and the deviation of the mean of the sne magnitudes in each bin as thick lines . the best - fit line to the 50 sne in fig .
[ figone ] has a slope @xmath46 , and is shown as a thin slanting line .
this fit does not include magnitude errors .
to include the errors we do the following .
we calculate the best - fit slope for 10,000 realizations of the data , where each data point s sn magnitude is replaced by a randomly picked magnitude from a gaussian distribution centred on the actual magnitude value , and having width equal to the quoted error .
this procedure correctly incorporates the information contained in the errors , and produces a distribution of best - fit slopes , which is shown as a solid line histogram in fig .
[ figextra ] .
this distribution shows that the median best - fit slope is @xmath52 , while @xmath53 ( i.e. a case of no correlation ) is ruled out at 99.80% confidence level .
had we used 53 sne ( i.e. had we not exluded the 3 sne with large distance errors ) , the median best - fit slope would gave been @xmath54 , while @xmath53 would have been ruled out at 99.37% confidence level .
the corresponding distribution of @xmath55 slopes is shown as the dashed line in fig .
[ figextra ] . ?
the histograms show the distribution of derived @xmath55 values in 10,000 realizations of the data .
the solid ( dashed ) line histogram represents results using 50 ( 53 ) sne .
the short vertical lines at the bottom of the plot indicate the medians of the distributions.,width=317 ] next , we determine the likelihood of this relation arising by chance . to that end
, we estimate the significance of the relation in two ways .
first , we assign random positions to sne ( keeping these control sne on the same apm plate as the original source ) , and redo all the analysis .
we repeat this 1000 times , and in 8 cases we find @xmath56 , implying statistical significance of @xmath57 .
second , we take the list of observed @xmath47 values and randomly reassign them to observed sn sky positions .
10,000 randomized sne samples are created in this fashion , and only @xmath58 of these have @xmath59 .
based on these two tests we conclude that the significance of the @xmath47@xmath38 relation is better than @xmath60 .
these results are consistent with weak gravitational lensing , which would amplify sne found behind more nearby mass concentrations , as traced by apm galaxies .
alternatively , the results could be due to the action of galactic dust , which will obscure certain directions of the sky making galaxies less numerous and sne fainter .
we consider galactic dust further in section [ disc ] ; in sections [ model ] and [ montecarlo ] we proceed on the assumption that weak lensing is responsible to the @xmath47@xmath38 relation .
the distribution of sne points in fig . [ figone ] depends on specific choices that we made for certain parameters , in particular we chose circles of radius @xmath34 and galaxy magnitude limit such that @xmath61 , where @xmath62 ( see section [ data ] ) .
how would the results change if different choices were made ?
in other words , how robust is our result , would it disappear had we picked a different set of parameters ? of circles around sne .
the vertical axis shows @xmath55 , the slope of the best - fit line to the relation of the type shown in fig [ figone ] .
the filled circles represent cases where galaxies were counted around sn ; stars symbols represent cases where galactic stars were used instead ( i.e. control experiment ) .
the dotted lines denote , _ approximately _ , the levels of statistical confidence .
the dashed vertical line indicates radius @xmath63 used in the analysis of section [ anal].,width=317 ] figure [ figthetatest ] shows the effect of changing @xmath63 .
the vertical axis is @xmath55 , the best - fit to the @xmath47@xmath38 relation in each case .
filled points represent cases where _ galaxies _ were counted and used to determine the @xmath38 rank , while star symbols represent cases where _ galactic stars _ were used instead .
as expected , @xmath64 do not correlate with sne magnitudes , and the values of the best - fit slopes are near zero .
however , because the apm star - galaxy classifier is not perfect , some ` stars ' are actually galaxies , which accounts for some signal being seen when using @xmath64 .
horizontal dotted lines mark the approximate location of @xmath65 , and @xmath60 confidence levels .
these are only approximate because every point in the plot will have its own significance level , but because the number of sne contributing to each point is the same in each case , and the total dispersion in sne magnitudes is the same , same @xmath55 values have about the same significance , regardless of @xmath63 .
we note that for very small @xmath63 the galaxy numbers become very small , and poisson noise drowns out any @xmath47@xmath38 correlation that might exist , so the upturn in the values of the best - fit slope at small @xmath63 is probably not real .
the dashed vertical line marks the @xmath63 value used in section [ anal ] .
we conclude that significant @xmath47@xmath38 anti - correlations occur only with galaxies and not with galactic stars ( which serve as a control sample ) , and only for @xmath66 . .
similar to fig .
[ figthetatest ] ; see section [ robust ] for details.,width=317 ] figure [ figmagtest ] shows the effect of changing the median redshift , @xmath61 of the apm galaxies , or equivalently , mag@xmath27 .
the hidden variable which is varied along the horizontal axis is @xmath67 .
since each sn has a different value of @xmath24 and hence mag@xmath27 , depending on its @xmath25 , there is no unique way of labeling the horizontal axis by using galaxy magnitudes , or redshifts .
we label that axis by assuming @xmath68 , the median of the sne redshift distribution . at the top of the plot
we show how galaxy magnitude limit on the horizontal axis translates into the median galaxy redshift ( for a specific case of @xmath68 ) .
we see that significant anti - correlation between @xmath47 and @xmath38 of foreground galaxies occurs for galaxies with @xmath69 . using galactic stars instead of galaxies
produces no significant results .
the value we used in section [ anal ] is shown with a vertical dashed line , and corresponds to @xmath70 . as in fig .
[ figthetatest ] the upturn in the best - fit slope values at bright mag@xmath27 is probably due to poisson noise . the range of redshifts of apm galaxies that act as the best lenses for sne are @xmath71 ; more distant galaxies show no signal , as expected , if lensing is the correct interpretation of the data , because more distant galaxies are either too close to sne in redshift or are actually at the same @xmath0 .
a number of other tests have been carried out as well .
for example , instead of using a @xmath25-dependent mag@xmath27 , we tried fixed values of @xmath72 , which gave , @xmath73 , respectively , comparable to , but somewhat smaller than those seen in fig .
[ figmagtest ] .
this is not surprising : @xmath25-dependent galaxy magnitude limits pick optimal lens redshifts for each source , thus maximizing the observed lensing signature .
we also reran the analysis with subsamples of the entire 50-source sample .
we split the sne according to the teams : 20 supernova cosmology project snia gave a @xmath47@xmath38 anti - correlation significant at 95.8% , while the corresponding significance level for the 30 sne from the high-@xmath0 supernova search team is 96% . because the quality of ukst apm ( southern hemisphere ) plates is higher than poss apm ( northern hemisphere ) plates , and because there is some overlap between plates along the equator , we used ukst plates whenever possible . redoing the analysis using only the 36 ukst sne we get the best fit slope @xmath74 at a significance level of 99.6% ,
while the 14 poss sne had @xmath75 and the correlation was not significant , which is not surprising given the size of this subsample . splitting the whole sample into low and high redshift groups we get the following results : 25 sne with @xmath76
have a slope @xmath77 at 96.8% , while 25 sne with with @xmath78 have a slope @xmath79 at 88.9% .
these tests suggest that the @xmath47@xmath38 relation has a physical origin ( weak lensing or galactic dust ) , and is not an artifact arising form one subset of the data .
in this section we adopt the weak lensing interpretation of the @xmath47@xmath38 relation .
our simple lensing model does not use the individual values of @xmath25 , and the optimal redshift distribution of the corresponding lenses . in lieu of these parameters
we use the lensing optical depth of the matter traced by the galaxies , @xmath80 . for a fixed global geometry
, @xmath80 depends on @xmath25 and the redshift distribution of the mass traced by the galaxies .
we assume that the apm galaxies faithfully trace the mass up to some redshift , @xmath81 , then @xmath82^{1/2}\ ; \sigma_{crit}(z , z_s ) } } \ ; dz,\end{aligned}\ ] ] in that case , projected fractional mass excess is @xmath83 , where @xmath84 is the convergence , with respect to a smooth universe ( filled beam ) for the corresponding line of sight . in the weak lensing regime source amplification @xmath85 ; @xmath86 for an unlensed source . the smallest value that @xmath84 can attain is minus the total optical depth , @xmath87 , which corresponds to emptying out all the mass along the line of sight from the observer to @xmath81 . in addition to the value for @xmath80 , our model has three ingredients , ( 1 ) mass distribution , i.e. a probability distribution ( pdf ) for @xmath88 , which is related to @xmath84 pdf , ( 2 ) a biasing scheme , which relates @xmath88 to the projected fractional galaxy number density excess , @xmath43 , and ( 3 ) the dispersion in the intrinsic magnitudes of snia s .
given this information we generate synthetic snia samples , with 50 sources each , then compute @xmath38 each sn , and the observed sn magnitude , @xmath47 .
the slope of the best - fit line to the @xmath47@xmath38 relation , @xmath55 is then used to test how well a given model reproduces the observations . because our model assumes the same @xmath24 distribution for all sources , the value of @xmath55 to compare our models predictions to should be the one obtained by using a constant mag@xmath89 , i.e. @xmath90 ( see fourth paragraph in section [ robust ] ) .
the specifics of the three model ingredients , and the associated model parameters are described later in this section .
the goal of the modeling is to determine what set of parameters can reproduce the observations , i.e. have @xmath91 .
we are particularly interested in what effect biasing has on the results .
it is often suggested that the large amplitude of qso - galaxy correlations mentioned in the introduction is , at least in part , due to the fact that biasing is not a simple linear , one - to - one mapping from @xmath88 to @xmath43 .
qso - galaxy correlation function ( @xmath92 ) , galaxy autocorrelation ( @xmath93 ) and the matter fluctuation power spectrum ( @xmath94 ) are related by @xmath95 where @xmath96 is the biasing parameter .
this means that @xmath92 probes a _ combination _ of @xmath96 and @xmath94 .
thus , for a given @xmath94 , @xmath92 can be enhanced with positive biasing , especially if it is non - linear on relevant spatial scales .
the important difference between weak lensing induced qso - galaxy correlations and the effect on standard candles we have studied here is that in the latter case biasing plays a minor role .
the slope of the @xmath47@xmath38 relation is most sensitive to the mass distribution .
our modeling demonstrates this later ; first , we described the specifics of the three model ingredients .
based on the results from cosmological n - body simulations @xcite the shape of the probability distribution function ( pdf ) of @xmath84 is very roughly gaussian , but asymmetric , with the most probable @xmath84 , which we call @xmath97 being less than 0 , i.e. most sources are deamplified .
the tail of the @xmath84 distribution extends to high positive values of @xmath84 .
we use these published results of @xcite to construct an approximate shape for our @xmath84 pdf : @xmath98 ^ 2/2\sigma_{\kappa_1 } ) , \quad { \rm for}\quad \kappa > \kappa_m\\ exp\;(-[\kappa-\kappa_m]^2/2\sigma_{\kappa_2 } ) , \quad { \rm for}\quad \kappa\le\kappa_m \end{array } \right .
\label{stdpdf}\ ] ] the pdf is a combination of two half - gaussians , with two widths , @xmath99 and @xmath100 , which describe the high @xmath84 and the low @xmath84 sides of the pdf respectively .
the location of the peak of the pdf , @xmath97 is adjusted such that the average @xmath84 is 0 , which implies @xmath101 @xmath97 is always negative , since the skewness of the pdf dictates that @xmath102 . from the numerically computed pdfs of @xcite we estimate that for a flat @xmath103 model , where the sources are at @xmath104 , and the smoothing scale is @xmath105 , the following approximations apply : @xmath106 , and @xmath107 .
thus we now have a realistic @xmath84 pdf , appropriate for a standard cosmology with standard mass distribution .
figure [ grid](d ) shows such a pdf as a short - dash line .
in addition to the standard @xmath84 pdf described above , we also try an extreme form for @xmath84 pdf obtained using the `` bifocal lens '' mass distribution proposed by @xcite : @xmath108 where @xmath109 is the kronicker s delta function , @xmath110 , and @xmath111 . for a fixed @xmath80 ,
this pdf can produce more pronounced lensing effects than the standard pdf , if @xmath112 is assigned the smallest allowable value , and @xmath113 is made very large .
this is because the standard pdf has a large probability near @xmath86 , while the bifocal pdf avoids such values altogether . the distribution of mass corresponding to eq .
[ bifocals ] is unrealistic .
one can make it somewhat realistic by allowing a range of @xmath113 and @xmath112 values .
we do the following : for any one line of sight we randomly pick @xmath113 and @xmath112 from specified ranges , using flat priors , and eq .
[ bifocals ] sets the probability distribution of the two amplifications .
we use two sets of ranges : ( 1 ) @xmath114 and @xmath115 ( we will call this model bifocal i model ) , ( 2 ) @xmath116 and @xmath117 ( we will call this bifocal ii ) . the corresponding @xmath84 pdfs , obtained by considering many lines of sight are shown in fig .
[ grid](d ) as a solid line for bifocal ii and as a long - dash line for bifocal i. in bifocal ii the minimum value of @xmath112 is the minimum allowed value ; the other limits on @xmath118 s were picked arbitrarily . for comparison ,
the standard @xmath84 pdf of section [ stdmass ] is also shown , as the short - dash line .
optical depth of @xmath119 was assumed for all three .
compared to a standard pdf , bifocal pdfs imply that most of the lines of sight are rather empty , and there are a few lines of sight with very high values of @xmath84 .
projected fractional mass excess , @xmath88 is related to the observationally accessible quantity , @xmath43 through biasing .
we chose `` power law '' biasing , motivated by numerical simulations of @xcite and @xcite , @xmath120 where the power law part allows the biasing to be non - linear , with different indexes depending on the sign of @xmath88 , while @xmath121 is a stochastic biasing component which is chosen randomly for each sn ; @xmath121 distribution has a gaussian shape and width @xmath122 .
factor @xmath123 multiplying @xmath121 ensures that the dispersion in @xmath43 is reduced in underdense regions .
qualitatively , the @xmath43 vs. @xmath88 relations produced by eq .
[ biasing ] look similar to those in fig.1 of @xcite .
we assume that the dispersion in magnitude about the perfect standard candle case has a gaussian shape , with width @xmath124 .
each model has five independent parameters : + @xmath125 . we assume flat priors for the three biasing parameters , @xmath126 , @xmath127 , and @xmath122 , as well as for @xmath124 .
each specific set of parameters together with one of the three mass distribution models generates a synthetic realization of the 50-source snia sample . from this entire ensemble of realizations , we only consider those that satisfy these observational constraints : ( 1 ) the total rms dispersion in the synthetic snia magnitudes ( which includes intrinsic and lensing induced contributions ) must be within 0.05 mag of the actual observed value of 0.3mag ; ( 2 ) the moments of the synthetic @xmath43 distribution ( average , standard deviation , skewness and kurtosis ) must be reasonably close to those of the observed distribution , which is characterized by ( 0.063 , 0.52 , 1.7 , 4.9 ) .
we chose `` reasonably close '' to mean that synthetic values must not be more than a factor of 2 away from the actual values . because of these constraints , the range from which the values of @xmath126 , @xmath127 , @xmath122 , and @xmath124 parameters are picked is not relevant , as long as it not too restrictive .
in other words , the observational constraints eliminate cases with very large values of these parameters .
the data suggests that demagnified sne are not lost due to the flux limit . if they were , we would have less sne at small values of @xmath38 , and more at high values of @xmath38 , whereas in the data ( fig . [ figone ] )
the sources are roughly equally spread over the entire @xmath38 range .
so our synthetic lensing models assume that we do not lose sne because of deamplification . of the @xmath47 vs. @xmath44 relation ) given different mass distribution models , and a range of optical depths .
solid dots assume standard @xmath84 pdf of section [ stdmass ] , empty squares and star symbols represent bifocal i and ii models , respectively , discussed in section [ bifocalmass ] .
each point was obtained using 10,000 synthetic realizations of snia 50-source sample .
see section [ montecarlo ] for details.,width=317 ] for each synthetic realization of the 50-source snia sample we derive the corresponding @xmath47@xmath38 relation . the slope of the best - fit line , @xmath55 is recorded . for a given value of @xmath80
we generate 10,000 realizations .
solid dots in fig .
[ figoptdepth ] show the percentage of realizations that have @xmath55 smaller than the observed value of -0.3 , as a function of total optical depth , for the standard @xmath84 pdf .
empty squares and star symbols represent the results for the bifocal i and ii .
figure [ figoptdepth ] considers a range of optical depths .
what is the appropriate value for @xmath80 ?
if the source is at the median source redshift for our sample , @xmath68 , then apm galaxies sample the mass distribution up to a redshift of about 0.3 . in this case
the lensing optical depth probed by apm galaxies is 0.011 ( left - most arrow in fig .
[ figoptdepth ] ) . a somewhat more optimistic estimate for @xmath80
is obtained if we assume that galaxies trace the mass fluctuations up to the median source redshift , 0.47 , which gives @xmath128 ( middle arrow ) .
the limiting case is obtained by considering the most distant source , at @xmath129 , and assuming that galaxies probe mass up to a redshift of 0.47 , which gives @xmath130 ( right - most arrow ) . the important conclusion from this figure is that if the standard mass distribution ( section [ stdmass ] ) is assumed , then the value of @xmath80 does not really matter , and the probability of reproducing the observations is @xmath131 , if @xmath80 is within a reasonable range . if , on the other hand , the bifocal pdf i or ii are assumed , the probability of reproducing observations depends sensitively on the assumed optical depth . if @xmath130 then these models yield 6 - 20% probability .
the bifocal pdfs produce more discernible lensing effects , for the same @xmath80 because they have a wider range in @xmath118 , as seen in fig .
[ grid](d ) . @xmath44 relation on the three biasing parameters ( see section [ biasscheme ] ) .
short - dash lines are for the standard @xmath84 pdf model of section [ stdmass ] , while the solid lines are for the bifocal ii @xmath84 pdf described in section [ bifocalmass ] .
both models assume optical depth @xmath132 .
the contours are based on 50,000 realizations , and are spaced a factor of 3 apart .
the shapes of the contours are partly determined by the observational constraints , described in section [ montecarlo ] ( second paragraph ) .
it is evident that biasing parameters have little impact on the @xmath55 .
_ panel ( d ) : _ short - dash and solid lines represent @xmath84 pdfs corresponding to the contours in panels ( a)-(c ) of the same line type .
the minimum @xmath84 in the bifocal ii pdf is @xmath87 ( indicated by an arrow ) .
the long - dash line represents bifocal i @xmath84 pdf .
, width=317 ] the results presented in fig .
[ figoptdepth ] marginalize over values of all the model parameters ( except @xmath80 ) , so the effects of the biasing parameters on @xmath55 are hidden .
however , it turns out that these parameters do not correlate with @xmath55 .
figure [ grid ] shows the dependence of @xmath55 on @xmath126 , @xmath127 and @xmath121 for the bifocal ii @xmath84 pdf model ( solid contour lines ) , and the standard @xmath84 pdf model ( dashed contour lines ) , both for @xmath132 .
fifty thousand realizations were created for each model , and the corresponding contour lines plotted , with adjacent contours separated by a factor of 3 .
it is apparent that for both models the dependence on the biasing parameters is weak .
we conclude that _ the slope of the @xmath47@xmath44 relation depends on the amount and distribution of matter , i.e. on @xmath80 and the shape of the @xmath84 pdf model , but does not depend on the specifics of the biasing scheme .
_ figure [ figoptdepth ] quantifies the dependence on amount and distribution of mass : as expected , higher optical depth results in more pronounced lensing effects .
standard @xmath84 pdf produces less lensing for the same @xmath80 compared to a much broader bifocal pdf .
figure [ grid ] demonstrates that biasing has little effect on the slope of @xmath47@xmath44 relation . in other words , the type of lensing signature considered here ( @xmath47@xmath44 relation for standard candles ) probes mass distribution , and is independent of biasing .
we detect the signature of weak lensing in the current sample of 50 high redshift snia taken from two teams : high-@xmath0 supernova search team and supernova cosmology project . after correcting sne magnitudes for cosmological distances ( assuming @xcite values )
, we find that brighter sne are preferentially found behind regions overdense in foreground galaxies .
this @xmath47@xmath44 relation has a slope @xmath133 to @xmath134 , when the angular radii of foreground regions are 5 - 15 arcmin .
the statistical significance is @xmath2 ( see fig . [ figthetatest ] and [ figmagtest ] ) .
the angular radii of 5 - 15 arcmin imply that the lensing structures are @xmath135 mpc across if they are located at a redshift of 0.1 , and so correspond to non - linearly evolved intermediate - scale structure . aside from the possibility that the observed @xmath47@xmath44 relation is a fluke ,
there is one other possible , non - lensing interpretation : galactic dust obscuration , which would make snia sources brighter and apm galaxies more numerous in the directions devoid of galactic dust .
if this interpretation is correct , then ( 1 ) galactic extinction corrections applied to the sne @xcite are much too small , and ( 2 ) galactic dust is able to change projected apm galaxy number density by @xmath136 , since the observed rms in @xmath43 is @xmath137 .
we can not comment on ( 1 ) , but can estimate the effect of ( 2 ) .
how much excess / deficit in galaxy density can galactic dust create ? using @xcite data we calculate the average @xmath138 galactic extinction for the 50 sne sample to be 0.18 mag .
the slope of the r - band apm galaxy number counts is @xmath139 , therefore typical `` dust - induced '' @xmath140 .
this is an upper bound because extinction in the @xmath141 band will be smaller than in @xmath138 , and because extinction averaged over 5 - 15 arcmin radius patches will be smaller than estimates for individual sne , given that @xcite extinction map resolution is [email protected] fwhm . if galactic dust is indeed the cause of the @xmath47@xmath44 correlation , then @xmath47 and @xmath44 should separately correlate with sne extinction . in reality
, @xcite extinction estimates do not show a correlation with either @xmath47 or @xmath44 .
overall , galactic dust interpretation of the observed @xmath47@xmath44 relation seems unlikely , but can not be ruled out . note that dust _
intrinsic _ to the @xmath143 structures probed by the apm galaxies can not be invoked to explain the observations , because such dust would produce an effect opposite to the one detected here , i.e. @xmath55 would be positive .
if dust is present in groups and clusters traced by the apm , it will diminish the amplitude of the effect we detect .
presence of dust in groups was suggested by @xcite who found that faint qso candidates are anti - correlated with foreground groups .
however , such an anti - correlation can also be explained by weak lensing @xcite .
current observations indicate that groups and clusters do not contain significant amounts of dust @xcite .
if the lensing interpretation in correct , then the standard models of mass distribution have some difficulty in reproducing @xmath55 ; the observed value would be detected only in @xmath131 of the cases .
we investigate how @xmath55 is affected by the amount and distribution of mass along the light of sight to the sources , and galaxy biasing schemes .
we find that larger optical depths and broader @xmath84 pdf result in steeper @xmath55 slopes ( fig .
[ figoptdepth ] ) .
optical depth is a function of global geometry , and no realistic cosmological model can give @xmath144 , which is what would be required to comfortably explain the results .
broader @xmath84 pdfs mean that mass fluctuations are more extreme than the standard cosmological models allow , a scenario which is in apparant conflict with other means of determining mass fluctuations , like cosmic velocity flows and weak shear lensing .
we also find that biasing has little effect on @xmath55 ( fig .
[ grid ] ) .
this insensitivity to biasing is in contrast to weak lensing induced qso - galaxy and grb - galaxy ( anti-)correlations , where biasing could , at least partly explain the higher than expected amplitude of the effect @xcite .
we conclude that weak lensing of standard candles provides a cleaner probe of the mass distribution at @xmath145 , on @xmath146few mpc scales , then lensing induced angular correlations .
in fact , snia can provide the perfect means of measuring mass inhomogeneities using gravitational lensing .
a set of standard candles at known redshifts can be analyzed using the complementary techniques of weak magnification and weak shear lensing .
the advantage of magnification lensing over the more commonly used shear lensing is that with the former one can chose the redshift range of the lenses , whereas the latter yields the cumulative effect of lensing along the entire line of sight to the source .
therefore a large uniform set of intermediate redshift snia , such as the ones that would result from the snap mission and the lsst missionhome.html ] would be invaluable for the studies of mass clustering in the nearby universe .
in addition to detecting intermediate and high - redshift sne for the purposes of estimating the global cosmological parameters and the equation of state of the dark energy , snap will also measure weak shear lensing signature due to large scale structure . combining magnification data ( of the type considered in this paper ) and shear information @xcite for
a large set of sne will allow the study of mass distribution at @xmath147 with unprecedented accuracy and 3-dimensional spatial resolution .
the final issue we address is the impact of weak lensing on the determination of global cosmological parameters , @xmath148 and @xmath149 using snia standard candles . in principle
, weak lensing can affect the derived values of @xmath148 and @xmath149 , if ( 1 ) demagnified sne are preferentially lost from the sample due to faint flux cutoff , and/or ( 2 ) the @xmath84 pdf is asymmetric and the sn sample size is small ( see also @xcite ) .
if demagnified sne were lost from the sample then the distribution of sne in @xmath44 would be skewed in the direction of larger @xmath44 values .
if @xmath84 pdf is asymmetric , ( and it is , on scales considered here ) , then most sne in a small sample will be slightly demagnified compared to average , and will have their @xmath44 skewed in the direction of smaller values .
so both ( 1 ) and ( 2 ) would make the distribution of sne in @xmath44 uneven , and would to some extent cancel each other . in the present sample of 50
, the distribution of sne in @xmath44 is indistinguishable from uniform , so the average @xmath47 corresponds to the average @xmath44 , and hence lensing effects by @xmath146mpc - size structures probably did not bias the derived values of @xmath148 and @xmath149 .
the authors would like to thank peter garnavich and jason rhodes for their careful reading of the manuscript and valuable comments .
99 amanullah r. , mortsell e. , goobar a. 2003 , a&a , 397 , 819 benitez n. , martinez - gonzalez e. 1997 , apj , 477 , 27 benitez n. , riess a. , nugent p. , dickinson m. , chornock r. , filippenko a. 2002 , apj , 577 , l1 benitez n. , sanz j.l . , martinez - gonzalez e. 2001 , mnras , 320 , 241 boyle , b. j. , fong , r. & shanks , t. 1988 , mnras , 231 , 897 cen r. , ostriker j.p .
1993 , apj , 417 , 415 croom , s. m. , & shanks , t. 1999 , mnras , 307 , l17 dekel a. , lahav o. 1999 , apj , 520,24 ferguson , h.c .
1993 , mnras , 263 , 343 garnavich p.m. et al .
1998 , apj , 493 , l53 holz d.e .
1998 , apj , 506 , l1 holz d.e . , wald r.m .
1998 , phys .
d , 58 , 063501 irwin m. , maddox s.j . , mcmahon r.g .
1994 , spectrum , 2 , 14 .
see also http://www.ast.cam.ac.uk/~mike/apmcat/ jain b. , scranton r. , sheth r.k .
2003 , + preprint , http://xxx.lanl.gov/abs/astro-ph/0304203 jain b. , seljak u. , white s.d.m .
2000 , apj , 530 , 547 kovner i. 1991 , apj , 376 , 70 lewis g.f .
, ibata r.a .
2001 , mnras , 324 , l25 maddox s. j. , efstathiou g. , sutherland w.j .
1996 , mnras , 283 , 1227 metcalf r.b .
1999 , mnras , 305 , 746 moertsell e. , gunnarsson c. , goobar a. 2001 , apj , 561 , 106 nollenberg , j.g . ,
williams , l.l.r .
& maddox , s.j .
2003 , aj , 125 , 2927 refregier a. 2003 , preprint , + http://lanl.arxiv.org/abs/astro-ph/0307212 rhodes j. , refregier a. , massey r. & the snap collaboration , priprint , + http://lanl.arxiv.org/abs/astro-ph/0304417 rodrigues - williams , l.l . , &
hogan , c. 1994 , aj , 107 , 451 riess a. et al .
1998 , aj , 116 , 1009 schlegel d.j .
, finkbeiner d.p . , davis m. 19998 , apj , 500 , 525 tonry j. l. et al .
2003 , preprint , + http://lanl.arxiv.org/abs/astro-ph/0305008 perlmutter s. et al . 1999 ,
apj , 517 , 565 wambsganss j. , cen r. , xu g. , ostriker j.p .
1997 , apj , 475 , l81 williams l.l.r .
, frey n. 2003 , apj , 583 , 594 williams l.l.r . ,
irwin m. 1998 , mnras , 298 , 378 | the current sample of high - redshift supernova type ia , which combines results from two teams , high-@xmath0 supernova search team and supernova cosmology project , is analyzed for the effects of weak lensing . after correcting sne magnitudes for cosmological distances , assuming recently published , homogeneous distance and error estimates , we find that brighter sne are preferentially found behind regions ( 5 - 15 arcmin radius ) which are overdense in foreground , @xmath1 galaxies .
this is consistent with the interpretation that sne fluxes are magnified by foreground galaxy excess , and demagnified by foreground galaxy deficit , compared to a smooth universe case .
the difference between most magnified and most demagnified sne is about 0.3 - 0.4 mag .
the effect is significant at @xmath2 level .
simple modeling reveals that the slope of the relation between sn magnitude and foreground galaxy density depends on the amount and distribution of matter along the line of sight to the sources , but does not depend on the specifics of the galaxy biasing scheme .
[ firstpage ] gravitational lensing
supernovae : general |
in the current hierarchical cosmological paradigm , galaxy mergers and interactions are a major route to galaxy formation .
the strong tidal fields experienced during close encounters between gas rich systems trigger large
scale nuclear inflows that can ultimately start the activity of the supermassive black holes ( smbhs ) residing in the central regions of the interacting systems ( e.g. , * ? ? ?
the existence of multiple , simultaneously active , smbhs represents a key observational test for this evolutionary scenario and for our understanding of the processes regulating the quasar ( qso ) activity and the co evolution of smbhs with their host galaxies .
a number of studies demonstrated the connection between qsos and ultra luminous infrared galaxies that are mostly found in merging systems ( e.g. , * ? ? ? * ) and confirmed the signature of recent merger events in qso hosts ( e.g. , * ? ? ? * ) .
a noticeable example is the discovery of a spatially resolved binary qso ( projected separation @xmath3kpc ) clearly hosted by a galaxy merger @xcite .
the search for candidate binary systems of qsos up to redshift @xmath4 unveiled an enhanced qso clustering signature at small scale ( @xmath5kpc ) relative to the simple extrapolation of the larger scale two point correlation function ( e.g. , @xcite ; but see @xcite ) .
this can be interpreted in terms of interactions that trigger the qso activity ( e.g. , * ? ? ?
* ; * ? ? ?
alternatively , the small scale excess could be a simple manifestation of the clustering properties of the halos that host qsos ( e.g. , * ? ? ?
little is known about systems with more than two physically associated qsos which are expected to be even more elusive objects .
the only physical qso triplet reported so far is qqq j1432 - 0106 at @xmath6 observed by @xcite . at smaller projected separations and lower luminosities
, @xcite recently discovered a triple agn in the galaxy sdss j1027 + 1749 at @xmath7 , and @xcite serendipitously observed three low mass ( @xmath8 ) accreting black holes in a galaxy at @xmath9 . in this paper
we present the discovery of , the second triplet of qsos that is known up to date .
this is the first result of our ongoing systematic search for qso triplets among the photometric and spectroscopic database of the sloan digital sky survey ( sdss ; * ? ? ?
the manuscript is organised as follows . in section [ sec:2 ]
we describe the spectroscopic and photometric data collected in our study .
the possible interpretations are discussed in section [ sec:3 ] , and we draw our conclusions in section [ sec:4 ] . throughout this paper we consider a concordance cosmology with h@xmath10mpc@xmath11 , @xmath12 , and @xmath13 .
all the quoted magnitude are expressed in the ab standard photometric system @xcite .
qso multiplets observed at close projected separations are rare ( e.g. , * ? ? ? * ) .
large spectroscopic surveys often fail to detect close qso systems due to the fiber collision limits .
for example in sdss it is not possible to obtain the spectrum for both sources in a pair with separation @xmath14 within a single plate @xcite . to overcome this limitation
, we started a programme to search for close qso triplets taking advantage of the large photometric sample of @xcite .
three qsos are considered a candidate triplet if ( i ) at least one of them has spectroscopic redshift , ( ii ) the other two reside within 500kpc from it , and ( iii ) they have coincident ( within the uncertainties ) photometric redshift ( estimated with the procedure suggested by * ? ? ?
we selected 13 triple qso _ candidates _ for spectroscopic follow up , in the redshift range @xmath15 , and with an average projected separation of @xmath16kpc .
further details on the selection procedure will be provided in a forthcoming paper ( farina et al .
, in preparation ) . here
we present the spectroscopic observations obtained for the first followed up target among our candidates . in particular , we report our discovery of two qsos within @xmath17 from the spectroscopic qso sdss j151947.3 + 062753 ( hereafter , * ? ? ?
* ) and located at a similar redshift ( radial velocity difference ) . to provide a rough estimate of the expected number of such systems , we derived the qso three point correlation function ( @xmath18 ) from the amplitude of the two point correlation function ( @xmath19 ) : @xmath20 $ ] , where @xmath21 ( e.g. , * ? ? ? * ; * ? ? ?
assuming the values of the projected two point correlation function from @xcite and integrating the qso luminosity function presented by @xcite above the magnitude range spanned by the photometric selected qsos ( i.e. , from @xmath22 to @xmath23 ) we calculated that in a perfect survey , with no source of incompleteness , given a qso , the probability of finding two companions within @xmath24kpc is @xmath25
. out of the @xcite catalogue , which consists in @xmath26 spectroscopically confirmed qsos in the sdss data release 7 footprint , we thus expect that @xmath27 objects have two companions .
the mere observation of one ( or more ) systems of this kind would substantially strengthen the argument of small scale enhancement of qso clustering ( e.g. , * ? ? ?
in table [ tab : triple ] we list the properties of the qsos belonging to our newly discovered triplet , labelled , and , for comparison , of the components of the only other triple qso system known so far : qqq j1432 - 0106 . in the following paragraphs
we describe the procedures and the results of the analysis of the spectroscopic and photometric data collected on . [ cols="<,^,^,^,^,^,^,^,^,^ " , ] spectra of and were gathered with the eso faint object spectrograph and camera 2 ( efosc2 ; * ? ? ?
* ) mounted on the new technology telescope ( ntt ) in la silla ( chile ) .
we performed long slit spectroscopy on february 17 , 2012 , using a slit width of @xmath28 and setting the position angle to @xmath29 , so that both the qsos were observed simultaneously .
grism # 16 was used in order to continuously cover the wavelength range 6015 - 10320 with a spectral resolution @xmath30 as measured on the night sky emission lines .
five frames of 900s were acquired , for a total exposure time of 75min on source .
standard ` iraf ` tools were used in the data reduction ( bias subtraction , flat fielding , wavelength and flux calibration , spectra extraction ) . as flux calibrator
we observed the spectrophotometric standard star hiltner 600 .
typical residuals in the wavelength calibration are @xmath31 . for the analysis of the qso spectra we followed the procedure presented in @xcite and @xcite .
namely , we model the qso spectra with a superposition of : ( i ) a power law non thermal component ; ( ii ) the emission of the host galaxy ( adopting the elliptical galaxy model by * ? ? ?
* ) ; ( iii ) the multiplets ( from the template of @xcite and from our own original spectrum of izw001 ) , and ( iv ) the broad emission lines fitted with two gaussian curves with the same peak wavelength ( see * ? ? ?
the two ntt spectra and the one from sdss are shown in figure [ fig:1 ] .
a bright emission line , identified as , is observed in all the spectra at @xmath32 .
the line identification is supported by the lack of other bright emission lines in the observed spectra , and by a tentative detection of the iron multiplets around the line ( see our fit results in figure [ fig:1 ] ) .
these pieces of evidence firmly pin down the redshifts of the qsos to @xmath33 , @xmath34 and @xmath35 for a , b and c respectively , where uncertainties are given by the positional accuracy of the line centroids .
we gathered nir broad band images of the field using omega2000 @xcite at the 3.5 m telescope in calar alto ( spain ) .
observations were performed on march 8 , 2012 as director discretional time .
we observed a region of @xmath36 around the system in z , j , and h bands , i.e. , sampling the rest
frame u , g , and r at @xmath37 .
we adapted usual jittering observing strategies in order to effectively subtract night sky emission .
we collected @xmath38s frames in z ( total integration time : 45min ) , @xmath39s frames in j ( total integration time : 30min ) , and @xmath40s frames in h ( total integration time : 30min ) .
images were processed with our package of nir image processing based on standard ` iraf ` tasks .
we compute the astrometric solution using the ` astrometry.net ` software @xcite .
photometric calibration is achieved by comparing the photometry of field stars observed in our images with the fluxes reported in the sdss ( z band ) and 2mass ( j and h band ) catalogues , using the following filter transformations : @xmath41 equations [ eq_z][eq_h ] are derived by computing the spectral magnitudes of template o to m type stars .
the seeing during the observations was @xmath42 ( as measured on the final images ) , and the @xmath43 limit magnitudes ( computed from the rms of sky counts over a seeing area ) are 23.90 , 23.38 and 22.25 in z , j and h respectively .
we use ` sextractor ` @xcite in order to identify and catalogue the sources in our images . within the limit fluxes of our observations , we are able to study the environment of the qso system down to @xmath44 , where @xmath45 is the rest
band characteristic luminosity of galaxies at redshift @xmath46 as derived from the fit of the galaxy luminosity function @xcite . in order to distangle between galactic and extragalactic objects we rely upon different colour
colour diagrams , also including sources from the sdss photometric database ( see figure [ fig:2 ] ) .
we classified a source as _ star _ if it is not resolved ( i.e. , fwhm @xmath47 seeing in all the considered images ) , and it lies within the locus of main sequence stars estimated through the prediction of the ` trilegal ` software @xcite , and from the distribution of the sdss spectroscopically confirmed stars detected also in 2mass @xcite .
our colour cuts are a simplified version of those adopted by @xcite on the sdss photometric database .
we also check the consistency of the sources not classified as stars with the reference colours of elliptical , sc galaxies and qsos obtained from the templates of @xcite and of @xcite . in summary in the @xmath36 region explored we detect 2048 sources , 807 of which are classified as star on the basis of their colours .
this yields a number density of @xmath48 stars per arcmin@xmath49 that roughly correspond to the prediction of the ` trilegal ` software in this region of the sky ( @xmath50arcmin@xmath51 ) .
the average surface density of extragalactic sources is @xmath52arcmin@xmath51 , that is consistent ( within the uncertainties ) with the @xmath53arcmin@xmath51 observed in the the great observatories origins deep survey ( goods , * ? ? ?
* ) once we consider our sensitivity limits .
in this section we discuss the possible scenarios for this peculiar system . and filter corrections .
we assume that the potential well of the lensing galaxy is well described by a sis model with @xmath54 , where @xmath55 is the stellar velocity dispersion as set by the faber - jackson relation ( see text for details ) . in the lensing scenario ,
a galaxy of [email protected] would be required in order to justify the observed separation between and c. thanks to the depth of our observations ( z@xmath57 ) , we can rule out the strong lensing scenario for these two sources . ] to test the lens hypothesis for we consider two different configurations : in case ( i ) and are images of the same , gravitationally lensed qso while is a different qso at similar redshift . in case ( ii ) , also is an image of the same qso .
we note that small differences are reported in the colours and in the spectra of the three qsos ( e.g. , the lines peak at slightly different redshifts ) .
this already disfavours the lensing scenario . however , due to the limited signal to noise ratio of the available spectra
, we will focus our analysis mostly on the images . in case ( i )
we model the potential well of the lensing object with a singular isothermal sphere ( sis ) .
multiple images of similar brightness are obtained only along the einstein ring , the radius of which is parametrised as @xmath58 ( e.g. , * ? ? ?
* ; * ? ? ?
* ) : @xmath59 here @xmath60 is the velocity dispersion of the sis , @xmath61 is the speed of light , @xmath62 and @xmath63 are the angular diameter distances between the source and the lens , and between the source and the observer , respectively .
the minimum value of @xmath58 allowed in this system corresponds to half of the separation in the sky between and c , i.e. , @xmath64 .
this implies that , for a given redshift of the lens , we can infer a lower limit for @xmath60 . if we adopt the working assumption that @xmath54 ( the velocity dispersion of stars in the galaxy ) , we can use the faber
jackson relation @xcite to convert the lower limit on @xmath60 into a luminosity limit for the lens galaxy .
we use the updated version of the faber
jackson relation reported by @xcite , @xmath65 where @xmath66 is the rest - frame r - band absolute magnitude .
we convert the observed z band into @xmath66 using the filter and @xmath67corrections computed basing on the elliptical galaxy template by @xcite .
figure [ fig:3 ] shows that a galaxy brighter than z=19.6 is necessary to explain the observed geometry of the system in terms of strong gravitational lensing .
the presence of such a galaxy is ruled out by our omega2000 images .
case ( ii ) is more extreme since wide separation lens systems are rare . in their search for lensed qsos with angular separation less than @xmath68 ,
@xcite find in the sdss only 3 out of 22 systems with @xmath69 .
@xcite estimated that in the sdss qso sample there should be only a few systems ( @xmath70 ) multiply lensed by galaxy clusters with separations @xmath71 . moreover ,
triple imaged qsos are not commonly observed ( e.g. , * ? ? ?
the three largest separation qso lenses known so far , sdss j1029 + 2623 ( @xmath72 , @xmath73 , * ? ? ? * ) , sdss j2222 + 2745 ( @xmath74 , @xmath75 , * ? ? ?
* ) , and sdss j1004 + 4112 ( @xmath76 , @xmath77 , * ? ? ?
* ) , have separation similar to the ones reported here . in all these cases
massive galaxy clusters act as lens @xcite . as a zeroth order test
, we estimate the velocity dispersion required to explain the separation of the images assuming again a sis profile for the lens . as in case ( i )
we consider a limit for the einstein radius @xmath78 that corresponds to @xmath79at @xmath80 and @xmath81 at @xmath82 . in our image
the presence of such clusters should be apparent , but in fact the surface density of galaxies within @xmath83 from ( in z band @xmath84arcmin@xmath51 ) is consistent with those detected in the whole frame ( @xmath52arcmin@xmath51 , see [ sec:2.2 ] and figure [ fig:4 ] ) .
for instance , we consider the case that in our field was present xmmu j100750.5 + 125818 : a galaxy cluster located at redshift @xmath85 with a line of sight velocity dispersion of @xmath86 @xcite .
if the cluster was centred on in the first @xmath83 there should be at least 10 galaxies in excess to the background .
this is in contrast to what observed in the z band where we detect only 12 galaxies instead of 22 .
we note that this estimate is conservative since we considered as reference only the ( few ) spectroscopically confirmed cluster members of xmmu j100750.5 + 125818 , while the cluster should be more populated . from the analysis of photometric data ,
we thus conclude that is not a case of gravitation lens . around the triplet as imaged in z band at the calar alto observatory .
blue diamonds are the three qsos .
the 13 sources identified as galaxies are shown as red squares , while the 9 stars as green circles .
the surface densities of galactic and extragalactic sources in this area is consistent with the detections in the rest of the image . ]
is the second triplet of qsos at similar redshift known , after qqq j1432 - 0106 @xcite . in section [ sec:2 ]
we have already shown that these are extremely rare objects . here
we estimate the expected number of qso triplets in our sample that are originated by chance superposition of otherwise ( physically ) disconnected qsos .
we start from the samples of spectroscopically confirmed qsos by @xcite and the sample of photometrically selected qso candidates by @xcite ( both based on sdss data ) and we rely upon a procedure similar to the redshift permutation method ( e.g. , * ? ? ?
* ; * ? ? ?
we assume that the qsos have the same position in the sky as in the original samples but a random redshift is assigned through a monte carlo simulation that takes into account the redshift distribution of our complete sample ( @xmath87 objects ) .
we find that @xmath88 qso triplets with @xmath89 and projected separations @xmath90kpc are expected .
this result is an upper limit for the number of chance triple qsos since in our computation most of the correlation was destroyed , but we kept the angular distribution of the qsos as in the original sample . in the hypothesis
that the system is virialised , the velocity differences among , and ( estimated from the broad emission line redshifts ) imply a dynamical mass of @xmath91 , i.e. the system would lie inside a massive structure . in order to test this scenario ,
we search for galaxies at redshift @xmath92 in the proximity of the system . in figure [ fig:5 ]
we show the observed colour magnitude diagram of the sources in our field .
we compare the photometry of these sources with the apparent magnitude and colours of a @xmath45 galaxy , where @xmath45 is the characteristic luminosity ( in absolute magnitudes ) of galaxies at a given redshift @xmath46 in the rest
frame b band , as computed by @xcite . for filter and @xmath67correction
, we refer to the elliptical galaxy template by @xcite .
we do not find evidence of a strong overdensity of sources with colours consistent with those of a red sequence at @xmath93 .
this suggests that is not located in a rich galaxy cluster or that the red sequence is not yet formed .
we also estimate the photometric redshifts of the sources in the field of with ` hyperz ` @xcite relying upon different galaxy templates ( i.e. , starburst , elliptical , lenticular , different type of spirals , and irregular galaxies ) and assuming the extinction law by @xcite .
it is worth noting that this choice of dust extinction model has only marginal effects on our results .
this analysis confirms that none of the objects detected within @xmath83 ( that corresponds to @xmath94kpc at the triplet redshift ) and @xmath95 ( @xmath31mpc ) have colours consistent with a galaxy at @xmath96 ( see figure [ fig:6 ] ) .
we thus propose that the physical structure , where the three qsos reside , may be caught in the act of formation ( similarly to what reported in e.g. : * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
are shown in orange , and those that lie within @xmath83 from the triplet are highlighted with red squares .
the vertical dotted line marks the 5@xmath60 detection limit ( note that ` sextractor ` is on average more conservative than the statistical limit magnitude ) .
typical error bars are shown as a cross in the bottom
right of the figure .
the dashed line shows the location of an @xmath45 galaxy at various redshifts , as derived from @xcite , and assuming the elliptical galaxy template by @xcite for @xmath67corrections . for reference
, we show with light violet crosses the location of the galaxies of the galaxy cluster xmmxcs j2215.9 - 1738 ( @xmath97 , velocity dispersion @xmath98 , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
dark violet crosses highlight the spectroscopic confirmed associations .
we do not find any evidence of an overdensity of galaxies with colours consistent with a red sequence at @xmath99 , thus suggesting that the environment of is not a rich cluster . ] ) .
sources that reside within @xmath95 and @xmath83 from the triplet are shown in purple and red , respectively .
the vertical dashed line marks the quasar triplet redshift .
typical errors associated to the photometric redshift determination are @xmath100 .
none of the galaxies close to has a sed consistent with that of a galaxy at @xmath96 , confirming that there is no strong galaxy overdensity around the quasar triplet . ]
we report the discovery of a triplet of qsos with angular separations between @xmath101 and @xmath102 at a similar redshift of @xmath93 .
the three sources have relative velocities within @xmath103 as derived from our measurements of the broad emission line and of the surrounding multiplets in the observed optical spectra . from the analysis of u , g , r ,
i , z , j , and h broad band images we conclude that this system consist of three distinct sources rather than being a multiply imaged lensed qso
. this would be the second triplet of qsos discovered so far after qqq j1432 - 0106 @xcite .
we estimate that the probability that these systems are due to chance superposition are negligible .
no clear galaxy overdensity is reported at photometric redshift consistent with the one of the triplet .
we therefore propose that these systems are part of an ongoing common physical structure formation .
the projected separation of from the other two members of the system is greater than the typical distances between interacting systems ( @xmath104kpc ; e.g. , * ? ? ?
* ; * ? ? ?
this implies that the observed nuclear activity of was probably not triggered during an interaction involving the three systems at the same time . on the other hand , due to
the smaller separation observed between and c , the triggering of nuclear activity by galactic interactions may be a viable scenario for these qsos .
more conclusive constraints on the dynamics of this peculiar system and on the properties of its environment will be obtained from future higher signal to noise ratio spectroscopic data in the nir band and deeper photometric images .
we thank the calar alto staff for the generous allocation of ddt time and the prompt execution of our programme .
we acknowledge a. treves for helpful suggestions and comments on the manuscript .
we would also like to acknowledge the support and the advertising given by m. colpi , g. gavazzi , and f. haardt .
for this work epf was supported by societ carlo gavazzi s.p.a . and by thales alenia space italia s.p.a .
rd acknowledges funding from germany s national research centre for aeronautics and space ( dlr , project fkz 50 or 1104 ) .
support for mf was provided by nasa through hubble fellowship grant hf-51305.01-a awarded by the space telescope science institute , which is operated by the association of universities for research in astronomy , inc .
, for nasa , under contract nas 5 - 26555 . for this work we use : ( i ) spectra from eso / ntt telescope in la silla ; ( ii ) imaging from caha/3.5 m telescope ; and ( iii ) data from the sloan digital sky survey .
funding for sdss - iii has been provided by the alfred p. sloan foundation , the participating institutions , the national science foundation , and the u.s .
department of energy office of science .
the sdss - iii web site is ` h`ttp://www.sdss3.org/. sdss - iii is managed by the astrophysical research consortium for the participating institutions of the sdss - iii collaboration including the university of arizona , the brazilian participation group , brookhaven national laboratory , university of cambridge , university of florida , the french participation group , the german participation group , the instituto de astrofisica de canarias , the michigan state / notre dame / jina participation group , johns hopkins university , lawrence berkeley national laboratory , max planck institute for astrophysics , new mexico state university , new york university , ohio state university , pennsylvania state university , university of portsmouth , princeton university , the spanish participation group , university of tokyo , university of utah , vanderbilt university , university of virginia , university of washington , and yale university .
aihara h. , et al .
, 2011 , apjs , 193 , 29 bennert n. , canalizo g. , jungwiert b. , stockton a. , schweizer f. , peng c. , lacy m. , 2008 , iaus , 245 , 235 bertin e. , arnouts s. , 1996 , a&as , 117 , 393 bolzonella m. , miralles j .-
, pell r. , 2000 , a&a , 363 , 476 blanton m. r. , lin h. , lupton r. h. , maley f. m. , young n. , zehavi i. , loveday j. , 2003 , aj , 125 , 2276 buzzoni b. , delabre b. , dekker h. , dodorico s. , enard d. , focardi p. , gustafsson b. , nees w. , et al .
, 1984 , messanger , 38 , 9 calzetti d. , armus l. , bohlin r. c. , kinney a. l. , koornneef j. , storchi - bergmann t. , 2000 , apj , 533 , 682 canalizo g. , stockton a. , 2001 , apj , 555 , 719 chieregato m. , miranda m. , jetzer p. , 2007 ,
a&a , 474 , 777 croom s. m. , et al . , 2009 ,
mnras , 399 , 1755 dahle h. , et al .
, 2012 , arxiv , arxiv:1211.1091 decarli r. , labita m. , treves a. , falomo r. , 2008 , mnras , 387 , 1237 decarli r. , reynolds m. t. , dotti m. , 2009 , mnras , 397 , 458 decarli r. , falomo r. , treves a. , kotilainen j. k. , labita m. , scarpa r. , 2010b , mnras , 402 , 2441 decarli r. , falomo r. , treves a. , barattini m. , 2010a , a&a , 511 , a27 de rosa g. , decarli r. , walter f. , fan x. , jiang l. , kurk j. , pasquali a. , rix h. w. , 2011 , apj , 739 , 56 di matteo t. , springel v. , hernquist l. , 2005 , natur , 433 , 604 djorgovski s. g. , courbin f. , meylan g. , sluse d. , thompson d. , mahabal a. , glikman e. , 2007 , apj , 662 , l1 faber s. m. , jackson r. e. , 1976 , apj , 204 , 668 farina e. p. , falomo r. , treves a. , 2011 , mnras , 415 , 3163 foreman g. , volonteri m. , dotti m. , 2009 , apj , 693 , 1554 francis p. j. , hewett p. c. , foltz c. b. , chaffee f. h. , weymann r. j. , morris s. l. , 1991 , apj , 373 , 465 girardi l. , groenewegen m. a. t. , hatziminaoglou e. , da costa l. , 2005 , a&a , 436 , 895 giavalisco m. , et al . , 2004 , apj , 600 , l93 green p. j. , myers a. d. , barkhouse w. a. , mulchaey j. s. , bennert v. n. , cox t. j. , aldcroft t. l. , 2010 , apj , 710 , 1578 hennawi j. f. , et al . , 2006 , aj , 131 , 1 hennawi j. f. , dalal n. , bode p. , 2007
, apj , 654 , 93 hennawi j. f. , et al .
, 2010 , apj , 719 , 1672 hilton m. , et al . , 2007 , apj , 670 , 1000 hilton m. , et al . , 2009 , apj , 697 , 436 hopkins p.f .
, hernquist l. , cox t.j . , kere d. , 2008 , apj suppl . , 175 , 356 ilbert o. , tresse l. , zucca e. , bardelli s. , arnout s. , zamorani g. , pozzetti l. , bottini d. , et al .
, 2005 , a&a , 439 , 863 inada n. , et al . , 2003 , natur , 426 , 810 inada n. , et al .
, 2006 , apj , 653 , l97 inada n. , et al .
, 2008 , aj , 135 , 496 kayo i. , oguri m. , 2012 , mnras , 424 , 1363 kochanek c. s. , falco e. e. , muoz j. a. , 1999 , apj , 510 , 590 kovcs z. , mall u. , bizenberger p. , baumeister h. , rser h .- j . , 2004 ,
spie , 5499 , 432 lang d. , hogg d.w .
, mierle k. , blanton m. , roweis s. , 2010 , aj , 139 , 1782 lawrence c. r. , schneider d. p. , schmidt m. , bennett c. l. , hewitt j. n. , burke b. f. , turner e. l. , gunn j. e. , 1984 , sci , 223 , 46 liu x. , shen y. , strauss m. a. , 2011 , apj , 736 , l7 mannucci f. , basile f. , poggianti b. m. , cimatti a. , daddi e. , pozzetti l. , vanzi l. , 2001 , mnras , 326 , 745 myers a. d. , brunner r. j. , richards g. t. , nichol r. c. , schneider d. p. , bahcall n. a. , 2007 , apj , 658 , 99 myers a. d. , richards g. t. , brunner r. j. , schneider d. p. , strand n. e. , hall p. b. , blomquist j. a. , york d. g. , 2008 , apj , 678 , 635 narayan r. , schneider p. , 1990
, mnras , 243 , 192 nigoche - netro a. , aguerri j.a.l . , lagos p. , ruelas - mayorga a. , snchez l.j . , machado a. , 2010 , a&a , 516 , 96 oguri m. , et al . , 2004 , apj , 605 , 78 osmer p. s. , 1981 , apj , 247 , 762 oke j. b. , 1974 , apjs , 27 , 21 padmanabhan n. , white m. , norberg p. , porciani c. , 2009 , mnras , 397 , 1862 peebles p. j. e. , 1993 , ppc .. book , richards g. t. , et al . , 2002 , aj , 123 , 2945 richards g. t. , et al . , 2009 , apjs , 180 , 67 richardson j. , zheng z. , chatterjee s. , nagai d. , shen y. , 2012 , apj , 755 , 30 schawinski k. , urry m. , treister e. , simmons b. , natarajan p. , glikman e. , 2011 , apj , 743 , l37 schneider d. p. , et al .
, 2010 , aj , 139 , 2360 schwope a. d. , et al .
, 2010 , a&a , 513 , l10 shen y. , et al . , 2010 , apj , 719 , 1693 skrutskie m. f. , et al . , 2006 , aj , 131 , 1163 stanford s. a. , et al . , 2006 , apj , 646 , l13 vestergaard m. , wilkes b. j. , 2001 , apjs , 134 , 1 weinstein m. a. , et al . , 2004 ,
apjs , 155 , 243 zhdanov v. i. , surdej j. , 2001 , a&a , 372 , 1 | we present the discovery of a triplet of quasars at @xmath0 .
the whole system is well accommodated within @xmath1 ( i.e. , 200kpc in projected distance ) .
the velocity differences among the three objects ( as measured through the broad emission line ) are @xmath2 , suggesting that the quasars belong to the same physical structure . broad band nir images of the field do not reveal evidence of galaxies or galaxy clusters that could act as a gravitational lens , ruling out the possibility that two or all the three quasars are multiple images of a single , strongly lensed source . is the second triplet of quasars known up to date .
we estimate that these systems are extremely rare in terms of simple accidental superposition .
the lack of strong galaxy overdensity suggests that this peculiar system is harboured in the seeds of a yet to be formed massive structure .
+ based on observations collected at the la silla observatory with the new technology telescope ( ntt ) of the european southern observatory ( eso ) and at the calar alto observatory with the 3.5 m telescope of the centro astrnmico hispano alemn ( caha ) . [ firstpage ]
quasars : general |
hadrons ; su(3 ) multiplets ; mass quantization .
although the standard model is in excellent agreement with the rich experimental data , it leaves many questions unanswered .
the observed mass spectrum of the elementary particles is one of the fundamental questions that has so far defied any reasonable explanation in the standard model@xcite@xmath0 @xcite .
the distribution of the elementary particle masses is quite bizzare and is spread from a fraction of ev s for neutrinos to hundreds of gev s for the top quark .
apart from few patterns based on su(3 ) symmetry that led to the gell - mann okubo@xcite and coleman - glashow@xcite formulae interelating the masses of pseudoscalar mesons and octet baryons respectively , the standard model has not revealed any general order in the elementary mass spectrum . on the other hand
the associated physical mass is the best known and most fundamental characteristic property of the elementary particles , the recognition of empirical systematics and of any general regularities in the mass spectrum of these particles , irrespective of their nature or scheme of classication is of tremendous importance in understanding the intricacies of mass spectrum of elementary particles .
the lowest mass states , as a general rule in physical systems , are considered to be the building blocks of more complex systems and hence in some sense the most fundamental .
the most stable and least massive particles i.e. electron , muon and pion to which other particles decay after a transient existence , are the natural candidiates to search for a systematic regularity within the mass spectrum of elementary particles@xcite .
empirical and theoritical investigations based on experimental data reveal the electron , muon and the pion to serve as basic units for exploring the discrete nature of the mass distribution of elementary particles@xcite@xmath1@xcite . to search for an order
, we perform a specific systematic analysis of the mass spectrum of elementary particles and reveal that mass differences among particles when arranged in the ascending order of mass have a general tendency to be close integral / half integral multiple of mass difference between a neutral pion and a muon i.e. 29.318 mev .
the mass differences between unstable leptons and between baryons were shown to be quantized as integral multiples of this basic unit of 29.318@xcite . in the present study
, we evaluate the applicability of this result to the su(3 ) hadron multiplets and to neutral hadrons .
we reveal that mass unit of about 29.318 mev is a determining factor for the distribution of mass of elementary particles by showing that 29.318 mev integral multiplicity of mass differences to be valid for these hadrons .
this reinforces our earlier result that elementary particles do not occur randomly and are linked through the mass difference between first two massive elementary particles i.e. a muon and a neutral pion .
the database for the present study is the latest version of the particle data group listings@xcite . here
we investigate relationship the pion - muon mass difference has with the mass structure of the 1 ) hadrons which are classified into multiplets on the basis of su(3 ) symmetry and 2 ) neutral hadrons .
the masses of the baryon octet members with spin j and parity p such that @xmath2=@xmath3@xmath1 are : @xmath4=938.27203 mev , @xmath5=939.56536 mev , @xmath6=1115.683 mev , @xmath7=1189.37 mev , @xmath8=1197.449 , @xmath9=1314.83 mev and @xmath10=1321.31 mev .
the successive mass differences are tabulated in column 1 of table 1 with numerical value in mevs given in column 2 .
the small mass differences between the different members of an isospin charge multiplet are known to arise from the electromagnetic interaction@xcite .
however , the masses of the members of different isospin multiplets differ considerably .
column 4 shows the integral multiples of 29.318 mev that are close to the observed mass difference between successive members of the octet .
the integers being shown in column 3 .
the deviations of the observed value from the closest integral multiple of 29.318 mev are given in column 5 .
it is observed that the mass difference between @xmath11 and @xmath12 i.e. 176.118 mev differs from the nearest predicted value of 175.908 mev by only 0.21 mev .
same is true of the mass difference i.e. 117.381 mev between the particles @xmath13 and @xmath14 which differs from the predicted value of 117.272 mev by only 0.109 mev .
however , observed mass interval of @xmath15 and @xmath11 differs from the predicted value by about 14.264 mev .
interestingly , this large value turns out to be half integral multiple of the mass difference between a @xmath16 and a @xmath17 .
as can be clearly seen from the row 3 of table 1 , the observed mass difference between @xmath15 and @xmath11 i.e. 73.69 differs from the half integral ( @xmath18 ) multiple of pion and muon mass difference by only 0.39 mev .
the maximum mass splitting within the baryon octet i.e. mass difference of 383.037 mev , between the heaviest member @xmath19 and the lightest baryon @xmath20 is close integral multiple of 29.318 mev , differing from the predicted value by only 1.904 mev .
it may be pointed out that 29.318 mev multiplicity also holds for the mass intervals among any of the octet members@xcite .
clearly the 29.318 mev multiplicity holds with great precision for the baryon octet members .
the analysis for the baryon decuplet members with @xmath2=@xmath21@xmath1 is detailed in table 2 .
it may be pointed out that while all the members of the baryon octet are non - resonant states , for the baryon decuplet all the members execpt for the @xmath22 are resonances . since the particle data group reports an average mass for the four charged states of the @xmath23 baryons and individual masses for the different charge states of @xmath24 and @xmath25 baryons , we cosider the average masses of each isospin multiplet which are as follows : @xmath26=1232 mev , @xmath27=1384.56 mev , @xmath28=1533.4 mev , and @xmath29=1672.45 mev
. the first three mass differences in the table 2 are those between the successive members of the decuplet .
the equal spacing rule for the su(3 ) decuplet predicts masses of successive isospin multiplets to be equidistant@xcite i.e. @xmath27 - @xmath26 = @xmath28 - @xmath27 = @xmath29 - @xmath28 .
however , this rule is not strictly obeyed in the decuplet , since the mass separations are not exactly same as evident from table 2 .
further , although the mass spacing among the successive decuplet members deviate from the closest integral ( 5 ) multiples of 29.318 mev by 5.97 , 2.25 and 7.54 mev respectively , it is important to note that the average mass spacing among successive members i.e. 146.816 mev is very close to 146.59 mev , a value obtained on integral ( 5 ) multiplication of 29.318 mev .
the difference between the observed and predicted values being 0.226 mev only .
this may be compared with 140 mev pionic mass interval@xcite among the decuplet members which deviates from the average mass spacing by about 6.816 mev@xcite .
the @xmath22 - @xmath23 is the difference between the mass of lightest resonance member @xmath23 and the heaviest non - resonant member @xmath22 of the decuplet .
this observed mass interval of 440.45 mev is very close to 439.77 mev , obtained on integral ( 15 ) multiplication of the mass difference between a neutral pion and a muon .
the difference between the observed and expected value being only 0.68 mev .
the mesons with spin zero and odd parity i.e. @xmath2 = @xmath30 are organized into a multiplet contaning nine states to form pseudoscalar meson nonet .
these mesons have the lowest rest energy . in column 1 of table 3 ,
the k@xmath31 - @xmath32 , @xmath33 - k@xmath31 and @xmath34 - @xmath33 are the mass difference between the successive members of the pseudoscalar meson nonet with numerical values given in column 2 .
the deviation of 0.182 mev between the observed and predicted @xmath34 - @xmath33 mass difference may be compared with that of 0.730 mev obtained from the integral multiple ( 3 ) of the 137 mev@xcite .
both the observed mass intervals of @xmath33 - @xmath16 and @xmath33 - @xmath32 deviate from the 14 multiples of 29.318 mev by about 2 mev .
however the difference 410.2364 mev between the mass of @xmath33 and average pion mass of 137.27339 mev differs by only 0.215 mev from predicted value 410.452 obtained by the multiplication of 29.318 mev by integer 14 . again
this departure of the observed mass difference between @xmath33 and average pion mass may be comapared with the difference of 0.736 mev bewteen the observed and that expected when mass intervals are taken as integral multiples of average pion mass i.e. 137 mev@xcite .
the observed mass interval between the @xmath34 and average pion mass is 820.506 mev .
this value deviates from the predicted value of 820.904 mev obtained from the integral multiple ( 28 ) of 29.318 mev by only 0.3973 mev only . whereas the predicted value 822 mev on the basis of integral pion mass differs from the observed value by 1.4933 mev@xcite .
the difference @xmath34 - @xmath16 is the mass difference between the lightest and heaviest member of the pseudoscalar meson nonet .
as can be seen from the table 3 , the observed mass spacings are close integral multiples of the mass difference between a neutral pion and a muon .
the masses for the pseudoscalar mesons taken from the particle data group listings are : @xmath35=134.9766 mev , @xmath36=139.57018 mev , @xmath37=493.677 mev , @xmath38= 497.648 mev , @xmath39=547.51 mev , @xmath40=957.78 mev . the nine vector mesons with spin one and odd parity i.e. @xmath2 = @xmath41 form a multiplet called vector meson nonet .
the analysis for the mass differences among the vector meson nonet members is detailed in table 4 .
the mass difference between successive members ( column 1 ) of the vector meson nonet is given in the column 2 .
as is evident from the table 4 , the observed mass differences k@xmath42 - @xmath43 and @xmath44 - @xmath45 are close integral multiples of 29.318 mev .
although the mass difference between isospin triplet @xmath43 meson and isospin singlet @xmath45 meson 7.15 mev is nonelectromagnetic in origin but is of the order of electromagnetic mass splitting .
the deviation of 2.226 mev of the observed @xmath44 - @xmath45 mass difference from the integral multiple of 29.318 mev is in better agreement than that of 8.19 mev obtained when mass interval is predicted as half integral multiple of a mass unit of about 70 mev i.e about half the pion mass@xcite .
since the particle data group lists average mass for the isospin triplet @xmath43 states and separate masses for two charge states of k@xmath42 meson , we consider the average mass of vector mesons for the analysis which are : @xmath46=893.83 mev , @xmath47=775.5 mev , @xmath48=782.65 mev and @xmath49=1019.460 mev .
the applicabilty of 29.318 mev multiplicity of elementary particle mass intervals extends beyond the mass differences between the successive members of a particular su(3 ) multiplet and applies to mass intervals among the members of multiplets with different spin and parity characteristics .
the analysis for the mass intervals between the octet and decuplet baryons is detailed in table 5 .
the lightest member of decuplet i.e @xmath23(1232 ) resonance is the first excited state of the proton and the observed mass interval between the two is accounted by the hyperfine splitting due to colour magnetic interaction among the quarks@xcite .
however , from table 5 it is seen that the observed mass interval of 293.728 mev between @xmath23(1232 ) and proton is very close to 293.180 mev , a value obtained on integral multiplication of 29.318 by 10 .
the difference between the two values being only 0.548 mev .
the relation of @xmath23(1232 ) and proton is expected as @xmath23 baryons are considered to be the excited states of nucleon but there is no relation between @xmath23(1232 ) and @xmath11 .
however , from our analysis it follows that the observed mass interval between the two 116.317 mev differs from 117.272 mev , the closest integral ( 4 ) multiple of 29.318 mev by 0.955 mev only .
thus @xmath23(1232 ) can be obtained by taking four excitations of 29.318 mev from the @xmath11 .
similarly the observed mass inteval of 556.767 bewteen two unrelated baryons @xmath22 and @xmath11 deviates from 557.042 mev , a value obtained as 19 multiples of pion - muon mass difference by only 0.275 mev .
further , the observed mass difference 351.14 mev between the heaviest member of baryon octet @xmath19 and that of baryon decuplet i.e. @xmath22 the only non - resonant member , differs from 351.816 mev , a value obtained on integral ( 12 ) multiplication of 29.318 mev , by only 0.676 mev .
the detailed anlysis of the mass differences among the pseudoscalar and vector mesons is given in table 6 .
as seen from the table 6 the 29.318 mev mass interval multiplicity is valid for mesons also .
the neutral hadrons are placed in the column 1 of the table 7 in the ascending order of their associated physical mass . since no bound system of top quarks has been detected , the neutral hadrons listed in table 7 span the quark structure from the light u , d , s quarks to the heaviest charm and bottom quarks .
the masses of the heavier mesons are : @xmath50 = 3096.916 mev , @xmath51 = 3686.093 mev , @xmath52 = 9460.30 mev .
the differences amongst mesons @xmath16 , @xmath33 , @xmath34 and @xmath45 reported in tables 3 and 6 are reproduced here for completeness .
the @xmath53 is the only baryon in the list . in column 2
we give the mass differences between the successive particles .
the integral ( column 3 ) multiples of 29.318 mev , the mass difference between a neutral pion and a muon , closest to the observed mass differences are tabulated in column 4 and in column 5 we give the deviation of the observed value from the predicted value . as evident from the table
the agreement of the predicted values as integral multiples of pion muon mass difference with the observed values is extraordinary .
the observed mass difference between @xmath53 and @xmath34 is 234.862 mev .
this value is very close to 234.544 mev obtained on multiplication of 29.318 mev by integer 8 .
the difference between the two values being only 0.318 mev .
similarly 1904.274 mev , the observed difference between the mass of j/@xmath54 and that of @xmath53 deviates from 29 multiples of 29.318 mev i.e. 1905.67 mev by only 1.396 mev .
the j/@xmath54 is the lowest - lying member of the charm - anticharm meson series and @xmath55 is its first excited state . from table 7
it can be seen that the mass difference between @xmath55 and j/@xmath54 is close integral multiple of the mass difference between a neutral pion and muon .
similarly @xmath56 is the first member of the bottom - antibottom meson series . from table 13
, it is seen that the observed mass difference between the @xmath56 and j/@xmath54 is 6363.384 which is very close to 217 multiples of 29.318 mev i.e. 6362.006 .
the difference between the two values being only 1.378 mev .
the fact that the mass differences of the lightest hadron @xmath16 with the heaviest mesons j/@xmath54 , @xmath55 and @xmath56 are close integral multiples of the mass difference between a neutral pion and muon indicates that the 29.318 mev mass systematics is valid over the wider range of the mass spectrum . in tables 8 to 13 we detail the analysis for the mass intervals of each neutral hadron with respect to the remaining hadrons .
the mass intervals for each neutral hadron already listed in table 7 have been excluded in the analysis detailed in 8 - 13 . from tables 1 to 13 it is clear that the hadron mass intervals fall into two classes 1 ) those integral multiples of the mass difference between a neutral pion and a muon and 2 ) those in which the difference between the observed and predicted values are large but turn out to be exact half integral multiples of 29.318 mev .
the inability of the standard model to explain the observed spectra of mass of elementary particle has led to the development of models and theories for particle masses based on direct analysis of the elementary particle data to understand different intricacies of the mass spectrum .
the recognition of the pionic mass intervals among the particles@xcite combined with the fact that ratios of life times of elementary particles scale as powers of @xmath57 , the fine structue constant , indicated the fundamental role of @xmath57 in the generation of mass of elementary particles and that its domain is not restricted to leptons only , but also extends to include hadrons@xcite .
the statistical analysis of masses of elementary particles when grouped according to their quark composition and @xmath58 quantum numbers reveals that the masses are quantized as integral multiples of a mass unit of about 35 mev .
this 35 mev mass quantization combined with the analysis of the distribution of particle lifetimes as a function of mass reveals a shell structure of hadrons@xcite .
a standing wave model assuming the particles to be held together in a cubic nuclear lattice has been proposed to explain empirical fact that meson and baryon masses are integral multiples of the neutral pion@xcite . from the preceeding discussion ,
it is clear that the understanding of any pattern in a physical system is a driving force for the development of a fundamental theory .
the development proceeds from the experimental measurements of the phenomena , first to a recognition of regularities amongst the measurements , then to the physical insight which gives some understanding of these regularities and finally to a fundamental theory , which allows the totality of the phenomena to be understood from a few general principles@xcite .
the failure to understand the regularities within a reasonable data lays the foundation for the neccessity of the underlying theory .
the above mentioned studies are among many other attempts dedicated to pin down the fundamental relation of least massive and most stable particles , pion and muon , to the elementary particle mass spectrum pointed by their presence in the decay products of almost all the heavier elemenatry particle mass states@xcite@xmath1@xcite .
the mass intervals among elementary particles have been shown to be integral multiples of the pion mass@xcite and some sequances of particles are also repoted with muonic mass intervals@xcite . the baryon and meoson masses are reported to be integral / half integral multiples of the pion mass@xcite@xmath1@xcite .
the mass unit of 35 mev equal to one - third of muon mass and one - fourth of the pion mass serves as basic unit for the quantization of the elementary particle masses@xcite . on the other hand we look at the elementary particle mass data from a different prespective and
reveal the fundamental nautre of mass difference between a neutral pion and a muon to the elementary particle mass spectrum .
the electron and muon mass are the first two levels in the mass spectrum .
indeed the muon is treated to be an excited state of the electron@xcite@xmath0 @xcite@xmath0 @xcite .
however , the mass of electron being much smaller than the muon mass , the mass interval between the two is obscured by the large mass of muon and is thus equal to muon mass itself .
hence we regard 29.318 mev , the mass interval between neutral pion ( a hadron ) and muon ( a lepton ) to be the first excitation within elementary particle mass spectrum .
we investigate whether the mass excitations among the elementary particles are related to the first excitation .
we have satisfactorily shown that mass differences among the elementary particles in general when arranged in the ascending order of mass have a striking tendency to be integral / half integral multiples of 29.318 mev@xcite .
same is shown to be true for the mass differences between any of the baryons and mass interval between the unstable charged leptons . from the present study
, it follows that the `` fine structure '' of mass i.e. mass differences betweend different isospin members of a given hadron multiplet are close integral multiples of the mass difference between a neutral pion and a muon .
this is also true for the mass separation of the lightest and heaviest member .
the mass intervals between the members of different su(3 ) multiplets , i.e. between octet and decuplet baryons and between the pseudoscalar and vector mesons are also integral multiples of the mass unit of 29.318 mev .
the 29.318 mev discretness of the mass intervals is applicable to elementary particles irrespective of their classification into leptons , mesons and baryons on the basis of their structure , associated quantum numbers and type of interaction .
it covers a wide range of mass spectrum , a fact borne out by the precise agreement of the observed and predicted mass intervals for the heavy charm - anticharm and bottom - antibottom mesons .
although the present findings are not in line with the current theories , yet the large scale agreement between the observed mass differences and those calculated as integral multiples of the mass difference between a neutral pion and a muon clearly rules out it to be a mere coincidence . the important intriguing result of the present study is that the mass difference between the lightest composite particle pion and the second lightest lepton muon interlinks the masses of both leptons and hadrons .
clearly the occurence of elementary particle states is not random but seems to follow a definite order such that the mass excitations from one particle to other are always in units of 29.318 mev i.e. the mass difference between a neutral pion and a muon .
the mass differences among the baryon octet members have been found to be close integral multiples of the mass difference between a neutral pion and a muon .
this also holds true for the average mass separation among the successive members of the baryon decuplet , for the mass intervals among the pseudoscalar meson nonet and among the vector meson nonet members .
we also pointed out mass intervals among neutral hadrons to be integral multiples of mass difference between a neutral pion and a muon .
our study reveals that inter particle mass excitations are quantized as integral multiples of 29.318 mev thereby indicating the fundamental nature of this mass quantum for the elementary particle mass spectrum .
00 v. christianto and f. smarandache , progress in physics , 4 , 112 ( 2007 ) h. fritzsch , arxiv:0802.0099 , hep - ph/0201198 s. okubo , prog .
27 , 949 ( 1962 ) s. coleman and s. l. glashow , phys .
review lett . 6 , ( 1961 ) m. h. mac gregor , il nuovo cimento .
a103 , 983 ( 1990 ) g. n. shah and t. a. mir , hep - ph/0702140 , mod .
a23 , 53 ( 2008 ) y. nambu , prog .
phys . 7 , 595 ( 1952 ) r. m. sternheimer , phys .
review lett .
10 , 7 ( 1963 ) , 13 , 1 ( 1964 ) , 13 , 10 ( 1964 ) r. m. sternheimer , phys .
b2 , 138 ( 1965 ) , 143 , 4 ( 1966 ) r. m. sternheimer , phys .
review , 143 , 1262 ( 1966 ) , 5 , 170 ( 1968 ) f. r. tangherlini , prog .
th . phys .
58 , 2002 ( 1977 ) b. g. sidharth , physics/0309037 , physics/0306010 m. h. mac gregor , il nuovo cimento .
a58 , 159 ( 1980 ) m. h. mac gregor , int
a20 , 719 ( 2005 ) p. palazzi , int . j. mod
a22 , 546 ( 2007 ) , ( url : http://www.particlez.org ) e. l. koschmieder , physics/0602037 , physics/0002179 r. frosch , nuovo cimento .
a104 , 913 ( 1991 ) a. o. barut , phys .
b73 , 310 ( 1978 ) , phys .
42 , 1251 ( 1979 ) w. a. rodrigues and j. vaz , hep - th/9607231 d. akers , hep - ph/0310239 , hep - ph/0311031 , hep - ph/0303139 , hep - ph/0303025 w.- m. yao et al , ( particle data group ) , j. phys .
g. 33 , 1 , ( 2006 ) ( url : http://pdg.lbl.gov ) d. h. perkins , introduction to high energy physics , pp .
126 - 129 , 4th edn .
( cambridge university press , 2000 ) y. oh , h. kim and s. h. lee , hep - ph/0310117 l. j. hall , hep - ph/9303217 | the study on the linkage of elementary particle mass differences with pion - muon mass difference is explored further . in the present study
we show this linkage to be equally true for the mass differences amongst the members of su(3 ) hadron multiplets and the hadrons belonging to multiplets having different spin and parity characteristics .
this reinforces our contention that inter particle mass excitations are quantized as integral multiples of basic mass unit of 29.318 mev and thereby indicates the fundamental nature of pion - muon mass difference in the elementary particle mass distribution . |
the study of the antiproton
nucleus interaction is an interesting issue which has attracted renewed interest in recent years at the prospect of future experiments at the fair facility .
the @xmath0nuclear bound states and the possibility of their formation have been studied in refs .
these considerations are supported by a strongly attractive potential that the @xmath0 feels in the nuclear medium . within the rmf approach the real part of the @xmath0nucleus potential derived using the g - parity transformation is re@xmath1 mev deep at normal nuclear density . however , the experiments with @xmath0 atoms @xcite and @xmath0 scattering off nuclei at low energies @xcite favor shallower real part of the @xmath0nucleus potential in the range of @xmath2 mev in the nuclear interior .
an important aspect of the @xmath0nucleus interaction is @xmath0 annihilation which appears to be the dominant part of the interaction .
nevertheless , the phase space for the annihilation products should be significantly suppressed for the antiproton bound deeply in the nuclear medium , which could lead to the relatively long living @xmath0 inside the nucleus @xcite . in this contribution
, we report on our recent fully self - consistent calculations of @xmath0nuclear bound states including @xmath0 absorption in a nucleus .
the calculations are performed within the rmf model @xcite .
dynamical effects in the nuclear core caused by the antiproton and the phase space suppression for the @xmath0 annihilation products are studied for various nuclei . in section 2
, we briefly introduce the underlying model .
few selected representative results of our calculations are discussed in section 3 .
the @xmath0nucleus interaction is described within the rmf approach . the interaction among ( anti)nucleons
is mediated by the exchange of the scalar ( @xmath3 ) and vector ( @xmath4 , @xmath5 ) meson fields , and the massless photon field @xmath6 .
the standard lagrangian density @xmath7 for nucleonic sector is extended by the lagrangian density @xmath8 describing the antiproton interaction with the nuclear medium ( see ref .
@xcite for details ) .
the variational principle yields the equations of motion for the hadron fields involved .
the dirac equations for nucleons and antiproton read : @xmath9\psi_j^{\alpha}=\epsilon_j^{\alpha } \psi_j^{\alpha } , \quad j = n,\bar{p}~,\ ] ] where @xmath10 are the scalar and vector potentials . here , @xmath11 stands for ( anti)nucleon mass ; @xmath12 , and @xmath13 are the ( anti)nucleon couplings to corresponding fields , and @xmath14 denotes single particle states . the klein
gordon equations for the boson fields acquire additional source terms due to the presence of @xmath0 : @xmath15 where @xmath16 and @xmath17 are the scalar , vector , isovector , and charge densities , respectively , and @xmath18 are the masses of the considered mesons . in this work , the nucleon meson coupling constants and meson masses were adopted from the nonlinear rmf model tm1(2 ) @xcite for heavy ( light ) nuclei .
the system of the coupled dirac and klein gordon equations is solved fully self - consistently by iterative procedure .
+ in the rmf model , the nucleon in a nucleus moves in mean fields created by all nucleons , i. e. , the nucleon feels repulsion as well as attraction also from itself . in ordinary nuclei
this nucleon self - interaction has only a minor ( @xmath19 ) effect .
however , the potential acting on @xmath0 in a nucleus is much deeper and the impact of the @xmath0 self - interaction could become pronounced . in order to exclude this unphysical @xmath0 self - interaction we omitted the antiproton source terms in the klein
gordon equations for the boson fields acting on the @xmath0 .
the @xmath0nucleus interaction is constructed from the @xmath20nucleus interaction with the help of the g - parity transformation : the vector potential generated by the @xmath21 meson exchange thus changes its sign and becomes attractive . as a consequence
, the total @xmath0 potential will be strongly attractive .
however , the g - parity transformation should be regarded as a mere starting point to determine the @xmath0meson coupling constants .
various many - body effects , as well as the presence of strong annihilation channels could cause significant deviations from the g - parity values in the nuclear medium . therefore ,
we introduce a scaling factor @xmath22 for the @xmath0meson coupling constants @xcite : @xmath23 the @xmath0 annihilation in the nuclear medium is described by the imaginary part of the optical potential in a ` @xmath24 ' form adopted from optical model phenomenology @xcite : @xmath25 where @xmath26 is the @xmath0nucleus reduced mass . while the density @xmath27 was treated as a dynamical quantity evaluated within the rmf model , the parameter im@xmath28 fm was determined by fitting the @xmath0 atom data @xcite .
the energy available for the @xmath0 annihilation in the nuclear medium is usually expressed as @xmath29 , where @xmath30 and @xmath31 is the @xmath0 and nucleon binding energy , respectively .
the phase space available for the annihilation products is thus considerably suppressed for the deeply bound antiproton . the nuclear core density distribution in @xmath32ca and @xmath32ca@xmath33 for @xmath34 , calculated in the tm1 model . the @xmath0 density distribution in @xmath32ca@xmath33 ( dotted line )
is shown for comparison . ] the nuclear core density distribution in @xmath32ca and @xmath32ca@xmath33 for @xmath34 , calculated in the tm1 model .
the @xmath0 density distribution in @xmath32ca@xmath33 ( dotted line ) is shown for comparison . ]
the phase space suppression factor @xmath35 for two body decay is given by @xcite @xmath36[s-(m_1-m_2)^2]}{[m^2-(m_1+m_2)][m^2-(m_1-m_2)^2]}}\theta(\sqrt{s}-m_1-m_2)~,\ ] ] where @xmath37 , @xmath38 are the masses of the annihilation products and @xmath39 . for channels containing more than 2 particles in the final state
the @xmath35 was evaluated with the help of the monte carlo simulation tool pluto @xcite .
the formalism introduced above was employed in the self - consistent calculations of the @xmath0 bound states in selected nuclei .
first , we did not consider the @xmath0 annihilation in the nuclear medium and focused on the study of the dynamical effects caused by the presence of the antiproton in the nucleus . during our calculations we noticed a pronounced effect of the @xmath0 self - interaction on the calculated observables . in fig .
[ fig . : selfinteraction ] , we present the @xmath0 density distribution in @xmath40pb , calculated dynamically in the tm1 model for different values of the scaling parameter @xmath41 . the central @xmath0 density @xmath42 calculated including the @xmath0 self - interaction ( left panel ) reaches its maximum for @xmath43 and then starts to decrease .
it is due to the interplay between the negative value of @xmath44 ( absolute value of which increases with @xmath41 ) , the @xmath0 single particle energy , and the @xmath0 rest mass , which affects the solution of the dirac equation for the @xmath0 wave function . on the other hand , when the @xmath0 self - interaction is subtracted ( right panel ) , the scalar and vector potentials are of comparable depth and the @xmath42 increases gradually with @xmath41 and saturates at much higher values .
it is to be stressed that the effect of the @xmath0 self - interaction is negligible for @xmath45 , which includes the @xmath0 potentials consistent with @xmath0 atom data .
our calculations revealed large polarization of the nuclear core caused by the @xmath0 in the nuclear @xmath46 state .
the nuclear core density in a @xmath0 nucleus reaches @xmath47 times the nuclear matter density as illustrated for @xmath32ca in fig .
[ fig . : density ] .
the @xmath0 is localized in the center of the nucleus and the resultant nuclear core density is substantially enhanced over a small region , @xmath48 fm .
the nucleon single particle energies are affected by the presence of the @xmath0 as well .
consequently the total binding energies of @xmath0 nuclei increase considerably , @xmath49 mev for @xmath32ca and @xmath50 mev for @xmath32ca+@xmath0 ( @xmath34 ) .
upper components @xmath51 ( left ) and the relation for lower components @xmath52 ( right ) of the @xmath53 and @xmath54 @xmath0 wave functions in @xmath55o , calculated dynamically in the tm2 model using the real ( no abs ) and complex ( + abs ) @xmath0 potential . ] upper components @xmath51 ( left ) and the relation for lower components @xmath52 ( right ) of the @xmath53 and @xmath54 @xmath0 wave functions in @xmath55o , calculated dynamically in the tm2 model using the real ( no abs ) and complex ( + abs ) @xmath0 potential . ] in order to account for @xmath0 annihilation , we performed calculations using the complex potential presented in section 2 .
we considered the suppression of the phase space for the annihilation products . in fig .
[ fig . : sf ] , the phase space suppression factors for the annihilation channels involved are presented as a function of @xmath56 . as @xmath56 decreases due to the @xmath0 and @xmath57 binding energies many channels become strongly suppressed or even closed .
moreover , the @xmath0nucleon annihilation takes place in the nuclear medium .
therefore , the momentum dependent term in the mandelstam variable @xmath58 , where @xmath59 , is non - negligible in contrast to the two body frame @xcite . our self - consistent evaluation of @xmath56 including @xmath60 and @xmath61 leads to an additional downward energy shift overlooked by many previous calculations .
in table [ tab . : results ] , we present the @xmath46 @xmath0 single particle energies and widths in @xmath55o+@xmath0 , calculated using the real and complex potentials consistent with @xmath0-atom data ( @xmath34 ) . to illustrate the role of the suppression factors @xmath62 we show the results of calculations without @xmath62 ( ` complex ' ) , as well as including @xmath62 for @xmath56 due to @xmath30 and @xmath31 ( ` complex+@xmath62 ' ) and for @xmath56 with the additional downward energy shift due to the momenta of annihilating partners ( ` + @xmath63 ' ) .
the static calculations , which do not account for the core polarization effects , give approximately the same values of the @xmath0 single particle energy for all cases .
the single particle energies calculated dynamically are larger , which indicates that the polarization of the core nucleus is significant . when the phase space suppression is taken into account the @xmath0 width is reduced by more than twice ( compare ` complex ' and ` complex+@xmath62 ' in the last row of table [ tab . : results ] ) .
when treating @xmath56 self - consistently including the @xmath0 and @xmath57 momenta , the @xmath0 width is reduced by additional @xmath64 mev , but still remains sizeable .
the corresponding lifetime of the @xmath0 in the nucleus is @xmath65 fm .
lcccccccc & & & & + & dyn & stat & dyn & stat & dyn & stat & dyn & stat + @xmath66 & -193.7 & -137.1 & -175.6 & -134.6 & -190.2 & -136.1 & -191.6 & -136.3 + @xmath67 & - & - & 552.3 & 293.3 & 232.5 & 165.0 & 179.9 & 144.7 + finally , we discuss spin symmetry in @xmath0 spectra .
static calculations with a real potential @xcite revealed that antinucleon spectra in nuclei exhibit spin symmetry .
we explored the @xmath0 spectra considering the dynamical effects as well as the @xmath0 absorption in the nucleus . in fig .
[ fig . : symmetry ] , the real parts of the upper components ( left ) and the differential relation @xcite @xmath68 for the real parts of the lower components ( right ) of the @xmath0 wave function in @xmath53 and @xmath54 states in @xmath55o are plotted for @xmath34 .
we found that spin symmetry is well preserved in the @xmath0 spectrum calculated fully self - consistently using the @xmath0 potential consistent with the @xmath0 atom data .
the @xmath0 annihilation causes only minor deviations .
when different values of @xmath41 for the scalar and vector potentials are considered , spin symmetry holds only approximately .
the deviations gradually increase with increasing difference between the scalar and vector potentials .
99 brvenich , t.j .
_ et al . _ : phys .
c * 71 * , 035201 ( 2005 ) larionov , a.b .
_ et al . _ : phys .
c * 82 * , 024602 ( 2010 ) serot , b.d . and
walecka , j.d . :
* 16 * , 1 ( 1986 ) hrtnkov , j. : master s thesis , czech technical university in prague ( 2013 ) , + sugahara , y. and toki , h. : nucl .
a * 579 * , 557572 ( 1994 ) friedman , e. , gal , a. and mare , j. : nucl .
a * 761 * , 283295 ( 2005 ) walker , g.e . ,
goodman , ch.d . and
olmer , c. ( eds . ) : antinucleon and nucleon nucleus interaction , plenum press , new york ( 1985 ) particle data group [ online ] , + a monte carlo simulation tool pluto [ online ] , + ciepl , a. _ et al .
b * 702 * , 402407 ( 2011 ) zhou , s. , meng , j. and ring , p. : phys .
91 * , 262501 ( 2003 ) ginocchio , j.n .
: phys . rep . *
414 * , 165261 ( 2005 ) | we performed fully self - consistent calculations of @xmath0nuclear bound states within the relativistic mean - field ( rmf ) model .
the g - parity motivated @xmath0meson coupling constants were adjusted to yield potentials consistent with @xmath0atom data .
we confirmed large polarization effects of the nuclear core caused by the presence of the antiproton .
the @xmath0 absorption in the nucleus was incorporated by means of the imaginary part of a phenomenological optical potential . the phase space reduction for the @xmath0 annihilation products
was taken into account .
the corresponding @xmath0 width in the medium significantly decreases , however , it still remains considerable for the @xmath0 potential consistent with experimental data . |
recent observations of type ia supernovae by two independent groups , the high - z supernova team ( riess et al .
1998 ) and the supernova cosmology project ( perlmutter et al . 1999 ) suggest that our universe is presently undergoing an accelerating expansion .
the highest redshift supernova observed so far , sn 1997ff at @xmath16 , not only support this accelerating view , but also glimpse the earlier decelerating stage of the expansion ( riess et al .
it seems that determining a convincing mechanism with a solid basis in particle physics that explains the accelerating universe is emerging as one of the most important challenges in modern cosmology .
it is well known that all known types of matter with positive pressure generate attractive forces and decelerate the expansion of the universe conventionally , a deceleration factor is always used to describe the status of the universe s expansion ( sandage 1988 ) . given this ,
the discovery from the high - redshift type ia supernovae may indicates the existence of a new component with fairly negative pressure , which is now generally called dark energy .
coincidently or not , a dark energy component could offset the deficiency of a flat universe , favoured by the measurements of the anisotropy of the cosmic microwave background ( de bernardis et al .
2000 ; balbi et al .
2000 , durrer et al . 2003 ; bennett et al .
2003 ; melchiorri and odman 2003 ; spergel et al .
2003 ) , but with a very subcritical matter density parameter @xmath7 , obtained from dynamical estimates or x - ray and lensing observations of clusters of galaxies(for a recent summary , see turner 2002 ) .
the simplest possibility for the dark energy component is cosmological constant @xmath17 ( weinberg 1989 ; carroll et al .
1992 ; krauss and turner 1995 ; ostriker and steinhardt 1995 ; chiba and yoshii 1999 ; futamase and hamana 1999 ) .
other candidates for the dark energy include : a decaying vacuum energy density or a time varying @xmath17-term ( ozer and taha 1987 ; vishwakarma 2001 ; alcaniz and maia 2003 ) , an evolving scalar field ( referred to by some as quintessence : ratra and peebles 1988 ; wetterich 1988 ; frieman et al .
1995 ; coble et al . 1997 ; caldwell et al . 1998 ; wang and lovelace 2001 ; wang and garnavich 2001 ; podarius and ratra 2001 ; li , hao and liu 2002 ; weller and albrecht 2002 ; li et al .
2002a , b ; chen and ratra 2003 ; mukherjee et al .
2003 ) , the phantom energy , in which the sum of the pressure and energy density is negative ( caldwell 2002 ; hao and li 2003a , b ; dabrowski et al . 2003 ) , the so - called x - matter " , an extra component simply characterized by an constant equation of state @xmath18 ( xcdm ) ( turner and white 1997 ; chiba et al . 1997 ; zhu 1998 , 2000 ; zhu , fujimoto and tatsumi 2001 ; yamamoto and futamase 2001 ; sereno 2002 ; alcaniz , lima and cunha 2003 ; jain et al .
2003 ; discus and repko 2003 ; lima , cunha and alcaniz 2003 ) , the chaplygin gas whose equation of state is given by @xmath19 where @xmath20 is a positive constant ( kamenshchik et al . 2001 ; bento et al .
2002 ; alam et al . 2003
; alcaniz , jain and dev 2003 ; dev , alcaniz and jain 2003a ; silva and bertolami 2003 ; dev , jain and alcaniz 2003 ) .
although a lot of efforts have been made to pin down the amount and nature of the dark energy , it is still far from reaching a convincing machanism with solid basics of particle physics for the accelerating universe . on the other hand , many models have appeared that make use of the very ideas of branes and extra dimensions to obtain an accelerating universe ( randall and sundrum 1999a , b ; deffayet , dvali and gabadadze 2002 ; avelino and martins 2002 ; alcaniz , jain and dev 2002 ; jain , dev and alcaniz 2002 ) .
the basic idea behind these braneworld cosmologies is that our observable universe might be a surface or a brane embedded in a higher dimensional bulk spacetime in which gravity could spread ( randall 2002 ) .
the bulk gravity see its own curvature term on the brane which accelerates the universe without dark energy .
here we are concerned with the mechanism proposed by freese and lewis ( 2002 ) , in which some kind of bulk stress energy changes the form of the friedmann equation as follows @xmath21 where @xmath22 is the hubble parameter as a function of cosmic time , @xmath23 is the scale factor of the universe , @xmath24 is the energy density containing only ordinary matter ( @xmath25 ) and radiation ( @xmath26 ) , i.e. , @xmath27 . since at present @xmath28
, @xmath24 can be considered consisting of @xmath25 only .
the second term , called cardassion term , would drives the universe accelerating at a late epoch when it is dominated ( or before or after it is dominated , which depends on the power index @xmath0 being equal to or less or larger than @xmath29 , see section 3 for more detail discussion ) , without seeking to another unknown dark energy component .
several authors have explored the agreement of the cardassian expansion model with various observations such as the compact radio source angular size versus redshift data ( zhu and fujimoto 2002 ) , the cosmic macrowave background anisotropy measurements ( sen and sen 2003a , 2003b ) , the distant type ia supernovae data ( zhu and fujimoto 2003 ; wang et al .
2003 ; cao 2003 ; szydlowski and czaja 2003 ; godlowski and szydlowski 2003 ) , optical gravitational lensing surveys ( dev , alcaniz and jain 2003b ) and structure formation ( multamaki et al . 2003 ) . in this work ,
we first constrain the parameters from the turnaround redshift , @xmath2 , at which the universe switches from deceleration to acceleration .
secondly we analyze this scenario with the sunyaev - zeldovich ( sz)/x - ray data compiled by reese et al .
( 2002 ) , which is so far the largest and homogeneous sample . after providing the basic equations and the angular diameter distance formulas relevant to our analysis ( section 2 )
, we discuss the constraints on the parameter space of the cardassian model from the turaround redshift @xmath2 in section 3 . in section 4
, we present our sz / x - ray data analysis to test the scenario .
finally , we finish the paper by summarizing our conclusions and discussion in section 5 .
we provide here the important equations resulting from the modified friedmann equation , eq.(1 ) , for more details , see freese and lewis ( 2002 ) . in the usual friedmann equation , @xmath30 . to be consistent with the standard comological result , one should take @xmath31 .
it is convenient to use the redshift @xmath32 , at which the two terms of eq.([eq : ansatz ] ) are equal , as the second parameter of the cardassian model . in this parameterization of ( @xmath33 )
, it can be shown that ( freese and lewis 2002 ) , @xmath34^{-1}$ ] , where @xmath35 is the matter density of the universe at the present time and @xmath36kms@xmath37mpc@xmath37 is the hubble constant .
evaluating the ansatz of eq.([eq : ansatz ] ) at the present time , we have ( freese and lewis 2002 ) @xmath38 .\ ] ] because in the cardassian model the universe is flat and contains only matter , the matter density at present , @xmath35 , should be equal to the ` critical density ' of this scenario . from eq.([eq : h0 ] ) , we have @xmath39^{-1 } \;\ ; , \ ] ] where @xmath40 is the critical density of the standard friedmann model .
as expected for a flat universe , the `` density parameter '' , @xmath41 , should be equal to @xmath42 .
but we are used to the standard density parameter , @xmath43 , which is defined in terms of the critical density in standard friedmann model . from equation
( 3 ) , we have @xmath44 , and hence we could alternatively use ( @xmath0 , @xmath1 ) to fully characterize the cardassian model .
now we evaluate the angular diameter distance as a function of redshift @xmath45 as well as the parameters of the model . following the notation of peebles ( 1993 )
, we define the redshift dependence of the hubble paramter @xmath46 as @xmath47 . for the ansatz of eq.([eq : ansatz ] ) and a flat universe with only matter ( baryonic and cold dark matter ) , we get @xmath48\ ] ] then , it is straightforward to show that the angular diameter distance is given by @xmath49
the cardassian scenario is motivated to provide a possible mechanism for the acceleration of the universe .
it is natually that one could use the observational constraints on the deceleration parameter , @xmath50 , to check the feasibility of the model . in terms of @xmath51 function , we get the deceleration parameter as a function of redshift as follows ( zhu and fujimoto 2003 ) @xmath52 from eq.([eq : newe ] ) , we could derive the turnaround redshift at which the universe switches from deceleration to acceleration , or in other words the redshift at which the deceleration parameter vanishes , which is as follows @xmath53^{1\over { 3(1-n ) } } \,\,.\ ] ] in figure 1 , we plot this redshift as a function the power index of the cardassian term , @xmath0 , for several values of @xmath1 . while @xmath54 is the current optimistic matter density ( see turner 2002 for the argument ) , @xmath55@xmath56 is a wider range
. the dotted lines of figure 1 correspond to the present observational constraints on the turnaround redshift ( at the @xmath57 level ) , @xmath58 , from the latest supernova data ( perlmutter et al .
1999 ; riess et al . 1998,2001
; turner and riess 2002 ; avelino and martins 2002 ) .
as figure 1 shows , for every value of @xmath1 , there exists a value for the power index of the cardassian term , @xmath59 , satisfying @xmath60 = { 1\over \omega_m } -1 \,\,,\ ] ] which makes the turnaround redshift @xmath2 reach the maximum value , @xmath61_{\rm max } = \exp\left[1/ ( 2 - 3n_{\rm peak } ( \omega_m))\right ] -1 \;\;.\ ] ] this @xmath59
@xmath4_{\rm max}$ ] relation is illustrated in figure 1 by the thick dotted - dash line .
for example , for @xmath62 , we have from eq.([eq : npeak ] ) @xmath63 and hence @xmath4_{\rm max } \sim 1.05 $ ] , or in other words a universe with @xmath62 can not switch from deceleration to acceleration at redshift higher than @xmath64 , no matter what the power index of the cardassian term is .
we could restate the constraints in an alternative way : for every value of the turnaround redshift @xmath2 , the matter density of the universe must satisfy @xmath65/\ln(1+z)_{q=0 } } + 1\right]^{-1 } \,\,,\ ] ] where the equal mark comes into existence if and only if the power index of the cardassian term is @xmath66/[3\ln(1+z)_{q=0}]$ ] .
for example , for @xmath5 , as we are considering , we have from eq.([eq : omegammax ] ) @xmath67 , where the equal sign happans at @xmath68 .
the problem is now apparent : a very low matter density would be always necessary for the cardassian expansion scenario if the turneraround redshift is larger than 0.6 , especially for the case of @xmath0 deviating from @xmath69 . in order to illustrate how the turnaround redshift efficiently constrains the cardassian parameter space ,
we explore the model with different values in the @xmath0@xmath1 plane .
the results are shown in figure 2 .
the thick solid curve delimits the parameter space of accelerating / decelerating universe as the down - left / up - right area of the curve . while the present observational constraints on the turnaround redshift with @xmath58 restrict the parameter space into the narrow shaded area , the observed matter density of the universe limits them further ( see the overlap part of the shaded and the cross - hatched areas ) .
for instance , if the density parameter of the universe takes value around @xmath7 , a reasonable value suggested by various observations , the power index should be around @xmath8 to satisfy @xmath70 .
last but not least , we would like to note that , in the accelerating area , a horizontal line with @xmath71 will delimit it into two parts , @xmath72 or @xmath73 , corresponding to a universe which switches from deceleration to acceleration later or earlier than the moment at which the cardassian term equals to the conventional term of the friedmann equation respectively .
this can be see from the relation of equation ( 7 ) . for clarity
, we do not show it in figure 2 .
therefore the expansion turnaround redshift is generally not equal to @xmath32 , the redshift at which the cardassian term reach the normal matter density term in friedmann equation . only for @xmath74 ,
the acceleration happans exactly when the cardassian term starts to dominate .
however for @xmath75 ( @xmath76 ) , the universe switches from deceleration to acceleration after ( before ) the cardassian term dominates .
it is widely believed that the universe only switches from deceleration to acceleration recently , which is based on two observational arguments ( amendola 2003 ) : ( 1)acceleration at high redshift might be in contrast with the observed large - scale structure , because it makes the gravitational instability ineffective ; ( 2)the highest redshift supernova sn1997ff at @xmath77 discovered so far seems to provide a glimpse of the epoch of deceleration ( riess et al . 2001 ; turner and riess 2002 ) .
however a cosmological model with a dark energy component strongly coupled to dark matter ( amendola 2000,2003 ) can be consistent with the above two observations but allow acceleration at high redshift , @xmath78 .
we can see from above analysis that the cardassian expansion model with reasonable matter density , @xmath79 , can hardly explain an acceleration happaned earlier than @xmath80 . in this sense ,
a directly observation of the acceleration might be one of the most crucial and efficient tests to discriminate different machanisms for acceleration .
there have been so far two proposals for this kind of measurements .
one is to monitor the redshift change of quasars during ten years or so ( loeb 1998 ) .
another one is to measure the gravitational wave phase of neutron star binaries at @xmath81 for ten years using a decihertz interferometer gravitational wave observatory ( decigo : seto , kawamura and nakamura 2001 ) .
they might provide an efficient test for various acceleration mechanisms .
it has long been suggested that a measurement of the thermal sz effect ( sunyaev and zeldovich 1972 ) of a cluster of galaxies , combined with x - ray observations , can be used to determine the cluster cosmological angular - diameter distance and hence the hubble constant @xmath10 and the deceleration parameter @xmath82 ( cavaliere , danese and de zotti 1977 ; silk and white 1978 ; birkinshaw 1979 ) .
benefiting from the improvements to the traditional single - dish observations ( myers et al .
1997 ) and thanks to the new bolometer technology ( holzapfel et al . 1997 ) and interferometry technique ( calstrom , joy and grego 1996 ) , several tens of clusters of galaxies have been selected for the sz measurements and have been used to determine the hubble constant ( for recent summaries , see birkinshaw 1999 ; calstrom , holder and reese 2002 ) . recently ,
reese et al . ( 2002 ) published the sz measurements of a sample of 18 galaxy clusters with redshifts ranging from 0.14 to 0.83 observed by the owens valley radio observatory ( ovro ) and the berkeley - illinois - maryland association ( bima ) interferometers .
the author used the maximum likelihood joint analysis method to analyze their sz measurements with archival rosat x - ray imaging observations , which provide the largest homogeneously analyzed sample of the sz / x - ray clusters with angular diameter distance determinations so far ( reese et al .
the database is shown in figure 3 . because the redshift range of the cluster sample is comparable to the distant type ia supernovae data compiled by riess et al .
( 1998 ) and perlmutter et al .
( 1999 ) that lead to the discovery of accelerating expansion , it provides well an independent cross check of the acceleration mechanism .
we use this sample to give an observational constraint on the cardassian model parameters , @xmath0 and @xmath1 ( or @xmath83 ) .
we determine the model parameters @xmath0 and @xmath83 through a @xmath84 minimization method .
the range of @xmath0 spans the interval [ -1 , 1 ] in steps of 0.01 , while the range of @xmath83 spans the interval [ 0,1-@xmath15 ] in 101 points .
@xmath86^{2}}{\sigma_{i}^{2}}},\ ] ] where @xmath87 are the redshifts of the galaxy clusters and @xmath88 are the theoretical predictions from equation ( [ eq : da ] ) .
@xmath89 are the angular diameter distances of the cluster sample determined by the sz / x - ray route .
@xmath90 are the symmetric root - mean - square errors of the cluster distance measurements ( reese et al .
2002 ; allen , schmidt and fabian 2002 ) .
the summation is over all the observational data points . in order to reduce the number of parameters to be determined ,
we specify the baryonic matter density parameter to be @xmath91 ( omeara et al . 2001 ; allen , schmidt and fabian 2002 ) . in order to make the analysis independent of choice of the hubble constant , we minimize eq.([eq : chi2 ] ) for @xmath10 , @xmath0 and @xmath83 , simultaneously , which gives @xmath9259.6@xmath93km@xmath93s@xmath94mpc@xmath37 , @xmath95 and @xmath96 as the best fit with @xmath97 and 15 degrees of freedom ( d.o.f ) . an inspection of terms of @xmath97 for the above best fit shows that in this summation a big amount , namely @xmath98 , is given by one cluster abell 370 with redshift 0.374 , which should be considered as an outlier .
following reese et al .
( 2002 ) , we also analyze the subsample of 17 galaxy clusters which excludes the outlier , abell 370 . from the view of statistics ,
an outlier appears when it is inconsistent with the remainder of the data set , and is better to be removed from the sample ( mszros 2002 ) .
a glimpse of figure 3 , one immediately find that abell 370 might be an outlier , because it is far above the distances expected from the trend given by other objects , in another word , it is `` too far '' .
our best fit to the rest 17 clusters occurs for @xmath9259.2@xmath93km@xmath93s@xmath94mpc@xmath37 , @xmath99 and @xmath96 with @xmath100 and 14 d.o.f . in figure 4
, we show contours of constant likelihood ( 68% and 95% c.l . ) in the @xmath0@xmath1 plane for the best fit .
the fitting results for samples with and without abell 370 are very similar , but the @xmath11 per d.o.f for the latter reduces significantly and indicates that no large statistical errors remain unaccounted for .
the similarity of the fitting results also indicates the robustness of our analysis .
the best fits of the cardassian expansion model to the current sz / x - ray data lead to a universe containing baryonic matter only , leaving no space for the huge amount of dark matter whose existence has been widely accepted among the astronomical community .
however , note from figure 4 that the allowed range for both @xmath0 and @xmath83 is reasonable large , showing that the constraints on cardassian expansion from the considered sz / x - ray cluster sample are not restrictive .
for example , the models with reasonable matter density ( @xmath7 ) but some negative values of @xmath0 are within @xmath57 of the best fit .
we have explored the constraints on the parameter space of the cardassian scenario from the turnaround redshift , @xmath2 , at which the universe switches from deceleration to acceleration .
we demonstrated that , for every matter density , @xmath1 , there exist a unique power index , @xmath3 , which makes @xmath2 reach its maximum value . if the acceleration happans earlier than @xmath5 , it is found @xmath8 for a reasonable matter density , @xmath7 .
it is also found that the cardassian scenario can hardly explain an acceleration happaned earlier than @xmath101 .
we further analyzed the scenario using the largest homogeneously sample of sz / x - ray clusters compiled by reese et al .
all best fitting results seem to give a universe with very low matter density ( @xmath102 ) .
our analysis is consistent with previous observational constraints ( sen and sen 2003a,2003b ; wang et al .
2003 ; cao 2003 ; zhu and fujimoto 2002 , 2003 ) .
however at present both the statistical and systematic uncertainties in the sz / x - ray route of distance determination are very large ( birkinshaw 1999 ; reese et al .
2002 ; calstrom , holder and reese 2002 ) . which loosen the constraints of the cardassian expansion scenario very much ( figure 4 )
therefore , the cardassian expansion scenario with reasonable matter density is also consistent with the current sz / x - ray data within @xmath57 .
the projection effect of aspherical clusters modeled with a spherical geometry is a large systematic error in the angular diameter distance determinations based on the sz / x - ray route ( hughes and birkinshaw 1998 ) .
other large systematic uncertainties are due to the departures form the isothermality , the possibility of clumping and possible point source contamination of the sz observations ( calstrom , holder and reese 2002 ; reese et al .
it is promising to improve the systematic uncertainties for both sz and x - ray observations within a few years .
the absolute calibration of the sz observation is now underway to improve significantly ( calstrom , holder and reese 2002 ) .
the chandra and xmm - newton x - ray telescopes not only reduce the uncertainty in the x - ray intensity scale , but also are providing temperature profiles of galaxy clusters ( markevitch et al .
2000 ; nevalainen , markevitch and forman 2000 ) .
the undergoing or planing large survey of sz clusters will perhaps provide a sz / x - ray sample of a few hundred clusters with redshift extending to one and beyond , which will improve our analysis and pin down the cardassian model parameters more accurately .
we would like to thank s. w. allen , a. g. riess , and r. g. vishwakarma for their helpful discussion .
our thanks go to the anonymous referee for valuable comments and useful suggestions , which improved this work very much .
this work was supported by a grant - in - aid for scientific research on priority areas ( no.14047219 ) from the ministry of education , culture , sports , science and technology .
alam , u. , sahni , v. , saini , t. d. and starobinsky , a. a. 2003 , , 344 , 1057 alcaniz , j. s. , jain , d. and dev , a. 2002 , , 66 , 067301 alcaniz , j. s. , jain , d. and dev , a. 2003 , , 67 , 043514 alcaniz , j. s. , lima , j. a. s. and cunha , j. v. 2003 , , 340 , l39 alcaniz , j. s. and maia , j. m. f. 2003 , , 67 , 043502 allen , s. w. , schmidt , r. w. and fabian , a. c. 2002 , , 334 , l11 amendola , l. 2000 , , 62 , 043511 amendola , l. 2003 , , 342 , 221 avelino , p. p. and martins , c. j. a. p. 2002
, , 565 , 661 balbi , a. et al .
2000 , , 545 , l1 bennett , c. l. et al .
2003 suppl .
148 , 1 bento , m. c. , bertolami , o and sen , a. a. 2002 , , 66 , 043507 birkinshaw , m. 1979 , , 187 , 847 birkinshaw , m. 1999 , phys .
, 310 , 97 caldwell , r. 2002 , phys.lett.b , 545 , 23 caldwell , r. , dave , r. , and steinhardt , p. j. 1998 , , 80 , 1582 calstrom , j. e. , joy , m. and grego , l. e. 1996 , , 456 , l75 calstrom , j. e. , holder , g. p. and reese , e. d. 2002 , , 40 , 643 cao , l. 2003 , chjaa , 3 , 341 carroll , s. , press , w. h. and turner , e. l. 1992 , , 30 , 499 cavaliere , a. danese , l. and de zotti , g. 1977 , , 217 , 6 chen , g. and ratra , b. 2003 , , 582 , 586 chiba , m. and yoshii , y. 1999 , , 510 , 42 chiba , t. , sugiyama , n. and nakamura , t. 1997 , , 289 , l5 coble , k. , dodelson , s. , and frieman , j. a. 1997 , , 55 , 1851 de bernardis , p. et al .
2000 , , 404 , 955 dabrowski , m. p. , stochowiak , t. , and szydlowski , m. 2003 , , accepted ( hep - th/0307128 ) deffayet , c. , dvali , g. and gabadadze , g. 2002 , , 65 , 044023 dev , a. , alcaniz , j. s. and jain , d. 2003 , , 67 , 023515 dev , a. , alcaniz , j. s. and jain , d. 2003 , astro - ph/0305068 dev , a. , jain , d. and alcaniz , j. s. 2003 , in preparation discus , d. a. and repko , w. w. 2003 , , 67 , 083520 durrer , r. , novosyadlyj , b. and apunevych , s. 2003 , , 583 , 33 freese , k. 2003 , nucl .
suppl . ) 124 , 50 freese , k. and lewis , m. 2002 , phys.lett.b , 540 , 1 frieman , j. , hill , c. t. , stebbins , a. and waga , i. 1995 , , 75 , 2077 futamase , t. and hamana , t. 1999 , prog.theor.phys . , 102 , 1037 godlowski , w. and szydlowski , m. 2003 , astro - ph/0309569 hao , j. -g . and
li , x. -z .
2003a , , 67 , 107303 hao , j. -g . and
li , x. -z .
2003b , , 68 , 043501 holzapfel , w. l. et al .
1997 , , 479 , 17 hughes , j. p. and birkinshaw , m. 1998 , , 501 , 1 jain , d. , dev , a. and alcaniz , j. s. 2002 , , 66 , 083511 jain , d. , dev , a. , panchapakesan , n. , mahajan , s. , bhatia , v. b. 2003 int.j.mod.phys .
d12 , 953 kamenshchik , a. , moschella , u. and pasquier , v. 2001 , phys.lett.b , 511 , 265 krauss , l. m. and turner , m. s. 1995 , gen .
grav . , 27 , 1137 li , m. , lin , w. , zhang , x. and brandenberger , r. 2002a , , 65 , 023519 li , m. , wang , x. , feng , b. and zhang , x. 2002b , , 65 , 103511 li , x. -z . ,
hao , j. -g . and liu , d. -j .
2002 , class.quan.grav .
19 , 6049 lima , j. a. s. , cunha , j. v. and alcaniz , j. s. 2003 , , 68 , 023510 loeb , a. 1998 , , 499 , l111 markevitch , m. et al .
2000 , , 541 , 542 melchiorri , a. and odman , c. 2003 , , 67 , 081302(r ) mszros , a. 2002 , , 580 , 12 mukherjee , p .
banday , a.j .
riazuelo , a .
gorski , k.m .
ratra , b .
2003 , apj , accepted ( astro - ph/0306147 ) multamaki , t. , gaztanaga , e. and manera , m. 2003 , , 344 , 761 myers , s. t. et al .
1997 , , 485 , 1 nevalainen , j. , markevitch , m. and forman , w. 2000 , , 536 , 73 omeara , j. m. et al .
2001 , , 552 , 718 ostriker , j. p. and steinhardt , p. j. 1995 , , 377 , 600 ozer , m. and taha , o. 1987 , nucl . phys . *
b287 * , 776 peebles , p. j. e. 1993 , principles of physical cosmology ( princeton university press , princeton , nj ) .
perlmutter , s. et al .
1999 , , 517 , 565 podariu , s. and ratra , b. 2001 , , 563 , 28 randall , l. 2002 , science , 296 , 1422 .
randall , l. and sundrum , r. 1999a , , 83 , 3370 randall , l. and sundrum , r. 1999b , , 83 , 4690 ratra , b. and p.j.e .
peebles , p. j. e. 1988 , , 37 , 3406 reese , e. d. et al .
2002 , , 581 , 53 riess , a. g. et al .
1998 , , 116 , 1009 riess , a. g. et al .
2001 , , 560 , 49 sandage , a. 1988 , , 26 , 561 sen , a. a. and sen , s. 2003a , , 588 , 1 sen , a. a. and sen , s. 2003b , , 68 , 023513 sereno , m. 2002 , , 393 , 757 seto , n. kawamura , s. and nakamura , t. 2001 , , 87 , 221103 silk , j. i. and white , s. d. m. 1978 , , 226 , l3 silva , p. t. and bertolami , o. 2003 , , submitted ( astro - ph/0303353 ) spergel , d. n. et al .
2003 suppl .
148 , 175 sunyaev , r. a. and zeldovich , y. a. 1972 , comm .
space phys . 4 , 173 szydlowski , m. and czaja , w. 2003 , astro - ph/0309191 turner , m. s. 2002 , , 576 , l101 turner , m. s. and riess , a. d. 2002 , , 569 , 18 turner , m. s. and white , m. 1997 , , 56 , r4439 vishwakarma , r. g. 2001 , class.quan.grav . 18 ,
1159 wang , y. , freese , k. , gondolo , p . and lewis , m. 2003 , , 594 , 25 wang , y. and garnavich , g. 2001 , , 552 , 445 wang , y. and lovelace , g. 2001 , , 562 , l115 weinberg , s. 1989 , rev .
61 , 1 weller , j. and albrecht , a. 2002 , , 65 , 103512 wetterich , c. 1988 , nucl.phys .
b302 , 645 zhu , z. -h .
1998 , 338 , 777 zhu , z. -h .
2000 , mod.phys.lett .
a15 , 1023 zhu , z. -h . and fujimoto , m. -k .
2002 , , 581 , 1 zhu , z. -h . and fujimoto , m. -k .
2003 , , 585 , 52 zhu , z. -h . , fujimoto , m. -k . and tatsumi , d. 2001 , , 372 , 377 | cosmic acceleration is one of the most remarkable cosmological findings of recent years .
although a dark energy component has usually been invoked as the mechanism for the acceleration , a modification of friedmann equation from various higher dimensional models provides a feasible alternative .
cardassian expansion is one of these scenarios , in which the universe is flat , matter ( and radiation ) dominated and accelerating but contains no dark energy component .
this scenario is fully characterized by @xmath0 , the power index of the so - called cardassian term in the modified friedmann equation , and @xmath1 , the matter density parameter of the universe . in this work ,
we first consider the constraints on the parameter space from the turnaround redshift , @xmath2 , at which the universe switches from deceleration to acceleration .
we show that , for every @xmath1 , there exist a unique @xmath3 , which makes @xmath2 reach its maximum value , @xmath4_{\rm max } = \exp\left[1/ ( 2 - 3n_{\rm peak})\right ] -1 $ ] , which is unlinearly inverse to @xmath1 .
if the acceleration happans earlier than @xmath5 , suggested by type ia supernovae measurements , we have @xmath6 no matter what the power index is , and moreover , for reasonable matter density , @xmath7 , it is found @xmath8 .
we next test this scenario using the sunyaev - zeldovich / x - ray data of a sample of 18 galaxy clusters with @xmath9 compiled by reese et al .
( 2002 ) .
we determine @xmath0 and @xmath1 , as well as the hubble constant @xmath10 , using the @xmath11 minimization method .
the best fit to the data gives @xmath12 , @xmath13 and @xmath14 ( @xmath15 is the baryonic matter density parameter ) .
however the constraints from the current sz / x - ray data is weak , though a model with lower matter density is prefered .
a certain range of the model parameters is also consistent with the data . |
the identification of a deterministic linear operator from the operator s response to a probing signal is an important problem in many fields of engineering .
concrete examples include system identification in control theory and practice , the measurement of dispersive communication channels , and radar imaging .
it is natural to ask under which conditions ( on the operator ) identification is possible , in principle , and how one would go about choosing the probing signal and extracting the operator from the corresponding output signal .
this paper addresses these questions by considering the ( large ) class of linear operators that can be represented as a continuous weighted superposition of time - frequency shift operators , i.e. , the operator s response to the signal @xmath2 can be written as @xmath3 where @xmath4 denotes the spreading function associated with the operator . the representation theorem ( * ? ? ?
14.3.5 ) states that the action of a large class of continuous ( and hence bounded ) linear operators can be represented as in . in the communications literature operators with input - output relation as in are referred to as linear time - varying ( ltv ) channels / systems and @xmath4 is the delay - doppler spreading function @xcite . for the special case of linear time - invariant ( lti ) systems , we have @xmath5 , so that reduces to the standard convolution relation @xmath6 the question of identifiability of lti systems is readily answered by noting that the system s response to the dirac delta function is given by the impulse response @xmath7 , which by fully characterizes the system s input - output relation .
lti systems are therefore always identifiable , provided that the probing signal can have infinite bandwidth and we can observe the output signal over an infinite duration . for ltv systems the situation is fundamentally different . specifically , kailath s landmark paper @xcite shows that an ltv system with spreading function compactly supported on a rectangle of area @xmath8 is identifiable if and only if @xmath9
. this condition can be very restrictive .
measurements of underwater acoustic communication channels , such as those reported in @xcite for example , show that the support area of the spreading function can be larger than @xmath10 . the measurements in @xcite exhibit , however , an interesting structural property : the nonzero components of the spreading function are scattered across the @xmath11-plane and the sum of the corresponding support areas , henceforth called `` overall support area '' , is smaller than @xmath10 .
a similar situation arises in radar astronomy @xcite .
bello @xcite shows that kailath s identifiability result continues to hold for arbitrarily fragmented spreading function support regions as long as the corresponding overall support area is smaller than @xmath10 .
kozek and pfander @xcite and pfander and walnut @xcite found elegant functional - analytical identifiability proofs for setups that are more general than those originally considered in @xcite and @xcite .
however , the results in @xcite require the support region of @xmath4 to be known prior to identification , a condition that is very restrictive and often impossible to realize in practice .
in the case of underwater acoustic communication channels , e.g. , the support area of @xmath4 depends critically on surface motion , water depth , and motion of transmitter and receiver . for wireless channels , knowing the spreading function s support region would amount to knowing the delays and doppler shifts induced by the scatterers in the propagation medium .
[ [ contributions ] ] contributions + + + + + + + + + + + + + we show that an operator with input - output relation is identifiable , without prior knowledge of the operator s spreading function support region and without limitations on its total extent , if and only if the spreading function s total support area satisfies @xmath0 .
what is more , this factor - of - two penalty relative to the case where the support region is known prior to identification @xcite can be eliminated if one asks for identifiability of _ _ almost all _ _ operators only .
this result is surprising as it says that ( for almost all operators ) there is no price to be paid for not knowing the spreading function s support region in advance .
our findings have strong conceptual parallels to the theory of spectrum - blind sampling of sparse multi - band signals @xcite .
furthermore , we present algorithms which , in the noiseless case , provably recover all operators with @xmath0 , and almost all operators with @xmath1 , without requiring prior knowledge of the spreading function s support region ; not even its area @xmath8 has to be known .
specifically , we formulate the recovery problem as a continuous multiple measurement vector ( mmv ) problem @xcite .
we then show that this problem can be reduced to a finite mmv problem @xcite .
the reduction approach we present is of independent interest as it unifies a number of reduction approaches available in the literature and presents a simplified treatment . in the case of wireless channels or radar systems ,
the spreading function s support region is sparse and typically contained in a rectangle of area @xmath10 . in the spirit of compressed sensing , where sparse objects are reconstructed by taking fewer measurements than mandated by their `` bandwidth '' ,
we show that in this case sparsity ( in the spreading function s support region ) can be exploited to identify the system while undersampling the response to the probing signal . in the case of channel identification
this allows for a reduction of the identification time , and in radar systems it leads to increased resolution .
[ [ relation - to - previous - work ] ] relation to previous work + + + + + + + + + + + + + + + + + + + + + + + + + recently , taubck et al .
@xcite and bajwa et al .
@xcite considered the identification of ltv systems with spreading function compactly supported in a rectangle of area @xmath12 . while @xcite assume that the spreading function consists of a finite number of dirac components whose delays and doppler shifts are unknown prior to identification , the methods proposed in @xcite do not need this assumption .
in the present paper , we allow general ( i.e. , continuous , discrete , or mixed continuous - discrete ) spreading functions that can be supported in the entire @xmath13-plane with possibly fragmented support region .
herman and strohmer @xcite , in the context of compressed sensing radar , and pfander et al .
@xcite considered the problem of identifying finite - dimensional matrices that are sparse in the basis of time - frequency shift matrices .
this setup can be obtained from ours by discretization of the input - output relation through band - limitation of the input signal and time - limitation and sampling of the output signal .
the signal recovery problem in @xcite is a standard _ single measurement _ recovery problem @xcite .
as we start from a continuous - time formulation we find that , depending on the resolution induced by the discretization through time / band - limitation , the resulting recovery problem can be an mmv problem .
this is relevant as multiple measurements can improve the recovery performance significantly . in fact , it is the mmv nature of the recovery problem that allows identification of _ almost all _ operators with @xmath14 .
[ [ organization - of - the - paper ] ] organization of the paper + + + + + + + + + + + + + + + + + + + + + + + + + the remainder of the paper is organized as follows . in section
[ sec : prform ] , we formally state the problem considered .
section [ sec : mainresults ] contains our main identifiability results with the corresponding proofs given in sections [ sec : proofmthm ] and [ sec : almostall ] . in sections [ sec :
reconstruct ] and [ sec : almostall ] , we present identifiability algorithms along with corresponding performance guarantees . in section [ sec : discior ] , we consider the identification of systems with sparse spreading function compactly supported within a rectangle of area @xmath10 . section [ sec : numerical_recres ] contains numerical results . [ [ notation ] ] notation + + + + + + + + the superscripts @xmath15 , @xmath16 , and @xmath17 stand for complex conjugation , hermitian transposition , and transposition , respectively .
we use lowercase boldface letters to denote ( column ) vectors , e.g. , @xmath18 , and uppercase boldface letters to designate matrices , e.g. , @xmath19 .
the entry in the @xmath20th row and @xmath21th column of @xmath19 is @xmath22_{k , l}$ ] and the @xmath20th entry of @xmath23 is @xmath24_k$ ] .
the euclidean norm of @xmath18 is denoted by @xmath25 , and @xmath26 stands for the number of non - zero entries in @xmath23 .
the space spanned by the columns of @xmath19 is @xmath27 , and the nullspace of @xmath28 is denoted by @xmath29 .
@xmath30 designates the cardinality of the smallest set of linearly dependent columns of @xmath28 .
@xmath31 stands for the cardinality of the set @xmath32 . for sets @xmath33 and @xmath34 ,
we define set addition as @xmath35 .
for a ( multi - variate ) function @xmath36 , @xmath37 denotes its support set . for ( multi - variate ) functions @xmath36 and @xmath38 , both with domain @xmath32 , we write @xmath39 for their inner product and @xmath40 for the norm of @xmath41 .
we say that a set of functions @xmath42 with domain @xmath32 is linearly independent if there is no vector @xmath43 , such that @xmath44 , where @xmath45}^t}$ ] .
we denote the dimension of the span of @xmath46 as @xmath47 .
the fourier transform of a function @xmath48 is defined as @xmath49 . the dirac delta function is denoted by @xmath50 and @xmath51 .
@xmath52 stands for the space of complex - valued square - integrable functions .
the random variable @xmath53 is proper complex gaussian with mean @xmath54 and variance @xmath55 . finally , for @xmath56
, we let @xmath57 be the largest integer not greater than @xmath58 .
given the normed linear spaces @xmath59 and @xmath60 , we consider linear operators @xmath61 that can be represented as a weighted superposition of translation operators @xmath62 , with @xmath63 , and modulation operators @xmath64 , with @xmath65 , according to @xmath66 with @xmath67 , where @xmath68 is a normed linear space .
this is a rather general setup , since according to ( * ? ? ?
14.3.5 ) , a large class of continuous ( and hence bounded ) linear operators can be represented as in .
for the theory to be mathematically precise , we need to consider suitable triplets of spaces @xmath69 inducing a space of operators @xmath70 .
the triplets @xmath71 have to be chosen to `` match '' ; specifically , @xmath68 may have to satisfy certain regularity conditions , depending on the choice of @xmath59 , for to be well - defined .
for example , if @xmath72 is a hilbert space and @xmath73 , then @xmath74 is the set of hilbert - schmidt operators ( * ? ? ? * , a.8 ) .
since our identifiability proof relies on the use of dirac delta functions as probing signals , we need to choose @xmath59 such that it contains generalized functions .
a corresponding valid choice is the following @xcite : let @xmath68 and @xmath59 be feichtinger s banach algebra @xmath75 @xcite and its dual @xmath76 , respectively , and @xmath77 .
the corresponding space @xmath74 is equipped with the hilbert - schmidt norm @xmath78 another valid triplet is obtained by setting @xmath79 , and @xmath80 , see @xcite . in this case
both @xmath59 and @xmath68 contain generalized functions , in particular dirac delta functions ; however , the norm of the corresponding space @xmath81 takes on a more complicated form than .
the norm the space @xmath81 is equipped with determines the definition of identifiability .
while our results , including the identification algorithms , hold true for the setup @xmath82 , to keep the exposition simple , we will work with the triplet @xmath83 and the corresponding norm .
we refer the interested reader to @xcite for a detailed description of the rigorous mathematical setup required for the choice @xmath82 . [
[ restrictions - on - the - spreading - function ] ] restrictions on the spreading function + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + following @xcite we consider spreading functions with compact support .
this assumption is not critical and can be justified , e.g. , in wireless and in underwater acoustic communication channels as follows .
the extent of the spreading function in @xmath84-direction is determined by the maximum delay ( doppler shift ) induced by the channel .
the maximum doppler shift will be limited as the velocity of objects in the propagation channel and/or the velocity of transmitter and receiver is limited .
while the maximum delay induced by scattering objects in the channel can , in principle , be arbitrarily large , contributions corresponding to large enough delay will be sufficiently small to be treated as additive noise , thanks to path loss @xcite .
while we do not present analytical results for the noisy case , the impact of noise on the performance of our identification algorithms is assessed numerically in section [ sec : numerical_recres ] .
following @xcite we , moreover , restrict ourselves to spreading functions with support regions of the form @xmath85 where @xmath86 is a `` cell '' in the @xmath13-plane and @xmath87 and @xmath88 are parameters whose role will become clear shortly .
the set of `` active cells '' is specified by @xmath89 . since @xmath90 and @xmath91
, it follows that , choosing @xmath92 and @xmath93 accordingly , @xmath94 and @xmath95 can be arbitrarily large ; the spreading function can hence be supported on an arbitrarily large , but finite , region .
we denote the area of @xmath96 as @xmath97 and note that @xmath98 .
a general , possibly fragmented , support region of the spreading function can be approximated arbitrarily well by covering it with rectangles @xmath99 ( see figure [ fig : supex ] ) , as in , with @xmath100 and @xmath92 chosen suitably .
note that @xmath92 determines how fine this approximation is , since @xmath101 , while @xmath93 controls the ratio of width to height of @xmath99 .
characterizing the identifiability of operators whose spreading function support region is not compact , but has finite area , e.g. , @xmath102 , is an open problem @xcite .
let us next define the notion of identifiability of a set of operators @xmath103 .
the set @xmath104 is said to be identifiable , if there exists a probing signal @xmath105 such that for each operator @xmath106 , the action of the operator on the probing signal , @xmath107 , uniquely determines @xmath108 .
more formally , we say that @xmath104 is identifiable if there exists an @xmath105 such that @xmath109 identifiability is hence equivalent to invertibility of the mapping @xmath110 induced by the probing signal @xmath111 . in practice , invertibility alone is not sufficient as we want to recover @xmath108 from @xmath112 in a numerically stable fashion , i.e. , we want small errors in @xmath112 to result in small errors in the identified operator .
this requirement implies that the inverse of the mapping must be continuous ( and hence bounded ) , which finally motivates the following definition of ( stable ) identifiability , used in the remainder of the paper .
[ def : defstableidentification ] we say that @xmath111 identifies @xmath104 if there exist constants @xmath113 such that for all pairs @xmath114 , @xmath115 furthermore , we say that @xmath104 is identifiable , if there exists an @xmath105 such that @xmath111 identifies @xmath104 . in @xcite the identification of operators of the form under the assumption of the spreading function support region being known prior to identification is considered .
the set @xmath104 in @xcite therefore consists of operators with spreading function supported on a given region @xmath96 , ( i.e. , @xmath116 for all @xmath117 ) , which renders @xmath104 a linear subspace of @xmath81 so that @xmath118 , for all @xmath114
. hence definition [ def : defstableidentification ] above is equivalent to the following : @xmath111 identifies @xmath104 if there exist constants @xmath113 such that for all @xmath106 , @xmath119 , which is the identifiability condition put forward in @xcite .
not knowing the spreading function s support region prior to identification will require consideration of sets @xmath104 that are not linear subspaces of @xmath81 , which makes the slightly more general definition [ def : defstableidentification ] necessary .
as detailed in appendix [ app : binv ] , the lower bound in guarantees that the inverse of @xmath120 in exists and is bounded and hence continuous , as desired .
the ratio @xmath121 quantifies the noise sensitivity of the identification process . specifically , suppose that @xmath111 identifies @xmath104 , but the measurement @xmath122 , @xmath123 , is corrupted by additive noise .
concretely , assume that instead of @xmath122 , we observe @xmath124 , where @xmath125 is bounded , i.e. , @xmath126 . now assume that @xmath127 is consistent with the noisy observation @xmath128 , i.e. , @xmath129 .
we would like the error in the identified operator , i.e. , @xmath130 , to be proportional to @xmath131 .
the lower bound in guarantees that this is , indeed , the case as @xmath132 since @xmath133 , it follows from that @xmath134 is optimal in terms of noise sensitivity .
we can also conclude from that larger @xmath135 leads to smaller noise sensitivity .
increased @xmath135 , however , simply amounts to increased power of the probing signal .
this can be seen as follows .
suppose we found an @xmath136 that identifies @xmath104 with constants @xmath137 in .
then , @xmath138 with @xmath139 identifies @xmath104 with constants @xmath140 .
choosing @xmath141 large will therefore lead to small noise sensitivity .
before stating our main results , we define the set of operators with spreading function supported on a given region @xmath96 ( with @xmath96 as defined in ):
@xmath142 kailath @xcite and kozek and pfander @xcite considered the case where @xmath96 is a ( single ) rectangle , and bello @xcite and pfander and walnut @xcite analyzed the case where @xmath96 is allowed to be fragmented and spread over the @xmath13-plane . in both cases the support region @xmath96 is assumed to be known prior to identification .
we start by recalling the key result in @xcite , which subsumes the results in @xcite , and @xcite .
let @xmath96 be given .
the set of operators @xmath143 is identifiable if and only if @xmath144 .
[ pr : necessity ] [ pro : knownsupport ] as mentioned earlier , knowing the support region @xmath96 prior to identification is very restrictive and often impossible to realize in practice .
it is therefore natural to ask what kind of identifiability results one can get when this assumption is dropped . concretely
, this question can be addressed by considering the set of operators @xmath145 which consists of all sets @xmath143 such that @xmath146 .
our main identifiability results are stated in the two theorems below .
the set of operators @xmath147 is identifiable if and only if @xmath0 .
[ pr : identifiabilitynecsparse ] [ th : sufficiency ] see section [ sec : proofmthm ] .
the main implication of theorem [ th : sufficiency ] is that the penalty for not knowing the spreading function s support region prior to identification is a factor - of - two in the area of the spreading function .
the origin of this factor - of - two penalty can be elucidated as follows . for operators
@xmath148 with spreading function supported on @xmath96 , i.e. , @xmath149 , we have @xmath150 , i.e. , @xmath143 is a linear subspace of @xmath81 . in the case of unknown spreading function support region
we have to deal with the ( much larger ) set @xmath147 , consisting of all sets @xmath143 with @xmath146 .
it is readily seen that @xmath147 is not a linear subspace of @xmath81 .
simply take @xmath151 such that the support regions of @xmath152 and @xmath153 both have area @xmath8 and are disjoint . while @xmath154 , we do , however , have that @xmath155 .
this observation lies at the heart of the factor - of - two penalty in @xmath8 as quantified by theorem [ th : sufficiency ] .
we can eliminate this penalty by relaxing the identification requirement to apply to `` almost all '' @xmath156 instead of `` all '' @xmath156 . to be specific , we consider identifiability of a subset @xmath157 , containing `` almost all '' @xmath158 .
the set @xmath159 is obtained as follows .
first , set @xmath160 for @xmath161 , and then define @xmath162 the motivation for this specific definition of the set @xmath159 will become clear in section [ sec : almostall ] .
at this point , it is only important to note that the condition on the @xmath163 in the definition of @xmath159 allows to eliminate the factor - of - two penalty in @xmath8 .
the set of operators @xmath159 is identifiable if @xmath1 .
[ th : almost_all ] see section [ sec : almostall ] . in order to demonstrate that `` almost all '' @xmath156 are in @xmath159
, suppose that can be represented by an expansion of the corresponding @xmath163 into a set of orthonormal functions . ]
@xmath164 , where @xmath165 is a set of functions orthogonal on @xmath99 and the @xmath166 are drawn independently from a continuous distribution .
then , the @xmath163 will be linearly independent on @xmath99 with probability one @xcite , if @xmath167 .
finally , note that the operator @xmath108 with spreading function @xmath168 where @xmath169 satisfies @xmath170 , is an example of an operator that is in @xmath147 but not in @xmath159 .
putting things together , we have shown that `` almost all '' operators @xmath156 can be identified if @xmath1 .
this result is surprising as it says that there is no penalty for not knowing the spreading function s support region prior to identification , provided that one is content with a recovery guarantee for `` almost all '' operators .
the factor - of - two penalty in theorem [ th : sufficiency ] has the same roots as the factor - of - two penalty in sparse signal recovery @xcite , in the recovery of sparsely corrupted signals @xcite , in the recovery of signals that lie in a union of subspaces @xcite , and , most pertinently , in spectrum - blind sampling as put forward by feng and bresler @xcite and mishali and eldar @xcite .
we hasten to add that theorem [ th : almost_all ] is inspired by the insight that in the context of spectrum - blind sampling
the factor - of - two penalty in sampling rate can be eliminated by relaxing the recovery requirement to `` almost all '' signals @xcite . despite the conceptual similarity of the statement in theorem [ th : almost_all ] above and
the result in @xcite , the technical specifics are vastly different , as we shall see later . [ [ generalizations ] ] generalizations + + + + + + + + + + + + + + + theorems [ th : sufficiency ] and [ th : almost_all ] can easily be extended to operators with multiple inputs ( and single output ) , i.e. , operators whose response to the vector - valued signal @xmath171}^t}$ ] is given by @xmath172 where @xmath173 is the spreading function corresponding to the ( single - input ) operator between input @xmath174 and the output .
for the case where the support regions of all spreading functions @xmath175 are known prior to identification , pfander showed in @xcite that the operator @xmath176 is identifiable if and only if @xmath177 .
when the support regions are unknown , an extension of theorem [ th : sufficiency ] shows that @xmath176 is identifiable if and only if @xmath178 .
if one asks for identifiability of `` almost all '' operators only , the condition @xmath178 is replaced by @xmath177 .
finally , we note that these results carry over to the case of operators with multiple inputs and multiple outputs ( mimo ) .
specifically , a mimo channel is identifiable if each of its miso subchannels is identifiable , see @xcite for the case of known support regions .
to prove necessity in theorem [ pr : identifiabilitynecsparse ] , we start by stating an equivalence condition on the identifiability of @xmath147 . this condition is often easier to verify than the condition in definition [ def : defstableidentification ] , and is inspired by a related result on sampling of signals in unions of subspaces (
2 ) . @xmath111 identifies @xmath147 if and only if it identifies all sets @xmath179 with @xmath180 and @xmath181 , where @xmath182 .
[ pr : blind_identification ] first , note that the set of differences of operators in @xmath147 can equivalently be expressed as @xmath183 from definition [ def : defstableidentification ] it now follows that @xmath111 identifies @xmath147 if there exist constants @xmath184 such that for all @xmath185 we have @xmath186 next , note that for @xmath187 , we have that @xmath188 .
we can therefore conclude that is equivalent to @xmath189 for all @xmath190 , and for all @xmath191 and @xmath192 with @xmath193 . recognizing that is nothing but saying that @xmath111 identifies @xmath194 for all @xmath191 and @xmath192 with @xmath193 , the proof is concluded .
necessity in theorem [ th : sufficiency ] now follows by choosing @xmath195 such that @xmath196 and @xmath197 .
this implies @xmath198 and hence application of theorem [ pr : necessity ] to the corresponding set @xmath194 establishes that @xmath194 is not identifiable . by lemma [ pr : blind_identification
] this then implies that @xmath147 is not identifiable .
we provide a constructive proof of sufficiency by finding a probing signal @xmath111 that identifies @xmath147 , and showing how @xmath199 can be obtained from @xmath107 .
concretely , we take @xmath111 to be a weighted @xmath200-periodic train of dirac impulses @xmath201 the specific choice of the coefficients @xmath202}^t}$ ] will be discussed later .
kailath @xcite and kozek and pfander @xcite used an unweighted train of dirac impulses as probing signal to prove that ltv systems with spreading function compactly supported on a rectangle ( known prior to identification ) of area @xmath9 are identifiable .
pfander and walnut @xcite used the probing signal to prove the result reviewed as theorem [ pr : necessity ] in this paper . using a _ weighted
_ train of dirac impulses will turn out crucial in the case of unknown spreading function support region , as considered here .
it was shown recently ( * ? ? ?
2.5 ) that identification in the case of known support region , i.e. , for @xmath143 , is possible only with probing signals that decay neither in time nor in frequency , making dirac trains a natural choice .
the main idea of our proof is to i ) reduce the identification problem to that of solving a continuously indexed linear system ( of @xmath92 equations in @xmath203 unknowns ) , and ii ) based on lemma [ pr : blind_identification ] to show that the solution of this underdetermined linear system of equations is unique whenever @xmath0 , provided that @xmath204 is chosen appropriately .
we start by computing the response of @xmath108 to @xmath2 in . from
we get @xmath205 next , we use the zak transform @xcite to turn into a continuously indexed linear system of equations as described in step i ) above . the zak transform ( with parameter @xmath200 ) of the signal @xmath206 is defined as @xmath207 for @xmath208 , and satisfies the following ( quasi-)periodicity properties @xmath209 it is therefore sufficient to consider @xmath210 on the fundamental rectangle @xmath211 .
the zak transform is an isometry , i.e. , it satisfies @xmath212 the zak transform of @xmath206 in is given by @xmath213 where we used the substitution @xmath214 in and follows from @xmath215 .
next , we split the fundamental rectangle @xmath211 into @xmath92 cells @xmath99 , where @xmath216 was defined in section [ sec : prform ] in the context of structural assumptions imposed on the spreading function . concretely , we substitute @xmath217 in , with @xmath218 and @xmath219 . this yields , for @xmath220 and @xmath221 , @xmath222 where is a consequence of @xmath223 for @xmath224 , by assumption .
we next rewrite in vector - matrix form . to this end
, we define the column vectors @xmath225 and @xmath226 according to @xmath227_p { \coloneqq}{t}l\ , z_{p}(t , f)e^{-j2\pi p { t}f } , \quad p=0, ... ,l-1 \label{eq : defzvect}\ ] ] and @xmath228^t$ ] with @xmath163 as defined in . since @xmath223 for @xmath224 , the vector @xmath226 , @xmath229 , fully characterizes the spreading function @xmath4 . with these definitions
can now be written as @xmath230 with the @xmath231 matrix @xmath232 , \quad { \mathbf } a_{{\mathbf } c , k } { \coloneqq}{\mathbf } c_{{\mathbf } c , k } { { { \mathbf } f}^h } \label{eq : defac}\ ] ] where @xmath233_{p , m } = e^{-j2\pi \frac{pm}{l } } , \ ; p , m=0, ... ,l-1 $ ] , and @xmath234 is the @xmath235 diagonal matrix with diagonal entries @xmath236 ( recall that the coefficient sequence @xmath237 is @xmath92-periodic ) . since @xmath225 is obtained from the operator s response to the probing signal and @xmath226 fully determines the spreading function @xmath199 , we can conclude that the identification of @xmath108 has been reduced to the solution of a continuously indexed linear system of equations .
conceptually , for each pair @xmath229 , we need to solve a linear system of @xmath92 equations in @xmath203 unknowns . the proof is then completed by showing that this continuously indexed linear system of equations has a unique solution if @xmath238 .
more formally , we need to relate identifiability according to definition [ def : defstableidentification ] to solvability of the continuously indexed linear system of equations . to this end
, we first note that thanks to lemma [ pr : blind_identification ] , it suffices to prove identifiability of @xmath194 for all pairs @xmath195 with @xmath239 and @xmath240 . by setting @xmath241
this is equivalent to proving identifiability of @xmath242 for all @xmath96 with @xmath243 . for @xmath244 , by definition , @xmath245 .
denote the restriction of the vector @xmath226 to the entries corresponding to the active cells , i.e. , the cells indexed by @xmath169 , by @xmath246 and let @xmath247 be the matrix containing the columns of @xmath248 that correspond to the index set @xmath169 .
the linear system of equations then reduces to @xmath249 solvability of can now formally be related to identifiability through the following lemma , proven in appendix [ app : lem2 ] .
let @xmath111 be given by .
then , the ( tightest ) bounds @xmath250 in for the set of operators @xmath143 are given by @xmath251 [ le : boundedness_classr ] the proof of sufficiency in theorem [ th : sufficiency ] is now completed by showing that for all @xmath96 with @xmath243 , @xmath143 is identifiable , i.e. , @xmath252 . by lemma
[ le : boundedness_classr ] , @xmath253 trivially , and showing that @xmath254 amounts to proving that @xmath247 has full rank for all @xmath255 with @xmath256 , i.e. , for all @xmath96 such that @xmath243 .
what comes to our rescue is ( * ? ? ?
4 ) which states that for almost all @xmath204 , each is sufficient for identifiability .
the crucial difference between @xcite and our setup is that we need _ each _ submatrix of @xmath92 columns of @xmath248 to have full rank , as we do not assume prior knowledge of the support region . ]
@xmath235 submatrix of @xmath248 has full rank . in the remainder of the paper @xmath204
is chosen such that each @xmath235 submatrix of @xmath257 , indeed , has full rank .
in other words , @xmath204 is chosen such that @xmath258 .
the philosophy of operator identification without prior knowledge of the spreading function s support region is related to the idea of spectrum - blind sampling of multi - band signals @xcite . in spectrum - blind sampling
the central problem is to recover a signal , sparsely supported on a priori unknown frequency bands , from its samples taken at a rate that is ( much ) smaller than the shannon - nyquist rate of the signal .
the conceptual relation between operator identification and spectrum - blind sampling is brought out by comparing to the recovery equation in spectrum - blind sampling , given by @xcite @xmath259 here @xmath260 , with @xmath261 , depends on the sampling pattern , @xmath262 , fully specifies the signal to be reconstructed , and @xmath263 , is obtained from the samples of the signal .
further , @xmath264 is a spectral `` cell '' , playing a role similar to the cell @xmath99 in our setup .
it is shown in @xcite that the penalty for not knowing the spectral support set is a factor - of - two in sampling rate .
the corresponding result in the present paper is theorem [ th : sufficiency ] .
it is furthermore shown in @xcite that there is no penalty for not knowing the spectral support set if one requires recovery of _ almost all _ signals only .
the corresponding result in this paper is theorem [ th : almost_all ] . despite this strong structural similarity
, there is a fundamental difference between spectrum blind sampling and the system identification problem considered here . in operator identification a function of
_ two _ variables , @xmath4 , has to be extracted from the _ univariate _ measurement @xmath265 .
moreover , in spectrum - blind sampling there is no limit on the cardinality of the spectral support set that would parallel the @xmath238 or @xmath1 thresholds .
we next present an algorithm that provably recovers all @xmath156 for @xmath238 from the operator s response @xmath107 to the probing signal @xmath2 in .
the algorithm first identifies the support set of @xmath4 , i.e. , the index set @xmath169 , and then solves the corresponding linear system of equations , which , based on , yields @xmath4 . starting from ,
an explicit reconstruction formula for @xmath4 is straightforward to derive and is given by @xmath266_{l , p } \ , z_{p}(t , f)e^{-j2\pi \left ( p { t}f + \left(f+ \frac{m}{{t}l } \right)t \right ) } \label{eg : endpiecwise}\end{aligned}\ ] ] for @xmath267 , @xmath268 , where @xmath269 is the pseudoinverse of @xmath270 and the index @xmath21 refers to the row of @xmath269 corresponding to the @xmath271th cell .
we now turn our attention to the main challenge , namely support set recovery .
formally , is a continuously indexed linear system of equations , whose solutions ( across indices @xmath229 ) share the support set @xmath169 .
this problem was studied before under the name of `` infinite measurement vector problem '' in @xcite as a generalization of the multiple measurement vector ( mmv ) problem @xcite , where the reconstruction of a finite number of vectors sharing a sparsity pattern , from a finite number of linear measurements , is considered .
starting from the observation that the cardinality of the index set @xmath169 is finite , and the matrix @xmath272 is finite - dimensional , it is perhaps not surprising to see that the infinite measurement vector problem at hand can be reduced to an mmv problem .
based on the recovery equation , this was recognized before in the context of spectrum - blind sampling in @xcite and , in a more general context , in @xcite .
we next present a general reduction method , which unifies the approaches in @xcite and is based on a simplified , and , as we believe , more accessible treatment .
the discussion in section [ sec : redtommv ] below is therefore of interest in its own right .
we assume throughout that @xmath273 ; this is w.l.o.g . as @xmath273 corresponds to @xmath9 and
we only consider the identification of operators satisfying @xmath238 or @xmath1 .
the index set @xmath169 can be recovered as follows : @xmath274 where the minimization is performed over all @xmath255 and all corresponding @xmath275 .
the proof of @xmath276 delivering the correct solution is deferred to section [ se : uniqueness ] .
we first develop a unified approach to the reduction of the infinite measurement vector problem @xmath276 to an mmv problem .
we emphasize , as mentioned before , that this reduction approach encompasses the settings in @xcite and hence applies to spectrum - blind sampling , inter alia .
our approach is based on a basis expansion of the elements of @xmath225 and @xmath246 .
we start with some definitions .
consider the linear space of functions @xmath277 equipped with the inner product @xmath278 , @xmath279 , and induced norm @xmath280 .
let @xmath281 be a basis ( not necessarily orthogonal ) for the space spanned by the functions @xmath282_p , p=0, ... ,l-1\}$ ] and set @xmath283_p , p = 0, ... ,l-1\}$ ] .
we can represent @xmath284 in terms of the basis elements @xmath285 according to @xmath286 where @xmath287}^t}$ ] and @xmath288 contains the expansion coefficients of @xmath225 in the basis @xmath289 .
it follows from @xmath290_p , p = 0, ...
,l-1 \}$ ] that @xmath291 has full rank @xmath292 . to see this ,
suppose that @xmath293 . then , each set of @xmath294 rows of @xmath291 is linearly dependent , i.e. , for each set of rows of @xmath291 , indexed by say @xmath295 , with cardinality @xmath296 , there exists an @xmath297 , @xmath298 , such that @xmath299 where @xmath300 is the matrix obtained by retaining the rows of @xmath291 in @xmath295 .
then , for each @xmath295 with @xmath301 , according to , there exists an @xmath298 such that @xmath302 where @xmath303 contains the entries of @xmath225 corresponding to the index set @xmath295 .
this would , however , imply @xmath304_p , p = 0, ...
,l-1 \ } < k$ ] , which stands in contradiction to @xmath305_p , p = 0, ... ,l-1 \ } = k$ ] .
expanding @xmath306 in in the basis having full column rank , @xmath307 ( cf . and ) . ]
@xmath308 , we can rewrite the constraint in @xmath276 as @xmath309 where @xmath310 contains the expansion coefficients of @xmath246 in the basis @xmath308 . since the elements of @xmath311 form a basis , is equivalent to @xmath312
we have therefore shown that ( ) is equivalent to @xmath313 where the minimization is performed over all @xmath255 and all corresponding @xmath314 .
we have thus reduced @xmath276 , which involves a continuum of constraints , to @xmath315 , which involves only finitely many constraints .
@xmath315 is known in the literature as the mmv problem @xcite , which is usually formulated equivalently as : minimize @xmath316 subject to @xmath317 , where the constraint is over all @xmath318 and @xmath319 is the number of non - zero rows of @xmath320 .
we are now ready to explain the reduction approaches in @xcite in the general reduction framework just introduced .
we start with the method described in @xcite in the context of spectrum - blind sampling .
this approach starts from a correlation matrix , which in our setup becomes @xmath321 with we can express @xmath322 as @xmath323 where @xmath324 .
analogously to the results in ( * ? ? ?
1 ) for signal recovery in spectrum - blind sampling , it can be shown that ( ) is equivalent to @xmath325 where the minimization is performed over all @xmath255 and all corresponding hermitian @xmath326 .
we next show that @xmath327 is equivalent to an mmv problem , and then explain this equivalence result in our basis expansion approach .
@xmath322 is a hermitian matrix and can hence be decomposed as @xmath328 ( * ? ? ?
4.1.5 ) , where the @xmath329 columns of @xmath330 are orthogonal .
analogously to ( * ? ? ?
1 ) , ( * ? ? ?
v - c ) , it can now be shown that ( ) ( and by induction ( ) ) is equivalent to the mmv problem @xmath331 where the minimization is performed over all @xmath255 and all corresponding @xmath332 . to see how the reduction to @xmath333 just described can be cast into the basis expansion approach described above , let @xmath334 , where @xmath311 is an orthonormal basis for @xmath335_p , p=0, ... ,l-1\}$ ] . by
, we then have @xmath336 \,{{{\mathbf b}}^h}_z = { \mathbf b}_{z } { { { \mathbf b}}^h}_z.\ ] ] from @xmath337 it follows that there exists a unitary matrix @xmath338 such that @xmath339 , which is seen as follows .
we first show that any solution @xmath340 to @xmath341 can be written as @xmath342 , where @xmath343 is unitary ( * ? ? ?
* exercise on p. 406 ) .
indeed , we have @xmath344 where the last equality follows since @xmath345 is self adjoint , according to ( * ? ? ?
7.2.6 ) . from it
is seen that @xmath346 is unitary , and hence @xmath347 , with @xmath343 unitary .
therefore , we have @xmath348 and @xmath349 , where @xmath350 and @xmath351 are unitary , and hence @xmath352 .
as @xmath353 is unitary , we proved that there exists a unitary matrix @xmath338 such that @xmath339 .
with @xmath339 , the minimization variable of @xmath315 is given by @xmath354 , where @xmath355 is the minimization variable of @xmath333 , hence @xmath315 and @xmath333 are equivalent .
another approach to reducing @xmath276 to an mmv problem was put forward in ( * ? ? ?
2 ) . in our setting and notation the resulting mmv problem
is given by @xmath356 where the minimization is performed over all @xmath255 and all corresponding @xmath332 . here
, the matrix @xmath357 can be taken to be any matrix whose column span is equal to @xmath358 . to explain this approach in our basis expansion framework , we start by noting that implies that @xmath359 .
we can therefore take @xmath360 to equal @xmath361 .
on the other hand , for every @xmath360 with @xmath362 , we can find a basis @xmath311 such that @xmath363 . choosing different matrices
@xmath360 in @xmath364 therefore simply amounts to choosing different bases @xmath311 .
we are now ready to study uniqueness conditions for @xmath276 .
specifically , we will find a necessary and sufficient condition for @xmath276 to deliver the correct solution to the continuously indexed linear system of equations in .
this condition comes in the form of a threshold on @xmath365 that depends on the `` richness '' of the spreading function , specifically , on @xmath366 .
let @xmath367 , with @xmath368
. then @xmath276 applied to @xmath225 recovers @xmath369 if and only if @xmath370 [ thm : uniqueness_imv ] since @xmath371 , theorem [ thm : uniqueness_imv ] guarantees exact recovery if @xmath372 , and hence by @xmath373 ( see section [ sec : prform ] ) , if @xmath238 , which is the recovery threshold in theorem [ th : sufficiency ] . recovery for @xmath1 will be discussed later .
sufficiency in theorem [ thm : uniqueness_imv ] was shown in ( * ? ? ?
* prop . 1 ) and in the context of spectrum - blind sampling in ( * ? ? ?
necessity has not been proven formally before , but follows directly from known results , as shown in the proof of the theorem below . the proof is based on the equivalence of @xmath276 and @xmath315 , established in the previous section , and on the following uniqueness condition for the mmv problem @xmath315 .
let @xmath374 with @xmath375 .
then @xmath315 applied to @xmath291 recovers @xmath376 if and only if @xmath377 [ prop : cond_uniq_mmv ] sufficiency was proven in ( * ? ? ?
1 ) , ( * ? ? ?
1 ) , ( * ? ? ?
2.4 ) , necessity in ( * ? ? ?
we present a different , slightly simpler , argument for necessity in appendix [ app : neccmmv ] . in section [ sec : redtommv ] , we showed that @xmath378_p , p = 0, ... ,l-1 \ } = k$ ] implies @xmath379 .
the converse is obtained by essentially reversing the line of arguments used to prove this fact in section [ sec : redtommv ] .
we have therefore established that @xmath378_p , p = 0, ... ,l-1 \ } = { \operatorname{rank}\ ! \left({{\mathbf b}_z}\right ) } $ ] .
analogously , by using the fact that @xmath380 contains the expansion coefficients of@xmath381 in the basis @xmath308 , it can be shown that @xmath382 .
it now follows , by application of proposition [ prop : cond_uniq_mmv ] , that @xmath315 correctly recovers the support set @xmath169 if and only if is satisfied . by equivalence of @xmath315 and @xmath276 , @xmath276 recovers the correct support set , provided that is satisfied .
once @xmath169 is known , @xmath246 is obtained by solving .
solving the mmv problem @xmath315 is np - hard @xcite .
various alternative approaches with different performance - complexity tradeoffs are available in the literature .
mmv - variants of standard algorithms used in single measurement sparse signal recovery , such as orthogonal matching pursuit ( omp ) and @xmath383-minimization ( basis - pursuit ) can be found in @xcite .
however , the performance guarantees available in the literature for these algorithms fall short of allowing to choose @xmath365 to be linear in @xmath92 as is the case in the threshold . a low - complexity algorithm that provably yields exact recovery under the threshold in is based on ideas developed in the context of subspace - based direction - of - arrival estimation , specifically on the music - algorithm @xcite .
it was first recognized in the context of spectrum - blind sampling @xcite that a music - like algorithm can be used to solve a problem of the form @xmath276 .
the algorithm described in @xcite implicitly first reduces the underlying infinite measurement vector problem to a ( finite ) mmv problem .
recently , a music - like algorithm and variants thereof were proposed @xcite to solve the mmv problem @xmath315 directly . as we will see below
, this class of algorithms imposes conditions on @xmath376 and will hence not guarantee recovery for all @xmath376 .
we will present a ( minor ) variation of the music algorithm as put forward in @xcite , and used in the context of spectrum blind sampling ( * ? ? ?
1 ) , in section [ sec : almostall ] below .
for @xmath384 , theorem [ thm : uniqueness_imv ] hints at a potentially significant improvement over the worst - case threshold underlying theorem [ th : almost_all ] whose proof will be presented next .
the basic idea of the proof is to show that @xmath276 applied to @xmath367 , recovers the correct solution if the set @xmath385 condition implies that @xmath386 .
therefore , with @xmath387 in theorem [ thm : uniqueness_imv ] , we get that @xmath276 delivers the correct solution if @xmath388 , i.e. , if @xmath389 , which is guaranteed by @xmath390 .
we next present an algorithm that provably recovers @xmath391 with @xmath14 , i.e. , almost all @xmath158 with @xmath14 . specifically , this low - complexity music - like algorithm solves is np - hard ( as noted before ) , since it `` only '' solves @xmath276 for _ almost all _ @xmath392 . ]
@xmath315 ( which is equivalent to @xmath276 ) and can be shown to identify the support set @xmath169 correctly for @xmath1 provided that condition is satisfied .
the algorithm is a minor variation of the music algorithm as put forward in @xcite , and used in the context of spectrum blind sampling ( * ? ? ?
the following algorithm recovers all @xmath391 , provided that @xmath14
. 1 . given the measurement @xmath284 , find a basis ( not necessarily orthogonal ) @xmath308 , for the space spanned by @xmath282_p , p=0, ... ,l-1\}$ ] , where @xmath393_p , p=0, ...
,l-1 \}$ ] , and determine the coefficient matrix @xmath291 in the expansion @xmath334 .
2 . compute the matrix @xmath394 of eigenvectors of @xmath395 corresponding to the zero eigenvalues of @xmath396 .
3 . identify @xmath169 with the indices corresponding to the columns of @xmath397 that are equal to @xmath398 .
[ thm : music_recovery ] in the remainder of the paper , we will refer to steps @xmath399 and @xmath400 above as the mmv - music algorithm . as shown next
, the mmv - music algorithm provably solves the mmv problem @xmath315 given that @xmath380 has full rank @xmath365 . the proof is effected by establishing that for @xmath14 under condition the support set @xmath169 is uniquely specified through the indices of the columns of @xmath397 that are equal to @xmath398 . to see this ,
we first obtain from ( where @xmath401 and @xmath402 are as defined in section [ sec : reconstruct ] ) @xmath403 next , we perform an eigenvalue decomposition of @xmath396 in to get @xmath404 where @xmath405 contains the eigenvectors of @xmath406 corresponding to the non - zero eigenvalues of @xmath406 . as mentiond in section [ sec : suff ] , each set of @xmath92 or fewer columns of @xmath248 is necessarily linearly independent , if @xmath407 is chosen judiciously .
hence @xmath247 has full rank if @xmath273 , which is guaranteed by @xmath408 .
thanks to condition , @xmath409 and hence @xmath410 ( this was shown in the proof of theorem [ thm : uniqueness_imv ] ) , which due to @xmath411 implies that @xmath412 .
consequently , we have @xmath413 where the second equality follows from .
@xmath414 is the orthogonal complement of @xmath415 in @xmath416 .
it therefore follows from that @xmath417 . hence , the columns of @xmath397 that correspond to indices @xmath267 are equal to @xmath398 .
it remains to show that no other column of @xmath397 is equal to @xmath398 .
this will be accomplished through proof by contradiction .
suppose that @xmath418 where @xmath419 is any column of @xmath248 corresponding to an index pair @xmath420 . since @xmath414 is the orthogonal complement of @xmath415 in @xmath416 , @xmath421 .
this would , however , mean that the @xmath92 or fewer columns of @xmath248 corresponding to the indices @xmath422 would be linearly dependent , which stands in contradiction to the fact that each set of @xmath92 or fewer columns of @xmath248 must be linearly independent .
the results presented thus far rely on probing signals of infinite bandwidth and infinite duration .
it is therefore sensible to ask whether identification under a bandwidth - constraint on the probing signal and under limited observation time of the operator s corresponding response is possible .
we shall see that the answer is in the affirmative with the important qualifier of identification being possible up to a certain resolution limit ( dictated by the time- and bandwidth constraints ) .
the discretization through time- and band - limitation underlying the results in this section will involve approximations that are , however , not conceptual .
the discussion in this section serves further purposes .
first , it will show how the setups in @xcite can be obtained from ours through discretization induced by band - limiting the input and time - limiting and sampling the output signal .
more importantly , we find that , depending on the resolution induced by the discretization , the resulting recovery problem can be an mmv problem .
the recovery problem in @xcite is a standard ( i.e. , single measurement ) recovery problem , but multiple measurements improve the recovery performance significantly , according to the recovery threshold in theorem [ thm : uniqueness_imv ] , and are crucial to realize recovery beyond @xmath423 .
second , we consider the case where the support area of the spreading function is ( possibly significantly ) below the identification threshold @xmath238 , and we show that this property can be exploited to identify the system while undersampling the response to the probing signal . in the case of channel identification , this allows to reduce identification time , and in radar systems it leads to increased resolution .
consider an operator @xmath158 and an input signal @xmath2 that is band - limited to @xmath424 , and perform a time - limitation of the corresponding output signal @xmath425 to @xmath426 .
then , the input - output relation becomes ( for details , see , e.g. @xcite ) @xmath427 for @xmath428 , where @xmath429 band - limiting the input and time - limiting the corresponding output hence leads to a discretization of the input - output relation , with `` resolution '' @xmath430 in @xmath431-direction and @xmath432 in @xmath433-direction .
it follows from that for a compactly supported @xmath4 the corresponding quantity @xmath434 will not be compactly supported .
most of the volume of @xmath434 will , however , be supported on @xmath435 , so that we can approximate by restricting summation to the indices @xmath436 satisfying @xmath437 . note that the quality of this approximation depends on the spreading function as well as on the parameters @xmath438 , and @xmath92 . here
, we assume that @xmath439 and @xmath440 .
these constraints are not restrictive as they simply mean that we have at least one sample per cell .
we will henceforth say that @xmath108 is identifiable with resolution @xmath441 , if it is possible to recover @xmath442 , for @xmath437 , from @xmath206 .
we will simply say `` @xmath108 is identifiable '' whenever the resolution is clear from the context . in the ensuing discussion @xmath443 , is referred to as the discrete spreading function .
the maximum number of non - zero coefficients of the discrete spreading function to be identified is @xmath444 .
next , assuming that @xmath95 , as defined in , satisfies @xmath445 , it follows that @xmath206 is approximately band - limited to @xmath424 . from @xcite we
can therefore conclude that @xmath446 lives in a @xmath447-dimensional signal space ( here , and in the following , we assume , for simplicity , that @xmath447 is integer - valued ) , and can hence be represented through @xmath447 coefficients in the expansion in an orthonormal basis for this signal space .
the corresponding basis functions can be taken to be the prolate spheroidal wave functions @xcite . denoting the vector containing the corresponding expansion coefficients by @xmath448 ,
the input - output relation becomes @xmath449 where the columns of @xmath450 contain the expansion coefficients of the time - frequency translates @xmath451 in the prolate spheroidal wave function set , and @xmath452 contains the samples @xmath453 for @xmath454 , of which at most @xmath444 are non - zero , with , however , unknown locations in the @xmath13-plane .
we next show that the recovery threshold @xmath238 continues to hold , independently of the choice of @xmath455 and @xmath456 .
[ [ necessity ] ] necessity + + + + + + + + + it follows from ( * ? ? ?
1 ) that @xmath457 is necessary to recover @xmath458 from @xmath448 given @xmath459 . with @xmath460 and @xmath461 , which follows trivially
, we trivially have @xmath462 . ]
since @xmath459 is of dimension @xmath463 , we get @xmath464 and hence @xmath465 . since
, by definition , @xmath146 we have shown that @xmath238 is necessary for identifiability .
[ [ sufficiency ] ] sufficiency + + + + + + + + + + + sufficiency will be established through explicit construction of a probing signal @xmath48 and by sampling the corresponding output signal @xmath206 . since @xmath446 is ( approximately ) band - limited to @xmath424 , we can sample @xmath446 at rate @xmath455 , which results in @xmath466 for @xmath467 . in the following ,
denote the number of samples of @xmath434 per cell @xmath468 , in @xmath431-direction as @xmath469 and in @xmath433-direction as @xmath470 ; see figure [ fig : disctnp ] for an illustration .
note that , since @xmath471 , and @xmath434 is sampled at integer multiples of @xmath430 in @xmath431-direction and of @xmath432 in @xmath433-direction , we have @xmath472 and @xmath473 .
the number of samples per cell @xmath474 will turn out later to equal the number of measurement vectors in the corresponding mmv problem . to have multiple measurements , and hence make identification beyond @xmath475 possible , it is therefore necessary that @xmath447 is large relative to @xmath92 .
as mentioned previously , the samples @xmath442 , for @xmath437 , fully specify the discrete spreading function .
we can group these samples into the active cells , indexed by @xmath169 , by assigning @xmath476 , for @xmath477 , to the cell with index @xmath478 , where @xmath479 , and @xmath480 .
discretization of the @xmath13-plane , with @xmath481 . ]
the probing signal @xmath48 is taken to be such that @xmath482 where the coefficients @xmath483 , are chosen as discussed in section [ sec : suff ] .
note that the sequence @xmath484 is @xmath485-periodic .
algebraic manipulations yield the discrete equivalent of as @xmath486 = { \mathbf } a_{{\mathbf } c } { { \mathbf s}}[n , r ] , \quad ( n , r ) \in { u_d}. \label{eq : syseqdisc}\ ] ] here , @xmath248 was defined in , and @xmath487 \right]_p { \coloneqq}{z}_p[n , r ] e^{-j2\pi \frac{rp}{{d}l } } , \quad p = 0 , ... , l-1\ ] ] with @xmath488 { \coloneqq}z_{y}^{({e}l,{d})}[n+{e}p , r ] $ ] , where @xmath489 { \coloneqq}\frac{1}{{d } } \sum_{q = 0}^{{d}-1 } y[n + { e}l q ] e^{-j 2 \pi \frac{qr}{{d } } } , \ ] ] for @xmath490 , is the discrete zak transform ( with parameter @xmath485 ) @xcite of the sequence @xmath491 $ ] . for general properties of the discrete zak transform we refer to @xcite .
further , @xmath492 { \coloneqq}[{s}_{0,0}[n , r],\allowbreak { s}_{0,1}[n , r ] , \allowbreak ... , \allowbreak { s}_{0,l-1}[n , r ] , { s}_{1,0}[n , r],\allowbreak ... , { s}_{l-1,l-1}[n , r ] ] ^t$ ] with @xmath493 { \coloneqq}\overline{{s_{\hspace{-0.02cm}h } } } \!\left(\frac{n+{e}k}{{b}},\frac { r+{d}m}{{v } } \right ) e^{j2 \pi \frac{n ( r+{d}m)}{{e}{d}l } } \label{def : skmdiscrete}\ ] ] for @xmath494 . note that @xmath495 , ( n , r)\in { u_d } , ( k , m ) \in \gamma$ ] , fully specifies the discrete spreading function .
the identification equation can be rewritten as @xmath496 where the columns of @xmath497 and @xmath498 are given by the vectors @xmath499 $ ] and @xmath500 $ ] , respectively , @xmath501 .
hence , each row of @xmath502 corresponds to the samples of @xmath503 in one of the @xmath203 cells .
since the number of samples per cell , @xmath504 , is equal to the number of columns of @xmath502 , we see that the number of samples per cell corresponds to the number of measurements in the mmv formulation .
denote the matrix obtained from @xmath502 by retaining the rows corresponding to the active cells , indexed by @xmath169 , by @xmath505 and let @xmath247 be the matrix containing the corresponding columns of @xmath248 . then becomes
@xmath506 once @xmath169 is known , can be solved for @xmath505 . hence , recovery of the discrete spreading function amounts to identifying @xmath169 from the measurements @xmath396 , which can be accomplished by solving the following mmv problem : @xmath507 where the minimization is performed over all @xmath255 and all corresponding @xmath508 .
it follows from proposition [ prop : cond_uniq_mmv ] that @xmath169 is recovered exactly from @xmath396 by solving @xmath509 , whenever @xmath510 , where @xmath511 .
correct recovery is hence guaranteed whenever @xmath512 . since @xmath513 and @xmath146
, this shows that @xmath238 is sufficient for identifiability . as noted before
, @xmath509 is np - hard . however , if @xmath514 then mmv - music provably recovers @xmath169 with @xmath515 , i.e. , when @xmath1 ( this was shown in the proof of theorem [ thm : music_recovery ] ) . for @xmath508 to have full rank @xmath365 ,
it is necessary that the number of samples satisfy @xmath516 . for @xmath516
almost all @xmath505 have full rank @xmath365 .
the development above shows that the mmv aspect of the recovery problem is essential to get recovery for values of @xmath8 beyond @xmath517 .
we conclude this discussion by noting that the setups in @xcite in the context of channel estimation and in @xcite in the context of compressed sensing radar are structurally equivalent to the discretized operator identification problem considered here , with the important difference of the mmv aspect of the problem not being brought out in @xcite . in the preceding sections
, we showed under which conditions identification of an operator is possible if the operator s spreading function support region is not known prior to identification .
we now turn to a related problem statement that is closer to the philosophy of sparse signal recovery , where the goal is to reconstruct sparse objects , such as signals or images , by taking fewer measurements than mandated by the object s `` bandwidth '' .
we consider the discrete setup and assume that @xmath8 is ( possibly significantly ) smaller than the identifiability threshold @xmath517 .
concretely , set @xmath518 for an integer @xmath519 .
we ask whether this property can be exploited to recover the discrete spreading function from a subset of the samples @xmath520 , @xmath521 only .
we will see that the answer is in the affirmative , and that the corresponding practical implications are significant , as detailed below . for concreteness , we assume that @xmath522 , with @xmath523 . to keep the exposition simple ,
we take @xmath524 , in which case @xmath525 becomes a vector .
note that , since @xmath526 ( this follows from @xmath527 ) , the operator can be identified by simply solving @xmath528 for @xmath525 , which we will refer to as `` reconstructing conventionally '' .
here @xmath529 and @xmath525 contain the columns of @xmath530 and rows of @xmath502 , respectively , corresponding to the indices in @xmath295 .
since @xmath518 , the area @xmath97 of the ( unknown ) support region @xmath96 of the spreading function satisfies @xmath531 .
we next show that the discrete spreading function can be reconstructed from only @xmath532 of the @xmath92 rows of @xmath396 .
the index set corresponding to these @xmath532 rows is denoted as @xmath32 , and is an ( arbitrary ) subset of @xmath533 ( of cardinality @xmath532 ) .
let @xmath534 and @xmath535 be the matrices corresponding to the rows of @xmath396 and @xmath536 , respectively , indexed by @xmath32 .
the matrix @xmath534 is a function of the samples @xmath537 only ; hence , reconstruction from @xmath534 amounts to reconstruction from an undersampled version of @xmath206 .
note that we can not reconstruct the discrete spreading function by simply inverting @xmath538 since @xmath539 is a wide matrix is of limited interest and will not be considered . ] .
next , implies ( see also ) that @xmath540 theorem 4 in @xcite establishes that for almost all @xmath204 , @xmath541 .
hence , according to proposition [ prop : cond_uniq_mmv ] , @xmath505 can be recovered uniquely from @xmath534 provided that @xmath542 and hence @xmath543 , by solving @xmath544 where the minimization is performed over all @xmath255 and all corresponding @xmath545 .
we have shown that identification from an undersampled observation @xmath206 is possible , and the undersampling factor can be as large as @xmath546 . a similar observation has been made in the context of radar imaging @xcite .
recovery of @xmath505 from @xmath534 has applications in at least two different areas , namely in radar imaging and in channel identification . [
[ increasing - the - resolution - in - radar - imaging ] ] increasing the resolution in radar imaging + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in radar imaging , targets correspond to point scatterers with small dispersion in the delay - doppler plane . since the number of targets is typically small , the corresponding spreading function is sparsely supported @xcite . in our model
, this corresponds to a small number of the @xmath547 in being non - zero .
take @xmath548 .
the discussion above then shows that , since only the samples @xmath549 , which in turn only depend on @xmath206 for @xmath550 $ ] , are needed for identification , it is possible to identify the discrete spreading function from the `` effective '' observation interval @xmath551 , while keeping the resolution in @xmath433-direction at @xmath432 .
if we were to reconstruct conventionally , given only the observation of @xmath206 over the interval @xmath551 , the induced resolution in @xmath433-direction ( see figure [ fig : disctnp ] ) would only be @xmath552 .
[ [ saving - degrees - of - freedom - in - channel - identification ] ] saving degrees of freedom in channel identification + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + next , consider the problem of channel identification , and take again @xmath548 . as discussed before
, @xmath534 is a function of the samples @xmath537 only , which , by careful inspection of , are seen to depend only on @xmath553 .
we can therefore conclude that it suffices to observe @xmath206 over the interval @xmath551 .
conceptually , this means that the time needed to identify ( learn ) the channel is reduced , which leaves , e.g. , more degrees of freedom to communicate over the channel .
we present numerical results quantifying the impact of additive noise and of the choice of @xmath407 on the performance of different identification algorithms . specifically , we consider the discrete setting and evaluate two probing sequences . the first one is obtained by sampling i.i.d . uniformly from the complex unit disc , the resulting sequence
is denoted by @xmath554 . since for almost all @xmath554 , each @xmath235 submatrix of @xmath530 has full rank for @xmath92 prime ( * ? ? ?
4 ) , @xmath554 will allow recovery for all operators with @xmath238 and for almost all operators with @xmath1 , in both cases , with probability one , with respect to the choice of @xmath554 .
the second probing sequence is the alltop sequence prime ) . ]
@xcite , denoted by @xmath555 , and defined as @xmath556_i = \frac{1}{\sqrt{l } } e^{\frac{j2\pi } { l } i^3 } , \quad i = 0, ... ,l-1.\ ] ] we compare two different algorithms for solving the mmv problem @xmath509 , namely the mmv - orthogonal matching pursuit ( mmv - omp ) algorithm as proposed in @xcite , and mmv - music - norm smaller than a certain threshold , which in turn depends on the noise level . ]
as introduced in section [ sec : almostall ] .
we generate the samples @xmath547 at random , as follows .
we choose is that we want @xmath92 to be prime , as by @xcite this guarantees that for almost all @xmath407 , each @xmath235 submatrix of @xmath530 has full rank . ]
@xmath557 , and vary the support set size @xmath558 and the number of samples per cell , @xmath504 . for fixed @xmath558 , and hence fixed @xmath365 , we draw @xmath255 uniformly at random from the set of all support sets with cardinality @xmath365 , and assign i.i.d .
@xmath559 values to each of the @xmath504 samples in each of the corresponding cells . to analyze the impact of noise
, we contaminate the measurement ( i.e. , @xmath560 in ) by i.i.d .
gaussian noise .
recovery performance in the noisy case is quantified through the empirical relative squared error in the discrete spreading function , abbreviated as ere , which is the empirical expectation of the relative squared error . in the noiseless setting , recovery success is declared if the relative squared error in the spreading function is less than or equal to @xmath561 .
recovery probabilities and the ere were obtained from 1000 realizations of @xmath169 .
[ [ impact - of - probing - sequence ] ] impact of probing sequence + + + + + + + + + + + + + + + + + + + + + + + + + + the results for the noiseless case , depicted in figure [ fig : emprec ] , show that the probing sequences @xmath555 and @xmath554 perform almost equally well .
we can see , as predicted by theorem [ thm : music_recovery ] , that mmv - music succeeds for @xmath14 , provided that @xmath562 . specifically , as shown in the proof of theorem [ thm : music_recovery ] , mmv - music succeeds if @xmath505 has full rank , which is the case with probability one if @xmath562 , as the entries of @xmath505 are i.i.d .
@xmath559 . for @xmath563
, mmv - music fails .
the performance of mmv - omp improves in @xmath504 ; however , the improvement stagnates at about @xmath564 .
for @xmath524 , mmv - omp outperforms mmv - music , for all other values of @xmath504 considered mmv - music outperforms mmv - omp .
recovery probabilities for the alltop sequence and for a randomly generated sequence , for the mmv - music algorithm in section [ sec : almostall ] and the mmv - omp @xcite algorithm . ]
[ [ impact - of - noise ] ] impact of noise + + + + + + + + + + + + + + + the results depicted in figure [ fig : noisy ] show that the identification process exhibits noise robustness up to @xmath565 . when @xmath562 , the error in recovering the spreading function is small for both identification algorithms , but mmv - music outperforms mmv - omp significantly .
the results in figure [ fig : noisy_2 ] quantify the noise sensitivity of mmv - music and mmv - omp .
ere for the alltop sequence and for a randomly chosen sequence , obtained by mmv - music and mmv - omp @xcite at @xmath566 . ] ere for a randomly chosen sequence , and @xmath567 , obtained by mmv - music and mmv - omp @xcite for different snr values . ]
the authors would like to thank g. pfander and v. morgenshtern for interesting discussions .
the inverse @xmath568 of the linear operator @xmath569 where @xmath570 is the range of @xmath93 , exists and is bounded if and only if @xmath93 is bounded below , in the following sense : there exists an @xmath571 such that for all @xmath114 , @xmath572 the proof corresponding to the case where @xmath104 satisfies @xmath573 for all @xmath114 is standard , see e.g. @xcite . for @xmath574
, the proof follows the same steps with minor modifications .
we first show that implies bounded invertibility of @xmath93 . if @xmath575 , then from @xmath576 and hence necessarily @xmath577 , which shows that @xmath93 is injective . since according to , the domain of the inverse is @xmath570 , @xmath93 is also surjective , and hence @xmath93 is invertible . to show boundedness of @xmath578 , set @xmath579 and @xmath580 for @xmath581
. using , we get @xmath582 which is @xmath583 and hence shows that @xmath578 is bounded .
we next show that bounded invertibility of @xmath93 implies . since @xmath578 exists and is bounded , we have , for @xmath571 , @xmath584
starting from , we get , for fixed @xmath268 , @xmath585 squaring @xmath586 and integrating over @xmath99 yields @xmath587_p \right|^2 d(t , f ) \nonumber \\ & = \sum_{p = 0}^{l-1 } \int _ { { u } } ( { t}l)^2 \left| z_p(t , f ) \right|^2 d(t , f ) \label{eq : usedefzv } \\ & = ( { t}l)^2 \int_0^{{t}l}\!\!\!\ !
\int_0^{1/({t}l ) } |{\mathcal } z_{y } ( t , f ) |^2 d(t , f ) \label{eq : usedefzp } \\ & = { t}l \ , { \ensuremath{{\left\vert h { x}\right\vert}_{}}}^2 \label{eq : hzeq}\end{aligned}\ ] ] where we used and for and , respectively , and follows since the zak transform is an isometry ( see ) .
similarly , based on we get @xmath588 where the last equality follows from . combining and with yields @xmath589 with @xmath590 which concludes the proof .
to prove necessity of , we show that one can construct a solution @xmath591 to @xmath315 applied to @xmath592 with @xmath593 . for any set @xmath295 of column indices of @xmath530 , with cardinality @xmath594 , we have that @xmath595 has full rank @xmath92 , as each set of @xmath92 columns of @xmath596 is linearly independent ( as discussed previously , according to ( * ? ? ?
4 ) this holds for almost all @xmath407 , and we assume that @xmath407 is chosen accordingly ) , and hence @xmath597 .
we can therefore conclude that there exists a matrix @xmath598 with @xmath599 such that @xmath600 we next construct index sets @xmath601 with @xmath602 and @xmath603 .
since @xmath599 , there exists a set of @xmath294 linearly independent rows of @xmath604 .
let @xmath605 be the index set corresponding to these rows augmented by the indices corresponding to @xmath606 arbitrary rows of @xmath604 , and set @xmath607 . by construction
, the matrix formed by the rows indexed by @xmath605 , @xmath608 , satisfies @xmath609 . from , with @xmath380 defined through @xmath610}^t}$ ] ,
we have @xmath611 it therefore follows from that ( @xmath612 ) is consistent with @xmath613 , which concludes the proof .
p. feng and y. bresler , `` spectrum - blind minimum - rate sampling and reconstruction of multiband signals , '' in _ proc . of ieee int .
speech sig .
( icassp ) _ , vol . 3 , atlanta , ga , usa , may 1996 , pp
. 16881691 .
g. taubck , f. hlawatsch , d. eiwen , and h. rauhut , `` compressive estimation of doubly selective channels in multicarrier systems : leakage effects and sparsity - enhancing processing , '' _ ieee j. sel .
topics signal process . _ ,
vol . 4 , no . 2 ,
pp . 255271 , 2010 .
w. u. bajwa , k. gedalyahu , and y. c. eldar , `` identification of parametric underspread linear systems and super - resolution radar , '' _ ieee trans .
signal process .
_ , vol . 59 , no . 6 , pp . 25482561 , jun
2011 .
h. g. feichtinger and g. zimmermann , `` a banach space of test functions for gabor analysis , '' in _ gabor analysis and algorithms _ , ser . applied and numerical harmonic analysis , h. g. feichtinger and t. strohmer , eds .
1em plus 0.5em minus 0.4embirkhuser boston , jan .
1998 , pp .
123170 .
s. cotter , b. rao , k. engan , and k. kreutz - delgado , `` sparse solutions to linear inverse problems with multiple measurement vectors , '' _ ieee trans .
signal process .
_ , vol .
53 , no . 7 , pp .
24772488 , 2005 . | we consider the problem of identifying a linear deterministic operator from its response to a given probing signal . for a large class of linear operators ,
we show that stable identifiability is possible if the total support area of the operator s spreading function satisfies @xmath0 .
this result holds for an arbitrary ( possibly fragmented ) support region of the spreading function , does not impose limitations on the total extent of the support region , and , most importantly , does not require the support region to be known prior to identification .
furthermore , we prove that stable identifiability of _ almost all operators _ is possible if @xmath1 .
this result is surprising as it says that there is no penalty for not knowing the support region of the spreading function prior to identification .
algorithms that provably recover all operators with @xmath0 , and almost all operators with @xmath1 are presented . |
lanthanide based heavy fermion ( hf ) metals constitute a major , long studied class of correlated electron materials @xcite .
their behaviour is quite distinct from conventional clean metals , the basic physics being driven by strong spin - flip scattering from essentially localised @xmath3-levels , generating the large effective mass and attendant low - energy scale indicative of strong interactions .
the low - temperature ( @xmath4 ) state is a lattice - coherent fermi liquid with well defined quasiparticles and coherently screened @xmath3-spins , crossing over with increasing @xmath4 to essentially incoherent screening via independent kondo scattering , before attaining characteristic clean metallic behaviour .
physical properties of hf are in consequence typically ` anomalous ' : e.g. the resistivity @xmath5 shows a strong non - monotonic @xmath4-dependence , while optics often exhibit rich structure from the microwave to the near infrared , and pronounced evolution on low temperature scales @xcite .
theoretical treatments of hf centre on the periodic anderson model ( pam ) , in which a non - interacting conduction band hybridizes locally with a correlated @xmath3-level in each unit cell of the lattice ; or on its strong coupling limit , the kondo lattice model .
the absence of exact results ( save for some in one dimension , see e.g. @xcite ) has long spurred the search for suitable approximation schemes .
one such framework , which has had a major impact in recent years , is provided by dynamical mean field theory ( dmft , for reviews see @xcite ) . formally exact in the large - dimensional limit
, the self - energy within dmft becomes momentum independent and hence spatially local , but still retains full temporal dynamics ; such that all lattice models map onto an effective single - impurity model with a self - consistently determined host @xcite .
that raises an immediate question , easier asked than answered : to what extent are the properties of real hf materials captured within a dmft approach to the pam ? to answer this clearly requires direct quantitative comparsion of theory to experiment . and
a prerequisite to that in turn is a method to solve the pam which dmft does not _
per se _ provide .
the latter has of course been studied extensively using a wide variety of techniques .
full scale numerical methods include the numerical renormalization group ( nrg ) @xcite , quantum monte carlo @xcite and exact diagonalization @xcite , while theoretical approaches encompass finite - order perturbation theory in the interaction @xmath6 @xcite , iterated perturbation theory @xcite , the lattice non - crossing approximation @xcite and the average @xmath7-matrix approximation @xcite , large-@xmath8 mean - field theory / slave bosons @xcite , the gutzwiller variational approach @xcite and the recently developed local moment approach @xcite .
all of these methods naturally have their own virtues .
but most possess significant , well known limitations @xcite , be it the general inability of perturbative approaches ( and in practice quantum monto carlo ) to handle strong interactions ; failure to recover fermi liquid behaviour at low - energies as arises in nca - based approaches , restriction to the lowest - energy fermi liquid behaviour as in large-@xmath8/slave boson mean - field theories , finite - size effects limiting exact diagonalization , and so on .
to enable viable comparison to experiment requires an approach that can adequately handle all experimentally relevant energy and/or temperature scales in the strongly correlated hf regime of primary interest ; and indeed ideally also across the full spectrum of interaction strengths , such that intermediate valence and related behaviour can likewise be treated .
one such is employed here , the local moment approach ( lma ) @xcite . via study of the generic asymmetric pam ,
our essential aims are ( i ) to provide a many - body description of dynamical and transport properties of paramagnetic hf , notably single - particle dynamics , d.c .
transport and optical conductivities ; as considered here .
( ii ) to make direct quantitative comparison with experiment . that is taken up in the following paper where comparison to transport / optical properties of @xmath9 , @xmath10 , @xmath11 and @xmath12 is made .
some remarks on the lma are apposite at this point since the paper will focus mainly on results obtained using the approach , with minimal technical details .
intrinsically non - perturbative and as such capable of handling strong interactions , the lma @xcite introduces the physically intuitive notion of local moments @xcite from the outset .
this leads directly to a ` two - self - energy ' description in which , post mean - field level , the key correlated spin - flip dynamics is readily captured ; corresponding in physical terms to dynamical tunneling between initially degenerate local moment configurations , which lifts the erstwhile spin degeneracy and restores the local singlet symmetry characteristic of a fermi liquid state . as with all techniques for lattice models within dmft , the lma originated in study of the single - impurity anderson model ( aim ) @xcite , where results for dynamics are known to give good agreement with nrg calculations @xcite , and for static magnetic properties with known exact results @xcite
the approach has recently been developed to encompass the anderson lattice ( pam ) ; initially for the particle - hole symmetric limit @xcite appropriate to the kondo insulating sub - class of heavy electron materials , where for all interaction strengths the system is an ` insulating fermi liquid ' that evolves continuously from its simple non - interacting limit of a hybridization - gap insulator @xcite . from this
a rich description of transport and optical properties of kondo insulators arises @xcite , particularly in strong coupling where the system is characterized by an exponentially small indirect gap scale @xmath13 , such that dynamics / transport exhibit scaling as functions of @xmath14 . exploiting that scaling enables direct comparison to experiment with minimal use of ` bare ' material / model parameters ; and in particular for three classic kondo insulators @xmath15 , @xmath16 and @xmath17 , leads to what we regard as excellent agreement between theory and experiment on essentially all relevant energy and temperature scales @xcite .
the particle - hole symmetric pam is of course special , confined as it is to the case of kondo insulators .
most recently the lma has been non - trivially extended to handle the generic asymmetric pam @xcite and hence hf metals ( with the insulating symmetric limit recovered simply as a particular case ) .
single - particle dynamics at @xmath18 were considered in @xcite , with a natural emphasis on the strongly correlated kondo lattice regime of localised @xmath3-electrons but general conduction ( ` @xmath19 ' ) band filling @xmath20 .
the problem was found to be characterized by a _
single _ low - energy coherence scale @xmath1 the precise counterpart of the insulating indirect gap scale @xmath21 , and likewise exponentially small in strong coupling in terms of which dynamics exhibit one - parameter universal scaling as a function of @xmath22 , independently of either the interaction strength or local @xmath23 hybridization . with increasing @xmath24 dynamics cross over from the low - energy quasiparticle behaviour required by and symptomatic of the coherent fermi liquid state , to essentially incoherent single - impurity kondo scaling physics at high-@xmath24
but still in the @xmath24-scaling regime and as such incompatible @xcite with a two - scale ` exhaustion ' scenario @xcite . in this paper
we extend the work of @xcite to finite temperature , thereby enabling access to d.c .
transport and optics .
our primary focus is again the strongly correlated hf regime and attendant issues of scaling / universality ( that play a key role in comparing to experiment ) , the paper being organised as follows . the model and a bare bones description of background theory is introduced in section 2 , together with preliminary consideration of transport / optics .
results for the thermal evolution of single - particle dynamics and scattering rates , and the connection between the two , are given in section 3 .
the d.c .
resistivity is considered in section 4 , with particular emphasis in this context on the crossover from the low-@xmath25 coherent fermi liquid to the high-@xmath26 incoherent regime , and explicit connection to single - impurity scaling behaviour .
optical conductivities on all relevant @xmath27- and @xmath4-scales are investigated in section 5 ; and the paper concludes with a brief summary .
the hamiltonian for the pam is given by @xmath28 : = _ c_i , c^_i c^_i & -&t_(i , j),c^_i c^_i + _
i,(_f+ f^_i - f^_i- ) f^_if^_i + & + & v_i , ( f^_i c^_i + ) [ 2.1 ] the first two terms represent the uncorrelated conduction ( @xmath19 ) band , + @xmath29 ( @xmath30 ) ; with @xmath19-orbital site energies @xmath31 and nearest - neighbour hopping matrix element @xmath32 , rescaled as @xmath33 in the large dimensional limit where the coordination number @xmath34 @xcite .
the third term describes the correlated @xmath3-levels , @xmath35 , with site energies @xmath36 and on site coulomb repulsion @xmath6 ; while the final term @xmath37 hybridizes the @xmath19- and @xmath3-levels locally via the matrix element @xmath38 , rendering the otherwise localised @xmath3-electrons itinerant .
the model is thus characterized by four independent dimensionless parameters , @xmath39 and @xmath40 ( @xmath41 sets the scale for the width of the free conduction band and is taken as the basic unit of energy , @xmath42 ) . an equivalent and somewhat more convenient set of ` bare'/material parameters ( with @xmath43 ) is @xmath44 and @xmath45 , where @xmath46 .
this parameter space is large , and as such encompasses a wide range of physical behaviour for the paramagnetic phases we consider .
the system is of course generically metallic , with non - integral @xmath3-level and @xmath19-band occupancies ( @xmath47 and @xmath48 respectively ) .
that in turn extends from the trivial case of weakly correlated , perturbative behaviour , through intermediate valence to the strongly correlated heavy fermion ( hf ) regime .
it is naturally the latter , characterized by a low - energy coherence scale @xmath49 , that is of primary interest .
the hf ( or kondo lattice ) regime corresponds to essentially localised @xmath3-electrons , @xmath50 , but with arbitrary conduction band filling @xmath51 , the latter being controlled by @xmath52 ( which determines the centre of gravity of the free ( @xmath53 ) conduction band relative to the fermi level ) .
it arises when @xmath54 , for @xmath55 and @xmath56 ( whence @xmath57 ) ; where @xmath58 , with @xmath59 the free conduction electron density of states as specified below and @xmath60 the fermi level .
the heavy fermion regime forms our main focus here ; intermediate valence behaviour will be discussed in an experimental context in the following paper .
the exception to the above behaviour arises when @xmath61 . here
the system is generically a kondo insulator ( see _ eg @xcite ) , with an indirect gap in both its @xmath18 single - particle spectrum _ and _ optical conductivity @xcite ; the canonical example being the particle - hole symmetric pam with @xmath62 and @xmath63 , where @xmath64 for all @xmath6 . just like its metallic counterpart arising for @xmath65 ,
the kondo insulator is however a fermi liquid , evolving continuously with increasing interaction strength from its non - interacting limit ( in this case a ` hybridization gap insulator ' @xcite ) . as such , the kondo insulating state
is obtained simply as a particular limit of the underlying theory . _
a knowledge of local single - particle dynamics and their thermal evolution is well known to be sufficient within dmft @xcite to determine transport properties ( see section 2.2 below ) .
our initial focus is thus on the local retarded green functions @xmath66 and likewise @xmath67 for the @xmath19-levels , with corresponding spectra @xmath68 ( and @xmath69 or @xmath3 ) .
some brief comments on the free conduction band are first required ( @xmath53 in equation ( [ 2.1 ] ) , where the @xmath19- and @xmath3-subsystems decouple ) ; specified by the local propagator @xmath70 with corresponding density of states ( dos ) @xmath59 .
this is given by g_0^c ( ) & = & ( ^+-_c ) [ 2.2a ] + & = & [ 2.2b ] with @xmath71 , where for arbitrary complex @xmath72 h(z ) = ^+_- d [ 2.3 ] denotes the hilbert transform with respect to @xmath73 ; such that from equation ( [ 2.2a ] ) , @xmath74 corresponds simply to a rigid shift of @xmath75 by @xmath52 .
equation ( [ 2.2b ] ) defines the feenberg self - energy @xmath76 @xcite as used below , with @xmath77 $ ] alone ( since @xmath78 from equations ( 2.2 ) ) . the free conduction band is thus determined by the non - interacting dos @xmath79 which , modulo the rigid @xmath52-shift , reflects the underlying host bandstructure , @xmath80 . while the formalism below holds for an arbitrary @xmath73 , explicit results will later be given for the hypercubic lattice ( hcl ) , for which within dmft @xcite @xmath81 is an unbounded gaussian ; and the bethe lattice ( bl ) , with compact spectrum @xmath82 for @xmath83 @xcite .
the hcl will in fact be the primary case , because the bloch states characteristic of it ultimately underlie the lattice coherence inherent to low - temperature metallic hf behaviour .
the major simplifying feature of dmft is that the self - energy becomes momentum - independent and hence site - diagonal @xcite ; and since we are interested in the homogeneous paramagnetic phase , the local green functions @xmath84 ( @xmath85 ) are also site - independent .
straightforward application of feenberg renormalized perturbation theory @xcite , then gives the @xmath86 as g^c()&=&[2.4a ] + g^f()&= & [ 2.4b ] + & = & [ 2.4c ] where @xmath87 is the retarded @xmath3-electron self - energy ( @xmath88 such that @xmath89 ) . in equations ( 2.4 )
, @xmath90 is the feenberg self - energy for the fully interacting case , with @xmath91 $ ] the _ same _ functional of @xmath92 as it is of @xmath93 in the @xmath53 limit . in consequence
, @xmath92 is given using equations ( [ 2.4a ] ) , ( 2.2 ) , ( [ 2.3 ] ) as g^c()=h ( ) [ 2.5a ] where @xmath94 ( @xmath95 ) is given by ( ; t)=^+-_c- .
[ 2.5b ] let us first point up the physical interpretation of equations ( 2.4 ) , ( 2.5 ) .
@xmath92 is a _ local _ propagator , and as such familiarly expressed as @xmath96 ; with the @xmath97-resolved conduction electron propagator @xmath98^{-1}$ ] and the usual conduction electron self - energy @xmath99 thus defined .
since @xmath80 , it follows directly that g^c ( ) = ^+_-d _ 0()g^c ( ; ) g^c ( ; ) _ [ 2.6 ] with ( @xmath100 ) @xmath101^{-1}$ ] . but equation ( [ 2.6 ] ) is precisely the form equation ( [ 2.5a ] ) ( with equation ( [ 2.3 ] ) for @xmath102 ) , showing that g^c ( ; ) = [ ( ; t)-]^-1 [ 2.7 ] with @xmath94 related to the conduction electron self - energy by ( ; t)=^+-_c - _ c(;t ) ; [ 2.8 ] and hence ( via equation ( [ 2.5b ] ) ) that @xmath103^{-1}$ ] in terms of the @xmath3-electron self - energy alone ( because the @xmath3-levels alone are correlated ) . for an arbitrary conduction band ( specified by @xmath104 ) equations ( 2.4 ) , ( 2.5 )
are central ; for given the self - energy @xmath87 , and hence @xmath94 from equation ( [ 2.5b ] ) , @xmath92 follows directly from the hilbert transform equation ( [ 2.5a ] ) , and @xmath105 in turn from equation ( [ 2.4c ] ) .
that statement hides however the truly difficult part of the problem : obtaining the self - energy @xmath87 .
this is not merely a calculational issue , e.g. the need to solve the problem iteratively and self - consistently ( any credible approximation to @xmath87 will in general be a functional of self - consistent propagators ) .
it reflects by contrast the longstanding problem of obtaining an approximate @xmath87 that , ideally : ( i ) handles non - perturbatively the full range of interaction strengths , from weak coupling ( itself accessible by perturbation theory or simple variants thereof @xcite ) all the way to the strongly correlated kondo lattice regime that is dominated by spin - fluctuation physics and typified by an exponentially small coherence scale @xmath1 .
( ii ) respects the asymptotic dictates of fermi liquid behaviour on the lowest energy ( @xmath27 ) and/or @xmath4 scales on the order of @xmath106 itself yet can also handle the _ full _
@xmath27 and/or @xmath4 range ; including the non - trivial dynamics that arise on energy scales up to many multiples of @xmath1 yet which remain universal ( and the existence of which we find to dominate transport and optics ) , as well as the non - universal energy scales prescribed by the bare material parameters of the problem .
the success of any theory naturally hinges on the inherent approximation to @xmath87 . in this paper
we employ the local moment approach ( lma ) @xcite , for it is known to satisfy the above desiderata and to our knowledge is currently the only theory that does .
it is based on an underlying two - self - energy description a natural consequence of the mean - field approach from which it starts , and from which the conventional single self - energy @xmath107 follows together with the concept of _ symmetry restoration _ that is central to the lma generally @xcite .
full details of the lma for the pam , including discussion of its physical basis and content , are given in @xcite .
in particular the generic asymmetric pam ( as considered here ) is detailed in @xcite for @xmath18 ; and extension of it to finite-@xmath4 , required to consider transport and optics , follows the approach of @xcite where the particle - hole symmetric pam appropriate to the case of kondo insulators was considered .
for that reason further discussion of the approach is omitted here .
the reader is instead directed to @xcite on the pam , from which appropriate results will be used when required ; and to @xcite for anderson impurity models _ per se _ where details of the lma , including its stengths and limitations in relation to other approaches , are fully discussed . as mentioned above
, a knowledge of single - particle dynamics is sufficient within dmft to determine @xmath108 transport properties @xcite .
this arises because the strict absence of vertex corrections in the skeleton expansion for the current - current correlation function means only the lowest - order conductivity bubble survives @xcite , and a formal result for it is thus readily obtained .
denoting the trace of the conductivity tensor by @xmath109 ( @xmath110 of which , denoted by @xmath111 , provides an approximation to the isotropic conductivity of a 3-dimensional system ) , this may be cast in the form ( ; t)=_0 f(;t ) [ 2.9 ] with @xmath112 merely an overall scale factor ( @xmath113 is the lattice constant and @xmath114 is typically of order @xmath115 ) .
the dimensionless dynamical conductivity @xmath116 naturally depends on the lattice type , and for a bloch decomposable lattice such as the hcl is given ( with @xmath43 ) by @xcite f_hcl(;t)=^_- d_1d^c(;_1 ) d^c(;_1 + ) _ [ 2.10 ] where @xmath117^{-1}$ ] is the fermi function . here
( as in equation ( [ 2.6 ] ) ) , the notation @xmath118 denotes an average with respect to the non - interacting conduction band dos @xmath73 ; and the spectral density @xmath119 with @xmath120^{-1}$ ] from equation ( [ 2.7 ] ) .
physically , @xmath121 ( @xmath122 ) represents the @xmath27-dependent conduction electron scattering rate ( inverse scattering time ) arising from electron interactions , @xmath123 ( @xmath124 from equation ( [ 2.8 ] ) ) .
it is given using equation ( [ 2.5b ] ) by = _
i(;t ) = [ 2.11 ] in terms of the @xmath3-electron self - energy ; a knowledge of which thus determines the scattering rates ( considered explicitly in section 3.1 ) , and in consequence the dynamical conductivity equation ( [ 2.10 ] ) ( noting that @xmath125 ^ 2 + [ \gamma_i(\om;t)]^2)$ ] )
. results for @xmath126 obtained using the lma will be considered in sections 4,5 .
here we simply point out an exact result , not apparently well known , for the weight of the drude peak in the @xmath18 conductivity . at @xmath18 , scattering at the fermi level is absent since the system is a fermi liquid , i.e. @xmath127 and hence @xmath128 .
the leading low - frequency behaviour of @xmath129 is given by ^r_f(;0 ) ~^r_f(0;0 ) - ( -1)[2.12 ] where @xmath130^{-1}$ ] is the usual quasiparticle weight / inverse mass renormalization ; hence ( from equation ( [ 2.5b ] ) ) @xmath131 , where ^*_f = _ f + ^r_f(0;0 ) [ 2.13 ] is the renormalized @xmath3-level energy .
a straightforward evaluation of equation ( [ 2.10 ] ) for @xmath18 and @xmath132 then shows that @xmath133 contains a @xmath134 drude ` peak ' ( as it must , reflecting the total absence of fermi level scattering and a vanishing @xmath18 resistivity ) .
denoted by @xmath135 , it is given explicitly by f_drude(;0 ) = ( ) _ 0(-_c+ ) [ 2.14a ] or equivalently f_drude(;0 ) = ( ) _ 0(-_c+ ) [ 2.14b ] where @xmath136 and _ l = zv^2 . [ 2.15 ] equations ( 2.14 ) are exact , and bear comment .
in the trivial limit @xmath53 where ( equation(2.1 ) ) the @xmath3-levels decouple from the conduction band , the total drude weight is naturally @xmath137 , the free conduction band dos at the fermi level ( recall @xmath138 ) . for any @xmath139 ,
the luttinger integral theorem requires ( n_c+n_f ) = ^-_c+1/_f^*_- _ 0 ( ) d + ( -_f^ * ) [ 2.16 ] ( with @xmath140 merely the unit step function ) .
this again is an exact result , proven in @xcite .
it holds for _ any _ interaction @xmath6 , reflecting the adiabatic continuity to the non - interacting limit that is intrinsic to a fermi liquid ; and shows in general that ( any ) fixed total filling @xmath141 determines @xmath142 entering equations ( 2.14 ) .
of particular interest is of course the strongly correlated hf regime , where @xmath143 . here
@xmath1 in equation ( [ 2.15 ] ) ( @xmath144 with @xmath145 ) is the coherence scale : exponentially small in strong coupling ( because @xmath146 is ) , it is the _ single _ low - energy scale in terms of which all properties of the system exhibit universal scaling ( as shown in @xcite and pursued below ) . in the hf regime , @xmath20 itself is moreover given ( see @xcite ) by n_c = ^-_c_-_0 ( ) d[2.17a ] showing that @xmath147 and @xmath20 are in essence synonymous , @xmath148 being determined by @xmath147 alone .
conjoining this with equation ( [ 2.16 ] ) gives n_f = ^-_c+1/_f^*_-_c _ 0 ( ) d + ( -_f^ * ) [ 2.17b ] so as @xmath143 ( the hf regime ) , @xmath149 is _ also _ determined by @xmath147 alone , and is typically of order unity .
it is this that determines @xmath150 entering equation ( [ 2.14b ] ) for @xmath135 ; showing in turn that the net drude weight is itself @xmath151 , and hence exponentially diminished compared to the free conduction band limit .
we add that kondo insulators , arising generically for @xmath152 as mentioned earlier , are also encompassed by the above . using @xmath153 , the luttinger theorem equation ( [ 2.16 ] )
shows that @xmath152 arises either for @xmath154 ( for an unbounded @xmath104 ) or for @xmath142 outside the band edges of a compact @xmath104 ; such that in either case the drude weight in equations ( 2.14 ) vanishes , symptomatic of the vanishing @xmath18 d.c . conductivity characteristic of the kondo insulating state .
our focus above has naturally been on the canonical case of a bloch decomposable lattice . for a bethe lattice by contrast , @xmath116 is given @xcite by ( _ cf _ equation ( [ 2.10 ] ) ) f_bl(;t)=^_- d_1d^c(_1 ) d^c(_1 + ) [ 2.18 ] where @xmath155 ( @xmath156 ) is the local conduction band spectrum .
in particular the d.c .
conductivity at @xmath18 follows as @xmath157 ^ 2 $ ] ; which , using @xmath158 together with equations ( [ 2.2a ] ) , ( 2.5 ) , ( [ 2.12 ] ) , is given by f_bl(0;0 ) = [ _ 0(-_c+)]^2 . [ 2.19 ] in contrast to equations ( 2.14 ) there is thus no drude @xmath134-peak and the @xmath18 d.c .
resistivity is in general finite , reflecting of course that the underlying one - particle states of the bl are not coherent bloch states .
hence , aside from the case of kondo insulators where the bl ( like the hcl ) does capture the vanishing @xmath18 d.c .
conductivity and indirect - gapped optics characteristic of the insulator @xcite , the ` joint density of states ' type formula equation ( [ 2.19 ] ) should not be taken seriously when considering transport / optics of real materials on sufficiently low @xmath4 and/or @xmath27 scales ( as discussed further in section 4 ) .
we turn now to lma results for single - particle dynamics at finite-@xmath4 . our natural focus will be the strong coupling kondo lattice regime ( where @xmath143 ) , characterized by the low - energy lattice scale @xmath159 .
this scale is of course a complicated function of the bare / material parameters , @xmath160 ( detailed lma results for it are given in @xcite , and nrg results in @xcite ) .
that dependence is however a subsidiary issue in comparison to the fact that , because @xmath1 becomes exponentially small in strong coupling , physical properties exhibit scaling in terms of it ; i.e. depend universally on @xmath161 , independently of the interaction strength .
universality in strong coupling single - particle dynamics at @xmath18 has been considered in @xcite for the generic pam ; the essential findings of which are first reprised for use below .
( i ) both the @xmath19-electron spectrum @xmath155 ( @xmath162 with @xmath43 ) and the @xmath3-electron spectrum @xmath163 ( with @xmath164 introduced in section 2 ) , exhibit universal scaling as a function of @xmath22 in a manner that is _ independent _ of both the interaction strength @xmath6 _ and _ hybridization matrix element @xmath38 .
( ii ) that scaling depends in general only on @xmath147 ( or equivalently the conduction band filling @xmath20 , see equation ( [ 2.17a ] ) ) which embodies the conduction band asymmetry ; and on @xmath165 reflecting the @xmath3-level asymmetry . more specifically ,
( iii ) in the coherent fermi liquid regime arising for @xmath166 , the @xmath3- scaling spectra depend only on @xmath147 and are in fact independent of @xmath45 as well as @xmath6 and @xmath38 . in this low-@xmath24 regime
the scaling spectra amount in essence to the quasiparticle behaviour ( equations ( 3.11 ) of @xcite ) required by the asymptotic dictates of low - energy fermi liquid theory .
( iv ) for @xmath167 by contrast the @xmath3- scaling spectra depend on the @xmath3-level asymmetry @xmath45 ( albeit rather weakly ) , but are now independent of @xmath147 and indeed also of the lattice type ; and the spectrum contains a long , logarithmically slowly decaying spectral tail .
( v ) the latter behaviour , which sets in progressively for @xmath168 , reflects in turn the crossover to incoherent effective single - impurity physics that one expects to arise for sufficiently high @xmath27 ( and/or @xmath4 ) : for @xmath167 the _ scaling form _ of the @xmath3-spectrum is found to be precisely that of an anderson impurity model ( aim ) . with increasing @xmath24 ,
dynamics thus cross over from the low - energy quasiparticle behaviour symptomatic of the lattice coherent fermi liquid state to single - impurity kondo scaling physics at high @xmath24 ( and that this crossover occurs in a single @xmath161 scaling regime is thus incompatible with the occurrence of ` two - scale exhaustion ' @xcite as explained in @xcite ) .
figure [ fig1 ] summarises representative results for @xmath18 scaling dynamics ( irrelevant non - universal energy scales such as @xmath6 , @xmath41 ( @xmath169 ) or @xmath170 are of course projected out in scaling spectra @xcite ) .
the main figure shows @xmath23 scaling spectra for the hcl as functions of the scaled frequency @xmath171 , for @xmath172 with @xmath173 ( dashed , and @xmath174 ) and @xmath175 ( solid , with @xmath176 ) .
the @xmath177 example corresponds to the particle - hole ( p - h ) symmetric kondo insulator , whose spectra are thus gapped at the fermi level @xmath178 ( with @xmath159 here corresponding to the insulating gap scale @xcite ) . for the asymmetric conduction band @xmath179 by contrast ,
the gap ( which is well developed in strong coupling @xcite ) moves above the fermi level ; and a sharp lattice - kondo resonance symptomatic of the hf metal , straddling the fermi level and of width @xmath180 , takes its place in the @xmath3-spectra .
the inset shows the @xmath3-spectra on a much larger @xmath24 scale ; displaying the @xmath31-independence of the slow logarithmic tails @xcite and reflecting the crossover to effective single - impurity behaviour ( which we emphasise arises whether the system is a hf metal or a kondo insulator ) . at finite temperatures , what one expects for the strong coupling scaling spectra
is clear : they should now depend universally on @xmath181 _ and _ @xmath182 .
that this arises correctly within the present lma is shown in figure [ fig2 ] . for a fixed @xmath183 , the @xmath3- and @xmath19-spectra are shown for progressively increasing interaction strengths @xmath184 and @xmath185 with @xmath186 ; for @xmath187 and @xmath172 ( corresponding results for the p - h symmetric limit have been obtained in @xcite ) .
the inset shows the @xmath3-spectra on an absolute scale ( _ vs _ @xmath188 ) , where the exponential reduction of the @xmath1-scale with increasing @xmath6 is clearly seen from the change in the width of the resonance .
the main figures by contrast show the spectra as functions of @xmath24 , from which the @xmath6-independent scaling collapse is evident ; repeating the calculations with different @xmath189 likewise shows the scaling to be independent of @xmath38 .
this behaviour is not of course confined to the chosen @xmath26 , and figure [ fig3 ] shows the resultant lma scaling spectra for a range of @xmath26 ( again for the representative @xmath190 ) .
figures [ fig3 ] and [ fig2 ] show clearly the thermal broadening and ultimate collapse of the @xmath3-resonance with increasing @xmath26 ; which is naturally accompanied by a redistribution of spectral weight leading to infilling of the ( @xmath191 ) spectral gap seen in figure [ fig1 ] for @xmath18 . in fact by @xmath192 this gap is already obliterated , and the lattice kondo resonance also significantly eroded .
this behaviour is typical of the metallic hf state .
by contrast , corresponding results for the p - h symmetric kondo insulator ( @xmath193 ) are shown in figure 4 of @xcite . in that case
the insulating gap at the fermi level fills up with increasing temperature , and the fermi level @xmath194 in particular increases monotonically with increasing temperature ; in contrast to the the asymmetric hf spectra shown above where @xmath195 diminishes with @xmath26 .
two further points regarding figure [ fig3 ] should be noted .
first , the thermal evolution of the @xmath3- and @xmath19-spectra differ somewhat in terms of the persistence of a pseudogap the @xmath3-spectrum shows no sign of the gap by @xmath196 , while a weak pseudogap structure persists in the @xmath19-spectrum up to @xmath197 ; this reflects the rapid spread of spectral weight caused by the meltdown of the sharp resonance in the @xmath3-spectra , of which there is no counterpart in the @xmath19-spectra .
second , the inset to figure [ fig3 ] shows the @xmath3-spectra on an enlarged frequency scale out to @xmath198 , from which it is seen that the high frequency behaviour of the finite-@xmath26 scaling spectra coincide with that for @xmath18 .
this is physically natural , since one expects the dominant influence of temperature to be confined to frequencies @xmath199 .
the corollary of course is that non - universal frequencies are affected only on non - universal , and thus in general physically irrelevant , temperature scales ( as shown in figure 5 of @xcite for the p - h symmetric case ) .
we consider now the scattering rates @xmath200 that underlie the evolution of the conductivity , and are given explicitly in terms of the @xmath3-electron self - energy by equation ( [ 2.11 ] ) . since the system is a fermi liquid with @xmath201 , at @xmath18 there is of course no scattering at the fermi level , @xmath202 .
the low frequency behaviour of the @xmath18 scattering rate can be understood qualitatively by using the low-@xmath27 expansion of @xmath203 ( equation ( [ 2.12 ] ) ) and simply neglecting the imaginary part @xmath204 ; leading to ^-1(;t=0 ) ( - _ f^ * ) [ 3.1 ] with @xmath205 the renormalized level .
restoring the small but strictly non - vanishing @xmath206 naturally implies a narrow resonance centred on @xmath207 instead of a pure @xmath208-function . at finite temperature
, we likewise expect the scattering rate to increase from zero in the neighbourhood of the fermi level , reflecting the finite-@xmath4 contribution to @xmath209 ; and that this will simultaneously lead to further , thermal broadening of the resonance at @xmath150 .
the above picture is corroborated by lma results as shown in figure [ fig4 ] , displaying the @xmath22 dependence of @xmath200 ( in units of @xmath42 ) arising in strong coupling for @xmath187 and @xmath172 , for a range of temperatures @xmath210 between @xmath211 and @xmath212 . in this case
the renormalized level is found to be @xmath213 , precisely where @xmath214 has a narrow resonance . with increasing temperature the resonance
is indeed seen to broaden and decrease in intensity ; and we reiterate that this occurs for temperatures @xmath4 set by the scale @xmath1 the sole low - energy scale characteristic of the problem in strong coupling . excepting the lowest @xmath26 we also note that scattering rates in the vicinity of the fermi level are on the order of @xmath215 of the bandwidth @xmath41 , values some two or so orders of magnitude higher than for conventional clean metals ( and indicative of the higher d.c .
resistivities that are typical of heavy fermion materials @xcite ) .
neither is this behaviour confined to a narrow @xmath26 regime since even for @xmath216 the scattering rates decay very slowly with @xmath26 ; the fermi level scattering rate for example is readily shown to decay as @xmath217 .
the scattering rates are also related to the @xmath3-electron scaling spectra considered above . for the kondo insulating p - h symmetric pam ,
it was shown in @xcite that the dimensionless scattering rate defined as = _
i(;t ) [ 3.2 ] coincides asymptotically with the @xmath3-spectral function , specifically ~_0 d^f ( ) [ 3.3 ] in the regime @xmath218 for any @xmath26 ( the spectral ` tails ' ) , and for all @xmath219 for sufficiently large @xmath216 . equation ( [ 3.3 ] ) is in fact readily shown to be quite general , and not dependent on p - h symmetry .
that it holds for hf metals embodied in the asymmetric pam is illustrated in figure [ fig5 ] , where for @xmath187 and @xmath172 the strong coupling @xmath220 and @xmath221 _ vs _ @xmath24 are compared , for @xmath222 in the left panel and @xmath183 and @xmath223 in the right panel .
the high - frequency behaviour of the scaling spectrum @xmath221 is itself known , being given ( here for @xmath224 explicitly ) by @xcite _ 0 d^f ( ) ( + ) [ 3.4 ] with @xmath113 a pure constant @xmath225 .
these slowly decaying logarithmic tails are evident in figure [ fig5 ] , and as mentioned in section 3 embody the connection to effective incoherent single - impurity physics on high energy scales .
they are independent of the interaction @xmath6 , local hybridization @xmath38 , underlying conduction band asymmetry @xmath31 , and even of the lattice type ; depending , albeit weakly , only on the @xmath3-level asymmetry @xcite .
the above discussion of scattering rates leads naturally to consideration of transport ; beginning with the d.c .
limit where ( section 2.2 ) the static conductivity @xmath226 , with @xmath116 given for the hypercubic lattice by equation ( [ 2.10 ] ) . in the strong coupling regime we expect static transport to exhibit universal scaling in terms of @xmath25 , and our aim here is to understand its thermal evolution across the full @xmath26 range .
transport on non - universal temperature scales @xmath227 ( @xmath228 ) or @xmath229 , will be discussed briefly at the end of the section . for the p - h symmetric kondo insulator ,
lma results for the @xmath26-dependence of the scaling resistivity have been considered in @xcite ( in this case @xmath230 is equivalently the insulating gap scale ` @xmath13 ' ) .
the @xmath18 resistivity is naturally infinite reflecting the gapped ground state , the scaling resistivity @xmath231 has an activated form @xmath232 for @xmath233 ( with @xmath234 a pure constant @xmath235 and hence a ` transport gap ' of @xmath236 ) ; and @xmath5 decreases monotonically with increasing @xmath26 , tending asymptotically to incoherent single - impurity scaling behaviour ( @xcite and figures [ fig7],[fig8 ] below ) . for the general case of heavy fermion metals
the situation is of course quite different , and what one expects in qualitative terms well known @xcite .
the @xmath18 resistivity vanishes , reflecting the absence of fermi level scattering and the underlying coherence generic to any bloch decomposable lattice .
with increasing temperature @xmath5 increases ( initially as @xmath237 for @xmath233 @xcite ) , passes through a maximum at @xmath238 a classic signature of hf compounds @xcite and decreases thereafter in the strong coupling , kondo lattice regime of interest .
figure [ fig6 ] shows lma results for @xmath5 _ vs _
@xmath26 for fixed @xmath239 , and with increasing interaction @xmath240 and @xmath185 for @xmath186 .
the scaling collapse is clearly evident : while the low - energy scale @xmath241 itself diminishes exponentially on increasing @xmath6 , universal scaling of @xmath5 as a function of @xmath25 indeed arises in strong coupling , independent of interaction strength ( and likewise readily shown to be @xmath38-independent on repeating the calculations varying @xmath189 ) .
this leads us first to comment briefly on the issue of ` the coherence scale ' , characterising the crossover from low - temperature lattice coherent behaviour to high - temperature effective single - impurity behaviour .
experimentally , many such identifications of the low - energy scale are commonly employed .
some groups use @xmath238 at which @xmath5 peaks , others identify the scale via the inflection points ( @xmath242 , marked by arrows in figure [ fig6 ] ) , via the leading @xmath237 behaviour of @xmath5 at low-@xmath4 , or via the onset of the ` log - linear ' regime @xcite ( shown in inset ( a ) to figure [ fig6 ] and seen in many experimental systems @xcite ) ; the inverse of the @xmath243 paramagnetic susceptibility , or the width of the lattice kondo resonance , are other possibilities .
this leads to what at first sight might seem a plethora of low - energy scales .
the key point however is that , because physical properties in strong coupling scale universally in terms of _ one _ low - energy scale , all the above definitions of ` the coherence scale ' are fundamentally equivalent : all are proportional to @xmath1 , and hence to each other in figure [ fig6 ] for example , the inflection points in @xmath5 lie at @xmath244 and @xmath245 , and the peak maximum at @xmath246 . as for single - particle dynamics and scattering rates considered in section 3 ,
the @xmath26-dependent scaling resistivity is independent of @xmath6 or @xmath38 ( as above ) but depends in general on @xmath147 ( reflecting the conduction band asymmetry and determining @xmath20 via equation ( [ 2.17a ] ) ) and @xmath45 ( reflecting the @xmath3-level asymmetry ) . to consider this figure [ fig7 ] shows the resultant scaling resistivities @xmath5 _ vs _
@xmath26 for @xmath172 , and a range of different @xmath147 @xmath247 and @xmath248 , corresponding respectively to conduction band fillings @xmath249 and @xmath250 .
the @xmath177 example is the kondo insulator @xcite , with its characteristic diverging @xmath5 as @xmath251 .
the others are all hf metals , and exhibit the same qualitative behaviour for all @xmath147 a positive slope for @xmath252 , going through the maximum and then decreasing monotonically for @xmath253 ; the coherence peak itself increasing monotonically with @xmath147 , albeit slowly such that @xmath254 for the @xmath147-range shown . qualitatively similar behaviour is found on varying the @xmath3-level asymmetry @xmath45 for fixed conduction band asymmetry embodied in @xmath147 , although quantitatively this effect is appreciably less .
the significant @xmath147-dependence of @xmath255 seen in figure [ fig7 ] for @xmath256 is intuitively natural : the strong coupling kondo lattice regime corresponds to @xmath257 , but with variable conduction band filling ( @xmath20 ) controlled by @xmath147 ( equation ( [ 2.17a ] ) ) ; and on decreasing @xmath20 ( increasing @xmath147 ) one expects the static conductivity to diminish and hence an increased @xmath5 , as found . to understand the @xmath147-dependence , and in turn to enable connection to incoherent effective single - impurity behaviour at high-@xmath26
, we first consider an approximate evaluation of @xmath258 ( equation ( [ 2.10 ] ) ) ; in which the energy dependence of the free conduction band dos @xmath259 is neglected , @xmath260 being replaced by its fermi level value . employing this
` flat band ' approximation in equation ( [ 2.10 ] ) ( where it enters via the @xmath261 average ) leads to f_hcl(0;t ) [ _ 0(-_c)]^2 ^+_-d ( ; t ) [ _ 0(-_c)]^2 [ 4.1 ] expressed as a physically intuitive thermal average of the dimensionless scattering time @xmath262 ( equation ( [ 3.2 ] ) ) . for kondo insulators
this approximation is qualitatively inadequate at low-@xmath26 @xcite , but as illustrated in figure [ fig7 ] ( inset ) it is entirely respectable for the hf metals and in particular recovers precisely the high-@xmath26 asymptotics of @xmath5 .
as shown in section 3.1 , the large @xmath24 and/or @xmath26 dependence of the reduced scattering rate @xmath220 coincides with the @xmath3-spectral function @xmath221 ( equation ( [ 3.3 ] ) ) ; and in section 3 ( see also @xcite ) the latter were shown to have common spectral tails , independently of @xmath147 .
this suggests that the primary effect of @xmath147 seen in figure [ fig7 ] for @xmath255 is contained in the @xmath263 ^ 2 $ ] of equation ( [ 4.1 ] ) .
that this is so is seen in figure [ fig8 ] where the results of figure [ fig7 ] are now shown as @xmath264 _ vs _ @xmath26 , where ^(t)=. [ 4.2 ] for @xmath265 or so in practice , @xmath264 is seen in particular to be _ independent _ of the conduction band filling embodied in @xmath147 ; including we note the kondo insulator , whose ` high ' temperature resistivity is thus seen to be that of a regular heavy fermion metal . indeed as readily demonstrated , and evident in part from the above discussion , the behaviour seen in figure [ fig8 ] is barely dependent on the details ( @xmath266-dependence ) of the host bandstructure embodied in @xmath104 .
[ h ] the obvious final question here concerns the high-@xmath26 form of @xmath264 for the pam . to that end
we consider the anderson single - impurity model ( aim ) , with @xmath267 denoting as usual the change of resistivity due to addition of the impurity to the non - interacting host , and @xmath268 .
this is given by @xcite ( _ _ cf__equations ( [ 4.1]),([4.2 ] ) ) = ^+_-d _ imp(;t ) [ 4.3 ] with the impurity scattering rate @xmath269 ; where @xmath270 is the impurity spectral function such that @xmath271 follows from the friedel sum rule @xcite in the singly occupied , strong coupling kondo regime of the aim . the lma scaling resistivity @xmath272 _ vs _ @xmath26 is also shown in figure [ fig8 ] , where @xmath273 and @xmath274 is the aim kondo scale ( with @xmath275 the impurity quasiparticle weight ) . from this
it is seen that the high-@xmath26 scaling behaviour of @xmath264 for the pam is precisely that of the aim ; in particular the leading @xmath216 behaviour of the lma @xmath264 is readily shown analytically to be given by @xmath276 , which is exact in the kondo limit of the impurity model @xcite .
this reflects again the crossover in the strong coupling pam from low - temperature lattice coherent behaviour to incoherent effective single - impurity scaling physics , here in the context of d.c . transport . as for its counterpart in the case of single - particle dynamics @xcite
, we point out ( a ) that since this connection is established from scaling considerations it is entirely independent of how the scales @xmath1 and @xmath277 for the two distinct models ( pam and aim ) depend on the underlying bare / material parameters of the respective problems ; and ( b ) the fact that it arises in the @xmath278 scaling regime precludes a two - scale description of the crossover from lattice - coherent to incoherent effective single - impurity physics .
our focus above has naturally been on the strong coupling , kondo lattice regime .
we now look briefly at d.c .
transport on non - universal scales .
what one expects here is that when the temperature is a not insignificant fraction of the hybridization @xmath170 or bandwidth scale @xmath279 , kondo screening will be washed out , and hence @xmath5 should cross over from the logarithmically decreasing single - impurity form at @xmath280 ( figures [ fig7 ] and [ fig8 ] ) to conventional metallic behaviour @xmath281 at non - universal temperatures ; and thus as such must go through a minimum .
that this indeed happens can be seen in the inset to figure [ fig8 ] where we show @xmath282 _ vs _
@xmath25 for @xmath187 , @xmath283 and @xmath284 ( solid line ) and @xmath285 ( point - dash ) .
for the lower @xmath6 example , a minimum is seen at @xmath286 , which corresponds in ` absolute ' units ( @xmath41 ) to a temperature @xmath287 an appreciable fraction of the hybridization @xmath288 .
the corresponding minimum does of course exist for the higher @xmath6 , but is pushed beyond @xmath289 ( and to concomitantly lower vales of @xmath5 ) ; and @xmath5 in this case lies on the universal scaling curve throughout the @xmath26-range shown in figure [ fig8 ] .
a final point is worth noting here . for @xmath290 , the @xmath5 _ vs _
@xmath25 for the two @xmath6 s shown in figure [ fig8 ] ( inset ) are in essence coincident ; each lies on the universal scaling curve .
what distinguishes different interaction strengths is of course the location of the minimum , occurring as it does on non - universal temperature scales .
no real hf material is however in the universal scaling regime ` for ever ' with increasing @xmath4 the scaling regime will be exited sooner or later .
and the temperature for which the experimental @xmath5 is a minimum ( once phonon contributions have been subtracted out ) can provide valuable information on the interaction strength , as we shall see in action in the following paper .
we have considered almost exclusively the hypercubic lattice , for the obvious reason that its one - particle bloch states ultimately underlie the low - temperature lattice coherence of the interacting problem . for the bethe lattice , the strong coupling scaling resistivity @xmath291 ( with @xmath292 from equation ( [ 2.18 ] ) ) is shown _ vs _ @xmath26 in inset ( b ) to figure [ fig6 ] , for @xmath293 .
in contrast to its counterpart for the hcl shown in the main figure , the @xmath222 resistivity is non - vanishing ( given by equation ( [ 2.19 ] ) ) , reflecting the absence of coherent bloch states for the bl .
further , the _ high_-@xmath26 asymptote of the bl @xmath5 in the scaling regime is likewise non - zero ; being given by the @xmath18 value of the free ( @xmath53 ) conduction band resistivity , namely @xmath294 ^ 2 = \pi^2/[4(1-\epsilon_c^2)]$ ] as marked by a cross in figure [ fig6 ] inset ( and arising for the same physical reasons discussed for kondo insulators in @xcite ) .
the qualitative contrast between @xmath5 for the canonically bloch decomposable hcl , and that for the bl , illustrates why the latter more specifically the associated ` joint density of states ' type formula equation ( [ 2.18 ] ) for @xmath295 that is not uncommonly employed in the literature gives a poor caricature of d.c .
transport for hf metals in which the lattice coherence is of central importance .
we turn now to the optical conductivity @xmath296 ( with @xmath126 given by equation ( [ 2.10 ] ) ) . in the strong coupling kondo lattice regime
@xmath126 is of course independent of @xmath6 and @xmath189 , and a universal function of @xmath22 and @xmath25 for fixed @xmath147 and @xmath45 .
lma results for @xmath126 are shown in figure [ fig9 ] , for @xmath179 and @xmath224 .
the right panel shows the thermal evolution of the optical conductivity ( on a linear @xmath24-scale ) for temperatures @xmath297 and @xmath223 ; while the left panel ( on a log - log scale ) shows the behaviour for a lower range of temperatures up to @xmath298 .
the latter in particular illustrates the thermal evolution of the optical drude peak , which at @xmath299 consists of an @xmath300 @xmath208-function given by equations ( 2.14 ) ( with net weight @xmath180 in strong coupling ) . on increasing @xmath26 from @xmath211
the drude peak naturally broadens , and is well fit by a lorentzian up to its half - width or so , after which it decays more slowly in @xmath24 . at
the lowest @xmath26 shown the drude peak is well separated from the ` optical edge ' in @xmath126 seen at @xmath301 ( although we add that @xmath126 is strictly non - zero for all @xmath24 ) , and with increasing @xmath26 is seen to persist as an essentially separate entity up to @xmath302 or so ; after which it is progressively destroyed as expected , merging into an optical pseudogap in the neighbourhood of @xmath303 , which is reasonably well filled up by @xmath304 and all but gone by @xmath305 ( see figure [ fig9 ] , right panel ) .
similar behaviour is naturally found on varying @xmath147 and/or @xmath45 .
figure [ fig10 ] shows in particular the influence of @xmath147 ( varying conduction band filling ) on the optical pseudogap for a fixed temperature @xmath306 , from which it is seen that the pseudogap becomes shallower with increasing @xmath147 .
the above behaviour should be compared to the p - h symmetric kondo insulator ( ki ) @xmath307 considered in @xcite . in that case
the @xmath18 optical conductivity is characterized by an indirect gap @xmath308 , and there is of course no @xmath18 drude peak .
instead a drude - like peak in the optical conductivity actually builds up on initially increasing @xmath26 from zero ( see figure 15 of @xcite ) , before being thermally broadened and subsumed into the optical pseudogap . for @xmath309 or so
the low - frequency optics of the ki are thus very different from those of the hf metal , as expected .
but for @xmath310 the optical behaviour of the two is qualitatively similar as shown by comparison of figure [ fig9 ] ( right panel ) and its counterpart for the ki , figure 14 of @xcite .
this too is physically natural , since the infilling of the indirect optical gap on temperature scales @xmath192 means that the ki behaves to all intents and purposes as a hf metal ; as seen also in figures [ fig7 ] or [ fig8 ] for the static transport .
a second point should be emphasised here , obvious though it is from the preceding discussion : whether for hf metals or kondo insulators , it is the low - energy scale @xmath159 that sets the intrinsic scale for both the @xmath27-dependence of the low - energy optical conductivity and its thermal evolution . and in strong coupling that scale is wholly distinct from the optical _ direct _ gap , @xmath311 .
the latter arises at its simplest in the commonly employed renormalized band picture ( see e.g. @xcite ) , as the minimum direct gap for which optical transitions are allowed . in this effective single - particle description the imaginary part of the @xmath3-electron self - energy and hence all scattering
is neglected entirely , and the corresponding real part @xmath312 is replaced by its leading low-@xmath27 behaviour equation ( [ 2.12 ] ) ( as also inherent to a slave boson mean - field approximation @xcite ) .
the two branches of the renormalized bandstructure , denoted by @xmath313 with @xmath314 , then follow from the zeros of @xmath315^{-1 } = [ \gamma(\om)-\epsilon_{\mathbf k}]$ ] ( see equation ( [ 2.7 ] ) ) with the approximate @xmath316 @xmath317^{-1}$ ] from equation ( [ 2.5b ] ) ; and the resultant @xmath266-dependent direct gap @xmath318 $ ] is given by @xmath319^{1/2}$ ] with @xmath320 the usual renormalized level .
the minimum direct gap , @xmath311 , occurs for @xmath321 ( @xmath322 in strong coupling ) and is thus _
[ 5.1 ] the corresponding result for the optical conductivity @xmath126 is readily determined from equation ( [ 2.10 ] ) . denoted by @xmath323
it is given for @xmath18 ( and all @xmath324 ) by f_o(;0 ) = [ 5.2 ] with @xmath325 the free conduction band dos and @xmath326 the unit step function ; and is thus non - zero only for frequencies @xmath327 _ above _ the direct gap ( which result is also readily shown to hold for _ all _ temperatures ) .
two points should be noted here .
first that the low - energy scale @xmath159 intrinsic to hfs or kis is qualitatively distinct from the direct gap @xmath311 .
in fact since @xmath328 it follows that in strong coupling where the quasiparticle weight @xmath146 and hence @xmath1 becomes exponentially small , optics on the direct gap scale do not even lie in the @xmath329 scaling regime ; although neither do they occur on truly non - universal scales ( because @xmath330 ) and in that sense belong to the ` low - frequency ' optical spectrum .
second , we emphasise the inherent naivet of interpreting optics in terms of renormalized single - particle interband transitions : it is scattering due to electron interactions that generates _ all _ the optical density below the direct gap scale . failure to include such , as in a renormalized band picture and regardless of how sophisticated the underlying band structure employed in practice inevitably leads to a qualitatively inadequate description of optics ( as illustrated explicitly in figure [ fig11 ] below ) .
neither is this situation ameliorated in materials application by the introduction of _ ad hoc _ @xmath27-dependent broadening factors , for that simply avoids the basic underlying physics .
lma results for optics on all frequency scales are given in figure [ fig11 ] , for @xmath190 ; where @xmath126 is shown _ vs _
@xmath22 on a log scale spanning five orders of magnitude , for the same range of temperatures @xmath25 employed in figure [ fig9 ] ( right panel ) . to encompass all @xmath27 including non - universal energies , the bare parameters @xmath6 and @xmath189 must of course be specified , @xmath331 and @xmath332 here being chosen for illustration ;
although note that the optical conductivity as a function of @xmath22 remains ` universal ' ( independent of @xmath6 or @xmath189 ) up to large but finite values of @xmath24 determined by the particular @xmath6 and @xmath189 chosen , in this example @xmath333 ( a directly analogous situation for the resistivity @xmath5 is shown in the inset to figure [ fig8 ] ) .
the inset to figure [ fig11 ] shows the renormalized bandstructure @xmath334 _ vs _ the free conduction band energy @xmath266 ; determined as above from solution of @xmath335 ( with @xmath336 the full @xmath337 ) .
this enables the notional direct gap to be determined , @xmath338 here well separated from the low - energy coherence scale @xmath1 in strong coupling and indeed seen to occur for @xmath339 .
the essential points from figure [ fig11 ] are clear .
as expected and well known ( see e.g. @xcite ) , significant optical absorption occurs in the vicinity of the direct gap ; strongly broadened to low - energies due to electron interactions as above , and all but ` dead ' on non - universal energy scales ( e.g. the hybridization @xmath340 for the chosen bare parameters ) .
regarding the thermal evolution of the optical conductivity note also that temperatures on the order of a few multiples of the coherence scale @xmath1 which control the thermal evolution of the low - energy optics have essentially no effect on frequencies of the order of the direct gap , reflecting the clean separation between @xmath1 and @xmath311 characteristic of strong coupling . as a corollary
the direct gap should be thermally eroded only for @xmath341 ; as indeed seen in figure [ fig12 ] where ( for the same parameters as figure [ fig11 ] ) the thermal evolution of @xmath126 is shown for temperatures up to @xmath342 .
significant thermal erosion sets in by about @xmath343 or so , and is well developed by the highest temperature shown .
the clear scale separation between @xmath1 and @xmath311 will not however be captured properly if one is restricted to relatively low interactions and high temperatures as e.g. in quantum monte carlo @xcite , or from theories in which the quasiparticle weight @xmath146 is algebraically rather than exponentially small in the interaction strength , such as iterated perturbation theory @xcite .
we have considered here the periodic anderson lattice , the canonical model for understanding heavy fermion metals , kondo insulators , intermediate valence and related materials .
optical conductivities , d.c .
transport and single - particle dynamics of the paramagnetic phase have been investigated , using the local moment approach within a dmft framework . for obvious physical reasons
our main focus has been the strongly correlated kondo lattice regime , where we find the problem to be characterised by a single , exponentially small coherence scale @xmath1 ; in terms of which the frequency and temperature dependence of physical properties scale being universally dependent on @xmath344 and/or @xmath25 regardless of the interaction or hybridization strengths .
all relevant energy / temperature scales are handled by the theory , from the low - energy coherent fermi liquid domain out to large ( and in the strict scaling limit arbitrarily large ) multiples of @xmath1 where incoherent many - body scattering dominates the physics ; followed by the crossover out of the scaling regime to non - universal , high energy / temperature scales dictated by ` bare ' model / material parameters . and
while our emphasis has been on strong correlations we add that all interaction strengths from weak to strong coupling are encompassed by the lma @xcite , such that intermediate valence behaviour in particular can also be addressed .
the first question posed in the introduction nonetheless remains : to what extent does the model , and our theory for it , capture experiment ?
we turn to that in the following paper where direct comparison of theory and experiment is made for three heavy fermion materials and a classic intermediate valence compound .
we are grateful to the epsrc for supporting this research . | the paramagnetic phase of heavy fermion systems is investigated , using a non - perturbative local moment approach to the asymmetric periodic anderson model within the framework of dynamical mean field theory .
the natural focus is on the strong coupling kondo - lattice regime wherein single - particle spectra , scattering rates , d.c .
transport and optics are found to exhibit @xmath0 scaling in terms of a single underlying low - energy coherence scale @xmath1 .
dynamics / transport on all relevant ( @xmath2)-scales are encompassed , from the low - energy behaviour characteristic of the lattice coherent fermi liquid , through incoherent effective single - impurity physics likewise found to arise in the universal scaling regime , to non - universal high - energy scales ; and which description in turn enables viable quantitative comparison to experiment . |
circumstellar disks , formed as a consequence of the star formation process , evolve from an initially gas rich phase to a system composed of small , planetary sized bodies , with little remaining gas .
the physical processes responsible for depleting circumstellar gas from the initial minimum mass solar nebula ( mmsn ) to the low quantities remaining in debris disks , those with significant dust evolution and gas depletion , have been the focus of gas dispersal models @xcite .
models for the evolution of the dust @xcite have been successfully utilized in the interpretation of a variety of infrared ( ir ) spectral shapes @xcite .
these models indicate that while some of the material is locked into planets and still more is accreted onto the star through viscous evolution , it is likely that photoevaporation is the most significant process playing a role in the dispersal of the gas disk .
there are several contributors to photoevaporation , including x - rays or extreme ultraviolet ( euv ) photons from the central star and far ultraviolet ( fuv ) emission produced in an accretion shock at the stellar surface of classical t tauri stars , or ctts @xcite . at
any given time , contributions from several different photoevaporation mechanisms will occur , but the overall timescales when each contribution is greatest may vary .
for example , accretion related radiation fields will decay as the mass accretion rate ( @xmath0 ) decreases .
alternatively , x - ray emission may increase as the accretion related radiation fields drop due to increased magnetic activity in non - accreting t tauri stars ( wtts ) , observed as a possible increase in x - ray luminosity from 1 - 100 myr in the _ chandra _ orion ultradeep project ( coup ) survey of young stars @xcite .
the changing environment may play a role , as fuv and x - rays dominate photoevaporation early at @xmath1 , while euv photons are absorbed in the accretion streams and do not become a significant contributor until @xmath0 drops to @xmath2 @xcite . with the large number of factors to consider in gas disk evolution ,
the intensity of uv and x - ray radiation fields and the timescales relevant to their evolution are essential parameters in the theory of photoevaporation .
not only observations of the radiation fields , but observations of the frequency of circumstellar disks place constraints on disk evolution theory , with only 10% of sources retaining disks at 10 myr @xcite .
full ( or primordial ) disks and debris disks are observed at both 5 and 10 myr @xcite , indicating that disk dispersal mechanisms must occur on a number of timescales in a single star forming region .
transitional and pre - transitional disks , which due to the presence of a gap are interpreted as an intermediary stage between full and debris disks @xcite , are observed to have a low frequency indicating that when disk dissipation begins it occurs quickly , taking as little as 0.5 myr @xcite .
alternatively , disk evolution may proceed along another path in which a similar decline in ir fluxes from all disk radii indicates that no gap or hole is formed @xcite . with
the diversity observed in disk evolution , different dispersal mechanisms may be necessary . in this paper
we investigate the relevant timescales for the evolution of disk dispersing emission from t tauri stars with new x - ray and fuv observations spanning 1 - 10 myr , and include previous observations of a larger sample of observations extending to young stars at 1 gyr .
while euv emission is almost impossible to observe due to interstellar hydrogen absorption , x - ray observations are numerous .
x - ray emission in t tauri stars is significantly stronger , up to 1000 times , than the x - ray emission from main sequence stars , and comes from hot plasma with temperatures up to 60 mk or energies up to 5 kev @xcite .
the current paradigm is that this hot plasma is produced through enhanced magnetic activity , with stronger magnetic reconnection events the likely cause of observed short period flaring activity @xcite .
there is some evidence that strong soft x - ray emission is produced in an accretion shock on the stellar surface @xcite .
currently observations of a soft x - ray component which dominates over the coronal emission are unique to the star tw hya , although other sources have exhibited accretion produced x - ray emission , though at a significantly lower level @xcite .
surveys of young , @xmath310 myr star forming regions have shown no evidence of a decline in the x - ray emission as @xmath0 decreases , indicating that accretion related x - ray emission is not a significant contributor to the total x - ray flux in most circumstances @xcite . while roughly constant between 1 and 10 myr , eventually the x - ray emission does decrease . when this decline begins has not been well characterized so far because the only 10 myr old x - ray sources studied were the eleven sources in the tw hydra association ( * ? ? ?
previously , the cluster nearest in age to twa with well characterized x - ray properties was the 68 myr @xmath4 chamaeleon cluster @xcite , with fourteen sources . in this paper
we analyze x - ray observations from another older cluster , the 25 ori aggregate in the orion ob1a subassociation @xcite . at 710 myr , 25
ori still has 4 - 5 remaining ctts and a large number of wtts , making it the most populous @xmath510 myr old star forming region known .
extinction in this off - cloud area is low , making it an ideal region for studying x - rays from t tauri stars in older , evolved populations .
less is known about fuv emission from t tauri stars than x - ray emission because , until recent instruments like the cosmic origins spectrograph ( cos ) on the hubble space telescope ( hst ) , most fuv detectors lacked the sensitivity to observe dim sources .
large fuv samples have been obtained with the international ultraviolet explorer ( iue ) through which fuv line emission has been well characterized @xcite , but little to no signal was achieved in the continuum at @xmath6 for the wtts .
space telescope imaging spectrograph ( stis ) observations of t tauri stars have also been analyzed ; however , long exposure times were required for all but the brightest sources , limiting the ability of stis to observe a large population @xcite . even though iue and stis observations are limited ,
the t tauri properties observed in these high resolution spectra were used to interpret low resolution spectra obtained in a large sample of ctts , wtts , and debris disks using the advanced camera for surveys solar blind channel ( acs / sbc ) prism on hst . while low in resolution , with its excellent sensitivity acs / sbc obtained some of the first observations of wtts in the fuv with the continuum detected from 1230 1800 @xcite .
the origin of fuv emission in t tauri stars has contributions from several sources . in ingleby
et al . ( 2011 ) , we show that in addition to an accretion shock component @xcite , 2 in the inner region of ctts disks contributes to the fuv line and continuum emission @xcite .
the evolution of fuv emission has not been well studied , as fuv spectra of t tauri stars greater than 1 - 3 myr are rare , and previously included only four t tauri stars in the twa region and older debris disks @xcite . in this paper
, we discuss the evolution of high energy radiation fields by analyzing the x - ray and fuv luminosities of a large sample of sources ( the largest fuv sample to date ) with ages between 1 myr and 10 myr , including a small sample of post t tauri objects with ages up to 1 gyr . in @xmath7[obs ] we discuss the new x - ray , fuv , and optical observations used in our analysis . in @xmath7[mage ] we present new x - ray selected t tauri stars in 25 ori and chamaeleon i. in @xmath7[xspec_text ] we present results of spectral analysis performed on high count rate x - ray sources and we discuss the calculation of x - ray and fuv luminosities in @xmath7[xraylum ] and @xmath7[fuvlum ] .
finally , in @xmath7[implications ] we analyze the high energy radiation fields and discuss whether the observed evolution is consistent with current disk evolution theories .
_ chandra _ observations of four fields in the 710 myr 25 ori star forming association were obtained in 2008 january with acis - i and s chips .
observations were done in faint mode with 10 ks exposures in go program 09200907 ( pi : hartmann ) . additionally , observations of two fields in the 2.5 myr chamaeleon i star forming region were obtained in 2009 february with acis - i and s chips .
again , the observations were taken in faint mode in go program 09200909 ( pi : calvet ) with 25 ks exposure times .
we analyzed the evt2 files provided from the _ chandra _ version 7.6.11.4 processing pipeline .
x - ray detections were identified using the _ chandra _ interactive analysis of observations software ( * ? ? ?
* ciao ) script wavdetect with the default significance threshold set to @xmath8 .
we cross - correlated the x - ray detections with the @xcite , @xcite , and the cida variability survey of orion ( * ? ? ? * cvso ) optical catalogs and identified previously studied t tauri sources . in chamaeleon
i , the selected fields focused on two ctts with disks showing evidence for inner clearing , sz cha ( t6 ) and t35 @xcite .
we detected 15 known t tauri stars in the two fields ( out of @xmath560 x - ray sources identified by wavdetect ) , five of which are accretors @xcite .
the sources are listed in table [ resultstable ] and their positions in the h - r diagram are shown in figure [ hrcham ] based on spectral types and luminosities from the literature ( see references in table [ resultstable ] ) . using the evolutionary tracks of @xcite , we determine that the sample covers a large range of t tauri masses from a 11.5 @xmath9 upper limit down to almost substellar masses , with a lower limit of 0.10.2 @xmath9 .
this mass range is similar to that used in @xcite , in which the evolution of x - ray emission in the low mass population of the coup sources was analyzed .
each of the four fields in the 25 ori aggregate was centered on a known ctts , specifically cvso 35 , cvso 206 , ob1a 1630 @xcite , and the transitional disk cvso 224 @xcite .
of @xmath5150 x - ray sources located by wavdetect , we identified 36 previously known t tauri stars , in addition to the 4 targeted sources , from the catalogs of @xcite , briceo et al .
( 2010 ) , and @xcite for a total of 40 710 myr classical and weak t tauri stars . of the remaining unidentified x - ray detections , six are possible newly detected t tauri stars ( see [ opt ] for further discussion ) .
the locations of the 25 ori sources in the h - r diagram are shown in figure [ hrdiagram ] .
sources published in @xcite had known spectral types and luminosities but sources published in @xcite , @xcite , and briceo et al .
( 2010 ) had only spectral types and optical colors . to get luminosities for those sources , we first used @xmath10 colors to get the reddening ( @xmath11 ) assuming photospheric colors from @xcite . with @xmath11 , we corrected two micron all sky survey @xcite @xmath12 magnitudes for reddening using the mathis extinction law with @xmath13 @xcite and calculated the bolometric fluxes @xcite .
our sample of 25 ori sources is similar to the ctts masses of taurus , from @xmath50.21.0 @xmath9 .
table [ resultstable ] lists the complete set of sources detected in the chamaeleon i and 25 ori fields which are discussed in this paper , along with published spectral types and disk classifications . after matching x - ray detections to previously identified t tauri stars
, we used the ciao script dmextract to obtain the counts within the extraction regions determined by wavdetect for each known t tauri source .
we included a background region defined as a 5 annulus around each extraction region , allowing us to determine the net counts and calculate the poisson errors which we used to estimate uncertainties in the x - ray luminosities ( @xmath7[xraylum ] ) .
of the 15 chamaeleon sources , 5 had greater than 250 counts and the spectra of these sources were extracted using the ciao script psextract , using the 3@xmath14 detection regions determined in wavdetect , and the 5 background annuli .
each extracted spectrum was re - binned to 20 counts per bin using the grppha task in ftools .
the re - binned spectra were then analyzed in the x - ray spectral fitting package , xspec @xcite .
only three of the 25 ori sources had counts in excess of 250 , likely due to the shorter exposure times and greater distance to the sources , and were analyzed with psextract and xspec .
we present new fuv observations of ten sources in the 2.5 myr chamaeleon i and two in the 4 myr chamaeleon ii star forming regions obtained with the pr130l grating on the advanced camera for surveys ( acs ) solar blind channel ( sbc ) .
these observations were completed between 2008 december and 2009 february in go program 11145 ( pi : calvet ) .
the fuv coverage extends from 12301800 with r@xmath5300 at 1230 and @xmath550 at 1700 .
chandra _ and acs / sbc chamaeleon i samples have some overlap but there are sources unique to each sample .
we show all the chamaeleon i and ii sources in figure [ hrcham ] , identifying the x - ray and fuv sources and those included in both samples . with acs / sbc we observed sources from 0.32.0 @xmath9 in chamaeleon i and approximately solar mass sources in chamaeleon ii .
we combine these new observations with previously observed acs / sbc and stis samples of t tauri stars analyzed in @xcite , @xcite , and @xcite .
the total sample of sources with fuv spectra consists of 56 sources , eleven of which are wtts or debris disks , and these are listed in table [ resultstable ] , with sources observed by stis distinguished in bold font .
a large number of x - ray detections corresponded to unidentified counterparts in ir and optical catalogs . using 2mass and cvso photometry @xcite , we produced color - magnitude diagrams in @xmath15 and @xmath16 to select sources which fell above the main sequence . in chamaeleon
i , one such object met the color requirements ( k. luhman , personal communication ) and in 25 ori there were six possible new t tauri stars detected . for three of the six t tauri candidates in 25 ori , we followed up the x - ray observations by obtaining optical spectra with the magellan echellete ( mage ) on the magellan- clay telescope at las campanas observatory @xcite .
mage provides r@xmath54100 spectra from 32009800 , including the li 6707 line , for confirmation that the sources are young , and h@xmath17 necessary to classify them as t tauri stars .
in addition to the mage spectra of new t tauri stars , we obtained r@xmath5 35,000 mike ( the magellan inamori kyocera echelle ) observations of 16 sources in our sample , primarily of accretors in the 25 ori and chamaeleon i and ii regions .
the optical data were reduced using the image reduction and analysis facility ( * ? ? ?
* iraf ) tasks ccdproc " , apflatten " and doecslit " .
bias corrections were completed with ccdproc " .
flats were created using apflatten " , in which the orders were traced using observations of a quartz lamp for the mike flat frames and the 10 reddest orders of mage .
for the five bluest orders of mage , we observed a xe lamp with a wide slit ( 5 as opposed to the science slit at 1 ) and out of focus for the trace . the task doecslit " was used to extract the science spectrum , extinction correct for airmass , subtract the background , and dispersion correct the spectra using a thar comparison lamp .
the blaze function was fit using a bright o or b standard star and used to correct for the blaze in the science exposures .
mike optical spectra were used to confirm status of wtts or ctts and obtain accretion rates for the ctts using the relation between @xmath0 and h@xmath17 10% width of @xcite .
we show h@xmath17 line profiles in figure [ halphs ] for the accreting sample of mike sources .
the relation between h@xmath17 width and @xmath0 is more appropriate for low values of the mass accretion rate ; more precise values of the accretion rate will be obtained in a forthcoming analysis of the optical veiling of these sources ( ingleby et al .
2011 ) .
the three candidate members of 25 ori with mage optical spectra , discussed in @xmath7[opt ] , showed li absorption and but only two ( 2mass j05243490 + 0154207 and 2mass j05270173 + 0139157 ) had weak h@xmath17 emission , indicating that they are newly identified wtts . the third source ( 2mass j05244265 + 0154116 )
has h@xmath17 in absorption but may be a young member of 25 ori still , due to its early spectral type ; additional analysis is needed for confirmation . this final source is not included in any further analysis of the 25 ori x - ray observations .
the li and h@xmath17 equivalent widths , along with the derived spectral types are listed in table [ tabxraynew ] for those sources which have optical spectra .
all of the candidate x - ray selected t tauri stars , in both chamaeleon i and 25 ori , are listed in table [ tabxraynew ] .
mage spectra will be obtained for the remaining sources in need of optical spectra .
eight sources in our chamaeleon i and 25 ori samples had strong x - ray fluxes allowing for spectral analysis to determine temperatures for the emitting plasmas .
three of these sources are accreting ( cr cha , sz cha , and tw cha ) and the remaining five sources are not @xcite . with this sample of high count sources
, we looked for any observable distinction between the ctts and wtts . in xspec , each source spectrum was fit using either one or two apec ( astrophysical plasma emission code ) models characterized by different temperatures ; the choice of models was motivated by the two - temperature plasma observed in x - ray emission from pleiades sources @xcite . assuming a model which includes multiple temperatures also allows for the distinction of a soft accretion related emission component , if present , similar to that observed in tw hya @xcite .
we included an absorption component using the xspec model , phabs , with the equivalent neutral hydrogen column density ) is the column density of hydrogen required to give the same opacity as that resulting from the true composition of the absorbing gas , which has contributions from he , c , n , o , and ne in addition to less abundant gasses .
] obtained from the heasarc ( high energy astrophysics science archive research center ) @xmath18 calculator @xcite .
the models with the best statistical fit to the observations ( or lowest @xmath19 ) are given in table [ xspec ] .
the best fit models for cr cha , sy cha , sz cha , and tw cha produced good @xmath19 s , but the remaining sources had either poor fits or model properties discrepant from the larger sample .
for chxr 30a , when the absorption model was forced to have a column density of @xmath20 @xcite , a 64 kev plasma was necessary to fit the spectrum , greater than ten times higher than the highest expected plasma temperatures @xcite . allowing the column density to vary
, we found a best fit model with @xmath21 , fifteen times higher than @xmath18 in the four chamaeleon sources with good fits . while high , a greater value for @xmath18 is consistent with chxr 30a being extincted by @xmath22 magnitudes while the other sources have @xmath23 @xcite .
the apec model fits to cvso 38 resulted in a large temperature for the hot component , a temperature typical of flares in t tauri stars @xcite , indicating that the observations may have caught it while flaring .
the apec components and best model fit to the highest count rate x - ray source ( cr cha ) are shown in figure [ crcha ] . for cr cha
we find that a two - temperature model improves the @xmath19 by a factor of 3 compared to a single temperature model . like cr cha ,
the models of the four sources with the best @xmath19 fits are each comprised of a cool , 0.3 - 0.9 kev , component and a hotter 2 - 5 kev component .
we note that the hottest emission ( excluding the possible flare component from cvso 38 ) comes from sz cha which , as a g2 , is the source with the earliest spectral type .
previous analyses of x - ray spectra from ctts have shown that the presence of accretion alters the x - ray spectrum by producing emission lines formed in the high densities of the accretion shock @xcite . also , cool
x - ray components with @xmath240.20.3 have been attributed to accretion @xcite .
we see no clear evidence of accretion related soft x - ray emission in cr cha , sz cha , and tw cha , the known accretors in our sample .
the cool apec components are @xmath5210 times hotter than the temperatures expected of plasmas heated in an accretion shock , even when considering material falling faster than the free fall velocity which is necessary to produce the soft x - ray shock emission in tw hya @xcite .
@xcite obtained high resolution chandra spectra of tw hya and identified line emission which is attributed to accretion .
additional observations obtained with xmm - newton of a sample of four ctts , including both tw hya and cr cha ( which we analyze in figure [ crcha ] ) , showed that shock formed line emission may be needed to confirm accretion for sources unlike tw hya , where the soft accretion component does not strongly exceed the coronal emission .
the resolution of our acis spectra prohibited accretion line analysis , which would be informative , particularly for tw cha , the accretor with a cool component most similar to that in tw hya .
interestingly , the sources with the coolest components ( sy cha , 2mass j05244498 + 0159465 and rx j0526 + 0143 ) are wtts and have been identified as such by observations of h@xmath17 , indicating that a detection of a cool component alone can not confirm accretion and likely line detections are needed as well @xcite .
instead , our results are consistent with the two - temperature coronal emission observed in pleiades sources @xcite ; however , for chamaeleon , the temperatures of the hot component are approximately 2 times higher than the hot components in the pleiades sources
. this may be indicative of theorized cooling coronal gas , producing a softer spectrum as x - ray sources evolve @xcite .
the results of our spectral analysis show that we can use the same treatment to find x - ray luminosities for both the wtts and ctts since we see no evidence for an accretion related component .
the majority of the x - ray sources had @xmath3250 counts , so instead of attempting detailed spectral analysis on dim sources we assume typical x - ray temperatures as determined for the sources discussed in @xmath7[xspec_text ] .
we assume there is no discernible difference between the x - ray properties of the ctts and wtts in the spectral region observed by acis , as indicated by the results shown in table [ xspec ] .
using webpimms on the heasarc website we calculated unabsorbed fluxes between 0.2 and 10 kev for the entire sample , correcting for @xmath25 and @xmath26 for chamaeleon i and 25 ori , respectively @xcite .
we assumed a single temperature apec model with kt=2.2 kev and solar abundance . from the unabsorbed fluxes ,
x - ray luminosities were calculated assuming distances of 160 and 330 pc for chamaeleon i and 25 ori , respectively ; the results are listed in table [ resultstable ] . for the sources in table [ xspec ] ,
which were fit with xspec , we find that the when using webpimms to find fluxes , the discrepancy is a factor of 23 . the distribution of unabsorbed x - ray luminosities is shown in figure [ lognlogs ] , which indicates that our chamaeleon i sample is not complete . to increase the sample ,
we include x - ray observations of this region from @xcite in our analysis .
the rosat count rates from @xcite were converted to acis - s count rates using webpimms before luminosities were calculated .
in addition to the x - ray luminosities calculated here , we include x - ray luminosities from the literature for sources with fuv spectra which were not observed with _
chandra _ , primarily in taurus , twa , and nearby debris disks ( shown in italics in table [ resultstable ] ) .
x - ray fluxes of taurus sources were calculated as part of the xmm newton extended survey of taurus ( xest ) project @xcite .
@xcite gives count rates for the pn camera with the medium filter and these were converted into _ chandra _ acis - s count rates using webpimms , assuming an input xmm - newton energy range of 0.4 - 10 kev and output _
chandra _ energy range of 0.2 - 10 kev , and using the same model parameters as were used to calculate the fluxes from our _ chandra _ observations .
x - ray fluxes for sources in taurus not provided by @xcite , along with sources in chamaeleon i and ii , twa , and field debris disks were converted from rosat pspc count rates @xcite into _ chandra _ acis - s count rates , again using webpimms but using the softer input energy of rosat ( 0.1 - 2 kev ) . once converted to acis - s count rates
, the unabsorbed x - ray fluxes were calculated as discussed for our new _
observations and the luminosities are given in table [ resultstable ] .
we also include x - ray luminosities from other young clusters in our analysis of x - ray evolution ; those from the @xmath5 4 myr cluster , trumpler 37 @xcite , and the 68 myr @xmath4 cha region @xcite .
our fuv sample contains sources in taurus , the chamaeleon i and ii clouds , 25 ori , twa , objects in the outskirts of the orion nebular cluster ( onc ) , and nearby wtts or debris disks .
we calculated fuv luminosities for these sources between 1230 and 1800 after degrading the resolution of the stis spectra to that of acs / sbc for comparison .
spectra were corrected for reddening using @xmath11 s provided in @xcite and @xcite for taurus , onc , 25 ori , and the nearby wtts and debris disks . for the chamaeleon
i and ii sources , @xmath11 s were calculated from published photometry @xcite .
for sources in taurus , the onc , and chamaeleon i and ii we corrected for reddening using the extinction curve observed along the line of sight towards hd 29647 with @xmath27 , appropriate for regions with associated molecular clouds @xcite . for sources in the dispersed 25 ori association , we used the mathis extinction law with @xmath13 @xcite .
we note that uv extinction laws , especially in the fuv , are very uncertain and introduce the largest error in our fuv luminosity calculations .
we assume a @xmath280.2 magnitude error in @xmath11 , which dominates other sources of error . for four chamaeleon sources , we failed to detect the fuv emission and upper limits are provided for these sources .
we calculated the flux in the detector noise and de - reddened the noise by the calculated @xmath11 s . in this manner
, we are accounting for any fuv emission washed out by extinction .
calculated fuv luminosities are given in table [ resultstable ] , with upper limits on the luminosities provided when appropriate .
the intensity of high energy emission from t tauri stars has important implications for the dispersion of circumstellar disks and the planets which may be forming within them . over a time period of 1 - 10 myr , during which these processes are expected to occur ,
we investigate the evolution of the x - ray and fuv radiation fields .
figure [ xrayvsage ] shows the median and full range of x - ray luminosities for the sources in each star forming region with spectral type k5 or later
. the sample of tr37 may include sources with earlier spectral types ( early k ) ; however , spectral types are not well established for this sample . in addition , it was confirmed that several of the tr37 sources were flaring during the observations @xcite , contributing to the high x - ray luminosities .
figure [ xrayvsage ] reveals that there is no decrease with age in either total x - ray luminosity or the x - ray luminosity normalized by the stellar luminosity , up to 10 myr .
most importantly , our addition of the 710 myr 25 ori x - ray observations supports previous claims that x - ray radiation from the central star remains strong throughout the disk dispersal phase @xcite .
while there does appear to be a slight increase in the normalized x - ray luminosity with age , the trend is statistically insignificant , with a pearson correlation coefficient of only 0.2 .
figure [ fuvvslaccage ] shows the evolution of fuv luminosities from 1 myr t tauri stars to 1 gyr young stars and debris disks . while there is an observed decrease in fuv luminosity out to 1 gyr , a decrease with age
can also be seen among the 1 - 10 myr sources , with the wtts characterized by the lowest fuv luminosities . by the wtts phase
, fuv luminosities are expected to be entirely chromospheric . with no additional accretion ,
fuv emission is no longer produced in an accretion shock and because the end of accretion indicates the depletion of inner disk gas , fuv emission from hot molecular gas no longer contributes to the fuv flux @xcite .
in addition to the decline with age , a correlation between fuv luminosity and @xmath29 is observed , with a pearson correlation coefficient of 0.78 ( calculated with the exclusion of sources with only upper limits ) , indicating that fuv emission is strongly correlated with @xmath29 .
values of @xmath29 were obtained from the literature ( see references in table [ resultstable ] ) or calculated from h@xmath17 10% line widths ( * ? ? ?
* @xmath7[mage ] ) .
this correlation between fuv luminosity and @xmath29 is not surprising , given results that fuv emission is produced in the hot gas of the accretion shock @xcite .
therefore , the decrease in fuv luminosity between 1 and 10 myr can be understood in terms of the decrease of @xmath0 in this time period @xcite .
we use the model of @xcite to predict the amount of fuv emission produced in the accretion column .
this model assumes that the circumstellar disk is truncated at a few stellar radii by the stellar magnetic field . at the truncation radius
, gas is channeled along the magnetic fields lines from the disk to the star , impacting the stellar surface at approximately the free fall velocity .
such fast moving material produces an accretion shock on the surface of the star which heats the gas to @xmath51 mk .
emission from this hot gas heats both the photosphere below the shock and the pre - shock material , producing an accretion column spectrum with a significant excess over the photospheric emission in the uv @xcite . as inputs to the accretion column model ,
we assume a 0.8@xmath9 source evolving along the @xcite evolutionary tracks , from which we obtain the radius , effective temperature , and luminosity of the source from 110 myr .
we assume the percentage of the stellar surface covered by accretion columns , the filling factor , remains constant at 1% and vary the energy flux in the accretion column for a given @xmath0 @xcite . using these quantities
we calculate the column emission and integrate the flux from 12301800 for the highest and lowest @xmath0 s observed at each age @xcite .
we also calculated the fuv accretion emission for the @xmath0 s predicted from a viscously evolving disk with an initial mass of 0.1@xmath9 @xcite . using this technique , we find we can approximate the range of observed fuv luminosities well ( left panel of figure [ fuvvslaccage ] ) , but for typical ctts masses , the predicted fuv fluxes from the viscously evolving disk are too low ( right panel of figure [ fuvvslaccage ] ) .
the missing fuv emission is probably due to several sources which are not included in our accretion column models .
first , emission from 2 gas was observed in the fuv spectra of ctts and was necessary to explain the fuv continuum between 1500 and 1700 . the contribution of gas emission can be a significant portion ( @xmath5 30% ) of the total fuv emission , although it does not contribute to the disk heating , since it likely originates in the upper layers of the disk @xcite .
in addition , we measure the flux in the strong fuv atomic lines of existing spectra , which our accretion column models do not attempt to reproduce and find that @xmath5 35% of the total fuv flux comes from hot lines .
another contribution necessary for re - producing the short wavelength fuv continuum ( @xmath30 ) could originate in post - shock emission escaping absorption by the pre - shock material .
the post shock spectrum peaks in the soft x - rays but extends into the fuv around 1300 @xcite .
this emission would be important for disk photoevaporation models as soft x - rays can efficiently heat disk gas @xcite , but may only play a role when the @xmath0 drops and the accretion columns become less dense @xcite .
additional observations are required to determine if this is a viable contribution to the fuv spectrum .
after accretion has ended and the gas has been dissipated at @xmath3110 myr , fuv luminosities continue to decline .
this is also observed in magnetically active young stars ; all high energy radiation fields decrease as the sources evolve towards and even along the main sequence @xcite .
given this fact , we looked for any relation between the fuv and x - ray emission in the ctts and non - accretors . in figure [ fuvvsxray ]
we distinguish non - accretors ( diamonds ) from ctts ( squares ) and indicate hotter stars by larger symbols .
we see no correlation between fuv and x - ray emission in the ctts ( with pearson correlation coefficients of 0.3 and -0.2 for total and normalized luminosities , respectively ) but we note that the sources with earlier spectral types tend to have the strongest total fuv and x - ray emission , as they are observed to occupy the upper right hand corner in the left panel of figure [ fuvvsxray ] .
fuv and x - ray luminosities normalized by the stellar luminosity are indistinguishable by spectral type ( right panel of figure [ fuvvsxray ] ) .
we do observe a weak correlation between both total and normalized x - ray and fuv luminosities when considering only the non - accreting sample , with pearson correlation coefficients of 0.74 and 0.80 , respectively .
we compare results from our sample of non - accreting young stars ( excluding the low mass twa sources ) to an analysis of solar analogs between 0.1 and 10 gyr @xcite , which showed that high energy fields between 1 and 1200 decreased with age , with emission from hotter plasmas diminishing more quickly . for our sample of non - accreting solar type sources between 15 myr and 1 gyr , classified as young suns in the formation and evolution of planetary systems survey ( feps ; meyer et al .
2006 ) , we compare the observed luminosities to those predicted assuming the power law relations between flux in a given energy range and age from @xcite . the predicted decay of x - ray luminosities is similar to that which we observe for the youngest of the solar type non - accretors in our sample but fail to re - produce the 1 gyr source , hd 53143 ( see figure [ ribas ] ) .
we also compare the slope of the predicted fuv fluxes between 9201180 to the slope of our observations which cover 12301800 .
the total predicted luminosities were scaled , due to the use of different wavelength bins , but the slope was held constant .
again , the younger sources follow the predicted slope but the 1 gyr source has an unexpectedly high fuv luminosity .
it is unclear why hd 53143 is discrepant in both x - rays and fuv fluxes for its age , with several different age indicators providing consistent values @xcite .
our young solar analogs extend the radiation evolution observed in @xcite to younger ages , and agree with their result that the higher energy radiation fields decline more quickly than the cooler fields . combining our results with those of @xcite , we see evidence that even past the 10 myr time period for disk dispersal and planet formation , high energy fields strongly in excess of the solar values persist and may have an effect on evolving disks and planets . using observations for a limited number of t tauri stars , @xcite have investigated the origin of the he ii @xmath321640 and c iv @xmath321549 lines and their relationship to the continuum flux , responsible for photoevaporation of the gas disk . in particular , @xcite propose that the ratio of the he ii to c iv is a probe of the euv radiation field .
they base their conclusion on the fact that single temperature and density models of collisionally ionized plasmas can not reproduce the observed strengths of the he ii line , from which they infer that the line must be produced by radiative recombination .
since the ionization threshold for he i is 228 , they conclude that the line is a probe of the strength of the radiation field around this wavelength , namely , the euv . here ,
we revisit these lines using our larger sample of fuv spectra of t tauri stars .
we measure the he ii and c iv fluxes in our acs / sbc and stis samples of sources with the stis observations degraded to the low resolution of acs / sbc . as a result of the resolution
, we are measuring a blend of emission lines in each line feature , including both components of the c iv doublet and numerous 2 lines present in the vicinity of c iv and he ii . while adding to the scatter in our measurements , the dominant line in each region is the he ii or the c iv line , with 2 line contributions @xmath33 . as discussed in @xmath7[fuvlum ] , the largest source of uncertainty comes from the reddening correction of the fuv spectra , so to estimate errors on the he ii and c iv line luminosities we assume a range of @xmath280.2 magnitudes in @xmath11 .
figure [ linevsage ] shows he ii and c iv line luminosities and their dependence on age and @xmath29 .
both he ii and c iv line luminosities are observed to decrease with age with the lowest values obtained for non - accretors .
we find strong correlations between line luminosities and @xmath29 , with pearson coefficients @xmath5 0.84 and 0.88 for he ii and c iv , respectively .
the strength of both lines is clearly related to accretion , as was previously found by @xcite and @xcite . in the left panel of figure [ ratiovsagelacc ]
we show the he ii to c iv ratio for the ctts and non - accretors of our sample as a function of age . the value of the ratio is higher in non - accretors than in the ctts , confirming the trend found by @xcite .
these authors suggest that this trend indicates that the ionizing flux is not powered by accretion , since the ratio would be expected to decrease if that were the case .
we suggest , instead , that the difference between the he ii to c iv ratio between the ctts and the non - accretors arises from the different formation mechanisms of the lines in each type of star .
the he ii @xmath321640 and c iv @xmath321549 lines in non - accretors must form in the transition region , as they do in the sun and in magnetically active stars , indicated by the similarity of the ratio in the two classes of stars @xcite .
in contrast , the correlations between line luminosity and @xmath29 found in ctts indicate that these lines must form in the accretion flows or at least in regions powered by accretion energy .
however , despite the line luminosity dependence on @xmath29 , we find no correlation between the he ii to c iv ratio and @xmath29 , as shown in the right panel of figure [ ratiovsagelacc ] .
studies of the optical and infrared lines of he i and he ii in ctts indicate that the lines have multiple components arising in the accretion shocks , the magnetospheric infall regions , and the winds @xcite .
no comparable analysis of the c iv line has been done , but it is also likely that this line has contributions from the accretion shock and from the infall region , given its large velocity width @xcite .
this diversity may be responsible for the large dispersion of the he ii to c iv ratios in ctts .
we therefore caution against using the he ii to c iv ratio as an indicator of euv emission without a careful treatment of accretion related emission in these lines .
regardless of the effectiveness of the he ii to c iv line ratio in tracing the euv emission , strong euv emission is likely present in the high energy spectrum of t tauri stars .
@xcite showed that euv emission decreases with age , or increases with the spin of the source .
they also showed that younger sources have enhanced euv line emission , in particular strong he ii and fe lines around 300 - 350 .
whether the euv emission continues to increase at @xmath310 myr like the fuv emission or flattens and remains constant like the x - ray emission is not clear , but in either case the euv emission is expected to be strong . in our model of the accretion column which contributes emission to the fuv spectrum ( [ xray_uv ] ) , we find that the euv luminosity produced in the shock is @xmath34 . in the current model ,
only @xmath52% of this emission escapes the high density material which is still accreting ; however , recent work has shown that varying accretion properties , like the accretion rate or infall velocity , affects the amount of shock emission which can be observed @xcite .
future models of the emission from the accretion column which take this into account will provide better estimates of the euv emission which is available for photoevaporation .
in contrast to early photoevaporation models in which the primary mechanism for dispersing the disk utilized euv radiation from the star producing winds with mass loss rates of @xmath35 @xcite , recent models stress that either fuv or x - ray radiation fields are significantly more important in dispersing the disk with mass loss rates 100 times higher than in the pure euv case . while in agreement on the values of the mass loss rate , @xcite achieve these high rates with x - ray radiation while @xcite find the fuv radiation drives the photoevaporation . with the large range of ages in our x - ray and fuv samples , we explore the photoevaporation theories in the context of our observations .
our results agree with previous findings by @xcite and @xcite showing that the x - ray luminosity remains roughly constant during the first 10 myr .
regarding models of x - ray photoevaporation , @xcite use @xmath36 for @xmath37 kev as their x - ray luminosity .
we find similar , though consistently lower , x - ray luminosities for the typical ctts mass sources in our sample .
@xcite note that their input x - ray luminosity is large and that further analyses will include lower x - ray luminosities .
an important new result in this work is the evolution of the fuv radiation , where we see that the luminosity between 1230 and 1800 clearly drops between 1 and 10 myr .
@xcite incorporate the evolution of fuv luminosity in their models for disk dispersal , assuming that @xmath0 evolves viscously and the fuv luminosity is 0.4 @xmath29 . in figure [ fuvvslaccage ]
we showed the evolution of the fuv luminosity along with the predicted fuv emission produced in an accretion column from a viscously evolving disk onto a 0.8@xmath9 star .
we found that the predicted fuv accretion emission is lower than the observed fuv fluxes , although , within an order of magnitude ( assuming typical mass ctts have @xmath38 ) and we discussed origins for the remaining fuv emission in @xmath7[xray_uv ] .
the fuv fluxes used by @xcite are similar to those we observe , but we found that @xmath51/3 of this emission may be produced by 2 in the circumstellar disk ( ingleby et al .
2011 ) and therefore not directly irradiating the gas in the disk .
our observations support the requirements of both the x - ray and the fuv dominated photoevaporation models of @xcite and @xcite ; however , several issues remain unresolved .
for example , circumstellar disks of gas and dust have been observed around much older sources , including the 10 myr ctts tw hya which maintains a 0.06 @xmath9 disk , albeit a transitional disk @xcite .
full disks of gas and dust are also observed around the 710 myr ctts in 25 ori @xcite . according to @xcite , the continued existence of full disks at late stages
could be explained by the range of x - ray luminosities observed , with lower x - ray luminosities increasing the disk lifetime and , alternatively , the presence of young wtts may be due to high x - ray luminosities in those sources . in our 25 ori observations
in which @xmath590% are wtts , we find that the median x - ray luminosity ( log @xmath39 ) is 29.7 erg @xmath40 and the four accreting sources have only marginally lower or , in one case , even higher x - ray luminosities . with similar x - ray luminosities to the wtts , these accreting sources should have cleared their inner gas disks in similar timescales .
if we look at the x - ray luminosities of the wtts in the 2.5 myr chamaeleon region , we find a median log @xmath41 @xmath40 , lower than all of the 25 ori accretors , indicating that if these x - ray luminosities may disperse an inner gas disk in @xmath52.5 myr , they would clearly be gone by 10 myr .
an additional parameter , besides a range of x - ray luminosities , is necessary to explain the diversity of disks observed at a given age .
when including the time evolution of the fuv radiation fields and their dependence on disk properties , like @xmath0 , @xcite find fuv fields evolving on a range of timescales related to differences in the accretion properties .
therefore , photoevaporation dominated by evolving fuv fields may disperse disks in a range of timescales .
with fuv driven photoevaporation and evolving fuv fields , the diversity of disks observed at any age may be expected .
this appears to be a plausible scenario given the large range of observed fuv luminosities and inferred @xmath0 s .
again , the full disks in the 25 ori association present an obstacle to the theory .
the observed fuv luminosities of the 25 ori accretors are @xmath5 2 orders of magnitude higher than the wtts in the 10 myr twa association , and given our results and those of @xcite , were likely stronger at 1 myr .
these fuv luminosities should be high enough to photoevaporate the gas disk , if fuv emission was responsible for the gas depletion in the twa wtts .
possible explanations for the remaining full disks in 25 ori , in the context of both x - ray and fuv photoevaporation theories , include low disk viscosities or high initial disk masses which increase the time for disk dispersal @xcite .
so far , we have focused on the gas in circumstellar disks ; however , it is unclear if gas and dust evolution proceed concurrently or independently @xcite .
@xcite modeled the affect of photoevaporation on the gas and dust in the circumstellar disk and found that the removal of gas from the disk has significant effects on the dust .
when photoevaporation creates a gap in the gas disk , the inner disk dust is also removed . also , pressure gradients formed when gas is removed cause the dust to migrate and may enhance grain growth .
for these reasons , we may expect to see dust evolution dependent on the strength of the photoevaporative fields . with constant x - ray fields throughout disk evolution , we focus on the accretion related fuv radiation fields
. @xcite find that dust dissipation and accretion are characterized by different timescales and to test this further , we compare fuv luminosities to 2mass @xmath42 @xcite , irac ( _ spitzer _ infrared array camera ) , and mips ( _ spitzer _ multiband imaging photometer ) colors .
irac and mips photometry , along with disk classifications for our source sample , were taken from the literature : taurus @xcite , chamaeleon i @xcite , chamaeleon ii @xcite , 25 ori @xcite , twa @xcite , and debris disks @xcite .
figure [ disk ] shows the relations between fuv luminosity and inner disk dust tracers . at @xmath43 m
we are tracing the dust in the inner regions of the disk ; as shown in @xcite , 60% of the 24 @xmath44 m emission is coming from inside 10 au when the dust has not settled towards the midplane .
the percentage of emission coming from within 10 au increases when observing at the shorter irac wavelengths , or when observing a disk in which some degree of dust settling has occurred . among the ctts ( circles ) in figure
[ disk ] , we find tentative correlations between fuv luminosity and each of the disk evolution tracers , with the most convincing correlation in @xmath45 .
however , we are likely observing two quantities which correlate with @xmath29 .
we showed in figure [ fuvvslaccage ] that the sources with highest fuv luminosities are also the strongest accretors .
disk properties also depend on the accretion rate through the disk and the ir spectral energy distributions reflect this ; models predict that lower @xmath0 s will produce bluer ir colors and that the ir spectrum becomes increasingly flatter as @xmath0 increases @xcite .
therefore , it is likely that these apparent trends are a result of accretion affecting both the fuv luminosities and the tracers of dust in the disk instead of a consequence of direct processing of the dust by fuv radiation .
this seems to indicate that if it is the fuv that is responsible for gas dispersal , the gas and dust are evolving independently . supporting this conclusion , the transitional @xcite and pre - transitional @xcite disks , the sources likely to have undergone dust processing and evolution , have the same fuv luminosities as the full disks , indicating that the fuv is not likely having a strong effect on the dust .
we present new observations of t tauri stars in the chamaeleon i and ii and the 25 ori star forming regions .
included are _ chandra _ x - ray and acs / sbc fuv observations which provide constraints on the photoevaporative fluxes present in young stars .
combined with previously published x - ray and fuv observations , we explored the evolution of high energy radiation fields with the following results : * with the addition of @xmath540 sources with x - ray observations in the 710 myr 25 ori star forming association , a significant addition to the @xmath525 previously observed 610 myr t tauri stars ( above the substellar limit ) , we confirm results that there is no decay in the x - ray luminosity between 1 and 10 myr . with consistently strong x - ray radiation fields , disk photoevaporation ,
if driven by x - rays , should proceed on short timescales .
however , observations of sources with remaining disk gas emission and strong x - ray luminosities at 710 myr and previous observations of tw hya ( a source with a transitional disk and strong x - ray luminosity ) still can not be explained by x - ray photoevaporation .
* we show , for the first time , the observed evolution of the fuv radiation fields in pre - main sequence stars from 1 myr to 1 gyr .
we find that the fuv luminosity decline can be explained , in part , by the decline of the accretion rate onto the star but note that some fuv emission must come from an additional source which is likely gas in the inner disk regions .
if fuv drives photoevaporation of the gas disk , the diversity of disks observed at 1 myr can be understood by the large range of @xmath0 s observed at 1 myr .
however , gas disks at 710 myr require additional explanation , as viscous @xmath0 evolution predicts that these sources should have had strong fuv radiation in the past , so disk dispersal by fuv radiation should have proceeded quickly . *
we show that the power law relations for the decay of high energy radiation fields in solar type active stars described by @xcite extend to the significantly younger sample discussed here . with both the x - ray luminosities and the slope of the fuv relation accurately predicted
, we expect that the difficult to observe euv radiation fields will similarly obey the power law trends .
this result tells us that strong radiation fields , necessary to remove the gas , are present throughout the disk dispersal phase .
* we find no correlation between fuv luminosities and tracers of dust evolution in the disk and furthermore , we find that sources with significant dust processing , the transitional disks , have similar fuv luminosities to full disk sources . therefore ,
if fuv radiation is the key emission field in gas evolution , it is not driving the dust depletion .
this would indicate that gas and dust evolution occur independently , though interestingly on similar timescales .
we thank kevin luhman for assistance in identifying possible new t tauri stars in chamaeleon i. we also thank jennifer blum and jun ji for their help with x - ray data reduction and analysis techniques .
we thank gregory herczeg for his assistance with reducing the acs chamaeleon data .
support for this work was provided by nasa through chandra awards go8 - 9028x and go8 - 9029x issued by the chandra x - ray observatory center , which is operated by the smithsonian astrophysical observatory for and on behalf of nasa under contract nas8 - 03060 .
support for this work was also provided by nasa through grant nos .
go-08317 , go-09081 , go-09374 , and go-11145 from the space telescope science institute , which is operated by aura , inc .
, under nasa contract nas 5 - 26555 .
c.e . was supported by the national science foundation under award no . 0901947 .
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 .
feigelson , e. d. , casanova , s. , montmerle , t. , & guibert , j. 1993 , , 416 , 623 feigelson , e. d. , broos , p. , gaffney , j. a. , iii , garmire , g. , hillenbrand , l. a. , pravdo , s. h. , townsley , l. , & tsuboi , y. 2002 , , 574 , 258 feigelson , e. d. , et al .
2005 , , 160 , 379 taurus + aa tau & & m0&@xmath46 & 0.016&_1.0_&_0.13_&1 , 15 + * bp tau&&k7&@xmath47&0.016&_1.3_&_0.23_&1 , 15 + ci tau & & k6&@xmath48 & @xmath49&_1.3_&_0.47_&1 , 15 + de tau & & m1&&0.013&_1.3 _ & _ 0.16_&1 + dl tau & & k7&&@xmath50&_1.0 _ & _ 0.32_&1 + * dm tau & & m1&@xmath51&@xmath52&_0.33_&_0.08_&1 , 15 + dn tau & & m0&@xmath47 & @xmath53&_1.3_&_0.04_&1 , 15 + do tau & & m0&&0.066&_1.3 _ & _ 0.29_&1 + dp tau & & m0&@xmath54 & @xmath55&_0.21_&_0.01_&1 , 15 + dr tau & & k7&&@xmath52&_1.6_&_1.03_&1 + fm tau & & m0&@xmath56&0.026&_0.52_&_0.30_&1 , 15 + fp tau & & m3&&@xmath57&_0.41_&_0.001_&1 + gk tau & & m0&@xmath58&@xmath55&_1.3_&_0.06_&1 , 15 + * gm aur & & k3&&0.016&_1.3_&_0.18_&1 + hn tau a@xmath59 & & k5&@xmath60&0.010&_0.21_&_0.02_&1 , 15 + hn tau b@xmath59&&m4&&@xmath61&_0.033_&&1 + ip tau & & m0&@xmath62 & @xmath49&_0.66_&_0.02_&1 , 17 + * lkca 15 & & k5&&@xmath50&_1.0_&_0.03_&1 + * ry tau & & g1&@xmath63&0.16&_10.4_&_1.6_&1 , 15 + * su aur & & g1&@xmath64&@xmath65&_8.3_&_0.10_&1 , 15 + * t tau & & g6&@xmath64&0.10&_8.3_&_0.90_&1 , 15 + uz tau a@xmath59 & & m1&&@xmath57&_0.33_&_0.02_&1 + uz tau b@xmath59 & & m2&@xmath66&@xmath67&_0.33_&_0.02_&1 , 15 + orion + * co ori & & g0&&0.16&_20.8_&_1.7_&1 + * ez ori & & g3&&0.10&_6.7_&_0.10_&1 + *
gw ori & & g0&@xmath68&0.26&_65.7_&_4.7_&1 , 16 + * p 2441 & & f9&&@xmath65&_10.4_&_0.4_&1 + * v1044 ori&&g2&&0.016&_6.7_&_0.6_&1 + chamaeleon i + chxr 8 & & k2&@xmath69 & & & &13 + chxr 30a & & k8&@xmath70 & & _ 1.3_&0&2 , 3 + chxr 30b & & m1&@xmath71 & & _ 0.21_&0&2 , 3 + cr cha & t8&k2&@xmath72 & @xmath73 & _ 2.6_&0.02&3 + cs cha & t11&k6&@xmath74&0.026 & _ 1.3_&0.01&2 , 3 , 18 + di cha & t26&g2&@xmath75&0.026 & _ 13.1_&0.15@xmath59&3 , 17 + fk cha & t29&k6& & @xmath35.2&_2.6_&1.2&3 + fl cha & t35&k8&@xmath76 & @xmath30.21&_0.52_&_0.04_&2 , 3 , 8 + fo cha & t47&m2&@xmath69 & & _ 0.41_&_0.01_&2 , 3 , 8 + hn 7 & & m5&@xmath71 & & _ 0.066 _ & 0&2 , 3 + hn 12w & & m6&@xmath71 & & _ 0.066_&0&2 , 3 + iso 86 & & &@xmath69 & && + iso 91 & & m3&@xmath77 & & _ 0.26_&&2 , 3 + iso 97 & & m1 & @xmath76 & & &&2 + iso 196 & & k7&@xmath78 & & & + sy cha & t4&k6&@xmath70 & & _ 0.52_&0.002&2 , 3 + sz cha & t6&k0&@xmath61 & @xmath49 & _ 2.1_&&3 + t25 & sz 18&m3& & @xmath3@xmath79&_0.26_&0.01&2 , 3 + t54 & & g8&@xmath80&@xmath53 & _ 4.1_&0.58@xmath59&3 , 18 + t56 & sz 45&m1&@xmath75&@xmath81 & _ 0.41_&0.003&2 , 3 , 17 + tw cha & t7&k8&@xmath70 & & _ 0.52_&0.01&2 , 8 + vw cha & t31&k8&@xmath63&0.041 & _ 3.3_&0.05&2 , 3 , 18 + chamaeleon ii + bf cha&sz 54&k5&@xmath58 & @xmath77&_2.1_&0.07&4 , 17 + bm cha&sz 61&k5&&@xmath82&_1.3_&_0.64_&4 + 25 ori + cvso 24&ob1a 499&m2 & @xmath83&&_0.16_&_0.01_&5 , 10 , 12 + cvso 29&ob1a 931&m3 & @xmath70&&_0.26_&&5 , 10 , 12 + cvso 34&cvso 221&m0&@xmath67&&_0.21_&0&5 , 9 + cvso 35&ob1a 1192&k7&@xmath84&@xmath55&_0.66_&0.05&5 , 10 , 12 + cvso 36&ob1a 1222&m2&@xmath84&&_0.21_&_0.03_&5 , 10 , 12 + cvso 38&ob1a 1317&m2&@xmath67&&_0.16_&_0.02_&5 , 10 , 12 + cvso 39&ob1a 1418&m2&@xmath73&&_0.33_&_0.01_&5 , 12 + cvso 43&ob1a 1716&m2&@xmath57&&_0.26_&_0.02_&5 , 12 + cvso 200&ob1a 489&m3&@xmath67&&0.21&&11 , 12 + cvso 206&ob1a 776&k7&@xmath57&@xmath49&0.26&2.3&10 , 11 , 12 + cvso 207&ob1a 864&k6&@xmath67&&0.66&&10 , 11 , 12 + cvso 211&ob1a 971&k4&@xmath84&&0.52&&10 , 11 , 12 + cvso 214&ob1a 1022&k2&@xmath73&&0.66&&10 , 11 , 12 + cvso 217&ob1a 1121&m1&@xmath73&&@xmath310.26&&11 , 12 + cvso 218&ob1a 1126&m4&@xmath57&&0.13&&10 , 11 , 12 + cvso 223&ob1a 1160&m3&@xmath73&&0.21&&10 , 11 , 12 + cvso 224&ob1a 1200&m4&@xmath85&@xmath61&0.13&_7@xmath86_&6 , 10 , 11 , 12 + cvso 225&ob1a 1235&m4&@xmath73&&0.26&&10 , 11 , 12 + cvso 228&ob1a 1316&m2&@xmath83&&0.16&&10 , 11 , 12 + cvso 230&ob1a 1326&m1&@xmath61&&0.33&&10 , 11 , 12 + cvso 233&ob1a 1417&m3&@xmath57&&0.052&&10 , 11 , 12 + cvso 235&ob1a 1548&m3&@xmath73&&0.16&&10 , 11 , 12 + ob1a 532&&m3&@xmath61&&0.26&&11 , 12 + ob1a 735&&m5&@xmath76&&@xmath310.21&&11 , 12 + ob1a 834&&m2&@xmath83&&0.16&&11 , 12 + ob1a 841&&m3&@xmath61&&0.16&&11 , 12 + ob1a 849&&m5&@xmath83&&0.16&&11 , 12 + ob1a 948&&m3&@xmath70&&0.13&&11 , 12 + ob1a 981&&k3&@xmath72&&0.83&&11 , 12 + ob1a 1088&&k6&@xmath72&&@xmath310.52&&11 , 12 + ob1a 1253&&m4&@xmath83&&0.10&&11 , 12 + ob1a 1586&&m5&@xmath85&&0.066&&11 , 12 + ob1a 1626&&k8&@xmath67&&@xmath310.83&&11 , 12 + ob1a 1630&&m2&@xmath87&@xmath88&0.16&0.09&11 , 12 + ob1a 1663&&m5&@xmath89&&0.10&&11 , 12 + ob1a 1755&&m3&@xmath70&&0.21&&11 , 12 + rx j0526.7 + 0143&&k0&@xmath88&&&&14 + j05244498 + 0159465&&k6&@xmath81&&0.66&&11 + j05260639 + 0137116&&m2&@xmath87&&0.041&&11 + j05263125 + 0141492&&m4&@xmath85&&0.033&&11 + twa + * tw hya & & k7&@xmath63&@xmath79&_0.32_&0.03&1 , 16 + hd 98800 & & k5&@xmath64&@xmath90&_0.66_&0&1 , 16 + twa 7 & & m1&@xmath56&@xmath78&_0.32_&0&1 , 16 + twa 13a & & m1&@xmath58&@xmath91&_0.16_&0&1 , 16 + twa 13b & & m1&@xmath58&@xmath91&_0.16_&0&1 , 16 + field + hd 12039 & & g4&@xmath62&@xmath89&_0.33_&0&1 , 16 + hd 202917 & & g5&@xmath75&@xmath87&_0.66_&0&1 , 16 + hd 53143 & & k1&@xmath92&@xmath71&_0.66_&0&7 , 16 + hd 61005 & & g8&@xmath48&@xmath91&_0.66_&0&1 , 19 + hd 92945 & & k2&@xmath93 & @xmath71&_0.41_&0&1 , 16 + mml 28 & & k2&@xmath75&@xmath76&_0.41_&0&1 , 16 + mml 36 & & k5&@xmath74&@xmath70&_1.0_&0&1 , 16 + * * * * * * * * * * * * * llcccccc 05243490 + 0154207 & k4k5&-0.46&0.10&29.0 + 05244265 + 0154116 & g9k0&1.54&0.32& + 05244320 + 0200355&&&& + 05251394 + 0143313&&&& + 05265823 + 0136078&&&& + 05270173 + 0139157 & k6&-0.39&0.17&29.3 + 10575375 - 7724495&&&& + lccccccc cr cha&1996&0.8&0.84&2.97&0.95&1.4@xmath94 + chxr 30a&324&12.2&&2.14&1.01&4.6@xmath95 + sy cha&283&0.8&0.34&2.94&0.81&2.4@xmath95 + sz cha&692&0.8&0.85&4.53&0.74&3.7@xmath95 + tw cha&313&0.8&0.41&1.95&1.08&1.8@xmath95 + cvso 38&266&1.1 & 0.98&6.98&1.42&4.1@xmath95 + j05244498 + 0159465&285&1.1&0.32&1.59&2.06&2.5@xmath95 + rx j0526 + 0143&369&1.1&0.31&1.48&2.29&2.6@xmath95 + | we present new x - ray and far ultraviolet ( fuv ) observations of t tauri stars covering the age range 1 to 10 myr .
our goals are to observationally constrain the intensity of radiation fields responsible for evaporating gas from the circumstellar disk and to assess the feasibility of current photoevaporation models , focusing on x - ray and uv radiation .
we greatly increase the number of 710 myr old t tauri stars observed in the x - rays by including observations of the well populated 25 ori aggregate in the orion ob1a subassociation . with these new 710 myr objects , we confirm that x - ray emission remains constant from 110 myr .
we also show , for the first time , observational evidence for the evolution of fuv radiation fields with a sample of 56 accreting and non - accreting young stars spanning 1 myr to 1 gyr .
we find that the fuv emission decreases on timescales consistent with the decline of accretion in classical t tauri stars until reaching the chromospheric level in weak t tauri stars and debris disks .
overall , we find that the observed strength of high energy radiation is consistent with that required by photoevaporation models to dissipate the disks in timescales of approximately 10 myr .
finally , we find that the high energy fields that affect gas evolution are not similarly affecting dust evolution ; in particular , we find that disks with inner clearings , the transitional disks , have similar levels of fuv emission as full disks . |
in 2008 the paul scherrer institut ( psi ) celebrated its 20@xmath0 anniversary and many years of delivering high intensity muon beams . several upgrades made the 590 mev/51 mhz ring cyclotron to be up to today the most powerful proton accelerator of its kind in the world , which delivers several 10@xmath1 muons per second to experiments .
the accelerator runs now routinely with 2.0 ma proton current and was already pushed to 2.15 ma for tests . in the near future running at 2.3 ma is foreseen , and an extensive program was launched to boost the operating proton current to 2.6 ma , by 2011 , and ultimately to 3.0 ma , envisaged for 2012 @xcite .
precision experiments should benefit from a correspondingly increased muon intensity .
charged lepton - flavor conservation has been empirically verified to a high precision , but is not a consequence of a known underlying symmetry . the decay @xmath3 is lepton - flavor violating and hence , excluding neutrino flavor mixing , forbidden within the standard model ( sm ) .
neutrino masses and mixing , which is established now , introduce a contribution to this decay within the sm , however , on an unmeasurably small level of order @xmath410@xmath5 @xcite . on the other hand , there are several attractive theories beyond the sm , such as supersymmetry , which generally predict lepton - flavor - violating processes at a level within today s experimental reach
. a corresponding experimental signal would be free of sm background and hence a clear indication for ` new physics ' .
the goal of the meg experiment at psi @xcite is to reach a sensitivity of 10@xmath6 , improving the present limit @xcite by almost 2 orders of magnitude .
consequently one needs a detector managing a challenging high muon stop rate up to @xmath7 muons / s .
the experimental principle is based on the simultaneous detection of the back - to - back emitted mono - energetic decay positron and gamma .
the positrons are detected in high rate drift - chambers located in a magnetic field for momentum determination and in scintillation counters for timing .
the gammas are detected in the world s largest liquid xenon scintillation counter , as sketched in fig.[meg - apparatus ] .
excellent timing , energy and spatial resolution for both reaction products are required to beat the main background caused by ordinary muon decay and pile - up .
2008 saw the first months of physics run of meg and the accumulated statistics looks promising to already improve the present limit on @xmath3 significantly .
the fermi constant @xmath8 describes the strength of the charged - current weak interaction . along with the fine structure constant @xmath9 and the @xmath10-boson mass , it is one of the three pillars of the electroweak standard model and directly related to the electroweak gauge coupling @xcite .
the most precise determination of @xmath8 is based on the mean lifetime of the positive muon , @xmath11 , and can be extracted from : @xmath12 with @xmath13 representing higher order qed and hadronic corrections as well as finite - lepton - mass phase space factors , which have only recently been computed to a sub - ppm level @xcite . a first computation of order @xmath14 using a finite electron mass shifted the value of @xmath13 by another 0.43ppm @xcite . hence , a comparably precise experimental determination of @xmath11 is highly desirable .
the mulan experiment @xcite installed a muon beam kicker on the pie3 beamline at psi , which allows after directing positive muons onto a target for a selectable time period ( e.g. 5 @xmath15s ) , to steer away the beam for the following , for instance , 22 @xmath15s , the decay positrons are recorded in a soccer - ball shaped detector ( see fig.[mu - lifetime - plot]b ) made from 170 double - layer scintillator tiles , which are read out via custom - made 500 mhz fadc modules able to separate pulse pile - up events on the ns level .
systematic issues , caused by positron detection differences in the counters , due to polarized muons precessing in the earth s magnetic field , are dealt with via measurements in different targets , which are in a homogeneous magnetic field and either fully maintain the muon polarization ( silver ) , depolarize the muons to a large extent ( sulphur ) , or cause a very fast muon precession due to an internal few tesla high magnetic field ( arnokrome@xmath16-iii ) .
several 10@xmath17 muon decays were recorded for each target .
the first mulan result , based on part of the data has set a new precision benchmark , as shown in fig.[mu - lifetime - plot]a .
additionally , several dedicated systematic measurements are presently under analysis .
the final precision goal on @xmath11 is 1ppm , which translates into a 0.5ppm precision on @xmath8 .
the fast experiment @xcite relies on the detection of the full decay sequence @xmath18 and corresponding times in a fast imaging target made of 32 x 48 pixels , constructed from plastic scintillator bars in a homogeneous b field .
this approach allows a good control of muon polarization effects .
fast is scheduled to achieve a statistics of several @xmath19 in 2008/2009 .
its goal is a 2ppm measurement of @xmath11 .
as a by - product , fast can also measure the @xmath20 lifetime and improve the present world average .
the determination of the proton s weak pseudoscalar coupling constant @xmath21 has been the driving force behind decades of muon capture measurements . the psi result on the muon capture rate in @xmath22he @xcite has set a precision landmark in this field .
however , with 3 involved nucleons some questions still remained in the precise theoretical prediction .
a specially exciting turn came with the precise triumf results from a measurement of radiative muon capture ( rmc ) in hydrogen @xcite , which disagreed with theory and results derived from ordinary muon capture ( omc ) measurements @xcite , as shown in fig.[gp - plot]a .
after decades of worldwide experimental efforts , the mucap experiment has achieved the first unambiguous determination of the proton s pseudoscalar coupling @xmath21 @xcite and has solved a longstanding discrepancy .
the result is in excellent agreement with recent calculations based on heavy baryon chiral perturbation theory ( hbchpt ) @xcite .
experimental determinations of @xmath21 depend on the ortho - para transition rate @xmath23 in the @xmath24 molecule .
the most precise previous measurement of ordinary muon capture ( omc ) @xcite and the rmc experiment @xcite both depend significantly on the value of @xmath23 , which itself is poorly known due to mutually inconsistent experimental @xcite and theoretical @xcite results . in contrast , the first mucap result for @xmath21@xcite is almost independent of molecular effects . sc/@xmath15pc - muon entrance counters , tpc - time projection chamber = target , epc , esc - electron tracking and timing counters.__,title="fig : " ] the mucap result was only possible with an enduring joint effort and a rigorous experimental technique @xcite . the setup is shown in fig.[gp - plot]b
. the active target , a time projection chamber , was filled with 10 bar of ultra - pure ( high @xmath10 contamination in the few ppb range ) and isotopically pure hydrogen @xcite .
muon stops and corresponding decay electron tracks were recorded in 3 dimensions , which allowed for very selective cuts and hence an unprecedented control and possible study of systematic effects .
specifically , muon capture events on high @xmath10 elements are even on the ppb contamination level visible in the mucap detector .
target conditions were selected in order to control effects due to muonic molecule formation and muon catalyzed fusion .
the muon capture rate was finally determined by comparing the lifetime of negative muons in hydrogen with the positive muon lifetime from the more precise result @xcite . in order to extract @xmath21 ,
the singlet muon capture rate was compared to two recent calculations @xcite adding the newly calculated radiative correction @xcite . presently
, the full data set of more than 10@xmath25 recorded @xmath26 stops in hydrogen is being analyzed in a blind analysis , with the final precision goal of 1@xmath27 on the singlet muon capture rate in hydrogen .
20 s@xmath28 to the total rate .
b ) estimations on the axial two - body current term @xmath29 .
all methods up to now include certain assumptions and approximations which might be questioned .
all references are given in @xcite _
, title="fig : " ] knowing @xmath21 from mucap facilitates the interpretation of the doublet muon capture rate on the deuteron ( @xmath30 ) , the measurement goal of the recently started musun experiment @xcite which also aims at 1@xmath27 precision .
such a result would allow a precise test of modern effective field theories and would represent the most stringent test of electro - weak interaction in a two - body system @xcite .
moreover , it would allow determination of the axial two - body current term which scales with the low energy constant denoted @xmath29 ( or d@xmath31 ) @xcite .
this parameter is of astrophysical interest , as it appears in the same form in the cross - section for i ) muon capture on the deuteron , ii ) @xmath32 fusion the main fusion reaction in the sun , and iii ) neutrino - deuteron reactions , which are the detection reactions used by the sudbury neutrino observatory @xcite .
hence , via the absolute neutrino rates , the precise @xmath15d capture rate determination would ` calibrate ' the sun .
existing experimental results on @xmath30 are not precise enough ( fig.[l1a - plot]a ) and also using other sources leaves the present experimental knowledge on @xmath29 rather sparse , as shown in fig.[l1a - plot]b . with precision neutrino physics on the horizon
the precise knowledge of @xmath29 will be necessary , as it also influences the determination of @xmath33 and @xmath34@xcite .
the experimental principle of musun will follow the successful mucap approach , but for control and optimization of muonic molecule formation and muon catalyzed fusion reactions in deuterium one has to use a high density cryo - target at @xmath430k @xcite .
@xmath35 fusion reactions occur at much higher rates than muon capture and hence represent a severe background , but also allow to monitor the hyperfine populations of the muonic atom @xcite .
high @xmath10 target purity will be even more critical as in mucap . in a first engineering run a new pad - based tpc
was successfully tested with high purity deuterium in late 2008 .
there are several other precision measurements of fundamental parameters ongoing or under discussion at psi , which either will test the standard model or search for new physics .
* the muonic lamb - shift experiment @xcite is preparing for its first physics data taking in 2009 , and wants to precisely determine the proton charge radius via observation of the 2p-2s energy difference in muonic hydrogen . * a precise test of the electron - muon universality
is being performed within the pen experiment and corresponding data are being analyzed @xcite .
* the search for the lepton - flavor - violating process @xmath36 would be a sensitive search for new physics and complement the present meg activity .
there is an ongoing discussion on 2 suggested experimental approaches , how to obtain a sensitivity which improves the present experimental limit by roughly 3 orders of magnitude @xcite . * a sensitive search for a cp violating muon electric dipole moment ( edm ) was suggested in @xcite using a compact storage ring , which could make use of psi s high muon intensity and reach a sensitivity of 5@xmath3710@xmath38 e@xmath39 cm . in this way
it would test ` new physics ' and pave the way for higher sensitivity tests of muon and other charged particle edms . * the discussion about dark matter and dark energy
has also put interest in particles decaying into mirror worlds , other dimensions or to other branes .
hence the decay products would be invisible .
a search for the invisible decay of muons was suggested in @xcite , and might be also searched for by using the mucap setup . *
high brightness muon beams would also allow a first test of the gravitational interaction of antimatter of a purely leptonic system , which involves second generation particles , namely muonium ( @xmath40 ) @xcite . given these and more ideas
, one can be sure that precision measurements using muons , at psi and other facilities in the world , will also in future contribute to a deeper understanding and testing of the standard model and provide a fair chance to first find ` new physics ' beyond our presently accepted theory .
i would like to express my sincere gratitude for advice , support and many stimulating discussions to k. kirch , p. kammel , d. hertzog , s. ritt , c. casella , c. petitjean , m. seidel , and especially all members of the mucap , mulan and musun collaborations .
p. kammel in : proceedings of the exa02 - international workshop on exotic atoms - future perspectives , vienna , austria , 2002 ; austrian academy of sciences press , vienna , 2003 ; arxiv : nucl - ex/0304019 ; b. lauss in : proceedings of the exa05 - intern .
conference on exotic atoms , vienna , austria , 2005 , austrian academy of sciences press , vienna 2005 ; arxiv : nucl - ex/0401005 . | muons can serve as probes to precisely determine fundamental parameters of the standard model or search for ` new physics ' .
the high intensity muon beams at the paul scherrer institut ( psi ) allow for precision measurements and searches for rare or forbidden processes . both types of experiments challenge the standard model in a way complementary to high energy physics .
we give a short overview of recent results and ongoing experiments at psi , and of ideas for the future .
muon physics , rare decay , muon lifetime , fermi constant , muon capture 14.60ef , 13.35bv , 12.15.-y , 21.45.bc , 25.30.mr , 13.15.+g |
recently , iwasawa _ et al . _
( 1996 ; hereafter i96 ) found an extremely broad iron k@xmath0 emission line with a particularly prominent red wing , extending down to @xmath4 kev , during an extended observation with the _ advanced satellite for cosmology and astrophysics _ ( _ asca _ ; see tanaka , inoue & holt 1994 ) .
the peculiar line profile , measured during what has been dubbed a ` deep minimum ' ( or dm ) in the x - ray light curve , had much more emission in the red wing , and much less around the line rest energy ( @xmath5 kev ) , compared to the fe k@xmath0 profiles thus far measured for mcg @xmath16@xmath130@xmath115 and other agns ( e.g. tanaka _ et al .
_ 1995 , yaqoob _ et al .
_ 1995 , nandra _ et al .
_ 1997 ) . also , the equivalent width of the line was unusually large , of the order of @xmath6 kev , a factor of @xmath7 larger than most such measurements of broad iron k lines in seyfert 1 galaxies . for the dm state of mcg @xmath16@xmath130@xmath115
, i96 rejected the standard model for the fe - k@xmath0 line profile ( e.g. see fabian _ et al .
_ 1989 ) , in which the line is emitted in an accretion disk rotating about a schwarzschild black hole , with the inner disk extending no closer than the radius for marginally stable orbits , ( i.e. 6 gravitational radii , or @xmath8 ) . instead
, the extreme gravitational redshifts implied by the bloated red wing of the line were explained either in terms of emission from inside the last stable orbit around a schwarzschild hole ( reynolds & begelman 1997 ) or emission in a disk rotating about a kerr black hole ( e.g. i96 ; dabrowski _ et al .
_ 1997 ; bromley , miller & pariev 1998 ) . in the latter case
the last stable orbit extends down to @xmath9 for a maximally rotating kerr black hole . in both cases ,
the red wing of the line is accounted for by emission closer to the black hole event horizon , so the photons can experience the effects of very strong gravity , resulting in extremely large redshifts .
we point out that in terms of fitting the dm data for mcg @xmath16@xmath130@xmath115 with the standard model ( schwarzschild hole , disk extending to @xmath8 ) and the kerr model ( disk extending to @xmath10 ) , the largest difference in the fitting statistic , @xmath11 , is 6.2 for the same number of free model parameters ( see rows 1 and 3 in table 3 of i96 who assume a disk inclination of 30@xmath12 and an outer radius of @xmath13 ) .
while this may be interpreted as being formally statistically significant , _ asca _ spectral fits do not in general ( and in this case , in particular ) include the effects of systematic errors which could reduce the overall significance of the result .
since the implications of really being able to observe x - rays inside of @xmath8 and even closer to a black hole event horizon are so far reaching ( e.g. see fabian 1997 ) it is important to investigate the robustness of the result for mcg @xmath16@xmath130@xmath115 , the only case thus far reported .
_ asca _ observed mcg @xmath16@xmath130@xmath115 for over 4 days starting 1994 , july 23 . _
asca _ has four identical , thin - foil , light - weight x - ray telescopes ( xrt ) which focus x - rays onto one of two solid - state imaging spectrometers ( sis ) or one of two gas imaging spectrometers ( gis , see ohashi _ et al .
see tanaka _
( 1994 ) for a summary of the _ asca _ mission and focal - plane detectors .
the sis sensors , each one consisting of four ccd ( charge coupled device ) chips were operated in a mode in which only one chip was exposed ( 1-ccd mode ) and the data were accumulated in faint mode .
hereafter the two sis sensors are referred to as sis0 and sis1 and the two gis sensors as gis2 and gis3 .
the data reduction and selection criteria are similar to those described in yaqoob _
( 1994 ) .
the lightcurve of the entire observation has been presented elsewhere ( i96 ; reynolds _ et al .
_ 1997 ; yaqoob _ et al .
we use exactly the same time intervals defined by i96 to extract spectra of the dm state ( interval @xmath14 in their figure 2 ) and the flare state ( interval @xmath15 ) , as well as the average ( total ) spectrum . for the dm
, we obtained @xmath16 kev count rates in the range 0.13 to 0.16 counts s@xmath17 and exposure times in the range 13.2 to 13.3 ks for the four instruments .
figure 1 shows the ratio of the data in the dm to the best - fitting power - law model ( @xmath18 = 1.92 , @xmath19 = @xmath20 @xmath21 ) using data only in the energy ranges @xmath22 kev and @xmath23 kev .
the excess above the underlying power - law is due to the fe - k@xmath0 line emission .
as pointed out by i96 , the emission on the blue side of the line is unusually diminished compared to the red side .
the portion of the lightcurve containing the dm state is shown in figure 2 .
using data between 3 and 10 kev from all four instruments , we fitted the fe - k@xmath0 line for the average , flare and dm spectra with our ` baseline ' model in which the line photons are emitted in a disk rotating around a central schwarzschild black hole ( e.g. , fabian _ et al .
the parameters are @xmath24 ( inclination angle of the disk normal relative to the observer ) , @xmath25 ( inner disk radius ) , @xmath26 ( outer disk radius ) , @xmath27 ( power - law index characterizing the line emissivity as @xmath28 ) , @xmath29 ( line intensity ) , and @xmath30 ( line energy in the disk rest frame ) .
the inner radius , @xmath25 , was fixed at @xmath8 where @xmath31 ( i.e. the last stable orbit ) .
there is interplay between @xmath30 and the other line parameters , so @xmath30 was fixed at 6.4 kev in the rest frame , corresponding to fluorescence in a cold disk .
the results are shown in table 1 , models sh1 , sh2 , and sh3 .
we repeated the above fitting , but this time replacing the schwarzschild black hole with a maximally rotating kerr black hole ( see e.g. , laor 1991 ) .
now the inner radius is fixed at @xmath10 , the minimum value , or last stable orbit for the kerr metric .
the spectral fit was repeated with @xmath30 fixed at 6.7 kev ( corresponding to he - like fe ) . for the average and flare spectra ,
a kerr profile does not improve the fit , but for the dm , the kerr model provides a statistically better fit than the schwarzschild model ( @xmath32 ) . the results for the dm are shown in table 1 , models kr1 and kr2 .
the basic reason why the fe - k@xmath0 line during the dm is better described by emission from matter closer than @xmath8 is that the observed strength of the blue peak of the line is too small for the observed strength of the red wing .
this is compensated by the extra gravitational redshifts obtained by placing the emitter closer to the black hole , increasing the intensity of the red wing relative to the blue .
our results confirm those of i96 , although the best - fitting parameters differ slightly because we fix @xmath27 at 2.5 and allow @xmath24 and @xmath26 to be free , the contrary to their technique .
we tested whether we could fit the dm fe - k@xmath0 line profile without placing line - emitting matter closer to the black hole than @xmath33 by obscuring part of the standard schwarzschild disk with optically - thick matter .
figure 3 illustrates how different parts of a schwarzschild disk affect the doppler and gravitational shift of line photons . defining @xmath34 ( including both doppler and gravitational effects ) , the shaded part labeled blue corresponds to @xmath35 and
the shaded part labeled red corresponds to @xmath36 . in our simple - minded model ,
the occultation of the disk is in the form of a strip running along the @xmath37 or @xmath38 directions , with half - width @xmath39 or @xmath40 respectively , passing symmetrically over the center of the disk .
thus , line emission from either @xmath41 or @xmath42 is blocked from view and the predicted line profiles for the two cases are shown in figures 4a and 4b respectively ( for several values of the half - width ) .
we use this model as a hypothesis , since we do not know the geometry of the optically - thick clouds ( there are plenty of arguments for their existence , e.g. guilbert & rees 1988 ) .
the clouds could be spherical , filament - shaped , or neither .
our simple model in some sense mimics an optically - thick cloud , whatever its shape , moving over the disk , resulting in some kind of average line profile .
clearly , the profiles in figure 4a , in which much of the blueshifted emission is obscured , are relevant for the dm data . for the spectral fits with the obscuration model , we fix @xmath43 kev and @xmath44 as before , and @xmath45 from the schwarzschild fit ( table 1 , sh3 ) .
the best - fitting parameters are shown in table 1 , model oc1 .
this fit is marginally better than the kerr model with @xmath46 kev ( model kr1 ) and marginally worse than the kerr model with @xmath47 kev ( model kr2 ) and so we conclude that the occulted , schwarzschild disk - line model describes the spectrum during the dm just as well as the standard kerr model .
the above fits include only galactic absorption because we assume that there is no significant absorption from the ` warm absorber ' present above 3 kev that will affect the shape of the continuum ( fabian _ et al .
_ 1994 ; reynolds _ et al .
_ 1995 ; otani _ et al .
this was also the assumption made by i96 and is based on observations of the warm absorber in mcg@xmath16 - 30 - 15 in higher flux states .
however , if absorption were present , this could also affect measurements of the fe - k@xmath0 line .
figure 5 shows the data / model ratio when the dm spectrum is fitted with galactic absorption and a power law with the best - fitting photon index of the average spectrum ( @xmath48 ) .
the deficiency of low - energy photons due to the ionized absorber is significant , even _ above _ 3 kev , and implies that the flattening of the continuum in the dm ( table 1 , models sh3 , kr1 , kr2 , oc1 ) is due to absorption .
thus , the _ intrinsic _ spectral slope during the dm is not required to change .
we therefore fitted the dm spectrum again with the occultation - line model with @xmath18 fixed at the average value of @xmath48 , but with free absorption ( table 1 , model oc2 ) .
the number of free parameters is the same as before , but @xmath49 compared to the best - fitting kerr model ( model kr2 ) .
the occultation description of the dm is thus as viable as the scenarios posited by i96 , reynolds & begelman ( 1997 ) , dabrowski _ et al . _
( 1997 ) , and bromley _
we have examined the fe - k@xmath0 line profile during the deep minimum ( dm ) flux state for mcg@xmath16 - 30 - 15 with an ` occultation model ' in which the continuum source and much of the blueshifted part of the emission from the disk is obscured by an optically - thick cloud .
this model describes the observed line profile as well as the kerr black - hole / disk model , but does not require emission from inside @xmath50 ( the last stable orbit for a schwarzschild black hole ) .
one difficulty with the kerr interpretation proposed by i96 and others is that it does nt explain why kerr effects should only manifest themselves in the dm .
the occultation model does not suffer from this problem .
also , the line intensity is _ not _ required to increase relative to the average line intensity , as has been previously claimed .
nor is the equivalent width ( ew ) of the fe - k line during the dm required to be unusually large .
we obtain @xmath51 ev for the dm _ and _ flare - states , and @xmath52 ev for the average line .
in addition to the optically - thick obscurer that blocks most of the continuum region , the data above 3 kev are affected by the o viii edge of the warm absorber that becomes stronger in the low - flux state ( otani _ et al .
we measure an additional column density of @xmath53 @xmath21 in the dm , which is a factor of @xmath54 increase compared to the high flux state ( fabian _ et al .
the increase in column density is consistent with a picture in which the material that is photoionized by the central source becomes less opaque to soft x - rays due to a reduction in ionization when the continuum is blocked .
note that the extra absorption we directly measure is _ not _ necessarily related to the optically - thick clouds responsible for obscuring the fe - k@xmath0 line emission . to conclude , we have shown that the unusual fe - k@xmath0 line profile in mcg @xmath16@xmath130@xmath115 during a deep minimum in intensity does not necessarily require line emission from within @xmath33 and that the data are not adequate enough to distinguish between the schwarzschild and kerr solutions .
( 1998 ) have also made the suggestion that a kerr black hole is not required , proposing that the true shape of the fe - k@xmath0 line is a double gaussian , but offer no physical model .
we have shown that the deep - minimum line profile can be explained by occultation of a standard schwarzschild disk .
this model is consistent with the detailed temporal profile of the continuum intensity dip .
our results serve to highlight the hazards of over - interpreting observational results which have low statistical significance , to the extent that theoretical implications can become generally accepted when the data do not provide a strong case for them .
llccccccc average & sh1 & 1.98@xmath55 & 15.6@xmath56 & @xmath57 & 1.88@xmath58 & @xmath59 & @xmath60 & 1417.9/1355 flare & sh2 & 2.00@xmath61 & 6.1@xmath62 & 39.4@xmath63 & 2.29@xmath64 & 410@xmath65 & @xmath60 & 925.6/899 deep min . & sh3 & 1.71@xmath66 & 18.5@xmath67 & @xmath68 & 1.52@xmath69 & @xmath70 & @xmath60 & 305.2/323 deep min .
& kr1 & 1.73@xmath71 & 19.8@xmath72 & 31@xmath73 & @xmath74 & 1070@xmath75 & @xmath60 & 299.6/323 deep min . &
kr2@xmath76 & 1.73@xmath77 & @xmath78 & 27@xmath79 & 2.42@xmath80 & 1130@xmath81 & @xmath60 & 297.6/323 deep min . &
oc1@xmath82 & 1.74@xmath83 & 18.5f & 46@xmath84 & 1.63@xmath85 & @xmath86 & @xmath87 & 298.2/323 deep min . &
oc2@xmath88 & 1.98f & 18.5f & @xmath89 & @xmath90 & 404@xmath91 & @xmath92 & 291.9/323 | we examine the unusual fe - k@xmath0 line profile in mcg @xmath16@xmath130@xmath115 observed by @xmath2 during a deep minimum in the source intensity .
the intense red wing and depressed blue wing of the line have been interpreted as evidence for extreme gravitational redshifts in terms of emission from within six gravitational radii of a black hole .
we find that the data do not uniquely support this interpretation and can be equally well explained by occultation of the continuum source and the putative line - emitting accretion disk , which we offer as an alternative hypothesis .
two problems with previous modeling were that the equivalent width of the line during the deep minimum was required to be unusually large ( @xmath3 kev ) and the line intensity was thought to increase as the source became dim .
the occultation model does not suffer from these problems .
our results serve to highlight the hazards of over - interpreting observational results which have low statistical significance , to the extent that theoretical implications can become generally accepted when the data do not provide a strong case for them . |
the kinetics of polymer loop formation has been studied for several decades and recently has attracted renewed attention due to the particular importance in biology .
the dna loop formation is a basic process that underlies genetic expression , replication , and recombination @xcite .
for example , in _ e. coli _ the _ lac _ repressor ( laci)-mediated loop is crucial for the repressive regulation of _ lac _ genes .
the hairpin loop formation is the elementary step in protein folding @xcite and structure formation in rna folding @xcite .
a cell is crowded with a multitude of subcellular structures including globular proteins and rnas @xcite , with which dna is constantly interacting . a dna fragment about to loop is often subject to temporally fluctuating forces due to its dynamic environment including the other part of the chain .
recently , the power spectrum of the fluctuating force exerted on cytoskeleton was measured to be an order of magnitude larger than that expected from thermal equilibrium condition @xcite .
this indicates that the cell interior is an active and nonequilibrium medium .
the advance of single molecule experiment techniques provides detailed information on the dna loop formation .
finzi and gelles @xcite observed laci - mediated dna loop formation and dissociation by monitoring nano - scale brownian motion of the micron - sized particle attached to one end of the dna .
et al . _
@xcite showed that in _ gal _ repressor and dna - bending protein hu mediated looping , mechanical constraints such as tension and torsion play a pivotal role .
gemmen _ et al . _
@xcite studied effects of tension in the presence of two - site restriction enzymes which can cut the dna upon binding on two sites simultaneously .
they found that the cleavage activity decreases approximately 10-fold as the tension increases from 30 fn to 700 fn .
they also found that the optimum loop size decreases with the tension , which is qualitatively in agreement with theoretical predictions @xcite .
more recently , chen _ et al . _
@xcite studied effects of tension in femtonewton range on the kinetics of laci - mediated dna looping .
they found that small tension of 100 fn scale on the substrate dna can not only increases the looping time @xcite but also found that the looping time is greatly reduced in the presence of a fluctuating tension @xcite .
these results suggest the ubiquitous roles of the static and temporally fluctuating tensions in regulation of the dna loop formation . yet
, there appears to be no unifying conceptual or theoretical framework that explains a variety of experiments including these . theoretically , on the other hand , yan _
et al . _
@xcite developed a transfer matrix method to calculate semiflexible polymer end - to - end distance distribution function and loop formation probability ( or @xmath0-factor ) .
they studied various effects of nonlinear elasticity arising from dna bending protein - induced kinks @xcite or thermal - fluctuation - induced bubbles on the @xmath0-factor .
their study provides a valuable insight to understand dna bending on short length scale @xcite , which has attracted much attention recently @xcite .
they also studied effects of tension on the @xmath0-factor @xcite , which is related to the free energy barrier for loop formation @xcite , thus to the loop formation rate .
similar results are obtained by using an elastic theory of a semiflexible polymer @xcite . however , since the loop formation rate is not proportional to the @xmath0-factor alone but depends on the free energy given an arbitrary chain end - to - end distance , it is hard to quantitatively compare these theories to the experiment @xcite .
independently , blumberg _ et al . _
@xcite studied effects of static tension on protein - mediated dna loop formation by modeling the dna conformation to be either one of two states , looped and unlooped states . in their appealing calculations of free energy change associated with the transition under the tension , they considered not only the stretching free energy of dna but also dna alignment constraint imposed by protein binding .
they found that for the loop size larger than 100 base pair distance ( @xmath1 nm ) , a tension of 0.5 pn can increase the looping time by more than two order of magnitude .
there is room for improvement in their approach , however , on the evaluation of the free energy that can be valid for short end - to - end distance of the chain as well as a description of detailed kinetic process using the mean first - passage time approach . in this paper , in an effort to understand the basic physical mechanism of the biopolymer looping in a coherent manner , we perform brownian dynamics simulation of semiflexible polymers treated as extensible wormlike chain , combined with one - dimensional theory of barrier crossing over the free energy of loop formation . for analytical understanding ,
we use , as an example , the mean - field wormlike chain model @xcite , which is shown to be a good approximation for the free energy for the chain lengths we consider here . with static tensions , we find that the looping time , defined as the mean first - passage time to cross the free energy barrier , steeply increases with the applied tension @xmath2 , in an agreement with our simulation results but distinct from the previous theoretical result @xcite . for the case of time - dependent tension , we consider dichotomically fluctuating tension , where the looping times are found to be reduced , consistent with the experiment @xcite .
most importantly , we find so - called the resonant activation , where the looping time is the minimum at an optimal flipping time of the dichotomic force . in this exploratory study
, we neglect the alignment constraint on the loop formation , which is minor effect for the chain lengths we consider here @xcite . in the following section
, we describe our polymer model and simulation method , whose results are discussed in sec .
[ sec : results ] .
we summarize our results in sec .
[ sec : conclusion ] .
we consider the semiflexible polymer looping in the presence of static and fluctuating tension .
the polymers are modeled as semiflexible chains of @xmath3 beads of diameter @xmath4 , with the interaction potential @xmath5 .
here @xmath6 , where @xmath7 and @xmath8 are the stretching and bending energy @xmath9 @xmath10 where @xmath11 is the stretching stiffness , @xmath12 is the natural bond length , @xmath13 is the bending stiffness , @xmath14 is the position of the @xmath15th bead , and @xmath16 is the angle of the @xmath15th bond .
.[fig : schematic],width=321 ] the dynamics of @xmath3 beads ( @xmath17 ) are described by the overdamped langevin equation @xmath18 where @xmath19 is the friction coefficient of the bead , and @xmath20 is the gaussian random noise with mean @xmath21 and variance @xmath22 . here
, @xmath23 denotes ensemble average , @xmath24 and @xmath25 are the cartesian coordinate indices , @xmath26 is the boltzmann constant , and @xmath27 is the absolute temperature .
the additional force @xmath28 is the tension which is applied only to the two end segments ( @xmath29 and @xmath3 ) along the chain end - to - end direction , @xmath30 ( see fig .
[ fig : schematic ] ) .
the bending stiffness of the short semiflexible chains we consider allows us to neglect the excluded volume effect .
furthermore , we neglect the hydrodynamic interactions , which is found to be small in the short chains we consider here @xcite .
we use the parameters @xmath12 , @xmath31 , and @xmath32 to fix the length , energy , and time scales , respectively .
the dimensionless parameters in our simulation are @xmath33 , @xmath34 , @xmath35 , and @xmath36 . in our model , we consider @xmath37 as 10 nm , so with @xmath36 the persistence length @xmath38 of the chain is 50 nm .
the diameter of the bead is set to be 5 nm to fit the hydrodynamic friction coefficient of a cylinder with 2 nm diameter and 10 nm length , which is about @xmath39 @xmath40 @xmath41 with the water viscosity @xmath42 @xmath40 @xmath43 @xmath41 at room temperature .
the time unit is then 1.14 @xmath44s .
the equations of motion are integrated by using a second - order stochastic runge - kutta algorithm @xcite with the time - step @xmath45 . in our simulation
, we consider that the looping occurs whenever the chain end - to - end distance @xmath46 is shorter than a cutoff distance @xmath47 .
the average were taken over at least 2,000 and 5,000 independent runs for static and fluctuating tension cases , respectively .
we first study the effects of static tension on semiflexible polymer looping . here
we consider the bead number @xmath48 , 18 , and 24 which respectively correspond to the chain lengths @xmath49 , 170 , and 230 nm , similar to the loop size considered in recent experiments @xcite .
figure [ fig : n12_static1 ] shows the looping time @xmath50 as a function of tension @xmath2 for the chain of @xmath48 and @xmath51 .
we apply the tension @xmath2 from -81 to 162 fn , comparable in magnitude to the typical entropic force on the double - stranded dna , @xmath52 @xmath53 80 fn .
the @xmath2 is much smaller than the piconewton scale forces those are typically involved in active molecules @xcite e.g. , molecular motors in a cell . as @xmath2 changes over the scale as small as 100 fn , the looping time @xmath54 dramatically changes ; for example , @xmath54 increases about 5 times as the tension @xmath2 increases 120 fn . as a function of tension @xmath2 for the chain of @xmath48 .
[ fig : n12_static1],width=321 ] to understand this behavior , we consider semiflexible polymer looping as a one - dimensional barrier crossing process @xcite , which is described by the langevin equation @xmath55 here , @xmath56 is the free energy of the chain given the chain end - to - end distance @xmath57 , where @xmath58 is the radial distribution function , and @xmath59 is a random force due to thermal fluctuation given by a gaussian and white noise that satisfies @xmath60 and @xmath61 . the shape of @xmath62 obtained by the @xmath3-beads simulation is shown in fig .
[ fig : n12_mfwlc ] ( solid line with circles ) for the chain of @xmath48 . in this one - dimensional description ,
the looping time is the mean first - passage time ( mfpt ) for the variable @xmath57 to reach the cutoff distance @xmath47 starting from the initial chain end - to - end distance @xmath63 .
it is given by @xcite @xmath64 where @xmath65 is the relative diffusion coefficient of two end beads , @xmath66 .
then the looping time is @xmath67 , where @xmath68 represents the average over the initial equilibrium distribution .
the mfpt in the presence of tension @xmath2 , also can be calculated by using the free energy , @xmath69 , where @xmath70 is the free energy without tension . for the @xmath58
, we use the mean - field wormlike chain model ( mf - wlc ) @xcite as an example .
the radial distribution function @xmath58 of this model is @xmath71^{-\frac{9}{2 } } \exp[-\frac{3l}{4 l_{p } } \frac{1}{(1-(r / l)^2)}]$ ] .
this formula yields a reasonable approximation to our simulation result for @xmath70 , except for @xmath72 , as shown in fig .
[ fig : n12_mfwlc ] .
the large deviation between two curves in the region near @xmath73 is because our model allows the chain extensibility while mf - wlc does not .
however , according to the single molecule experiment using optical tweezers @xcite , the dna is extensible , manifesting overstretching transition subject to a strong tension , thus showing a significant deviation from the inextensible wlc .
the optimal value @xmath11 to fit the force - extension curve is found to be @xmath74 2700 @xcite . however , the simulation using @xmath11 as large as @xmath75 demands computing time much longer than that using @xmath35 . unlike the free energy , the normalized looping time @xmath76 , the looping time in the presence of tension @xmath2 relative to the looping time in its absence , is expected to be quite insensitive to the value of @xmath11 @xcite .
for this reason we adopt @xmath35 , which was also employed in a number of studies on the looping @xcite . of @xmath48 chain obtained from the simulation ( solid line with circles ) and from the mean - field wormlike chain model
@xcite ( solid line).[fig : n12_mfwlc],width=321 ] to understand the effect of tension on the looping time , the mf - wlc provides a useful analytical model .
figure [ fig : normalized_looping_static ] shows @xmath77 as a function of @xmath2 for the chain lengths @xmath3=12 , 18 , and 24 . here
the symbols are simulation results and the dashed lines are from mfpt calculations ( eq . [ eq : mfpt ] ) . for the range of @xmath2 we study here ,
two results are in an excellent agreement .
this figure also shows that the normalized looping time increases exponentially with @xmath2 .
the reason is that , for the short chain lengths considered here , the free energy of the loop formation increases with @xmath78 , so @xmath54 increases approximately in exponential with @xmath2 .
in contrast , the theory of ref . @xcite considered that , for low tension ( @xmath79 fn ) , @xmath54 increases exponentially with @xmath80 because of gaussian force - extension relation they used . as a function of tension @xmath2 for @xmath48 , @xmath81 and @xmath82 .
the symbols are simulation data and the dashed lines are from mean first - passage time calculations using mean - field wormlike chain model @xcite.[fig : normalized_looping_static],width=359 ] the looping time in the presence of a tension @xmath2 can be written in dimensionless form @xmath83 where @xmath84 .
@xmath85 is a function of a dimensionless scaling variable @xmath86 .
it means for very large @xmath87 , a minute tension @xmath2 can dramatically change the looping time .
indeed , fig .
[ fig : normalized_looping_static ] shows that , for longer chains , the normalized looping time changes more sensitively with @xmath2 . if @xmath3 is 100 , corresponding to @xmath88 @xmath44 m , a change of the force as small as 4 fn can affect the looping time .
this is a consequence of the cooperativity of the long polymer chains arising from the chain connectivity , which is previously addressed in a study of polymer translocation through membrane @xcite .
the sensitivity of looping on tension is an emergent behavior that manifests beyond the complexity of real dna loop formation @xcite , e.g. , the details of chain conformations outside of the loop , orientation constraint , and associated proteins .
finally , we compare our result with the recent experimental data @xcite for the dna loop size @xmath89 nm . as shown by the circles in fig .
[ fig : cutoff_5 ] , the looping time rises steeply as a function of tension @xmath2 relative to the one with @xmath90 fn , which is the minimum tension used in the experiment .
we also plot the corresponding relative mfpt by a solid curve . with the cutoff distance chosen to be the bead diameter , @xmath47=5 nm
, our result is in good agreement with the experiment @xcite .
because our model does not consider many details of dna loop formation , this agreement could be somehow fortuitous but encouraging .
a theory including effectively the complexity of dna loop formation and further controlled experiments of various loop sizes are needed for quantitative comparison .
( @xmath90 fn ) as a function of tension @xmath2 with the cutoff distance @xmath91 nm for @xmath89 nm obtained by the mfpt calculation using mf - wlc model @xcite ( solid line ) .
the circles are experimental data including error bar from ref .
@xcite.[fig : cutoff_5],width=321 ] now suppose that tension temporally fluctuates due to the nonequilibrium noise inherent _ in vivo _ systems which can be generated by a variety of constituents of a cell , e.g. , protein like rna polymerase . also a dna fragment about to loop is influenced by the other part of the chain whose conformation is constantly fluctuating . as a simple example of nonequilibrium fluctuations , we consider a dichotomic noise , with which the tension @xmath92 flips between two level of the forces @xmath93 and @xmath94 with a mean flipping time @xmath95 .
the @xmath92 is a noise with zero mean and its time correlation function is @xmath96 .
we generate dichotomic noise @xmath92 using the algorithm described in @xcite . in the initial equilibration time , @xmath92 is given with either @xmath93 or @xmath94 with an equal probability 1/2 , and in a small time - step @xmath97 , @xmath92 can flip to the other value with the probability @xmath98 , which makes the mean flipping time be @xmath95 . in the presence of dichotomic tension as well as static tensions for the chain of @xmath48 .
the dichotomic force amplitude is @xmath99 fn , and the static tensions amplitudes are @xmath100 ( circle ) and @xmath101 fn ( rectangular ) .
the looping time shows a minimum at optimal flipping time @xmath102 and @xmath103 for @xmath100 and 161 fn , respectively .
( b ) the probability @xmath104 that the dichotomic force is negative ( i.e. , inward direction ) at the instant of looping as a function of @xmath95 .
[ fig : ran12_static],width=321 ] figure [ fig : ran12_static ] @xmath105 shows the normalized looping times , @xmath106 , as a functions of @xmath95 for the chain length @xmath3=12 in the presence of fluctuating tension @xmath92 added to the static tension @xmath2 .
we consider a value of dichotomic noise amplitude , @xmath99 fn and two values of static tensions amplitudes , @xmath100 ( circles ) and 161 fn ( squares ) , which are similar to those used in a recent experiment @xcite .
for very short @xmath95 , the dichotomic forces are averaged out , so the looping times converge to the values without dichotomic force .
they gradually decrease with @xmath95 until the minimum at @xmath107 and @xmath103 for @xmath100 and 161 fn , respectively . for very long @xmath95 ,
the dichotomic force rarely changes in a typical looping time , so the looping time goes to the average of the looping time with tension ( @xmath108 ) and the looping time with tension ( @xmath109 ) .
the average is dominated by the looping time with ( @xmath108 ) , so that the looping time sharply rises with @xmath95 .
related to this , we study the probability @xmath104 that the dichotomic tension @xmath92 is negative ( i.e. , is in inward direction ) at the instant of looping as a function of @xmath95 ( fig .
[ fig : ran12_static ] @xmath110 ) .
it has a maximum at @xmath111 and @xmath112 300 for @xmath100 and 161 fn , respectively , which means that , for @xmath95 near the @xmath113 , most of looping occurs when @xmath92 is in inward direction , and the looping time becomes the minimum .
in the limits of very short or very long @xmath95 , @xmath92 is positive or negative with equal probabilities at the moment of looping .
the maximum of @xmath104 is larger for larger @xmath2 . evidently the minimum looping time is closely associated with the maximum of the @xmath104 .
the minimum of @xmath54 is found to occur when the flipping time @xmath95 is comparable to the diffusion time of a brownian _ particle _ in the free energy @xmath62 , @xmath114 , where @xmath115 is a numerical value of order unity that increases with the tension @xmath2 @xcite .
this phenomenon is an extension of the resonant activation ( ra ) originally found in the single brownian particle crossing over a fluctuating barrier @xcite .
indeed , we can regard the polymer looping in the presence of dichotomic tension as the process of a brownian _ particle _ crossing over a fluctuating _ free energy _ barrier . to study how the dichotomic force affects the ra phenomena
, we consider different value of @xmath116 .
figure [ fig : n12_t ] shows the normalized looping time with @xmath117 ( circles ) , 121 fn ( squares ) _ in the absence _ of static tension ( @xmath118 ) .
the looping time has resonant minimum at the optimum flipping time @xmath119 which is smaller than the case with static tension @xmath2 . at the optimum flipping time , the looping time @xmath120 is smaller for larger @xmath116 , while , at very long @xmath95 , @xmath120 is larger for larger @xmath116 . therefore at certain @xmath95 in between , there is a crossing point . in the presence of dichotomic tension for the chain of @xmath48 .
the dichotomic force amplitudes are @xmath99 ( circle ) and @xmath121 fn ( rectangular ) .
the looping time shows a minimum at optimal flipping time @xmath119 . [
fig : n12_t],width=321 ] in contrast to the ra of single particle , the fluctuating barrier heights and thus the looping time depends sensitively on the chain length @xmath3 .
we obtain the normalized looping time @xmath122 for different chain lengths @xmath87 in the presence of dichotomic force of amplitude @xmath117 fn and static tension @xmath100 fn ( fig .
[ fig : compare ] ) . while the optimal flipping time @xmath123 increases with @xmath87 as implied by the relation @xmath124 , the minimum of @xmath125 decreases with @xmath87 .
this is because , similar to the static tension case , the fluctuating barrier height will be a function of @xmath126 , so that the longer chain tends to be more susceptible to the tension @xmath116 , and have the lower relative looping time . in the presence of dichotomic tension as well as static tensions for the chains of @xmath48 ( circle ) , 18 ( rectangular ) , and 24 ( triangle ) .
the dichotomic force amplitude is @xmath99 fn , and the static tension amplitude is @xmath100 fn .
as @xmath3 increases , the optimal flipping times @xmath113 increase and the minimum values of normalized looping time @xmath127 decrease.[fig : compare],width=321 ] a recent experiment @xcite has shown that a fluctuating tension on dna greatly reduces the looping time for small @xmath95 , in consistency with our results .
they attributed the phenomenon to an increase of the effective temperature .
this may be reasonable for the small @xmath95 , where the dichotomic noise adds to the thermal noise , but could not lead to the nonmonotonic resonant behavior emerging over the entire range of @xmath95 .
we have studied the effects of static or time - dependent tension on the semiflexible polymer looping using brownian dynamics simulation . for the case of static tension , we have found that a minute tension as small as 100 fn can dramatically change the looping time , especially for long chains .
this sensitivity is a consequence of the cooperativity of the chain arising from chain connectivity . for the case of time - dependent tension , we considered dichotomically fluctuating tension , where the tension in average changes its sign in time @xmath95 .
the looping time has a resonant minimum at an optimum flipping time @xmath123 , which is a nontrivial extension of the resonant activation ( ra ) of single brownian particle .
the effect of time - dependent tension is also more significant for longer chains .
our results are consistent with recent experiments for both static @xcite and a fluctuating tension @xcite cases .
although we neglect the details of chain conformation outside of the loop and orientation constraints , etc . , our model could be a basic step to understand the loop formation process @xmath128 where biopolymers are constantly subject to forces .
this study suggests a possibility that a biopolymer can self - organize optimally utilizing the ambient nonequilibrium fluctuations .
further experiment using dichotomically fluctuating tension with various @xmath95 is called for to establish the ra phenomena in polymer loop formation .
as the chain gets shorter , we observe small difference between the looping time with hydrodynamic interaction studied in @xcite and those without hydrodynamic interaction obtained by us using the path integral hyperdynamics method @xcite . | biopolymer looping is a dynamic process that occurs ubiquitously in cells for gene regulation , protein folding , etc . in cellular environments , biopolymers are often subject to tensions which are either static , or temporally fluctuating far away from equilibrium .
we study the dynamics of semiflexible polymer looping in the presence of such tensions by using brownian dynamics simulation combined with an analytical theory .
we show a minute tension dramatically changes the looping time , especially for long chains . considering a dichotomically flipping noise as a simple example of the nonequilibrium tension
, we find the phenomenon of resonant activation , where the looping time can be the minimum at an optimal flipping time .
we discuss our results in connection with recent experiments . |
an often neglected aspect of interaction - induced galaxy evolution is the possibility of formation of dwarf galaxies out of interaction debris .
the formation of self - gravitating objects in tidal tails was already announced by @xcite .
however some interest in these objects has started growing only in recent years .
numerical simulations of galaxy encounters support two possible scenarios of tidal dwarf formation .
the first one predicts the formation of massive bound condensations dominated by the stellar component along tidal tails and a subsequent infall of gas from the tail into the dwarf s potential well @xcite .
the second one , based on numerical simulations including dissipation for the gaseous component , proposes that the gas condenses first @xcite .
actually , both the stellar and gaseous component are observed in tidal tails and condensations along them .
the formation of condensations of gas and stars has been observed especially at the tips of tidal tails .
a spectacular example is the one of the `` antennae '' ( ngc4038/4039 , schweizer , 1978 ; mirabel , dottori & lutz 1992 ) , but other examples are shown e.g. in @xcite and @xcite .
these `` condensations '' in tidal tails of interacting galaxy systems are commonly known as tidal dwarf galaxies ( tdgs ) .
they usually share the typical properties of dwarf irregulars and blue compact dwarfs , concerning their surface brightness , blue colors , and strong bursts of star formation , but show somewhat higher metallicities confined in the narrow range 12+@xmath1(o / h ) @xmath2 8.4 - 8.6 @xcite .
this is consistent with their formation from material already processed and chemically enriched in the outer disk of the progenitor galaxies and can be a clue to disentangle dwarf galaxies born as tdgs from the population of field dwarf galaxies .
a more restrictive definition of tdgs has been proposed by @xcite , who consider a tdg `` an object which is a self - gravitating entity , formed out of the debris of a gravitational interaction '' .
this restricts such a category of objects to the ones having their own dynamics and a potential well deep enough to allow them to survive disruption for at least 1 gyr .
actually n - body simulations suggest that a substantial fraction of the material ejected during an interaction and accumulated in the tidal tails will slowly fall back into the remnant , allowing only for the outermost part ( e.g. , amounting to 20% in the simulation of ngc7252 by @xcite ) to gain enough kinetic energy to escape and evolve independently for a long time .
studies of poor galaxy groups have revealed the presence of a population of dwarf galaxies in them @xcite , sometimes considerably increasing the number of group members .
also isolated elliptical galaxies that are considered to be relics of compact groups ( cgs ) are surrounded by a population of dwarf galaxies @xcite .
a possible explanation is that part of the dwarf galaxies observed in groups is generated in the groups themselves as a by - product of galaxy interactions in these dense environments .
a search for tdgs in cgs led hunsberger , charlton , & zaritsky ( 1996 ) to suppose that up to half of the total dwarf population in cgs may be the product of interaction between galaxies .
however the study of @xcite suffers from the lack of kinematic information , which makes it difficult to distinguish `` real '' tdgs from non - self - gravitating condensations in tidal debris .
more restrictive identification criteria applied by @xcite led to the detection of a much lower number of tdgs in cgs .
however the study with most complete information on compact groups tdgs until present is the one by @xcite , which relies on both photometry and kinematics , but is limited to hcg 92 .
they identified in hcg 92 seven candidate tdgs , at least two of which are thought to be able to survive fall - backs into or disruption by the parent galaxy . in this paper
we focus our attention on the brightness enhancements and/or blue condensations we identified as possible tdg candidates during a detailed analysis of the compact galaxy group cg j1720 - 67.8 ( weinberger , temporin , & kerber 1999 ( paper i ) ; temporin et al .
2003 hereafter paper ii ) .
in particular we perform optical and near - infrared aperture photometry and investigate the physical properties of the identified knots and candidate tdgs through photoionization models calculated with the code cloudy 90.04 @xcite .
considerations about the tidal - tail kinematics are expressed on the bases of spectroscopic data obtained for discrete regions along the tail in addition to integral field spectra sampling its northern part .
the optical colors of the candidate tdgs are compared with the grids of spectrophotometric evolutionary synthesis models by weilbacher et al .
most of the data , on which the present study is based , have already been presented in paper ii , therefore we only briefly list here the observational material at our disposal .
broad band @xmath3 , @xmath4 , and @xmath5 images ( 900 s , 600 s , and 3@xmath6600 s exposures , respectively ) have been obtained in june 1999 at the eso 3.6 m telescope in la silla .
a number of broad band images was obtained in the near - infrared ( nir ) regime in june 2000 at the 1 m swope telescope in las campanas .
the total integration times were 1800 s in the @xmath7 band , 2000 s in @xmath8 and 2500 s in @xmath9-short ( @xmath10 ) .
details about the reduction steps and the photometric calibration are given in paper ii .
the nir photometric system is described in detail in @xcite . for clarity
we show in fig .
[ cmap ] the identification of the candidate tdgs and knots onto the contour maps of the @xmath3 and @xmath5-band images of cg j1720 - 67.8 .
spectra of the candidate tdgs were obtained at the 2.5 m du pont telescope of las campanas with the modular spectrograph and at the eso 3.6 m telescope in la silla , with efosc2 , in multi - object mode ( mos ) .
they all show emission lines with radial velocities in agreement with those of the three main galaxies and -region like diagnostic ratios ( see table 6 and fig . 10 in paper ii ) , although high values of the [ ] /h@xmath11 ratio suggest that shock heating might also be important .
fluxes , the extinction corrected intensities and the estimates of the internal extinction obtained from the balmer decrement were reported in table 3 of paper ii ( but see also 4 below ) .
the highest extinction value was found for object no .
3 , the candidate tdg at the northern tip of the group s tidal arc ( e ( ) = 0.78 @xmath12 0.42 ) . additionally ,
we make use of velocity information about object 3 + 9 obtained from recent observations ( june 2002 ) at the 3.9 m anglo - australian telescope with the integral field unit spiral ( segmented pupil / image reformatting array lens ) and a 600 g / mm grating in the wavelength range @xmath13 6500 - 7120 in combination with the 2048@xmath64096 13.5 @xmath14-sized pixels eev chip .
from an examination of the contour maps in fig . [ cmap ] , we note that objects 3 , 7 , 10 , and 12 along the tidal arc selected for showing peaks of intensity over the tidal feature where they are embedded show some sort of boundaries .
9 appears like a secondary peak inside the boundaries of no . 3 although this could be an effect of insufficient spatial resolution therefore we consider it as part of object 3 and we will refer to it as object 3 + 9 .
8 appears like a secondary peak adjacent to no . 7 and , although it does not show a clear brightness enhancement with respect to the tidal tail , it emerges as a sort of separated knot in the @xmath5-band image after the application of an adaptive laplacian filter ( fig .
[ laplace ] ) , which has the effect of suppressing the noise and enhancing faint structures @xcite .
5 emerges more clearly as a knot in fig .
[ laplace ] , while it appears as a diffuse structure connected to galaxy 4 in the contour maps of fig .
this object could be a star - forming small tidal tail or perhaps a vestige of a spiral arm .
its color does not differ from that of galaxy 4 , as can be seen in the color map of fig .
[ tdg_col ] ( obtained as described in paper ii ) .
finally , no .
11 appears as a knot inside a ring - like ( fig .
[ laplace ] ) diffuse structure without any obvious connection with tidal tails . in fig .
[ tdg_col ] all the objects identified along the tidal arc appear bluer than the underlying tidal tail and the adjacent galaxy 4 . since the features under investigation have irregular elongated shapes , we did not apply traditional circular aperture photometry , which would include too much flux from the surrounding parts of the tidal tail .
instead , we adopted a procedure analogous to the one applied by @xcite and measured total magnitudes inside polygonal apertures .
the apertures were defined on the contour plot of the calibrated @xmath3-band image following the faintest contour , which still allowed to separate each clump from the underlying tidal tail . therefore the surface brightness level at which the aperture was defined was different for each object and the values are listed in table [ size ] , together with peak surface brightnesses . the same surface brightness levels were taken as reference to measure the size of the clumps . due to the elongated shape of the objects , instead of a typical radius we list in table [ size ] their minor and major dimensions in kpc , adopting h@xmath15 = 75 km s@xmath16 mpc@xmath16 .
the same polygonal apertures were applied to the aligned images in the @xmath4 , @xmath5 , @xmath7 , @xmath8 , and @xmath10 bands .
the nir images were first re - binned to match the pixel scale of the optical ones .
we chose not to subtract the tidal tail contribution from the objects fluxes , but only the sky - background evaluated in several positions as near as possible to the objects of interest .
this choice will allow us to compare our photometric data with the evolutionary synthesis models by @xcite , which take into account also the old stellar population present in the underlying tidal tail .
the total magnitudes and average surface brightnesses inside the polygonal apertures and the optical and nir colors of the clumps are listed in table [ mag ] .
all the values are corrected for galactic extinction ( e ( ) = 0.088 ) .
we find optical colors ( @xmath2 0.65 - 0.85 ) consistent with those of hcg 92 s tdg candidates @xcite . due to the difficulty to define a clear boundary for the knot no .
5 we give only its surface brightness and colors , but not its total magnitude . for comparison purposes we list
also the average surface brightness and colors of two portions of the tail measured between the clumps .
the brightness of the tail increases slightly from north to south .
typical photometric errors in total optical magnitudes range from @xmath2 0.002 to 0.01 mag , therefore the calibration error ( @xmath2 0.05 , see paper ii ) is dominating . in the nir photometric errors
are typically @xmath2 0.1 - 0.2 mag but reach 0.6 mag for the faintest objects , nos . 8 and 11 , which could not be measured in the k@xmath17 frame .
although for the candidate tdgs nos . 3 + 9 and 7 we could estimate a significant amount of internal extinction from the balmer decrement ( see paper ii , table 3 ) , the error in this extinction measurement is high , therefore we preferred not to apply any internal extinction correction to the magnitudes of these objects .
we obtained the following luminosities of the candidate tdgs ( including the light from the underlying tidal tail ) : m@xmath18(3 + 9 ) = @xmath1917.46 , m@xmath18(7 ) = @xmath1917.06 , m@xmath18(8 ) = @xmath1913.93 , m@xmath18(10 ) = @xmath1915.86 ,
m@xmath18(12 ) = @xmath1915.35 , and m@xmath18(11 ) = @xmath1914.39 .
we note that the above luminosities are comparable or higher than those found by @xcite in the candidate tdgs sample of stephan s quintet and those of tdg candidates in interacting systems identified by @xcite .
the two brightest tdg candidates have luminosities higher than the brightest regions measured in sc galaxies by @xcite and more typical of giant extragalactic regions @xcite .
however , the comparison with region properties is made difficult by the different methods of measurement used in the present work and in the literature .
in particular , we need to emphasize that , unlike most published photometric measurements of extragalactic regions , we have used interactively defined polygonal apertures and we did not subtract the flux contributed by the tidal tail , where the knots are embedded , as explained above .
we will further discuss this point later , in 5 .
a number of additional irregularly shaped , faint knots have been identified in the close environment of cg j1720 - 67.8 , and in particular inside its halo .
their nature is unclear and we have not yet obtained spectral information on them .
however , at least for part of them , we could measure aperture magnitudes in the optical images . in this case
we used circular apertures and evaluated the sky background in an annulus around each object .
these knots are labeled with small - case letters onto the laplacian filtered r - band image in fig .
[ laplace ] .
their optical magnitudes and colors are listed in table [ knots ] .
some of these objects might belong to a population of dwarf galaxies associated with the compact group , however deeper data and spectral information are necessary to understand it . at present , the group s members among these can not be distinguished from background objects due to the lack of radial velocity information , although their colors could be consistent with background galaxies ( see 6 ) .
the spectra of the candidate tdgs ( nos . 3 , 7 , 8 , 9 , 10 ) and of additional knots of cg j1720 - 67.8 ( nos . 5 , 11 ) all show a blue continuum ( fig .
[ tdgspec ] ) , and some of them exhibit remarkable balmer absorption lines in addition to the emission lines typical of regions . therefore all their spectral properties agree to indicate a dominating young stellar population and an activity of current / recent star formation .
as already noticed in paper ii , veilleux & osterbrock s ( 1987 ) ( vo-)diagnostic diagrams applied to these objects confirm that their gas is photoionized by thermal sources although a certain degree of ionization by shock - heating can not be excluded .
also the spectrum of the portion no .
6 of the arc shows balmer absorption lines and weak emission lines ( h@xmath11 , [ ] , and [ ] ) , indicating that a low level of star formation is present in the tidal tail as well , together with an older stellar population .
given the thermal nature of the ionizing source , the h@xmath11 emission - line luminosity , after correction for internal extinction , could be used to evaluate the star formation rate ( sfr ) of these objects . in the spectra nos .
5 and 6 , only a few emission lines were detected , i.e. the blend h@xmath11- [ ] , the [ ] doublet , and for object 5 [ ] @xmath13 3727 .
these were insufficient to locate the regions in the vo - diagnostic diagrams and the lack of a measurable h@xmath20 prevented the estimate of the internal extinction . in this case
we used the uncorrected h@xmath11 luminosities for the sfr estimate .
the star formation rates calculated following @xcite are listed in table [ sfr ] .
they range from 0.004 to 0.82 m@xmath21 yr@xmath16 , with star formation rate densities ( sfrd ) of 0.05 to 14.15 m@xmath21 yr@xmath16 pc@xmath22 , the highest value being found for the candidate tdg no .
3 and the lowest for knot no .
these sfr values are comparable to those of the tdgs identified in hcg 92 @xcite .
the observed emission - line ratios were used to infer the physical and chemical properties of the ionized gas .
tdgs are expected to have higher metallicities than typical field dwarf galaxies , since they form out of already enriched material ejected by the parent galaxies . since in none of the candidate
tdgs the [ ] @xmath13 4363 emission - line was detected , a direct measurement of the electronic temperature of the emitting gas necessary for the calculation of the metal abundances of the gaseous component was not possible . therefore , in analogy with the method adopted for the three main galaxies of cg j1720 - 67.8 ( paper ii ) , we obtained a first estimate of the metal abundances of the candidate tdgs by comparing their emission line ratios with the empirical diagrams of @xcite and @xcite . from the comparison with the grid of models of @xcite in the plane [ ] / [ ] _ vs _ r@xmath23 ( r@xmath23 = @xmath1 [ ( [ ] @xmath13 3727 + [ ] @xmath13 4959,5007)/h@xmath24 ) we found that all the candidate tdgs have an ionization parameter in the range @xmath193.5 @xmath25 @xmath1u @xmath25 @xmath193 .
the metal abundances , represented by the o / h ratio , result in the range
@xmath193.9 @xmath25 @xmath1(o / h ) @xmath25 @xmath193.5 for objects nos .
7 , 8 , and 9 and have the two possible values @xmath1(o / h ) @xmath264.1 or @xmath1(o / h ) @xmath27 @xmath193.2 for no .
object no .
3 could not be compared with these model grids , because the [ ] @xmath13 3727 was not detected in its spectrum .
all the candidate tdgs could be compared with the monotonic relation 12+@xmath1(o / h ) _ vs _ @xmath1 ( [ ] @xmath13 6583/h@xmath11 ) @xcite . by referring to the model track with @xmath1u = @xmath193.0 (
* fig . 1 )
, we found metallicity values in rough agreement with the lowest values given above i.e . with the lower branch of the r@xmath23 - @xmath1(o / h ) relation in @xcite namely @xmath1(o / h ) @xmath2 @xmath193.7 for objects 3 , 8 , and 9 , @xmath1(o / h ) @xmath2 @xmath193.8 for object 7 , and @xmath1(o / h ) @xmath2 @xmath193.9 for object 10 , or , in other words , z @xmath2 0.2 - 0.3 z@xmath2810@xmath29 , n / h = 9.33@xmath610@xmath30 , and s / h = 1.62@xmath610@xmath30 . ] .
these estimates were further refined by fitting the observed emission line ratios with photoionization models by use of the code cloudy 90.04 @xcite . during the fits
the hydrogen density n@xmath31 and the ionization parameter u were let free to vary , while as a typical temperature of the thermal ionizing continuum we assumed t = 4@xmath610@xmath32 k. several models were tried with slightly different metallicity values , until the reproduction of the observed line ratio was satisfactory ( we chose the models with lowest @xmath33 ) .
the final photoionization model parameters are given in table [ photoion ] , while in table [ obs - mod ] the observed and modeled emission - line intensities relative to h@xmath20 are compared .
however , also with this procedure the metal abundances of objects 7 , 8 , and 9 could not be uniquely determined .
actually photoionization models with physical parameters similar to those given in table [ photoion ] but somewhat lower metal abundances would offer a comparably good , or even better fit to the observed line ratios .
the parameters of these alternative models are shown in table [ alt - photoion ] . to summarize ,
our candidate tdgs show recent / present star formation activity , and their gaseous component exhibits an ionization parameter u in the range @xmath2 4@xmath610@xmath29 - 7@xmath610@xmath29 , a hydrogen density n@xmath31 in the range @xmath2 1 - 2@xmath610@xmath34 @xmath35 , and metal abundances z @xmath2 0.1 z@xmath28 for object 10 and z @xmath2 0.3 z@xmath28 for objects 3 , 7 , 8 , and 9 , although somewhat lower abundances for the last three might be possible .
especially object 7 could have z @xmath2 0.1 z@xmath28 , but with an overabundance of oxygen and sulphur , which appear to be roughly a third of the solar value . as a comparison we recall here the metal abundances obtained for the three main galaxies of cg j1720 - 67.8 ( paper ii ) : z @xmath2 0.2 z@xmath28 for galaxy 1 , z @xmath25 0.5 z@xmath28 for galaxy 2 , and z @xmath2 0.1 z@xmath28 for galaxy 4 . during june 2002 observations at the anglo - australian telescope equipped with spiral we obtained integral field spectra of the upper part of the tidal arc , where the candidate tdg 3 + 9
is located .
the object was covered with an array of 14@xmath615 microlenses giving a field of view of [email protected]@[email protected] with a spatial scale of 0.7 arcsec per microlens per pixel .
the position of the array onto the r - band image of the group is shown in fig .
[ array ] ( left ) .
the observations were carried out with the nod and shuffle technique in order to provide an optimal subtraction of the sky - background .
the total exposure time was 1 hour , half of which was spent on target and the other half on sky .
the wavelength range was centered on the redshifted h@xmath11 line .
the 600r grating combined with the eev ccd - chip gave a dispersion of 0.612 pixel@xmath16 .
the average spectral resolution was in the range 1.8 - 2.3 with the lowest values in the central part of the array and highest values at the borders .
the basic reduction steps involving the tracing of the fibres by use of a flat - field frame , bias subtraction , flat - fielding , wavelength calibration , sky subtraction , and combination of subsequent exposures were done with the dedicated software 2dfdr , version 2.3 , provided by the anglo - australian observatory .
this software allows also the image reconstruction at a specified wavelength or in a specified wavelength range and was used to reconstruct an h@xmath37continuum image of the field in the range 6800 6900 for object identification purposes .
this image was resampled onto a grid of 56@xmath660 pixels ( i.e. 4 times larger than the original ) and smoothed with a 4@xmath64 pixel box for viewing purposes ( fig .
[ array ] , right ) .
further analysis of the data was carried out within iraf .
specifically , for each spectrum where the h@xmath11 emission line was detected we obtained a radial velocity measurement ( through gaussian fit of the emission - line ) in order to reconstruct the velocity field of the source .
the average calibration error , evaluated from position measurements of oh night - sky lines @xcite , is 2.9 km s@xmath16 . no systematic trends of the error were detected across the 14@xmath615 array of spectra .
the result is shown in fig .
[ vfield ] . on the left - hand panel
the original reconstructed continuum image ( without resampling ) shows object 3 + 9 in the center and portions of galaxies 2 and 4 at the right edge . on the central image the velocity field , with indicated the maximum and minimum measured radial velocities of the object 3 + 9 ,
shows a velocity gradient in direction ne - sw ( the north - eastern part being the receding one ) across the object , with a maximum velocity difference of @xmath2 200 km s@xmath16 over an extent of @xmath2 5 kpc .
the velocity field is shown also after magnification and resampling ( fig .
[ vfield ] , right - hand panel ) and can be compared with fig .
[ array ] .
there might be several possible explanations for such a gradient ( see e.g. * ? ? ?
* ) , like streaming motion along the tidal tail , projection effects , or even rotational motion .
we lack detailed information on the radial velocity gradient along the tidal tail , apart from the discrete measurements in correspondence of objects 7 , 8 , and 10 that indicate that the southern part of the tail is approaching us ( radial velocity @xmath2 13200 km s@xmath16 ) and showing a velocity difference @xmath25 400 km s@xmath16 between the tips of the tail .
however , the tdg candidate 3 + 9 is located at the base of the tail , where tidal streaming and geometric effects are most extreme ( see e.g. the case of the merger remnant ngc 7252 * ? ? ? * their fig . 1 ) .
such effects are likely to be responsible for the observed gradient .
even though , the presence of a rotational motion can not be ruled out .
if the velocity gradient of object 3 + 9 was actually entirely caused by rotation , it would imply a dynamical mass m@xmath38 6@xmath610@xmath39 m@xmath21 , comparable to the mass estimated for the group s galaxy no . 1 @xcite .
this mass , combined with the @xmath3-band luminosity given in 3 , would yield a mass - to - light ratio m@xmath40/l@xmath18 @xmath2 4 .
however , also in case of self - gravitation , the object is unlikely to be in a relaxed state , therefore the application of the virial theorem could lead to a considerable overestimate of its mass .
in the attempt to understand the real nature of the knots in cg j1720 - 67.8 and to investigate alternative possibilities to the tdg candidate hypothesis , we compare their optical properties with those of normal and giant extragalactic regions ( gehr ) present in the literature . as already stated in
3 , some difficulties in such a comparison arise as a consequence of different methods of measurement .
in particular the background subtraction is critical when comparing photometric properties .
since most of the published optical photometry of regions is obtained by subtracting the contribution of the underlying galaxy components ( i.e. disk , spiral arms ) , we have estimated the contribution of the tidal tail to the light of our tdg candidates in the @xmath41 bands and subtracted it from the measured fluxes of the knots .
the new set of background - subtracted magnitudes and colors is shown in table [ no_bkg ] .
the background of each knot was estimated by averaging the mean value inside a number of 10@xmath610 pixel boxes selected all around the polygonal aperture used for the photometry , taking care to avoid the inclusion of adjacent knots .
we stress that the background around the knots has strong gradients , especially at the tips of the tidal tail . therefore , although an as large as possible number of boxes was used for its estimate , the final values we adopted have uncertainties in the range @xmath2 12 40 per cent .
we found that the tidal tail contributes to the flux in the clumps by @xmath2 45 to 60 per cent in @xmath3 , 55 to 68 per cent in @xmath4 , and 52 to 70 per cent in @xmath5 .
the highest percentages are found in the central parts of the arc , while the lowest at the tips of the arc . the maximum contribution from the tail
is observed in the @xmath4 band and the minimum in the @xmath3 band .
such a differential contribution determines the difference between the colors given in table [ mag ] and those in table [ no_bkg ] .
the and colors given in table [ no_bkg ] can be compared with those measured by @xcite for gehr in a sample of nearby spiral and irregular galaxies .
median colors of gehrs and gehr groups in mayya s sample are = 0.21 , = 0.48 and = 0.30 , = 0.44 , respectively , which are consistent with the values we found for our clumps . actually in (
* table 3 ) there are several examples of gehrs with colors very similar to those in our table [ no_bkg ] . however , their h@xmath11 equivalent widths ( w@xmath42 ) are much higher than the ones we measured in the spectra of our clumps ( table [ sfr ] ) . furthermore the sizes of gehrs are @xmath25 1.5 kpc , while the clumps in cg j1720 - 67.8 are considerably more extended ( see table [ size ] ) .
the properties of the first ranked regions in a sample of nearby spiral and irregular galaxies were studied by @xcite through h@xmath11 photometry .
the brightest ( `` giant '' or `` supergiant '' ) regions in kennicutt s sample were found almost exclusively in late - type normal galaxies or in peculiar galaxies , and the question `` whether they simply represent the high - luminosity tail of the normal region luminosity function or are a physically distinct class of objects formed under special conditions '' appears to remain unanswered .
h@xmath11 luminosities of these regions are in the range 10@xmath4310@xmath44 ergs s@xmath16 and the masses of the embedded star clusters , calculated assuming a salpeter imf and including only stars with masses 10100 m@xmath21 , were expressed as m@xmath45 = 3600 l@xmath46 m@xmath21 ( where l@xmath46 is the h@xmath11 luminosity in units of 10@xmath43 ergs s@xmath16 ) , while the ionized gas mass was found to scale roughly linearly with h@xmath11 luminosity ( at least for the largest regions ) , although the actual value depends on the density @xmath47 according to the relation m@xmath48 = 2.3@xmath610@xmath49 l@xmath46/@xmath47 . by applying the above relations to our tdg candidates using their
extinction - corrected , spectroscopically determined , h@xmath11 luminosities given in table [ sfr ] and the hydrogen densities given in table [ photoion] we find masses of the embedded clusters ranging from @xmath2 4@xmath610@xmath32 m@xmath21 ( object 10 ) to @xmath2 5@xmath610@xmath50 m@xmath21 in the brightest knots at the tips of the tail and ionized gas masses ranging from @xmath51 ( object 10 ) to @xmath2 4@xmath52 m@xmath21 ( object 3 + 9 ) .
these are only rough estimates and the actual mass values could be larger , since we have used here h@xmath11 luminosities from long - slit spectra , which covered big fractions , but not the full extent of the objects . for comparison ,
the h@xmath11 luminosity of the position 6 along the tidal tail would yield m@xmath45 @xmath2 10@xmath32 m@xmath21 and m@xmath48 of order 10@xmath50 m@xmath21 ( in this case the density could not be determined and the ionized gas mass is estimated through @xcite empirical relation ) .
the equivalent ionizing luminosity of object 3 + 9 reaches 10@xmath53 photons s@xmath16 ( after correction for internal extinction ) .
the properties of the knots along ( and especially at the tips of ) the tidal arc are comparable to those of the giant and supergiant extragalactic regions of kennicutt s sample . we note that h@xmath11 luminosities given in @xcite are not corrected for internal extinction , unlike those in our table [ sfr ] . however , also before the extinction correction our tdg candidates have @xmath1 h@xmath54 ergs s@xmath16 . a comparison of h@xmath11 equivalent widths given in ( * ? ? ?
* their fig . 6 ) with those in table [ sfr ] shows that our tdg candidates are located at the lower limit of the range found for disk regions . furthermore the value measured in the spectrum no .
9 is extremely low and similar to the values measured in spectra nos . 5 and 6 .
this could indicate a non - negligible contribution to the stellar continuum from an underlying evolved stellar population .
the lack of detailed kinematic information ( except for the low - resolution velocity field of the ionized gas in object 3 + 9 ) prevent us from establishing which , if any , of the candidate tdgs we have identified are actually self - gravitating objects i.e .
tdgs in the most rigorous meaning of the term or simply clumps of gas and stars , which are still bound to the tidal tail where they have formed and might eventually evolve into tdgs .
even though , evolutionary synthesis models have proved useful to discriminate between candidate tdgs or tdg progenitors and background objects , and also to estimate the ratio of old to young stellar mass in them @xcite .
therefore we compare here the observational properties of our candidate tdgs with the grids of evolutionary synthesis models published in @xcite .
they have obtained models for two different metallicities , in the range expected for tdgs , i.e. z@xmath55 @xmath2 z@xmath28/18 and z@xmath56 @xmath2 z@xmath28/2.3 , by use of an evolutionary code based on the work of @xcite .
their models assume for tdgs the same undisturbed evolutionary history of the parent galaxies until the onset of the interaction - induced starburst . a @xcite initial mass function is assumed . in the burst the sfr
is set to a maximum value and decreases exponentially with a timescale @xmath57 .
the grids of evolutionary models obtained by @xcite for different burst strengths are represented by counter - clockwise loops in the @xmath58 two - color diagram .
since we have found that our candidate tdgs have metallicities in the range 0.1 z@xmath28 @xmath25 z @xmath25 0.3 z@xmath28 , and in particular oxygen abundances o / h @xmath59 0.35 ( o / h)@xmath28 , we have compared their optical colors with the z@xmath56 models from @xcite .
the observed colors of our candidate tdgs are marked in fig .
[ grid ] onto the two - color diagram adapted from fig .
1c of @xcite .
values corrected only for foreground galactic extinction are marked with crosses .
artificial data - points ( labeled `` a '' , `` b '' , `` c '' , `` d '' , and `` e '' ) from @xcite are plotted , as well ( full circles ) .
in particular point `` b '' corresponds to a burst strength @xmath60 0.1 and a burst age @xmath2 20 myr , point `` c '' indicates a medium to strong burst of much higher age ( @xmath2 80 myr ) , while point `` d '' corresponds to a burst age of @xmath2 7 myr or less , depending on the metallicity .
most of our data - points fall into a region of the two - color diagram where tracks of different models overlap , namely the part of the tracks with burst ages of only a few myrs .
this would indicate that such objects have just started their bursts of star formation .
however , the application of a correction for internal extinction would move the points in the diagram toward older burst ages , reaching the region occupied by the artificial data - point `` b '' , or it would imply stronger bursts
. older burst ages would be in agreement with the observed h@xmath11 equivalent widths ( table [ sfr ] ) , which are not as high as expected and observed in case of very young ( age @xmath61 2 - 3 myr ) regions ( see e.g. * ? ? ?
* ) . however , the balmer emission - line equivalent width could be lowered as an effect of the presence of an underlying population of stars older than those responsible for the ionization and contributing to the stellar continuum .
internal extinction affecting the emission lines but not the continuum could also play a role in lowering the equivalent width ( e.g. * ? ? ?
although it is not possible to establish the appropriate duration and strength of the bursts due to the similarity of models with @xmath62 in the range 0.05 0.18 and @xmath63 @xmath2 5@xmath610@xmath50
1@xmath610@xmath49 yr , object 8 appears consistent with a weaker burst with respect to the other tdg candidates .
object 11 , a condensation in a ring - like structure on the west side of the group , falls near the artificial datapoint `` c '' , suggesting a much higher burst age .
this could be a condensation formed in an already fading tidal tail . supposing that these tdg candidates will evolve following the tracks in fig
. [ grid ] , they will become still bluer during the next few myr , then they will become progressively redder , reaching after @xmath2 1 gyr the pre - burst colors .
unless new bursts of star formation are triggered , the objects will progressively fade during their evolution .
their fading can be quantified by comparison with the luminosity evolution modeled by @xcite , which shows a rapid luminosity decrease in the first 50 myrs after the maximum of the burst . in the hypothesis
that they have reached the maximum @xmath3 luminosity at the present epoch , they will fade by about 3.5 mag within 2 gyr .
therefore the brightest clump , no .
3 + 9 , would reach m@xmath18 @xmath2 @xmath1914 and it would be still observable , with an apparent magnitude @xmath3 @xmath2 22.3 . in the same time , the faintest object , no .
8 , would fade to @xmath3 @xmath2 25.9 mag and would be visible only in very deep exposures .
object , no .
11 , according to its estimated burst age of @xmath2 80 myr , should have already faded by @xmath2 2.2 mag with respect to the luminosity peak reached during the burst phase and its @xmath3 luminosity is expected to decrease by another 1.3 mag in the next 1.9 gyr , down to @xmath3 @xmath2 23.2 mag .
the observed colors of the additional knots ` a ' , ` b ' , ` c ' , ` d ' , and ` e ' ( fig . [ laplace ] ) identified in the group s halo would fall far away from the tdg model tracks , therefore , according to @xcite , they could be regarded as background objects .
we have presented photometric and spectroscopic observational data of a number of knots and candidate tdgs previously identified ( paper i and paper ii ) at the ends of and along the tidal features and in the halo of the ultracompact group cg j1720 - 67.8 .
although we are not able to establish whether the clumps along the tidal tail are already self - gravitating objects because of the lack of kinematic information with the possible exception of object 3 + 9 whose velocity field shows a regular velocity gradient we have found that their properties are consistent with those expected for tdgs or tdg progenitors .
namely , they show blue colors , recent and/or present - day star formation activity , and relatively high oxygen abundances with respect to field dwarf galaxies . the same properties would also be consistent with star formation within the tail .
indeed we have detected a low level of star formation in a region of the tail where no visible knots are present ( position no .
6 ) and the moderately blue colors of the tail suggest that some degree of star formation is present all along it .
however the star - forming clumps we have identified as candidate tdgs clearly appear as substructures and condensations within the tail and have colors significantly bluer than the surrounding regions .
moreover their luminosities are considerably higher than those typical of normal regions in spiral galaxies and more typical of gehrs .
optical colors as well , once a correction for contamination from the tidal tail material has been applied , are in agreement with those of gehrs @xcite .
actually , optical colors are not sufficient to discriminate between ( giant ) regions and tdgs .
the sizes of most of our candidate tdgs are considerably larger than those of the most extended gehrs , which reach at most @xmath2 1.5 kpc ( e.g. * ? ? ?
instead they have dimensions typical of dwarf galaxies ( see e.g. * ? ? ?
* ) and comparable to those of the candidate tdgs in stephan s quintet @xcite .
only object no .
8 , judging from its size , might be interpreted as a giant region . at the seeing of our observations ,
the spatial resolution element is @xmath25 1 kpc , which means that a giant region would barely be resolved , while the clumps we observe appear clearly extended .
some of them e.g .
nos . 10 and 12 that appear less concentrated than the objects at the tips of the tail
could be complexes of regions , which appear as an individual clump due to insufficient spatial resolution .
however , if such complexes form in the tail , their potential well might result deep enough for them to become bound objects that eventually can evolve as dwarf galaxies , provided their mass is sufficient .
in such a case they could be considered as tdgs in the process of formation .
the presence itself of structures in the tail suggests that self - gravitation might play some role .
whether or not the clumps of stars and ionized gas are to become bound objects and even accrete more matter from their surroundings , depends on their having enough mass to survive fall - back into the parent galaxies and tidal disruption @xcite .
the measured optical colors of our candidate tdgs are consistent with evolutionary synthesis tracks calculated for tdgs in the expected metallicity range @xcite .
the positions in the two - color diagrams , when compared with evolutionary tracks , suggests that the objects identified along the group s tidal arc have burst ages of a few myr , while the blue knot no .
11 embedded in a ring - like structure in the group s halo is consistent with a much older burst age ( @xmath2 80 myr ) .
however , the location of the objects in the two - color diagram might be strongly affected by internal extinction , which has not been taken into account due to the high uncertainty in its determination .
the effect could be relevant especially for the objects at the tips of the tail , where indications of higher extinction are found .
the application of an internal extinction correction would move the observational points toward higher burst ages ( up to @xmath2 20 myr ) in the model tracks , thus offering a possible explanation to the relatively low observed h@xmath11 equivalent widths .
if a luminosity evolution analogous to the one modeled by @xcite is assumed , the identified tdg candidates should fade to values of the @xmath3 luminosity still detectable after 2 gyr of evolution , provided they survive as bound objects .
a comparison of the metallicities of the candidate tdgs with those of the three main galaxies in the group does not easily allow to understand which one is the parent galaxy .
actually object no .
10 is the one with lowest metallicity , in agreement with the metal abundances of galaxy 4 ( paper ii ) . for the other objects
the metallicity value is uncertain , although it seems established that their oxygen abundance is approximately a third of the solar one , therefore higher than that of galaxy 4 .
their parent galaxy might be no .
1 , which exhibits somewhat higher abundances . a quantitative estimate of the possibility for these objects to survive as individual galaxies against internal motion and tidal forces exerted by
the parent galaxies would require an estimate of their virial and tidal masses @xcite .
the first can not be derived with the presently available data , the latter would require knowledge of the mass interior to their orbit .
since this mass is not known , we limit ourselves to a few qualitative considerations .
we have found that the object composed of no .
3 and no . 9 ( considered together ) is the brightest of the proposed candidate tdgs .
it s velocity gradient , if actually due to rotation , would imply a mass comparable to the one estimated for galaxy 1 @xcite .
if this was the case , its mass could be sufficient for it to escape tidal disruption , despite its unfavorable position in the tidal feature , very nearby to the parent galaxies .
however , the observed velocity gradient could be an effect of streaming motion in combination with projection effects , which can be extreme at the base of the tidal tail .
therefore , no conclusive argument can be given at present .
object no .
7 , although less bright and therefore presumably less massive , seems to be in a more favorable location to survive tidal disruption . in principle
nir photometry , when combined with optical photometry and equivalent width measurements , should help at least to understand whether the modest luminosity enhancements of our tdg candidates with respect to the surrounding regions of the tidal tail indicate an actual increase of the underlying mass density as expected in case of distinct dynamical entities and not only an increase of the luminosity as a consequence of the burst of star formation .
however the question can not be addressed in a straightforward manner .
not only the optical ( especially @xmath3 ) luminosity is influenced by the recent star formation activity with consequent changes in the m / l ratio , but also the nir luminosity is affected , although in smaller measure .
therefore accurate evolutionary synthesis models , with adequate time resolution , are necessary for a meaningful interpretation of the data .
additionally , the uncertainty in the burst strength and age determination , partly caused by uncertainty in the internal extinction determination as explained above , further complicates the matter .
we will attempt to address this question in a future work . at present , on the base of the analysis presented here , we can only conclude that the two objects at the tips of the tidal tail ( 3 + 9 and 7 ) appear to be the most promising tdg candidates of the galaxy group .
we thank the anonymous referee for useful comments and suggestions , which helped to considerably improve this paper .
st is grateful to p. m. weilbacher for fruitful discussions .
rw is grateful to the austrian bundesministerium fr wissenschaft und verkehr " for travel support .
gg thanks `` proyecto dipuc 2001/14-e '' .
st acknowledges support by the austrian science fund ( fwf ) under project
p15065 .
cccc 3@xmath649 & 22.35 & 21.80 & [email protected] + 5 & 22.47 & & [email protected] + 7 & 22.24 & 21.780 & [email protected] + 8 & 22.48 & 22.35 &
[email protected] + 10 & 22.25 & 22.03 & [email protected] + 11 & 23.54 & & [email protected] + 12 & 22.26 & 22.14 & [email protected] + ccccccccccccc 3@xmath649 & 18.72 & 0.44 & 0.41 & 2.16 & 0.66 & 0.25 & 22.07 & 21.63 & 21.22 & 20.13 & 19.47 & 19.22 + 5 & & 0.58 & 0.43 & 2.22 & 0.74 & 0.21 & 22.40 & 21.81 & 21.38 & 20.33 & 19.59 & 19.38 + 7 & 19.22 & 0.31 & 0.35 & 1.87 & 0.68 & 0.05 & 22.04 & 21.74 & 21.39 & 20.55 & 19.86 & 19.81 + 8 & 22.35 & 0.31 & 0.40 & 2.12 & 0.88 & & 22.44 & 22.14 & 21.73 & 20.90 & 20.02 & + 10 & 20.42 & 0.41 & 0.43 & 2.01 & 0.73 & 0.25 & 22.17 & 21.76 & 21.32 & 20.49 & 19.75 & 19.50 + 11 & 21.89 & 0.19 & 0.41 & 1.87 & 0.90 & & 23.36 & 23.17 & 22.75 & 22.20 & 21.30 & + 12 & 20.93 & 0.43 & 0.39 & 1.99 & 0.55 & 0.54 & 22.22 & 21.79 & 21.40 & 20.41 & 19.80 & 19.26 + arc@xmath55 & & 0.56 & 0.40 & 1.74 & 0.73 & 0.29 & 22.47 & 21.91 & 21.51 & 20.42 & 19.69 & 19.40 + arc@xmath65 & & 0.50 & 0.33 & 1.28 & 0.62 & 0.55 & 22.34 & 21.81 & 21.51 & 20.67 & 20.05 & 19.50 + cccccc a & 24.33 & 22.03 & 21.22 & 2.30 & 0.82 + b & 20.06 & 19.32 & 18.85 & 0.74 & 0.46 + c & 21.89 & 20.75 & 20.21 & 1.13 & 0.54 + d & 21.64 & 20.59 & 20.09 & 1.05 & 0.50 + e & 21.49 & 20.48 & 19.88 & 1.01 & 0.60 + [email protected]@[email protected]@.l 3 & 11&56 & 198 & 8&44 & 1688 & 0&82 & 14&15 + 5 & 0&06 & 28 & 0&05 & 9 & 0&004 & 0&05 + 6 & 0&32 & 24 & 0&23 & 46 & 0&02 & 0&31 + 7 & 5&34 & 90 & 3&90 & 780 & 0&38 & 3&46 + 8 & 1&30 & 94 & 0&95 & 190 & 0&09 & 1&41 + 9 & 2&14 & 26 & 1&56 & 313 & 0&15 & 1&67 + 10 & 1&03 & 59 & 0&75 & 150 & 0&07 & 0&71 + cccccc t@xmath66 ( 10@xmath32 k)&4.0 & 4.0 & 4.0 & 4.0 & 4.0 + z / z@xmath21 & 0.34 & 0.30 & 0.34 & 0.30 & 0.10 + n / n@xmath21 & 0.34 & 0.20 & 0.34 & 0.18 & 0.08 + o / o@xmath21 & 0.34 & 0.36 & 0.34 & 0.36 & 0.08 + s / s@xmath21 & 0.34 & 0.36 & 0.34 & 0.36 & 0.10 + u ( 10@xmath29 ) & 4.22 & 6.59 & 6.93 & 4.66 & 3.92 + n@xmath31 ( 10@xmath34@xmath35 ) & 0.94 & 1.31 & 1.32 & 0.57 & 2.27 + cccccc @xmath67oii@xmath68 $ ] 3727 & & [email protected] & [email protected] & [email protected]&[email protected] + & * * & * 4.70 , 2.3e+0 * & * 4.16 , 1.7e-2 * & * 5.11 , 3.2e-1*&*2.64 , 2.5e-3
* + h@xmath69 & & [email protected] & & & + & * * & * 0.48 , 3.3e-1 * & * * & * * & * * + @xmath67oiii@xmath68 $ ] 4959 & [email protected] & [email protected]&[email protected] & & + & * 0.37 , 2.4e-1 * & * 0.72 , 3.3e-1 * & * 0.69 , 5.7e-1 * & * * & *
* + @xmath67oiii@xmath68 $ ] 5007 & [email protected] & [email protected]&[email protected]&[email protected] & [email protected] + & * 1.07 , 1.4e-2 * & * 2.07 , 4.1e-1 * & * 1.99 , 4.8e-2 * & * 1.35 , 2.4e-1 * & * 0.60 , 1.2e-3 * + @xmath67nii@xmath68 $ ] 6548 & [email protected] & [email protected]&[email protected] & [email protected] & + & * 0.43 , 1.8e-1 * & * 0.20 , 2.3e-2 * & * 0.35 , 9.5e-2 * & * 0.23 , 2.2e-4 * & *
* + h@xmath11 & [email protected] & [email protected]&[email protected] & [email protected] & [email protected] + & * 2.86 , 2.9e-5 * & * 2.85 , 7.2e-5 * & * 2.85 , 1.6e-5 * & * 2.86 , 3.7e-5 * & * 2.88 , 4.3e-4 * + @xmath67nii@xmath68 $ ] 6583 & [email protected] & [email protected]&[email protected] & [email protected] & [email protected] + & * 1.26 , 1.7e+0 * & * 0.58 , 1.7e-1 * & * 1.04 , 4.2e-1 * & * 0.67 , 2.8e-3 * & * 0.50 , 9.5e-2 * + @xmath67sii@xmath68 $ ] 6716 & [email protected] & [email protected]&[email protected] & [email protected] & + & * 0.62 , 3.1e-1 * & * 0.53 , 4.4e-1 * & * 0.48 , 2.6e-1 * & * 0.65 , 5.8e-1 * & * * + @xmath67sii@xmath68 $ ] 6731 & [email protected] & [email protected]&[email protected] & [email protected] & + & * 0.46 , 3.1e-1 * & * 0.41 , 8.6e-1 * & * 0.37 , 4.0e-1 * & * 0.47 , 5.4e-2 * & * * + cccc t@xmath66 ( 10@xmath32 k)&4.0 & 4.0 & 4.0 + z / z@xmath21 & 0.10 & 0.15 & 0.15 + n / n@xmath21 & 0.10 & 0.23 & 0.15 + o / o@xmath21 & 0.34 & 0.38 & 0.05 + s / s@xmath21 & 0.34 & 0.38 & 0.05 + u ( 10@xmath29 ) & 5.91 & 6.45 & 4.66 + n@xmath31 ( 10@xmath34@xmath35 ) & 1.11 & 0.95 & 0.57 + ccccccccc 3 + 9 & 19.37 & 19.12 & 18.67 & 0.26 & 0.45 & @xmath1916.9 & @xmath1917.2 & @xmath1917.6 + 7 & 19.87 & 19.81 & 19.33 & 0.06 & 0.47 & @xmath1916.4 & @xmath1916.5 & @xmath1916.9 + 8 & 23.27 & 23.30 & 22.69 & @xmath190.03 & 0.60 & @xmath1913.0 & @xmath1913.0 & @xmath1913.6 + 11 & 22.54 & 22.87 & 22.26 & @xmath190.32 & 0.61 & @xmath1913.7 & @xmath1913.4 & @xmath1914.0 + 10 & 21.27 & 21.07 & 20.46 & 0.20 & 0.61 & @xmath1914.2 & @xmath1914.4 & @xmath1914.9 + 12 & 22.05 & 21.84 & 21.42 & 0.21 & 0.42 & @xmath1915.0 & @xmath1915.2 & @xmath1915.8 + | this is the second part of a detailed study of the ultracompact group cg j1720 - 67.8 : in the first part we have focused the attention on the three main galaxies of the group and we have identified a number of candidate tidal dwarf galaxies ( tdgs ) .
here we concentrate on these candidate tdgs .
absolute photometry of these objects in @xmath0 bands confirms their relatively blue colors , as we already expected from the inspection of optical and near - infrared color maps and from the presence of emission - lines in the optical spectra .
the physical conditions in such candidate tdgs are investigated through the application of photoionization models , while the optical colors are compared with grids of spectrophotometric evolutionary synthesis models from the literature .
although from our data self - gravitation can not be proved for these objects , their general properties are consistent with those of other tdg candidates .
additionally we present the photometry of a few `` knots '' detected in the immediate surroundings of cg j1720 - 67.8 and consider the possibility that these objects might belong to a dwarf population associated with the compact group . |
restarts have been shown to boost the performance of backtracking sat solvers ( see for example @xcite,@xcite ) . a restart strategy ( @xmath0,@xmath1,@xmath2 , ... ) is a sequence of restart lengths that the solver follows in the course of its execution .
the solver first performs @xmath0 steps ( in case of sat solvers a step is usually a conflict ) .
if a solution is not found , the solver abandons its current partial assignment and starts over . the second time it runs for @xmath1 steps , and so on .
luby , sinclair and zuckerman @xcite show that for each instance there exists @xmath3 , an optimal restart length that leads to the optimal restart strategy ( @xmath3,@xmath3,@xmath3 , ... ) . in order to calculate @xmath3
, one needs to have full knowledge of the runtime distribution ( rtd ) of the instance , a condition which is rarely met in practical cases .
since the rtd is not known , solvers commonly use `` universal restart strategies '' .
these strategies do not assume prior knowledge of the rtd and they attempt to perform well on any given instance .
huang @xcite shows that when applied with conflict driven clause learning solvers ( cdcl ) , none of the commonly used universal strategies dominates all others on all benchmark families .
he also demonstrates the great influence on the runtime of different restart strategies , when all its other parameters are fixed . in this paper
we show that the recent success in applying machine learning techniques to estimate solvers runtimes can be harnessed to improve solvers performance .
we start by discussing the different universal strategies and recent machine learning success in sect .
[ sec : background ] . in sect .
[ sec : restart_strategy_portfolio ] we present _ lmpick _ , a restart strategy portfolio based solver .
experimental results are presented and analyzed in sect .
[ sec : results ] .
we conclude and suggest optional future study in sect .
[ sec : conclusion ] .
competitive dpll solvers typically use restarts .
most use `` universal '' strategies , while some use `` dynamic '' restart schemes , that induce or delay restarts ( such as the ones presented in @xcite and @xcite ) . currently , the most commonly used universal strategies fall into one of the following categories : * _ fixed strategy - _ ( @xcite ) . in this strategy
a restart takes place every constant number of conflicts .
while some solvers allow for a very short interval between restarts , others allow for longer periods , but generally fixed strategies lead to a frequent restart pattern .
examples of its use can be found in berkmin @xcite ( where the fixed restart size is 550 conflicts ) and seige @xcite ( fixed size is 16000 conflicts ) . *
_ geometric strategy - _ ( @xcite ) . in this strategy
the size of restarts grows geometrically .
this strategy is defined using an initial restart size and a geometric factor . wu and
van beek @xcite show that the expected runtime of this strategy can be unbounded worse than the optimal fixed strategy in the worst case .
they also present several conditions which , if met , guarantee that the geometric strategy would yield a performance improvement .
this strategy is used by minisat v1.14 @xcite with initial restart size of 100 conflicts and a geometric factor of 1.5 . *
_ luby strategy - _ ( @xcite ) . in this strategy
the length of restart @xmath4 is @xmath5 when @xmath6 is a constant `` unit size '' and @xmath7 + the first elements of this sequence are 1,1,2,1,1,2,4,1,1,2,1,1,2,4,8,1,1 , ... luby , sinclair and zuckerman @xcite show that the performance of this strategy is within a logarithmic factor of the true optimal strategy , and that any universal strategy that outperforms their strategy will not do it by more than a constant factor .
these results apply to pure las vegas algorithms , and do not immediately apply to cdcl solvers in which learnt clauses are kept across restarts .
the effectiveness of the strategy in cdcl solvers appears mixed ( @xcite,@xcite ) and there is still no theoretical work that that analyzes its effectiveness in such solvers .
however , luby s restart strategy is used by several competitive solvers including minisat2.1 and tinisat . * _ nested restart strategy - _ ( @xcite ) this strategy and can be seen as a simplified version of the luby strategy . after every iteration the restart length grows geometrically until it reaches a higher bound , at this point the restart size is reset to the initial value and the higher bound is increased geometrically .
this strategy is used by picosat @xcite and barcelogic .
previous work shows that restart strategies perform differently on different data sets .
huang @xcite compares the performance of different strategies both for benchmark families and different benchmarks .
he shows that there is no one strategy that outperformed all others across all benchmark families which suggests that adapting a strategy to a benchmark family , or even a single benchmark , could lead to performance gain .
this suggests that choosing the best strategy from a set of strategies could improve the overall runtime , and for some benchmark families , improves it significantly .
machine learning was previously shown to be an effective way to predict the runtime of sat solvers .
satzilla @xcite is a portfolio based solver that uses machine learning to predict which of the solvers it uses is optimal for a given instance .
satzilla uses an hierarchical approach @xcite and can use different evaluation criteria for solver performance .
satzilla utilizes runtime estimations to pick the best solver from a solver portfolio .
the solvers are used as - is , and satzilla does not have any control over their execution .
satzilla was shown to be very effective in the sat competition of 2007 .
two other machine learning based approaches for local search and cdcl are presented in @xcite and @xcite respectively .
ruan et al .
@xcite suggest a way to use dynamic programming to derive dynamic restart strategies that are improved during search using data gathered in the beginning of the search . this idea corresponds with the `` observation window '' that we will discuss in the next section .
there are several differences between this work and ours .
one important difference is that their technique chooses a different instance from the ensemble at each restart .
while our intention is to solve each of the instances in the ensemble , it seems their technique is geared towards a different goal , where the solver is given an ensemble of instances and is required to solve as many of them as possible .
another approach for runtime estimation was presented in our previous work @xcite . in that paper
we introduce a linear model predictor ( lmp ) which demonstrates that runtime estimation can also be achieved using parameters that are gathered in an online manner , while a search is taking place , as opposed to the mostly static features gathered by satzilla .
another difference between the methods is the way training instances are used .
while satzilla uses a large number of instances that are not tightly related , lmp uses a much smaller set of problems , but these should be more homogeneous .
since restart strategies are an important factor in the performance of dpll style sat solvers , a selection of a good restart strategy for a given instance should improve the performance of the solver for that instance .
we suggest that by using supervised machine learning , it is possible to select a good restart strategy for a given instance .
we present _ lmpick _ , a machine learning based technique which enhances cdcl solvers performance .
_ lmpick _ uses a portfolio of restart strategies from which it chooses the best one for a given instance .
following @xcite we recognize several restart strategies that have shown to be effective on one benchmark family or more .
we chose 9 restart strategies that represent , to our understanding , a good mapping of commonly used restart strategies . * _ luby-32 - _ a luby restart strategy with a `` unit run '' of 32 conflicts .
this strategy represents a luby restart strategy with a relatively small `` unit run '' .
this technique was shown to be effective by huang @xcite . * _ luby-512 - _ a luby restart strategy with a `` unit run '' of 512 conflicts .
this strategy represents a luby restart strategy with a larger `` unit run '' .
this is the original restart scheme used by tinisat , the solver we used in this study . *
_ fixed-512 - _ a fixed restart scheme with a restart size of 512 conflicts .
similar restart schemes that are used by solvers are berkmin s _ fixed-550 _ @xcite , and the 2004 version of zchaff , _ fixed-700
_ @xcite . *
_ fixed-4096 - _ a fixed balance scheme with a restart size of 4096 conflicts .
we chose this restart scheme because it balances the short and long fixed schemes , and because it performed very well in our preliminary tests . * _ fixed-16384 - _ a fixed balance scheme with a restart size of 16384 conflicts .
a longer fixed strategy , similar to the one used by siege @xcite ( _ fixed-16000 _ ) . * _ geometric-1.1 - _ a geometric restart scheme with a first restart size of 32 conflicts and a geometric factor of 1.1 .
inspired by the one used by hunag @xcite .
* _ geometric-1.5 - _ a geometric restart scheme with a first restart size of 100 conflicts and a geometric factor of 1.5 .
this is the restart scheme used in minisat v1.14 @xcite . *
_ nested-1.1 - _ a nested restart strategy , with an inner value of 100 conflicts , an outer value of 1000 values and a geometric factor of 1.1 .
this strategy is parameterized as the one used by picosat @xcite . *
_ nested-1.5 - _ a nested restart strategy , with an inner value of 100 conflicts , an outer value of 1000 values and a geometric factor of 1.5 .
inspired by the results presented in @xcite .
satsifiable and unsatisfiable instances from the same benchmark family tend to have different runtime distributions @xcite . a runtime prediction model that is trained using both sat and unsat instances performs worse than a homogeneous model .
it is better to train a layer of two models , one trained with satisfiable instances ( @xmath8 ) and the other with unsatisfiable instances ( @xmath9 ) . since in most cases
we do not know whether a given instance is satisfiable or not we need to determine which of the models is the correct one to query for a given instance according to its probability to be satisfiable .
previous work ( @xcite,@xcite ) suggests that machine learning can be successfully used for this task as well .
a classifier can be trained to estimate the probability of an instance to be satisfiable .
some classification techniques perform better than others , but it seems that for most benchmark families , a classifier with 80% accuracy or more is achievable . using supervised machine learning , we train models offline in order to use them for predictions online . for every training example @xmath10 , where @xmath11 is the training set
, we gather the feature vector @xmath12 using the features presented in section [ sec : featurevector ] . once the raw data
is gathered , we perform a feature selection .
we repeatedly remove the feature with the smallest standardised coefficient until no improvement is observed based on the standard aic ( akaike information criterion ) . we then searched and eliminate co - linear features in the chosen set .
the reduced feature vector @xmath13 is then used to train a classifier and several runtime prediction models .
the classifier predicts the probability of an instance to be satisfiable and the runtime models predict cpu - runtime .
_ lmpick _ trains one classifier , but two runtime models for each restart strategy @xmath14 ( where @xmath15 is the set of all participating strategies ) to the total of @xmath16 models .
each training instance is used to train the satisfiability classifier , labeled with its satisfiability class , and @xmath17 runtime models , for each model it is labeled with the appropriate runtime . as the classifier , we used a logistic regression technique .
any classifier that returns probabilities would be suitable .
we found logistic regression to be a simple yet effective classifier which was also robust enough to deal with different data sets .
we have considered both sparse multinomial linear regression @xcite ( suggested to be effective for this task in @xcite ) , and the classifiers suggested by devlin and osullivan in @xcite , but the result of all classifiers were on par when using the presented feature vector on our datasets . for the runtime prediction models we used ridge linear regression .
using ridge linear regression , we fit our coefficient vector @xmath18 to create a linear predictor @xmath19 .
we chose ridge regression , since it is a quick and simple technique for numerical prediction , and it was shown to be effective in the linear model predictor ( lmp ) @xcite . while lmp predicts the log of number of conflicts , in this work we found that predicting cpu - runtime is more effective as a selection criterion for restart strategies . using the number of conflicts as a selection criterion tends to bias the selection towards frequent restart strategies for large instances .
this is because an instance with many variables spends more time going down the first branch to a conflict after a restart .
this work is unaccounted for when conflicts are used as the cost criterion .
hence a very frequent restart strategy might be very effective in the number of conflict while much less effective in cpu - time .
there are 4 different sets of features that we used in this study , all are inspired by the two previously discussed techniques - satzilla @xcite and lmp @xcite .
the first set include only the number of variables and the number of clauses in the original clause database .
these values are the only ones that are not normalized .
the second set includes variables that are gathered before the solver starts but after removing clauses that are already satisfiable , shrinking clauses with multiple appearances and propagating unit clauses in the original formula .
these features are all normalized appropriately .
they are inspired by satzilla and were first suggested in @xcite . the third set include statistics that are gathered during the `` observation window '' , this is a period where we analyze the behavior of the solver while solving the instance .
the `` observation window '' was first used in @xcite . the way the observation window
is used in this study will be discussed shortly .
the variables in this set are the only ones which are dpll dependent .
the last set includes the same features as the second , but they are calculated at the end of the observation window . a full list of the features is presented in fig .
[ fig : featurvector ] . for further explanation about these features see @xcite and @xcite .
[ fig : steps ] .,title="fig:",width=309 ] once all runtime models are fitted and the satisfiability classifier is trained , we can use them to improve performance for future instances .
the steps that are taken by _
lmpick _ are presented in fig .
[ fig : steps ] . since no prediction can be made before the observation window is terminated , and since we favor an early estimation , it is important that the observation window should terminate early in the search . in our preliminary testings
we have noticed that the first restart tends to be very noisy , and that results are better if data is collected in the second restart onwards .
we have tried several options for the observation window location and size , eventually we opted for a first restart which is very short ( 100 conflicts ) , followed by a second restart ( of size 2000 ) which hosts the observation window .
hence the observation window is closed and all data is gathered after 2100 conflicts .
once the feature vector @xmath20 is gathered it is used with the classifier to determine the probability of the instance to be satisfiable , @xmath21 . for each of the strategies ,
both models are queried and a best strategy @xmath22 is picked using @xmath23.\ ] ] the restart strategy which is predicted to be the first to terminate is picked , and the solver starts following this strategy from the next restart onwards .
although restart strategies are usually followed from the beginning of the search , we do not want to lose the learned clauses from the first 2100 conflicts .
therefore , we continue the current solving process and keep the already learnt clauses .
we denote the restart sequence that takes place from the first restart to termination as @xmath24 .
it is important to note that @xmath25 .
[ fig : script : bmc : sat ] dataset .
different instances are made using different array_size values and a different number of unwinding iterations.,title="fig:",width=196 ] [ fig : script : bmc : unsat ] dataset .
different instances are made using different array_size values and a different number of unwinding iterations.,title="fig:",width=196 ]
in this study we used tinisat ( version 0.22 ) @xcite as the basic solver .
tinisat is a lightweight dpll style solver that was first presented in the sat race of 2006 .
tinisat is a modern solver that uses clause learning and a unique decision heuristic that generally favours variables from recent assignments ( as in berkmin @xcite ) and uses vsids @xcite over the literals as a backup .
we chose to use tinisat since ( i ) it is tuned in a way that would make comparison of restart strategies more meaningful @xcite and ( ii ) it is a compact and straightforward implementation which allows for greater ease of use .
tinisat is not equipped with a pre - processor , and we have not used any in our study . by default , tinisat uses a luby restart strategy with a run unit of 512 conflicts .
all our experiments were conducted on a cluster of 14 dual intel xeon cpus with em64 t ( 64-bit ) extensions , running at 3.2ghz with 4 gb of ram under debian
gnu / linux 4.0 . by implementing a runtime cutoff of 90 minutes per instance
, we managed to complete all experiments in approximately 290 cpu days .
.summary of features of datasets . for each dataset
the following details are presented : the dataset s classification ( class ) , the number of instances ( ins . ) and its size in mb ( all file are zipped ) . also , we present the time ( in hours ) it took for all 9 restart strategies to solve these datasets .
we present the mean time ( mean ) , the standard deviation ( sd ) and the minimal and maximal time .
runtime cutoff is 5400 seconds and it is the maximal runtime per instance . [ cols="<,<,<,<,^,^,^,^ " , ] in order for _ lmpick _ to enhance the solver s performance , runtime estimation models need to perform well .
nevertheless , it is not crucial that each model s prediction is accurate .
the important factor is the relative order of these predictions .
we would prefer the chosen strategy ( @xmath26 ) to be amongst the best strategies for that instance .
table [ table : bestpick_location ] demonstrates the quality of the chosen strategy @xmath26 .
the table presents the percentage of strategies that outperform @xmath26 .
this table shows that in all cases @xmath26 is better than picking a random strategy , and that for the _ crypto - unsat _ instances it performs very well .
the difference in the performance on the _ crypto - sat _ data set is not easily explained by the data presented so far . while both the classifier and runtime prediction models perform better than on the _ velev - sat _ , the overall performance is worse .
we conjecture that the cause of this difference is the differing likelihood of an instance being solved by multiple strategies .
[ fig : compare - sat - datasets : bars ] [ fig : compare - sat - datasets : plots ] figure [ fig : compare - sat - datasets ] compares the 3 non - random satisfiable data sets . in fig .
[ fig : compare - sat - datasets : bars ] bars represent the percent of instances in the data set that were solved within the cutoff time by each number of strategies .
there is a clear difference between _ crypto - sat _ and the other 2 data sets . while for both industrial verification based data sets , most instances were solved by the majority of strategies , for
_ crypto - sat _ many of the instances are solved by a small set of strategies .
the effect of this difference is demonstrated in fig .
[ fig : compare - sat - datasets : plots ] .
this plot presents the probability of a randomly picked instance being solved within the cutoff time given the quality of selected strategy . from left to right , the picked strategy shift from best to worse .
if _ lmpick _ picks one of the the two best strategies , the probability of all three data sets is quite similar , but when the chosen strategy is 3rd or 4th , the probability of solving a _ crypto - sat _ instance drops significantly compared to the other two .
many instances in the _ crypto - sat _ data set are only solved within the cutoff by a small subset of strategies , making this data set harder as a sub - optimal selection is likely to lead to a timeout .
restart strategies have an important role in the success of dpll style sat solvers .
the performance of different strategies varies over different benchmark families .
we harness machine learning to enhance the performance of sat solvers .
we have presented _ lmpick _ , a technique that uses both satisfiability classification and solver runtime estimation to pick promising restart strategies for instances . we have demonstrated the effectiveness of _ lmpick _ and compared its results with the most commonly used restart strategies .
we have established that in many cases _ lmpick _ outperforms any single restart strategy and that it is never worse than a randomly picked strategy .
we have also discussed the influence of different components of _ lmpick _ on its performance .
while universal restart strategies are more commonly used than dynamic ones , dynamic strategies are getting more attention lately . an interesting continuation to
this work would be to use machine learning to develop a fully dynamic restart strategy .
such a strategy could use unsupervised machine learning algorithm to develop a dynamic restart policy for a benchmark family of problems .
restart strategies are not the only aspect of sat solving influencing performance on different benchmark families .
machine learning can be also used to tune other parameters that govern the behavior of modern dpll based solvers , namely , parameters that are commonly set manually as a result of a trial and error process , such as decision heuristic parameters , clause deletion policy , etc .
nicta is funded by the department of broadband , communications and the digial economy , and the arc through backing australia s ability and the ict centre of excellence program .
hutter , f. , hamadi , y. , hoos , h. , leyton - brown , k. : performance prediction and automated tuning of randomized and parametric algorithms .
proc . of the 12th int .
conf . on principles and practice of constraint programming ( 2006 ) krishnapuram , b. , figueiredo , m. , carin , l. , hartemink , a. : sparse multinomial logistic regression : fast algorithms and generalization bounds . ieee transactions on pattern analysis and machine intelligence , 27:957 - 968 ( 2005 ) nudelman , e. , leyton - brown , k. , hoos , h.h . ,
devkar , a. , shoham , y. : understanding random sat : beyond the clauses - to - variables ratio . proc . of the 10th int .
conf . on principles and practice of constraint programming ( 2004 ) | restart strategies are an important factor in the performance of conflict - driven davis putnam style sat solvers . selecting a good restart strategy for
a problem instance can enhance the performance of a solver .
inspired by recent success applying machine learning techniques to predict the runtime of sat solvers , we present a method which uses machine learning to boost solver performance through a smart selection of the restart strategy .
based on easy to compute features , we train both a satisfiability classifier and runtime models .
we use these models to choose between restart strategies .
we present experimental results comparing this technique with the most commonly used restart strategies .
our results demonstrate that machine learning is effective in improving solver performance . |
nature of the mechanism generating order out of randomness constitutes one of the most fundamental issues of the contemporary physics .
theories based on various versions of ensembles of the random matrices provide one possible theoretical frame for studying such effects .
the gaussian orthogonal ensemble ( goe ) constitutes the most common reference .
the related concepts originate @xcite from nuclear physics and prove very useful in the area of strongly interacting fermi systems or in quantum chaos @xcite . at present
they finds even broader applications in such diverse fields like the brain research @xcite , econophysics @xcite and most recently in the `` real - world '' networks or graphs @xcite .
utility of the standard random matrix theory ( rmt ) results form the fact that a potential agreement reflects the generic properties of a system and thus in many cases it provides an appropriate null hypothesis . from this perspective
the deviations are even more interesting as they can be used to quantify some properties which are nonrandom and thus system specific . in this context
the recently identified @xcite preponderance of the @xmath1 ground states in strongly interacting fermi systems , such as atomic nuclei , arising from random two - body interactions seems to indicate the effect reflecting a sparser connectivity than just pure random .
several closely related issues have also been addressed in the context of mesoscopic @xcite and randomly interacting many - spin systems @xcite .
one purpose of the present investigation is to identify the origin of the related physically relevant deviations from standard rmt and to quantify their character . since it was nuclear physics which gave birth to rmt
we believe that the present example , even though addressed in the nuclear context , may also stimulate much broader activity and understanding of similar effects in other areas .
our theoretical framework is thus analogous to this of ref . @xcite .
then schematically , indicating nevertheless all the relevant ingredients , the interaction matrix elements @xmath2 of good total angular momentum @xmath0 in the shell - model basis @xmath3 can be expressed as follows @xcite : @xmath4 the summation runs over all combinations of the two - particle states @xmath5 coupled to the angular momentum @xmath6 and connected by the two - body interaction @xmath7 .
@xmath8 denote the radial parts of the corresponding two - body matrix elements while @xmath9 globally represent elements of the angular momentum recoupling geometry .
structures analogous to eq .
( [ eqv ] ) can be identified in various other areas .
the quantum open systems @xcite or the neural nets @xcite provide immediate examples . in statistical ensembles of matrices
the crucial factor determining the structure of eigenspectrum is the probability distribution @xmath10 of matrix elements @xcite . especially relevant
are the tails of such distributions since they prescribe the probability of appearance of the large matrix elements . from the point of view of the mechanism producing the energy gaps they are most effective in generating a local reduction of dimensionality responsible for such effects . in principle , the probability distribution of the shell model matrix elements is prescribed by their general structure expressed by the eq .
( [ eqv ] ) , provided the probability distributions of both @xmath8 and @xmath9 are known . in general terms
this structure can be considered to be of the form @xmath11 and each @xmath12 to be a product of another two variables denoted as @xmath13 and @xmath14 . by making use of the convolution theorem @xcite the probability distribution @xmath10
that @xmath15 assumes a value equal to @xmath16 can be expressed as : @xmath17 , \label{eqpv}\ ] ] where @xmath18 denotes a fourier transform , @xmath19 its inverse and @xmath20 the probability distributions of individual terms .
taking in addition into account the fact that @xmath21 one can explicitely derive the form of @xmath22 in several cases .
assuming for instance that all the above constituents are identically gaussian distributed ( then , according to eq .
( [ eqpvi ] ) , @xmath23 and thus @xmath24 ) one arrives at @xmath25 where @xmath26 stands for the modified bessel function .
asymptotically , for large @xmath16 , this leads to @xmath27 for such a global estimate the identical gaussian distribution of @xmath8 is consistent both with the two - body random ensemble ( tbre ) @xcite and with the random quasiparticle ensemble ( rqe ) @xcite .
the only anticipated difference originates from the fact that in the second case the variance of the distribution drops down with @xmath6 like the inverse of @xmath28 which is expected to result in a smaller effective @xmath29 as compared to tbre .
by contrast , in both versions of the above random ensembles the geometry expressed by @xmath9 enters explicitely . however , the complicated quasi - random coupling of individual spins is believed @xcite to result in the so - called geometric chaoticity @xcite . for the extreme values of @xmath0
the underlying selection rules may however impose severe constraints in achieving such a limit .
below we therefore explicitly verify its range of applicability .
the model to be quantitatively explored here consists , similarly as in ref .
@xcite , of 6 identical particles ( all single - particle energies are set to zero ) operating in the @xmath30 shell .
1 shows distributions of the corresponding geometrical factors @xmath9 for @xmath31 and for several relevant values of @xmath0 and @xmath6 .
as one can see , the gaussian may be considered quite a reasonable representation of the distribution of such factors for all combinations of @xmath0 and @xmath6 shown , with one exception , for those which involve @xmath1 . in this later case
the distribution of @xmath32 resembles more a uniform distribution over a finite interval located symmetrically with respect to zero .
one principal reason for this fact is that the @xmath33 symbols which enter @xmath9 are here more selective .
these empirical facts justify well the estimates of @xmath10 based on eq .
( [ eqpr ] ) for @xmath34 and not so well for @xmath1 .
more appropriate in this particular case is to assume a uniform distribution of @xmath32 over an interval confined by say @xmath35 and @xmath36 , i.e. , @xmath37 , retaining @xmath38 in its original gaussian form of course . by making use of eqs .
( [ eqpv ] ) and ( [ eqpvi ] ) one then obtains @xmath39^n \cos ( \omega v ) d\omega \label{eqpvu}\ ] ] which for large @xmath16 behaves like @xmath40 an explicit calculation of the distribution of the shell model off - diagonal matrix elements for the various @xmath0-values based on the present model with two - body matrix elements drawn from rqe ( tbre results in similar relations among different @xmath0-sectors though the distributions are somewhat broader as compared to rqe ) confirms the above analytical estimates as is illustrated in fig . 2 .
indeed , such a distribution in the @xmath1 sector resembles more a gaussian and the large @xmath16 tails of this distribution drop down faster as compared to the @xmath34-sectors where this asymptotics is exponential ( eq . ( [ eqpa ] ) ) . at the same time
the @xmath34 sectors are dominated by very small matrix elements to a larger degree than @xmath1 .
the probability of appearance of a large off - diagonal matrix element which in magnitude overwhelms the remaining ones is thus greater for @xmath34 than for @xmath1 .
such an effective reduction of the rank in the former case is expected to result in a stronger tendency to localization as compared to goe @xcite .
the corresponding characteristics can be quantified in terms of the information entropy @xmath41 of an eigenstate labelled by @xmath42 from the @xmath0-sector .
the coefficients @xmath43 denote the eigenvector components in the basis @xmath44 . such a mean field basis offers an appropriate reference @xcite for the present purpose . since the definition of @xmath45 involves the total number of states @xmath46 which differ for different @xmath0 s , before relating the result to the goe we normalise @xmath45 to the goe limit of this quantity @xcite @xmath47 where @xmath48 is the digamma function . within our model
the so - calculated and rqe ensemble averaged quantity for all the states versus their corresponding energies @xmath49 is illustrated in fig .
as anticipated , it is not @xmath1 whose lowest eigenstate comes out most localised , i.e. , most regular .
the lowest states for several higher @xmath0 values ( like 2 and especially 4 ) deviate much more from goe .
this thus indicates more favorable conditions for the emergence of energy gaps for larger @xmath0 than for @xmath1 .
3 provides one more information which turns out helpful to properly interpret the results .
the @xmath1 states are spread over the broadest energy interval even though the number of states @xmath50 is here significantly smaller than for several larger @xmath0 values @xmath51 . as a result ,
the average level spacing is a factor of few larger for @xmath1 than for the remaining ones . in fig .
4 ( dashed line ) we therefore show distributions of the ground state @xmath52 gaps @xmath53 where similarly as in ref .
@xcite , @xmath54 though here for each @xmath0 individually .
as it is clearly seen the @xmath1-sector does not significantly distinguish from the remaining ones .
in view of our investigations presented above one would however expect a reduced probability for occurance of the large ground state energy gaps in this particular sector . as the solid lines in fig .
4 indicate such an effect does indeed take place when @xmath55 in eq .
( [ eqrho ] ) is replaced by @xmath56 in fact , it seems more appropriate and more consistent with the above global considerations to relate the ground state energy gap just to the average global level spacing among the remaining states , characteristic for a given @xmath0 , as expressed by eq .
( [ eqdbar ] ) .
finally one may ask a question why this tendency does not extend to the highest @xmath0-values . in this connection one has to remember that the off - diagonal matrix elements is not the only relevant element .
these are the diagonal matrix elements which constitute the driving term .
irrespective of the value of @xmath0 their distribution is always gaussian - like .
this can be observed numerically and is consistent with arguments formulated in terms of eqs .
( [ eqv ] - [ eqpa ] ) since the geometrical factors @xmath9 entering the diagonal matrix elements are always nonnegative . as it is shown in fig . 5 ( dashed lines ) ,
increasing however @xmath0 beyond 4 results in a significant reduction of the variance of @xmath10 for the diagonal matrix elements and consequently a larger fraction of the off - diagonal matrix elements becomes effective in mixing the basis states .
in addition , due to a smaller number of terms entering the eq .
( [ eqv ] ) , for the stretched high-@xmath0 states the effect of geometric chaoticity is no longer effective and in this respect the conditions become similar to those for @xmath1 . as a result
, the distribution of off - diagonal matrix elements converts back towards more gaussian - like shaped , i.e. @xmath57 becomes larger ( for @xmath58 not shown in fig .
2 @xmath59 ) . superposing the above two effects one thus obtains even smaller gaps and even more delocalized states at the edges of the spectra as compared to @xmath1 .
in fact the spectral density ( solid lines in fig . ( 5 ) ) becomes even somewhat closer to semicircular in this case .
the present investigation based both on theoretical as well as on numerical arguments clearly shows that the many - body problems described in terms of various variants of the two - body random ensembles ( like rqe or tbre ) develop quantitatively well identified deviations from the goe .
these deviations can be linked to differences in the distribution of matrix elements and quantified in terms of the localisation or of energy gaps in eigenspectra .
contrary to the common belief they point to the intermediate total angular momenta as those @xmath0-sectors whose ground states are ordered most . from this perspective
, a predominance of the @xmath1 ground states @xcite can be viewed as a result of mixing states with different characteristic energy scales from different @xmath0-sectors .
it seems also appropriate to notice here that the arguments formulated in terms of eqs .
( [ eqv]-[eqpa ] ) provide a more adequate approach towards understanding the distribution of matrix elements in realistic nuclear shell - model calculations than the ones based on multipole expansion @xcite .
finally , similar parallels between the distribution of matrix elements and the structure of eigenspectra relative to the goe can be set in the recent econophysics @xcite applications .
we acknowledge useful discussions with j. okoowicz , m. poszajczak and i. rotter at the early stage of this development .
this work was partly supported by kbn grant no .
2 p03b 097 16 and by the german - polish dlr scientific exchange program , grant no .
pol-028 - 98
. e.p .
wigner , ann . math . * 53 * ( 1951 ) 36 m.l .
mehta , _ random matrices _ ( academic press , boston , 1991 ) v. zelevinsky , b.a .
brown , n. frazier , m. horoi , phys . rep .
* 276 * ( 1996 ) 85 t. guhr , a. mller - groeling , h.a .
weidenmller , phys . rep .
* 299 * ( 1998 ) 189 j. kwapie , s. drod , a.a .
ioannides , phys .
e * 62 * ( 2000 ) 5557 l. laloux , p. cizeau , j .-
bouchaud , m. potter , phys .
* 83 * ( 1999 ) 1467 v. plerou , p. gopikrishnan , p. rosenow , l.a.n .
amaral , h.e .
stanley , phys . rev
* 83 * ( 1999 ) 1471 s. drod , f. grmmer , a.z .
grski , f. ruf , j. speth , physica a * 287 * ( 2000 ) 440 i. farkas , i. dernyi , a .-
barabsi , t. vicsek , phys .
e * 64 * 026704:1 - 12 ( 2001 ) c.w .
johnson , g.f .
bertsch , d.j .
dean , phys .
* 80 * ( 1998 ) 2749 r. bijker , a. frank , s. pittel , phys .
c * 60 * ( 1999 ) 021302 ; c.w .
johnson , g.f .
bertsch , d.j .
dean , i. talmi , phys .
c * 61 * , 014311(2000 ) 014311 ; r. bijker , a. frank , phys . rev .
* 84 * ( 2000 ) 420 ph .
jacquod , a.d .
stone , phys .
* 84 * ( 2000 ) 3938 l. kaplan , t. papenbrock , phys .
rev . lett .
* 84 * ( 2000 ) 4553 i. talmi , _ simple models of complex nuclei _ , ( harwood academic publishers , 1993 ) j.j .
hopfield , proc .
usa * 79 * , 2554(1982 ) s. drod , a. trellakis , j. wambach , phys .
76 ( 1996 ) 4891 s. drod , s. nishizaki , j. speth , m. wjcik , phys .
e * 57 * ( 1998 ) 4016 r. bracewell , _ the fourier transform and its applications _ , ( mcgraw - hill , new york , 1999 ) j.b .
french , s.s.m .
wong , phys .
b * 33 * ( 1970 ) 449 t. ericson , adv
* 9 * ( 1960 ) 425 p. cizeau and j.p .
bouchaud , phys .
e * 50 * ( 1994 ) 1810 s. drod , s. nishizaki , j. wambach , j. speth , phys .
* 74 * ( 1995 ) 1075 f.m .
izrailev , phys . rep .
* 196 * ( 1990 ) 299 s. drod , j. kwapie , f. grmmer , f. ruf , j. speth , physica a , * 29 * 9 ( 2001 ) 144 + * fig . 1 * the normalised distribution of geometrical factors @xmath9 entering the off - diagonal matrix elements ( eq . ( [ eqv ] ) ) for the model of 6 particles in the @xmath30-shell . +
* fig . 2 * the probability distributions of nonzero many - body off - diagonal matrix elements in different @xmath0-sectors drawn from one thousand of rqe samples of two - body matrix elements .
the energy scale is set by @xmath60 , where @xmath61 and @xmath62 determines the rqe mean square variance .
these distributions are fitted ( solid lines ) in terms of eq .
( [ eqpr ] with @xmath29 treated as a fitting parameter .
the corresponding best @xmath29 s @xmath63 for each @xmath0 are listed . by increasing @xmath29 the distribution prescribed by eq .
( [ eqpr ] ) quickly approaches ( as a consequence of the central limit theorem ) the gaussian distribution . in this way
the @xmath1 distribution is demonstrated to be much closer to the gaussian than the remaining ones whose asymptotic behaviour is consistent with a slower , exponential fall - off . + * fig . 3 * the information entropy normalised to its goe limit @xmath64 for all the states @xmath42 from various @xmath0-sectors ( all positive parity ) versus energies @xmath65 of those states .
all the quantities are ensemble averaged .
the energy units are the same as in fig . 2 . +
4 * distributions of ground state energy gaps @xmath66 as defined by the eq .
( [ eqrho ] ) for successive @xmath0 s .
the dashed line uses @xmath55 defined by eq .
( [ eqd ] ) while the solid line the one defined by eq .
( [ eqdbar ] ) . + * fig . 5 * the probability distributions of the diagonal matrix elements in different @xmath0-sectors drawn from one thousand of rqe samples of two - body matrix elements ( dashed lines ) and the corresponding spectral densities ( solid lines ) .
the energy units are the same as in fig . | we investigate the origin of order in the low - lying spectra of many - body systems with random two - body interactions .
our study based both on analytical as well as on numerical arguments shows that except for the most @xmath0-stretched states , the ground states in the higher @xmath0-sectors are more orderly and develop larger energy gaps than the ones in the @xmath1-sector . due to different characteristic energy scales in different @xmath0-sectors
the @xmath1 ground states may predominate only when all the states are taken together .
pacs numbers : 05.30.fk , 21.60.-n , 24.60.lz , 21.10.re |
in the first part of our paper @xcite , extending an approach formulated in ref .
@xcite to the relativistic limit , we have studied systematically the structure of the singularities formed in a relativistically large amplitude plasma wave close to the wavebreaking in a thermal plasma .
we have shown that typically the electron density distribution in the breaking wave has a peakon " form with a discontinuous coordinate dependence of its first derivative , similar to the profiles of nonlinear water waves @xcite and that in the above breaking limit the derivative becomes infinite .
this results in a finite reflectivity of an electromagnetic wave interacting with nonlinear plasma waves .
in particular , this is an important property because nonlinear langmuir waves play a key role in the
relativistic flying mirror " concept @xcite . in this concept ,
very high density electron shells are formed in the nonlinear wake wave generated by an ultrashort laser pulse propagating in an underdense plasma with a speed close to the speed of light in vacuum .
the shells act as mirrors flying with relativistic velocity .
when they reflect a counterpropagating electromagnetic pulse , the pulse is compressed , its frequency is upshifted and its intensity increased .
it is the singularity in the electron density distribution that allows for a high efficiency in the reflection of a portion of the counterpropagating electromagnetic pulse .
if the langmuir wave is far below the wave - breaking threshold , its reflectivity is exponentially small . for a nonlinear langmuir wave the singularity formed in the electron density breaks
the geometric optics approximation and leads to a reflection coefficient that is not exponentially small @xcite . in the present paper
we address the problem of the interaction of an electromagnetic wave with a nonlinear plasma wave which is of interest for the photon accelerator " concept @xcite and for the
relativistic flying mirror " paradigm @xcite .
we calculate the reflection coefficients of an electromagnetic wave at the singularities of the electron density in the most typical regimes of a strongly nonlinear wave breaking in thermal plasmas .
as we have seen in the first part of our paper @xcite , in a strongly nonlinear wake wave the electron density is modulated and forms thin shells ( singularities or caustics in the plasma flow ) moving with velocity @xmath0 . in the introduction , in a way of refs .
@xcite , we have discussed how a counterpropagating electromagnetic wave can be partially reflected from these density shells which play the role of relativistic mirrors . while in the case of a cold plasma the electron density at the singularity tends to infinity ( see eq .
( 59 ) of part i @xcite and refs .
@xcite ) , in a thermal plasma the density is limited by the expressions given by eqs . ( 41 ) and ( 42 ) of part i @xcite .
although in this case the density profile is described by a continuous function of the variable @xmath1 , its derivatives with respect to @xmath1 are discontinuous .
this discontinuity results in the breaking of the geometric optics approximation and leads to a reflectivity that is not exponentially small . in order to calculate the reflection coefficient
, we consider the interaction of an electromagnetic wave with the electron density shell formed at the breaking point of a langmuir wave in a thermal plasma similarly to what has been done in refs .
the electromagnetic wave , described by the @xmath2 component of the vector potential @xmath3 , evolves according to the linearized wave equation @xmath4 where we have reverted to dimensional units and @xmath5 the last term in the l.h.s .
( [ eq25-weq ] ) is the @xmath6component of the electric current density generated by the electromagnetic wave in a plasma with the electron distribution function @xmath7 . in the limit @xmath8 for the electromagnetic wave frequency larger than the langmuir frequency calculated for the maximal electron density , @xmath9
, we can neglect the finite temperature effects on the electromagnetic wave dispersion , which have been analyzed in ref .
@xcite , in the limit of homogeneous , stationary plasmas . for the water - bag distribution function @xmath10
@xmath11 takes the form @xmath12 where @xmath13 , @xmath14 and @xmath15 are now dimensionless ( normalized on @xmath16 ) .
the wake wave modulates the electron density and temperature increasing them in the compression regions and decreasing them in the rarefaction regions . in fig .
[ fig14 ] we illustrate the dependence of @xmath17 on @xmath1 for the parameters of a wakewave corresponding to @xmath18 and @xmath19 at @xmath20 and for @xmath21 . on the coordinate @xmath1 for the parameters of a wakewave corresponding to @xmath18 and @xmath19 at @xmath20 and for @xmath21 . in the inset
the ratio @xmath22 is shown in the vicinity of the maximum.,width=302,height=188 ] from eqs .
( 35 ) and ( 59 ) of part i @xcite in the ultrarelativistic case , @xmath23 , using eq .
( [ eq26-ompe ] ) we find for a relatively cold distribution such that @xmath24 that near the wavebreaking point @xmath11 is given by @xmath25 the propagation of a sufficiently short electromagnetic wave packet in the plasma with electron density modulated by the langmuir wave can be described within the framework of the geometric optics approximation .
the electromagnetic wave is represented as a particle ( photon " ) with coordinate @xmath26 and momentum @xmath27 ( wave vector ) .
the interaction of a photon " with a langmuir wave that propagates with a relativistic phase velocity @xmath28 can be accompanied by a substantial frequency upshift called photon acceleration `` @xcite . using the dispersion equation @xmath29 where @xmath30 with @xmath31 and @xmath32 the wave vector components parallel and perpendicular to the propagation direction of the langmuir wave , we obtain the ' ' photon " hamiltonian function which depends on the canonical variables @xmath33 and @xmath31 ( see ref .
@xcite ) @xmath34 the transverse component of the wave vector is constant @xmath35 and @xmath36 is assumed for the sake of simplicity . the phase portrait of the photon for @xmath37 given by eq .
( [ eq26-ompe ] ) for the parameters corresponding to fig .
[ fig14 ] is shown in fig .
[ fig15 ] . .
the dashed line corresponds to the trajectory of photons that have appeared due to the over - barrier reflection at the crest of the breaking wave . in the inset
the photon trajectories in the vicinity of the saddle point are shown.,width=340,height=226 ] along an orbit corresponding to the value @xmath38 of the hamiltonian ( [ eq25-ham - phtn ] ) the photon frequency is given by @xmath39 .
\label{eq25-omega - phtn}\ ] ] photons , for which @xmath40 are trapped inside the region encircled by the separatrix . along the orbit
their frequency changes from @xmath41 and @xmath42 corresponding to the plus and minus signs in the r.h.s . of eq .
( [ eq25-omega - phtn ] ) at the minimum of @xmath37 .
photons with @xmath43 are not trapped and for them the sign in the r.h.s .
( [ eq25-omega - phtn ] ) does not change . for trajectories far above the separatrix the photon frequency variations are relatively weak .
however a sufficiently strong wakefield can reflect a counterpropagating photon , @xmath44 due to above - barrier reflection ( this trajectory is shown in fig . [ fig15 ] by a dashed line ) .
such a photon acquires a frequency @xmath45 according to the einstein formula for the frequency of the electromagnetic wave reflected by a relativistic mirror @xcite .
the geometric optics approximation fails when the wakefield is close to wave breaking and this provides the appropriate conditions for a not exponentially weak wave scattering .
in order to find the reflectivity of the nonlinear wake wave we make a lorentz transformation to the frame of reference moving with the phase velocity of the langmuir wave . in the boosted frame , eq . ( [ eq25-weq ] ) for the electromagnetic wave interacting with the nonlinear langmuir wave can be written as @xmath46 with @xmath47 } \label{eq34-az}\ ] ] and in the neighbourhood of the breaking point @xmath48 can be written as @xmath49 here @xmath50 and @xmath51 , @xmath52 , @xmath53 , @xmath54 are the coordinate and time and the wave number and frequency in the boosted frame of reference .
the coefficient @xmath55 is equal to @xmath56 we seek for the solution to the above - barrier scattering problem for eq .
( [ eq33-ax ] ) writing its solution in the form ( see refs .
@xcite ) @xmath57 , \label{eq40-sol}\ ] ] where the phase integral is defined as @xmath58 the above - barrier scattering geometry is illustrated in fig .
[ fig16 ] . for constant @xmath59 and
@xmath60 eq . ( [ eq40-sol ] ) corresponds to the wkb " solution @xcite . in the following the coefficients @xmath59 and
@xmath60 are considered as functions of @xmath61 instead of @xmath62 , because , as explained in ref .
@xcite , the mapping between @xmath61 and @xmath62 given by eq .
( [ eq40-w ] ) is one - to - one on the real axis .
far from the breaking point , i.e. formally for @xmath63 the function @xmath64 reduces to a constant and the solutions ( [ eq40-sol ] ) are exact , so that @xmath65 as @xmath66 . in other words ,
the boundary conditions at @xmath63 are @xmath67 since in the representation ( [ eq40-sol ] ) , the single unknown function @xmath68 has been replaced by the two unknown functions @xmath69 , a subsidiary condition is necessary . we shall impose the condition @xmath70 differentiating eq .
( [ eq40-sol ] ) with respect to @xmath62 and taking into account the constraint ( [ eq43-subs ] ) , we find @xmath71 while differentiating eq .
( [ eq43-subs ] ) with respect to @xmath62 and substituting @xmath72 into eq .
( [ eq33-ax ] ) yields @xmath73 the system of eqs .
( [ eq45-bpm1 ] ) and ( [ eq45-bpm2 ] ) is equivalent to eq .
( [ eq33-ax ] ) .
it can be rewritten in the form @xmath74 with @xmath75 for @xmath76 given by eq .
( [ eq35-q2 ] ) we have @xmath77 \ , { \rm sign}(\zeta ) , \label{eq45-sw11}\ ] ] where @xmath78 if @xmath79 and @xmath80 for @xmath81 , and @xmath82 so that @xmath83 it follows that @xmath84 is discontinuous at @xmath85 ( @xmath86 ) @xmath87 integrating both sides of eq .
( [ eq44-b ] ) and using the above formulated boundary conditions for @xmath88 , we can obtain the reflection coefficient @xmath89 in the form of the infinite series @xcite @xmath90 with the product equal to unity for @xmath91 .
the function @xmath92 defined by eq .
( [ eq35-q2 ] ) has a discountinous first derivative at @xmath93 . in the vicinity of the singularity point
it can be represented in the form @xmath94 with @xmath95 and @xmath96 .
expanding @xmath97 and @xmath84 in powers of @xmath62 and substituting them into eq .
( [ eq46-rho ] ) we can find ( see eq . ( 27 ) of ref .
@xcite ) that the first term yields the dominant contribution to the reflection coefficient , with the result @xmath98 and @xmath99 applicability of the wkb theory implies that @xmath100 . similarly
( see also ref .
@xcite ) we can find the reflection coefficient at the electron density singularity formed in the above breaking regime discussed in part i @xcite . in this case the electron density distribution is given by eq .
( 106 ) of part i. using this relationship we obtain @xmath101}\left(\theta ( \zeta ) \sqrt{\zeta } -\theta ( \zeta-\delta \zeta ) \sqrt{\zeta-\delta \zeta}\right ) d\zeta } , \label{eq - rho - half - half}\ ] ] where @xmath102 @xmath103 and @xmath104 . calculating the integral ( [ eq - rho - half - half ] )
we find @xmath105 consequently , we write @xmath106 from eqs .
( [ eq - r-1 ] ) and ( [ eq - r - half - half ] ) we can see that in thermal plasmas the reflection coefficient is @xmath107 times larger in the above breaking regime than for a wake wave approaching the wavebreaking threshold
. generalizing eqs .
( 106 ) and ( 108 ) of part i @xcite , we can write the electron density dependence on the coordinate @xmath62 in the form @xmath108\ ] ] with @xmath109 and @xmath110 an even number , and @xmath111\ ] ] for @xmath110 an odd number .
it is easy to show that for the reflection coefficient , @xmath112 , we have @xmath113 if @xmath110 is even , and @xmath114 if @xmath110 is odd . comparing eqs ( [ eq - r - half - half ] , [ eq - r - odd ] , [ eq - r - even ] ) for the reflection coefficient with the corresponding coefficients obtained in ref .
@xcite , we find that the effects of a finate temperature enter eqs .
( [ eq - r - half - half ] , [ eq - r - odd ] , [ eq - r - even ] ) as a form - factor @xmath115 . in the limit
@xmath116 this form factor tends to unity while for @xmath117 decreases as @xmath118 .
since the frequency , @xmath119 , and the number of reflected photons , @xmath120 , are related to that incident on the relativistic mirror @xmath121 and @xmath122 as @xmath123 and @xmath124 , the energy of the relected photon beam is given by @xmath125 , where @xmath126 is the energy of the laser pulse incident on the mirror .
comparing @xmath127 with the energy of the electrons in the first period of the wake wave ( e.g. see @xcite ) , @xmath128 , where @xmath129 is the laser driver energy , we find that the photon back reaction ( the ponderomotive pressure ) on the wake wave can be neglected provided @xmath130 . as a typical reflection coefficient value we can take @xmath131 ( see refs.@xcite ) and obtain the condition of relative weakness of the incident laser pulse @xmath132 .
as we see owing to the weakness of the photon - wakewave interaction the incident pulse energy can be of the order of that in the driver laser pulse .
in the first part of our paper @xcite we found the structure of the typical singularities that appear in the electron density during the wave breaking in a thermal plasma .
the singularity in the electron density , moving along with the wake wave excited by a high intensity ultra - short pulse laser , can act as a flying relativistic mirror for counterpropagating electromagnetic radiation , leading to coherent reflection accompanied by the upshift of the radiation frequency .
this process implies finite ( not exponentially small ) reflectivity at the electron density singularities .
this is provided by the structure of the singularity formed in a relativistically large amplitude plasma wave close to the wavebreaking limit that leads to a refraction coefficient with discontinuous coordinate derivatives .
we found the reflection coefficients of an electromagnetic wave at the singularities of the electron density in the most typical regimes of strongly nonlinear wave breaking in thermal plasmas .
the efficiency of the photon reflection can be substantially increased by using the above breaking limit regimes which lead to the formation of high - order singularities .
we acknowledge support of this work from we acknowledge the support from the mext of japan , grant - in - aid for scientific research , 23740413 , and grant - in - aid for young scientists 21740302 from mext .
we appreciate support from the nsf under grant no .
phy-0935197 and the office of science of the us doe under contract no .
de - ac02 - 05ch11231 .
99 s. v. bulanov , t. zh .
esirkepov , m. kando , j. k. koga , a. s. pirozhkov , t. nakamura , s. s. bulanov , c. b. schroeder , e. esarey , f. califano , and f. pegoraro , _ phys .
plasmas _ ( 2012 ) - submitted for publication ; [ arxiv e - print : 2012arxiv1202.1903b ] .
g. g. stokes , _ trans .
cambridge philos .
_ * 8 * , 441 ( 1847 ) ; g. g. stokes , mathematical and physical papers , vol .
197 - 219 , ( cambridge , 1880 ) ; j. wilkening , _ phys .
* 107 * , 184501 ( 2011 ) .
r. camassa and d. d. holm , _ phys .
* 71 * , 1661 ( 1993 ) ; a. degasperis and m. procesi , in : symmetry and perturbation theory// eds .
a. degasperis and g. gaeta ( river edge , nj : world scientific , 1999 ) , pp .
23 - 37 .
s. v. bulanov , i. n. inovenkov , v. i. kirsanov , n. m. naumova , a. s. sakharov , _ sov .
rep . _ * 6 * , 9 ( 1991 ) [ _ kratk
. soobshch .
fiz . _ * 6 * , 9 ( 1991 ) ] ; 8 . s. v. bulanov , f. califano , g. i. dudnikova , t. zh .
esirkepov , i. n. inovenkov , f. f. kamenets , t. v. liseikina , m. lontano , k. mima , n. m. naumova , k. nishihara , f. pegoraro , h. ruhl , a. s. sakharov , y. sentoku , v. a.vshivkov , v. v. zhakhovskii , _ reviews of plasma physics _ , edited by v. d. shafranov ( kluwer academic / plenum , new york , 2001 ) , vol . 22 , p. 227 .
m. kando , y. fukuda , a. s. pirozhkov , j. ma , i. daito , l .- m .
chen , t. zh .
esirkepov , k. ogura , t. homma , y. hayashi , h. kotaki , a. sagisaka , m. mori , j. k. koga , h. daido , s.v .
bulanov , t. kimura , y. kato , and t. tajima , _ phys .
_ * 99 * , 135001 ( 2007 ) .
a. s. pirozhkov , j. ma , m. kando , t. zh .
esirkepov , y. fukuda , l .- m .
chen , i. daito , k. ogura , t. homma , y. hayashi , h. kotaki , a. sagisaka , m. mori , j. k. koga , t. kawachi , h. daido , s. v. bulanov , t. kimura , y. kato , and t. tajima , _ phys .
plasmas _ * 14 * , 123106 ( 2007 ) .
m. kando , a. s. pirozhkov , k. kawase , t. zh .
esirkepov , y. fukuda , h. kiriyama , h. okada , i. daito , t. kameshima , y. hayashi , h. kotaki , m. mori , j. k. koga , h. daido , a. ya .
faenov , t. pikuz , j. ma , l .- m .
chen , e. n. ragozin , t. kawachi , y. kato , t. tajima , and s. v. bulanov , _ phys .
lett . _ * 103 * , 235003 ( 2009 ) .
s. s. bulanov , t. zh .
esirkepov , f. f. kamenets , f. pegoraro , _ phys .
e _ * 73 * , 036408 ( 2006 ) ; s. s. bulanov , a. maximchuk , c. b. schroeder , a. g. zhidkov , e. esarey , w.p .
leemans , _ phys
. plasmas _ * 19 * , 020702 ( 2012 ) .
v. p. silin , _ sov .
. jetp _ * 11 * , 1136 ( 1960 ) ; b. kurunolu , _ nuclear fusion _ * 1 * , 213 ( 1961 ) ; a. b. mikhajlovskii , _ plasma phys . _ * 22 * , 133 ( 1980 ) ; d. b. melrose , _ aust . j. phys .
_ * 35 * , 41 ( 1982 ) ; j. bergman and b. eliasson , _ phys . plasmas _ * 8 * , 1482 ( 2001 ) .
v. a. mironov , a. m. sergeev , e. v. vanin and g. brodin , _ phys .
rev . a _ * 42 * , 4862 ( 1990 ) ; s. v. bulanov and a. s. sakharov , _ jetp lett .
_ * 54 * , 203 ( 1991 ) ; v. a. mironov , a. m. sergeev , e. v. vanin , g. brodin and j. lundberg , _ phys .
a _ * 46 * , r6178 ( 1992 ) ; j. t. mendonca and l. o. silva , _ phys .
e _ * 49 * , 3520 ( 1994 ) ; r. bingham , j. t. mendonca , j. m. dawson , _ phys .
lett . _ * 78 * , 247 ( 1997 ) ; l. o. silva and j. t. mendonca , _ phys .
e _ * 57 * , 3423 ( 1998 ) ; a. a. solodov , p. mora , and p. chessa , _ phys .
plasmas _ * 6 * , 503 ( 1999 ) ; a. spitkovsky and p. chen , _ phys . lett .
a _ * 296 * , 125 ( 2002 ) ; g. raj , m. r. islam , b. ersfeld , and d. a. jaroszynski , _ phys . plasmas _ * 17 * , 073102 ( 2010 ) . c. w. siders , s. p. leblanc , d. fisher , t. tajima , m. c. downer , a. babine , a. stepanov , and a. sergeev , _ phys .
lett . _ * 76 * , 3570 ( 1996 ) ; j.
m. dias , c. stenz , n. lopes , x. badiche , f. blasco , a. dos santos , l. o. silva , a. mysyrowicz , a. antonetti , and j. t. mendonca , _ phys .
lett . _ * 78 * , 4773 ( 1997 ) ; c. d. murphy , r. trines , j. vieira , a. j. w. reitsma , r. bingham , j. l. collier , e. j. divall , p. s. foster , c. j. hooker , a. j. langley , p. a. norreys , r. a. fonseca , f. fiuza , l. o. silva , j. t. mendonca , w. b. mori , j. g. gallacher , r. viskup , d. a. jaroszynski , s. p. d. mangles , a. g. r. thomas , k. krushelnick , and z. najmudin , _ phys .
plasmas _ * 13 * , 033108 ( 2006 ) ; r. m. g. m. trines , c. d. murphy , k. l. lancaster , o. chekhlov , p. a. norreys , r. bingham , j. t. mendonca , l .o silva , s. p. d. mangles , c. kamperidis , a. thomas , k. krushelnick , and z. najmudin , _ plasma phys .
control . fusion _ * 51 * , 024008 ( 2009 ) . | the structure of the density singularity formed in a relativistically large amplitude plasma wave close to the wavebreaking limit leads to a refraction coefficient which has a coordinate dependence with discontinuous derivatives .
this results in a non - exponentially small above - barrier reflection of an electromagnetic wave interacting with the nonlinear plasma wave . |
the swift increase of the number of absorptions ( and the average opacity ) with increasing redshift is the most impressive property of the ly-@xmath4 forest . fig .
[ dndz ] shows the number density evolution of the ly-@xmath4 lines @xcite in the column density interval has been chosen to allow a comparison with the hst key - programme sample at @xmath5 @xcite for which a threshold in equivalent width of 0.24 was adopted . ]
the long - dashed line is the maximum - likelihood fit to the data at @xmath7 with the customary parameterization : @xmath8 .
the uves @xcite observations imply that the turn - off in the evolution does occur at @xmath9 , not at @xmath10 as previously suggested .
-0.2 cm the evolution of the @xmath11 is governed by two main factors : the hubble expansion and the metagalactic uv background ( uvb ) . at high @xmath12 both the expansion ,
which decreases the density and tends to increase the ionization , and the uvb , which is increasing or non - decreasing with decreasing redshift , work in the same direction and cause a steep evolution of the number of lines . at low @xmath12 ,
the uvb starts to decrease with decreasing redshift , due to the reduced number and intensity of the ionizing sources , counteracting the hubble expansion . as a result
the evolution of the number of lines slows down . up to date numerical simulations
@xcite have been remarkably successful in qualitatively reproducing the observed evolution , however they predict the break in the @xmath13 power - law at a redshift @xmath14 that appears too high in the light of the new uves results .
this suggests that the uvb implemented in the simulations may not be the correct one : it was thought that at low redshift qsos are the main source of ionizing photons , and , since their space density drops below @xmath15 , so does the uvb .
however , galaxies can produce a conspicuous ionizing flux too , perhaps more significant than it was thought@xcite .
the galaxy contribution can keep the uvb relatively high until at @xmath9 the global star formation rate in the universe quickly decreases , determining the qualitative change in the number density of lines . under relatively general assumptions ,
it is possible to relate the observed number of lines above a given threshold in column density or equivalent width to the expansion , the uvb , the distribution in column density of the absorbers and the cosmology : @xmath16^{\beta-1 } h^{-1}(z ) , \label{eq : dndz}\ ] ] where @xmath17 is the photoionization rate and the @xmath18 distribution is assumed to follow a power - law of index @xmath19 , as discussed in the next section . -0.7
fig . [ nhi ] shows the differential density distribution function measured by uves @xcite , that is the number of lines per unit redshift path and per unit @xmath18 as a function of @xmath18 .
the distribution basically follows a power - law @xmath20 extending over 10 orders of magnitude with little , but significant deviations , which become more evident and easy to interpret if the plot is transformed in the mass distribution of the photoionized gas as a function of the density contrast , @xmath21 , @xcite : 1 ) a flattening at @xmath22 is partly due to line crowding and partly to the turnover of the density distribution below the mean density ; 2 ) a steepening at @xmath23 , with a deficiency of lines that becomes more and more evident at lower z , reflects the fall - off in the density distribution due to the onset of rapid , non - linear collapse : the slope @xmath19 goes from about @xmath24 at @xmath25 to @xmath26 at @xmath27 and recent hst stis data @xcite confirm that this trend continues at lower redshift measuring at @xmath28 a slope of @xmath29 ; 3 ) a flattening at @xmath30 can be attributed to the flattening of the density distribution at @xmath31 due to the virialization of collapsed matter .
hydrodynamical simulations successfully reproduce this behaviour , indicating that the derived matter distribution is indeed consistent with what would be expected from gravitational instability .
the last ingredient to be determined in eq . [ eq : dndz ] is the ionization rate . in a recent computation
@xcite we have investigated the contribution of galaxies to the uvb , exploring three values for the fraction of ionizing photons that can escape the galaxy ism , @xmath32 and @xmath33 ( the latter value corresponds to the lyman - continuum flux detected by @xcite in the composite spectrum of 29 lyman - break galaxies ) .
estimates of the uvb based on the proximity effect at high-@xmath12 and on the @xmath34 emission in high - latitude galactic clouds at low-@xmath12 provide an upper limit on @xmath35 , consistent with recent results on individual galaxies both at low-@xmath12 @xcite and at @xmath36 @xcite . introducing a contribution of galaxies to the uvb ,
the break in the ly-@xmath4 @xmath13 can be better reproduced than with a pure qso contribution @xcite .
the agreement improves considerably also at @xmath37 . besides , models with @xmath38 describe the flat evolution of the absorbers much better than @xmath39 .
a consistency check is provided by the evolution of the lower column density lines .
for @xmath40 the @xmath18 distribution is flatter , and according to eq .
[ eq : dndz ] this translates directly into a slower evolutionary rate , which is consistent with the uves observations@xcite : @xmath41 .
another diagnostic can be derived from the spectral shape of the uvb and its influence on the intensity ratios of metal lines @xcite .
given the cosmological scenario , the amount of baryons required to produce the opacity of the lyman forest can be computed @xcite and a lower - bound to the cosmic baryon density derived from the distribution of the ly-@xmath4 optical depths .
applying this approach to the effective optical depths measured in the uves spectra , the estimated lower bound @xmath42 is consistent with the bbn value for a low d / h primordial abundance .
most of the baryons reside in the lyman forest at @xmath43 with little change in the contribution to @xmath44 as a function of @xmath12 .
conversely , given the observed opacity , a higher uvb requires a higher @xmath45 .
as pointed out by @xcite , the large escape fraction measured by @xcite would result in an @xmath46 in strong conflict either with the d / h or in general with the bbn or with the ly-@xmath4 opacity measurements .
99 bianchi s. , cristiani s. , kim .
, 2001 , 376 , 1 dodorico s. , cristiani s. , dekker h. et al . , 2000 ,
spie 4005 , 121 dav r. , tripp t.m .
, 2001 , apj 553 , 528 deharveng j .-
m . , buat v. , le brun v. , et al . , 2001 , 375 , 805 giallongo e. , cristiani s. , fontana a. , dodorico s. , 2001 , in preparation haehnelt m.g . , madau p. , kudritzki r. , haardt f. , 2001 , 549 , l151 heckman t.m . , sembach k.r . , meurer g.r .
et al . , 2001 , 558 , 81 kim t .-
s . , cristiani s. , dodorico s. , 2001 , 373 , 757 kim t .-
carswell r.f . , cristiani s. , dodorico s. , giallongo e. , 2001 , mnras submitted savaglio s. , cristiani s. , dodorico s. et al . , 1997 , 318 , 347 schaye j. , astro - ph/0104272 steidel c.c . , pettini m. , adelberger k.l . , 2001 , 546 , 665 theuns t. , leonard a. , efstathiou g. , 1998 , mnras 297 , l49 weinberg d.h . , miralda - escud j. , hernquist l. , katz n. , 1997 , 490 , 564 weymann r.j .
, jannuzi b.t .
, lu l. et al . , 1998 , 506 , 1 | a sample of 8 qsos with @xmath0 has been observed with vlt / uves at a typical resolution @xmath1 and s / n @xmath2 .
thanks to the two - arm design of the spectrograph , a remarkable efficiency has been achieved below 400 nm and above 800 nm , which translates immediately in the possibility of obtaining new results , especially at @xmath3 .
we report here new insight gained about the evolution of the number density of ly-@xmath4 lines , their column density distribution , the ionizing uv background and the cosmic baryon density . |
antiferromagnetism ( af ) appears in the hubbard and related models for an arbitrary interaction strength @xmath0 provided we are close to the half - filled - band situation ( @xmath1 in the orbitally nondegenerate case ) @xcite .
this is easy to understand qualitatively , since the intraatomic hubbard interaction @xmath2 is diminished by keeping apart the electrons with the opposite spins . in the strong - correlation limit
@xcite this interaction leads to an antiferromagnetic kinetic exchange for an arbitrary band filling @xmath3 . at the same time , the band energy is not increased because the concomitant nesting condition for the quasiparticle states achievable for bipartite lattices only leads to the energy decrease of the occupied states , even when going beyond the hartree - fock picture . in effect , the regime of the band filling @xmath4 for which the af state is stable at given @xmath0 has been determined for variety of theoretical approaches @xcite - @xcite .
the reliability of the results for the half - filled case is not in question as long as they reduce to those in the hartree - fock and to the mean - field ( heisenberg ) approximations in the weak- and strong - correlation limits , respectively , as we will discuss in the following .
the theoretical results are in accord with the fact that _ all _ known mott insulators with the half - filled band configuration are also antiferromagnetic insulators .
the main purpose of this paper is to overview the situation by discussing the crossover behavior from the hartree - fock to the mean - field approximation for the heisenberg antiferromagnet .
similar , though not equivalent results can be reached within the dynamic mean - field theory ( dmft , see below ) . in the orbitally degenerate situation ( with the degeneracy @xmath5 ) the same type of magnetic(slater ) gap
is generated by an alternant orbital ordering @xcite , as is also discussed at the end of the paper .
explicitly , we concentrate our attention to two specific features of quasiparticle states not elaborated in detail so far , namely , ( i ) an evolution of the magnetic gap ( renormalized by the electronic correlations ) into the mott - hubbard gap , and ( ii ) a rather weak renormalization of the effective mass for the half - filled - band case , which is in contrast with that calculated in the paramagnetic ( para ) case @xcite . a rather strong mass enhancement is retained in the @xmath6 case , in direct analogy to the paramagnetic case @xcite . these results are obtained within the slave - boson approach in the saddle - point approximation , which we compare with the corresponding analysis in the infinite - dimension limit @xcite .
in particular , we introduce the concept of a nonlinear staggered molecular field , which shows up as the effective ( nonlinear - in - magnetization ) magnetic gap , evolving at temperature @xmath7 continuously with increasing @xmath0 from the slater ( hartree - fock ) gap @xmath8 into the hubbard gap . in connection with this evolution , we single out the magnetic and coulomb parts of the localization energy . these particular features
resolve explicitly the old question about the difference between the slater and mott - hubbard insulators in the sense that only the mott - hubbard gap survives when antiferromagnetism disappears ( i.e. above the nel temperature ) .
most of the results are contained in the mathematical formulation established earlier @xcite - @xcite . here , we only discuss those points in an explicit manner .
we believe that these points are relevant to the general physics of correlated systems .
this reason is also behind publishing , perhaps belatedly , such a simple paper .
additionally , since the effective mass is only weakly renormalized for the @xmath9 case and it never reaches zero for @xmath6 , the role of the quantum fluctuations is relatively not as crucial as for the discussion of the paramagnetic state .
the structure of the paper is as follows . in section
ii we provide the analytic details of the af solution , which are followed by the numerical analysis ( section iii ) and the discussion in physical terms in section iv , where we also compare the results with those for a doubly degenerate - band case for @xmath9 .
to stress the role of the molecular field coming from the electronic correlations , we start from the extended hubbard model , with intersite exchange interactions included , i.e. write @xmath10 where the labels @xmath11 and @xmath12 represent two interpenetrable sublattices , each containing half @xmath13 of available atomic sites .
the first term represents single - particle hopping of electrons between the sublattices ( nearest neighbors ) , the second and the third express the intraatomic interaction of the same magnitude on _ all _ sites , the fourth includes the heisenberg exchange between the sublattices , and @xmath14 is the reference energy with @xmath15 being the chemical potential , and @xmath16the total number of fermions .
the heisenberg term has been added only to provide an illustration of the concept of the exchange field coming from the electron correlations ( they will add to one another ) . in the mean - field ( saddle - point ) approximation for the slave bosons , the rotationally invariant approach of li et al . @xcite and
the original kotliar - ruckenstein @xcite formulations can be brought to an equivalent form @xcite .
explicitly , the six bosons @xmath17 and @xmath18 appearing in the approach fulfill the following five constrains in the mean - field approximation @xmath19 @xmath20 where the subscript @xmath21 characterizes the sublattices , and fermion operators @xmath22and @xmath23 represent the new ( quasiparticle ) fermion operators appearing in the theory .
the effective hamiltonian in the saddle - point approximation takes the form in reciprocal ( @xmath24 ) * * * * space @xmath25 where @xmath26 contains the constraints appearing in the theory in the form of lagrange multipliers @xmath27 + \notag \\ & \frac{n}{2}\lambda ^{(1b)}\left ( e^{2}+d^{2}+p_{b\uparrow } ^{2}+p_{b\downarrow } ^{2}-1\right ) - \notag \\ & \frac{n}{2}\left [ \lambda _ { \uparrow } ^{(2b)}\left ( p_{b\uparrow } ^{2}+d^{2}\right ) + \lambda _ { \downarrow } ^{(2b)}\left ( p_{b\downarrow } ^{2}+d^{2}\right ) \right ] \notag\end{aligned}\ ] ] the summation over ( @xmath24 * * ) * * comprises the states in the halved brillouin zone and @xmath28 is the number of nearest neighbors .
the lagrange multipliers @xmath29 and @xmath30 correspond to the constraint ( [ con1 ] ) and , since they are spin symmetric we can put @xmath31 the spin - dependent lagrange multipliers @xmath32 and @xmath33 represent spin - resolved constraint ( [ con2 ] ) . in the antiferromagnetic
( af ) case the magnetic moments @xmath34 @xmath35 have the opposite sign , i.e. @xmath36 this means that the constraints @xmath37 obey the relations @xmath38 the narrowing factor @xmath39 renormalizing the bare band energy @xmath40 assumes the form @xmath41 the hamiltonian ( [ hf ] ) can be diagonalized with the help of the bogolyubov transformation @xcite . in effect , the free energy functional of the system ( per atom ) takes the form @xmath42
-\frac{k_{b}t}{n}% \underset{\mathbf{k}\sigma}{\sum}\ln\left [ 1+e^{-\beta\left ( -e_{\mathbf{k}% } -\mu_{eff}\right ) } \right ] \quad \label{hb } \\ & + \frac{1}{2}jzm^{2}+ud^{2}+\mu\frac{n_{e}}{n}+h_{con } , \notag\end{aligned}\ ] ] where now @xmath43 here the quasiparticle energies are @xmath44 , with the magnetic gap @xmath45 , and the effective chemical potential @xmath46 with @xmath47 .
the quantity @xmath48 plays a role of the correlation induced molecular field , since it adds to the effective heisenberg field @xmath49 in the case @xmath50 ( taken in the numerical analysis ) @xmath51 constitutes the entire gap induced by the magnetic ordering ( it is the _
magnetic _ gap ) . on the whole , the first two terms in ( [ hb ] )
provide the contribution to the thermodynamics coming from the single particle excitations in the magnetic ( slater ) subbands caused by af ordering and having energies @xmath52 these quasiparticle energies comprise the effective band narrowing or the mass renormalization @xmath53 and the molecular field @xmath54 both to be determined in a self - consistent manner detailed below . the field @xmath51 arises from the local constraint ( [ con2 ] ) .
thus , one can say that the molecular field is induced by the correlations .
the functional ( [ hb ] ) must be minimized with respect to all appearing bose fields and the chemical potential @xmath55 effectively , one can reduce ( [ hb ] ) to the form with two variables @xmath56 and @xmath18 . for the purpose of simplicity and clarity of our points
we take constant density of states @xmath57 for @xmath58 for which the ground state energy takes the form @xmath59 where @xmath60 is the ground - state energy , and @xmath61 in this expression we have already connected @xmath62 to @xmath63via the relation @xmath64 for the sake of completness , we list also the explicit form of the band narrowing factor for @xmath9 : @xmath65 .\ ] ] note that the variable @xmath66 has a physical meaning of the ratio of the slater gap to the renormalized band energy . in other words , it provides a relative strength of the effective local field with respect to the renormalized kinetic energy . the growing ratio @xmath67 drives the system towards localization induced by the formation of staggered magnetic moments arranged in two sublattices , whereas the growing ratio @xmath68 drives the system towards localization independent of magnetic ordering .
therefore , the present formulation will allow us to single out the contributions coming from the two factors .
the magnetic energy is thus measured with respect to the band energy @xmath69 renormalized by the coulomb interaction .
we now show that the results obtained above provide correctly a mean - field solution in the @xmath72 limit . for the sake of simplicity
consider the half - filled band case with @xmath50 and for symmetric form of the density of states , @xmath73 .
we can then write the ground state energy ( per site ) in the form @xmath74 for large @xmath0 , the gap @xmath75 is also large . in that limit
the energy has the form @xmath76 where @xmath77 . minimizing this expression with respect to @xmath75 we
obtain that @xmath78 minimization with respect to @xmath62 and @xmath79 gives the relations @xmath80 and @xmath81 where @xmath82 . combining eqs.([m1])-([m3 ] )
we obtain that @xmath83 , and @xmath84 .
explicitly , from the fact that @xmath85 and that numerically @xmath86 , we obtain that @xmath87 , and @xmath88 . thus finally , for the featureless form of the density of states we have @xmath89 and @xmath90 in other words , in the @xmath72 limit the hubbard model reduces to the heisenberg model @xcite with the hubbard gap @xmath91 , since in the mean - field approximation the ground - state energy is then given by the kinetic exchange contribution @xcite @xmath92 with @xmath93 . also , the magnetic gap reduces to the atomic value of the hubbard gap , as @xmath94 with @xmath95 .
the hartree - fock @xmath96 limit is recovered once one notices that the method has been constructed to obtain @xmath97 in the weak coupling limit @xcite , @xcite . under that circumstance eq.([eg ] ) reduce to the usual hartree - fock form if we assume that now @xmath98 .
this limit was checked out also numerically , but the results are not reproduced here .
in fig.1 we have displayed both the effective magnetic ( slater - type ) gap @xmath99 and the mott - hubbard gap for the paramagnetic phase , both for @xmath9 [ ptb ] those characteristics are plotted for the ground state .
the chemical potential is then @xmath100 the mott- hubbard gap is expressed through the difference in the chemical potential in the paramagnetic case @xmath101 for @xmath102(the upper part ) and for @xmath103 ( the lower part ) and was discussed earlier @xcite . for @xmath9 the slater split - band picture appears for arbitrary small @xmath0 and @xmath104 increases with increasing @xmath105 , where @xmath106 . in the limit @xmath107 the gap is composed of the slater and the mott - hubbard parts , and when @xmath108 the former merges gradually with the latter .
this can be seen explicitly in fig.2 , [ ptbptb ] ffig2.eps where we have shown the ground state energy @xmath109versus @xmath110 in the strong - correlation limit the energy is determined by the kinetic - exchange contribution @xmath111 @xcite , as shown explicitly .
the fitted coefficient @xmath112 to the numerical results can not be attributed to any particular 3d structure , since we have used in numerical calculation a constant value of @xmath113 .
the energies of para- and antiferro - magnetic states merge for @xmath9 and in the @xmath114 limit .
the inset illustrates another interesting characteristic of the solution , namely the magnetic gap _ is not _
proportional to the magnetization , as one would expect from the hartree - fock ( slater ) solution .
in other words , the molecular field @xmath51 is a nonlinear function of @xmath115 since from the condition @xmath116 we obtain the relation @xmath117 @xmath118 this is the reason why the antiferromagnetic gap for almost localized fermions can not be regarded as just the slater gap . also , the af solution disappears at the critical band filling @xmath119 to visualize the difference between the magnetic gap and the magnetization we have plotted in fig.3 both quantities as a function of the interaction strength @xmath120 @xmath35for different band filling parameter @xmath121 [ ptb ] ffig3.eps while for @xmath9 the magnetic moment saturates gradually with growing @xmath122 @xmath104 keeps increasing .
the magnetic gap for @xmath9 increases and eventually @xmath123 ; this circumstance indicates again that the magnetic gap merges with the hubbard gap , which can be estimated analytically and is @xmath124 @xmath125 for @xmath126 .
the double occupancy probability @xmath127 is shown in fig.4 for different band fillings .
[ ptb ] ffig4.eps it decreases continuously with growing @xmath105 , i.e. the charge fluctuations are gradually suppressed , while the magnetic moment grows ( cf .
the difference in the behavior of @xmath128and @xmath129is caused by the circumstance that the @xmath79 is of intraatomic nature , whereas @xmath62 is determined from the competition between the magnetic energy @xmath130 ( also of intraatomic nature ) and the renormalized band energy @xmath131 .
the inset to fig.4 exemplifies the difference between the diminution of @xmath79 with growing @xmath105 for @xmath9 in two situations : for the paramagnetic ( para ) case @xmath132 for @xmath133 this feature is concurrent with the well - known effective mass divergence at the mott - hubbard localization boundary ( the _ brinkman - rice point _ ) .
this divergence _ does not emerge _ in the antiferromagnetic state as @xmath79 approaches zero gradually , in the same manner , as @xmath62 approaches saturation @xmath134 in fig.5 we have displayed the inverse band narrowing factor @xmath135 ( for @xmath136 ) which turns into the effective mass renormalization @xmath137 ( for @xmath138 ) as a function of @xmath139 and for different @xmath140 values .
[ ptb ] ffig5.eps again , the inset illustrates the difference with the @xmath9 case .
one should note that the enhancement factor in af state is very small compared to that in para state , which is equal to @xmath141 ^{-1}.\,\ $ ] the difference between af and para states diminishes with decreasing @xmath142as in that situation the magnetic moment is reduced rapidly .
so , the weak band narrowing in the @xmath9 case can be associated with the circumstance that the fermi level falls in the gap .
this is the reason why @xmath143 raises rather steeply around @xmath144 .
thus , the appearance of the itinerant antiferromagnetism with an isotropic gap changes the brinkman - rice scenario for the mott - hubbard transition , as it has been underlined before @xcite .
also , the physical parameters @xmath145 and @xmath146 are all of the same magnitude .
this is easy to envisage when estimating e.g. the band narrowing @xmath147 which is in the af state roughly @xmath148 and is of the order of unity .
in fig.6 we have displayed the stability regime @xmath149 of the af phase .
[ ptb ] ffig6.eps the full circles has been obtained @xcite in the limit of infinite dimension with the help of quantum monte carlo simulation .
one should note the range of the filling @xmath3 of stable af phase is the broadest for @xmath150 i.e. when the molecular field is the strongest ( cf .
fig.3 ) . note also that the monte - carlo results did note provide the asymptotic behavior for @xmath1 , as it does not reduce correctly to the hartree - fock limit .
for the sake of completeness we display in fig.7 the ground state energy as a function of @xmath105 , for different @xmath151close to the half filling .
[ ptb ] ffig7.eps in each case ( for @xmath138 ) the system contains the contribution @xmath152 for @xmath153and the contribution @xmath111 for @xmath154 in other words , the solution interpolates between the hartree - fock and the kinetic exchange limits .
moreover , the shift of the curves with diminishing @xmath155 @xmath3 @xmath35 in the @xmath70 limit is due to the presence of the band term proportional to @xmath156 one should note that in the paramagnetic state the direct coulomb interaction contribution @xmath157 competes with band energy @xmath158 in the antiferromagnetic state the local magnetic contribution @xmath159 is of the opposite sign than the coulomb part .
however , it changes also the band energy . to specify the role of the staggered field we have plotted in fig.8 the difference @xmath160 as a function of @xmath110 [ ptbptb ]
ffig8.eps the magnetic contribution exceeds the coulomb part for @xmath161 ( i.e. away from the hartree - fock limit ) , and is particularly strong ( and comparable to the band energy ) in the regime @xmath162 where the mott - hubbard localization would take place for the paramagnetic case .
this circumstance indicates why the metal - insulator transition present at @xmath7 in the paramagnetic phase is wiped out when the itinerant af sets in .
simply , the magnetic contribution @xmath163 freezes gradually the spatially separated electrons in the antiferromagnetic phase , taking also advantage of nonvanishing kinetic energy ( the kinetic exchange , @xmath164 , contribution ! ) . in the para phase we have instead a direct competition between band and coulomb energies . to summarize ,
the mott - localization is achieved gradually at @xmath7 in the af state . in other words ,
the present approach provides a crossover behavior from slater to mott insulator , as shown e.g. in fig.2 .
the same holds true even when we include the intersite exchange @xmath165 which may originate from the quantum gaussian fluctuations .
the continuous evolution with growing @xmath166does not preclude the first order transition at nonzero temperature , as has been demonstrated some time ago for the paramagnetic state @xcite and subsequently reconfirmed in the limit @xmath167 @xcite .
the correlated state builts in gradually with increasing @xmath105 , as can be seen from the gradual increase of the mass enhancement in @xmath168case , i.e. when the ground state is always metallic .
the same holds true for the ground state energy shown in fig.7 .
we have addressed the question of crossover from slater ( hartree - fock ) to mott - hubbard ( atomic ) picture in the half - filled band case , as well as have analyzed the behavior of quasiparticle physical properties in af state in the half- and non half - filled - band cases . although our analysis is based on the saddle - point solution within the slave - boson functional - integral approach , the results can serve as a mean - field analysis , since they interpolate between those in the hartree - fock approximation in the limit @xmath169 and those in mean - field approximation for the heisenberg model ( for @xmath170 in the @xmath72 limit .
they also represent basis for inclusion of gaussian fluctuations @xcite in a magnetically ordered ( af ) state close to the mott - hubbard localization .
however , one should realize that for the half - filled band case the effective mass is only weakly renormalized so the renormalization factor @xmath171 . also , for @xmath172 @xmath173 remains always finite so , perhaps the role of the quantum fluctuations is not as crucial for af state , as it is for the paramagnetic state . in any case , it will be much more involved that in the paramagnetic case @xcite . the full analysis of the mott - hubbard boundary should also include the disordered local - moment phase @xcite , so far not included within the present scheme .
the physical meaning of the results obtained within the slave - boson approach ( sba ) can be characterized as follows .
it is well known @xcite , @xcite - @xcite , @xcite that the saddle point solution of the approach reproduces the results of the gutzwiller approach ( ga ) , as far as the overall ( ground - state ) behavior is concerned .
moreover , sba leads to an improvement of the ga by incorporating both the quasiparticle picture of those systems and the quantum gaussian fluctuations ( not considered in this paper ) . in general , the pseudo - fermion fields @xmath174 representing the quasiparticle states are in one - to - one correspondence to the original fermion fields in the physical fock space .
the mapping is quite obvious within the above analysis .
specifically , the quasiparticle energy @xmath175 leads to the following density of states @xmath176 where @xmath177is the bare density of states in pm phase .
thus the enhancement factor due to the correlations is distinct ( and disappear in the first factor ) from the change of the density of states caused by the appearance of the magnetic gap . by analogy , for the paramagnetic state ( @xmath178 ) , for which @xmath179 can be written as @xmath180 and where the self - energy @xmath181 leads to the mass enhancement @xmath182 we can not single out the factor ( @xmath183 ) in ( [ ro ] ) as the corresponding enhancement also in af state . in the doubly - degenerate - band case ( for @xmath9 ) and under the same - type of approximation scheme , the role of the magnetic gap is played by the gap formed by an alternant orbital ordering in the ferromagnetic state @xcite .
the gap @xmath184 in the latter case is reproduced in fig.9 for the magnitude of the intraatomic ( hunds - rule ) exchange @xmath185 .
again , the gap approaches asymptotically the mott - hubbard gap , which in this case @xcite is equal to @xmath186 .
however , here there is a critical value of @xmath68 , at which the system exhibits antiferromagnetic orbital ordering . in the half - filled case for the doubly degenerate case
@xmath187 the gap appears for an arbitrary small @xmath0 @xcite in the af case and for the critical value of @xmath0 for the paramagnetic phase @xcite .
one should also note that within the present sb scheme the antiferromagnetic slate - type state ( afs ) evolves for @xmath9 gradually into antiferromagnetic mott ( afi ) state with increasing @xmath68 ratio .
for @xmath138 , we observe only antiferromagnetic metallic ( afm ) state .
this is in contrast with the results obtained with the composite - operator method @xcite , as well as with an original analysis @xcite where a phase border line between itinerant- and localized - staggered moment - bearing phases is drawn .
localized - moment phase for @xmath188 @xmath189 can be obtained only when the polaronic effects due to antiferromagnetic moments surrounding the hole in the mott insulator are included @xcite .
obviously , such asf - afi border line at temperature @xmath190 ( for @xmath9 ) is present @xcite and is induced by the difference in entropy of afs and afi states .
* acknowledgment * j.s .
is grateful to claudine lacroix and m. avignon for discussions during his stay at drfmc - ceng in grenoble .
this research was supported by kbn grant no .
2 p03b 050 23 .
two of the authors ( ml and js ) acknowledge also the support of the project franco - polonais polonium in the period 1998 - 2000 .
99 see : j. des cloizeaux , j. phys
. radium . *
20 * , 606 ( 1959 ) , and d.r .
penn , phys .
rev.*142 * , 350 ( 1966 ) , - for the stability of antiferromagnetic ( af ) phases in the hartree - fock ( hf ) approximation .
the stability of af in the present context ( i.e. in the gutzwiller approximation ) has been obtained in : k.kubo and m. uchinami , prog .
. phys . * 54 * , 1289 ( 1975 ) . for a comparative analysis
see : e.g. a.m. ole and j. spaek z. phys.b * 44 * , 177 ( 1981 ) ; see also : g. seibold et al .
b * 57 * , 6937 ( 1998 ) .
the phase diagram recovering both hf and strong correlation limits correctly has been discussed in g. kotliar and a.e .
ruckenstein , phys .
* 57 * , 1362 ( 1986)-within the slave - boson approach ; and in : w. metzner and d. vollhardt , phys .
* 62 * , 324 ( 1989 ) ; p. fazekas , b. menge and e. mler - hartmann , z. phys .
* 78 * , 69 ( 1990)-in the limit of infinite dimension @xmath191 .
the corresponding discussion in two - spatial dimensions was performed in : e. arrigoni and g.c .
strinati , phys .
b**44 * * , 7455 ( 1991 ) ; w. ziegler et al . , phys .
b * 53 * , 1231 ( 1996 ) . in that respect
the af stability should be considered for @xmath192 against the onset of the ferromagnetism , for which the pauli principle plays the same role as @xmath72 for antiferromagnetism . however , in the large-@xmath0 limit , the virtual hopping processes , particularly for @xmath9 , lead to a stable af state ; see e.g. j.spaek , a.m. ole , and k.a .
chao , phys .
solidi ( b ) * 108 * , 329 ( 1981 ) .
cf . also : k. a.chao , j. spaek , and a.m. ole , j. phys . c**10 * * , l271 ( 1977 ) , where the effective kinetic exchange has been derived in the @xmath193 limit for an arbitrary @xmath3 .
asymptotic values of the magnetic moment and the molecular - field parameter @xmath194 are discussed in : b. mller et al .
, j. phys . : condens .
matter * 5 * , 4847 ( 1993 ) .
m. jarrell , phys .
lett . * 69 * , 168 ( 1992 ) ; m. jarrell and t. pruschke , z.phys . b**90 * * , 187 ( 1993 ) . for the comparison the men - field slave - boson and the quantum monte - carlo approaches see : l. lilly et al .
* 65 * , 1379 ( 1990 ) .
the gaussian fluctuations in para state are treated in ref .
6 , and in : p. wlfle and t. li , z. phys .
* 78 * , 45 ( 1990 ) ; r. raimondi and c. castellani , phys .
rev . b**48 * * , 11453 ( 1993 ) . for recent treatment
see : r. frsard and t. kopp , nucl .
b * 594 * , 769 ( 2001 ) ; a. tandon et al .
lett . * 83 * , 2046 ( 1999 ) .
these effects are considered mainly in the context of large - u situation for a two dimentional case , cf .
e.g. p. wrbel and r. eder , phys .
b**58 * * , 15160 ( 1998 ) , and references therein ; cf .
also : m. imada et al . ,
70 * , 1039 ( 1998 ) .
the slater gap @xmath195 and the hubbard gap as the functions of interaction strength .
the slater gap merges with the mott - hubbard gap as @xmath196 * fig.2 .
ground state energy for af state and @xmath9 versus @xmath197 the lower curve is the fit to the expression @xmath198the inset displays the difference in behavior of magnetic moment @xmath62 and half of the slater gap @xmath199 both plotted as a function of the band filling .
the slater gap parameter @xmath200(top panel ) and the magnetic moment @xmath201 ( bottom ) versus @xmath105 and for different values of @xmath121 * fig.4 .
the double occupancy @xmath202 vs @xmath203and for the different @xmath3 values.the inset display the difference in behavior for para- and antiferro - magnetic cases for @xmath204 * fig.5 . the effective mass enhancement @xmath205 ( with respectto the band value @xmath206 ) vs @xmath105 and for the @xmath3 values shown .
the inset shows a rather weak enhancement close to the mott - hubbard limit . *
fig.6 . the stability regime of af solution ; the inset : results ofmonte - carlo calculations in the@xmath207limit @xcite .
ground state energy vs @xmath208for different band filling @xmath209 .
in each case the energy is @xmath152 for @xmath210 and @xmath211 for @xmath212 * fig.8 .
comparison of the magnetic @xmath213and the coulomb @xmath214terms in the ground state energy versus @xmath110 * fig.9 .
the half - gap parameter @xmath215 ( in units of the bare bandwidth @xmath216 ) vs @xmath68 for a quarter - filled doubly - degenerate band .
the hund s rule exchange is @xmath217 . | we supplement ( and critically overview ) the existing extensive analysis of antiferromagnetic solution for the hubbard model with a detailed discussion of two specific features , namely ( i ) the evolution of the magnetic ( slater ) gap ( here renormalized by the electronic correlations ) into the mott - hubbard or atomic gap , and ( ii ) a rather weak renormalization of the effective mass by the correlations in the half - filled - band case , which contrasts with that for the paramagnetic case .
the mass remains strongly enhanced in the non - half - filled - band case .
we also stress the difference between magnetic and non - magnetic contributions to the gap .
these results are discussed within the slave boson approach in the saddle - point approximation , in which there appears a _ non - linear _ staggered molecular field due to the electronic correlations that leads to the appearance of the _
magnetic gap_. they reproduce correctly the ground - state energy in the limit of strong correlations .
a brief comparison with the solution in the limit of infinite dimensions and the corresponding situation in the doubly - degenerate - band case with one electron per atom is also made .
pacs nos .
71.10.fd , 75.10.lp , 75.50.ee |
one of the most mysterious of our milky - way galaxy is the `` missing satellite problem '' that cold dark matter scenario predicts its number of satellite galaxies to be around hundred , but only dozens are observed ( e.g. moore et al . 1999 ) .
a few solutions have been proposed for this contradictory .
one instant question is if we are comparing the right things , i.e , how to relate the observed stellar velocity dispersion to the measured circular velocity from n - body simulation ( hayashi et al .
another possible solution is that photoionization suppress the gas accretion in small halos , and only halos formed before reionization can form stars .
( e.g. gnedin 2000 ) .
also there is worry about the incompleteness of observation as very faint satellites have too low surface brightness to be observed . in recent years , more faint satellites are observed along with their kinematics and mass measurements , and the satellite luminosity function is also well determined from the sdss ( koposov et al .
one the other hand , theoretical modelling of dark matter halo formation history ( cole et al .
2008 ) and galaxy merging time - scales ( jiang et al . 2008 , boylan - kolchin et al .
2008 ) are also improved .
given these progress , it is deserved to revisit the `` missing satellite problem '' and there have been a few papers to address this ( e.g. simon & geha 2007 ) . here
we use the semi - analytical model of galaxy formation ( kang et al .
2005 ) to predict the luminosity function , mass - to - light ratios for satellite galaxies and compare them with the data .
one of the main ingredients to model the satellite galaxy population is to produce their formation and assembly history . here
we use the monte - carlo merger tree code from parkinson et al .
( 2008 ) to obtain the formation history of a milk - way type galaxy with mass around @xmath4 .
this new monte - carlo algorithm is still based on the eps formula , but revised to match the millennium simulation ( springel et al .
2005 ) results at different mass scales .
cole et al .
( 2008 ) have shown that for halo with mass around @xmath4 , the merger tree from the previous monte - carlo algorithm ( cole et al .
2000 ) produces too strong evolution and too many major mergers at lower redshift .
we produce 1000 realizations of the merger trees using the new monte - carlo algorithm , and in fig.1 we show their formation history with comparisons to the n - body simulation results ( stewart et al .
2008 , giocoli et al .
it can be seen that both the evolution of the main progenitors and the mass function of accreted subhalos agree well with the simulation results . at z=0 .
left panel : the mass evolution of the main progenitors .
right panel : the mass function of accreted subhalos by the main progenitors .
good match are found with n - body simulations ( stewart et al .
2008 , giocoli et al .
2008).,scaledwidth=100.0% ] we then use the semi - analytical model of kang et al .
( 2005 ) to model the formation of satellite galaxies along the merger trees .
the semi - analytical model includes a few important physical processes governing galaxy formation : hot gas cooling in halos , star formation from cold gas , supernova feedback to reheat the inter - stellar medium , stellar evolution , galaxy merger .
we update the model of kang et al .
( 2005 ) by using an improved fitting formula for the galaxy merging time - scales from jiang et al .
( 2008 ) , who have shown that for massive mergers , the survival time of satellite galaxies in sph simulation is longer than the prediction from lacey & cole ( 1993 ) .
here we also include a simple model for photoionization from kravtsov et al .
( 2004 ) . in case of heating from ionized photons , the gas content in halos formed after reionization
are suppressed and can be simply described by a filter mass which increases with redshift .
the filter mass increase from @xmath5 at z=8 to @xmath6 at z=0 ( okamoto et al .
( 2008 ) recently argue that the filter mass should be smaller ) .
the gas fraction in a filter mass halo is about half of the universal fraction .
with photoionization the amount of gas available for star formation is decreased significantly in less massive halos formed after reionization . in this paper , we take the reionization redshift as z=7 .
in fig.2 we show the model luminosity function ( lf ) of satellites with comparison to the recent results of koposov et al .
( 2008 ) from sdss dr5 .
koposov et al measured the lf up to @xmath7 , and found that lf can be described by a power law with slope of 0.1 .
at the very faint end ( @xmath8 ) the solid circle points in fig.2 are extrapolated assuming the satellite galaxies following a nfw density distribution , and empty circles are assumed with an isothermal distribution ( see koposov et al .
it can be seen that if there is only supernova feedback ( dashed line ) , the predicted total number of satellites are more than observed by a factor of 3 . with the suppression of gas accretion by photoionization , the lf ( solid line ) can be brought into abroad agreement with the data .
this is expected that the decrease of gas content produce less stellar mass .
compared to the model prediction of benson et al .
( 2002 ) , our model produces more luminous satellites with @xmath9 .
this success is credited to the combination of improved models for halo merger tree and galaxy merging time - scales .
the merger tree used by benson et al .
( 2002 ) is based on cole et al .
( 2000 ) , which produces too many recent major mergers . as the galaxy merging time is shorter for major merger ,
so less is the number of survived massive satellites .
also we use the new fitting formula from jiang et al .
( 2008 ) for galaxy merging time - scales , which is longer than the often used dynamical friction time scale from lacey & cole ( 1993 ) . as we can see that without photoionization , there are only a few satellites fainter than @xmath10 .
this is because hot gas can not cool via hydrogen line emission in halos with virial temperature below @xmath2 ( @xmath11 ) and
@xmath12 cooling is very inefficient .
the solid line shows that those faint satellites formed in halos with virial temperature just over @xmath2 , but have their gas content strongly suppressed by photoionization ( similar conclusion was also obtained by kravtsov et al . 2004 ) . in our model , it is difficult to produce satellite ( @xmath13 ) with number around 30 , and this favors the satellites to have an isothermal density distribution . with the advent of accurate measurements of satellites kinematics , it is possible to obtain the total mass of satellite galaxies inside their luminous radii .
it is found that most satellites are dominated by dark matter inside their luminous radii .
here we show the mass - to - light ratio of satellites in fig.3 , where the data points are from compilation by simon & geha ( 2007 ) .
we model the dark matter mass evolution of satellite galaxies using the model of giocoli et al .
( 2008 ) , and we further assume that about 5% of the total dark matter mass are inside their luminous radii . the model prediction ( solid line ) along with 20th and 80th percentiles of the distribution ( dashed lines ) are shown in fig.3 .
we can find good agreement with the data from @xmath14 up to @xmath15 . ) .
dashed line is their mass distribution before accretion , and solid line is the distribution of their present mass after evolution .
the distribution has a sharp peak at @xmath16 before accretion , where hydrogen line emission cooling is efficient .
the present mass has a broad distribution around @xmath17 , and the wide spread is from the dispersion of accretion times .
, scaledwidth=50.0% ] the above results of lf and mass - to - light ratio are encouraging as they indicates that we can model the luminosity and mass of satellites simultaneously .
now we make prediction for the dark matter mass of satellites faint than @xmath18 . in fig.4 , we show their host halo mass before accretion ( dashed line ) and the present dark matter mass after evolution ( solid line ) . as we can see that the faint satellites ( @xmath19 ) formed in halos with mass peak at @xmath16 , and they have a broad distribution for their present mass with peak at about @xmath17 .
the broad distribution is from the spread of their accretion times .
we revisit the `` missing satellite problem '' using a semi - analytical model of galaxy formation combined with a high - resolution merger tree from monte - carlo algorithm ( parkinson et al .
we model the luminosity function and mass - to - light ratio for the satellites .
the model luminosity function agrees well with the recent results of koposov et al .
( 2008 ) from the sdss dr5 only if the photoionization effect is included to suppress the gas fraction in less massive halos .
our ability to produce more luminous satellite galaxies than previous semi - analytical models ( e.g. , benson et al .
2002 ) is from the improvement on modelling of halo merger tree and galaxy merging time - scales .
very faint satellites ( @xmath20 ) form in halos with virial temperature above @xmath2 , but their gas content are strongly suppressed by photoionization .
in addition their number density favors an isothermal density distribution in the milky - way .
we model the mass evolution of subhalos using the model of giocoli et al .
( 2008 ) , and find that the measured total kinematic mass inside the luminous radii of satellite galaxies are about 5% of their present dark matter mass .
i thank nicolas martin , jeta t. de jong and simon white for helpful discussions . | we revisit the milky way satellite problem using a semi - analytical model of galaxy formation and compare the predicted luminosity function to recent result from the sdss . with cosmic photoionization , the luminosity function can be brought into broad agreement with the data between @xmath0 .
this improvement over previous semi - analytical model results ( e.g. , benson et al .
2002 ) is from our adoption of improved models for galaxy merger history and galaxy merging time - scales .
the very faint satellites ( @xmath1 ) formed in halos with virial temperature over @xmath2 ( mass around @xmath3 before accretion ) , but their baryon content are strongly suppressed by photoionization .
we model the mass evolution of the subhalos , and compare the predicted mass - to - light ratio with the data .
we find that the measured total mass inside the luminous radii of satellites are about 5% of their present total dark matter mass . |
two `` modes '' of star formation are observed in ( optically or uv selected ) starburst galaxies ( e.g. meurer 1995 ) : a young unresolved population responsible for emission of diffuse uv light ( meurer 1995 , also calzetti these proceedings ) , and compact stellar clusters , losely termed super star clusters ( sscs ) hereafter .
sscs have been the focus of numerous recent studies related in particular to the possibility that these clusters may represent the progenitors of globular clusters ( cf .
fritze von alvensleben , miller , these proceedings ) .
a different aspect is emphasized in the present work .
we use spectroscopic observations of young star forming ( sf ) regions to determine their massive star content with the aim of providing constraints on stellar evolution models and the upper end of the imf .
sscs and similar compact young sf regions have the following properties : _ a ) _ numerous such objects are known . _
b ) _ they represent clusters rich enough ( @xmath0 10@xmath1 o stars ) such that the imf can be well populated and stochastical effects ( cf .
lanon these proceedings ) are negligible . _
c ) _ a priori the clusters cover a wide range of metallicities , and _
d ) _ consist likely of a fairly coeval population . given these properties ,
sscs resemble `` normal '' galactic of local group clusters which represent fundamental test - cases for stellar evolution .
the only disadvantage is that their stellar content can only be studied through observations of their integrated light . on the other hand b ) and c ) represent important advantages for studies focussed on massive stars over using `` local '' clusters .
_ this shows that young sscs provide ideal samples for studies of massive star evolution in different environments , such as e.g. extreme metallicities largely inaccessible in local group objects . _ after a brief introduction on the type of objects used here ( wolf - rayet rich sf region ) we will summarise recent work along these lines .
we will concentrate on the so - called wolf - rayet ( wr ) galaxies ( cf.schaerer 1999b for the latest catalogue ) , which are objects where broad stellar emission lines ( called `` wr bumps '' , mostly at and ) in the integrated spectrum testify to the presence of wr stars . for the study of massive star populations
these objects are ideal since wr stars are the descendents of the most massive stars in a short - lived phase ( @xmath2 , @xmath3 yr ) .
their detection is also a good age indicator for young systems ( @xmath4 myr ) , and allows good measure of the burst duration and the best direct probe of the upper end of the imf .
an overview of studies on wr populations in starburst regions can be found in the reviews of schaerer ( 1998 , 1999 ) . in the context of the present workshop
it is important to note that the objects broadly referred to as wr `` galaxies '' are found among a large variety of objects including bcd , massive spirals , iras galaxies , seyfert 2 , and liners ( see schaerer 1999b ) .
the `` wr rich '' regions contained in the spectroscopic observations will thus in general cover quite a large scale of sizes , different morphologies etc . in the case of blue compact dwarfs ( bcd ) , one is however mostly dealing with one or few individual compact regions or ssc dominating the observed light .
although this statement can not , with few exceptions , be quantified so far for the objects studied below ( but see e.g. conti & vacca 1994 ) we will mostly assume that the spectroscopic observations correspond closely to light from one young compact sf region or ssc .
the spectroscopic sample of dwarf galaxies from izotov , thuan and collaborators , obtained for the main purpose of determining the primordial he abundance and other abundance studies , has proven to be very useful for analysis of massive star populations especially at very low metallicities . indeed , @xmath0 20 wr rich regions are found in this sample at metallicities below the smc ( @xmath5 o / h @xmath6 8.1 ) extending to i zw 18 with @xmath0 1/50 solar metallicity .
no _ bona fide _ massive star of such low a metallicity is known in the local group ! the analysis of the wr and o star content in these objects has been presented by guseva et al .
( 1999 , hereafter git99 ) .
some of their main results are summarised in fig . 1 , which shows ( left panel ) the derived wr/(wr+o ) number ratio as a function of metallicity from their objects and observations of kunth & joubert ( 1985 ) , vacca & conti ( 1992 ) , and schaerer et al . ( 1999a , hereafter sck99 ) .
the left fig . considerably extends the previous samples ( cf .
vacca & conti 1992 , meynet 1995 ) .
the trend of increasing wr / o with metallicity is well understood ( arnault et al .
1989 ) the comparison with appropriate evolutionary synthesis models ( schaerer & vacca 1998 , sv98 ; shown as solid lines ) calculated for a `` standard '' salpeter imf with @xmath7 and using the high mass loss geneva tracks shows a good agreement .
this and more direct comparisons of the observed wr features ( see schaerer 1996 , de mello et al .
1998 , sck99 , git99 ) indicate that the bulk of the observations are compatible with short ( `` instantaneous '' ) bursts with a salpeter imf extending to large masses .
the short burst durations derived by sck99 for the metal - poor objects are also in agreement with the study of mas - hesse & kunth ( 1999 ) . of particular interest for evolutionary models is the relative number of wr stars of the wn ( exhibiting h - burning products on their surface ) and wc subtypes ( he - burning products ) .
the relative lifetimes vary strongly with initial mass and metallicity and are sensitive to various mass loss prescriptions and mixing scenarios currently not well known ( see maeder & meynet 1994 , meynet these proceedings ) .
the recent high s / n spectra of sck99 and git99 have now allowed to establish number ratios of wc / wn stars in a fair number of wr rich regions .
the determinations of git99 are shown in fig . 1 ( right panel ) ;
similar values are derived by sck99 .
the comparison with synthesis models shows a reasonable agreement . to reproduce sufficiently large wc / wn ratios the use of the stellar tracks based on the high mass loss prescription are , however , required as shown by de mello et al .
( 1998 ) and sck99 .
it is understood that part of the `` requirement '' for the high mass loss ( cf .
schaerer 1998 ) may be compensated by additional mixing processes leading to a similar prolongation of the wr phase ( cf .
meynet , these proceedings ) . in any case
in addition to the well known stellar census in the local group ( cf .
maeder & meynet 1994 and references therein ) the present new data from integrated populations place important constraints on the evolutionary models which have to be matched by successful stellar models .
especially the new studies extend the range of available metallicities to very low @xmath8 , well beyond the smc .
a small sample of metal - rich ( o / h @xmath9 solar ) starbursts ( 4 objects from git99 and mrk 309 ) have recently been analysed in detail by schaerer et al .
( 2000 ) . in this case
the observations ( high s / n , intermediate resolution optical spectroscopy ) correspond to compact nuclear sf regions .
despite this complication we use these objects as a first step to probe the upper end of the imf at high - metallicity .
subsequent studies of more isolated and simple , cluster - like objects will be undertaken in the future . for our comparison with evolutionary synthesis models ( cf .
below ) we use following main observational constraints : h@xmath10 and h@xmath11 equivalent widths serving as age indicator , h@xmath11/h@xmath10 determining the gaseous extinction , intensities and equivalent widths of the main wr features ( 4650 bump , ) , tio bands at @xmath0 6250 and 7200 indicating the presence of red supergiants from a population with ages @xmath9 710 myr , and the overall sed provide an important constraint on the population responsible for the continuum .
model calculations have been done using the sv98 synthesis code .
the basic model parameters we consider are : stellar metallicity , imf slope and upper mass cut - off , star formation history ( instantaneous or extended bursts , constant sf ) , fraction of lyc photons absorbed by the gas , stellar extinction ( which may differ from gaseous ) .
the comparison of the observed and predicted wr features is shown in fig . 2 .
the observations are well reproduced by z=0.02 models with a `` standard imf '' ( salpeter , m@xmath12120 ) assuming extended burst durations of @xmath13 410 myr .
the longer burst durations found here are physically plausible and likely explained by the different nature of the metal - rich objects analysed here ( larger nuclear regions vs. compact clusters in bcd , cf .
the corresponding ages of our objects , as indicated by w(h@xmath10 ) , are between @xmath0 7 and 15 myr , also in agreement with the presence of the tio bands .
a very good fit is also obtained to the overall sed .
this requires , however , an extinction of the stellar continuum which is less than that derived from the gas .
the differences are of similar amount that found by other methods in other studies ( e.g. calzetti 1997 , mas - hesse & kunth 1999 ) . in short
, we conclude that all the given observational constraints can be well reproduced by models with a salpeter imf extending to high masses for a burst scenario with star formation extending over @xmath0 410 myr .
this solution is not unique .
therefore a variety of alternate models have been considered ( cf .
schaerer et al .
regarding the imf we find that steeper ( with slope @xmath0 miller - scalo ) imfs are very unlikely . in view of several studies indicating a possible lack of massive stars in metal - rich environments ( e.g. bresolin et al . 1999 ,
goldader et al .
1997 ) we have used the present set of metal - rich wr galaxies to put a _ lower limit _ on the value of m@xmath14 from the strength of the wr features .
as mentioned above our data is compatible with a large upper mass cut - off .
the real range of values our data is sensitive to is , however , limited ; intrinsically younger regions need to be sampled to probe the most massive stars . adopting a conservative approach ( cf .
3 ) we obtain m@xmath15 30 ( for a salpeter slope ) .
we also find similar values ( m@xmath16 3550 ) from comparisons of h@xmath10 equivalent width measurements in metal - rich regions ( see schaerer et al .
2000 ) .
in contrast with previous studies of metal - rich starbursts and related objects based on properties of the ionized gas our work provides first constraints on the upper end of the imf measured directly from stellar signatures . future work on larger samples and using detailed coupled stellar population and photoionisation models should be of great interest .
arnault , p. , kunth , d. , schild , h. 1989 , , 224 , 73 calzetti , d. , 1997 , in `` the ultraviolet universe at low and high redshift : probing the progress of galaxy evolution '' , eds .
waller et al .
, aip conference proceedings , v.408 . , 403 conti , p.s . , vacca , w.d . , 1994 , , 423 , l97 de mello , d.f .
, schaerer , d. , heldman , j. , leitherer , c. , 1998 , , 507 , 199 goldader , j. , et al . , 1997 , , 474 , 104 guseva , n.g . , izotov , y.i . ,
thuan , t.x . , 1999 , , in press ( git99 ) kunth , d. , & joubert , m. 1985 , , 142 , 411 maeder , a. , meynet , g. , 1994 , , 287 , 803 mas - hesse , j.m . ,
kunth , d. , 1999 , , in press meurer , g. , et al . , 1995 , , 110 , 2665 meynet , g. 1995 , , 298 , 767 schaerer , d. , 1996 , , 467 , l17 schaerer , d. , 1998 , in `` wolf - rayet phenomena in massive stars and starburst galaxies '' , iau symp .
193 , 539 , schaerer , d. , contini , t. , kunth d. , 1999a , , 341 , 399 ( sck99 ) schaerer , d. , contini , t. , pindao m. , 1999b , , 136 , 35 schaerer , d. , guseva , n. , green , r. , izotov , y.i . ,
thuan , t.x . , 2000 ,
, in preparation schaerer , d. , vacca , w. d. 1998 , , 497 , 618 ( sv98 ) vacca , w.d . , conti , p.s . , 1992 , , 401 , 543 * q. ( manfred pakull ) : * some years ago it was thought that wrs do not form in very low z environments because the massive stellar winds were too weak .
is that still the state of the art in stellar evolution models ? or are the broad emission features due to a population of h - rich , very young objects like in 30 dor and ngc 3603
? * a. ( daniel schaerer : * indeed the metallicity dependence of mass loss causes an important decrease of the wr population toward low z. despite this evolutionary models applying recent mass - loss prescriptions predict _ some _ wr stars at the lowest metallicities ( 1/50 solar ) corresponding to i zw 18 ( see de mello et al .
1998 ) , quite in agreement with the observations .
the role of other formation channels ( massive close binaries , rotation ) at these low z remains unexplored in schaerer et al .
( 1999 , sck99 ) we have explored the effect of such r136-like wr stars during h - burning in addition to other wr and of stars with broad emission lines . given their high initial mass their integrated contribution to the wr bump should in most cases not be very important compared to `` normal '' wr stars .
an exception could be very young ( @xmath17 myr ) and strongly coveal populations .
few such observational cases seem , however , known to date ( see sck99 ) . | young super star clusters and young compact massive star forming regions can provide useful information on their burst properties ( age , burst duration , sfr ) , the upper end of the imf and yield new constraints on the evolution of massive stars . through the study of their stellar populations we can
in particular extend our knowledge on massive stars to extreme metallicities unavailable for such stars in the local group .
here we summarise the main results from recent studies on two metallicity extremes : wolf - rayet and o star populations in very metal - poor bcd , and metal - rich compact nuclear sf regions . |
recent observations ( and especially with hst ) demonstrate that mutual interactions and mergers between galaxies at early stages of evolution of the universe were probably among the main processes leading to the observed properties of galaxies ( e.g. , keel 1996 ) . even at the present epoch , at least 5 - 10% of galaxies are members of interacting systems .
many other galaxies keep signs in their structure of past interactions and mergings ( for example , elliptical and s0 galaxies with inclined gaseous disks , galaxies with faint shells and ripples , galaxies with kinematically decoupled nuclei , etc . ) .
polar - ring galaxies ( prgs ) , consisting of large - scale rings of stars , gas and dust orbiting around major axes of early - type galaxies , may be considered as extreme samples of possible interaction relics . indeed , in the case of prgs , the remnants of merged galaxies are not mixed in one smooth object but stay separately in a quasi - steady state for a long time .
prgs are very rare objects .
for example , the polar ring catalogue by whitmore et al .
( 1990 ) ( prc ) lists only 6 classic kinematically - confirmed polar - ring galaxies .
the unique geometry of prgs attracts the attention of astronomers trying to test the 3d shape of galactic potentials and to study the stability of inclined rings and disks ( see recent review articles by tohline 1990 , sackett 1991 , sparke 1991 , combes 1994 , cox & sparke 1996 ) .
such an important question as the origin of these peculiar objects it is still not adequately investigated .
it is usually suggested that collapse of a single protogalactic cloud can not create an object with two nearly - orthogonal large - scale systems ( but see curir & diaferio 1994 ) , and so some secondary event must occur in the history of prgs .
summarizing possible scenarios of polar - ring formation , one can enumerate the following : the accretion of matter from a nearby system or the capture and merging of a gas - rich companion ; the delayed inflow of a primordial intergalactic cloud ; the accretion of matter from the outskirts of the host galaxy itself ; the polar - ring formation from the return of the tidal material during the merging of two gas - rich spirals ( toomre 1977 , shane 1980 , schweizer et al . 1983 , sackett 1991 , sparke 1991 , hibbard & mihos 1995 ) . probably , all the above mechanisms can create ring - like structures around early - type galaxies . to our mind , the most straightforward scenario is the first one .
recent observations of several binary interacting systems clearly demonstrate such rings in the making ( for instance , ngc 7464/65 - li & seaquist 1994 , ngc 3808a , b and ngc 6285/86 - reshetnikov et al .
1996 ) .
another unclear question is the nature of central objects in prgs and a possible correlation of host galaxy properties with characteristics of a surrounding polar ring . in this paper , we present sph simulations of polar ring formation around target galaxies of different structures due to gas accretion during the encounter with a comparable - mass spiral galaxy . in our simulations
, we try to answer the following main questions : does this mechanism work ? what determines the size of the resulting ring , and what is its spatial structure ? on what timescale does the ring form ? what is the mass fraction of the gas captured into a ring ?
the paper is organized as follows : in section 2 , we examine observational properties of all known kinematically confirmed prgs and formulate some observational constraints on numerical simulations ; in section 3 , we discuss previous attempts to model the prgs formation , describe our modelling technique and results of simulations ; and finally we give our conclusionsin section 4 . throughout the paper , all distance - dependent quantities
are calculated using @xmath275 km / s / mpc .
as a definition of a polar - ring galaxy , we will use the definition of category a objects in the prc : spectroscopic evidence must exist for two nearly - perpendicular kinematical subsystems ; centers of the two components must be aligned , and both subsystems must have similar systemic velocities ; the ring must be comparable in size to the host galaxy , must be luminous and nearly planar .
this definition allows to separate dust - lane ellipticals , galaxies with inclined hi rings etc . from prgs . using this rigorous definition
, one can now consider only three additional galaxies to 6 the classic prgs listed in the prc : am 2020 - 504 ( whitmore & schweizer 1987 , arnaboldi et al . 1993 ) , ic 1689 ( reshetnikov et al .
1995 , hagen - thorn & reshetnikov 1997 ) and ngc 5122 ( cox et al .
( we do not consider eso 603-g21 here due to the puzzling kinematics of the central galaxy ( arnaboldi et al .
1995 ) . )
an examination of the optical images of prgs ( e.g. in the prc ) allows one to divide them into two groups ( whitmore 1991 ) : galaxies with extended disk - like rings with the central region cut out and galaxies with relatively narrow rings , not extended in radius .
this division is quite distinct since the first group of galaxies - a0136 - 0801 ( a-1 ) , ugc 7576 ( a-4 ) , ngc 4650a ( a-5 ) , ugc 9796 ( a-6 ) , and ngc 5122 ( b-16 ) - possess optical rings extended out to 2 - 3 diameters of the central galaxies , while the second group - eso 415-g26 ( a-2 ) , ngc 2685 ( a-3 ) , ic 1689 ( b-3 ) , and am 2020 - 504 ( b-19 ) - demonstrate optical rings with size not exceeding the diameter of the host galaxy . in table 1
, we generalize the main observational characteristics of the two groups of prgs .
( note that , due to the absence of optical data about ngc 5122 , we did not consider this galaxy in the table . ) in the case of incomplete data or large scatter of characteristics , we give in the table only limits or indicate the range of parameter changes .
absolute luminosities and colors in the table are corrected for galactic absorption .
we discuss table 1 in detail .
lll + parameter & extended rings & short rings + + _ central galaxy _ : + bulge - to - disk ratio & @xmath3 & @xmath41 + absolute luminosity in the @xmath5 band & @xmath6 & @xmath7 + @xmath8 & @xmath9 & @xmath10 + colour gradient & yes & yes + @xmath11 ( km / s ) & @xmath12 & @xmath13 + @xmath14/@xmath11 & @xmath15 & @xmath16 + decoupled nuclei & ? & yes + + _ ring _ : + diameter & @xmath17 & @xmath18 + & @xmath19 kpc & @xmath20 kpc + absolute luminosity in the @xmath5 band & @xmath21 & @xmath22 + @xmath8 & @xmath23 & @xmath24 + colour gradient & yes & ?
+ hi mass ( @xmath25 ) & @xmath26 & @xmath27 + ring inclination & @xmath28 & @xmath29 + + the problem of the bulge - to - disk ratio determination for the prgs is not quite simple .
reshetnikov et al .
( 1994 ) have noticed that bulge effective parameters ( @xmath30 and @xmath31 ) of prgs with wide extended rings ( our first group ) lie in the plane of effective parameters below the mean relationship for normal galaxies ( this means that at the same value of @xmath30 , bulges of prgs have smaller radii in comparison with bulges of normal galaxies ) - see figure 7 in the cited paper . but
bulge characteristics of the second group of galaxies are located exactly along the standard relationship .
such a dichotomy suggests two explanations : the unusual compactness of the first group of prg bulges or the underestimation of their sizes and luminosities due to the projection of gas and dust rich extended rings on central regions of the galaxies ( reshetnikov et al .
1994 ) .
a comparison of photometric cuts of the galaxies in different color bands allows one to solve this problem . in figure
[ fcuts ] , we present surface brightness distributions for four prgs with extended rings along the major axes of their central galaxies .
we compare two profiles for each galaxy : one in the @xmath5 passband and other in the red filter ( @xmath32 , @xmath33 , or @xmath34 ) .
as one can see from figure , absorption in the rings barely changes the shape of the profiles .
therefore , prgs with wide extended rings in fact possess unusually compact and faint bulges in comparison with normal early - type galaxies .
as is evident from figure [ fcuts ] , the exponential disk dominates in the photometric structure of all these galaxies .
the ratio of the total bulge luminosity to the luminosity of the exponential disk in the first group of prgs is about 0.1 ( we denoted this as @xmath3 in table 1 ) .
the photometric structure of prgs with short rings ( the second group of galaxies ) looks usual for early - type galaxies .
as was mentioned earlier , their characteristics in the plane of effective parameters follow the mean relation for normal galaxies . according to original papers , bulge - to - disk ratios in the @xmath5 band for these galaxies
are : eso 415-g26 @xmath35 0.52 ( whitmore et al .
1987 ) , ngc 2685 @xmath35 0.9 ( makarov et al .
1989 ) , ic 1689 @xmath35 1.9 ( reshetnikov et al .
am 2020 - 504 is an elliptical galaxy ( arnaboldi et al .
therefore , the second group prgs are normal bulge - dominated galaxies by their photometric structure ( conditionally , @xmath36 in the table ) . rings sizes in the two groups of galaxies are different , as well as their absolute luminosities ( @xmath37 in table 1 denotes a diameter of the central galaxy measured at the surface brightness level @xmath3825 ) .
both groups of rings show a large scatter of optical colors although , as it was shown by reshetnikov et al .
( 1995 ) , there is a general trend with a blue surface brightness ; bluer rings have , on average , higher surface brightnesses .
extended rings also show large - scale color gradients , that is , they become bluer at larger radii ( see , for instance , figures 4,6 in reshetnikov et al .
1994 and arnaboldi et al .
both group of rings contain a large amount of neutral hydrogen ( we assume in table 1 that all detected hi belongs to the rings ) , with larger scatter of hi mass in prgs with narrow rings ( schechter et al .
1984 , van gorkom et al . 1987 , richter et al . 1994 ) .
the last line of table 1 shows the range of ring inclinations ( angular distance between the ring and the perpendicular to the central galaxy plane ) according to whitmore ( 1991 ) .
both groups of prgs demonstrate rings to be very close to perpendicular , with a somewhat larger deviation for extended rings .
it should also be noted that global characteristics of extended rings resemble the disks of spiral galaxies ( this was previously pointed out by reshetnikov et al .
1994 and reshetnikov & combes 1994a , b ) .
we will discuss this analogy in section 4 . summarizing the above analysis
, one can conclude that _ there is a correlation between general properties of optical polar rings and characteristics of host galaxies .
_ extended , disk - like rings exist preferably around galaxies which have global photometric structure ( and probably mass distribution ) similar to late - type spiral galaxies .
host galaxies of prgs with extended rings demonstrate a remarkable similarity @xmath35 their characteristics fall in a relatively narrow range ( see table 1 ) . from the other side ,
bulge - dominated galaxies have tighter and narrow rings .
host galaxies and rings of prgs with non - extended rings demonstrate significantly larger scatter of intrinsic properties . although the statistics of prgs are not great ( 8 - 9 objects only ) , the mentioned tendency is quite distinct @xmath35 there are no bulge - dominated galaxies with extended luminous polar rings .
it is significant since the known prgs were selected on the base of optical morphology only , without any attention to the structure of the central galaxies .
one can note also that the prg dichotomy is quite analogous to a recently - found difference of ionized gas distribution in elliptical and s0 galaxies . according to macchetto et al .
( 1996 ) , more than half of elliptical galaxies with detected ionized gas have their gas concentrated in small ( @xmath394 kpc ) nuclear disks , while most s0 galaxies demonstrate more extended ( up to 18 kpc ) distribution of the gas .
a similar tendency was mentioned for the first time by whitmore ( 1991 ) , who found that only rapidly - rotating s0 disks have extended luminous polar rings , while dust - lane ellipticals rarely show an extended luminous component .
he suggested that a more flattened potential of the s0 galaxy is able to stabilize the ring at greater radii than in an elliptical galaxy . in the next section
, we will demonstrate that the observed dichotomy could be caused by another reason .
there are only a few investigations that illustrate several proposed scenarios of polar ring formation .
this lack of theoretical study devoted to the question of ring origin is connected with the necessity of the usage of complicated 3d gas dynamics models , since polar rings contain a high amount of neutral gas ( up to a few @xmath40 ) , as seen from table 1 .
one of the most fit tools to construct such models is a smoothed particle hydrodynamics algorithm sph , originally proposed by lucy ( 1977 ) and gingold & monaghan ( 1977 ) and considerably developed in the late eighties ( e.g. , hernquist & katz 1989 ) .
it is fully three - dimensional , flexible regarding the geometry of the gas distribution and consequently well - suited to modelling both the donor and recipient galaxies and the empty extended space between them .
applied to a range of polar - ring investigation problems , this method has already lead to many important results .
the main result of the ring stability investigations is that the time for gas in an axisymmetric or triaxial potential to settle into a steady state is small compared to the age of the universe ( see christodoulou et al .
1992 and references therein ) .
so there is sufficient time for accreted gas to form polar rings .
however , these studies did not take into account time - dependent effects and realistic initial conditions .
some efforts to include these effects have been made by rix & katz ( 1991 ) and weil & hernquist ( 1993 ) .
rix & katz ( 1991 ) have treated the polar - ring formation process as a gradual smearing of a small gaseous blob captured by a galaxy of spherically - symmetric structure on a distant circular orbit . naturally , the diameter of the forming ring ( up to 80 kpc ) was quite completly dictated by the initial position of a gaseous satellite ( the orbit radius ) .
the approach of weil & hernquist ( 1993 ) was more self - consistent .
they have considered a parabolic encounter of a low - mass gas - rich companion with a more massive galaxy and consequent merging of the former .
the gaseous component of the satellite was settling into the ring after the total disruption of the satellite in the vicinity of the target galaxy .
the forming ring turned out rather small ( about 6 - 7 kpc in diameter ) and its size was determined in the end by a hand - introduced description of details of the companion disruption .
the ability of the accreting material to dissipate energy is the crucial factor for ring formation .
test ( noninteracting ) particles ( which imitate molecular clouds from intergalactic space or clouds belonging to a companion galaxy ) lose energy due to ram - pressure in an extended halo of a spiral galaxy and form a gaseous ring rotating around the disk of the spiral ( sofue & wakamatsu 1993 , sofue 1994 ) .
the main feature of the present investigation is a description of the full history of the gas stripping of the spiral galaxy outskirts and its consequent capture by a satellite during a parabolic encounter . keeping in mind the recent observations of several interacting pairs of galaxies of comparable luminosities which demonstrate evidently ring - like structures in the making ( see introduction ) , we have considered a distant encounter of equal - mass systems . in the case of a distant encounter the effect of self - gravity
may be ignored .
the treatment of the gas hydrodynamics is described below .
we did not explore in a comprehensive manner the orbital parameter space and choose only one set of impact parameters for which the modelling encounter of an s0 galaxy with a gas - rich system of comparable mass unambiguously results in the polar - ring formation around the former .
motivated by the above observational data analysis , we have undertaken a numerical investigation of the ring - forming process for galaxies with different structures .
our simulations are based on a rather standard variant of the sph code , which is the same as that in sotnikova ( 1996 ) . for simplicity , we adopt the smoothing length @xmath41 ( analogue of the particle size ) to be independent of the local gas density .
the smoothed values of hydrodynamical parameters are estimated by using a grid with cell length equal to @xmath42 .
we assume an isothermal equation of state , leaving thermal effects completely unexplored .
the gas is always at a temperature @xmath43 , and the corresponding equation of state is @xmath44 where @xmath45 and @xmath46 are the gas pressure and density , and @xmath47 is the isothermal speed of sound . for @xmath48 , @xmath49 .
the motion of @xmath50 particles , which represent elements of the continous gaseous medium , is traced by means of equations of momentum conservation : @xmath51 where @xmath52 and @xmath53 are the spatial coordinate and velocity of the @xmath33-th particle , and @xmath54 is the gravitational potential .
the term @xmath55 describes the acceleration due to viscosity @xmath56 : @xmath57 where @xmath58 .
we adopt one of the standard forms for the artificial viscosity @xmath56 which serves to represent the sudden deceleration of the gas motion when strong shocks arise in the gas ( e.g. , hernquist & katz 1989 ) : @xmath59 in the expression ( [ visc ] ) , the viscosity depends on the divergence of the velocity field .
the first term is analogous to a bulk viscosity , the second , indroduced to prevent particle interpenetration at high mach number , is of the neumann - richtmyer type .
parameters @xmath60 and @xmath61 are free parameters . according to hernquist & katz ( 1989 ) , one can satisfactorily reproduce the change of the density and pressure across the shock front if the values of @xmath60 and @xmath61 are equal to 0.5 and 1.0 , respectively .
we used just these numbers in our simulations .
the procedure of a smoothed - value estimation of the gas density is quite usual , as well as the transition from hydrodynamical equations ( [ coord ] ) and ( [ vel ] ) to sph - equations ( for a detailed description of the adopted numerical scheme , see sotnikova 1996 ) .
_ donor galaxy _ to minimize the number of free parameters , we chose a very simple model for the donor galaxy its potential was taken as that of a softened point mass @xmath62 where @xmath63 is the gravitational constant , @xmath64 the mass , and @xmath65 the softening scale length of the potential .
@xmath50 particles are used to represent the gaseous medium .
they are initially gathered in a thin disk with a @xmath66 distribution from the center of the donor galaxy up to an outward radius @xmath67 and are placed in dynamically - cold circular orbits .
the total gas mass is 0.2 @xmath64 .
the gaseous particle size @xmath41 , which determines the numerical spatial resolution , is 300 pc . throughout , we employ the following system of units : the gravitational constant @xmath68 , the mass of the donor galaxy @xmath69 , the outer radius of the initial gas distribution in the donor galaxy @xmath70 . in terms of physical units ,
a unit mass can be defined as @xmath71 , and a unit distance corresponds to 15 kpc .
combined together , they give a unit time to be 86.6 myrs , and a unit velocity to 169.3 km @xmath72 .
+ + _ accreting galaxy _
we have considered the two - component model of a galaxy , consisting of a bulge and a disk .
the light distribution of the bulge is well - provided by the model used , for example , in weil & hernquist ( 1993 ) . for this model ,
the potential is @xmath73 where @xmath74 is the bulge mass , and @xmath75 the scale length of the potential .
the gravitational potential of the disk component of the galaxy is approximated by a miyamoto & nagai ( 1975 ) potential @xmath76 where @xmath77 is the disk mass , and @xmath78 and @xmath79 are scale radii .
this model was favored by the simplicity of its practical usage . in our simulations , @xmath80 .
the ratio @xmath81 determines the range of models from bulge - dominated systems ( large values of the ratio ) to disk - dominated galaxies ( small values of the ratio ) .
four models for an accreting galaxy have been considered , with values of @xmath81 equal to : 2.0 , 1.0 , 0.5 , 0.2 .
the total mass of the galaxy is taken the same for all models and equal to that of the donor galaxy : @xmath82 .
as shown in section 2 , main galaxies of prgs with extended rings are similar to spiral galaxies in global photometric structure ( and probably mass distribution ) .
according to de jong ( 1996 ) , the bulge effective surface brightness shows the best correlation with morphological type of spiral galaxies .
therefore , the total bulge luminosity ( mass ) also correlates well . scaling radii
show a large scatter for all types of spiral galaxies .
thus , we fixed the scale lengths of galaxy components , leaving only the mass of the bulge to be variable .
scale lengths of the bulge @xmath75 and disk @xmath79 potentials are 1 kpc and 5 kpc , respectively . for these values ,
the half - mass radius of the bulge is @xmath83 kpc , and the maximum of the rotation curve of the disk component falls on the radius @xmath84 kpc .
+ + _ orbit _ before modelling a ring - forming encounter , we first solved the two - body problem for the donor and ring - host galaxies .
the numerical procedure employed is quite similar to that used in weil & hernquist ( 1993 ) .
two interacting galaxies are initially separated by a rather large distance ( @xmath85 75 kpc ) , so that tidal effects be negligible .
the initial velocities are taken as for parabolic encounter .
the primary galaxy passes in a zero - inclination ( in the plane of the gaseous disk of the donor galaxy
@xmath86 plane ) , prograde orbit ( that is , the orbital angular momentum is parallel to the spin of the donor galaxy ) with a pericenter distance @xmath87 initially set as @xmath88 .
the polar axis of the ring - host galaxy disk component ( parallel to the @xmath89-axis ) lies in the orbital plane , so the orbit is polar regarding the ring - host galaxy .
the calculated orbit somewhat differs from parabolic ( due to non - keplerian potentials of galaxies ) , but its form is nearly the same for all models of a ring - host galaxy . during the encounter ,
the primary galaxy strips the outskirts of the donor object and a ring , rotating in the direction of the orbital motion , eventually forms around the galaxy in the encounter plane . as the equatorial plane of the disk component of s0
is taken to be perpendicular to that of the orbital motion , the ring is polar .
the total amount of accreted gas is about 10% of all gas in the donor galaxy ( or about @xmath90 ) and there is not any significant difference in the stripped mass for all considered models of the ring - host galaxy .
some amount of the gas fell into the central part of galaxy .
the mass of this gas is estimated as @xmath91 within 1 kpc from the center .
the timescale of the ring formation is a few @xmath92 years .
this time is somewhat shorter for rings forming around bulge - dominated galaxies and reaches up to @xmath93 years for a disk - dominated model .
the interacting galaxies are separated by a sufficiently large distance ( more than 120 kpc ) by the time of the ring steady - state settling , so that there are no any direct evidences of interaction between the galaxies and the ring - host system as an isolated object .
the simulations have revealed an interesting feature which appears during the ring - forming process in different galaxian potentials . as a steady state sets in
, ring sizes begin to diverge .
the results for four runs are presented in figure [ frings ] .
figure [ frings ] gives a final stage of polar - ring formation for all considered models .
this is the orbital plane projection of the gaseous rings .
the marked difference in the ring structure is the difference in ring size , which rises when the bulge mass decreases and ranges from about 7 kpc in diameter for bulge - dominated systems ( see table 1 ) to approximately 13 kpc for disk - dominated objects .
the ring radius is finally determined by the angular momentum of donor galaxy gaseous particles forming the ring with respect to the galaxy at the moment of the pericenter passage .
as the impact parameters are nearly the same for all runs , the value of the total angular momentum of ring - forming particles is almost the same for all models .
but one can see from figure [ fmom ] that the positions of particles with the same angular momentum are different for galaxies with different structures .
particles are closer to the center of galaxy if there is a more concentrated mass distribution the bulge - dominated model .
this implies , in particular , that under the same conditions rings forming around elliptical galaxies with a strong concentration of mass to the center ( see , for instance , the curves marked as ( @xmath94 ) and ( @xmath95 ) in figure [ fmom ] ) should be less extended ( on the average ) rather internal than those around galaxies with a more gently sloping density profile ( this fact was first mentioned in sotnikova 1996 ) . in figure [ f12 ] ,
the surface density profiles along the major axes of the rings are shown .
one can see one more interesting feature of simulated rings : rings farther from the center of the galaxy are more extended .
these results give the post factum justification of our impact parameter choice .
indeed , a closer passage results in less extended rings . the angular momentum of a gaseous particle at the moment of the closest approach of galaxies with respect the ring - host system is significantly affected by the term proportional to @xmath96 , where @xmath97 is the ring - host galaxy velocity relative to the donor galaxy . for parabolic encounters
, it leads to @xmath98 .
as the ring size monotonically decreases , the angular momentum decreases ; a closer encounter gives a ring lying rather inside the galaxy .
it is known that s0 galaxies possess some original gas , so the interaction between two gaseous systems would result in ring destruction .
the further decreasing of pericenter distance leads to merging of the galaxies and total change of their structures . as to the more distant encounters
, they do not allow capture of substantial amounts of gas , due to the sharp decrease of disturbing forces .
we have repeated our runs with @xmath99 kpc and have obtained that the mass of the resulting very diluted rings did not exceed @xmath91 .
hyperbolic galaxy encounters are not efficient for ring formation , as the time of strong galaxy interaction is short .
let us note also that we discuss the properties of classic prgs with optical star - forming rings .
so we discuss the origin of relatively dense rings - with gas density large enough for star formation .
this gives an additional restriction on the impact parameter , since the matter captured during the distant encounter will spread along a more extended orbit and will have lower mean density .
thus , galaxy encounters favourable to extended optical ring formation are extremely rare events .
such interactions must be between galaxies of specific types and within restricted geometry ( polar encounters and a very narrow range of impact parameter ) .
we found in our simulations that , with the @xmath100 impact parameter , the rings forming around bulge - dominated galaxies are less extended ( by factor two ) than the rings forming around disk - dominated galaxies .
therefore , having the same accretion history in the samples of elliptical and disk - type galaxies , we will have _ on average _ more extended rings around disk - dominated galaxies .
this inference may explain in a quite natural manner the absence of extended polar rings around elliptical galaxies .
it can also explain a difference in morphology of ionized gas in elliptical and s0 galaxies ( macchetto et al .
1996 ) , assuming external origin of gas in these galaxies .
one can remark , however , that the obtained segregation of ring sizes ( see figure [ frings ] ) is not as clear as the observed dichotomy ( see table 1 ) .
the simple bulge - disk model giving internal rings for bulge - dominated objects fails to explain the existence of very extended ( up to 30 kpc in diameter ) polar rings around disk - dominated galaxies .
probably , there exists one more physical factor leading to the strong differences of the prg structural properties .
as it was found from the observational data analysis , there exists for prgs with extended rings a remarkable structural resemblence of host galaxies to late - type spiral galaxies .
it is known that dynamical properties of late - type spiral galaxies are determined to a considerable extent by invisible massive halos ( e.g. , freeman 1992 ) .
hence , we can suppose that prgs with extended rings possess a third global component ( besides the bulge and the disk ) - a dark massive halo .
the existence of dark massive halos also follows from the analysis of ring kinematics ( e.g. , schweizer et al .
1983 , reshetnikov & combes 1994a ) . to account for the rotational velocities in the rings of ugc 7576 and ugc 9796 inside radii of 17 kpc and 21.4 kpc , respectively ,
one requires amounts of dark mass reaching 1.6 and 3 times the luminous masses ( reshetnikov & combes 1994a ) .
what are the possible consequences of taking into account this structural component in the context of our investigation ? as the gravitating mass of the primary galaxy increases , the orbital velocity of this galaxy relative to the donor object rises also
. then angular momentum arguments for a galaxy with such a structure lead to more distant orbits for captured particles .
we changed our bulge - disk model by adding a smoothed third component - a halo .
the structure and the shape of dark halos in early - type galaxies appears to be , at present , an open question ( sackett et al .
1994 , combes & arnaboldi 1996 ) . for lack of unambiguous knowledge about the halo shape
, we suppose the halo mass distribution to be spherical . as usual ,
halos are characterized by isothermal spheres over some radial interval . for simplicity ,
the folowing phenomenological cumulative mass profile is used to represent a halo in the present study : @xmath101 where @xmath102 is the total halo mass , @xmath103 a cutoff radius , and @xmath104 a core radius .
our choice of halo parameters was determined by characteristic values obtained by reshetnikov & combes ( 1994a ) from the ring kinematics investigation .
the core radius @xmath104 was taken as 9 kpc .
the value of the cutoff radius @xmath105 is somewhat arbitary and was taken to be equal to @xmath106 .
we choose this value to give a reasonable halo mass outside the radius comparable with @xmath87 .
we have let the dark mass to be equal to twice the luminous mass ( bulge @xmath107 disk ) inside a radius of 15 kpc .
the total mass of the bulge and disk have been reduced , that is , @xmath108 , and we let @xmath109 , @xmath110 kpc , @xmath111 kpc .
the geometry of the encounter was the same as in previous numerical experiments .
expecting the capture of gaseous matter on distant orbits and formation of a more diluted object , we have increased the total number of particles ( up to 20000 ) as well as their size @xmath41 up to 375 pc to keep the gas treatment as a continuous medium .
the morphology of the new run is shown in figure [ fhalo ] . by the time of @xmath112 after the pericenter passage , the captured gaseous matter
is azimuthally smeared in a annular configuration around the galaxy .
this structure gradually evolves into a closed ring .
we have followed its evolution up to @xmath113 . by this time , the ring was completly closed but not quite symmetric . according to rix & katz ( 1991 ) for a ring forming as a result of the gaseous satellite disruption , the smearing process takes up to seven orbital periods - up to 3 - 4 gyrs or two times the final time of our run .
the total mass gathered in the ring - like structure does not significally differ from that obtained in haloless runs .
the ring has a total mass of about @xmath114 . to explore the sensitivity of this value to the structural properties of the donor galaxy ,
we have constructed an additional model in which the spiral galaxy has its own halo .
the parameters of this halo were taken the same as for the host galaxy .
the mass of the captured gas turned out to be slightly smaller
( this value is somewhat small in comparison with that obtained from observations .
we omit the discussion of this question , because this value is obtained for one set of orbital parameters and we did not investigate the role of initial conditions , impact parameter , and structure of the donor galaxy . it may be impossible to understand the formation of very massive rings in the frame of the pure accretion scenario and the merging event is needed . )
the most remarkable feature of this annular structure is its size .
the ring material lies well outside the luminous material of the host galaxy .
we estimated the diameter of the ring of about 30 kpc .
this value is typical of the objects of the first group of prgs ( see table 1 ) .
the annular configuration obtained is rather narrow .
its further long - lived evolution will lead not only to azimuthal smearing of the gas but also to flattening of the radial density profile , forming a disk - like structure .
two factors promote the formation of rings with extended density distributions - the viscosity ( see rix & katz 1991 ) and the nonsphericity of the potential ( see fig.5 in this paper and fig.1 in katz & rix 1992 ) .
we have analyzed the data on global structural properties of all known prgs , and came to a conclusion that there is a correlation between the main characteristics of polar rings and host galaxies .
there exist two different classes of prgs : bulge - dominated galaxies with internal ( relative to the optical size of a galaxy ) rings and disk - dominated systems with extended rings .
the sample studied is somewhat small ( 8 - 9 objects ) , but when we combine our preliminary inference with the results of numerical experiments , a plausible picture of prg properties emerges .
we have presented the investigation of the ring - forming process by studying the stripping of gas - rich galaxy outskirts during the encounter with galaxies of different structures and the subsequent settling of the captured material in the potential of a companion galaxy using an sph method .
the simulations have demonstrated that the ring sizes depend on galaxy potentials .
_ under the same interaction conditions , rings twisting around galaxies with a strong concentration of mass to the center ( bulge - dominated systems ) are less extended ( on the average ) than that around disk - dominated galaxies .
_ moreover , we have found that it is impossible to obtain very extended rings ( with a diameter of about 30 kpc ) taking into account the distribution of the luminous mass only .
to obtain such extended rings , the galaxy mass distribution should be less concentrated than observed luminosity distribution .
therefore the presence of dark halos is required to explain the existence of extended rings .
one can suppose that the observational dichotomy is the result of selection .
indeed , interactions between galaxies accompanied by matter accretion occur with participation of various types of galaxies .
but we have more chances to create long - lived extended rings when the accreting galaxy is a gas free , disk - like galaxy with extended massive dark halo .
so such extended polar rings are expected to be observed preferably around specific types of galaxies .
the existence of the observational dichotomy suggests that prgs with bulge - dominated central galaxies may possess less pronounced dark haloes in comparison with the first group of objects ( another possibility - bulge - dominated galaxies have more centrally - concentrated halos - does not change our results as follow from sect.3.1.3 ) .
these conclusions do not contradict the present - day observational data about dark halos in early - type galaxies ( e.g. , bertin & stiavelli 1993 , de zeew 1995 ) .
nevertheless , we do not exclude other explanations of the revealed observational dichotomy ( for instance , environmental influence etc . ) . as was mentioned in section 2
, there is a close similarity between extended rings and disks of late - type spiral galaxies .
extended rings have the same total luminosities , integral colors , ratios of hi mass to the blue luminosity corrected for internal absorption , gas - to - dust ratios as normal spiral galaxies ; there is evidence for the gaseous medium in polar rings to be near ( or above ) the gravitational stability limit , as the gaseous disks in spiral galaxies ( reshetnikov et al .
1994 ) . two polar rings ( in ugc 7576 and ugc 9796 )
demonstrate h@xmath60-derived global star formation rates to be typical for normal spiral galaxies ; general properties of hii regions in these galaxies are normal for late - type spirals ( reshetnikov & combes 1994a ) .
hii regions in the ring of prototype polar - ring galaxy ngc 2685 demonstrate nearly solar oxygen abundances ( eskridge & pogge 1994 ) . extended rings show radial color gradients with bluer colors at larger radii ( reshetnikov et al .
1994 , arnaboldi et al .
1995 ) . the hi/@xmath116 mass ratio in the rings of ngc 660 and ngc 4650a is usual for spiral galaxies ( combes et al . 1992 , watson et al .
there is evidence for the spiral structure in the rings of ugc 7576 ( reshetnikov & combes 1994b ) and ngc 4650a ( arnaboldi et al . 1996 ) .
there is an indication of the presence of a large - scale magnetic field in the ring of ngc 660 ( reshetnikov & yakovleva 1991 ) .
moreover , according to combes & arnaboldi ( 1996 ) and arnaboldi et al .
( 1996 ) , dark matter in prgs could co - exist with the hi component , leaving its mark on the kinematics of the polar ring itself similar to the situation with a late - type galaxy gaseous disk .
one can note finally that prgs lie close to the tully - fisher relation for spiral galaxies ( knapp et al .
therefore , prgs with extended rings could be undistinguishable from normal spiral galaxies at less advantageous orientation ( with a more face - on ring ) - see the discussion of this question in prc .
probably , they will look like early - type spiral galaxies ( maybe barred ) with extended low surface brightness disks .
the striking similarity between extended rings and spiral disks suggests that prgs @xmath117 giant low surface - brightness spiral galaxies with decoupled bulges .
although this can not be completely ruled out in some special cases , we suppose that the ring capture due to external accretion is a more frequent process ( see section 1 about observations of forming rings ) .
therefore , in contrast to the usual opinion that galactic interactions lead to fast morphological changes towards s0/e , the observational data on prgs and our modelling simulations provide the example of a rare opposite shift - from an early - type gas - free galaxy through the capture of a gaseous disk to a spiral galaxy .
one can suppose that prg formation due to external accretion can imitate in some aspects the formation of disk galaxies in the hierarchically clustering model of the universe ( e.g. , steinmetz 1996 ) . | we present an analysis of structural features of all known galaxies with optical polar rings . we find a clear dichotomy for objects of this peculiar class .
bulge - dominated s0 galaxies possess only short narrow rings , while disk - dominated objects always have wide extended polar rings .
we try by gas dynamical simulations to explain such a segregation by dependence of the ring - forming process on different galaxy potentials .
it is found that the total mass captured into the ring during an encounter of a host - ring system with a gas - rich spiral galaxy of comparable mass exceeds @xmath0 ( or about 10% of all gas in the donor galaxy ) , which is of the order of that found by observation . the process of gas to gather into a steady - state ring
takes approximately @xmath1 years .
this time is somewhat shorter for rings forming around bulge - dominated galaxies .
we also present observational arguments for s0 galaxies with extended rings to be similar to late - type spirals by their photometric properties , while numerical modelling of the extended ring formation suggests that these galaxies must possess massive dark halos as well . in this case
, the sizes of the modelled rings turn out large enough ( up to 30 kpc in diameter ) , and the time scale for ring formation is prolonged up to several gyrs . |
in the accordance with the conventional regge - gribov approach , the one - pomeron contribution to the differential cross section of shdid can be expressed at @xmath25 in the form : @xmath26 where @xmath27 and @xmath28 are cms total interaction energy , total cross section , transverse momentum transferred , and invariant masses of final diffractively excited states respectively , @xmath29 is three - pomeron vertex and @xmath30 is the pomeron trajectory ; the parameter @xmath31 is to be chosen to single out diffraction processes from other ones [ 7 ] . since the mean slope of the pomeron trajectory is the only dimensional parameter which can be responsible for the decrease of the function @xmath29 as @xmath32
is increased , the domain where @xmath29 is expected to be nearly constant is estimated as @xmath33 where @xmath34 is an effective mean value of the derivative @xmath35 there which is reasonably evaluated to be @xmath36 .
it is why this domain is expected to be remarkably large , from @xmath37 to @xmath38 or even larger ( it has been observed long ago by comparison of the elastic and single inelastic diffraction differential cross sections that @xmath39 at @xmath40 [ 7 ] , wherefrom , in particular , a rather slow @xmath41-dependence of double inelastic diffraction differential cross section at @xmath42 follows ) .
the double inelastic diffraction is the only type of hadron interaction which is expected to exhibit such slow transverse momentum dependence . at still larger values of squared 4-momentum transferred pomeron is expected to be dissolved to its constituents [ 6 ] that begin to interact independently , so that the `` normal '' qcd regime @xmath43 is to be approached gradually . in
what follows the logarithmic dependence on @xmath41 and rather ambiguous but definitely slow decrease of @xmath44 in the right - hand side of eq.(1 ) are accounted on the average as @xmath45 .
the rough estimate of screening corrections to the one - pomeron shdid scattering amplitude @xmath46 associated with diagrams depicted in fig.3 shows that @xmath47 , @xmath48 being the corrected amplitude .
it is reasonable to adopt @xmath49 and enhance the above correction ( i.e.,to multiply the denominator in eq.(2 ) ) by the phenomenologically approved ( for forward elastic scattering amplitude ) factor about 1.5 , accounting the shadowing by the inelastic intermediate states .
then the corrected shdid amplitude is expected to be @xmath50 and the corresponding differential cross section is @xmath51 after integration of eq.(1 ) over the region @xmath52 one obtains the total cross section of shdid @xmath53 if one chooses a reasonable values @xmath54 , @xmath55 and the experimental value of @xmath56 , @xmath57 , then the fraction of shdid is expected to be @xmath58 and 0.10 at @xmath59 and @xmath60 respectively .
it can be several times less or larger , since the above estimate is rather rough , but its smooth logarithmic threshold - like energy increase is independent of the choice of parameters .
it seems reasonable to expect that hadronization of diffractively excited final states produced by shdid is dominated by mechanism of string rupture as shown in fig.4 , string been formed between scattered colored hadron constituent ( quark , diquark or gluon ) and remnant of the same hadron . any alternative string configuration would be unfavorable since it implies formation of some strings of a very high energy ( it is worthy to mention that diffractively produced state associated with target particle was always out of the game in cosmic ray experiments under discussion because it is never seen within the area of observation ; it is why the projectile inelastic diffraction only is thought of throughout the paper ) . at the same time , transferred momentum @xmath61 is insufficiently large for the fragmentation mechanism of hadronization to prevail .
let us consider the above string in its own cms and adopt that secondary particle rapidity and transverse momentum distributions in pomeron - proton interaction is similar to that in real hadron one at cms energy @xmath62 ( as to the rapidity distribution , it is supported by the well known result of ua4 collaboration [ 8 ] ) . since what is observed is nothing else , than transverse plane projection of the picture which
is resulted from its rupture , it becomes obvious that the typical ratio of a secondary transverse momentum projection normal to reaction plane ( i.e. , to the plane of draft ) to `` transverse momentum string length '' ( i.e. to ls relative transverse momentum of leading particles oppositely directed in string cms ) is about @xmath63 where @xmath64 is mean transverse momentum of secondaries in hadron interactions , and mean leading particle energy is experimentally proved to be about half of incident particle one . at @xmath65
this ratio is about 0.13 .
the only point what remains to be discussed to compare the above consideration to the experimental data is an obvious estimate of the role of atmospheric cascade . since the atmosphere thickness above the altitude where the calorimeter is mounted corresponds to about 3.5 nuclear mean free paths , the probability of at least one shdid collision is about @xmath66 at @xmath67 .
if it does happen , then the subsequent soft collisions can not , most probably , blur essentially the target plane picture it initiates , especially for energy distinguished cores .
it is why the additional assumption suggested by experimenters [ 2 ] seems to be not necessary , that alignment is caused by some peculiarities of the lowest nuclear collision above the chamber only . at the same time ,
the threshold - like dependence of alignment on core energies is associated , may be , with the violating role of nuclear cascade .
thus , the main puzzling experimental features of alignment phenomenon , namely , the fraction of alignment events about ( 20 - 40)% and the ratio of mean value of normal to reaction plane projection of core transverse momentum to maximal value of core relative transverse momenta ( @xmath68 ) ( string `` half - thickness '' to its `` length '' in transverse momentum space ) are compatible qualitatively with the above theoretical consideration ( 30% and 0.13 respectively ) , if one adopts that each core is originated ( due to electromagnetic cascade ) from a hadron created along with string rupture .
the threshold - like dependence of shdid cross section on interaction energy can elucidate why the phenomenon has not been noticed at lower energies ( especially , accounting a poor statistics and other ambiguities of cosmic ray experiments ) .
however , this point as well as some other features of the phenomenon , such as its threshold - like dependence on core energies , core energy distribution , their energy sequence along the alignment line , etc . , needs both the enrichment of statistics and mc simulation of cascade and shdid collisions themselves ( especially , accounting that hadrons of different masses can be produced at the end of string and along its length ) which are in progress .
unfortunately , it is rather questionable , whether an attempt to observe the alignment phenomenon will be undertaken in accelerator experiments soon .
99 pamir collaboration , `` analysis of structure of halo in families with energy @xmath69 '' , proceedings 5th international symposium on very high energy cosmic ray interactions , lodz , 1988 , v. contributed papers , p. 9 .
amineva , g.f .
fedorova et al . , `` alignment of increased background region in gamma - hadron superfamilies '' , proceedings 6th international symposium on very high energy cosmic ray interactions , tarbe , 1990 v. contributed papers , p. 264 ; i.p .
ivanenko , v.k .
kopenkin , a.k .
managadze and i.v .
rakobolskaya , pisma jetf , 1992 , v. 56 , p. 192 . f. halzen and d.a .
morris , phys .
, 1990 , v. d42 , p. 1435 .
ua1 collaboration , g. arnison et al .
lett . , 1985 , v.158b , p. 494
a. managadze , private communication .
a. mironov and i. royzen , sov .
fiz . , 1988 ,
v. 47 , p. 1125 ; v. 48 , p. 194 .
a. alberi and g. goggi , phys . rep .
1984 , v. 74 , no .
ua4 collaboration , d. bernard et al .
lett . , 1986 , v. 166b , p.459 fig.1 .
the example of target plane picture with energy distinguished cores for event with alignment , @xmath70 ; figures stand for energy in tev ( already multiplied by factor 3 for hadrons);andor stand for electromagnetic halo and hadrons of high energy respectively .
other particles of the family are marked as(@xmath1 -quanta ) and ( hadrons ) .
one - pomeron exchange approximation to shdid .
wavy lines refer to pomeron exchange , @xmath71 and @xmath28 are invariant masses of diffractively excited states , q is 4-momentum transferred , @xmath72 is triple - pomeron vertex function . | an explanation of the puzzling alignment effect observed in cosmic ray experiments is suggested * theoretical approach to alignment phenomenon * + * i. royzen * + p.n .
lebedev physical institute of russian academy of sciences 53 leninsky prospect , 117924 moscow , russia + few years ago the observation has been made [ 1 ] in cosmic ray experiments that the alignment of the main energy fluxes along a straight line in target ( transverse ) plane exceeds significantly the background level .
more precisely , at superhigh energies of initial particle ( @xmath0 ) the secondary particle superfamilies detected by deep lead x - ray emulsion chamber appeared to be situated almost along straight line in target plane ( fig.1 ) .
the coplanar scattering of such a type was so surprising that an attempt has been made to revise the result but instead they were confirmed with much better confidence level [ 2 ] .
the analysis of the alignment effect for 74 high energy @xmath1-families induced by hadrons above and within the chamber has been carried out .
their energies energies are selected to be @xmath2 ( hadron energies being restored , accounting that the energy of induced @xmath1-family is about @xmath3 of the hadron energy it is originated from ) .
this analysis suggested that superfamily production happened predominantly rather low above the chamber ( at the altitude @xmath4 , since it seemed that nuclear - electromagnetic cascade development would blur alignment , if several interactions contributed ) .
it confirmed a coplanar scattering and scaling - like fragmentation spectrum of energy distinguished cores .
the alignment parameter @xmath5 , @xmath6 is used as the alignment criterion where @xmath7 stands for a number of centers of highest energy and @xmath8 is the angle between the two - dimensional vectors @xmath9 and @xmath10 in target plane , an event being recognized to have alignment , if @xmath11 .
actually , events with @xmath12 were chosen only because of too high statistical background for @xmath13 and rather poor statistics for @xmath14 .
the threshold - like behavior of the effect has been observed : no alignment at @xmath1-family energies @xmath15 , then its gradual increase within energy range @xmath16 to manifest itself finally in ( 20 - 40)% of total number of events .
14 events with @xmath17 have been observed , exhibiting most striking alignment structure ( @xmath18 ) .
core transverse momentum @xmath19 was estimated by rough relation @xmath20 , @xmath21 being the distance of a spot from the interaction axis .
the mean ratio of value of maximal relative core transverse momentum to its normal to the alignment line projection ( in target plane ) @xmath22 is @xmath23 .
no other peculiarities of alignment events compared to `` usual '' cascade have been noticed .
the first attempt of theoretical consideration of the above alignment phenomenon has been made by f. halzen and d.a .
morris [ 3 ] , whose approach was based on the assumption that semihard gluon jets is a feature of all events at energies above @xmath24 .
it was shown that within this approach the cosmic ray observations were associated probably with the jet alignment in three - jet events observed already in the collider experiment [ 4 ] .
i would like to suggest an alternative treatment which makes it possible to understand many features of cosmic ray alignment observations quite naturally , including the threshold - like energy behavior and fraction in extensive atmospheric showers as well as the typical projections of core transverse momenta to the alignment line and normal to it , and allowing for events with more , than four cores aligned , that have been extracted recently from cosmic ray data [ 5 ] .
the main point of the approach under consideration is that the alignment events are assumed to be associated with semihard double inelastic diffraction ( shdid ) of hadrons [ 6 ] .
let us trace them step - by - step . |
here we present a two - loop renormalisation - group analysis of the multifractal properties of a disordered weyl semimetal in @xmath19 dimensions using the minimal subtraction scheme@xcite .
a similar scheme has been applied recently @xcite to compute the correlation - length and dynamical exponents for a wsm to the two - loop order , following previous studies of graphene@xcite and of the equivalent gross - neveu model@xcite .
also , a similar multifractality analysis has been carried out in ref .
@xcite for 2d ( @xmath124 ) dirac fermions on the surface of a 3d topological insulator .
although there is no disorder - driven phase transition in such a 2d system , the wavefunctions display multifractal behaviour on sufficiently short length scales .
@xmath125 \psi\,d\br , \label{l0minimal } \\ \cl_{\text{int}}=&\frac{1}{2}\varkappa_0\int(\psi^\dagger\psi)^2d\,\br \\
\cl_{v_0}= & v_0\int\left[s_r(\br)s_r^*(\br)\right]^{q-1 } s_a(\br)s_a^*(\br ) d\br .
\label{lv0}\end{aligned}\ ] ] here we have introduced a positive matsubara frequency @xmath126 that ensures the convergence of the superintegral ( [ zpartition ] ) with respect to the bosonic components of the supervectors @xmath50 and @xmath49 and , in the below calculation , also regularises infrared divergences of momentum integrals in @xmath19 dimensions for @xmath127 .
the lagrangian of the system in the minimal subtraction scheme is separated into the effective lagrangian @xmath128 that describes the long - wave behaviour of the physical observables and the counterterm lagrangian @xmath129 that cancels contributions divergent in the powers of @xmath130 : @xmath131 \psi\,d\br \nonumber \\ & + \frac{1}{2}\varkappa\int(\psi^\dagger\psi)^2d\,\br \label{le } \\ & + v\int\left[s_r(\br)s_r^*(\br)\right]^{q-1 } s_a(\br)s_a^*(\br ) d\br , \label{lv}\end{aligned}\ ] ] where the velocity of the renormalised weyl fermions ( the coefficient before @xmath132 ) is set to unity , without loss of generality , by appropriately choosing the field @xmath133 .
the scale @xmath134 sets the characteristic momentum of the long - wave behaviour .
the coefficients @xmath67 and @xmath135 before the source terms ( [ lv0 ] ) and ( [ lv ] ) are considered infinitesimal in the calculation below .
we note , that in general the renormalisation generates additional terms @xmath136^{m } s_a(\br)s_a^*(\br ) d\br $ ] with @xmath137 , which we neglect here because their contributions to the partition function ( [ zpartition ] ) are less singular @xmath138 at @xmath139 than that of the term ( [ lv ] ) , and , thus , they do not contribute to the iprs ( [ iprminimals ] ) .
the minimal subtraction scheme@xcite consists in calculating perturbative corrections to the lagrangian ( [ le ] ) and choosing the counterterms @xmath129 to cancel divergent in powers of @xmath130 contributions .
the rg equations can then be derived by relating the `` observable '' parameters @xmath133 , @xmath140 , @xmath33 , and @xmath67 to the `` bare '' ones @xmath50 , @xmath141 , @xmath142 , and @xmath135 .
all momentum integrals in such a calculation are convergent in low dimensions @xmath19 with @xmath127 .
the results have to be analytically continued to higher dimensions , @xmath17 , at the end of the calculation ( dimensional regularisation ) .
the `` dephasing rate '' @xmath36 , sent to zero at the end of the calculation [ cf .
( [ iprminimals ] ) and ( [ l0minimal ] ) ] , may be assumed to be significantly smaller than the scale @xmath140 and neglected when computing the parameters of the renormalised lagrangian .
the one - loop perturbative correction to the vertex @xmath67 is described by the diagrams in fig .
2c - e and the topologically equivalent diagrams .
since the `` dephasing rate '' @xmath36 may be neglected when considering the respective high - momentum scattering processes , the advanced and retarded green s functions may be taken identical @xmath143 when evaluating these diagrams .
the sum of diagrams 2c - e is given by @xmath144 where the prefactors @xmath99 and @xmath42 account for the numbers of topologically equivalent diagrams , @xmath145 , and @xmath94 , and @xmath146 is the tensor product of the two subspaces of the two @xmath147 propagators connected by impurity lines in figs .
2d and 2e ; the structure of the correction is trivial in the other propagators subspaces .
the perturbative corrections to the disorder strength @xmath33 and the frequency @xmath140 have been calculated in detail in ref . . the velocity of the weyl fermions [ the coefficient before the @xmath132 term in eq .
( [ le ] ) ] does not receive first - order corrections .
the presence of the one - loop counterterms ( [ lcounter1 ] ) requires introducing the respective diagrammatic elements , fig .
[ oneloopcounter ] , in addition to the propagators and impurity lines , when calculating diagrammatically perturbative corrections beyond the one - loop order .
diagrams in fig .
[ diagrams](a)-(m ) are computed similarly to the two - loop diagrams for the renormalisation of the disorder strength , obtained by replacing the zigzag line by an impurity line and considered in detail in ref . .
the values of diagrams [ diagrams](a)-(m ) , together with the numbers of equivalent diagrams , are provided in table [ table : samediagr ] .
diagrams [ diagrams](n)-(p ) are regular in @xmath14 due to the mutual cancellation of the singularities coming from blocks with vertical and diagonal impurity lines . for instance ,
the sum of the diagrams in fig .
[ diagrams](n ) ( see also fig . [ diagramn ] ) can be evaluated ( in units @xmath152 ) as @xmath153 } { ( \omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp+\bq)^2 ] } \nonumber\\ = 4\omega^2\int\frac{(\hbsigma\bp)(\hbsigma\bq ) } { ( \omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp+\bq)^2 ] } \nonumber\\ + 4\omega^2\int\frac{1 } { ( \omega^2+p^2)(\omega^2+q^2)[\omega^2+(\bp+\bq)^2 ] } \nonumber\\ -8\omega^4\int\frac{1 } { ( \omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp+\bq)^2 ] } \nonumber\\ = \co(1).\end{aligned}\ ] ] ( for detailed calculations of the last integrals see ref . ) .
the values of diagrams 3(n)-(q ) are given in table [ diagrnq ] .
diagrams [ diagrams](r)-(w ) are the two - loop diagrams for the corrections to the vertex @xmath67 that contain one - loop counterterms and are equivalent to similar diagrams in ref . , up to replacing the zigzag line or its counterterm by the impurity line or its counterterm , with the values provided in table [ table : counter ] .
the one - loop and two - loop corrections to the lagrangian ( [ le ] ) are cancelled by the counterterm lagrangian @xmath157\psi\ : d\br \nonumber\\ & + \frac{1}{2}\delta\varkappa\int\left(\psi^\dagger\psi\right)^2d\br \nonumber\\ & + \delta v\int\left[s_r(\br)s_r^*(\br)\right]^{q-1 } s_a(\br)s_a^*(\br ) d\br,\end{aligned}\ ] ] where the values of @xmath158 , @xmath159 , and @xmath160 have been calculated previously@xcite ( see also refs . ):
@xmath161 , \label{omegadivergent } \\ & \delta(\hbsigma\hat\bk ) = \frac{1}{4\varepsilon}(\varkappa c_{2-\varepsilon}\omega^{-\varepsilon})^2\hbsigma\hat\bk , \\ & \delta\varkappa = \varkappa\left[-\frac{2}{\varepsilon}\varkappa c_{2-\varepsilon}\omega^{-\varepsilon } + \left(\frac{4}{\varepsilon^2}-\frac{1}{2\varepsilon}\right)\left(\varkappa c_{2-\varepsilon}\omega^{-\varepsilon}\right)^2\right ] .
\label{kappadivergent}\end{aligned}\ ] ] introducing the dimensionless disorder strength @xmath163 the full lagrangian ( [ lfull ] ) can be rewritten as @xmath164\psi \,d\br \nonumber\\ + \frac{\gamma\omega^\varepsilon}{4c_{2-\varepsilon } } \left(1-\frac{\gamma}{\varepsilon}-\frac{\gamma^2}{8\varepsilon } + \frac{\gamma^2}{\varepsilon^2}\right)\int(\psi^\dagger\psi)^2d\br \nonumber\\ + v\left[1-\frac{q}{2\varepsilon}\gamma + \frac{1}{\varepsilon^2}\left(\frac{q^2}{8}+\frac{q}{4}\right)\gamma^2 -\frac{1}{\varepsilon}\left(\frac{3}{16}q^2-\frac{3}{16}q\right)\gamma^2\right ] \nonumber\\ \int\left[s_r(\br)s_r^*(\br)\right]^{q-1 } s_a(\br)s_a^*(\br ) \,d\br .
\label{lagrangianfinal}\end{aligned}\ ] ] by comparing the lagrangian ( [ lagrangianfinal ] ) , that depends on the renormalised observables @xmath133 , @xmath140 , @xmath33 , and @xmath67 , with the lagrangian ( [ l0minimal])-([lv0 ] ) expressed in terms of the `` bare '' variables @xmath50 , @xmath141 , @xmath142 , and @xmath135 , we can relate the `` bare '' and the renormalised observables : the input parameters @xmath135 and @xmath142 of the lagrangian are independent of the characteristic momentum scale @xmath168 at which the long - wave properties of the system are observed , which gives @xmath169 eqs .
( [ cs ] ) , analogous to the callan - symanzik equation@xcite , immediately lead to the rg equations ( 11c ) and ( 11a ) with @xmath170 .
the rg equation ( 11b ) follows from eq .
( [ omegabare ] ) using that @xmath171 . | systems with the power - law quasiparticle dispersion @xmath0 exhibit non - anderson disorder - driven transitions in dimensions @xmath1 , as exemplified by weyl semimetals , 1d and 2d arrays of ultracold ions with long - range interactions , quantum kicked rotors and semiconductor models in high dimensions .
we study the wavefunction structure in such systems and demonstrate that at these transitions they exhibit fractal behaviour with an infinite set of multifractal exponents .
the multifractality persists even when the wavefunction localisation is forbidden by symmetry or topology and occurs as a result of elastic scattering between all momentum states in the band on length scales shorter than the mean free path .
we calculate explicitly the multifractal spectra in semiconductors and weyl semimetals using one - loop and two - loop renormalisation - group approaches slightly above the marginal dimension @xmath2 .
= 1 after half a century of studies , disorder - driven transitions in conducting materials still motivate extensive research efforts .
anderson localisation ( al ) transition is responsible for turning a metal into an insulator when increasing the disorder strength in dimensions @xmath3 and was believed for several decades to be the only possible disorder - driven transition in non - interacting systems .
al continues to fascinate researchers by its peculiar and universal properties , such as , e.g. , _
multifractality_ fractal behaviour of the wavefunctions at the transition with an infinite set of multifractal exponents@xcite .
a broad class of systems with the power - law quasiparticle dispersion @xmath0 in dimensions @xmath1 displays , however , another single - particle disorder - driven transition distinct from al@xcite . this transition , unlike al , occurs only near a band edge or at a nodal point ( in a semimetal ) .
it reflects in the critical behaviour of the disorder - averaged density of states ( in contrast with al ) , as well as in other physical observables , e.g. , conductivity .
such a transition has first been proposed@xcite for the specific case of dirac semimetals ( @xmath4 , @xmath5 ) and has recently sparked vigorous studies@xcite@xcite @xcite of its critical properties in 3d weyl and dirac systems@xcite .
other playgrounds for the observation of this non - anderson disorder - driven transition are 1d and 2d arrays of trapped ultracold ions with long - range interactions@xcite , quantum kicked rotors@xcite ( mappable onto high - dimensional semiconductors ) , and numerical simulations of schroedinger equation in @xmath6 dimensions@xcite . despite these comprehensive studies ,
the wavefunction structure at these non - anderson disorder - driven transitions is rather poorly understood .
such transitions are not necessarily accompanied by localisation ; they can occur between two phases of localised states [ like in 1d ( non - chiral ) chains of trapped ions@xcite ] or between two phases of delocalised states [ e.g. , in single - node weyl semimetals ( wsms ) ] or between localised and delocalised states ( in a high - dimensional semiconductor@xcite ) .
particle wavefunctions in all of these cases are characterised by a correlation length that diverges from both sides of the transition . _
results . _ in this paper we study microscopically wavefunctions @xmath7 at the non - anderson disorder - driven transitions and demonstrate their _ multifractal _ nature .
when delocalised states are allowed by symmetry , dimensions , and topology , the typical wavefunctions at the critical disorder strength have a fractal structure and are characterised by a universal non - linear multifractal spectrum @xmath8 , defined@xcite by the inverse participation ratios ( iprs ) @xmath9 in the limit of an infinite system size @xmath10 .
such multifractal behaviour persists even if the wavefunctions are delocalised on both sides of the transition ( like in a single - node wsm ) . unlike the al transition , here
the multifractal spectrum @xmath8 is determined by the elastic scattering on length scales shorter than the mean free path . in systems that allow for localised states near the transition ,
these states scale as @xmath11 when approaching it , where @xmath12 is the localisation length divergent at the transition . in this paper
we also calculate the multifractal spectrum @xmath8 explicitly for several systems .
we find the multifractal spectrum of a disordered system with the power - law quasipatricle dispersion @xmath13 in dimensions @xmath1 in the orthogonal symmetry class , in the expansion in powers of @xmath14 , to be @xmath15 where @xmath16 ( and @xmath17 ) . for weyl semimetals , @xmath18 ,
in @xmath19 dimensions @xmath20 we note that on sufficiently short length scales the multifractality in a weyl semimetal is similar ( with @xmath14 replaced by the dimensionless disorder strength ) to that of 2d dirac fermions on the surfaces of a 3d topological insulator , studied in ref . @xcite .
although there is no phase transition in such 2d systems , their wavefunctions display non - universal short - length multifractal behaviour@xcite , in contrast with the universal multifractality ( [ weylresult ] ) that persists at all length scales . ) near a non - anderson disorder - driven transition .
the transition occurs at the energy @xmath21 near the ( renormalised@xcite ) band edge or near a nodal point ( for semimetals).,scaledwidth=40.0% ] _ the phase diagram _ of a finite - size system near a non - anderson disorder - driven transition is shown in fig .
[ phasediagr ] . in what follows we measure all energies from a nodal point or a ( renormalised@xcite ) band edge[45 ] set to @xmath21 . for disorder strengths weaker than a critical value ( `` weak - disorder phase '' in fig . [ phasediagr ] ) , @xmath22 , the disorder is perturbatively irrelevant@xcite with the dimensionless disorder strength @xmath23^{-1}$ ] vanishing at small energies @xmath24 , where @xmath25 is the elastic scattering time .
unlike the case of low dimensions , the lowest - energy levels of a sufficiently large high - dimensional system are _ discrete _ for @xmath22 , as the `` level width '' @xmath26 vanishes faster than the spatial - quantisation gaps @xmath27 between the lowest levels at @xmath28 .
the interplay of the level discreteness with multifractality will be discussed below . for supercritical disorder strength , @xmath29 , the dimensionless disorder strength grows at low energies , until reaching the value @xmath30 that marks the boundary of the `` strong - disorder phase '' ( that for @xmath31-dimensional semiconductors in the orthogonal symmetry class also matches the mobility threshold ) in fig .
[ phasediagr ] .
_ inverse participation ratios .
_ the wavefunction structure at energy @xmath32 and disorder strength @xmath33 is conveniently characterised by the disorder - averaged iprs@xcite @xmath34^{q-1}g_a(\br,\br , e,\eta)\right>_{\text{dis } } , \label{iprmain}\end{aligned}\ ] ] where @xmath35 are the retarded and advanced green s functions with an artificially introduced `` dephasing rate '' @xmath36 , @xmath37 are the energies of the eigenstates @xmath38 for a given disorder realisation , and @xmath39 is the ( disorder - averaged ) density of states . below we compute the disorder - averaged iprs ( [ iprmain ] ) near the critical point @xmath40 as a function of the system size @xmath41 or the localisation length @xmath12 .
the iprs , eq .
( [ iprmain ] ) , are mimicked by the diagram in fig .
[ basicdiagr]a .
green s functions before disorder averaging , cf .
eq .
( [ iprmain ] ) .
b ) vertex that connects @xmath42 green s functions .
c)-e ) diagrams for the one - loop renormalisation of the vertex . for @xmath43 , diagrams c)-e ) have equal values . in weyl semimetals and other odd - spectrum ( @xmath44 ) systems diagrams
d ) and e ) cancel each other . ] unlike the case of low dimensions , @xmath45 , the low - energy properties of high - dimensional materials under consideration are affected by elastic scattering between all momenta in the band@xcite . such ultraviolet processes are qualitatively important on length scales shorter than the mean free path @xmath46 , and , in particular , determine the criticality and multifractality near the transition point due to the diverging @xmath46 .
this leads to the critical properties and multifractal spectrum different from those at the usual al transition , that occurs for states away from nodes and band edges and is described by non - linear sigma - models@xcite on length scales longer than the mean free path .
_ renormalisation procedure .
_ the effects of the ultraviolet scattering can be addressed in a controlled way by means of a perturbative renormalisation - group ( rg ) controlled by the small parameter@xcite@xcite @xcite @xmath16 .
the results obtained from this approach are expected to hold qualitatively also in systems with @xmath47 . to perform renormalisations , we rewrite the disorder - averaged iprs using a supersymmetric@xcite field theory : @xmath48^{q-1 } \nonumber\\ & s_a(\br)s_a^*(\br ) \:\exp({-\cl_0-\cl_{\text{int } } } ) , \label{pqsuper } \\
\cl_0=&-i\int \psi^\dagger\lambda^\frac{1}{2 } \left[e\,\lambda(k)+i\eta\lambda\lambda(k)-\epsilon_{\hat\bk}\right ] \lambda^\frac{1}{2}\psi\,d\br , \label{l0 } \\
\cl_{\text{int}}&=\frac{1}{2}\varkappa(k)\int(\psi^\dagger\lambda\psi)^2d\,\br , \label{lint}\end{aligned}\ ] ] where @xmath49 and @xmath50 are @xmath51-component supervectors in the @xmath52 ( fermion - boson @xmath53 retarded - advanced ) space , @xmath54 and @xmath55 are the bosonic components of the supervectors , and @xmath56 . the factors with @xmath57-matrices in eq .
( [ l0 ] ) ensure the convergence of the supersymmetric integral with respect to the bosonic variables@xcite . upon repeatedly integrating out shells of highest momenta ,
the action ( [ l0])-([lint ] ) reproduces itself with renormalised disorder strength @xmath58 and the parameters @xmath59 and @xmath60 that `` flow '' with the running cutoff @xmath61 and initial values @xmath62 , @xmath63 , and @xmath64 , where @xmath65 is the ultraviolet momentum cutoff set by the bandwidth or the impurity size@xcite .
the parameters @xmath66 and @xmath67 grow upon coarse - graining and exhibit singular behaviour with @xmath68 at the critical disorder , @xmath69 .
the iprs ( [ iprmain ] ) can be rewritten as @xmath70 , \label{pqrenormgen}\end{aligned}\ ] ] where @xmath71 $ ] is the ipr of an effective renormalised system with the same quasiparticle dispersion @xmath0 , but with renormalised disorder strength @xmath33 and energy @xmath72 and that excludes scattering into momentum states @xmath73 that were removed by the rg procedure .
the rg has to be stopped either when the spatial quantisation effects become important or if it runs into the regime of strong disorder , @xmath74 .
the renormalised system is then equivalent to a simple ( low - dimensional ) system with discrete energy levels or with a constant density of states and unaffected by scattering into high - momentum modes ; the ipr @xmath75 of such a system can be found using conventional methods developed for low - dimensional systems@xcite .
_ fractality of delocalised states .
_ in what immediately follows we consider a system with delocalised finite - energy states at @xmath69 [ along path @xmath76 in fig .
( [ phasediagr ] ) ] , as , e.g. , in a @xmath77-dimensional system with potential disorder .
the rg procedure at critical disorder is terminated when either the momentum @xmath61 reaches @xmath78 or the spatial quantisation effects become important ( i.e. the energy levels become discrete ) .
the renormalised system is then either ballistic ( for small @xmath14 , that ensures weak disorder at the critical point ) or equivalent to a usual weakly disordered metal ( for larger @xmath14 ) , with@xcite @xmath79 .
the characteristic energy @xmath80 of terminating the rg is related to the system size @xmath41 as @xmath81 , where @xmath82 is the dynamical critical exponent . using that @xmath83 , we find @xmath84 , which , together with eq .
( [ pqrenormgen ] ) gives the multifractal spectrum @xmath85 _ localised states .
_ for trivial - topology systems in the orthogonal symmetry class the states in the `` strong - disorder phase '' ( fig . [ phasediagr ]
) are localised .
also , all states on the phase diagram are localised in systems in @xmath86 dimensions ( if allowed by symmetry / topology ) .
near the critical point of the non - anderson disorder - driven transition localised states are still multifractal on length scales shorter than the localisation length @xmath12 ( that diverges at the transition ) . for zero - energy states at supercritical disorder ,
@xmath29 , ( along path 2 in fig .
[ phasediagr ] ) the rg is stopped when reaching strong disorder , @xmath87 .
such @xmath21-states are then characterised by only one length scale @xmath88 that gives the localisation length @xmath12 .
similarly to the case of delocalised states in a size-@xmath41 system , we find that @xmath89 , and the participation ratio @xmath90 with the multifractal spectrum ( [ deltageneral ] )
. the non - trivial scaling of the ipr ( [ iprloc ] ) with the size of the localisation cell reflects the multifractality of the wavefunctions on length scales @xmath91 . _ orthogonal semiconductors .
_ for a disordered system with the quasiparticle dispersion @xmath92 the rg flow of the system parameters is given in terms of the dimensionless disorder strength @xmath93 , with @xmath94 , by the rg equations @xmath95 where @xmath96 and @xmath97 are the terms of higher orders in @xmath98 .
eqs .
( [ rglambdao ] ) and ( [ rggammao ] ) for the renormalisation of the energy and the disorder strength have been obtained previously in refs . and .
eq .
( [ rgvo ] ) describes the flow of the preexponential in eq .
( [ pqsuper ] ) .
the renormalisations ( [ rgvo])-([rggammao ] ) can be also easily obtained diagrammatically .
for instance , the one - loop renormalisation of the vertex @xmath67 , fig .
[ basicdiagr]b , is given by @xmath42 diagrams equivalent to [ basicdiagr]c , @xmath99 diagrams equivalent to [ basicdiagr]d , and @xmath99 diagrams equivalent to [ basicdiagr]e .
all these diagrams have the same value for the dispersion under consideration , hence the prefactor @xmath100 in eq .
( [ rgvo ] ) .
the renormalisation of the disorder strength @xmath98 and the parameter @xmath66 for the dispersion under consideration has been described in detail in ref . .
in the one - loop order we find from eqs .
( [ rgvo])-([rggammao ] ) that @xmath101 and@xcite @xmath102 , which , together with eq .
( [ deltageneral ] ) , gives the multifractal spectrum ( [ semicondresult ] ) .
_ chiral systems , _ such as weyl semimetals or chiral chains with long - range hopping@xcite , often have odd quasiparticle spectra , @xmath44 , which leads to the mutual cancellation of diagrams [ basicdiagr]d and [ basicdiagr]e .
the rg flow of the vertex @xmath67 is given by eq .
( [ rgvo ] ) with the replacement @xmath103 and with the dimensionless disorder strength @xmath104 .
the flow of @xmath66 is given by eq .
( [ rglambdao ] ) with the replacement @xmath105 . from the rg equations we find @xmath106 [ cf .
eq .
( [ zeta ] ) ] , which , according to eq . , gives vanishing multifractality @xmath107 in the one - loop order .
thus , finding the multifractal behaviour in such chiral systems requires rg analysis in higher orders . in what follows
we present the result for a weyl semimetal ( see appendix for a detailed two - loop rg analysis of multifractality using the minimal - subtraction scheme ) .
_ weyl semimetals
_ are 3d systems with the quasiparticle dispersion @xmath108 , where @xmath109 is a vector of pauli matrices .
wsm properties near the non - anderson disorder - driven transition can be studied by performing rg analysis in @xmath19 dimensions with setting @xmath110 at the end of the calculation .
although the rg procedure is controlled by small @xmath14 , it is known@xcite to give good agreement with numerical results even for @xmath111 .
the flows of the parameters of a disordered wsm in @xmath112 dimensions are given by ( see appendix ) @xmath113 eqs .
( [ lambdargwsm ] ) and ( [ gammargwsm ] ) for the renormalisation of the energy and disorder strength in a disordered wsm have been derived previously in refs . and and in refs .
for the equivalent gross - neveu model .
an equation equivalent to eq .
( [ vrgwsm ] ) has also been derived in ref .
@xcite to describe the wavefunctions of 2d dirac fermions on the surface of a 3d topological insulator@xcite . from eqs .
( [ vrgwsm])-([gammargwsm ] ) we find to the two - loop order @xmath114 , which , together with @xmath115 and eq . ( [ deltageneral ] ) , gives the multifractal spectrum ( [ weylresult ] ) .
_ level discreteness and observability of multifractality .
_ observation of multifractality at the critical point ( @xmath21 , @xmath69 ) requires that the disorder - averaged energy spectrum of the system at this point is continuous , i.e. smeared by disorder and unaffected by the spatial quantisation .
this condition is always met in systems with @xmath116 as the `` level width '' @xmath117 for the lowest levels @xmath118 is of the order of their energies @xmath37 .
however , for some systems , e.g. , chains of ultracold ions@xcite , it is possible to realise@xcite @xmath119 , that corresponds to weak disorder @xmath120 at the critical point and the existence of a large number @xmath121 of energy levels that remain discrete [ @xmath122 at the critical disorder , although the energies and the spacings between these levels vanish in the limit @xmath10 .
the multifractal behaviour in such systems is observable only on sufficiently short length scales @xmath123 that correspond to the wavelengths of higher levels belonging to the continuous part of the energy spectrum .
_ rare - region effects .
_ the perturbative rg that we used in this paper neglects non - perturbative instantonic contributions [ , ] to the field theory ( [ pqsuper])-([lint ] ) , that , e.g. , result in the formation of lifhsitz tails near band edges and always lead to a finite density of states@xcite near nodes .
the effects of such instantons on physical observables , such as the density of states and conductivity , are rather small near the critical point in high - dimensional systems and were undetectable in all numerical studies@xcite so far except ref . .
another potential consequence of rare - region ( instantonic ) effects , albeit currently not demonstrated analytically , may be the `` rounding '' of the criticality , i.e. preventing the divergence of the correlation length near the critical point , and thus converting the phase transition of the type discussed here into a sharp crossover , with the latter scenario advocated in ref . .
in our view , the plausibility of this scenario deserves further investigation , in particular , in systems that disallow localisation by symmetry and topology . in the case
the criticality does get smeared in a system , the wavefunction multifractality studied here is observable on distances shorter than a large characteristic length set by the rare - region effects . _
acknowledgements .
_ we are grateful to a.w.w .
ludwig for useful motivating discussions at the initial stages of this work and to p.m. ostrovsky for previous collaboration . also , we thank m.s .
foster for valuable discussions .
our work was financially supported by the nsf grants dmr-1001240 ( lr and svs ) , dmr-1205303(vg and svs ) , phy-1211914 ( vg and svs ) , phy-1125844 ( svs ) and by the simons investigator award of the simons foundation ( lr ) . |
magnetic reconnection is the fundamental plasma process where a change in magnetic topology facilitates the conversion of magnetic energy to plasma kinetic energy and heat .
it plays a fundamental role in causing eruptions in the coronae of the sun and other stars , in the interaction between the solar wind and the magnetospheres of earth and other planets , for confinement in toroidal fusion devices , and in a large collection of astrophysical settings @xcite
. there has been increased interest in the properties of reconnection in realistic systems going beyond the simplifying assumptions of the sweet - parker model @xcite . in this classical model ,
the magnetic fields , densities , and temperatures are equal on either side of the reconnection site , and the upstream plasmas has no bulk flow other than the inflow in the reference frame of the reconnection site .
one example of going beyond this model is to allow asymmetries in the reconnecting magnetic fields , densities and temperatures on either side of the reconnecting current sheet .
a second example is including the effect of a bulk flow in the upstream plasma , whether in the direction of the reconnecting magnetic field or out of the reconnection plane . understanding how these effects impact the reconnection process , both quantitatively and qualitatively , is often of great importance for applying our understanding of reconnection to real systems .
one example is reconnection at the dayside magnetopauses of earth and other planets .
the plasmas at the magnetopauses of earth @xcite and mercury @xcite differ on the two sides and the solar wind drives a bulk flow in earth s magnetosheath @xcite and undoubtedly does at mercury s , as well .
when the interplanetary magnetic field ( imf ) is northward , the magnetosheath flow is parallel / anti - parallel to the reconnecting magnetic field in the polar regions .
when the imf is southward , magnetosheath flow at the flanks is mostly out of the reconnection plane .
the effect of upstream bulk flow is even more dramatic at the magnetospheres of jupiter and saturn , where rotation of the magnetosphere is much stronger of an effect than at earth @xcite .
a second example where upstream asymmetries and bulk flow are important is in tokamaks .
the density and temperature profiles are peaked in the plasma core with a spatially varying magnetic field , so the plasma profiles at the reconnection site are non - uniform @xcite .
further , there are often bulk flows causing the toroidal and poloidal rotation of the plasma @xcite , especially those driven by neutral beam injection .
therefore , both asymmetries and flows effects are present and are important to the dynamics in magnetospheres and tokamaks .
while the effect of asymmetries and flow shear have separately received much attention @xcite , only a few studies have treated systems that simultaneously contain both effects .
studies of the shock structure far downstream of the reconnection site were carried out analytically @xcite and using magnetohydrodynamic ( mhd ) modeling @xcite .
particle - in - cell simulations were used to study systems simultaneously including asymmetries , flow shear , and an out - of - plane ( guide ) magnetic field @xcite .
it was shown that the flow shear and diamagnetic drifts set up with the pressure gradient and the guide field can either reinforce or counteract each other .
more recently , a scaling analysis for systems including both asymmetries and upstream flow in the reconnection plane was performed @xcite .
it was argued that the reconnection site ( the x - line ) typically convects in the outflow direction .
the convection speed of the x - line and the rate of reconnection was predicted as a function of arbitrary upstream plasma parameters for isolated systems ; the results will be reviewed in sec .
[ sec : theory ] . in symmetric
reconnection with a flow shear , reconnection does not occur if the flow is super - alfvnic because the tension in the reconnecting magnetic field can not overcome the energy of the flow @xcite .
there is also a critical flow speed above which reconnection does not occur for asymmetric reconnection ; a generalization of the symmetric result for the asymmetric case was also derived @xcite .
these predictions were successfully tested with two - dimensional numerical simulations using the two - fluid model ( mhd with the hall term and electron inertia ) .
however , it is known that the fluid model is not well - suited to describe systems with asymmetric density and temperature as the fluids do not mix in the absence of thermal conduction @xcite ; even if conduction is present , the fluid model may not faithfully describe mixing in a nearly collisionless plasma as is the case in many applications .
these shortcomings are not present in kinetic simulations , such as the particle - in - cell numerical technique @xcite where macro - particles are evolved in time and plasma mixing naturally occurs .
thus , it is important to investigate the scaling of the reconnection rate and the drift speed of isolated x - lines within a fully kinetic model . in this study , we perform a systematic numerical study of magnetic reconnection with asymmetries and in - plane upstream flow using the particle - in - cell ( pic ) technique .
we measure relevant quantities in independent simulations in which all quantities are held fixed other than the upstream flow .
we find that the theoretical predictions previously tested in fluid simulations @xcite are consistent with the results of the pic simulations . in sec .
[ sec : theory ] , we review the predictions for the convection speed of isolated x - lines and the reconnection rate in terms of upstream parameters . sec .
[ sec : simulations ] discusses the simulations we perform as well as our methodology for analyzing the simulation data . sec .
[ sec : results ] presents the simulation results and compares them to the predictions .
we summarize our results and discuss applications of the results in sec .
[ sec : discussion ] .
scaling laws for the dissipation region s convection speed and the reconnection rate were derived @xcite for isolated configurations including asymmetric ( but anti - parallel ) magnetic fields and asymmetric densities and temperatures , along with arbitrary in - plane upstream flow .
we define the magnetic field strengths of the two upstream regions as @xmath0 and @xmath1 , the plasma mass densities as @xmath2 and @xmath3 , and the upstream flow speeds as @xmath4 and @xmath5 , where the @xmath6 subscript is borrowed from boundary normal coordinates in magnetospheric applications to denote the reconnecting component , and the 1 and 2 subscripts denote the two upstream sides of the reconnection site .
the reconnecting magnetic fields are treated as positive quantities , and the speeds are defined as positive if in the direction of @xmath0 and negative in the direction of @xmath1 .
the convection speed @xmath7 of the x - line along the current sheet , the reconnection rate @xmath8 , and the upstream flow speed @xmath9 required for steady state reconnection to be prevented scale ( in cgs units ) as @xmath10 @xmath11 and @xmath12 in writing these expressions , the asymmetric alfvn speed @xmath13 , the asymmetric reconnection rate in the absence of a flow shear @xmath14 , and the velocity shear @xmath15 are @xmath16 @xmath17 and @xmath18 where @xmath19 and @xmath20 are the half - thickness in the normal direction and half - length in the outflow direction of the dissipation region and @xmath21 is the speed of light .
equation ( [ eqn::drift_eqn ] ) was derived using conservation of momentum in the @xmath6 direction into and out of the dissipation region .
equation ( [ eqn::recon_eqn ] ) follows from treating the energetics of the release of magnetic tension on the outflow jets while including the upstream flow , keeping track of the fact that the x - line and stagnation point are not colocated in asymmetric reconnection @xcite .
equation ( [ eqn::vcrit ] ) is the condition that makes @xmath22 in eq .
( [ eqn::recon_eqn ] ) , which is the condition for when reconnection shuts off .
equations ( [ eqn::alfven_outflow ] ) and ( [ eqn::asym_rate ] ) follow from the analysis of asymmetric reconnection with no upstream flow @xcite .
equation ( [ eqn::shear_vel ] ) is merely shorthand for the quantity of import for the reconnection rate in eq .
( [ eqn::recon_eqn ] ) .
we emphasize a few important assumptions made in this analysis .
it assumed no upstream out - of - plane ( guide ) magnetic field .
it also assumed that the outflow speeds in the two downstream directions are equal and opposite in the reference frame of the moving x - line . as a scaling analysis
, it assumes a single characteristic value represents each quantity in the appropriate regions .
we perform 2d kinetic particle - in - cell simulations using the p3d code @xcite to test the predictions .
particles are stepped forward using a relativistic boris algorithm , while electromagnetic fields are updated with a second order trapezoidal leapfrog .
magnetic field strengths
are normalized to an arbitrary strength @xmath23 and plasma number densities are normalized to an arbitrary density @xmath24 . values of length and speed are normalized to the ion inertial length @xmath25 and the alfvn speed @xmath26 respectively , where @xmath27 is the ion charge and @xmath28 is the ion mass .
the unit of time is therefore @xmath29 . as in the fluid simulations
@xcite , the boundary conditions are doubly periodic , and the magnetic field profile is initialized as a double harris sheet : @xmath30 where @xmath31 is the initial current sheet width and @xmath32 is the domain size in the inflow direction .
( the @xmath33 direction in the simulations corresponds to the @xmath6 direction in boundary normal coordinates . )
there is no out - of - plane guide field .
the temperature profile of species @xmath34 , which can denote electrons @xmath27 or ions @xmath35 , is initialized as @xmath36 where @xmath37 and @xmath38 are selected asymptotic initial temperatures .
we use @xmath39 for all simulations .
initial electron and ion densities are chosen to be equal with asymptotic values of @xmath40 and @xmath41 .
the density profiles initially enforce pressure balance across the current sheet in the fluid sense .
there is no known general asymmetric kinetic equilibrium @xcite ( although there are approximations @xcite that are not employed here ) . as in many previous studies , our system rings at early times but settles to a steady state by the time of interest for this study . by this time
, the initial kinks in the initial magnetic field and bulk velocity profiles at the current sheets also smooth out and therefore do not present any problems .
.initial upstream plasma parameters for the simulations in this study .
the set labeled `` @xmath42 '' have asymmetric fields with a symmetric density .
the set labeled `` @xmath43 '' has a symmetric magnetic field and an asymmetric density .
the set labeled `` @xmath44 '' is representative of earth s magnetopause .
the predicted critical flow shear to shut off reconnection is given for each set . [ cols="^,^,^,^,^,^,^,^,^,^",options="header " , ] [ table::runs ] for all simulations , the speed of light is @xmath45 and the electron mass is @xmath46 .
the time step for particles is @xmath47 and the electromagnetic fields have a time step half as much .
the grid scale is @xmath48 .
the simulations performed for this study are summarized in table [ table::runparms ] .
the set labeled `` @xmath42 '' employ asymmetric magnetic fields with symmetric density , the set labeled `` @xmath43 '' have symmetric magnetic fields with asymmetric density , and the set labeled `` @xmath44 '' are for representative magnetospheric conditions @xcite .
the domain size is @xmath49 for the @xmath42 simulations and @xmath50 for the @xmath43 and @xmath44 simulations .
the initial number of particles - per - grid cell is @xmath51 for the @xmath42 and @xmath43 simulations and @xmath52 for the @xmath44 simulations .
the upstream flow speeds are varied for each set ; table [ table::runs ] shows the values used in the present study . to reach the nonlinear phase of reconnection more rapidly , the simulations are initialized using a coherent divergence - free sinusoidal perturbation to the magnetic fields of amplitude 0.1 with one full wavelength of the perturbation in the @xmath33 direction and two full wavelengths in the @xmath53 direction .
each simulation is evolved until magnetic reconnection reaches a steady - state .
since the reconnection rate differs for different upstream flow speeds , the steady state is reached at different times for different simulations .
consequently , we use the half - width of the primary magnetic island as a common indicator across the simulations ; a range of island widths of 4 - 6 @xmath54 is used to identify comparable times .
if needed , the interval is slightly adjusted to ensure the system is in a steady state . for each time step and each current sheet ,
the x - line and o - line are found using standard techniques by calculating the flux function @xmath55 as @xmath56 .
the saddle point of @xmath55 is the x - point and the extremum is the o - point .
the convection velocity of the reconnection site is measured as the time derivative of the x - line position .
the reconnection rate is the time rate of change of the magnetic flux difference between the x - line and o - line .
these values are averaged over the steady state interval to provide a representative value for that simulation .
2d plots of data from the magnetopause - like simulation with @xmath57 and @xmath58 . ( a ) out - of - plane current density @xmath59 , with magnetic field lines in blue and ( b ) out - of - plane magnetic field @xmath60 .
plots show only a small portion of the computational domain .
( c ) reconnection rate @xmath61 as a function of time @xmath62 for this simulation . ]
we begin by showing an overview of the plasma parameters in a simulation with representative magnetospheric conditions , specifically the @xmath44 simulation with @xmath57 and @xmath58 .
figure [ fig::panel_c ] contains ( a ) the out - of - plane current density @xmath59 with magnetic field lines overplotted and ( b ) the out of plane magnetic field @xmath60 at a time of 67.5 when the reconnection rate has reached a steady state , with the coordinate system shifted so that the x - line is at the origin .
only a fraction of the total computational domain is plotted .
interestingly , the results are quite similar to standard systems without flow shear .
in particular , the hall magnetic field in ( b ) is mostly bipolar , as is the norm in strongly asymmetric systems @xcite .
the normal electric field @xmath63 ( not shown ) shows the typical asymmetric hall electric field dominated by a positive @xmath63 on the strong ( magnetospheric ) field side of the dissipation region and a negative larmor electric field upstream of it @xcite .
panel ( c ) shows the reconnection rate @xmath61 as a function of time from this simulation , showing that the system reaches a steady - state by approximately @xmath64
. cuts in the inflow ( @xmath53 ) direction of plasma parameters from the simulation shown in fig .
[ fig::panel_c ] . plotted are the ion density @xmath65 and ion bulk flow @xmath66 in the direction of the reconnecting field @xmath67 to the ( a ) left and ( b ) right of the x - line .
( c ) normal electric field @xmath63 ( blue ) and out - of - plane magnetic field @xmath60 ( red ) @xmath68 to the left of the x - line . ] figure [ fig::panel_f ] shows cuts 4 @xmath54 downstream of the x - line from the same simulation .
the ion density @xmath65 and ion bulk flow velocity @xmath66 to the left and right of the x - line are shown in panels ( a ) and ( b ) , respectively .
they reveal the negative and positive deflections from the background flow profile due to the reconnection exhausts .
panel ( c ) shows @xmath63 on the left axis and @xmath60 on the right axis in a cut 4 @xmath54 to the left of the x - line .
the dashed horizontal lines mark zero for the two plots .
the hall magnetic ( @xmath69 ) , hall electric ( @xmath70 ) and larmor electric ( @xmath71 ) fields are present .
these results suggest that the upstream bulk flow largely does not alter the kinetic signatures of reconnection for typical magnetospheric conditions .
another feature of reconnection with an upstream flow shear is the tilting of the current sheet near the x - line @xcite .
we see the current sheet tilt in the present simulations , as well .
the tilting in the @xmath43 simulations with asymmetric density is more pronounced than similar upstream flows from the @xmath42 simulations with asymmetric magnetic field ( not shown ) .
the tilt is more pronounced for higher flow shear , as is to be expected .
convection speed @xmath7 of the x - line as a function of flow speed @xmath15 for simulations with ( top ) asymmetric magnetic field strengths @xmath72 and @xmath73 and symmetric density 0.2 and ( bottom ) asymmetric densities @xmath74 and @xmath75 and symmetric magnetic field strength @xmath76 .
the speeds in the top plot are negative because the x - line convects to the left . triangles and squares are for the top and bottom current sheets , respectively . the predicted convection speed from eq .
( [ eqn::drift_eqn ] ) is given by the dashed line.,title="fig : " ] convection speed @xmath7 of the x - line as a function of flow speed @xmath15 for simulations with ( top ) asymmetric magnetic field strengths @xmath72 and @xmath73 and symmetric density 0.2 and ( bottom ) asymmetric densities @xmath74 and @xmath75 and symmetric magnetic field strength @xmath76 .
the speeds in the top plot are negative because the x - line convects to the left .
triangles and squares are for the top and bottom current sheets , respectively . the predicted convection speed from eq .
( [ eqn::drift_eqn ] ) is given by the dashed line.,title="fig : " ] next , we test the predictions for the x - line drift speed , reconnection rate , and cutoff speed . raw measured and predicted values for all simulations are given in table [ table::runs ] .
we first consider the @xmath42 simulations containing asymmetries in magnetic field strength . from eq .
( [ eqn::vcrit ] ) , the predicted cutoff speed is about @xmath77 .
we vary the upstream flow speed up to this cessation condition , with @xmath78 for simplicity , and measure the drift speed in each simulation .
the results are shown in the top plot in fig .
[ fig::magdrift ] . here and throughout ,
blue triangles and red squares are for the two current sheets in the equilibrium .
error bars are determined as the standard deviation during the steady state intervals . a linear trend in drift speed for increasing shear flow speed
is observed , qualitatively consistent with eq .
( [ eqn::drift_eqn ] ) .
the dashed black line is the prediction of eq .
( [ eqn::drift_eqn ] ) , so the quantitative agreement is good as well .
we carry out the same analysis on the asymmetric density simulations , with results shown in the bottom plot of fig .
[ fig::magdrift ] .
the predicted cessation condition for these simulations is approximately @xmath79 .
again , the trend and absolute agreement is quite good .
note that some of the measurements from the simulation are slightly below the prediction for both the @xmath42 and @xmath43 simulations ; this could be due to the inertia of moving larger primary islands in this finite sized and periodic domain .
reconnection rate @xmath80 as a function of flow speed @xmath15 for simulations with ( top ) asymmetric magnetic field strengths @xmath72 and @xmath73 and symmetric density @xmath81 and ( bottom ) asymmetric densities @xmath74 and @xmath75 and symmetric magnetic field strength @xmath76 .
triangles and squares are for the top and bottom current sheets , respectively . the predicted reconnection rate from eq .
( [ eqn::recon_eqn ] ) is given by the dashed line and is normalized to the average reconnection rate in the simulation with @xmath82.,title="fig : " ] reconnection rate @xmath80 as a function of flow speed @xmath15 for simulations with ( top ) asymmetric magnetic field strengths @xmath72 and @xmath73 and symmetric density @xmath81 and ( bottom ) asymmetric densities @xmath74 and @xmath75 and symmetric magnetic field strength @xmath76 .
triangles and squares are for the top and bottom current sheets , respectively . the predicted reconnection rate from eq .
( [ eqn::recon_eqn ] ) is given by the dashed line and is normalized to the average reconnection rate in the simulation with @xmath82.,title="fig : " ] next , we test the reconnection rates obtained in both sets of simulations .
the results are plotted in fig .
[ fig::magrate ] for ( top ) asymmetric magnetic field and ( bottom ) asymmetric density sets , respectively .
the dashed lines denote the predicted values ; in calculating the predictions , we use the average reconnection rate from the two sheets in the zero flow shear case as @xmath14 in eq .
( [ eqn::recon_eqn ] ) .
we find excellent agreement between these measurements and the predictions .
a similar analysis is carried out for the representative magnetospheric simulations . from the prediction , the convection speed should double and the reconnection rate should drop by @xmath83 as @xmath4 is increased from @xmath84 to @xmath85 .
the measured convection speed is 0.88 for @xmath86 and 1.64 for @xmath87 , which differ by a factor of @xmath88 ; this agrees well with the prediction .
the reconnection rate drops from 0.143 to 0.123 , a decrease of @xmath89 . while ostensibly greater than the prediction , the uncertainty in these measurements is enough where the difference is not expected to be significant .
the key is that the significant increase in upstream flow speed only slightly impacts the reconnection rate , as predicted for reconnection in systems with a strongly asymmetric density @xcite . as a test of the prediction for the cutoff flow shear speed to shut off asymmetric reconnection , we report results from simulations beyond the predicted cessation conditions .
we run simulations with @xmath90 = 2.4 and 2.8 for the asymmetric field and @xmath90 = 2.0 and 2.4 for the asymmetric density simulations .
the current sheets in these systems tend to contort and become sinusoidal rather than flat , as if beginning to go kelvin - helmholtz unstable .
this is qualitatively different than the simulations below the threshold , where reconnection clearly starts from the beginning .
we also point out that some of them nonlinearly experience reconnection in the strongly bent fields .
we argue this form of reconnection is different than the robust form of reconnection discussed earlier , so we assert that the simulation results are consistent with eq .
( [ eqn::vcrit ] ) . finally , we comment on whether the predictions remain valid when the upstream flow on the two sides is in the same direction , _
i.e. , _ both @xmath4 and @xmath5 are positive .
we do an asymmetric magnetic field simulation with @xmath91 .
the predictions are that the x - line will drift with the common upstream drift speed and the reconnection rate is the same as if there was no flow .
as shown in table [ table::runs ] , this is borne out in the simulations . in summary ,
the simulation results agree quite well with the predictions discussed in sec .
[ sec : theory ] .
we use particle - in - cell simulations to study the scaling of 2d asymmetric anti - parallel reconnection with an in - plane upstream flow .
the particle - in - cell approach is necessary to faithfully capture the effect of plasma mixing between the two disparate plasmas in the exhaust region , which is not well - described in the fluid description .
we find very good agreement with the scaling predictions as a function of upstream plasma parameters for the drift speed of the x - line , the reconnection rate , and the critical upstream flow speed necessary to suppress reconnection found in a recent study @xcite .
one area in which this study goes beyond the previous fluid simulation study is by testing the theory for systems in which the flow is in the same direction on each upstream side , which has been studied for solar wind applications @xcite .
the results confirm the theory @xcite works in this case as well , and in particular confirms that the figure of merit is @xmath92 .
physically , for a case with @xmath4 and @xmath5 in the same direction , the upstream plasma on both sides enter the diffusion region with momentum in the direction of the reconnecting field , and this momentum makes the x - line convect at the weighted average of the two speeds . in the special case of equal flows on the two sides , the upstream plasmas are stationary in the reference frame of the x - line , so there is no effect on the reconnection rate .
this reveals that reconnection in the solar wind should not be suppressed by flow shear , which is consistent with the observation of active reconnection in the solar wind @xcite . a important consideration before applying the results here to naturally occurring reconnection or reconnection in the laboratory is that the x - lines in question must be `` isolated '' in the sense that they are free to convect in the external flow for the theory to apply .
this is essentially satisfied in the solar wind , for example , and may also be the case for neoclassical tearing modes ( ntms ) in tokamaks .
however , for the magnetopause of earth and other planets , one should proceed with caution .
a primary x - line is undoubtedly controlled by global considerations such as being line - tied to the ionosphere . therefore , it is not clear if a single x - line would follow the predictions of the theory tested here .
however , a flux rope or flux transfer event ( fte ) @xcite could be considered isolated , so the theory may apply .
there are differences between the predictions tested here and the leading model of open flux ( _ i.e. _ , fte ) motion @xcite , so future work on this topic would be interesting .
an area of potential interest for future study are the properties of the system in regimes where the flow is higher than the cessation condition . in our simulations ,
the current sheet contorts as if beginning to undergo kelvin - helmholtz instability .
then , reconnection as a secondary process on the bent current sheets begins .
the current sheets flatten and reconnect robustly .
it is not clear whether this is physical or only a function of the finite system size in the simulations , so it is worth future study .
we point out that a limit with such strong flows is not likely to apply at the magnetosphere except possibly when the reconnection site interacts with the dense , cold plasmas in plasmaspheric drainage plumes @xcite , but may be important in tokamaks .
one limitation of this study is that it is in 2d .
any 3d dynamics , including drift waves set up with wave vector in the out - of - plane direction due to the in - plane pressure asymmetry being normal to the reconnecting magnetic field , are artificially suppressed in 2d .
it is not expected drift waves will change the bulk properties of the reconnection , but this is worth future study .
another limitation of the present study is that it does not take into account an out - of - plane component of bulk flow velocity or an out - of - plane ( guide ) magnetic field .
it is known that diamagnetic effects arise in systems with a guide field and a pressure asymmetry @xcite , and that the effects of flow shear and diamagnetic drifts compound @xcite .
the effect of out - of - plane flow has been studied @xcite .
incorporating diamagnetic effects into the theoretical predictions @xcite should be the subject of future work .
support from west virginia university s summer undergraduate research experience ( sure ) program ( ced ) , the nasa west virginia space grant consortium ( ced ) , nsf grant ags-0953463 ( pac ) , and nasa grants nnx16af75 g and nnx16ag76 g ( pac ) are gratefully acknowledged .
this research uses resources of the national energy research scientific computing center ( nersc ) , a doe office of science user facility supported by the office of science of the u.s .
department of energy under contract no .
de - ac02 - 05ch11231 .
we thank k. malakit for use of his color table in some of the plots , and we thank m. t. beidler , c. m. komar , and y. h. liu for stimulating discussions .
the simulation data used to produce the results of this paper are available from the authors by request .
49ifxundefined [ 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 * * , ( ) in @noop _
_ , ( , ) p. @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) `` , '' ( , ) chap . @noop ( ) @noop * * , ( ) `` , '' ( , ) chap . @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) `` , '' ( , , ) chap . @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1029/2004ja010809 [ * * , ( ) , 10.1029/2004ja010809 ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) | we investigate magnetic reconnection in systems simultaneously containing asymmetric ( anti - parallel ) magnetic fields , asymmetric plasma densities and temperatures , and arbitrary in - plane bulk flow of plasma in the upstream regions . such configurations are common in the high - latitudes of earth s magnetopause and in tokamaks .
we investigate the convection speed of the x - line , the scaling of the reconnection rate , and the condition for which the flow suppresses reconnection as a function of upstream flow speeds .
we use two - dimensional particle - in - cell simulations to capture the mixing of plasma in the outflow regions better than is possible in fluid modeling .
we perform simulations with asymmetric magnetic fields , simulations with asymmetric densities , and simulations with magnetopause - like parameters where both are asymmetric . for flow speeds below the predicted cutoff velocity , we find good scaling agreement with the theory presented in doss et al . , _
j. geophys .
res .
_ , * 120 * , 7748 ( 2015 ) .
applications to planetary magnetospheres , tokamaks , and the solar wind are discussed . |
the advent of smartphones and tablets has made the use of touchscreen keyboards pervasive in modern society .
however , the ubiquitous qwerty keyboard was not designed with the needs of a touchscreen keyboard in mind , namely accuracy and speed .
the introduction of gesture or stroke - based input methods significantly increased the speed that text could be entered on touchscreens [ @xcite ] .
however , this method introduces some new problems that can occur when the gesture input patterns for two words are too similar , or sometimes completely ambiguous , leading to input errors .
an example gesture input error is illustrated in figure [ fig : example - swipe - collision ] .
a recent study showed that gesture input has an error rate that is about 5 - 10% higher compared to touch typing [ @xcite ] . with the fast and inherently imprecise nature of gesture input
the prevalence of errors is unavoidable and the need to correct these errors significantly slows down the rate of text entry .
the qwerty keyboard in particular is poorly suited as a medium for swipe input .
characteristics such as the `` u '' , `` i '' , and `` o '' keys being adjacent lead to numerous gesture ambiguities and potential input errors .
it is clearly not the optimal layout for gesture input .
the rise of digital keyboard use , first on stylus based keyboards in the 90 s and then on modern touchscreens a decade later , has led to a lot of research and development in breaking away from qwerty to a layout that s statistically more efficient .
this work resulted in various improved layouts for digital stylus keyboards such as the opti keyboard [ @xcite ] , the metropolis and hooke keyboards [ @xcite ] , and the atomik keyboard [ @xcite ] .
in addition to statistical efficiency , attempts were also made to improve statistical efficiency while simultaneously making the new layout as easy to use for novices as possible [ @xcite ] .
more recently , a few keyboards have been introduced that improve text input for certain situations on modern smartphones and tablets : optimizing for the speed of two - thumb text entry on tablets [ @xcite ] ; optimizing tap - typing ambiguity ( the swrm keyboard ) and simultaneously optimizing single - finger text entry for speed , reduced tap - typing ambiguity , and familiarity with the qwerty keyboard ( the sath keyboard ) [ @xcite ] ; and optimizing the autocorrect feature itself to simultaneously increase the accuracy of word correction and completion [ @xcite ] .
most of the aforementioned work was done specifically for touch typing since that is the most common form of text input on touchscreen devices .
however , the relatively recent rise in popularity of gesture typing has led to some interesting new keyboard layouts that were specifically optimized for improved gesture typing performance .
the square osk and hexagon osk keyboards were optimized to maximize gesture input speed using fitt s law [ @xcite ] .
various optimizations were also done by smith , bi , and zhai while maintaining some familiarity with qwerty by using the same layout geometry and only changing the letter placements .
this resulted in four new keyboards : the gk - c keyboard , which is optimized to maximize gesture input clarity ; the gk - s keyboard , which is optimized for gesture input speed ; the gk - d keyboard , which was simultaneously optimized for gesture clarity and speed using pareto optimization ; and the gk - t keyboard , which was simultaneously optimized for gesture clarity , gesture speed , and qwerty similarity [ @xcite ] . ]
evaluating and comparing various keyboard layouts is a difficult problem given the complexity and variability associated with text entry .
measuring text entry and error rates from user based trials is typically done to evaluate or directly compare the effectiveness of various keyboards and input methods .
these studies usually require the test subjects to transcribe a set of predefined phrases using a specified input device .
text entry evaluations of mini - qwerty keyboards [ @xcite ] , chording keyboards [ @xcite ] , handwriting recognition systems [ @xcite ] , and various gesture input systems [ @xcite , @xcite ] have all been done using this approach .
the main downside of this approach is the fact that in day - to - day use most users spend very little time transcribing phrases .
the majority of text entry is done by composing original phrases .
therefore , text entry evaluations from transcription based user studies are not realistic and can introduce unintended biases into the results .
vertanen and kristensson showed how these biases can be mitigated by including composition tasks in user trials to complement the standard transcription tasks [ @xcite ] . despite the recent work done to improve text entry evaluations with user based studies the metrics used for optimization
are typically based on surrogate models of the actual performance characteristic of interest .
for example , the gesture clarity metric used by @xcite is correlated with how frequently words are correctly reconstructed but does not measure this directly .
the reason that these approximate measures have been used is that accurately evaluating real keyboard reconstruction error rates would require an immense amount of user input data .
modern optimization techniques typically evaluate and compare hundreds of thousands of different keyboard layouts , making it completely infeasible to obtain the necessary user data .
the methodology that we propose allows for the direct evaluation of gesture reconstruction error rates , or any other desired metric , by simulating realistic user interactions with a keyboard .
this is similar to the approach used by fowler et al when they simulated noisy tap typing input to estimate the effect of language model personalization on word error rate [ @xcite ] . to demonstrate the effectiveness of our methodology
we will show how it can be used to find a keyboard layout that minimizes gesture input errors .
this requires accurately modeling gesture input for a given layout , interpreting the input , and quantifying how frequently typical inputs would be misinterpreted .
we employ several different models for gesture input and a dictionary of the most common words in the english language to simulate realistic usage and take into account variations between users .
we also attempt to develop a highly accurate algorithm for recognizing gesture inputs that is not limited to a specific keyboard layout .
it should be noted that although this paper focuses on the error rate performance , the overall methodology can be used to evaluate and compare keyboard layouts based on any performance measure . finally , in order to address the problem we designed and built an open source software framework , called dodona , for exploring different input methods .
this framework is well suited for examining a wide range of possible keyboard designs and input models .
it was built with optimization in mind and has a focus on efficient implementations and extensibility .
the library is freely available on github [ @xcite ] and was used to perform the analysis and generate all keyboard related graphics presented here .
an extremely large dataset of gesture inputs is needed in order to accurately evaluate the error rate of a given keyboard layout .
the only way to obtain such a dataset on a reasonable time - scale is to generate gesture input data based on models of user input . to accomplish this we developed several models which can take a word and produce what we refer to as a gesture input vector , a sequential series of @xmath0 points that represent discrete samples along a gesture input pattern .
we then used words that were randomly generated based on their frequency of use in the english language to feed into these models and generate realistic sets of input .
in general , our input model can produce either a `` random vector '' or a `` perfect vector '' .
the former is used for realistic , inexact gesture input while the latter represents the ideal input pattern that is free from variation . to construct random vectors
we begin by drawing control points for each letter in a given word from a two dimensional gaussian distribution that s centered around each corresponding key on the keyboard .
the @xmath1 and @xmath2 widths of the gaussian , in addition to the correlation in the offsets between subsequent control points , can be changed as parameters of the input model .
we then interpolate between these control points for each letter to produce a continuous gesture input as a function of time .
this is then sampled at evenly spaced intervals along the entire interpolation in order to produce an input vector with a set number of points .
perfect vectors are constructed in the same way but use the centers of the keys as control points . the idea that their exists a unique perfect vector for each word in the lexicon was first introduced by kristensson and zhai in their seminal paper about the shark@xmath3 text input system for stylus keyboards [ @xcite ] . in their work
they refer to perfect vectors as _
+ we chose to implement a variety of different interpolations to account for the variations in individual gesture input style .
we settled on five different interpolation techniques : a straight - line spatial interpolation , a natural cubic spline , a cubic hermite spline [ @xcite ] , a monotonic cubic hermite spline [ @xcite ] , and a modified natural cubic spline where the first and last segments are required to be straight lines . using randomly generated control points with various interpolation techniques allows us to capture a large range of input possibilities .
this is demonstrated in figure [ fig : five - different - random ] , which shows five different possible swipe patterns corresponding to the word `` cream '' .
each pattern was constructed with a different interpolation and random variations of the control points .
we used the google web trillion word corpus as compiled by peter norvig to provide the dictionary for our input model [ @xcite ] .
this dictionary contains approximately 300,000 of the most commonly used words in the english language and their frequencies . however , more than half of the entries are misspelled words and abbreviations .
our final dictionary only contained words from the google web trillion word corpus that also occurred in the official scrabble dictionary , the most common boys / girls names , or winedts us dictionary [ @xcite ] .
the result was a dictionary containing 95,881 english words and their frequencies .
the individual word frequencies ( magenta ) and the associated cumulative distribution ( green ) are shown in figure [ fig : individual - word - frequencies ] . in order to reduce the computational needs
associated with using a dictionary this large we elected to only use the top @xmath4 words in the optimization procedures described later . even though this is only @xmath5 of the words contained in the dictionary it accounts for @xmath6 of the total word occurrences .
furthermore , the average vocabulary size of a college educated adult in the u.s . is @xmath7 words with a range extending from about @xmath8 to @xmath9 words , which is consistent with the size of the dictionary used in this analysis [ @xcite ] .
word marker , which is where we cut off for inclusion in the analysis dictionary.[fig : individual - word - frequencies ] ]
in order to evaluate the gesture clarity of a given keyboard layout , a quantitative metric must be defined . in the recent paper by smith , bi , and
zhai they define gesture clarity as the average distance between a vector representation of each word and its nearest neighbor on a given keyboard layout [ @xcite ] .
the vector used to represent each word corresponds to what they call its _ ideal trace _
( identical to _ perfect vectors _ defined in this paper ) .
this definition is naturally related to how effective a keyboard will be since keyboards with smaller distances between nearest neighbors will tend to have more frequent reconstruction errors .
however , there are a number of important factors that it does not take into account : more than one neighboring word can be a source of recognition errors , there are threshold distances beyond which the impact on gesture clarity is negligible , and there are countless subtle interplays between the specific layout of a keyboard and the way that users input imperfect gestures .
therefore , a different procedure for computing something akin to gesture clarity is required if we want to take these effects into account and more accurately reflect the frequency of words being successfully reconstructed .
for this reason we decided to use the _ gesture recognition error rate _ for a given keyboard layout and lexicon as our metric . ideally , this would be measured experimentally but this is time consuming , expensive , and essentially impossible when the goal is to evaluate a large number of keyboards as would be done in an optimization . in the absence of real user testing , modeling and simulating how users input gestures allows for an approximate evaluation of the actual error rate .
the error rate , @xmath10 , for a given keyboard layout can be approximated by generating @xmath11 random input vectors from random words in the lexicon .
the generated input vectors can then be reconstructed using a realistic algorithm and checked against the words that they were meant to represent . if @xmath12 input vectors were reconstructed as the wrong word then
the recognition error rate is simply @xmath13 .
this quantity can very roughly be thought of as relating to the gesture clarity , @xmath14 , according to @xmath15 , though this relationship is just a heuristic .
as mentioned in the previous paragraph , there are subtle nuances that contribute to gesture input errors that can affect the error rate of a given keyboard layout but not its gesture clarity .
since our error rate metric depends on how accurately we can recognize gesture inputs on a given keyboard layout we need a gesture input recognition algorithm that can accurately recognize input vectors on any keyboard layout . using our input model
we can define gesture input recognition as the ability correctly match a random input vector with the word that generated it .
this is a difficult problem to solve because as a user passes their finger / stylus over the keyboard they typically pass over many more characters than those required to spell the desired word .
this means that we must be able to pick out a particular word from an ambiguous input .
if you look at the example gesture input pattern for the word `` whole '' in figure [ fig : example - swipe - collision ] , it is easy to see how even differentiating between two words can be a challenge .
our first approach to this problem was to simply take the euclidean distance between two input vectors .
this requires each input vector to be normalized so that they each have the same number of interpolation points , which are equally spaced along the interpolation .
implementing the euclidean distance approach is then straightforward and given by the equation , @xmath16},\label{eq : cartdist}\ ] ] where @xmath17 is the total number of interpolation points in the gesture input vector and @xmath18 is the @xmath1-component of the @xmath19th interpolation point in the first of the two input vectors being compared .
this is very similar to the _ proportional shape matching _
channel used in the shark@xmath3 writing system [ @xcite ] and in the gesture clarity metric used by @xcite .
although this method can correctly identify when two gesture inputs match exactly , it could also return a large distance between two input vectors that are qualitatively very similar .
for example , a user may start near the bottom of the key for the first letter of the word and end up shifting the entire gesture input pattern below the centers of the subsequent keys .
this input pattern could pass over all of the correct keys but still result in a large euclidean distance when compared to the perfect vector for the word .
the shortcomings of this approach made it clear that we were not utilizing all of the useful information contained in the input vectors . if a poor distance measure were to cause misidentifications that would not happen in practice then this could introduce significant biases during the optimization procedure , resulting in a keyboard that is not optimal for real use .
kristensson and zhai accounted for this in the shark@xmath3 writing system by incorporating language and location information in addition to proportional shape matching in their recognition algorithm . similarly , in order to reduce the impact of these systematic effects , we needed to identify additional features that would improve our gesture input recognition .
our first step was to uncouple the @xmath1 and @xmath2 coordinates and treat them individually . given the anisotropic nature of most keyboards , the relative importance of biases between the two spatial dimensions is not clear a priori .
therefore , we decided that the first two features should be the euclidean distance ( eq .
[ eq : cartdist ] ) between two input vectors for the @xmath1 and @xmath2 components individually , @xmath20 are also shown in each plot.[fig : a - closer - look],title="fig : " ] are also shown in each plot.[fig : a - closer - look],title="fig : " ] are also shown in each plot.[fig : a - closer - look],title="fig : " ] are also shown in each plot.[fig : a - closer - look],title="fig : " ] are also shown in each plot.[fig : a - closer - look],title="fig : " ] are also shown in each plot.[fig : a - closer - look],title="fig : " ] in order to address the issue with offset input vectors , translational symmetry needed to be taken into account . to do this
we decided to look at the time derivatives of the input vectors with respect to their @xmath1 and @xmath2 coordinates . since the input vectors are sets of sequential , discrete points we can easily estimate the derivative at each point .
we can then construct a distance measure by taking the euclidian distance between the time derivatives of two swipe patterns at each point .
the equation for the derivative distance in the @xmath1-dimension is given by : @xmath21^{2}}},\label{eq : ddist}\ ] ] where @xmath22 and @xmath23 are the @xmath1-components of the two input vectors being compared .
we assume a constant input velocity for the models considered here so we ve implicitly rescaled the time coordinated such that @xmath24 to simplify the equations .
we also wanted a distance measure that would be more sensitive to the positions of sharp turns in a gesture input pattern .
this led us to include the distance between the second derivatives of the input vectors in a similar fashion to the first derivatives ( eq . [ eq : ddist ] ) .
the quantity @xmath25 is rotationally invariant as well so we can see how these might help allow for more leniency in situations where there might be some variation in the orientation of a touchscreen device relative to a users hand .
the utility of these features in regards to correctly identifying gesture input is apparent when you take a closer look at the differences between a random input vector and a perfect vector for a given word .
the @xmath1 and @xmath2 values as a function of time for a random input vector and a perfect vector for the word `` cream '' , as well as their first and second derivatives , are shown in figure [ fig : a - closer - look ] .
this example illustrates how the first and second derivatives can be useful for finding the correct match even when the swipe pattern is shifted significantly from the center of the appropriate keys .
two additional distinguishing features of each gesture input pattern are the start and end positions .
these points carry information about the overall position of an input vector while being less sensitive to the shape than @xmath26 and @xmath27 .
the addition of this information to the euclidean distance was shown by @xcite to reduce the number of perfect vector ambiguities by 63% in their 20,000 word lexicon .
consequently , the distance between the @xmath1 and @xmath2 components of the first and last points of the input vectors were included in the feature set .
this gives us four additional features to add to the six previously discussed .
finally , we realized that the length of each gesture input pattern can be a very distinguishing characteristic , leading us to include the difference in the length of each gesture input pattern as the final addition to our feature set , giving us a total of eleven features which are related to the difference between two gesture input patterns . however , in order avoid repeatedly calculating square roots we decided to put the squared value of each distance in the feature set .
to summarize , the set contains : the squared euclidean distance between the @xmath1 and @xmath2 components , the squared distance between the @xmath1 and @xmath2 components of the first derivatives , the squared distance between the @xmath1 and @xmath2 components of the second derivatives , the squared distance between the @xmath1 and @xmath2 components of the first point , the squared distance between the @xmath1 and @xmath2 components of the last point , and the difference in the squared length of the two gesture input patterns being compared . the last step in creating our desired distance measure was figuring out a way to combine the eleven elements in our feature set to a single output representing the `` distance '' between two gesture input vectors . despite the intuitive basis of our feature set
, the relationship of each element to the overall performance of a classifier is highly non - trivial given the complexity of gesture input vectors .
fortunately , this is a problem that is well suited for a neural network based solution . to build a classifier using our feature set
, we created a deep , fully - connected artificial neural network with eleven input nodes ; one for each of the variables in the previously discussed feature set .
the network architecture consists of three hidden layers with 11 nodes each and a fourth hidden layer with only two nodes .
the activation function for each hidden node and the output node is an elliot function , @xmath28 where @xmath29 is the steepness and is set to 0.5 for each layer .
this is a computationally efficient sigmoid function that asymptotes slower than a standard sigmoid .
this was necessary since we employed the rprop algorithm [ @xcite ] to train the network , which is susceptible to the flat - spot problem when using steep sigmoid activation functions .
the artificial neural network used in the analysis was implemented using the fast artificial neural network software library ( fann ) [ @xcite ] and is available in the repository as a fann binary file and as a text file listing the weights of every connection in the deep neural network [ @xcite ] .
-axis ) and the neural network distance measure ( @xmath2-axis ) .
the dashed line is just @xmath30 and is only meant to help the comparison . the average error rate for each method is represented by the bold green dot.[fig : the - performance - of - nn ] ] the neural network was trained on a dataset of @xmath31 pairs of random input vectors and perfect vectors .
the random input vectors were constructed with each of the different interpolation models used ( @xmath32100,000 for each model ) but the perfect vectors were restricted to be linear interpolations .
this was done to make the algorithm as realistic as possible since a practical gesture recognition algorithm would not be able to make any assumptions about a user s input style .
the training set was divided up so that @xmath33 of the input pairs corresponded to the same word , @xmath34 corresponded to completely random words , and the remaining @xmath35 corresponded to different but similar words . the exact definition of `` similar words '' is given in the next section
. the performance of the neural network recognition algorithm can be seen in figure [ fig : the - performance - of - nn ] .
this plot shows a comparison between the neural network method and the euclidean distance method . in this study
, 50 random keyboards layouts were created and for each layout 5,000 random gesture input vectors were generated .
the input vectors were then matched to a word using the two methods .
the fraction of attempts that each method got wrong is shown as the error rate ( as discussed in much more detail in the next section ) .
it is easy to see that for each keyboard layout the neural network recognition algorithm outperformed the euclidean distance algorithm .
the average error rate using the neural network algorithm on random keyboard layouts is @xmath36 compared to @xmath37 for the euclidean distance measure , which is a reduction of @xmath38 in the gesture input recognition error rate . with a more accurate gesture recognition algorithm
we can confidently evaluate a given keyboard s gesture recognition error rate .
the general approach is to use a monte carlo based algorithm to determine the error rate .
this technique can be described as follows : a random word is chosen from the lexicon with a probability proportional to its frequency of occurrence .
a random gesture input vector is then generated for this word based on a given input model .
the gesture recognition algorithm then determines the word that is the best match for that specific random input vector .
if the selected word matches the original word then the match is considered a success . this process is repeated @xmath11 times so that the error rate is given by the ratio of successful matches to the total number of attempts . due to the statistical nature of this technique
there will be an uncertainty in each measurement .
as with most efficiency calculations , the uncertainty is given by the variance of a binomial distribution scaled by @xmath39 .
although effective , this method is very computationally intensive .
a reasonable optimization procedure will require around @xmath40 matching attempts in each efficiency calculation to reduce the effects of statistical fluctuations ( specifically , this produces error rate calculations with statistical uncertainties of @xmath320.7% ) .
each matching attempt requires a comparison for every word in the lexicon , which contains @xmath4 words , so every efficiency determination will require @xmath41 distance measure calculations .
since the goal is to use the error rate calculation in an optimization procedure , increasing the total time by several orders of magnitude , another approach was needed .
consider the case where a user is trying to input the word `` pot '' on a standard qwerty keyboard .
clearly words such as `` cash '' and `` rules '' are not going to be calculated as the best match by the distance measure because they have dramatically different gesture input patterns .
therefore , there is no need to spend time comparing the perfect vectors of these words as part of the error rate calculation .
the error rate calculation can be made much faster without sacrificing much accuracy by comparing only to the perfect vectors of more plausible candidate words such as pit " , put " , `` lit '' , etc .
the difficulty lies only in determining which words are plausible candidates . ] to determine what words should be included in the candidate list we created what we call the `` string form '' for each gesture input vector .
the string form is just the sequential list of letters that the gesture input pattern traversed .
for example , if a user inputs the word `` pot '' the corresponding string form might be `` poiuyt '' .
if we were implementing an algorithm for determining candidates given a fixed keyboard layout we could first generate a large number of random input vectors for every word in the dictionary .
we could then build a lookup table where each observed string form corresponds to a list of all words that had been observed to ever produce that string form .
that approach is unfortunately not possible when optimizing the keyboard layout because the lookup table would need to be rebuilt for every new keyboard .
instead , we generate a large number of random vectors for the word that would be the correct match and find the string form for each of those .
we then allow any words that are contained within any of these string forms to be potential candidates .
this would not be possible if we did nt know the intended word but it results in a superset of the candidates we would find using the precomputed lookup table given that a sufficient number of random vectors are used .
the words that are consistent with each string form are determined by recursively searching a radix tree representation of all of the words in the dictionary as shown in figure [ fig : a - radix - tree ] . in this example
we would start by looking at the radix tree that begins with the letter `` p '' .
then we would look to see if this tree has a child node corresponding to the letter `` p '' ; repeated letters are not always distinguishable in a swipe pattern so we have to always explicitly check for the possibility . since there is no word that begins with `` pp '' we then move on to the next letter in the string form , `` o '' , and look for a child node corresponding to this letter .
the search is then done recursively for the sub - tree beginning on the child node corresponding to `` o '' with the string form `` oiuyt '' .
the search continues in this recursive manner , traversing all of the branches of the subtree that are consistent with the string form and returning any leaf that is found as a candidate word .
this will effectively find all candidate words beginning with `` po '' that could potentially be contained in the string form .
once this subtree is exhausted we move on to the next letter of the original string form , `` i '' , and recursively search the subtree corresponding to this letter with the string form `` iuyt '' for candidate words beginning with `` pi '' .
this process continues until the final subtree corresponding to the letter `` t '' is searched , thus finding all candidate words contained in the string form . ]
this approach , which we call _ radix tree pruning _ ( abbreviated as `` radixmc '' when combined with the standard monte carlo algorithm ) , reduces the number of comparisons to make for each input vector and subsequently speeds up the calculation significantly .
the time required scales roughly linearly with the number of random vectors which are used to find candidates so some balance between efficiency and accuracy is required . in order to determine a suitable threshold we calculated the error for a given random keyboard as a function of the number of random vectors used in the radix tree pruning as shown in figure [ fig : radix_vs_nvectors ] . the flattening out of the error rate
is expected since the words that are most similar are typically found in the earlier iterations .
we chose to use 20 random vectors in the pruning step since they allow for the vast majority of nearest neighbors to be found and can be generated in a reasonable amount of time .
we conducted a small study where we calculated the error rate while varying the number of random vectors from one to 25 for the qwerty keyboard and four random keyboard layouts .
the results showed that the relative error rate of the random keyboards to qwerty remained roughly constant across the entire range of the number of random vectors .
the relationship between the error rate and the number of input vectors is largely independent of the keyboard layout , which means the relative error rates remain approximately unchanged .
when performing an optimization , where only relative error rates are important , this effect becomes largely negligible .
when used in the error rate calculation this algorithm outperforms the standard monte carlo approach in terms of computational efficiency by two orders of magnitude when the full dictionary is used , as seen in figure [ fig : the - cpu - time ] .
it is also obvious that the radix tree based monte carlo algorithm scales much more favorably with the size of the dictionary .
a similar problem was faced by kristensson and zhai during the development of shark@xmath3 .
however , they opted for a different filtering technique , which they called _
template pruning_. this required that the start and end positions of the input vector and a perfect vector be less than a given threshold in order to be considered for a potential match [ @xcite ] . ]
both of these approaches have the same goal : to drastically reduce the number of comparisons necessary to determine the intended input word .
the main difference lies in the trade - off between the computational complexity of the filtering and the number of comparisons made during the matching stage .
while computationally very efficient , template pruning has the consequence of passing many words to the comparison step that would be filtered using the radix tree approach . using kristensson s and zhai s method with a pruning threshold large enough to allow for consistent word reconstruction
, we were able to prune the list of comparison words by 95.08% on average .
the radix tree pruning technique with 20 random vectors was able to reduce the necessary list of comparison words by 99.92% . by removing so many additional candidate words that have nearby starting and ending keys but dissimilar input vectors , we were able to significantly reduce the time required for the combined filtering and matching stages .
in order to demonstrate the utility of our methodology and the dodona open source framework we optimized permutations of a standard keyboard layout to the minimize gesture recognition error rate . the specific optimization algorithm we employed begins with generating a random keyboard layout and calculating its error rate . at each successive iteration
a new keyboard is created by swapping @xmath12 pairs of keys in the previous keyboard .
the error rate of the new keyboard is calculated for each interpolation method and averaged .
the average error rate is then compared to the previous keyboard .
if the error rate has decreased then the new keyboard is kept and the previous one is discarded , otherwise the new keyboard gets discarded .
this process is repeated @xmath11 times where the number of key swaps , @xmath12 , is repeatedly decreased by one at set intervals .
this results in successive keyboards differing by only one swap at the end of the optimization procedure . for our final analysis
we ran 256 separate optimizations , each running through 200 iterations , and starting with @xmath42 .
the average error rate for all of the input models at each step in the optimization procedure is shown in figure [ fig : the - average - error ] . the minimum and maximum error rate at each step and the error rate of the qwerty keyboard
are also shown in the figure .
interestingly , we see that the qwerty keyboard error rate of @xmath43% is less efficient for gesture input than the average randomly generated keyboard layout ( error rate : @xmath3213% ) .
however , the optimization procedure quickly finds keyboard layouts with even lower error rates .
after two hundred iterations the average error rate found in each trial is approximately 8.1% .
this represents an improvement in the error rate of 47% over the qwerty keyboard .
] the optimal keyboard for gesture input clarity found in the analysis is shown in figure [ fig : the - most - optimal ] and we will refer to it by its first four letters : _ dghp_. the keys are colored to represent their relative frequency in the lexicon .
this keyboard is found to have an error rate of @xmath44 , which is a @xmath45 improvement compared to the qwerty keyboard .
this is higher than the minimum shown in figure [ fig : the - average - error ] because of the limited resolution of the error rate measurement used in the optimization procedure
. the quoted error rate of the optimal keyboard was determined by a final high precision measurement .
it is important to note that the values of the error rates computed by our method depend heavily on the parameters of the input model .
thus , the error rates themselves hold little general meaning . instead , it is more meaningful to speak of the relative change in error rate compared to the standard qwerty keyboard .
we have found the ratios of the error rates produced from different keyboards to be largely independent of the input model parameters .
this permits us to state , in general , whether the keyboard layout resulting from the above procedure is more efficient than the qwerty keyboard for swipe input and quantify the relative improvement . ]
one major feature of the dghp keyboard is the appearance of the most frequently used keys along the edges .
the qwerty keyboard has many of the most frequently used keys near its center , which results in large number of gesture input errors . in this gesture
input optimized keyboard the keys at the center of the layout are less frequently used .
this makes sense because having the most frequently used keys at the edges will decrease the probability that the user passes over them without intending to . by removing the most common letters from appearing arbitrarily in gesture input patterns we naturally reduce the number of errors
however , there are more subtle characteristics of the keyboard that arise due to the way words are structured in the english language . for example , the letters `` i '' , `` o '' , and `` u '' are no longer clustered together , which eliminates the ambiguity between words like `` pout '' , `` pit '' , and `` put '' .
in addition , another notable feature is the separation of the letters `` s '' , `` f '' , and `` t '' .
this helps to distinguish between the words `` is '' , `` if '' , and `` it '' , which are very common in the english language .
it s interesting to try and understand some of the reasons why the keyboard has such a low error rate but , in reality , it is a finely tuned balance that depends on the structure and frequency of every word used in the analysis . out of curiosity
we also decided to see what would happen if we optimized a keyboard to maximize swipe errors .
we ran five similar optimization procedures through 100 iterations to find the least optimal keyboard layout for swipe input .
the worst keyboard we could find is shown in figure [ fig : the - least - optimal ] and has an error rate of 27.2% , which is about 78% worse than the qwerty keyboard . in this keyboard
the most frequently used keys are all clustered together , making swipe patterns more ambiguous and resulting in more swipe errors . ] besides running optimization procedures , evaluating and comparing existing keyboard layouts is another important reason to have a robust framework for evaluating keyboards .
we evaluated the gesture recognition error rate for a number of existing virtual keyboard layouts , similarly to what was done by @xcite when he compared the tapping speed for almost the same set of keyboard layouts .
in addition to the ones presented in his paper we also included the four keyboards from @xcite ( gk - c , gk - s , gk - d , gk - t ) and the optimal keyboard found here that was presented in the previous section , dghp .
the results are displayed in table [ table : keyboard_evaluations ] .
the error rate and statistical uncertainty associated with each calculation are displayed for each keyboard .
the error rates were calculated using 20 , 000 monte carlo iterations , 20 random vectors in the radix tree pruning step , and random input vectors containing 50 spatially ( linearly ) interpolated points . as has been noted previously ,
the absolute value of the error rates is not meaningful unless training data is used to constrain the model .
when only the permutations of keys on a single keyboard geometry are considered , however , then the relative error rates of the different layouts can be directly compared because the degree of randomness in the input model scales all of the error rates in a very similar way .
the comparison of error rates becomes far more subtle when considering entirely different keyboard geometries , particularly if they included keys of different shapes ( e.g. squares and hexagons ) .
when working with multiple keyboard geometries , there are several different possible approaches for adjusting the input model uncertainty and scaling the relative sizes of the keyboards . in the results shown here ,
each keyboard was scaled such that the total area of its keys would be the same for all keyboards .
we then set the input model uncertainties , @xmath46 and @xmath47 , to be the same for all keyboards with values equal to @xmath48 , where @xmath49 is the area of a single key .
it is quite possible , if not likely , that the size and shape of different keys may have an affect on the magnitude of @xmath46 and @xmath47 in real usage . as an alternative approach
, we also considered the case where @xmath46 and @xmath47 scale directly with the horizontal and vertical extent of each key , respectively .
this resulted in a 10 - 25% increase in the error rates of the hexagonal keyboards relative to those with square keys and so this should be taken into account when considering the results in table [ table : keyboard_evaluations ] .
l*2c _ keyboard layout _ & _ gesture recognition error rate _ & _ error rate uncertainty _ + gk - c & @xmath50 & @xmath51 + swrm & @xmath52 & @xmath53 + gk - d & @xmath54 & @xmath55 + gk - t & @xmath56 & @xmath57 + hexagon qwerty & @xmath58 & @xmath59 + atomik & @xmath60 & @xmath61 + metropolis ii & @xmath62 & @xmath63 + square alphabetic & @xmath64 & @xmath65 + gag i & @xmath66 & @xmath55 + wide alphabetic & @xmath67 & @xmath68 + metropolis i & @xmath69 & @xmath70 + sath - trapezoid & @xmath71 & @xmath72 + sath - rectangle & @xmath73 & @xmath74 + chubon & @xmath75 & @xmath72 + getschow et al . &
@xmath76 & @xmath77 + opti i & @xmath78 & @xmath65 + square osk & @xmath79 & @xmath80 + square atomik & @xmath81 & @xmath82 + hexagon osk & @xmath83 & @xmath57 + qwerty & @xmath84 & @xmath85 + fitaly & @xmath86 & @xmath87 + quasi - qwerty & @xmath88 & @xmath89 + hooke & @xmath90 & @xmath91 + gk - s & @xmath92 & @xmath53 + gag ii & @xmath93 & @xmath94 + lewis et al . &
@xmath95 & @xmath96 + opti ii & @xmath97 & @xmath94 + dvorak & @xmath98 & @xmath99 + dghp & @xmath100 & @xmath101 + we can see that the results vary widely , which is expected given the drastic differences between some of the keyboard layouts .
the first thing to note is that the top two performing keyboards - dghp and gk - c - were the only two keyboards that were optimized for gesture input clarity or gesture recognition error rate and show a reduction in error rate compared to qwerty by @xmath102 and @xmath103 , respectively .
the fact that dghp outperforms gk - c is also expected since dghp was optimized using the exact metric used in this evaluation .
the swrm keyboard came in third even though this keyboard was optimized for input clarity with respect to tap input , not gesture input .
the gk - d and gk - t keyboards were optimized for gesture clarity but they were simultaneously optimized for one and two other performance metrics , respectively , so it is not surprising that they came in fourth and fifth .
although the difference between the two is not statistically significant , the order they appear is exactly what we would expect .
the worst keyboard by far is dvorak which has an error rate that is @xmath104 higher than qwerty .
we can also see that the hexagon keyboards tend to perform better than the square keyboards , although this comparison is not necessarily accurate given the subtleties mentioned in the previous paragraph .
it also interesting to see that just using an alphabetic keyboard will give you a much lower error rate than the majority of the keyboards listed in table [ table : keyboard_evaluations ] .
the results presented here show a clear demonstration of our proposed methodologies effectiveness with respect to calculating gesture recognition error rates and performing keyboard optimizations .
in contrast to recent work , such as the gesture clarity optimization performed by smith , bi , and zhai , this new approach allows for the direct estimation of the error rate [ @xcite ] . this distinction may appear subtle , but it allows for ambiguities other than nearest neighbors to be taken into account .
these secondary ambiguities have a sizable impact on the real - world performance of any particular keyboard .
additionally , threshold effects are more realistically taken into account with the reconstruction error rate estimation .
words that have distant nearest neighbors will all have reconstruction rates that are effectively @xmath105 . increasing the distance between these already distant words will increase the gesture clarity metric while having no effect on the actual reconstruction error rate . in an optimization setting , these ineffectual changes might be preferred at the expense of making words with closer nearest neighbors more ambiguous , resulting in keyboards that are less effective overall .
although these changes are significant , we consider the primary advancement of this methodology to be its ability to extend even further in quantifying metrics that accurately reflect the performance of different keyboards . in this section
, we will discuss some of the directions where this work and the dodona framework can be built upon in future research .
one of the key advantages of the monte carlo evaluation approach is that any desired metric can be computed . in the smith ,
bi , and zhai paper , linear combinations of gesture clarity and gesture speed are optimized and then pareto optimal configurations are found .
this is effective for finding keyboards that have relatively high gesture clarity and gesture speed metrics but requires making assumptions about the relative importance of each metric when choosing a keyboard that is optimal overall . in past work
, an optimal keyboard has typically been chosen from along the pareto front under the assumption that each metric has equal importance [ @xcite , @xcite , @xcite ] . while this is a reasonable approach ,
there are certain cases where two metrics can be combined to more directly measure the overall performance .
this is the case with gesture clarity / error rate and gesture speed since the overall goal of minimizing the error rate is to speed up gesture input by minimizing the time required to correct mistakes . with the monte carlo approach it is possible to do this by estimating the time it takes to input a gesture and then adding the correction time when a word is incorrectly reconstructed .
for example , the gesture input time could be estimated with the model used in @xcite and the correction time could be estimated by adapting the approach used in @xcite for gesture input .
this would allow for the direct estimation of how many words per minute can be entered on any given keyboard layout , leading to a metric that intuitively relates much more closely to how effective that layout would be in real - word usage .
the ability to reduce gesture speed and error rate into a single metric has the additional benefit of reducing the problem to one of scalar , rather than vector , optimization . roughly speaking ,
this allows you to focus the search in the direction that will directly maximize the underlying desired metric instead of optimizing in many different directions to extend the pareto front ( as was done in @xcite ) .
this significantly reduces the computational requirements of the optimization procedure .
extensions like this are another possible area for future work and can be made very easily with the dodona framework while adding virtually nothing to the overall computation time .
another major strength of this approach is that it can be applied to an extremely wide variety of input methods with only the input models , and possibly the keyboards , needing to be modified .
detailed discussion is beyond the scope of this paper , but the dodona framework includes a model for touch typing which can be used to evaluate autocorrect on traditional keyboards or disambiguation in text entry on a t9 touchpad ( on a touchpad keyboard which is also included in the framework ) . new keyboard designs that use any sort of touch or gesture typing , with or without disambiguation , can be evaluated using the same overall methodology and framework .
this allows for extremely quick prototyping and testing of novel text input methods .
reconstruction algorithms can also be easily modified to reflect more state of the art techniques . instead of considering individual word frequencies , word pair frequencies
could be used to generate pairs of words and the previous word could modify the prior probabilities of the reconstructed words .
this would more closely mimic the behavior of commercially available gesture typing software and allow for more realistic estimation of metrics .
the algorithms could also be extended to include forgiveness of certain types of mid - gesture mistakes or allowing for common spelling mistakes .
this degree of flexibility is something that can only be achieved in a framework where actual reconstruction is taking place instead of relying on heuristic metrics .
one of the disadvantages of the monte carlo methodology is that it depends on the accuracy of the model .
a key improvement that could be made in future analyses is to utilize user data to tune and validate the input models .
the input models could also be extended to include more subtle aspects of user behavior such as mid - gesture mistakes and misspellings . with well trained models , user behavior
could be incorporated that would otherwise be impossible to take into account .
this would additionally make the exact values of the metrics meaningful while the untrained models used in this paper can only be used for relative keyboard comparisons .
one way this could be done is to incorporate touch accuracy into the gesture input models .
the model presented in this paper assumes that the accuracy of each key is identical .
however , henze , rukzio , and boll used a crowd sourced dataset to show that tap accuracy is systematically skewed on mobile devices [ @xcite ] .
furthermore , they showed that the frequency of touch errors for a specific target is correlated with the absolute position of the target on the touchscreen .
this could lead one to believe that certain keys are more susceptible to touch , and possibly gesture , inaccuracies than others .
therefore , an obvious improvement to our user input model would be to systematically incorporate these results to adjust the accuracy at each location for a given keyboard geometry .
in addition , language model personalization could be incorporated and compared to the overall effect of keyboard layout on the error rate .
fowler et al showed that including language model personalization can dramatically reduce the word error rate for touch typing [ @xcite ] .
we have described a new way to model gesture input and a procedure to evaluate the error rate of any touchscreen keyboard layout . using this method
we have evaluated and compared the error rates for numerous existing virtual keyboards and , specifically , shown that the qwerty keyboard is far from optimal for using a gesture input mechanism .
we have also described an optimization procedure for finding a keyboard that is better suited for gesture input .
we presented the most optimal keyboard that was found in our analysis and showed that it decreased the error rate for gesture input by @xmath106 when compared to the qwerty keyboard .
we thank duncan temple lang and paul baines of the statistics department at uc davis for allowing us access to the computing cluster which provided the necessary computational power to run the final optimizations .
in addition , we thank daniel cebra and manuel calderon de la barca sanchez of the physics department at uc davis for allowing us to make use of the nuclear physics group computing resources while designing and testing the analysis software .
dunlop:2012:mpo:2207676.2208659 mark dunlop and john levine .
multidimensional pareto optimization of touchscreen keyboards for speed , familiarity and improved spell checking . in _ proceedings of the sigchi conference on human factors in computing systems _ _ ( chi 12)_. acm , new york , ny , usa , 26692678 . 978 - 1 - 4503 - 1015 - 4 http://dx.doi.org/10.1145/2207676.2208659 shark2 per - ola kristensson and shumin zhai .
shark2 : a large vocabulary shorthand writing system for pen - based computers . in _ proceedings of the 17th annual acm symposium on user interface software and technology _ _ ( uist 04)_. acm , new york , ny , usa , 4352 . 1 - 58113 - 957 - 8 + http://dx.doi.org/10.1145/1029632.1029640 fastswiping jochen rick .
performance optimizations of virtual keyboards for stroke - based text entry on a touch - based tabletop . in _ proceedings of the 23nd annual acm symposium on user interface software and technology _ _ ( uist 10)_. acm , new york , ny , usa , 7786 .
978 - 1 - 4503 - 0271 - 5 http://dx.doi.org/10.1145/1866029.1866043 googlekeyboard brian a. smith , xiaojun bi , and shumin zhai .
2015 . optimizing touchscreen keyboards for gesture typing . in _ proceedings of the 33rd annual acm conference on human factors in computing systems
_ _ ( chi 15)_. acm , new york , ny , usa , 33653374 .
978 - 1 - 4503 - 3145 - 6 http://dx.doi.org/10.1145/2702123.2702357 shark shumin zhai and per - ola kristensson .
shorthand writing on stylus keyboard . in _ proceedings of the sigchi conference on human factors in computing systems
_ _ ( chi 03)_. acm , new york , ny , usa , 97104 .
1 - 58113 - 630 - 7 http://dx.doi.org/10.1145/642611.642630 shapewriter shumin zhai , per ola kristensson , pengjun gong , michael greiner , shilei allen peng , liang mico lui , and anthony dunnigan .
shapewriter on the iphone - from the laboratory to the real world _
( chi 2009)_. acm .
http://dx.doi.org/10.1145/1520340.1520380 splines richard h. bartels , john c. beatty , and brian a. barsky .
hermite and cubic spline interpolation .
ch . 3 in _ an introduction to splines for use in computer graphics and geometric modeling . _
( 1998 ) monotonicsplines randall l. dougherty , alan edelman , and james m. hyman .
positivity- , monotonicity- , or convexity - preserving cubic and quintic hermite interpolation .
_ mathematics of computation_. * 52 * ( 186 ) : 471 - 494 .
http://www.ams.org/journals/mcom/1989-52-186/s0025-5718-1989-0962209-1/home.html touchaccuracy niels henze , enrico rukzio , and| susanne boll .
100,000,000 taps : analysis and improvement of touch performance in the large . _
mobilehci 11 - proceedings of the 13th international conference on human computer interaction with mobile devices and services_. pages 133 - 142 .
http://doi.acm.org/10.1145/2037373.2037395 composition keith vertanen and per ola kristensson .
complementing text entry evaluations with a composition task .
_ acm transactions on computer - human interaction_. volume 21 issue 2 , february 2014 .
miniqwerty edward clarkson , james clawson , kent lyons , and thad starner .
an empirical study of typing rates on mini - qwerty keyboards . in _
chi05 : extended abstracts on human factors in computing systems_. http://doi.acm.org/10.1145/1056808.1056898 twiddler kent lyons , thad starner , and brian gane . 2006 .
experimental evaluations of the twiddler one - handed chording mobile keyboard . _
human - computer interaction_. volume 21 , pp .
343 - 392 .
http://dx.doi.org/10.1207/s15327051hci2104_1 graffiti steven j. castellucci and scott i. mackenzie .
graffiti vs. unistrokes : an empirical comparison . in _
chi08 : proceedings of the sigchi conference on human factors in computing systems .
_ http://doi.acm.org/10.1145/1357054.1357106
joystick jacob o. wobbrock , duen horng chau , and brad a. myers . 2007 .
an alternative to push , press , and tap - tap - tap : gesturing on an isometric joystick for mobile phone text entry . in _
chi07 : proceedings of the sigchi conference on human factors in computing sytems_. http://doi.acm.org/10.1145/1240624.1240728 handwriting per ola kristensson and leif c. denby .
text entry performance of state of the art unconstrained handwriting recognition : a longitudinal user study . in _
chi09 : proceedings of the sigchi conference on human factors in computing systems_. http://doi.acm.org/10.1145/1518701.1518788 fowler andrew fowler , kurt partridge , ciprian chelba , xiaojun bi , tom ouyang , and shumin zhao .
effects of language modeling and its personalization on touchscreen typing performance . in _
chi15 : proceedings of the 33rd annual acm conference on human factors in computing systems_. http://doi.acm.org/10.1145/2702123.2702503 arif ahmed sabbir arif and wolfgang stuerzlinger .
predicting the cost of error correction in character - based text entry technologies . in em
chi10 : proceedings of the sigchi conference on human factors in computing systems . | gesture typing is a method of text entry that is ergonomically well - suited to the form factor of touchscreen devices and allows for much faster input than tapping each letter individually .
the qwerty keyboard was , however , not designed with gesture input in mind and its particular layout results in a high frequency of gesture recognition errors . in this paper , we describe a new approach to quantifying the frequency of gesture input recognition errors through the use of modeling and simulating realistically imperfect user input .
we introduce new methodologies for modeling randomized gesture inputs , efficiently reconstructing words from gestures on arbitrary keyboard layouts , and using these in conjunction with a frequency weighted lexicon to perform monte carlo evaluations of keyboard error rates or any other arbitrary metric .
an open source framework , dodona , is also provided that allows for these techniques to be easily employed and customized in the evaluation of a wide spectrum of possible keyboards and input methods . finally , we perform an optimization procedure over permutations of the qwerty keyboard to demonstrate the effectiveness of this approach and describe ways that future analyses can build upon these results .
touchscreen keyboards , gesture input , model - based design , monte carlo simulation |
tidal interactions play an important role in dynamical processes in the two - body problem in close binary systems : star star ( binary star ) or star planet .
they can lead to such phenomena as synchronization and orbital circularization ( hut 1981 ; zahn 1977 ) as well as to the tidal capture ( press and teukolsky 1977 ) or disruption of an object ( star ) ( ivanov and novikov 2001 ) and the fall of the object onto the star ( rasio et al .
1996 ; penev et al . 2012 ; bolmont and mathis 2016 ) . in this paper
we consider the tidal interaction of two bodies : a star and a point source .
the point source can be both a star ( a neutron star , a white dwarf , etc . ) and a planet .
below we will call the point source a planet by implying that this can also be a star .
since the evolution time scales of the eccentricity or semimajor axis strongly depend on the orbital period of the binary system and for some stellar models can take values up to 108 yr or more for periods of about 5 days ( ivanov et al . 2013 ; chernov et al . 2013 ) , the evolution of the star itself should be taken into account on such time scales .
as the star evolves , the orbital parameters change due to tidal interactions . in this paper
we investigate the dynamical tides by taking into account the stellar evolution . for our study
we chose three types of stars with masses of one , one and a half , and two solar masses .
we consider all stars without allowance for their rotation and magnetic field and touch on the stellar physics itself superficially , as far as this problem requires .
the star of one solar mass at lifetimes @xmath3 yr closely corresponds to our sun and has a radiative core and a convective envelope on the main sequence .
the other two stars of one and a half and two solar masses are more massive and have a more complex structure .
these stars have a convective core and a radiative envelope on the main sequence ( a more precise structure is presented below ) .
the problem of determining the tidal evolution is reduced to the problem of determining the normal modes of stellar perturbations and to calculating the energy and angular momentum exchange in the star.planet system ( ivanov and papaloizou 2004 , 2010 ; papaloizou and ivanov 2010 ; lanza and mathis 2016 ) .
the low - frequency g - modes of the stellar oscillations play an important role in the theory of dynamical tides .
the tidal interactions are fairly intense at small periastron distances of the planet . for a periastron distance @xmath4 au , the dimensionless excitation frequency is @xmath5 , which corresponds to g - modes .
a large number of exoplanets in stellar systems have been discovered in the last few years owing to the kepler , superwasp , and other observational programs . in particular ,
short - period massive planets with an orbital period of a few days , the so - called hot jupiters , have been detected ( winn 2015 ) . as a rule ,
the hot jupiters have low eccentricities , which points to the importance of tidal interactions ( ogilvie 2014 ) .
the results of this paper can be directly applied to some of such systems .
for example , the system ybp1194 is a solar twin ( brucalassi et al .
a planet with a mass of 0.34@xmath6 and a period of only 6.9 days revolves around this star .
for such a short - period planet the dynamical tides must be fairly intense and must affect the orbital evolution .
predictions about the subsequent evolution of this planet can be made by analyzing this system .
one of the results of this evolution is the fall of the planet onto the star .
the possibility of such a fall has been considered in many papers ( see , e.g. , rasio et al .
1996 ; penev et al . 2012
; weinberg et al . 2012 ; essick and weinberg 2016 ) . rasio et al .
( 1996 ) considered the possibility of the fall of the planet onto the star due to quasi - static tides and provided a plot for solar - like stars that shows the threshold , as a function of planetary mass and orbital period , below which the planet falls onto the star .
penev et al .
( 2012 ) considered tides with a constant tidal quality factor @xmath7 specified phenomenologically . in reality , this factor will depend on the planet s orbital period ( ivanov et al .
2013 ) and stellar age .
essick and weinberg ( 2016 ) took into account the energy dissipation due to nonlinear interaction of modes with one another .
in contrast to our approach ( ivanov et al .
2013 ; chernov et al .
( 2013 ) , the simultaneous solution of a large number of ordinary differential equations for each stellar model is suggested , with only solar - type stars having been considered .
in this paper we consider the evolution of stars with masses of one , one and a half , and two solar masses .
data on the stars are presented in tables 1 3 .
a novelty of this study is a consistent allowance for the stellar evolution . for each moment of the star
s lifetime we calculated the spectra of normal modes , the overlap integrals ( press and teukolsky 1977 ) , which are a measure of the intensity of tidal interactions ( for a generalization to the case of a rotating star , see papaloizou and ivanov 2005 ; ivanov and papaloizou 2007 ) , and the time scales of the orbital parameters ( the tidal quality factor @xmath7 ) .
the overlap integrals are directly related to the tidal resonance coefficients that were introduced by cowling ( 1941 ) ( see also rocca 1987 ; zahn 1970 ) .
all of the quantities marked by a tilde are dimensionless ; the normalization is presented in appendix a.
such a quantity as the overlap integral q is of great importance in the theory of dynamical tides .
it is specified by the expression ( press and teukolsky 1977 ; zahn 1970 ) @xmath8 where m is the stellar mass , r is the stellar radius , and @xmath9 is the dimensionless overlap integral ; the remaining quantities and their dimensionless forms are defined in appendix a. the overlap integral serves as a measure of the intensity of the tidal forces and plays a crucial role in the physics of dynamical tides ( press and teukolsky 1977 ; zahn 1970 , 1975 , 1977 ) . here
, we will consider only the quadrupole part of the tidal forces @xmath10 , because it is this part that makes the greatest contribution to the overlap integral ( press and teukolsky 1977 ) .
the change in orbital semimajor axis and eccentricity with time is determined from the energy and angularmomentum conservation laws and is specified by the following formulas ( ivanov et al .
2013 ) : @xmath11 here , @xmath12 is the orbital eccentricity , @xmath13 is the semimajor axis , and the time scales of the change in semimajor axis @xmath14 and eccentricity @xmath15 in the case of low eccentricities ( @xmath16 ) are specified by the relations ( ivanov et al .
2013 ) @xmath17 where @xmath18 is the planetary mass .
the rate of change of energy @xmath19 , @xmath20 in the case of a dense spectrum and moderately large viscosities is specified by eqs .
( 51 ) from ivanov et al .
the spectrum is deemed to be dense enough if @xmath21 . for simplicity
, we will consider here only the time scales of the changes in semimajor axis .
using formulas from subsection 7.1 of the paper by ivanov et al .
( 2013 ) , we obtain @xmath22 where @xmath23 is the ratio of the planetary mass to the stellar mass , @xmath24 is the difference between adjacent frequencies , @xmath25 is the frequency number , and @xmath26 ( ivanov et al .
. the orbital period of the planet around the star is specified by the relation @xmath27 below , using eqs .
( [ ta ] ) and ( [ porb ] ) and solving eq .
( [ tate ] ) , we will present the change in the planet s orbital period as a function of time by taking into account the stellar evolution . as is clear from eq .
( [ overlapsintegral ] ) , the eigenfunctions and , accordingly , eigenfrequencies of the stellar oscillations should be known to calculate the overlap integral .
there exist three types of eigenfrequencies for a nonrotating star : the so - called p- , f- , and g - modes .
we will be concerned only with the low - frequency g- modes .
the properties of the g - modes are determined by the brunt - v@xmath28is@xmath29l@xmath29 frequency .
it is specified as follows ( christensen - dalsgaard 1998 ) : @xmath30 where @xmath31 is the pressure , @xmath32 is the density , @xmath33 is the gravitational acceleration , @xmath34 is the adiabatic index .
however , this definition of the brunt - v@xmath28is@xmath29l@xmath29 frequency is difficult to apply to many stars .
this is related to the numerical errors , to the calculation of the derivative of the density , and to the fact that there exist regions in stars where both terms in parentheses can be of the same order of magnitude .
therefore , a different definition is used ( brassard et al . 1991 ) : @xmath35,\end{aligned}\ ] ] where @xmath36 the temperature gradient @xmath37 @xmath38 is the mass fraction of atoms of type i , and @xmath39 it is convenient to divide the brunt - v@xmath28is@xmath29l@xmath29 frequency into two parts : the structural part ( structure terms ) brunt - v@xmath28is@xmath29l@xmath29 @xmath40\end{aligned}\ ] ] and the compositional part @xmath41 the compositional part is related to the change in stellar chemical composition due to nuclear reactions , and it is of great importance at the boundary of the convective and radiative regions .
another important stellar characteristic is the acoustic frequency defined by the formula @xmath42 the low - frequency g - modes are excited at @xmath43 .
the high - frequency p - modes are excited at @xmath44 .
we will not consider the p - modes , because they play a secondary role in the theory of dynamical tides .
thus , the problem is reduced to finding the eigenfunctions and eigenfrequencies of the stellar oscillations , calculating the overlap integrals , and calculating the characteristic orbital parameters as a function of time .
the equations that describe the eigenfrequencies and eigenfunctions of a star in the adiabatic and cowling approximations are presented in appendix a. the stars were modeled with the mesa software package ( paxton et al .
2011 , 2013 , 2015 ) ; the data obtained were then interpolated by the steffen ( 1990 ) method to two million points . using these data
, we solved the eigenfrequency and eigenfunction problem and calculated the overlap integrals .
the fourth - order runge
kutta method was used to solve the differential equations .
the data for the stars and the overlap integrals were compared with those from ivanov et al .
( 2013 ) and chernov et al .
in this section we present the results of our calculations of the change in the planet s orbital period for three types of stars with masses of one , one and a half , and two solar masses .
the course of stellar evolution for these models is shown on the hertzsprung russell diagram depicted in fig .
data for each star are presented in the tables in appendix b. the main phase of stellar evolution is the main - sequence phase .
it is at this phase that any star spends most of its lifetime .
this phase is indicated in fig . 1 by the nearly horizontal straight line . in this section we present the results of our calculations of the overlap integrals for various lifetimes of the star of one solar mass and the change in the planet s orbital period with time .
the star was modeled from @xmath45 to @xmath46 yr . at
times @xmath47 yr this star closely corresponds to our sun , and below we will call this star the sun . at the initial time the sun was completely convective , hydrogen burning has not yet begun , and its metallicity is z = 0.02 .
figure 2 shows a dimensionless dependence of the solar density on radius at the initial and subsequent times .
is@xmath29l@xmath29 frequency ( b ) for the sun versus radius for various times : the dashed , dotted , solid , and dash dotted lines are for the ages @xmath45 , @xmath48 , @xmath49 , and @xmath46 yr , respectively.,title="fig:",scaledwidth=45.0% ] is@xmath29l@xmath29 frequency ( b ) for the sun versus radius for various times : the dashed , dotted , solid , and dash dotted lines are for the ages @xmath45 , @xmath48 , @xmath49 , and @xmath46 yr , respectively.,title="fig:",scaledwidth=45.0% ] is@xmath29l@xmath29 frequency versus radius for the solar age @xmath49 : the solid , dashed , and dash - dotted lines indicate the total brunt - v@xmath28is@xmath29l@xmath29 frequency @xmath50 , the structural part @xmath51 , and the compositional part @xmath52 , respectively.,scaledwidth=45.0% ] the course of stellar evolution and the excitation of eigenmodes can be described as follows . at times
@xmath53 yr the sun was completely convective ( @xmath54 ) and , therefore , no low - frequency g - modes were excited , but only the high - frequency f- and p - modes were excited . subsequently ,
as the sun evolves ( at times @xmath55 yr ) a radiative region ( radiative core ) where the brunt - v@xmath28is@xmath29l@xmath29 frequency is greater than zero ( @xmath56 ) appears and , consequently , the excitation of low - frequency g - modes begins .
only the structural part makes a major contribution to the brunt - v@xmath28is@xmath29l@xmath29 frequency ; the compositional part is zero .
this is because at such lifetimes of the sun hydrogen burning makes a minor contribution to the energetics and evolution dynamics of the sun .
in the course of subsequent evolution the radiative region expands , the number of lowfrequency g - modes increases , and their spectrum becomes dense .
the low - frequency g - modes begin to play an increasingly important role . at times
@xmath57 yr hydrogen burning begins to make an increasingly large contribution to the total energy of the star and , consequently , the compositional part of the brunt - v@xmath28is@xmath29l@xmath29 frequency also begins to contribute to the total frequency .
in the course of subsequent evolution , as hydrogen burns , this contribution becomes progressively larger .
one of the problems being investigated here is to take into account the evolution of the brunt - v@xmath28is@xmath29l@xmath29 frequency and the influence of this evolution on the tidal interactions .
the radiative region expands approximately to a radius @xmath58 . for the present sun
the boundary between the radiative and convective regions is @xmath59 . in fig .
2 the brunt - v@xmath28is@xmath29l@xmath29 frequency is plotted against radius for various stellar evolution times .
it can be seen from the plot that the brunt - v@xmath28is@xmath29l@xmath29 frequency is subject to significant changes as the solar age increases ; both the frequency spectrum and its density change accordingly , which can directly affect the tidal forces . in fig .
3 the brunt - v@xmath28is@xmath29l@xmath29 frequency is plotted against radius for the present sun .
it follows from this plot that the brunt - v@xmath28is@xmath29l@xmath29 frequency exhibits a double - humped structure .
the first and second humps are associated with the compositional and structural parts of the brunt - v@xmath28is@xmath29l@xmath29 frequency , respectively . at a solar age
@xmath60 yr the two contributions become comparable . the overlap integrals ( 1 ) are plotted against frequency in the cowling approximation in fig .
4 . for a solar age @xmath53 yr ,
as has been said above , only the f- and p - modes exist ; therefore , there is no low - frequency part of the spectrum .
it can be seen from the plot that the absolute value of the overlap integrals decreases in the low - frequency part of the spectrum as the sun evolves .
this decrease is gradual , because the stellar structure does not change in the evolution time of the sun . on the main sequence
the sun always has a radiative core and a convective envelope .
in the high - frequency part of the spectrum , as the stellar age increases , the absolute value of the overlap integrals increases and a characteristic structure appears in the spectrumin the form of kinks . , @xmath48 , @xmath49 , and @xmath46 yr , respectively.,scaledwidth=45.0% ] figure 5 plots the change in the orbital period ( in days ) of a planet with a mass of one jupiter mass around the sun with time ( in years ) . the solar evolution models for which the overlap integrals were calculated are presented in table 1 . at lifetimes of the sun @xmath61
yr the spectrum at frequencies @xmath62 is insufficiently dense . at other lifetimes of the sun
the spectrum may be deemed sufficiently dense with a good accuracy for all frequencies .
therefore , eq . ( 4 ) may be used .
equation ( 2 ) was integrated over time from @xmath63 to @xmath64 yr .
the initial orbital periods of the planet around the sun were taken to be @xmath65 days ( solid curves ) .
it follows from fig .
5 that in the main - sequence lifetime of the sun a planet with a mass of one jupiter mass with the initial orbital period @xmath66 days will reduce its orbital period to @xmath67 days due to dynamical tides .
the time scale of the change in semimajor axis for this orbital period @xmath67 days and this solar age @xmath64 yr is @xmath68 yr .
consequently , it can be said with confidence that this planet will fall onto the sun within several tens of millions of years .
the same analysis showed that a planet with the initial orbital period @xmath69 days would reduce its orbital period to @xmath70 days and would fall onto the sun only within approximately half a billion years .
for the initial orbital period @xmath69 days fig . 5 compares the changes in orbital period with allowance for the solar evolution ( solid curve ) , without allowance for the solar evolution ( dash dotted curve ) , and the one calculated using eq .
( 25 ) from essick and weinberg ( 2016 ) ( dotted curve ) . to calculate the change in orbital period without allowance for the solar evolution
, we took the time scale at the solar age @xmath0 yr .
all curves qualitatively coincide , and the influence of solar evolution may be disregarded in the estimation .
this is because hydrogen burning in the sun is gradual , and no change in structure occurs .
quantitatively , the dotted curve from essick and weinberg ( 2016 ) gives smaller changes in the planet s orbital period .
this may be related to both the neglect of the change in the time scales of the orbital parameters as the sun evolves and to a different approach in this paper to the problem of the energy dissipation of tidally excited modes associated with their nonlinear interactions . in this subsection
we present the results of our calculations of the overlap integrals and the change in orbital period for the star with a mass of one and a half solar masses .
the star is considered from the time the protostar begins to collapse .
this time is quite arbitrary and is @xmath72 yr ; the initial radius is @xmath73 .
the initial metallicity of the star is @xmath74 . at lifetimes of the star up to @xmath75 yr it is completely convective ( @xmath54 ) . in such a star no lowfrequency g - modes
are excited , and only the highfrequency f- and p - modes are present ; therefore , the effective energy and angular momentum transfer from the orbit to the star will be suppressed .
as the stellar age increases further , hydrogen burning begins at the stellar center and a radiative core appears ( @xmath56 ) , which begins to expand , encompassing progressively newer layers .
the star has a well - defined convective envelope and a radiative core .
the radiative region expands up to @xmath76 for the age @xmath77 yr . a fairly dense spectrum of low - frequency g - modes appears very rapidly in such a star
. at times @xmath78 yr the stellar structure changes completely .
a convective core appears in the star , while its envelope becomes completely radiative .
the star abruptly contracts from a radius @xmath79 to @xmath80 .
the evolution time scales of the orbital parameters ( 2 ) increase sharply .
the star passes from a solar - type star that has a radiative core and a convective envelope to a star with a convective core and a radiative envelope .
is@xmath29l@xmath29 frequency ( b ) for the star of one and a half solar masses versus radius for various times : the dashed , dotted , solid , and dash dotted lines are for the ages @xmath81 , @xmath82 , @xmath83 , and @xmath84 yr , respectively.,title="fig:",scaledwidth=45.0% ] is@xmath29l@xmath29 frequency ( b ) for the star of one and a half solar masses versus radius for various times : the dashed , dotted , solid , and dash dotted lines are for the ages @xmath81 , @xmath82 , @xmath83 , and @xmath84 yr , respectively.,title="fig:",scaledwidth=45.0% ] in fig .
6 the stellar density and brunt - v@xmath28is@xmath29l@xmath29 frequency are plotted against radius for various times . from the figure
we see how the central density increased as the star contracted .
it can be seen from the figure for the brunt - v@xmath28is@xmath29l@xmath29 frequency that the star was initially completely convective ( dashed curve ) and subsequently a solar - type one ( the envelope was convective , and the core was radiative ; the dotted curve ) . at the subsequent times
the star has a more complex structure , the core becomes convective , and the envelope is completely radiative ( solid curve ) .
a thin convective envelope then again appears in its envelope ( dash dotted curve ) .
is@xmath29l@xmath29 frequency versus radius for the star of one and a half solar masses for the age @xmath84 yr : the solid , dashed , and dash - dotted lines indicate the total brunt - v@xmath28is@xmath29l@xmath29 frequency @xmath50 , the structural part @xmath51 , and the compositional part @xmath52 , respectively.,scaledwidth=45.0% ] figure 7 shows the brunt - v@xmath28is@xmath29l@xmath29 frequency as well as its structural and compositional parts . it follows from the figure that the compositional part makes the greatest contribution at the boundary between the convective and radiative regions . , @xmath82 , @xmath83 , and @xmath84 yr , respectively.,scaledwidth=48.0% ] the overlap integrals are plotted against frequency in fig .
. the dashed line does not extend to the lowfrequency part of the spectrum , because the star at this time is completely convective ( @xmath54 ) .
it can also be seen from the figure that the absolute value of the overlap integrals rapidly drops by more than two orders of magnitude as the star s lifetime increases .
such catastrophic changes are due to the change in stellar structure .
in the high - frequency part the absolute value of the overlap integrals increases by an order of magnitude , and characteristic kinks appear in the spectrum .
figure 9 plots the change in the orbital period ( in days ) of a planet with a mass of one jupiter mass around the star of one and a half solar masses with time ( in years ) .
the stellar evolutionmodels for which the overlap integrals were calculated are presented in table 2 .
equation ( 2 ) was integrated over time from @xmath85 to @xmath86 yr .
for such a stellar lifetime interval the g - mode spectrum is sufficiently dense .
the initial orbital periods of the planet around the sun are @xmath87 days ( solid curves ) .
it follows from the figure that all planets with a mass equal to one jupiter mass and with an initial orbital period of 2 days or shorter will fall onto the star in the main - sequence lifetime of the star .
a planet with the initial orbital period @xmath88 days will fall onto the star , while a planet with the initial orbital period @xmath89 days will reduce its orbital period to @xmath90 days in the main - sequence lifetime of the star .
the dashed curve indicates the change in orbital period without allowance for the stellar evolution .
the time scale of the orbital change was taken at the star s lifetime @xmath91 yr .
it can be seen from fig .
9 that the planet does not fall onto the star in the mainsequence lifetime of the star without allowance for its evolution .
the last star that we will consider is the star with a mass of two solar masses .
the time the protostar begins to collapse is @xmath92 yr , the initial stellar radius is @xmath93 , and the stellar metallicity is @xmath74 . at initial times up to @xmath94
yr the star is completely convective ( @xmath54 ) .
no low - frequency g - modes are excited ; only the high - frequency f- and p - modes are present .
the dynamical tides play no significant role in such a star .
however , as the star evolves further , at times @xmath94 yr a radiative region ( radiative core ) appears in the star , which begins to expand .
the low - frequency g - modes are well excited in this region ( @xmath56 ) , and the dynamical tides begin to play an increasingly significant role as the star evolves further .
such a star resembles solartype stars that have a radiative core and a convective envelope .
the radiative region expands quite rapidly , and this region almost reaches the stellar surface @xmath95 already at times @xmath96 yr . the star becomes almost completely radiative ; the brunt - v@xmath28is@xmath29l@xmath29 frequency in the entire region becomes positive ( @xmath56 ) in the entire region , except for the stellar surface .
a sufficiently dense g - mode spectrum is generated in such a star . in fig .
10 the stellar density and brunt - v@xmath28is@xmath29l@xmath29 frequency are plotted against radius for various times .
is@xmath29l@xmath29 frequency ( b ) for the star of two solar masses versus radius for various times : the dashed , dotted , solid , and dash dotted lines are for the ages @xmath97 , @xmath98 , @xmath99 , and @xmath100 , respectively.,title="fig:",scaledwidth=45.0% ] is@xmath29l@xmath29 frequency ( b ) for the star of two solar masses versus radius for various times : the dashed , dotted , solid , and dash dotted lines are for the ages @xmath97 , @xmath98 , @xmath99 , and @xmath100 , respectively.,title="fig:",scaledwidth=45.0% ] as the lifetime increases further ( @xmath101 yr ) , a convective core ( @xmath54 ) appears in the star , which begins to expand as the star evolves .
the compositional part of the brunt - v@xmath28is@xmath29l@xmath29 frequency @xmath52 begins to play an increasingly large role at the boundary between the convective and radiative regions ( fig .
. the stellar structure does not undergo significant changes as the lifetime increases further .
only the convective core evolves .
initially , the convective region expands , and then its expansion is replaced by contraction ( fig .
is@xmath29l@xmath29 frequency versus radius for the star of two solar masses for the age @xmath100 yr : the solid , dashed , and dash - dotted lines indicate the total brunt - v@xmath28is@xmath29l@xmath29 frequency @xmath50 , the structural part @xmath51 , and the compositional part @xmath52 , respectively.,scaledwidth=45.0% ] , @xmath98 , @xmath99 , and @xmath100 yr , respectively.,scaledwidth=45.0% ] the overlap integrals are plotted against frequency in fig .
. the dashed line does not extend to the low - frequency part of the spectrum , because the star at this time is completely convective ( @xmath54 ) . as the star s lifetime increases , the absolute value of the overlap integral rapidly drops by more than an order of magnitude .
such catastrophic changes are due to the change in stellar structure . when the stellar structure undergoes no significant changes , the overlap integrals change gradually and insignificantly ( see fig .
12 , the solid and dash - dotted lines ) .
figure 13 plots the change in the orbital period ( in days ) of a planet with a mass of one jupiter mass around the star of two solar masses with time ( in years ) .
the evolution models of the star of two solar masses for which the overlap integrals were calculated are presented in table 3 .
equation ( 2 ) was integrated over time from @xmath102 to @xmath103 yr .
for such a stellar lifetime interval the g - mode spectrum is sufficiently dense .
the initial orbital periods of the planet around the sun were taken to be @xmath104 days ( solid curves ) .
it follows from the figure that all planets with a mass equal to one jupiter mass and with an initial orbital period of 2 days or shorter will fall onto the star in the main - sequence lifetime of the star ( fig .
a planet with the initial orbital period @xmath89 days will reduce its orbital period only to @xmath105 days in the main - sequence lifetime of the star .
the dashed curve indicates the change in orbital period without allowance for the stellar evolution .
the time scale of the orbital change was taken at the star s lifetime @xmath106 yr . as can be seen from fig .
13 , the planet does not fall onto the star in the main - sequence lifetime of the star without allowance for its evolution .
q just as for the star of one and a half solar masses , the evolution time scales ( 3 ) of the planet decrease as the star begins to move off the main sequence .
this is because the hydrogen reserves at the stellar center begin to be depleted , and the convective core becomes a helium one .
this leads to an expansion of the star and , consequently , to an increase in the number of g - modes and the density of the spectrum , which leads to an increase in the pumping of orbital energy into the energy of stellar oscillations and to the fall of short - period planets onto the star .
we considered the change in the orbital period of a planet around a star with a mass of one , one and a half , and two solar masses .
the behavior of the overlap integrals for these models as a function of the main - sequence lifetime of the star was analyzed .
we showed that for the same star with a different structure during its evolution the overlap integrals could differ by two or more orders of magnitude .
we showed that all planets with a mass of one jupiter mass that revolve around the sun with an orbital period @xmath1 days or shorter should fall onto the stellar surface in the main - sequence lifetime of the star .
all planets with a mass of one jupiter mass that revolve around the stars with a mass of one and a half and two solar masses with an orbital period @xmath2 or shorter will fall onto the stellar surface in the main - sequence lifetime of the star .
such a fall may not occur without allowance for the evolution of the star itself .
these results qualitatively agree with the previous results from bolmont and mathis ( 2016 ) and penev ( 2012 ) .
calculations for the stellar models considered in this paper should be made for a more accurate quantitative estimate .
i am grateful to p.b . ivanov who read the paper for a number of valuable remarks and to j. papaloizou for a fruitful discussion .
this work was financially supported by the russian foundation for basic research ( project nos .
140200831 , 150208476 , 160201043 ) , grant no .
nsh-6595.2016.2 from the president of the russian federation for state support of leading scientific schools , and program 7 of the presidium of the russian academy of sciences .
the system of equations describing the adiabatic perturbations in stars is written in dimensionless form ( christensen - dalsgaard 1998 ) : @xmath107 where the following dimensionless variables are introduced : @xmath108 is the radius , @xmath109 is the radial perturbation , @xmath110 is the horizontal perturbation , @xmath111 and @xmath112 are the perturbations of the gravitational potential and its derivative , respectively .
the subscript @xmath113 refers to an unperturbed quantity , and the prime @xmath114 refers to a perturbed quantity .
a tilde over a quantity means that this quantity is dimensionless .
the normalization is done as follows : @xmath115 where @xmath116 is the sound speed squared , @xmath31 is the pressure , @xmath32 is the density , and @xmath50 is the brunt - v@xmath28is@xmath29l@xmath29 frequency squared .
four boundary conditions should be added to the four first - order differential equations .
two boundary conditions are specified at the stellar center , and the other two are specified on the stellar surface ( christensen - dalsgaard 1998 ) .
the boundary conditions at the stellar center @xmath117 are @xmath118 the boundary conditions on the stellar surface @xmath119 are @xmath120 there exists the cowling ( 1941 ) approximation , in which the perturbed gravitational potential is neglected , to investigate the low - frequency g - modes .
this approximation holds good and simplifies considerably the system of equations for the adiabatic perturbations ( 14 ) . in the cowling approximation
the system of equations ( 14 ) will be rewritten in a simpler form : @xmath121 in the cowling approximation the system of equations ( 18 ) ceases to depend explicitly on the stellar density .
thus , in the low - frequency limit the radial and horizontal displacements do not depend on the stellar density .
in contrast , the overlap integrals ( 1 ) directly depend on the density . in the cowling approximation not four but two
boundary conditions should be added to the system of equations ( 18 ) .
one boundary condition is specified at the stellar center , and the other one is specified on the stellar surface .
the boundary condition at the stellar center @xmath117 is @xmath122 the boundary condition on the stellar surface @xmath123 is @xmath124 to determine the eigenfrequency spectrum of the system of equations ( 14 ) or ( 18 ) , this system should be redefined .
this requires that the determinant composed of the eigenfunctions of the system of equations ( 14 ) or ( 18 ) be equal to zero .
this will hold only for some frequencies ( eigenfrequencies ) of the system of equations ( 14 ) or ( 18 ) , which determines the perturbation spectrum .
the tables present the models for which the overlap integrals ( 1 ) and the evolution time scales of the orbital parameters ( 2 ) were calculated .
the notation is as follows : @xmath125 is the time in years , @xmath126 is the radius measured in solar radii , @xmath127 is the luminosity measured in solar luminosities , @xmath128 is effective surface temperature measured in kelvins , h1 is the hydrogen mass fraction at the stellar center , and @xmath129 is the mean stellar density measured in @xmath130 . | we investigate the change in the orbital period of a binary system due to dynamical tides by taking into account the evolution of a main - sequence star .
three stars with masses of one , one and a half , and two solar masses are considered .
a star of one solar mass at lifetimes @xmath0 yr closely corresponds to our sun .
we show that a planet of one jupiter mass revolving around a star of one solar mass will fall onto the star in the main - sequence lifetime of the star due to dynamical tides if the initial orbital period of the planet is less than @xmath1 days .
planets of one jupiter mass with an orbital period@xmath2 days or shorter will fall onto a star of one and a half and two solar masses in the mainsequence lifetime of the star . |
the current version of the standard model ( sm ) consists of three generations of quarks and leptons .
recently we proposed @xcite a four generation lepton model based on the non - abelian discrete symmetry @xmath0 ( icosahedron ) , in which the best features of the three family @xmath2 ( tetrahedral ) model survive . besides the new heavy degrees of freedom in the @xmath3 model , which satisfy the experimental constraints , we retain tribimaximal neutrino mixings , three light neutrino masses , and three sm charged lepton masses in the three light generation sector . in this paper
, we will explore a generalization of our @xmath4 model to include four generations of both quarks and leptons .
but before launching into that discussion we must first discuss the viability of models with four generations given recent experimental developments .
a fourth generation is now being constrained @xcite by precision electroweak data @xcite , by flavor symmetries @xcite , and by the higgs - like particle at 125 gev recently reported at the lhc @xcite .
the new data provide an important step forward in distinguishing various four generation models , and in particular eliminating some from consideration .
in particular , four sequential generation models are now highly disfavored @xcite ; however , it would be premature to dismiss all four generation models .
while tension between four generation models and data has developed , a fourth generation is not excluded by the electroweak precision data @xcite , so the existence of a fourth generation is still a viable phenomenological possibility which can provide an explanation of the observed anomaly of cp asymmetries in the b meson system @xcite , and the baryon asymmetry of the universe @xcite , with additional mixings and cp phases . also , there are a number of way to relieve this tension . for example two higgs doublet models ( see e.g. , @xcite and references therein ) can accommodate a fourth generation of fermions and current data . for a comprehensive review
typically these two higgs doublet models are low energy effective field theories that require composite higgses similar to top quark condensate models @xcite .
for some recent examples see ref .
another possibility is to add electroweak doublets that are in color octets @xcite .
further discussion can be found in ref . @xcite .
while the model we will discuss has an extended higgs structure , a full exploration of the possible composite nature of the scalar sector is beyond the scope of our present study . to generalize our @xmath4 model to include four generations of quarks and leptons
, we first recall the three family scenarios in which the binary tetrahedral group @xmath5 is capable of providing a model of both the quarks and leptons with tribimaximal mixings and a calculable cabibbo angle @xcite . the @xmath6 group is the double covering group of @xmath7 .
it has four irreducible representations ( irreps ) with identical multiplication rules to those of @xmath7 , one triplet @xmath8 and three singlets @xmath9 , and @xmath10 , plus three additional doublet irreps @xmath11 , and @xmath12 .
the additional doublets allow the implementation of the @xmath13 structure to the quark sector @xcite , thus the third family of quarks are treated differently and are assigned to a singlet .
hence they can acquire heavy masses @xcite .
one should note that @xmath7 is not a subgroup of @xmath6 , therefore , the inclusion of quarks into the model is not strictly an extension of @xmath7 , but instead replaces it @xcite . based on the same philosophy , we study the model of four families of quarks and leptons by using the binary icosahedral group @xmath14 .
the relation between @xmath1 and @xmath3 is similar to that for @xmath6 and @xmath7 .
the icosahedral group @xmath15 has double - valued representations that are single - valued representations of the double icosahedral group @xmath16 .
hence , besides the irreps of @xmath1 that are coincident with those of @xmath3 , there are four additional spinor - like irreps @xmath17 , and @xmath18 of @xmath1 . we shall be able to assign quarks to the spinor - like representations , but to discuss model building using @xmath1 , we must first review our lepton model based on @xmath3 , which will remain essentially unchanged when generalized to @xmath1 .
some useful group theory details have been relegated to the appendix .
the irreps of @xmath3 are one singlet @xmath19 , two triplets @xmath8 and @xmath20 , one quartet @xmath21 , and one quintet @xmath22 .
the model is required to be invariant under the flavor symmetry of @xmath23 and the particle content is given by table [ a5 ] .
the most general form of the higgs potential containing the scalar fields @xmath24 , @xmath25 , @xmath26 and @xmath27 , invariant under the discrete @xmath23 symmetries is given by @xmath28 where the individual terms are written as @xmath29 , \\\end{aligned}\ ] ] @xmath30 , \\
v(h'_{4 } , \phi_{3 } ) & = & \lambda^{h'\phi}_{\beta}(h'^{\dag}_{4}h'_{4})_{\textbf{$\beta$}}(\phi^{\dag}_{3}\phi_{3})_{\textbf{$\beta$ } } \nonumber \\ & & + \lambda'^{h'\phi}_{\gamma}(h'^{\dag}_{4}\phi_{3})_{\textbf{$\gamma$}}(\phi^{\dag}_{3}h'_{4})_{\textbf{$\gamma$ } } \nonumber \\ & & + \left [ \lambda''^{h'\phi}_{\gamma}(h'^{\dag}_{4}\phi_{3})_{\textbf{$\gamma$}}(h'^{\dag}_{4}\phi_{3})_{\textbf{$\gamma$ } } + \rm{h.c . }
\right ] , \\
v(h_{4 } , h'_{4 } , \phi_{3 } ) & = & \lambda^{hh'\phi}_{\gamma}(h^{\dag}_{4}\phi_{3})_{\textbf{$\gamma$}}(h'^{\dag}_{4}h'_{4})_{\textbf{$\gamma$ } } \nonumber \\ & & + \lambda'^{hh'\phi}_{\gamma}(h'^{\dag}_{4}\phi_{3})_{\textbf{$\gamma$}}(h'^{\dag}_{4}h_{4})_{\textbf{$\gamma$ } } \nonumber \\ & & + \lambda''^{hh'\phi}_{\gamma}(h^{\dag}_{4}\phi_{3})_{\textbf{$\gamma$}}(h^{\dag}_{4}h_{4})_{\textbf{$\gamma$ } } + \rm{h.c .. }\end{aligned}\ ] ] here we have introduced the @xmath1 group representation indices @xmath31 ; @xmath32 ; and @xmath33 = @xmath34 respectively .
the first stage of ssb takes us from the initial @xmath1 discrete symmetry to @xmath6 , and this is accomplished with a vev for @xmath35 .
consider the pure @xmath35 sector of the higgs potential @xmath36 where @xmath37 and @xmath38 are @xmath1 group matrices .
since @xmath39 in @xmath1 contains three terms there are three singlets in @xmath40 ^ 2 $ ] and hence in @xmath41 .
there is also a potential cubic term that can be suppressed , either by imposing an addition @xmath42 symmetry , or by making @xmath35 a complex field . for @xmath43 and @xmath44 sufficiently small and of the proper signs
, the ssb is dominated by the @xmath45 term and we have @xmath46 such that @xmath47 with the sm gauge group unaffected and with the scalars decomposing as in eq . ( 6 ) .
the sector of the higgs potential of the @xmath1 model that depends only on @xmath48 , @xmath49 , @xmath50 , and @xmath51 is given in the appendix of @xcite which only involves @xmath6 invariant terms and is identical to our form of the potential up to constraints due to residual @xmath1 relations on the coupling constants .
the work of @xcite demonstrates how @xmath6 is broken completely and how the light quark and lepton masses and mixings arise .
our only other additional scalars are the @xmath1 triplet , sm doublet fields @xmath52 . since @xmath53 under @xmath47 , a @xmath52 breaks @xmath6 .
the pure @xmath52 scalar sector can be rewritten as @xmath54 where we have included the triplet indices @xmath55 and find two quartic terms due to the fact that @xmath56 in @xmath1 contains two terms and hence @xmath57 ^ 2 $ ] contains two singlets .
given the @xmath6 level vevs for @xmath50 , and @xmath51 , there is no @xmath6 symmetry remaining to rotate the @xmath52 vev .
hence the choice @xmath58 is stable for @xmath59 and of the form needed in the model .
a similar analysis can be applied to the investigation of the ssb in our @xmath4 model .
finally we note that typically , only an @xmath60 fine tuning of scalar quartic coupling constants is needed to maintain the stability of such patterns of ssb .
o. eberhardt , a. lenz and j. rohrwild , phys .
d * 82 * , 095006 ( 2010 ) [ arxiv:1005.3505 [ hep - ph ] ] ; o. eberhardt , g. herbert , h. lacker , a. lenz , a. menzel , u. nierste and m. wiebusch , phys .
d * 86 * , 013011 ( 2012 ) [ arxiv:1204.3872 [ hep - ph ] ] ; o. eberhardt , a. lenz , a. menzel , u. nierste and m. wiebusch , phys . rev .
d * 86 * , 074014 ( 2012 ) [ arxiv:1207.0438 [ hep - ph ] ] ; o. eberhardt , g. herbert , h. lacker , a. lenz , a. menzel , u. nierste and m. wiebusch , phys .
lett . *
109 * , 241802 ( 2012 ) [ arxiv:1209.1101 [ hep - ph ] ] . c. lazzeroni
_ et al . _
[ na62 collaboration ] , phys .
b * 698 * , 105 ( 2011 ) [ arxiv:1101.4805 [ hep - ex ] ] .
h. lacker and a. menzel , jhep * 1007 * , 006 ( 2010 ) [ arxiv:1003.4532 [ hep - ph ] ] .
g. aad _ et al . _ [ atlas collaboration ] , phys .
b * 716 * , 1 ( 2012 ) [ arxiv:1207.7214 [ hep - ex ] ] .
s. chatrchyan _ et al . _
[ cms collaboration ] , phys .
b * 716 * , 30 ( 2012 ) [ arxiv:1207.7235 [ hep - ex ] ] .
s. chatrchyan _ et al . _
[ cms collaboration ] , arxiv:1302.1764 [ hep - ex ] .
e. gates and j. terning , phys .
* 67 * , 1840 ( 1991 ) ; b. a. kniehl and h. g. kohrs , phys .
d * 48 * , 225 ( 1993 ) ; b. holdom , phys .
d * 54 * , 721 ( 1996 ) [ arxiv : hep - ph/9602248 ] ; h. j. he , n. polonsky and s. su , phys .
d * 64 * , 053004 ( 2001 ) [ arxiv : hep - ph/0102144 ] ; b. holdom , w. s. hou , t. hurth , m. l. mangano , s. sultansoy and g. unel , pmc phys . a * 3 * , 4 ( 2009 ) [ arxiv:0904.4698 [ hep - ph ] ] .
w. s. hou , m. nagashima and a. soddu , phys .
* 95 * , 141601 ( 2005 ) [ arxiv : hep - ph/0503072 ] ; w. s. hou , h. n. li , s. mishima and m. nagashima , phys .
* 98 * , 131801 ( 2007 ) [ arxiv : hep - ph/0611107 ] ; a. soni , a. k. alok , a. giri , r. mohanta and s. nandi , phys
. lett .
b * 683 * , 302 ( 2010 ) [ arxiv:0807.1971 [ hep - ph ] ] .
s. bar - shalom , m. geller , s. nandi and a. soni , arxiv:1208.3195 [ hep - ph ] .
w. a. bardeen , c. t. hill and m. lindner , phys .
d * 41 * , 1647 ( 1990 ) .
p. q. hung and c. xiong , nucl .
b * 847 * , 160 ( 2011 ) [ arxiv:0911.3890 [ hep - ph ] ] ; p. q. hung and c. xiong , phys .
b * 694 * , 430 ( 2011 ) [ arxiv:0911.3892 [ hep - ph ] ] ; p. q. hung and c. xiong , nucl .
b * 848 * , 288 ( 2011 ) [ arxiv:1012.4479 [ hep - ph ] ] ; c. m. ho , p. q. hung and t. w. kephart , jhep * 1206 * , 045 ( 2012 ) [ arxiv:1102.3997 [ hep - ph ] ] .
p. h. frampton , _
the third family is different _ , in _ proceeding of the fourth international symposium on particles , strings and cosmology _ , k.c .
, world scientific ( 1995 ) , [ arxiv : hep - ph/9409331 ] and _ treating top differently from charm and up _ , in _ particle theory and phenomenology _
lassila et al .
eds . , world scientific ( 1996 ) , [ arxiv : hep - ph/9507351 ] .
m. a. schmidt and a. y. smirnov , nucl .
b * 857 * , 1 ( 2012 ) [ arxiv:1110.0874 [ hep - ph ] ] .
h. ishimori , t. kobayashi , h. ohki , h. okada , y. shimizu and m. tanimoto , prog .
suppl . * 183 * , 1 ( 2010 ) [ arxiv:1003.3552 [ hep - th ] ] .
y. abe _ et al .
_ [ double - chooz collaboration ] , phys .
lett . * 108 * , 131801 ( 2012 ) [ arxiv:1112.6353 [ hep - ex ] ] .
f. p. an _ et al . _
[ daya - bay collaboration ] , phys .
* 108 * , 171803 ( 2012 ) [ arxiv:1203.1669 [ hep - ex ] ] .
j. k. ahn _
[ reno collaboration ] , phys .
lett . *
108 * , 191802 ( 2012 ) [ arxiv:1204.0626 [ hep - ex ] ] .
g. l. fogli , e. lisi , a. marrone , a. palazzo and a. m. rotunno , phys .
d * 84 * , 053007 ( 2011 ) [ arxiv:1106.6028 [ hep - ph ] ] ; t. schwetz , m. tortola and j. w. f. valle , new j. phys . * 13 * , 109401 ( 2011 ) [ arxiv:1108.1376 [ hep - ph ] ] .
l. l. everett and a. j. stuart , phys .
d * 79 * , 085005 ( 2009 ) [ arxiv:0812.1057 [ hep - ph ] ] ; f. feruglio and a. paris , jhep * 1103 * , 101 ( 2011 ) [ arxiv:1101.0393 [ hep - ph ] ] ; g. -j .
ding , l. l. everett and a. j. stuart , nucl .
b * 857 * , 219 ( 2012 ) [ arxiv:1110.1688 [ hep - ph ] ] ; l. l. everett and a. j. stuart , phys .
* b698 * , 131 - 139 ( 2011 ) [ arxiv:1011.4928 [ hep - ph ] ] . | to include the quark sector , the @xmath0 ( icosahedron ) four generation lepton model is extended to a binary icosahedral symmetry @xmath1 flavor model .
we find the masses of fermions , including the heavy sectors , can be accommodated . at
leading order the ckm matrix is the identity and the pmns matrix , resulting from same set of vacua , corresponds to tribimaximal mixings . |
it is well - known that environmental processes play a significant role in shaping the evolution of galaxies as they assemble onto clusters . with the aid of hubble space telescope ( _ hst _ )
imaging and deep optical spectroscopy , recent studies have quantified this evolution in galaxy properties , painting a picture where the fraction of early - type ( elliptical and s0 ) galaxies and the fraction of passive non - star - forming galaxies both grow with time , and at a rate that seems to depend sensitively on the local density of galaxies @xcite .
yet there are a wide variety of physical processes that may be responsible for these evolutionary trends including galaxy mergers , galaxy - galaxy harassment , gas stripping by the icm , or tidal processes @xcite .
observationally , it has so far been impossible to fully separate the effects of the various physical processes , in large part due to the overlapping regimes of influence for each of the proposed mechanisms ( see * ? ? ?
further complicating the picture , the large scale assembly states of clusters show considerable variety @xcite , such that the dominant forces acting on galaxies are likely to vary from cluster to cluster , or over the course of an individual cluster s assembly history . but
gaining an understanding of the complex interplay between a variable icm , the properties of assembling galaxies , and the overall cluster dynamical state is crucial if we are to have a complete picture of the growth and evolution of galaxies in a hierarchical universe . in this paper
, we combine optical ( _ hst _ ) and uv ( _ galex _ ) imaging of two @xmath4 galaxy clusters with ground - based spectroscopy of member galaxies , in an attempt to trace directly the buildup of passive early - type galaxies via a detailed ` case study ' of the galaxy population across each cluster .
the two studied clusters , cl 0024 + 17 ( @xmath5 ) and ms 0451 ( @xmath3 ) , are part of a long - term campaign to trace the evolution of galaxies in wide fields ( @xmath6 mpc diameter ) centered on both clusters , using a variety of methods . by undertaking an in - depth , wide - field comparative study of two prominent clusters ,
we hope to provide a complement to other observational ( e.g. , * ? ? ?
* ; * ? ? ?
* ) and theoretical investigations ( e.g. , * ? ? ?
* ) which trace with a broad brush the evolution in star formation rate and the buildup of structure in the universe . the first paper in our series , @xcite ( hereafter paper i ) , introduced our panoramic _
hst _ imaging of cl 0024 and began our ongoing discussion of the physical processes that may be acting on galaxies within clusters .
in several subsequent papers , whose results are summarized in 2 , we have added extensive optical spectroscopy to the program , allowing targeted investigations of galaxy stellar populations and star formation rates as a function of cluster - centric radius , local density , and morphology .
our goal for this paper is to bring our complete survey data set to bear on the question of how galaxies are affected by their environment , as a function of both the overall cluster properties and of local environment within each cluster . for maximum clarity and deductive power
, we focus our investigation on several key populations of ` transition galaxies ' in the clusters galaxies whose stellar populations or dynamical states indicate a recent or ongoing change in morphology or star formation rate .
lcclcccc cl 0024 & 6.65125 & 17.162778 & @xmath7 & @xmath8 & 0.395 & @xmath9 & 3.5@xmath10 + ms 0451 & 73.545417 & -3.018611 & 2.6 & @xmath11 & 0.540 & @xmath12 & 10.0@xmath13 + in evaluating cluster galaxies for signs of evolution , we have adopted a strategy to make maximal use of our _ hst_-based morphologies by characterizing signs of recent evolution in spirals and early - types separately .
this approach is similar to using the color - magnitude relation to divide our sample into ` red sequence ' and ` blue cloud ' galaxies , but it provides additional leverage to identify galaxies in transition .
early - type galaxies that have either been newly transformed or prodded back into an active phase , or spiral galaxies where star formation is being suppressed or enhanced will all stand out in our sample . at the same time ,
their morphologies reveal important information about their formation histories prior to their current transition state , information that colors alone do not provide .
our strategy also has the benefit of allowing us to directly investigate the hypothesis that many cluster spirals transform into s0s between @xmath4 and today @xcite an investigation that will form the basis of this paper . in the next section
, we outline our rationale for selecting cl 0024 and ms 0451 , describe the large - scale properties of each cluster , and give a summary of what we have concluded so far in our study of galaxy evolution in both clusters . in 3 , we describe new data not covered in previous papers in our series . in 4 , we will investigate the properties of ` passive spirals ' across the two clusters , suggesting that they are in the process of transforming into s0 galaxies .
we confirm in 5 that this is the case , via identification of newly created s0s that we believe reflect the distinct passive spiral populations found in each cluster . in 6 , we consider the environments of these galaxies in transition , and begin to investigate the physical mechanisms that may be responsible for these transformations . in 7
, we outline a model of how galaxy evolution proceeds in each cluster .
we consider the fundamental plane as a way to further constrain the physical mechanisms at work , and derive similar constraints from the populations of compact emission line galaxies in both clusters .
finally , in 8 , we summarize our conclusions about the transformation of spirals into s0s at @xmath4 . in this paper
, we adopt a standard @xmath14cdm cosmology with @xmath15 km s@xmath16 mpc@xmath16 , @xmath17 , and @xmath18 .
cl 0024 and ms 0451 were chosen for study primarily because of their comparable total masses , similar galaxy richness , and strong lensing features , while exhibiting x - ray properties that are quite distinct ( see table [ tab : basic ] ) . while ms
0451 is one of the most x - ray luminous clusters known @xcite , cl 0024 is somewhat under - luminous in the x - ray , with a mass inferred from _
xmm / newton _ observations that significantly underestimates the mass derived from other methods @xcite .
ms 0451 has x - ray luminosity seven times larger than cl 0024 , with a corresponding gas temperature nearly three times as high .
this implies a large difference in the density and radial extent of the intracluster medium ( icm ) between the two clusters . as a result
, icm - related physical processes are naively expected to be more important in the evolution of currently infalling ms 0451 galaxies than in cl 0024 .
schematic diagram indicating the cluster - centric radius over which each of several listed physical mechanisms may be effective at fully halting star formation or transforming the visual morphology of a radially infalling galaxy .
each physical mechanism listed can act effectively over the radial range indicated by the solid ( cl 0024 ) or dashed line ( ms 0451 ) .
arrows indicate the virial radius for each cluster .
note that tidal processes here refer to interactions with the cluster potential , while tidal forces during galaxy galaxy interactions are a component of the harassment mechanism .
each regime of influence was calculated according to the simple models described in paper i for cl 0024 , using the global cluster properties given in table [ tab : basic ] . ] in the schematic diagram shown in figure [ mechanisms ] , we apply simple scaling relations to estimate the regimes of influence for several key physical processes which could be acting on infalling galaxies , following the procedure described in paper i. icm - related processes , such as gas starvation @xcite and ram - pressure stripping @xcite begin to affect galaxies at much larger radius in ms 0451 . as in paper
i , we expect that other icm - related processes such as thermal evaporation @xcite or viscous stripping @xcite operate with at most similar strength to ram pressure , and so we do not consider them separately . an important caveat , however , is that the role of difficult - to - observe shocks in the icm are unknown , and are not accounted for in figure [ mechanisms ] ( but see 7 ) .
similarly , the two clusters differing masses set the radial regions where galaxy merging will be effective ; because of the @xmath19 higher mass of ms 0451 , typical galaxy relative velocities become too fast for mergers to occur at a higher radius than in cl 0024 .
the differing regimes of influence for the physical mechanisms illustrated in figure [ mechanisms ] provide the key template for our attempt to disentangle the relative importance of the various processes .
by surveying the galaxies of both clusters for signs of recent transformation or disturbance , across the entire radial range to @xmath20 mpc , we hope to associate the sites and characteristics of galaxies in transition with the likely causes from figure [ mechanisms ] .
an additional factor not reflected in figure [ mechanisms ] , however , is the overall assembly state and level of substructure in each cluster .
therefore , figure [ mechanisms ] can only be used as a guide , and we must carefully consider the effects that large - scale cluster assembly and irregularities may have on their galaxies as well .
for example , the effects of tidal processes and galaxy - galaxy harassment , which according to figure [ mechanisms ] should occur in much the same regions across both clusters , may very well differ greatly between clusters depending on how well each is virialized . as we will discuss below ,
there are marked differences in the levels of substructure between the two clusters , and these may drive important differences between the galaxy populations .
distribution of redshifts for both cl 0024 ( top ) and ms0451 ( bottom ) , including all galaxies with spectroscopically confirmed redshifts from our observations as well as previously published redshifts ( see 3 ) . in cl 0024
, we identify 508 cluster members , defined to lie in the range @xmath21 . in ms 0451
, we count 319 cluster members in the range @xmath22 . ]
while it is evident from their different x - ray luminosities that ms 0451 and cl 0024 provide quite different environments for their constituent galaxies , our detailed study of environmental effects on cluster galaxies requires a comprehensive characterization of the two clusters and their respective environments .
here , we study the radial velocities and spatial distributions of galaxies in each cluster , in order to evaluate the global kinematic structure of each cluster and identify significant substructures . our extensive
_ hst _ imaging and keck spectroscopy , which will be described more fully in 3 , readily reveal marked differences in the distributions of galaxies between the two clusters , suggesting that the clusters have quite dissimilar recent assembly histories . in figure
[ redshifts ] , we display the distribution of redshifts for members of both cl 0024 and ms 0451 .
the redshift distribution of ms 0451 members is broadly consistent with a gaussian distribution .
a somewhat better fit to the data is given by a two - component double gaussian function , but the distribution splits into these two peaks only for galaxies at large radius ( @xmath23 mpc ) , suggesting that we are observing two filaments feeding galaxies into the virialized center of ms 0451 .
cl 0024 galaxies likewise exhibit a double - peaked structure in redshift space , with two components that are widely separated and asymmetric in height .
this feature was discovered by @xcite , and is thought to be the remnant of a high - speed , face - on encounter between the main cluster and a smaller subcluster or large group @xcite .
the effects of this collision , even an estimated several gyr after it occurred @xcite , are still important in the core of cl 0024for example , as shown by intriguing recent claims of a ` dark matter ring ' in the core of cl 0024 @xcite .
the measured line of sight velocity dispersion in the core of @xmath24 km s@xmath16 implies a 3050@xmath25 lower mass than has been directly measured through strong and weak lensing constraints @xcite . the fact that the galaxy velocity distribution and x - ray emitting gas both underestimate the mass leads to the conclusion that the core of cl 0024 is not in virial nor hydrostatic equilibrium , an assumption that was made for each of these mass estimates .
the lower relative velocities of galaxies in the cl 0024 core may importantly affect the action of physical processes whose strengths vary with galaxy velocity . for example , gas stripping by the icm may be even less effective in cl 0024 than naively predicted by figure 1 .
the effects of galaxy galaxy harassment may also be different , as close galaxy encounters may happen at both lower frequency and lower relative velocities than have previously been modeled in detail ( e.g. , * ? ? ?
the spatial distribution of galaxies in ms 0451 and cl 0024 also show key differences . in figure
[ dstests ] , we construct modified dressler - shectman ( d - s ) plots for the region within @xmath26 of each cluster @xcite . in such plots ,
each cluster member is indicated by a circle , with the size of the circle proportional to that galaxy s ` dressler - shectman statistic ' : @xmath27\ ] ] where line - of - sight velocity @xmath28 and dispersion @xmath29 are measured with respect to each galaxy s ten nearest neighbors , and @xmath30 and @xmath31 are the global cluster values .
this statistic measures each galaxy s local deviation from a smooth , virialized velocity and spatial distribution .
in other words , groups of large circles on the plot tend to indicate the presence of an infalling group . in figure
[ dstests ] , we further color - code each galaxy according to its velocity with respect to the cluster center .
the d - s plot for cl 0024 reveals the presence of at least two significant groups near to the cluster core : an infalling group at high velocity nearly along the line of sight to the cluster center , and a previously noted large group to the northwest of the cluster core , presumably assembling onto the core in an orientation almost in the plane of the sky . the large structure to the nw
is also detected in both the weak lensing map ( * ? ? ?
* hereafter , paper ii ) and _ xmm / newton _ observations @xcite , which additionally suggest the presence of a shock front at the interface between the group and the main cluster .
in contrast , the d - s plot for ms 0451 reveals an elongated but largely smooth distribution of galaxies .
the cluster s elliptical shape is clearly visible , but the segregation of red points to the southeast and blue points to the northwest suggests a uniform contraction of the cluster along its major axis . while it is possible that the observed velocity field is due to cluster - scale rotation , previous analyses of the x - ray and sunyaev - zeldovich observations prefer a prolate or triaxial shape for ms 0451 @xcite , whereas the cluster would necessarily be oblate if the observed velocities are due to rotation .
we may be seeing this spread in velocities reflected in the redshift distribution of ms 0451 , which splits into two peaks at large radius , likely indicating infall from two directions .
one might worry that the elongated distribution gives ms 0451 an artificially large velocity dispersion , @xmath31 , but cluster mass estimates derived from @xmath31 are consistent with those derived from the x - ray and weak lensing analysis ( * ? ? ?
* and references therein ) .
the origin of the elliptical shape is unclear , but may represent the remnant of a past major merger .
nevertheless , still - bound infalling groups appear to be absent or insignificant within the virial radius .
the overall increased level of substructure observed in cl 0024 is likely to have implications for the galaxy population . in comparison to ms 0451 , cl 0024 galaxies are more likely to have been a member of an infalling group in the recent past , and so we may expect to see more signs of recent group ` pre - processing ' in cl 0024 . furthermore , in the chaotic environment of cl 0024 the icm is more likely to be disturbed , and so shocks , cold fronts , or other features in the icm may be present , and could play an important role in the star formation histories of cluster members @xcite . between the extremely dense icm in ms 0451 and the active assembly state of cl 0024 , these two clusters provide very distinct environments for their member galaxies . in the course of this paper , we will examine how the properties of transition galaxies in each cluster reflect these distinct environments . however , as our new results rely on and incorporate the findings of our previous papers in this series , we present here a brief summary of our investigations so far .
we highlight in particular several results which , in the initial interpretation , hint at the action of one or more physical mechanisms from figure [ mechanisms ] .
@xmath32 by constructing the fundamental plane ( fp ) of cl 0024 , we observed in @xcite ( hereafter paper iii ) that elliptical and s0 galaxies ( e+s0s ) exhibit a high scatter in their fp residuals , equivalent to a spread of @xmath33 in mass to light ratio ( @xmath34 ) .
the high scatter occurs only among galaxies in the cluster core , suggesting a turbulent assembly history for cluster early types , perhaps related to the recent cluster
subcluster merger @xcite . around the virial radius of cl 0024
, we observed a number of compact , intermediate
mass ellipticals undergoing a burst of star formation or weak agn activity , indicated by strong [ o@xmath35ii ] emission ( paper iii ) .
the [ o@xmath35ii ] emitters reside in relatively low density and high speed regions , and so we deemed that they are not likely the remnants of mergers .
though we will revisit the merger hypothesis in 7 , we tentatively concluded that the observed activity is caused by a rapidly acting physical process : two candidates are galaxy harassment and shocks in the icm , perhaps generated by the sub - cluster merger in cl 0024 .
we searched for disruptions in the internal structures and star formation rates of disk galaxies due to tidal effects or galaxy galaxy interactions , by measuring emission line rotation curves and constructing the tully - fisher relation @xcite .
we find that the cluster tf relation exhibits significantly higher scatter than the field relation , echoing the high scatter seen in the cl 0024 fp .
we argue that the high scatter in both @xmath36- and @xmath37-band tf relations demonstrate that cluster spirals are kinematically disturbed by their environment .
we proposed that such disturbances may be due to galaxy merging and/or harassment .
we combined _ galex _ uv observations with key spectral line indices to place strict constraints on the recent star formation histories of `` passive spiral '' galaxies , an important class of transition object @xcite .
passive spirals show spiral morphology in _ hst _ images , but reveal weak or no [ o@xmath35ii ] emission in their spectra , suggesting a lack of current star formation . through _ galex _ uv imaging , we find that passive spirals in cl 0024 exhibit uv emission nearly as strong as regular star - forming spirals , implying the presence of young stars .
their unusual combination of uv emission with weak h@xmath38 strength supports a picture where passive spirals have experienced a rapid decline and eventual cessation of star formation over the last @xmath39 gyr .
the timescale of this decline suggests `` starvation '' by the icm as a possible cause @xcite a process where the intracluster medium ( icm ) strips gas from a galaxy s halo , causing star formation to decline due to the absence of new cold gas accretion onto the disk . each of these previous investigations has revealed a partial view of environmental evolution across the studied clusters . through these investigations
, we have identified several physical mechanisms that we could call ` likely suspects ' for driving galaxy evolution in clusters , and these seem to fall into two classes : galaxy icm interactions and galaxy galaxy interactions . however
, a unified picture is still lacking .
both of these flavors of interaction have been implicated before in the decline of star formation and the possible conversion of spirals into s0s ( see * ? ? ?
* and references therein ) , yet a detailed evaluation of their importance remains elusive .
we aim in this paper to complete our accounting of galaxy evolution across these two quite distinct clusters , building on and linking our earlier work into what we hope will be a more comprehensive picture of how cluster galaxies are affected by their environment at intermediate redshift . to accomplish this , we will first document what we believe to be direct evidence for the transformation of spirals into s0s : through an analysis of their stellar populations and recent star formation rates ( sfrs ) , we link the passive spiral galaxies in both clusters to their eventual end states as newly - generated cluster s0 galaxies . only then , as we examine the physical mechanisms responsible for this transformation , will we draw on the above summarized results to place extra constraints on the physics driving the transformation
. this discussion will include an extension of our analysis of the fundamental plane and the strongly - emitting compact e+s0s in cl 0024 to include ms 0451 , in order to further strengthen the constraints we can place on galaxy galaxy interactions and galaxy icm interactions , respectively .
before turning to our new analysis , we must specify the characteristics of our now - complete data set . while most imaging and spectroscopic data on cl 0024 has been fully or partially described in previous papers (
* paper i , paper iii ) , much of the following relies on new data for ms 0451 . in this section , we describe our new observations of ms 0451 , and provide updated information on cl 0024 , including _ galex _ imaging and additional spectroscopy .
we make use of _ hst _ imaging of cl 0024 and ms 0451 from the comprehensive wide - field survey described in paper i and smith et al .
( 2007 , in preparation ) . in cl 0024 ,
_ hst _ coverage consists of a sparsely - sampled mosaic of 39 wfpc2 images taken in the f814w filter ( @xmath40 band ) , providing coverage to a projected radius @xmath41 mpc at exposure times of 4 - 4.4ks .
ms 0451 observations were taken with the acs , also in f814w , and provide contiguous coverage within a 10mpc@xmath4210mpc box centered on the cluster , with single orbit ( 2ks ) depth over the field .
for both clusters , reliable morphological classification is possible to rest frame absolute magnitude @xmath43 , corresponding to @xmath44 in ms 0451 and @xmath45 in cl 0024 .
broader classification as early or late type is possible to a fainter limit , @xmath46 .
all galaxies that have spectroscopically confirmed redshifts and are brighter than this limit are classified visually following the procedure described in paper i for cl 0024 . in ms 0451 ,
galaxies were classed by one of us ( rse ) , and we expect that the typing is accurate to the quoted limits based on previous experience with acs imaging of similar depth ( e.g. , * ? ? ?
morphologies were assigned according to the medium deep survey scheme introduced by @xcite : t=-2=star , -1=compact , 0=e , 1=e / s0 , 2=s0 , 3=sa+b , 4=s , 5=sc+d , 6=irr , 7=unclass , 8=merger , 9=fault . in the following ,
all galaxies assigned types t=0,1,2 are together labeled as ` early types ' or e+s0s , and all galaxies with t=3,4,5 are labeled as spirals .
cl 0024 and ms 0451 were respectively observed for 15ks and 80ks with
_ galex _ @xcite in 2004 october ( go-22 ; cycle 1 ; pi treu ) , reaching comparable depths in rest frame fuv ( observed nuv ) .
galaxy fluxes were measured within @xmath47 circular apertures , centered on the optical position , on images reduced and sky - subtracted using the standard automated _ galex _ reduction pipeline .
the aperture size is chosen to be comparable to the measured nuv fwhm ( @xmath48 ) to avoid difficulty with source confusion , which can be a significant issue for _ galex _ imaging .
this problem limits our ability to measure total fluxes in larger ( @xmath49 ) apertures centered on the optical position , as contamination from nearby sources becomes much higher . as a result
, we apply an aperture correction to bring our aperture fluxes into agreement with sextractor - derived total magnitudes ( mag_auto ) , and for comparison to mag_auto magnitudes in f814w @xcite .
we note that at the redshifts of the clusters , most galaxies are effectively point sources at _ galex _ resolution , so we do not expect that the aperture corrections introduce significant errors into the photometry . while observed f814w provides a good match to rest - frame @xmath37 for both cl 0024 and ms 0451 , we also make use of supplemental ground - based imaging to aid in determining k corrections to transform observed magnitudes to the nearest rest frame bands .
ground based data include panoramic @xmath50-band imaging of both clusters , along with @xmath51-band imaging of cl 0024 , obtained with the wirc camera on the hale 200 telescope at palomar observatory @xcite .
these near - infrared data are supplemented with optical imaging in _ bvri _ bands from either the 3.6-m canada - france - hawaii telescope ( cl 0024 ) or the subaru 8-m telescope ( ms 0451 ) , which have been described in @xcite .
we use the _ kcorrect _ software v.4_1_2 @xcite to estimate the necessary k corrections ; we make use of imaging in all available bands from nuv to @xmath50 , modeling each galaxy s spectral energy distribution and deriving the best - fit correction on a galaxy by galaxy basis . however , we find that the derived k correction is mostly insensitive to the omission of one or several bands , with scatter typically @xmath52 magnitudes .
we assume a galactic extinction of e(_b
v_)=0.056 for cl 0024 and 0.033 for ms 0451 @xcite , and all absolute magnitudes will be expressed on the ab system .
we note also that , in all cases where we quote rest - frame colors , we have applied a k correction equal to the _ median _ k correction for all galaxies of the same morphological type and in the same cluster , rather than the individually fit k - corrections .
we do this to avoid introducing any additional scatter into the rest - frame colors ( due to uncertainties in the k corrections ) , above that present in the observed - frame colors . in general ,
the difference between the two methods is @xmath53 mag , and we do not expect the method of k correction to impact our analysis significantly .
cccc 73.321434 & -3.022260 & 0.667 & 1 + 73.328568 & -3.033381 & 0.370 & 1 + 73.325325 & -3.024078 & 0.300 & 1 + 73.334717 & -3.043395 & 0.128 & 1 + 73.331795 & -3.002323 & 0.539 & 1 + 73.344551 & -3.028507 & 0.725 & 1 + 73.346550 & -3.024269 & 0.371 & 1 + 73.339874 & -3.009066 & 1.160 & 1 + 73.337769 & -3.004604 & 0.388 & 1 + 73.343781 & -3.007794 & 0.995 & 1 + ... + we have used the deimos spectrograph on keck ii to obtain deep spectra of @xmath54 galaxies / cluster in the fields of cl 0024 and ms 0451 . in both clusters , we observe with @xmath55 wide slits with lengths much larger than the galaxy size ( typically @xmath56 ) . for each cluster
, spectral setups were chosen to span rest frame wavelengths from @xmath57 to @xmath58 , covering optical emission lines [ o@xmath35ii ] , [ o@xmath35iii ] , h@xmath59 , and , more rarely , h@xmath60 .
observations of cl 0024 from 2001 october to 2003 october are described in paper i and paper iii .
integrations totaled 2.5hr per mask and targets were selected from the cfht @xmath61band mosaic , with priority given to known cluster members with _
morphologies , followed by galaxies in the _ hst _ survey without a known redshift , to @xmath62 .
we undertook an initial redshift survey of ms 0451 during 2003 october , observing 14 slitmasks for one hour each , with the 600 l mm @xmath16 grating set to central wavelength 7500@xmath63 . this allowed redshift identification for 1300 objects , including 250 cluster members .
targets were selected from the i - band subaru image , as the acs mosaic was not yet available .
we first selected objects randomly from the set with @xmath64 , and then filled in slit mask gaps with fainter objects ( @xmath65 ) .
we have additionally observed both clusters again in 2004 december and 2005 october . for the 2004 december run , spectroscopic targets in cl 0024
were selected in the same manner as previous runs .
integrations were 2.5hrs long , but used the 600 l mm@xmath16 grating set at @xmath66 central wavelength . in 2004 ,
most targeted objects in ms 0451 were selected for deeper followup after already being identified as cluster members in our previous year s redshift survey .
however we excluded bright cluster members that had already yielded spectra of sufficient signal to noise in the 1-hr integrations of 2003 .
we observed 3 masks for 4 hrs each , with gaps between high - priority targets filled as before with random objects to @xmath65 .
observations used a 600 l mm@xmath16 grating set to a central wavelength of 6800@xmath63 . in 2005 october ,
objects in both clusters were observed with the same spectral set up and integration times as in 2004 , with six masks observed per cluster .
however , in 2005 our goal was to follow up on sources detected in both _
galex _ imaging and _ spitzer space telescope _ mips imaging @xcite , and so we filled masks with these sources preferentially . while targets in these latest observations were not selected randomly to a fixed magnitude limit , the total spectroscopic sample is still representative of the cluster as a whole , with the bulk of cluster members identified prior to 2005 .
ccccccccc 6.443974 & 17.141319 & 0.3792 & e / s0 & 21.9 & -18.8 & @xmath67 & @xmath68 & @xmath69 + 6.462808 & 17.145109 & 0.3978 & s0 & 19.1 & -21.7 & @xmath70 & @xmath71 & @xmath72 + 6.477189 & 17.274000 & 0.3967 & s0 & 20.0 & -20.7 & @xmath73 & @xmath74 & @xmath75 + 6.481423 & 17.247549 & 0.3969 & sc+d & 20.7 & -20.1 & @xmath76 & @xmath77 & @xmath78 + 6.499555 & 17.336269 & 0.3951 & s0 & 19.9 & -20.8 & @xmath79 & @xmath72 & @xmath80 + 6.508250 & 17.056641 & 0.3992 & sa+b & 20.5 & -20.2 & @xmath81 & @xmath82 & @xmath83 + 6.510695 & 17.325680 & 0.3931 & s0 & 19.2 & -21.5 & @xmath84 & @xmath85 & @xmath86 + 6.511346 & 17.324520 & 0.3913 & s0 & 19.7 & -21.0 & @xmath87 & @xmath88 & @xmath89 + 6.514315 & 17.317190 & 0.3946 & e & 18.8 & -21.9 & @xmath90 & @xmath91 & @xmath72 + 6.517361 & 17.105089 & 0.4059 & sa+b & 18.8 & -21.9 & @xmath92 & @xmath93 & @xmath94 + ... + deimos data were reduced using the deep2 deimos data reduction pipeline @xcite , which produce sky - subtracted , wavelength - calibrated one- and two - dimensional spectra .
redshifts for all galaxies were determined from the one - dimensional spectra and verified by eye .
we further correct spectra for the instrumental response curves before measuring line strengths , but we do not perform flux calibration . in total , we have obtained spectra of over 300 member galaxies per cluster , to @xmath46 , boosting the total known cluster members to 504 in cl 0024 and 319 in ms 0451 .
the spectroscopic sample is @xmath95 complete for objects with f814w@xmath96 in cl 0024 ( @xmath97 at the cluster redshift ) . in ms 0451 , completeness to the same absolute magnitude limit
is lower , @xmath98 , because of the deeper observations required ( to f814w@xmath99 ) . in both clusters ,
the spectroscopic sample remains representative of the cluster population as a whole to @xmath97 , roughly 1.5 magnitudes below @xmath100 .
however , in the range @xmath101 , we are biased toward detection of emission line cluster galaxies over absorption line galaxies , and completeness is lower in ms 0451 than cl 0024 . in the following , we will restrict our analysis to cluster members with @xmath102 , except where specified otherwise .
our full redshift catalog for cl 0024 was published in paper iii , excluding the small number of additional redshifts obtained in 2005 . in table
[ 0451cat ] , we publish a sample of the equivalent full catalog for ms 0451 , containing redshifts for 1562 objects .
the full catalog is available in the online edition . as cl 0024 and ms 0451
have now been observed in more detail than perhaps any others at these redshifts , we aim to make as much of our data as practical available to other investigators through our web site .
we will therefore publish updated versions of both the cl 0024 and ms 0451 catalogs to our web sitesmm / clusters/ ] , including positions matched with photometric measurements , _ hst _ morphologies , and redshifts .
we measure spectral line indices for several key emission and absorption lines following the lick system @xcite .
we focus in this paper on indicators of recent or ongoing star formation , including [ o@xmath35ii ] 3727 and the balmer lines h@xmath103 and h@xmath59 , which have been described in paper iii .
we also measure the @xmath104 index , which indicates the strength of the 4000@xmath63 break ; we adopt the definition from @xcite . in order to make maximal use of our spectra for the investigation of stellar populations and star formation rates in cluster transition galaxies
, we took great care to measure the equivalent widths ( ews ) of key spectral lines in an optimal way .
we developed a code in idl which measures ews via an inverse - variance weighted integration of the flux across the line .
the inverse variance ( @xmath105 ) for each pixel in a spectrum is automatically generated as output from the deimos reduction pipeline .
weighting by inverse variance minimizes the effects of poorly subtracted sky lines or other low s / n regions of the spectrum in the calculation of ews , and allows for accurate uncertainty estimates on the indices .
the code also robustly measures _ lower limits _ on emission line ews in spectra where emission lines are bright but the underlying stellar continuum is undetected . as in earlier papers , we adopt the convention that emission lines have negative equivalent widths and absorption lines positive . hereafter , we refer to the equivalent widths in the [ o@xmath35ii ] line as simply `` [ o@xmath35ii ] '' , and we refer to h@xmath38 equivalent widths by the index name , `` h@xmath103 '' . in total , we measure ews for samples of 116 and 124 spiral galaxies in ms 0451 and cl 0024 , respectively , as well as samples of 130 and 109 e+s0s in ms 0451 and cl 0024 .
photometric characteristics and key spectral line measurements for each of these galaxies are listed in table [ tab : data ] .
while our spectroscopic sample of cluster members contains no broad - lined agn , narrow - lined agn are harder to identify due to the lack of h@xmath60 in most spectra , complicating identification based on spectral line ratios . in 4 , we therefore must consider the possibility of agn contamination through other means .
in determining the fundamental plane of early type galaxies ( 6 ) , we require measurement of the line of sight velocity dispersion from the spectra of e+s0 cluster members . despite a variety of spectral setups ,
as detailed above , all of our deimos observations yield spectra with a velocity resolution of @xmath106 km s@xmath16 .
as described in paper iii for cl 0024 , we measure velocity dispersions for ms 0451 early types using the gauss - hermite pixel - fitting code of @xcite , limiting our sample to those observed galaxies with spectral signal to noise ( s / n ) @xmath107 ( @xmath108 ) , with a median s / n@xmath109(@xmath108 ) .
extensive testing of this code discussed in paper iii and @xcite imply errors of @xmath110 in the derived velocity dispersions for galaxies above this s / n cutoff . in total , 60 e+s0s in ms 0451 and 71 in cl 0024 have high s / n spectra suitable for measurement of the velocity dispersion .
the discovery of passive spirals in clusters at intermediate redshift @xcite has led to their recognition as promising candidates for the role of ` transition object ' during the theorized conversion of cluster spirals into s0s between @xmath4 and today .
such a conversion , though controversial ( e.g. , * ? ? ?
* ) , has been expected because of the striking contrast between the large number of spirals observed in clusters at @xmath4 and the correspondingly large population of s0s found in clusters locally . in our initial analysis of these passive spirals in cl 0024 @xcite , we found evidence in their stellar populations for a slow decline in star formation , confirming that passive spirals at least in cl 0024exhibit the prerequisite cessation of star formation needed for any transformation into s0s .
perhaps more importantly , we found that the _ abundance _ of passive spirals in cl 0024 is quite high , over 25@xmath25 of the total spiral population
enough , perhaps , to account for the _ entire _ buildup of s0s in clusters ( but see * ? ? ?
however , several questions remain about the nature of passive spirals and their purported s0 end states . among them : apart from their lack of emission lines , how do the passive spirals as a class differ from the star forming spirals , and are their properties uniform between clusters ?
if they are truly destined to transform into s0s , can we identify directly these new s0s , perhaps just after their new morphologies become firmly in place ? finally , in what environments do the passive spirals reside , and with what implications for the physical mechanisms driving their creation ? in this section , we attempt to answer the first of these questions , specifying more fully the properties and recent star formation histories of the passive spirals . in 5 and 6
, we will then consider the next two questions , in our attempt to piece together the full evolutionary history of cl 0024 and ms 0451 galaxies .
we define as ` passive ' any spiral with an [ o@xmath35ii ] equivalent width [ o@xmath35ii]@xmath111 .
though this definition necessarily allows some contamination from star - forming galaxies that are either moderately dusty or forming stars at a low rate , we will demonstrate below via coadded spectra that all emission lines are weak or absent in the passive spirals .
we note that we would classify as ` passive ' many of the spiral galaxies in local clusters that are observed to be hi - deficient but retain a low level of star formation @xcite . by examining multi - color ( nuv and f814w ) imaging of the cluster by eye
, we remove from the sample all galaxies where the uv flux is likely to be significantly contaminated by neighboring objects or image artifacts .
lcccc cl 0024 & @xmath112 & @xmath113 & @xmath114 & @xmath115 + ms 0451 & @xmath116 & @xmath117 & @xmath118 & @xmath119 + we also restrict our passive spiral sample to those objects with h@xmath120 ; we assign otherwise passive spirals with h@xmath103 stronger than this to the ` post - starburst ' category .
we note that this h@xmath38 limit differs from the definition adopted by some other authors ( e.g. , * ? ? ?
* ) , who label all passive galaxies with h@xmath121 as post - starburst .
however , the difference is primarily due to different measurement methods for h@xmath38 : we find that a large number of star - forming and passive spirals exhibit @xmath122h@xmath123 in our sample , which is within the range where @xcite report that star formation histories are consistent with continuous star formation .
therefore , spiral galaxies in our sample with [ o@xmath35ii]@xmath111 and h@xmath103 between @xmath124 and 5@xmath63 are perhaps more properly called ` post-_starforming _ ' , as there is no need to invoke a starburst to explain their h@xmath103 values .
we therefore group them with the remainder of the passive spirals .
distributions of fuv _
v _ colors in ms 0451 ( top ) and cl 0024 ( bottom ) .
open blue histograms indicate the colors of normal star - forming spirals .
green hatched histograms display the colors for those passive spirals that are detected in the uv , and solid red histograms likewise indicate the colors of e+s0s that have been detected in the uv . no correction for dust extinction
is applied .
uv - detected passive spirals in both clusters have intermediate colors between those of star forming spirals and those few early types detected in the uv.,title="fig:",width=312 ] + distributions of fuv _
v _ colors in ms 0451 ( top ) and cl 0024 ( bottom ) .
open blue histograms indicate the colors of normal star - forming spirals .
green hatched histograms display the colors for those passive spirals that are detected in the uv , and solid red histograms likewise indicate the colors of e+s0s that have been detected in the uv . no correction for dust extinction
is applied .
uv - detected passive spirals in both clusters have intermediate colors between those of star forming spirals and those few early types detected in the uv.,title="fig:",width=312 ] after culling contaminated objects and dividing our sample according to their [ o@xmath35ii ] and h@xmath103 strengths , we find that @xmath125 of ms 0451 spirals with [ o@xmath35ii ] emission ( brighter than @xmath97 ) are detected in our _ galex _ imaging , similar to the detection rate in cl 0024 .
conversely , only @xmath126 of spectroscopically passive spirals are similarly detected ( dropping to @xmath127 if we were to include the post - starburst spirals ) , significantly less than the uv - detected fraction in cl 0024 ( see table [ tab : uv ] ) .
as the rest - frame luminosity limits for detection of fuv are virtually the same between the two _ galex _ exposures ( @xmath128 in ms 0451 and @xmath129 in cl 0024 ) , this represents an important difference in uv luminosity between the two clusters passive spiral populations . in figure [ uvcolor ] , we display the distributions of rest - frame fuv _
colors for star - forming spirals , passive spirals , and early types in each cluster .
it is clear that , despite the overall lower fraction of uv - detected passive spirals in ms 0451 , those that are detected exhibit the same intermediate colors as in cl 0024 . to further investigate the colors of passive spirals , and to include those that were not uv - detected
, we also utilize our ground - based _
r _ and _ i_-band imaging to measure rest - frame _ b v _ , for passive and normal spirals .
we observe a trend similar to that seen in the uv optical colors : the median passive spiral is @xmath52 magnitudes redder than the median star - forming spiral in cl 0024 , and 0.2 magnitudes redder in ms 0451 .
ks tests indicate that passive spirals have a color distribution significantly different from that of normal spirals , in both clusters .
in fact , we can verify visually that passive spirals appear different from normal star forming spirals . in figure [ montage
] , we display color postage stamp images of nine spirals in cl 0024 , derived from deep multicolor acs imaging of the center of cl 0024 which we retrieved from the _ hst _ archive ( gto proposal 10325 , pi : ford ) . in the top row , we display three normal star - forming cluster spirals .
the next two rows show cluster passive spirals , which appear to fall into two basic types . shown in the middle row
, some passive spirals appear to retain blue disks , but at surface brightnesses lower than that of star - forming spirals .
while some star formation may be occurring in these galaxies , we note that our spectroscopic slits are large enough that we do not believe we are simply ` missing ' regions of star formation in our observations . in the bottom row , we see three examples of passive spirals with distinctly red disks , yet with spiral arms and dust lanes still present . these appear similar to the @xmath130 ` anemic spirals ' first identified by @xcite ; indeed , inspection of the integrated spectrum of a prototypical ` anemic spiral ' , ngc 4569 @xcite , indicates that it would likely satisfy our definition of ` passive spiral ' .
it seems , then , that passive spirals could genuinely be disk - dominated systems where star formation is on the decline or has halted . in counterpoint to their lower incidence of uv emission , the overall frequency of passive spirals in ms 0451
is quite similar to that found in cl 0024 ( table [ tab : uv ] ) . in both clusters ,
more than a quarter of all spirals are passive important confirmation that passive spirals are a significant component of both clusters despite the large difference between the clusters global properties .
yet their presence in two quite distinct clusters leads us to question whether the passive spirals are a cluster - related phenomenon at all , or if they could instead represent some fraction of all spirals that have internally exhausted their star formation .
if they are indeed generated in the cluster environment , then we would not expect to find passive spirals in the field at these same intermediate redshifts .
we have therefore examined a sample of 105 field spirals in the redshift range @xmath131 , identified in the course of our spectroscopic campaign .
out of 62 galaxies where [ o@xmath35ii ] falls within our wavelength coverage , we measure a passive spiral fraction of only @xmath132 .
this low incidence in the field confirms that the generation of passive spirals is a cluster - related phenomenon at these redshifts @xcite .
the @xmath132 of passive spirals found in the field may indicate that some passive spirals can be formed in groups . despite their overall similar abundance in both clusters , the weaker uv emission in ms 0451 passive spirals
presents a puzzle that leads us to again consider the different assembly states and icm properties of the two clusters . to more
precisely quantify the nature of the passive spirals in each cluster , we introduce as a key diagnostic plot the fuv _
v _ vs @xmath104 diagram .
@xmath104 is sensitive to stellar populations with ages of @xmath133 gyr , with a dependence on metallicity that only becomes apparent for old stellar populations @xcite , while the fuv _
v _ color is sensitive to star formation on a much shorter timescale ( 10@xmath13410@xmath135 yr ) . as we will see , comparing @xmath104 to fuv _
v _ color provides a valuable tool to discriminate between different star formation histories for passive and star - forming spirals . in figure
[ hd_uvnew ] , we plot @xmath104 versus rest frame fuv _
color for ms 0451 ( top ) and cl 0024 ( bottom ) .
it is clear from the diagram that most star - forming spirals in both clusters exhibit blue fuv _
v _ color ( median of @xmath136 ) combined with weak @xmath104 strength ( median @xmath137 ) .
such values indicate young stellar populations , and are expected for galaxies with ongoing star formation .
fuv _
v _ colors versus @xmath104 strength for spirals in ms 0451 ( top ) and cl 0024 ( bottom ) .
star - forming and passive spirals are indicated by open blue triangles and solid green triangles , respectively .
arrows denote 3-@xmath31 lower limits for fuv _
colors for those galaxies not detected in the uv , colored blue for star - forming spirals and green for passive spirals .
small triangles or lower limit arrows indicate spirals with h@xmath138 , similarly color coded .
we plot only galaxies with especially small uncertainty in @xmath104 , @xmath139 .
solid , dashed , dotted , and dash - dotted lines respectively indicate model tracks for exponentially declining , dust - enshrouded , truncated , or ` starved ' star formation , as described in the text .
tracks are colored red for regimes where [ o@xmath35ii]@xmath111 . most uv - detected passive spirals are consistent with a starvation model , while most undetected passive spirals are better fit by a rapid truncation of star formation.,title="fig : " ] fuv _
v _ colors versus @xmath104 strength for spirals in ms 0451 ( top ) and cl 0024 ( bottom ) .
star - forming and passive spirals are indicated by open blue triangles and solid green triangles , respectively .
arrows denote 3-@xmath31 lower limits for fuv _
colors for those galaxies not detected in the uv , colored blue for star - forming spirals and green for passive spirals . small triangles or lower limit arrows indicate spirals with h@xmath138 , similarly color coded .
we plot only galaxies with especially small uncertainty in @xmath104 , @xmath139 .
solid , dashed , dotted , and dash - dotted lines respectively indicate model tracks for exponentially declining , dust - enshrouded , truncated , or ` starved ' star formation , as described in the text .
tracks are colored red for regimes where [ o@xmath35ii]@xmath111 . most uv - detected passive spirals are consistent with a starvation model , while most undetected passive spirals are better fit by a rapid truncation of star formation.,title="fig : " ] turning to examine the locations of passive spirals , we note that passive spirals in ms 0451 largely occupy a different region of the plot than cl 0024 passive spirals .
cl 0024 passive spirals with uv detections the majority
cluster quite tightly on the plot at moderate fuv _
v _ colors ( median @xmath140 ) and moderately strong @xmath104 ( median @xmath141 ) .
in contrast , as we have already noted , many of the ms 0451 passive spirals have only upper limits on their fuv emission , some at quite red fuv _
v _ colors . across both clusters ,
these uv - undetected objects seem also to exhibit a wider spread in @xmath104 than the uv - detected passive spirals ( rms of @xmath142 c.f .
@xmath143 ) .
the dominance of uv - detected passive spirals in cl 0024 , compared to the uv - undetected type that are predominant in ms 0451 , begs the question of whether passive spirals are created in different ways in each of the two clusters . as we will show , the two types of passive spirals
uv detected and undetected may have fundamentally different star formation histories . to help decipher the star formation histories of passive spirals in cl 0024 and ms 0451 , we overlay in figure [ hd_uvnew ] evolutionary tracks from the population synthesis code of @xcite , adopting fixed solar metallicity for all models . using these
, we can test several classes of models that could explain the origin of passive spirals .
specifically , can passive spirals be created by simply adding dust to a star - forming spiral ? can they be created by suddenly switching off star - formation in a galaxy , which we will call ` truncation ' models ? or , finally , can we create the passive spirals by switching a star - forming spiral into a phase where its star formation rate declines more rapidly over a timescale of 0.52 gyr ( a ` starvation ' model ) ? in figure [ hd_uvnew ] , solid lines indicate the track followed in fuv _
v _ vs. @xmath104 space by several model stellar populations with exponentially - declining sfrs , with characteristic timescales @xmath1447 gyr .
such models have been found to reproduce the optical spectra of sa - sc type star forming galaxies @xcite . dashed
lines add dust extinction of @xmath145 to these same models , corresponding to the mean difference in fuv _
v _ between the ( uv - detected ) passive and active spirals .
active spirals largely occupy the region in between tracks with zero and moderate extinction .
tracks are in red for regimes where [ o@xmath35ii]@xmath146 , estimated from the models in the same way as in @xcite . only models with a relatively fast exponential decline in sfr ( @xmath147 gyr , upper solid and dashed lines in figure [ hd_uvnew ] ) can match the peculiar combination of intermediate fuv _
v _ and strong @xmath104 exhibited by the bulk of the uv - detected passive spirals .
any star forming spirals that transition into this sort of rapid decay of star formation , eventually entering a spectroscopically passive phase , will reproduce the positions of the uv - detected passive spirals in the diagram ( thick dash - dotted ` starvation ' line in figure [ hd_uvnew ] ) .
conversely , passive spirals with lower limits on their color , which dominate the population in ms 0451 , seem to be most consistent with models where star formation is rapidly truncated at an age @xmath148 gyr , indicated by the dotted lines in figure [ hd_uvnew ] .
stacking together the uv images for undetected passive spirals , we still detect no significant uv emission .
though confusion noise becomes significant in the stacked image , the nondetection implies that the median fuv _
v _ is at least @xmath149 magnitude fainter than the upper limits indicated in figure [ hd_uvnew ] .
the positions of these uv - undetected spirals on the plot could also be explained by strong levels of dust obscuration .
however , 24@xmath150 m imaging of both clusters with mips on the _ spitzer space telescope _ indicates that there is a deficit of obscured dusty starbursts in ms 0451 , in comparison to cl 0024 @xcite .
if the uv - undetected passive spirals were simply dusty , we would expect to see the _ opposite _ trend in the mips observations , since there are so many of these objects in ms 0451 .
we therefore believe that the rapid truncation of star formation is the most likely explanation for the uv - undetected passive spirals .
the population of spirals in ms 0451 includes a number of post - starburst galaxies that also seem to be consistent with rapidly - truncated star formation ; in fact , rapid truncation of a starburst is thought to be the primary way that these galaxies achieve such high h@xmath103 values @xcite .
indeed , the post - starburst spirals may be closely related to the uv - undetected passive spirals in ms 0451 .
several apparently star - forming galaxies also reside along the ` truncated ' track .
together , these could represent a continuum of galaxies in various stages of having their star formation halted .
however , the supposedly rapid timescale for the cessation of star formation begs the question of why we would see any star - forming spirals in the region of the plot where models indicate that they should already be passive .
some mix of models , with increased dust obscuration combined with the truncation of star formation , could provide an explanation .
while we have found passive spirals to be abundant in both cl 0024 and ms 0451 , it appears that they are largely formed through different mechanisms in each cluster . the more rapid cessation of star formation required to explain the ms 0451 passive spirals is likely due to some physical mechanism that exerts a stronger force on galaxies in ms 0451 than in cl 0024 .
this once again brings to mind the hot , dense icm of ms 0451 , which at least has the potential to apply a much stronger force ( ram - pressure stripping ) on infalling galaxies than in cl 0024 . in the next section
, we will examine the stellar populations of early type galaxies across the two clusters , in an attempt to identify the expected end products of these rapidly and slowly quenched passive spirals : s0s with signatures of recent star formation on varying timescales .
while careful analysis of local s0 galaxies has revealed some signs that their stellar populations are younger than those of ellipticals ( e.g. , * ? ? ?
* ) , such signatures have so far been elusive at intermediate redshift , despite expectations that s0s with newly halted star formation would be abundant .
now , however , under the hypothesis that passive spirals in ms 0451 and cl 0024 must be disappearing from our spiral sample as they fade in uv and increase in @xmath104 , we turn again to a study of the stellar populations of early type galaxies . the different characteristics of passive spirals in cl 0024 and ms 0451 help to provide a key for uniquely linking the passive spirals to their potential s0 descendants . because the properties of passive spirals largely differ between the two clusters , so too will any s0s that have just recently changed their morphology . with this in mind
, we wish to examine recent star formation as well as any ongoing star formation in the cluster e+s0s . in figure
[ dn4000_oii ] , we plot @xmath104 strength versus [ o@xmath35ii ] equivalent width for all early types in cl 0024 ( left ) and ms 0451 ( right ) , with symbols coded to indicate their morphologies ( as shown in the legend ) .
most bright e+s0s in both clusters exhibit the signatures of old stellar populations : weak or absent [ o@xmath35ii ] and a strong balmer break ( @xmath151 ) . in cl 0024 , however , we observe a striking tail of galaxies extending from the locus of old e+s0s toward weaker @xmath104 , coupled with significant [ o@xmath35ii ] emission .
ms 0451 exhibits a similar tail of early type galaxies toward weak @xmath104 strength , but _ without _ any associated [ o@xmath35ii ] emission .
if we apply a cut at @xmath152 , indicated by the horizontal dotted lines in the figure , we find that 1015@xmath25 of the total e+s0 population in each cluster has @xmath104 below 1.55 . while somewhat arbitrary , the chosen @xmath104 dividing line lies well below the expected value for an early type galaxy that has been passively evolving since formation at @xmath153 .
in fact , in the case of a prototypical elliptical whose stars were created in a short burst , @xmath104 should fade from @xmath154 to @xmath155 in less than @xmath133 gyr , according to @xcite models less than the time from @xmath156 to @xmath4 .
there are essentially two possible explanations for finding e+s0 galaxies with such low @xmath104 : either they have undergone recent star formation , or they harbor an agn that contributes to the spectrum . if they have formed stars recently or are continuing to form stars , then they must have either formed a significant population of stars at @xmath157 while already displaying early type morphology , or else transformed morphology from a star - forming spiral to an early type ( either singularly or through a merger ) . under each of these scenarios , there must be some way to account for the presence of emission lines in the cl 0024 objects but a lack thereof in ms 0451 . a third possibility that we have simply misclassified some spiral galaxies as s0s is easy to discount based on this last requirement .
if the low @xmath104 s0s are truly just spirals , then we would not expect in ms 0451 to _ exclusively _ miscast as s0 just those spirals with no [ o@xmath35ii ] emission , while at the same time misclassifying spirals with a range of [ o@xmath35ii ] in cl 0024 .
on the other hand , if we are biased towards mixing up the morphological classifications of passive spirals and s0s , this only strengthens the notion that one type may be transforming into the other . to better assess the star formation histories of these peculiar objects , we overplot in figure [ dn4000_oii ] two characteristic model tracks equivalent to those plotted for the passive spirals in figure [ hd_uvnew ] . in cl 0024
, we plot a starvation - like track ( @xmath158 gyr with internal @xmath145 ) , while in ms 0451 , we plot the track of a galaxy with star formation truncated rapidly at an age of 5 gyr . in each case
, the model track reproduces the positions of the galaxies very well .
we note that most physically plausible truncated - spiral tracks are inconsistent with the distribution of points in cl 0024 , as no such models reach @xmath159 before truncation , yet we still see galaxies with [ o@xmath35ii ] emission in this range .
remarkably , then , a single starvation track reproduces the positions of both the low-@xmath104 early types and most passive spirals ( figure [ hd_uvnew ] ) in cl 0024 , while a truncated star formation track similarly matches the star formation histories of both classes in ms 0451 .
these similarities in star formation history suggest an evolutionary link between the passive spirals and the low-@xmath104 early types in each cluster .
we can further evaluate the likelihood of a connection between the passive spirals and the low-@xmath104 early types through a consideration , for each cluster , of the coadded spectra of both classes . in figure [ coadded ] , we plot in the top panel the coadded spectra of all e+s0s in the clusters with @xmath160 , and in the bottom panel , the coadded spectra of passive spirals in each cluster .
in each case , we only include galaxies with overall spectral @xmath161 ( @xmath108 ) in the summation .
the locations of several key spectral lines are marked .
top : normalized , coadded spectra of e+s0s in cl 0024 and ms 0451 with @xmath160 , including only spectra with @xmath162 ( @xmath108 ) .
cl 0024 spectrum is shifted upward in flux ( arbitrary units ) for display purposes .
the locations of several key spectral lines are marked .
bottom : normalized , coadded spectra of passive spirals in cl 0024 and ms 0451 , also restricted to spectra with @xmath163 . in each cluster ,
the [ o@xmath35ii ] , h@xmath103 , and h@xmath59 strengths of passive spirals closely resemble those of the low-@xmath104 early types in the same cluster . ,
title="fig:",width=312 ] + top : normalized , coadded spectra of e+s0s in cl 0024 and ms 0451 with @xmath160 , including only spectra with @xmath162 ( @xmath108 ) .
cl 0024 spectrum is shifted upward in flux ( arbitrary units ) for display purposes .
the locations of several key spectral lines are marked .
bottom : normalized , coadded spectra of passive spirals in cl 0024 and ms 0451 , also restricted to spectra with @xmath163 . in each cluster ,
the [ o@xmath35ii ] , h@xmath103 , and h@xmath59 strengths of passive spirals closely resemble those of the low-@xmath104 early types in the same cluster .
, title="fig:",width=312 ] considering the coadded early - types first , we can immediately cast doubt on the idea that agn are responsible for the low @xmath104 . while a substantial contribution from a flat agn continuum can serve to weaken the balmer break in a galaxy ,
the same effect should also dilute the observed depth of stellar absorption lines , such as h@xmath103 .
yet it is clear from the coadded spectra that the low-@xmath104 early types in both clusters exhibit moderately strong h@xmath103 : @xmath164 and @xmath165 for ms 0451 and cl 0024 , respectively .
as both values are much higher than the median h@xmath166 of cluster ellipticals , their balmer strengths are consistent with the interpretation that they contain young stars , and _ inconsistent _ with a significant agn ( dust obscured or otherwise ) contributing to the galaxy s spectrum .
though the spectra make clear that these ` young s0s ' ( as we will call them ) truly contain a population of young stars , is it possible that these galaxies have recently undergone a ` rejuvenation ' of star formation through some interaction ?
serious doubt is cast on this hypothesis by the lack of ms 0451 e+s0s currently containing emission lines in their spectra .
after a moderate starburst on top of an established stellar population , @xcite models predict a return to a passive spectrum with strong @xmath104 in as little as 100 myr after the burst . to have so many s0s experience a rejuvenation episode so recently , yet observe none of them currently in a starburst phase , is implausible .
we are left , then , with the possibility that these young s0s truly represent the end states of the passive spirals .
the contrast between young s0s in cl 0024 , whose [ o@xmath35ii ] strengths are consistent with a gradual decline in star formation , and those in ms 0451 , where the lack of [ o@xmath35ii ] indicates recent truncation , evokes the similar dichotomy between passive spirals with a slow truncation of star formation ( mostly in cl 0024 ) and those exhibiting a more rapid truncation ( mostly in ms 0451 ) .
comparing the coadded spectra of the two classes in more detail , we find further evidence to support an evolutionary link between passive spirals and young s0s .
first , the passive spiral spectrum for cl 0024 exhibits a weak [ o@xmath35ii ] emission line of @xmath167 , despite the population having been selected for their low [ o@xmath35ii ] .
the presence of some [ o@xmath35ii ] emission in both the passive spirals and the young s0s ( [ o@xmath35ii]@xmath168 ) is consistent with the slow decay in star formation thought to be acting .
it may be that , in some cases , the transformation to s0 morphology occurs _ before _ the final cessation of star formation .
ms 0451 passive spirals , in contrast , have an [ o@xmath35ii ] equivalent width consistent with zero , as do their s0 counterparts . again , this is consistent with the rapid truncation of star formation in ms 0451 passive spirals , followed by an almost simultaneous transformation into s0 morphology .
the morphological transformation can not be significantly delayed after the halt in star formation , because the s0s low @xmath104 and strong h@xmath103 strictly limit the time since last star formation to @xmath169 myr , according to the model tracks in figures [ hd_uvnew ] and [ dn4000_oii ] . secondly , the h@xmath103 values of passive spirals in each cluster are similar to those of their counterpart young s0s . in cl 0024 , passive spirals exhibit weaker
h@xmath103 than normal star forming spirals
h@xmath170 in the coadded spectrum compared to a median h@xmath171 for star - forming spirals ( see also * ? ? ?
the coadded spectrum of young s0s exhibits a similarly moderate h@xmath172 , between the typical values for spirals and for ellipticals . in ms 0451 ,
passive spirals ( h@xmath173 ) and young s0s ( h@xmath174 ) exhibit h@xmath103 strengths that also mirror each other closely . yet
they exhibit overall higher strengths than in cl 0024 , another indication that the halt in star formation was quite recent for galaxies in ms 0451 .
the close similarities between passive spirals and young s0s _ within _ each cluster , coupled with the clear differences in populations _ between _ the two clusters strongly argues in favor of a model where passive spirals and young s0s represent an evolutionary sequence . at the same time
, we have not yet fully explored possible explanations for why this evolutionary sequence behaves differently in the two clusters . in the next section ,
we begin consideration of this issue by examining the local environments in which passive spirals and young s0s are found .
so far , we have treated the populations of passive spirals and young s0s largely as a uniform population within each cluster . the reality , however , is more complex .
while the passive spirals in ms 0451 were largely found to have different star formation histories than cl 0024 passive spirals , the segregation is imprecise : ms 0451 contains a proportion of uv - detected spirals ( @xmath175 ) consistent with a slow decline in star formation , while cl 0024 conversely contains a similar fraction of passive spirals whose star formation may have been truncated more rapidly . to help disentangle this puzzle ,
we focus now more closely upon the local environments in which the passive spirals and their remnant s0s are found .
we start by examining the spatial distribution of transition galaxies across both clusters . in figure
[ spatial_dist ] , we plot the positions of all passive spirals ( green squares ) and young s0s ( red triangles ) .
we also plot the populations of compact narrow emission line galaxies ( cnelgs ) found in each cluster ( blue pentagons ) , which will be discussed further in 7 .
the first striking feature to note in figure [ spatial_dist ] is that passive spirals and young s0s occur both in the cluster core and at higher radius , even beyond @xmath26 , with no obvious segregation between the passive spirals and s0s .
secondly , in ms 0451 both passive spirals and young s0s appear to be more spread out across the cluster than in cl 0024 , where more than half are located within 1 mpc of the cluster center .
both features reveal important clues to the physical processes at work .
careful inspection of those passive spirals found beyond the cluster cores gives the impression that many have other cluster members nearby .
could passive spirals be preferentially found in infalling groups ?
some evidence for passive spirals in groups at these redshifts has already be noted @xcite . to evaluate quantitatively this possibility ,
we calculate the projected local densities of these passive spirals , using the 10 nearest neighbors method , following paper i and @xcite . considering only galaxies outside of the cluster cores ( @xmath176 mpc in cl 0024 and @xmath177 mpc in ms 0451 , @xmath19 of @xmath26 )
, we find that passive spirals are found at a median local density of @xmath178 mpc@xmath179 in ms 0451 and @xmath180 mpc@xmath179 in cl 0024 , both somewhat higher than the density of a typical infalling spiral of @xmath181 mpc@xmath179 across both clusters .
passive spirals , therefore , seem to be forming both within infalling groups and closer to the cluster core two distinct regions where the dominant physical forces acting on galaxies are likely to be quite different . remarkably , we find that four out of the five uv - detected passive spirals in ms 0451 reside in groups outside the cluster core .
recalling from figure [ hd_uvnew ] that these uv - detected passive spirals closely follow a ` starvation - like ' gradual cessation of star formation , we conclude that _ nearly all _ passive spirals within infalling groups are experiencing a cessation of star formation spread over a @xmath149 gyr timescale .
in contrast , virtually all ms 0451 passive spirals within 1.5 mpc of the cluster center have undergone a rapid truncation in star formation . in cl
0024 it appears that rapid truncation of star formation can also occur , but only extremely close to the cluster core : @xmath182 passive spirals without uv detection are located within 300 kpc of the cl 0024 center . while passive spirals are abundant in the region around the ms 0451 center , there is a hole of @xmath183 kpc radius , within which we observe almost no spirals of any type . unlike in cl 0024 , where spirals are observed even within @xmath184 kpc of the central galaxies ,
no spirals survive to reach the center of ms 0451 . even accounting for the larger virial radius of ms 0451 ,
this spiral - free zone is significantly larger than that seen in cl 0024 : within @xmath185 ( @xmath186 kpc in ms 0451 and @xmath187 kpc in cl 0024 ) , only @xmath188 of ms 0451 members in our sample are spirals ( @xmath189 passive ) , compared to @xmath190 ( @xmath191 passive ) in cl 0024 . as the time required for a galaxy to travel @xmath183kpc across the core of ms 0451 is only 0.4 gyrs , it appears that spirals in this environment must indeed be converted to early morphology quite rapidly .
both the central hole in ms 0451 and the central concentration of uv - undetected passive spirals in cl 0024 can be better understood by considering the local icm densities of the transition galaxies in each cluster . in figure
[ rampressure ] , we make use of _ chandra _ x - ray data for each cluster to plot the expected strength of ram - pressure as a function of radius for cl 0024 and ms 0451 .
each track is generated by calculating the gas density @xmath192 from the best - fit isothermal @xmath59-model from @xcite and @xcite , for ms 0451 and cl 0024 , respectively .
we then estimate ram pressure @xmath193 by adopting each cluster s line of sight velocity dispersion , @xmath31 , as the characteristic velocity of a galaxy in that cluster .
strength of ram pressure as a function of radius , for both cl 0024 and ms 0451 .
solid lines indicate ram pressure @xmath194 as described in the text .
dotted line indicates the pressure required to strip a spiral with characteristics of the milky way ( paper i * ? ? ?
overplotted on each track are the radial positions of passive spirals ( uv - detected and undetected ) and young s0s in that cluster .
where the ram pressure falls below the lower bound of the plot , we mark the positions of galaxies along the lower edge of the plot .
most uv - undetected passive spirals are found in regimes of high ram pressure , while no passive spirals survive in regimes where ram pressure is more than a few times that required to strip a milky way - like spiral.,width=312 ] figure [ rampressure ] reveals that virtually all uv - undetected passive spirals are found in regimes where the ram pressure is significant ( @xmath195 of that required to strip a milky way analogue ) , while uv - detected passive spirals are largely confined to regions of lower gas density . while the corresponding radial ranges differ between cl 0024 and ms 0451 , the uv - undetected passive spirals span a nearly - identical range of ram pressure strengths in each cluster .
this argues strongly in favor of ram pressure stripping as the mechanism responsible for the uv - undetected passive spirals .
we note that no spirals survive in environments where the ram pressure is more than a few times that required to strip the milky way , and a natural consequence of this limit is that no spirals in ms 0451 survive to reach @xmath196 kpc ( though , as expected , several young s0s appear at slightly lower radii ) . in 7.2 , we will discuss in more detail the relation between this presumed ram pressure stripping and other mechanisms that may be acting to transform morphology .
so far , we have shown in 4 and 5 that the large population of passive spiral galaxies in both cl 0024 and ms 0451 are transforming into s0s on a variety of timescales . building on this , in 6 , we have successfully matched the timescales of transformation with specific environments within each cluster .
consequently , we are now able to construct a road map describing the sites and timescales of transformations across both clusters . to summarize , we find that some passive spirals are generated within infalling groups , experiencing a gradual decline in star formation rate and eventual transformation to s0 morphology . in the central regions of both clusters , a high proportion of spirals are seen to be passive . in cl 0024 , star formation within these spirals
continues to decay at a relaxed pace , with few signs of rapid interaction with the cluster environment except near the very center where the effects of ram pressure begin to be important . in contrast
, passive spirals all across the core of ms 0451 have had their star formation truncated rapidly , with a subsequent rapid transformation into s0 morphology . in this section
, we use this road map to consider the constraints that we can place on the physical processes operating in each of these three regimes : groups in the cluster outskirts , the cl 0024 core , and the ms 0451 core . in the hierarchical assembly of clusters ,
groups infalling from the cluster outskirts can be thought of as the building blocks .
paper i argued that the existence of the morphology density relation implies that a galaxy s evolutionary history is tightly linked to its group s history , until the group is absorbed into the cluster .
furthermore , the observation that mass and light trace each other tightly on large to small scales ( paper ii ) suggests that groups of roughly constant mass to light ratio rather than individual galaxies coalesce to build up the cluster .
the ` preprocessing ' of galaxies within these infalling groups is a crucial feature of evolution within both clusters , as here we are probing galaxies in a key density range at a key point in time : at @xmath197 , observations indicate that the early type fraction is growing only in the cores of clusters , while below @xmath4 evolution begins to accelerate in lower density regions @xcite . as such , gaining an understanding of the physical processes driving galaxy transformation within the infall regions of clusters at @xmath4 can shed light on the observed evolution in the morphology density relation . by virtue of the fact that passive spirals in the cluster outskirts are preferentially found in groups , we do not expect that the clusters icm plays a direct role in suppressing star formation . even in ms 0451
, the force of the icm should not be significant beyond the virial radius ( figure [ mechanisms ] ) . and
while we had previously argued in favor of starvation by the icm @xcite , which can be effective at large cluster radius , there is little reason to expect it to act on galaxies in groups more effectively than isolated spirals .
however , we can not rule out the presence of gaseous interactions _ within _ groups . indeed
, x - ray observations of groups in the local universe reveal that they often harbor an intra - group medium @xcite , and simulations by @xcite suggest that the starvation mechanism can operate similarly within large groups as it would in the overall cluster ( see also * ? ? ?
* ; * ? ? ?
intriguingly , x - ray bright groups seem to contain a higher early - type fraction than x - ray undetected groups @xcite , consistent with the notion that the gas plays a role in shaping galaxy morphology .
the predicted timescale for cessation of star formation under starvation is consistent with our observed decline over @xmath198 gyr @xcite ; in simulations by @xcite , a timescale of 12 gyr for the halt in star formation after gas removal provided a good fit to observations .
we emphasize , however , that the @xmath149 gyr decline in star formation represents only one observable phase of the predicted total timescale for conversion of a spiral to an s0 via starvation ( @xmath199 gyr , * ? ? ?
* ; * ? ? ?
the slow action of starvation would preserve recognizable spiral structure for over 1 gyr after star formation stops , and we have shown in @xcite that these lifetimes for each phase are consistent with the hypothesis that all such passive spirals will become s0s by @xmath130 . yet direct observations of the intra - group gas are challenging at these redshifts , especially for small groups ( but see * ? ? ?
* ) , and other mechanisms may also be able to reproduce the observed timescale for the cessations of star formation .
in addition , some models suggest that modification of star formation rates due to starvation only begin to be visible many gyr after the halo is stripped @xcite , and so the effectiveness of starvation in groups would depend in some sense on the dynamical age of each group , which is difficult to quantify .
gentle gas stripping or starvation by a diffuse intra - group medium therefore remains an attractive , but difficult to test , possibility .
alternatively , the increased frequency of galaxy galaxy interactions among members of a group , compared to the interaction rate between isolated galaxies , could be driving the creation of passive spirals in the cluster outskirts .
we note that , at the group scale , the effects of impulsive galaxy galaxy interactions are difficult to separate from galaxy galaxy tidal interactions that act over a longer timescale , and so we consider them together here .
while galaxy galaxy harassment at the typical velocity dispersions of bound groups has not been widely studied , there are some advantages to this explanation .
first , on a larger cluster scale , simulations indicate that harassment of massive cluster spirals results in gas being funneled to the central bulge @xcite , providing a natural mechanism to accelerate star formation and deplete a spiral s gas supply ; a similar gas funneling effect due to tidal effects during group preprocessing has also been proposed ( e.g. , * ? ? ?
while simulations would be necessary to test the hypothesis , one could imagine a ` gentle harassment ' mechanism that induces a transport of gas to the galaxy center without immediately destroying the spiral arms .
a slow group - scale harassment could possibly reproduce the timescales for quenching of star formation indicated by the ` starvation ' model tracks in figure [ hd_uvnew ] .
we note that the timescale for final conversion to s0 morphology may be different from the timescale under starvation , and could provide an important discriminator between the two possibilities in the future .
depending on the rate of gas transport , it is possible that a brief increase in star formation rate could be detected before the final quenching @xcite , and this could provide another way to distinguish between this mechanism and starvation .
however , within the uncertainties , we can not confirm any small starbursts among galaxies within infalling groups in our sample .
dividing our total sample of star - forming galaxies by radius into three equally - sized bins , we do not see evidence for an enhancement of either the mean or median [ o@xmath35ii ] of a star forming galaxy in any of the three radial zones . nor do we see any change in the fraction of star - forming galaxies with strong ( @xmath200 ) [ o@xmath35ii ] .
however , as the expected lifetime for a burst is short @xcite , we can not rule out the presence of small bursts that change the median specific star formation rate by less than 20% .
a definitive distinction between gas - related processes and galaxy interactions therefore remains elusive .
importantly , however , we have already established that galaxy galaxy harassment is affecting spirals across both clusters , and at a level that does not destroy their spiral structure : @xcite found that the high scatter in the cluster tully - fisher relation compared to the field was most readily understood if galaxy harassment is acting to perturb the kinematics of spirals .
@xcite also observed a strange deficit in the cluster outskirts of spirals with high central mass concentrations ; one possible explanation is that these galaxies have dropped out of our emission line sample as they become passive due to the aforementioned funneling of gas to their centers .
while hardly a settled issue , we expect that the action of harassment is more likely to be at the root of the galaxy evolution in infalling groups , as the mechanism is already known to be acting in cl 0024 and ms 0451 .
yet some contribution from the intra - group gas is also possible , and the two effects could act in combination .
future study of the bulges of spirals and s0s could help distinguish between the mechanisms by establishing whether significant mass has been funneled to centers of galaxies during their evolution , as predicted in the case of harassment .
we turn now to the cores of both clusters , where spirals in ms 0451 appear to have their star formation quenched at a more rapid pace than in cl 0024 .
the most obvious explanation for the accelerated truncation of star formation within the central region of ms 0451 is , of course , its much denser icm . as we saw in figure [ rampressure ] , rapid truncation of star formation appears to occur in both clusters only in regimes where the ram pressure is high .
due to the dense icm in ms 0451 , such an environment is much more widespread in this cluster , and so , unlike in cl 0024 , most of the passive spirals in its core are experiencing a rapid truncation of star formation .
ram pressure stripping , in regimes where it is effective , can strip an entire spiral disk in less than 100 myr @xcite .
detailed analysis of local analogs of passive spirals thought to be undergoing ram - pressure stripping suggest a timescale of @xmath201 myr for the process to produce a galaxy similar to the uv - undetected passive spirals we observe @xcite . both of these are consistent with the maximum timescale of 400 myr implied by the 600 kpc hole devoid of spirals at the center of ms 0451 .
further , there is some evidence that post - starburst / post - starforming galaxies in the coma cluster similar to , but fainter than those in ms 0451are generated via ram pressure stripping @xcite .
the fact that we observe passive spirals at all implies that the transformation to s0 is not precisely simultaneous with the halt in star formation @xcite . however , the ` truncated ' star formation history models presented in figure [ hd_uvnew ] suggest that @xmath104 strength will increase at a fast rate after truncation , such that we must be seeing young s0s within 100 myr of the halt in star formation . the similar fraction of galaxies observed in both the passive spiral phase and the young s0 phase support the idea that the lifetimes for each phase are comparable , at about 100 myr .
yet there are several remaining uncertainties in this simple picture .
ram pressure stripping alone does not alter disk kinematics and spiral structure on such a short timescale , so there must be an additional mechanism to speed the conversion . in the cluster cores , harassment and/or tidal interactions could provide this additional impetus ( e.g. , * ? ? ? * ; * ? ? ? * ) . in cl 0024 , in the absence of rapid truncation by the icm , a similar harassment or tidal process must be at work .
however , the complex substructure of cl 0024 , and particularly the effects of the recent cluster collision , may generate importance differences between the two clusters . in order to better constrain the mechanisms affecting passive spirals in the cores of cl 0024 and ms 0451 ,
we are aided by two supplementary observations , one of which illuminates the action of harassment , with the other providing further constraints on icm interactions particularly the hypothesis that shocks in the icm could be important .
as discussed in 2 , we found in cl 0024 a population of relatively faint elliptical galaxies with strong emission , concentrated in a narrow range in radius close to the virial radius .
we cast doubt on the idea that major mergers were the cause of these apparent starbursts , and we speculated that another rapidly acting physical interaction must be triggering these bursts of activity .
strong shocks in the icm of cl 0024 could easily have been generated @xcite during the cluster - subcluster merger @xcite , and these shocks could be responsible for the triggered starbursts / agn we observe .
if this were the case , we hypothesized that we would not see such bursts in ms 0451 , as its smoother large scale structure makes it less likely that such shocks would be generated .
referring back to figure [ spatial_dist ] , however , we see that objects of this type are indeed found in the outskirts of both clusters . while only half as many are identified in ms 0451 as in cl 0024 , this detection rate is consistent with the lower spectroscopic completeness of our ms 0451 campaign for galaxies at these low luminosities : in ms 0451 , 10% of detected cluster members are at @xmath202 , while in cl 0024 , the fraction is 20% . the properties of these objects compact size ,
[ o@xmath35ii]@xmath203 , and typical luminosities @xmath204 , associate them with a class of objects dubbed compact narrow emission line galaxies ( cnelgs ) @xcite , or their more luminous cousins the luminous blue compact galaxies ( lbcgs ) ( e.g. , * ? ? ?
analysis by @xcite of deep acs imaging of lbcgs reveals that more than @xmath205 show signs of a recent merger .
their positions outside of dense regions in both ms 0451 and cl 0024 support the notion that these galaxies may simply represent the remnants of a merger between small , gas - rich galaxies . in this interpretation
, two important constraints can be put on the physical processes acting on passive spirals .
first , we have eliminated the possibility that strong shocks in the cl 0024 icm have a significant effect , even on low mass galaxies .
secondly , the minimum cluster radii at which these cnelgs / lbcgs are found identifies empirically the point at which merging becomes impossible and harassment must begin to dominate any galaxy - galaxy interaction . in cl 0024
, this occurs nearly at the virial radius , while in ms 0451 it occurs at a similar projected radius of 1.5 mpc .
we expect , then , that harassment within 1.5 mpc should scale in strength similarly between the two clusters . if harassment is responsible for the slow conversion of spirals to s0 in the cl 0024 core , it should also be acting on ms 0451 spirals with similar strength .
an analysis of the fundamental plane can provide sensitive constraints on the star formation and kinematic histories of e+s0s ( paper iii ) . in cl 0024 , we observed that elliptical and s0 galaxies form a clear fp ( figure [ fp ] ) , yet they exhibit a high scatter , equivalent to a spread of @xmath33 in mass to light ratio ( @xmath34 ) .
, where @xmath59 is the slope of the fp and @xmath206 is each galaxy s deviation from the cluster s overall fp intercept @xmath207 . ] upon closer inspection , this high scatter appeared to occur only among galaxies within 1 mpc of the cluster core .
we suggested in paper iii that the enhanced scatter in the cluster core may somehow be related to the recent cluster - subcluster merger in cl 0024 @xcite . by examining the fp and its residuals for ms 0451 as well
, we can test the hypothesis that galaxy interactions in the core of cl 0024 are enhanced over that expected for a relaxed cluster . in constructing the fp for both clusters , surface brightness ( @xmath208 ) and effective radius ( @xmath209 )
are determined via the galfit software @xcite , which fits a 2d model to the _ hst _ imaging .
we fit models following the devaucoleur s function to each galaxy ( see paper iii ) , and apply a k correction from observed f814w magnitudes to rest - frame @xmath37-band surface brightness , @xmath208 , as discussed in 3 .
the resulting fp for our sample of 60 e+s0 galaxies in ms 0451 is shown in figure [ fp ] , alongside the previously published fp for cl 0024 .
for both cl 0024 and ms 0451 , we overplot the local fp from the coma cluster @xcite as a solid line , adapted to our chosen cosmology .
in addition , we plot two parallel dashed lines indicating the galaxies expected fp intercept if they had formed in a single burst at @xmath210 or @xmath211 , calculated using the @xcite models .
it is clear that the e+s0 populations in both clusters have largely formed their stars at @xmath212 .
nevertheless , the high scatter in the cl 0024 fp also seems to be present in ms 0451 . limiting our sample to those galaxies with @xmath213 km s@xmath16 as in paper iii
, the overall scatter is @xmath214 , expressed in terms of a spread in @xmath215 , compared to @xmath216 for cl 0024 .
a small portion of this large scatter in @xmath34 can be attributed to the higher redshift of ms 0451 , where an equivalent spread in galaxy formation ages translates into a larger fp scatter at @xmath3 than at @xmath217 . in figure [ fp ] , this effect is most easily seen by noting the larger separation between the @xmath210 and @xmath211 model lines in the ms 0451 panel , compared to the cl 0024 panel . if ms 0451 early - types were to passively evolve between @xmath3 and @xmath217 , the single stellar population ( ssp ) models of @xcite predict a decrease in the fp scatter of only up to @xmath218 .
this still leaves an unexpectedly high spread in mass to light ratios for ms 0451 early - types . in figure
[ deltaml ] , we plot each galaxy s residual from the fp as a function of its dynamical mass , calculated according to @xmath219 ( see paper iii ) . in both clusters ,
we observe a clear ` downsizing ' trend that has been described by several authors ( e.g. , * ? ? ?
* ) , the tendency for less massive e+s0s to exhibit lower mass to light ratios , implying younger stellar populations . to judge the effects of ` downsizing ' ,
we fit a straight line to the fp residuals as a function of mass , limiting the fit to masses above @xmath220m@xmath221 , where selection effects are minimal . in ms 0451 , subtracting off this extra ` tilt ' to the fp greatly tightens the observed scatter , to @xmath222 , a level of scatter consistent with that of other well - virialized clusters at lower redshift ( e.g. , * ? ? ?
@xmath223 versus dynamical mass , m , for ms 0451 ( top ) and cl 0024 ( bottom ) .
symbols denote different morphologies as indicated on the legend .
we are insensitive to galaxies in the area to the lower left of each solid black line , due to the luminosity limit of our sample . above , @xmath224m@xmath221 , we are nearly unbiased .
lower split panels show the residuals from fitting and subtracting a straight line to galaxies at m@xmath225m@xmath221 .
@xmath226 for s0s in cl 0024 do not form a tight sequence in mass , while their counterparts in ms 0451 and the es in both clusters do form a tight sequence .
many cl 0024 galaxies also occupy unphysical locations at @xmath227 . , title="fig:",width=312 ] @xmath223 versus dynamical mass , m , for ms 0451 ( top ) and cl 0024 ( bottom ) .
symbols denote different morphologies as indicated on the legend .
we are insensitive to galaxies in the area to the lower left of each solid black line , due to the luminosity limit of our sample . above , @xmath224m@xmath221 , we are nearly unbiased .
lower split panels show the residuals from fitting and subtracting a straight line to galaxies at m@xmath225m@xmath221 .
@xmath226 for s0s in cl 0024 do not form a tight sequence in mass , while their counterparts in ms 0451 and the es in both clusters do form a tight sequence .
many cl 0024 galaxies also occupy unphysical locations at @xmath227 . , title="fig:",width=312 ] performing the same exercise on cl 0024 , however , results in no improvement in the fp scatter .
close inspection of the morphologies of galaxies in figure [ deltaml ] uncovers the reason .
while ellipticals in cl 0024 display a tight sequence in mass , s0s , especially at intermediate masses of @xmath228m@xmath221 , exhibit a remarkable scatter in their fp residuals .
these residuals do not appear to correlate with any typical measures of star formation rate or mass to light ratio , including @xmath104 , h@xmath103 , or even broadband optical colors .
it seems , then , that the previously described fp scatter in the core of cl 0024 can be ascribed almost entirely to s0s whose properties do not obey any simple trends with mass or star formation .
furthermore , a significant number of cl 0024 galaxies exhibit fp residuals @xmath227 , an unphysical quantity that either indicates stellar populations older than those of coma cluster early types ( impossible due to the much younger age of the universe at @xmath217 and the extreme mass to light ratios implied ) , or else indicates the breakdown of the fundamental plane for these objects .
in contrast , all of the scatter in ms 0451 early types is in the expected direction , in the sense that some galaxies have younger stellar populations with @xmath229 , while the most massive ellipticals and s0s lie neatly on the local fp . what could be generating these abnormal galaxies in cl 0024 but not in ms 0451 ?
enhanced levels of galaxy - galaxy interaction due to the complex substructure in cl 0024 is the most likely explanation .
the difference between a slow conversion of spirals to s0s ( in cl 0024 ) and a rapid truncation of star formation ( in ms 0451 ) could also affect the positions of s0s on the fp .
future simulations to test this possibility would be welcome .
we note that the actual ` young s0s ' are only sparsely represented in figure [ deltaml ] , due to the smaller sample of galaxies that can be plotted on the fp , and are not driving the observed difference by themselves .
what , then , are the implications for the physical processes driving galaxy evolution in the cores of both clusters ? in short , we can conclude that the complex recent assembly history of cl 0024 _ does not _ enhance the ability of the icm to transform galaxies via shocks , but it _ does _ appear to enhance the observed kinematic disturbance of galaxies within the cluster core .
some mix of galaxy galaxy interaction and tidal interaction therefore must dominate in the core of cl 0024 , driving the steady decline in star formation rate within its spiral population . only for
galaxies that reach the inner @xmath201kpc will ram pressure stripping become important .
in contrast , the quenching of star formation in ms 0451 is almost certainly driven by ram pressure from the icm , probably coupled with galaxy and tidal interactions that serve to erase spiral structure and complete the morphological conversion . in ms 0451
, the effects of harassment / tidal interaction must complete the morphological conversion more rapidly than in cl 0024 , where the stellar populations of passive spirals indicate that they are long - lived ( @xmath230 gyr ) . at the same time
, the effects of this conversion must somehow decay rapidly enough in ms 0451 s0s to be unobserved in the residuals from the fp .
future investigations into the effects of harassment on the s0 fundamental plane could help shed light on this issue .
in this paper , we have presented new results from our comprehensive comparative survey of two massive , intermediate redshift galaxy clusters , cl @xmath0 ( @xmath1 ) and ms @xmath2 ( @xmath3 ) .
we have identified and studied several key classes of transition objects in the clusters : the passive spirals and the young s0s . through uv imaging and measurements of spectral line indices ,
we have concluded that some passive spirals have experienced a decline in star formation over a @xmath149 gyr timescale , mostly in cl 0024 , while others , mostly in ms 0451 , have experienced a more rapid truncation in star formation .
for the first time , we have been able to conclusively identify spiral galaxies in the process of transforming into s0 galaxies , by directly linking the passive spirals in each cluster with their descendant s0s .
having established that the transformation from spiral to s0 galaxies is taking place in each cluster , we have leveraged the differences between clusters in the timescales and spatial location of the conversion process , in order to evaluate the relative importance of several proposed physical mechanisms that could be responsible for the transformation .
combined with other diagnostics that are sensitive to either icm - driven galaxy evolution or galaxy - galaxy interactions including the residuals from the fundamental plane and the properties of ` signpost ' compact emission line galaxies we paint a tentative but remarkably self - consistent picture of galaxy evolution in clusters .
we find that spiral galaxies within infalling groups have already begun a slow process of conversion into s0s primarily via gentle galaxy - galaxy interactions that act to quench star formation , perhaps aided by interaction with the intra - group gas .
the fates of spirals upon reaching the core of the cluster depend heavily on the cluster icm , with rapid conversion of all remaining spirals into s0s via ram - pressure stripping in clusters where the pressure of the icm is @xmath231 of that needed to strip a canonical milky way - like spiral . in the presence of a less - dense icm
, the conversion continues at a slower pace , with galaxy - galaxy interactions continuing to play a role , perhaps along with ` starvation ' or gentle stripping by the icm .
several authors have raised objections to a scenario where s0s are created via simple fading of spiral disks .
first , s0s are observed to have higher bulge to disk ratios than their supposed spiral progenitors ( e.g. , * ? ? ?
* ) , suggesting that if the spiral to s0 conversion takes place , significant redistribution of mass or significant new star formation is required @xcite .
second , other authors have noted that the local abundance of s0s is only weakly correlated with environment , such that processes like ram pressure stripping that act only in cluster cores can not be responsible for the entire buildup of s0s @xcite . we have here demonstrated ,
however , that simple stripping of gas does _ not _ build up the entire population of cluster s0s . rather , a combination of gas effects and galaxy interactions are responsible , in ratios that vary widely and depend on both the cluster dynamical state and the location within the cluster or its outskirts .
the buildup of bulges through new star formation @xcite that is expected from harassment - like encounters could potentially balance out the expected fading of disks after ram pressure stripping has run its course .
it may still be necessary to invoke significant obscured star formation to bridge the observed gap between the bulges of spirals and s0s @xcite ; the higher detection rate of such systems in cl 0024 compared to ms 0451 @xcite may indicate that such obscured starbursts are part of the pre - processing that occurs before galaxies encounter a dense icm .
furthermore , the wide distribution of passive spirals across both clusters , and their association with groups in the cluster outskirts , indicates that the resulting s0s will likely be spread across a wide range of environments .
the typical densities of infalling groups at @xmath4 are likely realized even in isolated groups by @xmath130 @xcite , and so it seems that , though a variety of mechanisms are responsible , passive spirals could be the progenitors of most local s0s .
in @xcite , a rough assessment of the frequency of passive spirals and their expected lifetimes in cl 0024 suggested that passive spirals could account for the entire buildup of s0s between @xmath217 and @xmath130 . in ms 0451
, we observe a similar number of spirals currently in the passive phase , but we have shown that the timescale for conversion is much shorter than in cl 0024 .
this suggests that all existing spirals should be converted to s0 in ms 0451 even _ before _
@xmath130 , and the rate of any future buildup should be limited by the infall rate of new galaxies . as ms 0451
is a remarkably well - evolved system at @xmath3 , comparable in mass to virgo or coma ( @xmath232 and @xmath233 , respectively , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , this is perhaps not surprising .
it suggests , however , that there may exist a generic tipping point in the evolution of a massive cluster : beyond the threshold in ram pressure strength that we have identified here ( which is reached through some combination of cluster mass and icm density ) , spirals are transformed so rapidly upon infall that they will be essentially absent . by @xmath130 , many clusters have likely reached this threshold , as evidenced by their low spiral fractions @xcite .
this sort of dichotomy between well - evolved and still - assembling clusters could , for example , explain why some clusters , like coma , reveal few signs of evolution in their massive galaxies @xcite , while others , like virgo , have a rich population of passive spirals and galaxies with other signs of recent evolution ( e.g. , * ? ? ?
regardless , the results presented here indicate that the abundant cluster spirals found at intermediate redshift , _ do _ , in fact , transform into the equally abundant s0 population seen today . the transformation process at @xmath234
, as we have seen , is beginning to operate even in groups at the cluster outskirts . as a result ,
the conversion of passive spirals to s0s can account for the evolution of the morphology density relation at both the cluster and group scale from intermediate redshift to today .
we thank the referee , a. boselli , for insightful comments , especially on the various physical processes and their timescales , which were helpful in revising the discussion of these issues .
smm would like to thank t. heckman , g. kauffmann , s. yi , & members of the _ galex _ science team for valuable discussions .
faint object spectroscopy at keck observatory is made possible with lris and deimos thanks to the dedicated efforts of j. cohen , p. amico , s. faber and g. wirth .
we acknowledge use of the gauss - hermite pixel fitting software developed by r. p. van der marel . the analysis pipeline used to reduce the deimos data was developed at uc berkeley with support from nsf grant ast-0071048 .
tt acknowledges support from the nsf through career award nsf-0642621 , and from the sloan foundation through a sloan research fellowship .
rse acknowledges financial support from nsf grant ast-0307859 and stsci grants hst - go-08559.01-a and hst - go-09836.01-a .
irs and gps acknowledge support from the royal society .
abraham , r. g. , et al .
1996 , apj , 471 , 694 balogh , m. l. , morris , s. l. , yee , h. k. c. , carlberg , r.g . , & ellingson , e. 1999 , apj , 527 , 54 bekki k. , couch , w.j . , & shioya , y. 2002 , apj , 577 , 651 bertin , e. & arnouts , s. 1996 , a&as , 117 , 393 biviano , a. girardi , m. , giuricin , g. , mardirossian , f. & mezzetti , m. 1993 , apj , 376 , 458 blanton , m. r. , et al .
2003 , aj , 125 , 2348 boselli , a. & gavazzi , g. 2006 , pasp , 118 , 517 boselli , a. , boissier , s. , cortese , s. , gil de paz , a. , seibert , m. , madore , b. f. , buat , v. & martin , d. c. 2006 , apj , 651 , 811 bruzual , g. , & charlot , s. 2003 , mnras , 344 , 1000 burstein , d. , ho , l. c. , huchra , j. p. , & macri , l. m. 2005 , apj , 621 , 246 .
christlein , d. & zabludoff , a. i. 2004 , apj , 616 , 192 chung , a. , van gorkom , j. h. , kenney , j. d. p. , & vollmer , b. 2007 , apj , 659 , l115 cooper , m. c. , et al .
2007 , mnras , 376 , 1445 cortese , l. , et al .
2007 , mnras , 376 , 157 couch , w. j , barger , a. j. , smail , i. , ellis , r. s. & sharples , r. m. 1998 , apj , 497 , 188 cowie , l. l. & songaila , a. 1977 , nature , 266 , 501 czoske , o. , kneib , j .-
p . , soucail , g. , bridges , t. j. , mellier , y. , & cuillandre , j .- c .
2001 , a&a , 372 , 391 czoske , o. , moore , b. , kneib , j .-
, & soucail , g. 2002 , a&a , 386 , 31 davis , m. , et al .
2003 , spie , 4834 , 161 de filippis , e. , sereno , m. , bautz , m. , & longo , g. 2005 , apj , 625 , 108 de lucia , g. , springel , v. , white , s. d. , croton , d. , & kauffmann , g. 2006 , mnras , 366 , 499 donahue , m. , gaskin , j. a. , patel , s. k. , joy , m. , clowe , d. , & hughes , j. p. 2003
, apj , 598 , 190 dressler , a. 1980 , apj , 236 , 351 dressler , a. 2004 , in carnegie observatories centennial symposia , eds : j.s .
mulchaey , a. dressler , & a. oemler , ( cambridge university press : cambridge , uk ) , 206 dressler , a. & gunn , j. e. 1983 , apj , 270 , 7 dressler , a. & shectman , s. a. 1988 , aj , 95 , 985 dressler , a. , smail , i. , poggianti , b. , butcher , h. , couch , w. j. , ellis , r. s. , & oemler , a. , jr .
1999 , apjs , 122 , 51 dressler , a. , et al .
1997 , apj , 490 , 577 ellingson , e. , yee , h. k. c. , abraham , r. g. , morris , s. l. , & carlberg , r. g. 1998 , apjs , 116 , 247 fouqu , p. solanes , j. m. , sanchis , t. & balkowski , c. 2001 , a&a , 375 , 770 fujita , y. 1998 , apj , 509 , 587 fujita , y. 2004 , pasj , 56 , 29 fujita , y. & nagashima , m. 1999 , apj , 516 , 619 gavazzi , g. , boselli , a. , cortese , l. , arosio , i. , gallazzi , a. , pedotti , p. , & carrasco , l. 2006 , a&a , 446 , 839 gavazzi , g. , zaccardo , a. , sanvito , g. , boselli , a. , & bonfanti , c. 2004 , a&a , 417 , 499 geach , j. e. , et al .
2006 , apj , 649 , 661 geller , m. j. , diaferio , a. , & kurtz , m. j. 1999 , apj , 517 , l23 gunn , j. e. & gott , j. r. 1972 , apj , 176 , 1 hester , j. a. 2006 , apj , 647 , 910 jee , m. j. , et al .
2007 , apj , in press jeltema , t. e. , mulchaey , j. s. , lubin , l. m. , & fassnacht , c. d. 2007 , apj , 658 , 865 kauffmann , g. , et al .
2003 , mnras , 341 , 33 kelson , d. d. , illingworth , g. d. , van dokkum , p. g. , & franx , m. 2000 , apj , 531 , 184 kneib , j - p . ,
2003 , , 598 , 804 ( paper ii ) kodama , t. & smail , i. 2001 , mnras , 326 , 637 koo , d.,guzmn , r. , faber , s. m. , illingworth , g. d. , bershady , m. a. , kron , r. g. , & takamiya , m. 1995 , apj , 440 , 49 larson , r. b. , tinsley , b. m. , & caldwell , c. n. 1980 , apj , 237 , 692 lucey , j.r . , guzmn , r . ,
carter , d. , & terlevich , r. j. 1991 , mnras , 253 , 584 martin , d. c. , et al .
2005 , apj , 619 , l1 mihos , c. 2004 , in iau symposium no .
217 , recycling interstellar and intergalactic matter , ed .
duc , j braine , e brinks , ( san francisco : asp ) , 390 moore , b. , lake , g. , quinn , t. , & stadel , j. 1999 , mnras , 304 , 465 moran , s. m. , ellis , r. s. , treu , t. , salim , s. , rich , r. m. , smith , g. p. & kneib , j - p .
2006 , apj , 641 , l97 moran , s. m. , ellis , r. s. , treu , t. , smail , i. , dressler , a. coil , a. l. & smith , g. p .. 2005 , apj , 634 , 977 ( paper iii ) moran , s. m. , miller , n. , treu , t. , ellis , r. s. & smith , g. p. 2007 ,
apj , 659 , 1138 mulchaey , j. 2000 , ara&a , 38 , 289 noeske , k. g. , koo , d. c. , phillips , a. c. , willmer , c. n. a. , melbourne , j. , de paz , a. g. , & papaderos , p. 2006 , apj 640 , l143 nulsen , p. e. j. 1982 , mnras , 198 , 1007 okamoto , t. & nagashima , m. 2003 , apj , 587 , 500 ota , n. , pointecouteau , e. , hattori , m. , & mitsuda , k. 2004 , apj , 601 , 120 peng , c. y. , ho , l. c. , impey , c. d. , & rix , h. 2002 , aj , 124 , 266 poggianti , b. m. & barbaro , g. , 1997 , a&a , 325 , 1025 poggianti , b. m. & barbaro , g. , 1996 , a&a , 314 , 379 poggianti , b. m. , bridges , t. j. , komiyama , y. , yagi , m. , carter , d. , mobasher , b. , okamura , s. , & kashikawa , n. 2004 , apj , 601 , 197 poggianti , b. m. , smail , i. , dressler , a. , couch w. j. , barger , a. j. , butcher , h. , ellis , r. s. , & oemler , a. , jr . 1999 ,
apj , 518 , 576 poggianti , b. m. , et al .
2001 , apj , 563 , 118 poggianti , b. m. , et al .
2006 , apj , 642 , 188 postman , m. , et al .
2005 , apj , 623 , 721 quilis , v. , moore , b. , & bower , r. 2000 , science , 288 , 1617 rawat , a. , kembhavi , a. k. , hammer , f. , flores , h. , & barway , s. 2007 , a&a , in press roettiger , k. , burns , j. , & loken , c. 1996 , , 473 , 651 schlegel , d.j . ,
finkbeiner , d. p. , & davis , m. 1998 , apj , 500 , 525 smith , g. p. , treu , t. , ellis , r. s. , moran , s. m. , & dressler , a. 2005a , apj , 620 , 78 smith , g. p. , et al .
2005b , mnras , 359 , 417 treu , t. , ellis , r. s. , kneib , j - p . , dressler , a. , smail , i. , czoske , o. , oemler , a. , & natarajan , p. 2003 ,
apj , 591 , 53 ( paper i ) treu , t. , et al .
2005 , apj , 633 , 174 van den bergh , s. 1976 , apj , 206 , 883 van der marel , r. 1994 mnras , 270 , 271 wilson , j. c. , et al .
2003 , proc .
spie , 4841 , 451 worthey , g. , faber , s. m. , gonzalez , j. j. , & burstein , d. 1994 , apjs , 94 , 687 zabludoff , a. & mulchaey , j. 1998 , apj , 496 , 39 zhang , y. -y . ,
bhringer , h. , mellier , y.,soucail , g. , & forman , w. 2005 , a&a , 429 , 85 | we present new results from our comprehensive comparative survey of two massive , intermediate redshift galaxy clusters , cl @xmath0 ( @xmath1 ) and ms @xmath2 ( @xmath3 ) . combining _
hst _ optical and _ galex _ uv imaging with keck spectroscopy of member galaxies , we identify and study several key classes of ` transition objects ' whose stellar populations or dynamical states indicate a recent or ongoing change in morphology and star formation rate . for the first time , we have been able to conclusively identify spiral galaxies in the process of transforming into s0 galaxies .
this has been accomplished by locating both spirals whose star formation is being quenched as well as their eventual successors , the recently created s0s .
differences between the two clusters in both the timescales and spatial location of this conversion process allow us to evaluate the relative importance of several proposed physical mechanisms that could be responsible for the transformation . combined with other diagnostics that are sensitive to either icm - driven galaxy evolution or galaxy - galaxy interactions including the residuals from the fundamental plane and the properties of ` signpost ' compact emission line galaxies we describe a self - consistent picture of galaxy evolution in clusters .
we find that spiral galaxies within infalling groups have already begun a slow process of conversion into s0s primarily via gentle galaxy - galaxy interactions that act to quench star formation .
the fates of spirals upon reaching the core of the cluster depend heavily on the cluster icm , with rapid conversion of all remaining spirals into s0s via ram - pressure stripping in clusters where the icm is dense . in the presence of a less - dense icm
, the conversion continues at a slower pace , with galaxy - galaxy interactions continuing to play a role along with ` starvation ' by the icm .
we conclude that the buildup of the local s0 population through the transformation of spiral galaxies is a heterogeneous process that nevertheless proceeds robustly across a variety of different environments from cluster outskirts to cores . |
in the last few years there has been a resurgence of interest in the spin structure of the nucleon .
this was largely initiated by the measurements of the polarized structure function @xmath1 by the slac - yale @xcite and emc @xcite collaborations . in the ` naive ' parton model ,
@xmath2 can , like the unpolarized structure function @xmath3 , be expressed in terms of the probability distributions for finding quarks with spin parallel or antiparallel to the longitudinally polarized parent proton : @xmath4 \\ g_1(x , q^2 ) & = & \half \sum_q\ ; e_q^2\ ; [ \delta q(x ) + \delta\bar q(x ) ] \ ; , \label{naiveg1}\end{aligned}\ ] ] where @xmath5 the renewed interest in the subject was triggered by the first precision measurement of the integral of @xmath1 by the emc collaboration @xcite , @xmath6 which was significantly lower than the ` ellis - jaffe sum rule ' value of 0.18 @xcite .
this latter prediction is obtained by assuming that the net contribution of strange quarks to the proton spin is negligible . in the context of this model , @xmath7 and @xmath8 ,
the net spin carried by up and down quarks respectively , can be obtained from the @xmath9-decay rates of the octet hyperons : @xmath10 attributing the difference between the ellis - jaffe prediction and the slac - emc measurement to a non - zero strange sea polarization leads to the rather unusual result that the net spin carried by the quarks , @xmath11\ ; , \ ] ] is very small .
this so - called ` spin crisis ' precipitated an enormous amount of theoretical discussion a clear and comprehensive review can be found in ref .
@xcite .
one of the most compelling explanations for the violation of the ellis - jaffe sum rule is that a substantial amount of the proton s spin is carried by _
gluons_. as first pointed out in @xcite , the polarized gluon contributes to @xmath2 via the @xmath12-triangle anomaly .
thus in perturbative qcd , the naive parton model result of eq .
( [ naiveg1 ] ) is replaced by @xmath13\ ; \nonumber \\ & & \times \left\{\delta(1-y ) + { \alpha_s(q^2 ) \over 2\pi } \delta c_q(y ) + \ldots \right\ } \nonumber \\ & & + \ ; \frac{1}{9 } \int_x^1 \ ; { dy\over y}\ ; \delta g(x / y , q^2 ) \ ; \left\ { n_f\ ; { \alpha_s(q^2 ) \over
2\pi}\;\delta c_g(y ) + \ldots \right\ } \label{qcdg1}\end{aligned}\ ] ] where @xmath14 is the polarized gluon distribution . now according to the altarelli - parisi evolution equations @xcite , @xmath15 which implies that the gluon contribution to @xmath2 is formally of the same order as the quark contribution .
the splitting of the structure function into two conserved quantities associated with quark and gluon contributions to the net spin is natural in the sense that it allows the former to be identified ( up to finite mass corrections ) with the su(6 ) constituent " quarks . retaining only the leading terms in eq .
( [ qcdg1 ] ) , assuming three quark flavours , and using a factorization scheme where @xmath16 ( see section 2 ) , gives @xmath17 - \frac{1}{3 } { \alpha_s(q^2 ) \over 2\pi}\delta g(x , q^2 ) , \label{eq : g1bare}\\ \gamma_1^p & = & \int_0 ^ 1 dx\ ; g_1^p = \frac{2}{9 } \eta_u + \frac{1}{18 } \eta_d + \frac{1}{18 } \eta_s - \frac{1}{3 } { \alpha_s(q^2 ) \over 2\pi}\eta_g .
\label{eq : gamma}\end{aligned}\ ] ] it is now straightforward to calculate the amount of net gluon spin needed to explain the slac - emc sum - rule data . for example , with @xmath18 and @xmath7 , @xmath8 again obtained from hyperon decays , we find @xmath19 at @xmath20 @xcite .
although this might be considered a surprisingly large number there is no violation of spin conservation , since there is also a contribution to the proton s spin from the orbital angular momentum of the partons : @xmath21 according to eq .
( [ glueevo ] ) , the second and third terms on the right - hand side increase as @xmath22 in such a way that their sum is constant . in ref .
@xcite hereafter referred to as as a simple leading - order qcd model , with no polarized strange sea at @xmath23 ,
was used to extract polarized quark and gluon distributions from the slac and emc data on @xmath24 .
although the integrated parton distributions were reasonably well determined , the shapes of the distributions , especially those of the @xmath25-quark and the gluon , were poorly constrained . as a result ,
counting - rule and regge arguments for the large and small @xmath0 behaviour had to be invoked .
the main purposes of the exercise were ( i ) to show that a consistent set of distributions could be derived , and ( ii ) to present parton distributions which could be used for polarized lepton - hadron and hadron - hadron collision phenomenology .
since the as analysis was performed , there has been a dramatic increase in the amount of polarized structure function data available . using a polarized @xmath26he target ,
the e142 collaboration at slac has measured the neutron structure function @xmath27 @xcite .
the smc collaboration at cern first measured the deuteron structure function @xmath28 @xcite and more recently has measured @xmath1 @xcite , improving on the earlier slac - emc measurements .
these new results allow a test of another important sum rule due to bjorken @xcite @xmath29 .\ ] ] this sum rule follows from su(2 ) isospin invariance and is a rigorous prediction of qcd .
any disagreement between the measured and predicted values would invalidate the qcd - improved parton model approach on which the present study is based .
the most recent experimental measurement using all available data @xcite gives @xmath30 to be compared to the third - order qcd prediction @xcite @xmath31 the agreement is acceptable , especially considering that higher - twist contributions might still give a small contribution at this @xmath32 . in our parton distribution analysis , therefore , we shall use isospin symmetry to relate the distributions in the proton to those in the neutron .
it is interesting that the new measurement of the integrated proton structure function @xcite , again using all available data , @xmath33 is larger than the old slac - emc result .
this means that our polarized gluon distribution ( which accounts for the difference between the measured value and the ellis - jaffe prediction ) will now be somewhat smaller than previous estimates : in fact we shall show below that and @xmath34 values ] @xmath35 , compared to the as result @xmath19 .
the goal of the present study is to use the latest available data to update and improve the as parton distribution analysis .
following as , we adopt the point of view that there should be no significant polarization of the strange quark sea .
although the sum - rule measurements give an indication of the size of @xmath36 , there is a large arbitrariness in the _ shape _ of the gluon distribution .
we will explore several qualitatively different shapes allowed by the data .
the outcome of the analysis will again be a set of distributions which will provide benchmarks for future deep inelastic scattering experiments and which will be useful for phenomenological analyses of other types of polarized scattering experiment .
as we have already indicated , the first moments of our parton distributions are constrained by hyperon decay and sum - rule data . in the remainder of this section
, we describe the input data which we use , and the corresponding first moments which obtain .
the structure function @xmath1 occurs in the antisymmetric part of the hadronic tensor in deep inelastic scattering , which can be expressed as the matrix element of the axial vector current between two proton states @xcite .
these matrix elements can be further decomposed into the su(3)@xmath37 singlet and octet pieces . in order to compare the sum rules with experiment and to perform a @xmath32 evolution of the polarized parton densities , we rewrite eq .
( [ eq : g1bare ] ) as @xmath38 with similar expressions for the neutron and deuteron : @xmath39 here we have defined octet and singlet quark distributions : @xmath40 the first moments of ( [ eq : g1p])-([eq : g1d ] ) are : @xmath41 the axial - vector current matrix elements , @xmath42 and @xmath43 , are related to the @xmath44 and @xmath34 couplings : @xmath45 assuming @xmath46 gives @xmath47 , leading to separate predictions for the quark contributions to the @xmath48 s @xcite . measuring @xmath44 and @xmath34 in hyperon @xmath9-decays gives their values at a momentum transfer scale of @xmath49 .
however @xmath44 and @xmath34 , being related to the octet axial vector current matrix elements , are @xmath32 dependent . to fix @xmath7 and @xmath8 for our fits
, we incorporate these perturbative corrections by replacing @xmath44 and @xmath34 by @xmath50 and @xmath51 , which are obtained by comparing the parton model expressions for @xmath42 and @xmath52 with their values corrected to first order in perturbative qcd .
demanding that @xmath50 and @xmath51 reproduce the corrected values of @xmath42 and @xmath52 , one has , following @xcite , @xmath53
\nonumber\\ \tilde{i}_8 + \tilde{i}_0 = \frac{5}{36}(3\tilde{f}-\tilde{d } ) & = & \frac{1}{36}(3f - d)\left[(1-\frac{\alpha_s}{\pi } ) + 4(1-\frac{\alpha_s}{3\pi})\right]\ ; . \label{eq : oct}\end{aligned}\ ] ] this leads to @xmath54 taking @xmath55 and @xmath56 ( see below ) , we calculate @xmath50 and @xmath51 at the starting point of our perturbative evolution @xmath57 ( @xmath58 ) , taking the experimental values for @xmath44 and @xmath34 from ref .
@xcite : @xmath59 which fixes the first moments of the polarized @xmath60- and @xmath25-quark distributions : @xmath61 finally , comparing the parton model expression for @xmath62 ( [ eq : gamma ] ) with the experimental value ( [ eq : smc ] ) gives @xmath63 the above result for the moments and the sum rules are summarized in table ( [ tab : sums ] ) . ' '' '' & @xmath7 & @xmath64 + ' '' '' & @xmath8 & @xmath65 + ' '' '' partons & @xmath66 & @xmath67 + ' '' '' & @xmath68 & @xmath69 + ' '' '' & @xmath36 & @xmath70 + ' '' '' & @xmath62 & @xmath71 + ' '' '' @xmath2 & @xmath72 & @xmath73 + ' '' '' @xmath74 & @xmath75 & @xmath76 + ' '' '' & @xmath77 & @xmath78 + the remainder of the paper is organized as follows . in the next section we discuss the different possible definitions of the polarized gluon distribution . in section 3
we describe our fit to the polarized structure function data , in section 4 we discuss the implications of our fits for the polarized sea quark distributions at higher @xmath32 , and in section 5 we summarize our results and present our conclusions .
the form of @xmath79 in eq .
( [ qcdg1 ] ) has been a matter of some dispute during the past few years @xcite .
kodaira @xcite first pointed out that the anomalous dimension of the singlet axial vector current was non - zero . according to ref .
@xcite , the first moment of @xmath79 is @xmath80 based on these ideas , altarelli and ross @xcite evaluated the anomalous gluonic contribution to @xmath2 .
a crucial point in this calculation is the regularization of collinear singularities ( from @xmath81 ) in the photon - gluon fusion process .
depending on the regularization procedure used , one obtains different expressions for @xmath79 . in ref .
@xcite the physical quark mass was used as a regulator , giving @xmath82 the scheme dependence of @xmath79 was further discussed by bodwin and qiu @xcite .
they obtained the surprising result that , in dimensional regularization and for finite quark masses , the first moment of @xmath79 vanishes .
only the ( unphysical ) approach of assuming a small but finite gluon mass reproduced the result of @xcite
. a careful reanalysis @xcite of the polarized photon - gluon fusion process showed that in fact this disagreement between different regularization schemes comes from the soft non - perturbative region .
contributions from this region should be absorbed into the light quark density , and should therefore not contribute explicitly to @xmath79 .
this leaves us with two different possible @xmath79 , both with first moment equal to @xmath83 .
the first is obtained by introducing a non - zero gluon mass , while the second derives from either introducing finite quark masses or from applying dimensional regularization , always subtracting the soft contribution : @xmath84 the gluon contribution to @xmath2 , eq .
( [ qcdg1 ] ) , involves the convolution of @xmath79 with @xmath85 . rather than tying ourselves to a particular theoretical scheme ,
we adopt the as procedure @xcite and define a @xmath2-scheme " gluon with @xmath86 .
since only the first moments of the polarized quark and gluon densities are precisely defined , we are free to fix the higher moments in this way to obtain the simplest form for the gluon term in eq .
( [ qcdg1 ] ) . of course , in a consistent next - to - leading order analysis , any change in the definition of @xmath87 is compensated by a corresponding change in the @xmath88 correction to the quark contribution , eq .
( [ qcdg1 ] ) , in such a way that the physical quantity @xmath2 is scheme independent .
a different choice of @xmath87 will then yield a different gluon _ and _ a different set of quark distributions .
we are aware of the fact that our mixture of leading - order quark and next - to - leading order gluon contributions in @xmath2 is ambiguous .
nevertheless we consider the incorporation of the anomalous gluon , although formally of order @xmath89 , to be motivated by the scaling behaviour of @xmath90 and by the magnitude of this contribution to @xmath2 .
when complete next - to - leading order corrections become available it will of course be necessary to perform calculations of different processes within a consistent scheme . within our leading - order approach ,
it is straightforward to relate gluon distributions defined according to different schemes .
suppose that one uses a different coefficient function , @xmath91 say , in the definition of @xmath2 .
then the corresponding polarized gluon distribution @xmath92 ( i.e. the distribution which reproduces the same @xmath2 at @xmath93 ) is related to ours by @xmath94 this equation can be inverted in mellin moment space . applying a mellin transformation to eq .
( [ baregl ] ) and defining @xmath95 we obtain @xmath96 for reference , we list in table [ tab : mellin ] the mellin moments of the coefficient functions of eqs .
( [ cg1 ] ) and ( [ cg2 ] ) . in the following section
, we will determine a polarized gluon distribution at @xmath93 from a fit to the structure function data .
the gluon defined in any other scheme can then be obtained by numerical integration of eq .
( [ invert ] ) . as an example
, figure [ fig : mellin ] shows our gluon set a ( see below ) together with the two distributions which correspond to the different coefficient functions defined in eqs .
( [ cg1 ] ) and ( [ cg2 ] ) . ' '' '' @xmath79 & @xmath97 & mellin transformation @xmath98 + ' '' ''
@xmath99 & @xmath83 & @xmath83 + ' '' '' ( as assumption ) & & + ' '' '' @xmath100 & @xmath83 & @xmath101 + ' '' '' ( @xmath102 ) & & + ' '' '' @xmath103 & @xmath83 & @xmath104 + ' '' '' ( dr , @xmath105 ) & & +
as a result of over twenty years of experiments , the _ unpolarized _ parton distributions in the proton are very precisely known , see for example ref .
some twenty independent parameters are needed to specify the various quark and gluon distributions at @xmath93 . in comparison ,
measurements of the polarized distributions are still rather imprecise .
it is therefore necessary to build in to the parametrizations a significant amount of theoretical prejudice .
as measurements improve in the future , these expectations will be tested and refinements can be made . at present , the only constraints on the distributions are ( i ) the specification of the first moments of the distributions , as described in the introduction and summarized in table [ tab : sums ] , and ( ii ) the requirements of _ positivity _ of the individual spin - parallel and spin - antiparallel distributions : @xmath106 therefore , the absolute value of a polarized parton disrtibution always has to be smaller than the corresponding unpolarized distribution .
we therefore require a consistent set of leading - order unpolarized distributions to provide the bounds those of ref .
@xcite are ideal for this purpose .
the starting distributions are , at @xmath107 , @xmath108 and the evolution is performed with @xmath109 .
the parameters @xmath110 and @xmath111 are fixed by the requirement that @xmath112 . for consistency ,
we choose the same @xmath113 and @xmath114 values as @xcite , and similar starting parametrizations .
only our choice of performing the whole evolution with three quark flavours differs from @xcite .
as explained in the introduction , we start at @xmath114 with valence - like forms for the up and down quarks and with an unpolarized sea : @xmath115 with normalization factors @xmath116 ( @xmath117 ) to ensure that @xmath118 : @xmath119 non - zero polarized su(3 ) symmetric sea quark distributions are generated dynamically for @xmath120 ( see below ) .
note that ( i ) the positivity constraints ( [ eq : pos ] ) demand that the @xmath121 parameter values are at least as big as the corresponding unpolarized parameters of eq .
( [ eq : unpol ] ) , and ( ii ) the convergence of the first moment integrals requires @xmath122 . even with these constraints , there are still too many free parameters in ( [ eq : form ] ) for the amount of data available .
we therefore impose additional theoretical constraints at large and small @xmath0 , based on dimensional counting and coherence / regge arguments respectively @xcite .
a good discussion of these can be found in ref .
@xcite . in the present context , they can be summarized as follows : 1 .
the assumption that the small @xmath0 behaviour of the quark distributions is controlled by regge behaviour requires @xmath123 .
coherence arguments at small @xmath0 give @xmath124 for the gluon distribution .
spectator quark counting rules for the @xmath125 behaviour suggest @xmath126 and require @xmath127 and @xmath128 to be the same as in the unpolarized case .
we have already mentioned that the most accurate data at present are for @xmath1 .
this means that the detailed shape of the @xmath25-quark and gluon distribution can not yet be determined .
thus for the former we impose @xmath129 , and for the latter we suggest three qualitatively different forms , defined by @xmath130 the reasoning behind these choices is as follows : @xmath131 ( the ` standard ' form ) will give a fit with @xmath132 at large @xmath0 , which corresponds to a fast - moving gluon being equally likely to have its spin parallel or antiparallel to the parent proton ; @xmath133 large and positive will lead to @xmath134 at large @xmath0 , i.e. a fast - moving gluon which is preferentially polarized in the direction of the proton s spin ; @xmath133 large and negative will give a large-@xmath0 gluon whose spin is anticorrelated with that of the proton .
for the latter two cases , we limit the absolute size of the parameter @xmath133 to ensure positivity of both @xmath135 and @xmath136 , and in each case we fit for the remaining parameter @xmath137 .
table [ tab : fixed ] lists the parameters which are fixed by the above theoretical considerations . in section [ sec : fitproc ] we describe how the remaining parameters @xmath138 , @xmath139 and @xmath140 are determined from the data . ' '' '' parameter & value & comments + ' '' '' @xmath127 & 3.64 & unpolarized @xmath127@xcite + ' '' '' @xmath128 & 4.64 & unpolarized @xmath128@xcite + ' '' '' @xmath141 & 1.000 & coherence arguments@xcite + ' '' '' @xmath7 & 0.848 & @xmath142 + ' '' ''
@xmath8 & @xmath1430.294 & @xmath144 + ' '' '' @xmath36 & 1.971 & from @xmath62 + having fixed the starting distributions at @xmath145 , the distributions at higher @xmath32 are obtained from the leading - order evolution equations @xcite : @xmath146 \nonumber \\
\frac{d\delta g(x , q^2)}{d \ln q^2 } & = & \frac{\alpha(q^2)}{2\pi } \int_x^1\frac{dy}{y}\left[\delta \sigma(y , q^2 ) \delta
p_{gq}\left(\frac{x}{y}\right)+\delta g(y , q^2)\delta p_{gg } \left(\frac{x}{y}\right)\right ] \nonumber \\
\frac{d\delta q_{val}(x , q^2)}{d \ln q^2 } & = & \frac{\alpha(q^2)}{2\pi } \int_x^1\frac{dy}{y}\left[\delta q_{val}(y , q^2 ) \delta p_{qq}\left(\frac{x}{y}\right)\right ] \label{eq : evo}\end{aligned}\ ] ] where the splitting functions are @xmath147 \nonumber\\ \delta p_{qg } & = & \frac{1}{2 } \left[z^2-(1-z)^2\right]\nonumber\\ \delta p_{gq } & = & \frac{4}{3 } \frac{1-(1-z)^2}{z}\nonumber\\ \delta p_{gg } & = & 3\left[(1+z^4)\left(\frac{1}{z}+ \frac{1}{(1+z)_{+}}\right)-\frac{(1-z)^3}{z}+\left ( \frac{11}{6}-\frac{n_f}{9}\right)\delta(1-z)\right ] \ ; .\end{aligned}\ ] ] with the first moments @xmath148 and @xmath149 both being equal zero it follows that @xmath150 are both independent of @xmath32 . with @xmath151 and @xmath152 ,
one finds , using the renormalization group equation , @xmath153 and so @xmath62 , @xmath72 and @xmath75 defined in eq .
( [ eq : moments ] ) are independent of @xmath32 in leading order . in principle , five sets of @xmath2 structure function data are available for a global fit : the proton data from slac - yale @xcite , emc @xcite and smc @xcite , the neutron data from slac - e142 @xcite , and the deuteron data from smc @xcite . our analysis , however , is based on a leading - twist approximation , and it is therefore necessary to impose a minimum @xmath32 cut . based on similar fits to unpolarized structure function data , see for example @xcite , and on recent theoretical estimates @xcite , we believe that @xmath154 is sufficient .
this means that the bulk of the slac data ( @xmath155 ) falls outside our fit region .
only the emc and smc proton data ( @xmath156 ) cover a broad range of @xmath0 for @xmath154 .
the results of the fit are shown in figs .
[ fig : proton ] and [ fig : deuteron ] .
it turns out that the errorbars on the deuteron data are so large that there is no statistically significant constraint .
the fit to @xmath1 , fig .
[ fig : proton ] , is equally good for the three different types of gluon distribution described in section [ sec : parform ] ( see table [ tab : fitted ] ) , and so only the curves corresponding to set a are shown .
each data point is fitted at its appropriate @xmath0 and @xmath32 value , but since data from different beam energies are shown on the same figure , the curves correspond to fixed @xmath157 . for the smc and emc data ,
these correspond to the small- , medium- and large-@xmath0 data respectively .
for the slac data , only the largest @xmath0 data points have @xmath158 .
the lower @xmath0 data points are not used in the fit but are displayed for completeness .
figure [ fig : neutron ] shows our _ predictions _ for the neutron structure function , together with the slac - e142 data @xcite .
only the largest @xmath0 data point ( with @xmath159 ) can be compared with the theoretical curves . at lower @xmath0
, we might expect that higher - twist contributions are important , although it is interesting that the @xmath160 curve still gives a fair description of the data .
the values of the fitted parameters and their associated errors are listed in table [ tab : fitted ] .
not surprisingly , the quark distributions are almost independent of the choice of gluon distribution and the quality of the fit is similar for sets a , b and c. note that the @xmath138 ( @xmath161 ) parameter is reasonably constrained to have a value @xmath162 , close to the corresponding unpolarized values , see ( [ eq : unpol ] ) . ' '' '' & gluon a & gluon b & gluon c + ' '' '' @xmath133 & 0.0 & 18.0 & @xmath1433.5 + ' '' ''
@xmath163 & 18.8/30 & 18.9/30 & 20.8/30 + ' '' ''
@xmath138 & [email protected] & [email protected] & [email protected] + ' '' '' @xmath139 & [email protected] & [email protected] & [email protected] + ' '' ''
@xmath140 & [email protected] & [email protected] & [email protected] + figures [ fig : partons1]-[fig : partons3 ] show the gluon , quark - singlet , @xmath60-valence and @xmath25-valence distributions at @xmath93 obtained from the fits . for comparison ,
we show also the unpolarized distributions of ref .
@xcite . a final comment concerns the shape of the gluon distribution at small @xmath0 . for all of our sets ,
the gluonic contribution to @xmath2 is smooth and negative at small @xmath0 .
if one uses a different gluon coefficient function , for example either of those in table [ tab : mellin ] , then the same smooth single - signed behaviour can only be reproduced by using an oscillating gluon , see fig .
[ fig : mellin ] .
conversely , a smooth gluon combined with either of these alternative coefficient functions would give a gluonic contribution to @xmath2 which changes sign at small @xmath0 .
in our model , a non - zero antiquark distribution @xmath165 is generated at higher @xmath32 by the evolution of the quark singlet , eq .
( [ eq : evo ] ) .
the resulting @xmath165 is sensitive to the initial shape of the gluon . a measurement of the polarized sea - quark distribution for @xmath166 ( recall that we generate an su(3 ) symmetric sea )
would therefore probe the gluon distribution .
there are several ways that this could be done .
perhaps the most promising in the short term is to identify specific flavours of hadrons in the final - state .
the extension of eq .
( [ naiveg1 ] ) to hadron inclusive structure functions for @xmath167 is @xmath168 \\
g_1^{p\to h}(x , z , q^2 ) & = & \half \sum_q\ ; e_q^2\ ; [ \delta q(x)d^{q\to h}(z ) + \delta\bar q(x ) d^{\bar q \to h}(z)]\ ; , \label{naivefrag}\end{aligned}\ ] ] where @xmath169 are fragmentation functions . as explained , for example , in ref .
@xcite , the identification of @xmath170 in the final state allows the sea - quark polarization to be probed directly .
our predictions for the polarized sea - quark distribution are shown in fig .
[ fig : sea ] .
independent of the choice of gluon , all the distributions have the common feature of being _ negative _ for small @xmath171 . at large @xmath0 , the fact that @xmath172 is less than zero ( see fig . [
fig : partons3 ] ) is reflected in the corresponding @xmath173 distribution .
a preliminary study of the final - state hadron charge asymmetry in polarized deep inelastic scattering by smc @xcite has given a first indication of the size and shape of the @xmath165 distribution .
although the experimental errors are still rather large , all the distributions of fig .
[ fig : sea ] are consistent with these data .
in this paper we have described an attempt to perform a global fit to polarized deep inelastic structure function data on proton targets in order to extract a consistent set of polarized parton distributions .
we have adopted the point of view that there is no significant polarization of the sea quarks at low @xmath32 , and therefore that the gluon carries a substantial fraction of the proton s spin .
we have constrained the net spin carried by the various parton flavours using deep inelastic sum - rule and hyperon decay measurements .
recognizing that the quality of the data is not yet high enough to fully constrain the shapes of the distributions , we have imposed reasonable theoretical constraints at large and small @xmath0 .
we have performed fits with three qualitatively different gluon distributions , chosen to reflect the extent to which fast - moving gluons ` remember ' the spin of the parent proton .
our distributions have several applications : * they provide a benchmark for future deep inelastic structure function measurements in particular , measurements using neutron and deuteron targets , * they can be compared with theoretical , non - perturbative model calculations of the spin structure of the proton , * they can be used to predict cross sections for the production of various flavours of hadrons in the final state @xmath174 and @xmath175 for probing the quark sea , and the production of heavy quarks ( in particular @xmath176 and open charm ) for probing the gluon distribution , and * they can be used as input for the calculation of hard scattering cross sections in polarized hadron - hadron collisions . in conclusion ,
polarized lepton - hadron scattering experiments are beginning to provide precision information on how the spin of the proton is shared among its parton constituents .
this will be invaluable in testing the predictions of quantum chromodynamics and in designing future experiments .
our distributions should be a useful tool in both these areas .
we are grateful to r. voss and w. wislicki for useful discussions concerning the data .
one of us ( tg ) would like to thank e. reya for drawing his attention to this subject and for numerous clarifying and motivating discussions .
this work was supported in part by studienstiftung des deutschen volkes , bonn , and by the eec programme `` human capital and mobility '' , network `` physics at high energy colliders '' , contract chrx - ct93 - 0537 ( dg 12 coma ) . | in the past year , polarized deep inelastic scattering experiments at cern and slac have obtained structure function measurements off proton , neutron and deuteron targets at a level of precision never before achieved .
the measurements can be used to test the bjorken and ellis - jaffe sum rules , and also to obtain information on the parton distributions in polarized nucleons .
we perform a global leading - order qcd fit to the proton deep inelastic data in order to extract the spin - dependent parton distributions . by using parametric forms which are consistent with theoretical expectations at large and small @xmath0 , we find that the quark distributions are now rather well constrained .
we assume that there is no significant intrinsic polarization of the strange quark sea .
the data are then consistent with a modest amount of the proton s spin carried by the gluon , although the shape of the gluon distribution is not well constrained , and several qualitatively different shapes are suggested .
the spin - dependent distributions we obtain can be used as input to phenomenological studies for future polarized hadron - hadron and lepton - hadron colliders .
dtp/94/38 + may 1994 + 1.cm * spin - dependent parton distributions from + polarized structure function data * 1.cm t. gehrmann .2 cm _ department of physics , university of durham + durham dh1 3le , england _
+ .4 cm and .4 cm w.j .
stirling .2 cm _ departments of physics and mathematical sciences , university of durham + durham dh1 3le , england _ |
corrosion of solids is an everyday phenomenon of evident practical importance @xcite . the recent development of theoretical tools for the study of disordered systems and fractals in the context of statistical mechanics @xcite has triggered an outburst of activity in this subject .
when an etching solution is put in contact with a disordered etchable solid , the solution corrodes the weak parts of the solid surface , while the hard , stronger , parts stay uncorroded . during this process
new regions of the solid , both hard and weak , are discovered and come into contact with the etching solution . as corrosion proceeds the etching power of the solution may diminish : indeed , if the etchant is consumed in the reaction , etching becomes more and more unlikely until , finally , the solution is so impoverished and the solid surface so hardened that the corrosion process is arrested . at that moment
all solid points in contact with the solution are too hard to be etched by the weakened etching solution .
one of the most interesting aspects of this type of phenomenon is that the final solid - liquid interface has , in general , a fractal geometry , at least up to a certain scale @xcite .
this is precisely the qualitative phenomenology observed in a nice experiment on pit corrosion of aluminum thin films @xcite .
recently , a simple dynamical model of etching , capturing the aforementioned phenomenology , has been proposed @xcite .
this model has been studied using both computational and analytical tools in @xcite , and from these studies strong evidence has been provided that the fractal properties of the solid surface , once the dynamics has stopped , are related to isotropic percolation . in principle , this is not an obvious result ; in fact , at first sight , one could think that the interface should be anisotropic as there is a preferential direction in which the solution advances by etching the solid .
the purpose of this paper is to provide further theoretical evidence that indeed the critical behavior of the model dynamics is related to isotropic percolation .
we also extend the previous relation to spatial dimensions larger than @xmath2 . to this aim ,
we shall first review ( section 2 ) two known percolation models that will be useful in the forthcoming discussion : ( i ) dynamical percolation , and ( ii ) gradient percolation ( gp ) . afterwards
( section 3 ) , we will define the dynamical etching model @xcite in a circular ( spherical ) geometry , and will derive a phenomenological field theory for it ( section 4 ) . from the analysis of this field theory the parallelism with percolation will be set up in a rather clear way , and this will provide further theoretical evidence on the connection between etching and percolation phenomena .
the approach presented in this paper will allow us to study the system _ self - organization _ from a field theoretical point of view , and to verify that , in certain limit , the system is self - driven to the neighborhood of a critical point without need of any parameter fine - tuning .
this is a new path to self - organized criticality @xcite as will be discussed in the last section .
in this section we review two different well - known percolation models that will be useful in the discussion of the etching processes under consideration .
dynamical percolation is a model proposed for the study of the propagation of epidemics in a population , and/or for the analysis of forest fires .
it is defined as follows @xcite .
let us consider a regular square lattice ; at each site there is a variable that can be in one of three possible states ( we borrow the language from epidemiology @xcite ) : ( i ) infected sites , ( ii ) healthy sites susceptible to be infected , and ( iii ) immune sites ( non susceptible to be re - infected ) . at time
@xmath3 a localized seed of infected sites is located at the center of an otherwise empty ( healthy ) lattice .
the dynamics proceeds in discrete time steps either by parallel or by sequential updating as follows : at each time - step every infected site can infect a ( healthy ) randomly chosen neighbor with probability @xmath4 or , alternatively , heal and become immune to reinfection with complementary probability @xmath5 .
any system state with no infected site is an _ absorbing configuration _ , i.e. , a configuration in which the system can get trapped and from which it can not escape @xcite .
it is clear that depending on the value of @xmath4 the epidemics generated by the initial infection seed will either spread in the lattice ( for large values of @xmath4 ) or die out ( for small values of @xmath4 ) . in all cases
, the epidemics will leave behind a cluster of ( healed ) immune sites , infinite or finite respectively for the two aforementioned cases .
separating the two previous phases , there is a critical value of @xmath4 , @xmath6 , at which the epidemics propagates marginally , leaving behind a fractal cluster of immunized sites .
it can be shown using field theoretical tools ( see below ) that this is a percolation cluster @xcite . in this way we have a dynamical model which at criticality reproduces the ( static ) properties of standard percolation .
needless to say , the dynamical properties of the dynamical percolation equation , do not correspond to any known property of static percolation .
the dynamical percolation model can be cast into the following langevin equation @xcite ( or equivalently into a field theory @xcite ) : @xmath7 where @xmath8 ( the `` mass '' in a field theoretical language ) and @xmath9 are constants , @xmath10 an activity field describing at a coarse grained level the density of infected sites , and @xmath11 a gaussian white noise .
note the multiplicative nature of the noise , because of which the state @xmath12 defines an absorbing state , i.e. @xmath13 . note also the presence of a non - markovian term , that constitutes the key difference between this equation and the reggeon field theory , characteristic of many other systems with absorbing states .
this non - markovian term stems from the existence of immunized sites , of which the system keeps indelible memory @xcite .
the field theoretical and renormalization group analysis of eq.([dyp ] ) can be found in the literature @xcite .
the critical dimension is @xmath14 , and the exponents , calculated in an epsilon expansion , coincide with the well known values for percolation calculated using other techniques @xcite .
apart from the static exponents , also a dynamical exponent @xmath15 can be derived from this analysis of dynamical percolation @xcite .
gradient percolation @xcite ( gp ) is defined in the following way : let us consider a bidimensional rectangular lattice of lateral sizes @xmath16 and @xmath17 respectively , as shown in fig .
[ fig1 ] . an occupation probability given by @xmath18
is assigned to sites in column @xmath19 ; this defines a transversal constant gradient , @xmath20 for the occupation probability .
then , at each lattice site @xmath21 a random number @xmath22 $ ] , extracted from an homogeneous distribution , and representing the site reluctance to be occupied , is assigned .
all sites with @xmath23 are declared occupied , while the remainings are empty . at the first column , @xmath24 ,
all sites are occupied , while there is zero occupancy at the last one , @xmath25 ( see fig .
[ fig1 ] ) . after identifying all sites
as occupied or empty , one detects two connected regions ( clusters ) : one ( leftmost ) with a majority of occupied sites possibly surrounding `` lakes '' of empty sites , and another one ( rightmost sea ) possibly surrounding islands .
separating these two regions there is an interface ( the frontier of the connected cluster of occupied sites ; it corresponds to the dark sites in figure 1 ) .
the average position of this interface can be shown to be at the square lattice site percolation threshold , @xmath26 @xcite .
in fact , gradient percolation has been used as a computational tool to obtain accurate values of percolation thresholds in different geometries by identifying the average position of the interface in sufficiently large lattices @xcite . in the case that we are considering ,
the fractal dimension of the interface , @xmath27 , can be identified as the hull fractal dimension of the critical percolating cluster in a two - dimensional lattice @xcite .
there is an upper cut - off up to which this fractal behavior is observed ; it is fixed by the width @xmath28 which , in its turn , is determined by @xmath17 , and therefore by @xmath29 . it can be shown using percolation theory that @xmath30 where @xmath31 @xcite .
in order to have a well defined percolation system , with negligible finite size effects , the limit @xmath32 has to be used . in this way , the length @xmath17 , determining the value of @xmath28 , is the parameter that controls the finite size effects ; the `` thermodynamic limit '' corresponds to @xmath33 and @xmath34 with both limits taken in the proper way @xcite .
one can also estimate the variation of @xmath4 from on the leftmost to the rightmost extremes of the wandering interface , @xmath35 : @xmath36 the identity @xmath37 imposes the following scaling relation among exponents : @xmath38 , and therefore , @xmath39 let us observe that gradient percolation can also be defined in a circular geometry , in which the gradient changes with the radial distance to the origin , and the cut - off is determined by the width of the roughly circular crown in which the interface is inscribed . summarizing , in this section , we have reviewed two well - known percolation models .
dynamical percolation is a model that , at its critical point , generates dynamically a percolation cluster . on the other hand ,
gradient percolation is a static model , in which an interface appears with the same hull fractal dimension of the percolation cluster , but with no intrinsic dynamics defined .
having introduced the previous two percolation models , we go ahead by reviewing the dynamical etching model ( dem ) at the focus of our study @xcite .
it is defined by the following ingredients ( see figure 2 ) .
\(i ) the random solid is mimicked by a two - dimensional square lattice of finite linear width @xmath16 and depth @xmath40 ( @xmath40 can be arbitrarily large , or even infinite ) .
periodic boundary conditions in the finite direction are imposed , leading to a cylindrical geometry .
\(ii ) a random quenched number @xmath41 $ ] ( extracted from a uniform distribution ) , assigned to each solid site @xmath42 , represents the site resistance to etching .
\(iii ) the etching solution occupies a fixed volume @xmath0 and is initially in contact with the solid through the bottom boundary , as depicted in fig .
[ fig2 ] , defining a solid - solution interface advancing on average in the upward direction .
the solution contains an initial number @xmath43 of dissolved etchant molecules .
its concentration at time @xmath44 is @xmath45 .
it is assumed that the etching power of the solution @xmath46 is proportional to @xmath47 . without loss of generality the proportionality constant
can be fixed to unity .
following @xcite , we assume that etchant particles diffuse infinitely fast in the solution ( at least much faster than the characteristic time scale of etching ) and , hence , @xmath46 is taken as spatially homogeneous , i.e. , the etching power does not depend on the spatial position in the solution . at each discrete time - step all solid sites located at the solid surface and satisfying @xmath48 are dissolved ( see fig .
[ fig2 ] ) , i.e. , they are removed from the solid , and a particle of etchant is consumed for each dissolved site , reducing in this manner the total etching power .
calling @xmath49 the number of dissolved solid sites ( or equivalently the number of consumed etchant particles ) at time - step @xmath44 , and @xmath50 the total number of etched solid sites up to time @xmath44 , one can write @xmath51 as @xmath52 , a site having endured the etching attack at time @xmath44 will also resist at any time @xmath53 @xcite . furthermore , as a consequence of the corrosion process at time @xmath44 , @xmath54 new solid sites , previously in the solid bulk , come into contact with the solution at time @xmath55 .
note that they are the sole candidates for corrosion at the next time - step .
finally , since the solution has the possibility to detach finite solid islands , the global solid surface is composed both by the surfaces of the detached islands , and by the set of solid sites separating the solution from the bulk .
this interface is called the _
corrosion front_. a more detailed description of the model phenomenology can be found in @xcite .
here we simply summarize the main features of the corrosion front at the arrest time @xmath56 .
they are well represented by gp with @xmath57 : ( i ) the corrosion front shows fractal features with @xmath58 up to a characteristic scale ( front thickness ) @xmath28 ; \(ii ) @xmath59 ; \(iii ) @xmath60 , with @xmath61 ( therefore , in the right thermodynamic limit @xmath62 ) .
let us introduce here a slight geometrical modification of the dem which makes more clear the connection to dynamical percolation . instead of considering a cylindrical geometry with the etchant solution invading the cylinder from the bottom ( as in fig .
2 ) , we consider a flat infinite lattice , in which the volume @xmath0 of the etching solution is poured at time @xmath3 at an arbitrarily chosen central site as schematically shown in fig .
the volume @xmath0 of the etching solution is constant .
observe that with this geometry the model has some clear analogies with dynamical percolation .
the main difference is that , in the spherical dem , the control parameter ( the corroding or infecting probability ) is not a constant but decreases in time as the etching process proceeds . as in cylindrical geometry , the dynamics can be roughly divided into two regimes @xcite : a _ smooth _ one when @xmath46 is much larger than @xmath26 , and a _ critical _ one when @xmath46 approaches @xmath26 . in the smooth regime ,
fluctuations around the average behavior are small while in the critical regime fluctuations dominate the dynamics @xcite .
indeed , at early time - steps , the etching power being sufficiently larger than @xmath26 , it is simple to show @xcite that the corrosion front is an approximate expanding circumference centered at the origin , and the number of new solid sites coming into contact with the solution at time @xmath44 satisfies the approximate relation @xmath63 , where @xmath64 is the maximal radius reached by the corrosion up to time time @xmath44 . as the etchant power is reduced , the corrosion front becomes rougher and rougher , until the dynamics is finally arrested ( see fig .
[ fig3 ] ) .
since in the smooth regime @xmath65 , we can write @xmath66 . hence , within this approximation , it is possible to write down the following equation : @xmath67 since in this regime obviously @xmath68 , we can write eq . [ smooth ] in a differential form as follows : @xmath69 whose solution is @xmath70 from this , the characteristic time of the dynamics is seen to be proportional to @xmath71 ; i.e. the etching power of the solution reaches the value @xmath26 in a time @xmath72 proportional to @xmath71 .
moreover , as @xmath68 , at the cross - over between the smooth and the critical regime ( i.e. when @xmath73 ) the solution reaches a distance @xmath74 from the origin . by differentiating this expression
we conclude that the gradient of values of @xmath4 at which different sites have been corroded in the radial direction is proportional to @xmath75 .
finally , as @xmath76 , the gradient @xmath29 at @xmath72 is proportional to @xmath77 . in this way , in analogy with the cylindrical case , we expect that the geometrical properties of the final corrosion front are well represented by gp where the gradient of @xmath4 is dynamically generated . replacing @xmath29 with @xmath77 ,
the scaling relations studied for gradient percolation can be extended to the present case .
the previous description is valid only for the smooth regime , i.e. up to the time at which @xmath78 .
however , since the critical regime is shorter than the smooth one , we have @xmath79 , where @xmath80 is the average radius of the final corrosion front , and the previous estimations remain valid . in order to check that also in the critical regime the radial gradient of the solution etching power
is given by @xmath75 , it is sufficient to assume that during this regime the corrosion front changes from a quite smooth geometry to a rougher one , with a final thickness @xmath28 . because of the much shorter duration of the critical regime one has @xmath81 . in this way , during the critical regime the solution etches a number of solid sites proportional to @xmath82 .
therefore , from eq .
( [ pp ] ) , the variation of the etching power in this regime in average is @xmath83 . in conclusion , we have defined a spherical version of the dem , and seen its connection with gradient percolation : given the time - diminution of @xmath4 , the system generates dynamically a spatial gradient of the values of @xmath4 at which the different sites were etched .
let us finally emphasize that if , after the process is arrested , more etchant solution is added then the process continues until it is stopped again at a value of @xmath4 around @xmath26 . in this way
the disordered solid plays the role of a chemical buffer . in the next section we present a more theoretical treatment allowing us to draw even more precise connections between dem and percolation theory .
in order to construct a field theoretical description for the dynamical etching model , a possibility would be to write down the master equation defining the dynamics and then ( using a poissonian transformation @xcite or alternatively a fock - space formalism @xcite ) , derive a generating functional@xcite . instead of following that strategy
, we prefer here to present a phenomenological set of stochastic langevin equations describing the model at a mesoscopic scale .
this direct approach , `` la landau '' , based on the analysis of the main symmetries and conservation laws of the discrete model , has proven very efficient in describing many other systems related to percolation , directed percolation and , in general , systems with absorbing states @xcite .
let us consider the following three different local densities or coarse - grained fields : * @xmath84 describing the local density of material susceptible to be etched at any time after @xmath44 .
in the discrete model there are two types of sites contributing to this density : ( i ) bulk solid sites and ( ii ) `` fresh '' interface sites i.e. , solid sites freshly arrived to the solid - liquid interface ( susceptible to be etched at the next time - step ) .
* @xmath85 is the local density of passivated and inert material . in the microscopic model
this is the density of interface sites having already resisted an etching trial , i.e. , immune or not - susceptible to be corroded at any future time - step .
* @xmath86 is the local density of corroded and replaced by solution sites , i.e. , the local density of etchant .
the mean field equations ( rate equations ) describing the evolution of the averaged mean values of these magnitudes are : @xmath87 where @xmath46 is the probability to etch an active site at time @xmath44 , and @xmath88 is a positive constant . in
what follows , and without loss of generality we fix @xmath89 .
the interpretation of the first equation is : in order for the density of susceptible sites to change ( decrease ) in a region , it is necessary to have locally both a non - vanishing density of etchants and raw solid material susceptible to be etched .
this restricts the dynamics to _ active _ regions , i.e. , zones in the interface separating the etchable - solid and the solution , in which non - vanishing local densities of @xmath90 and of @xmath91 coexist .
moreover , the second and the third relations in eq . [ deterministic ] express the fact that an active site becomes either a @xmath91-site , with probability @xmath92 ( the corrosion power at time @xmath44 ) , or alternatively , after healing , a @xmath93-site with complementary probability @xmath94 . note that , as @xmath95 , the total number of sites is conserved during the dynamics .
it is worth stressing that eq .
( [ deterministic ] ) captures the fact that sites resisting an etching attempt remain un - corroded indefinitely ( as occurs in the microscopic model ) .
in fact , the number of @xmath93-sites grows monotonously until the etching process is arrested .
observe that we have written so far mean field equations in which spatial dependence and fluctuations are not taken under consideration . at this point
, it is convenient to introduce an activity field @xmath96 ( or @xmath97 as long as spatial dependences are omitted ) . from eq .
[ deterministic ] it follows immediately that @xmath98 in order to implement in our theoretical description the diminution of the etching power as the corrosion process proceeds , we write @xmath46 ( analogously to eq.(4 ) ) as @xmath99 where now @xmath100 $ ] is the number of consumed etchant particles up to time @xmath44 and , as before , @xmath0 is the solution volume . integrating in time the equation for @xmath91 in eq .
[ deterministic ] , integrating then in space , and substituting the result into eq .
[ p ] , we obtain @xmath101 .
\label{p3}\ ] ] plugging this result into eq . [ det2 ] , and re - introducing the spatial dependence of the fields , it is a matter of simple algebra to obtain @xmath102 \rho(\xv , t ) \nonumber \\ & - & \rho(\xv , t ) \left [ \int_0^t dt ' ( 1-p(t ' ) ) \rho(\xv , t ' ) \right ] .
\label{det3}\end{aligned}\ ] ] in order to go beyond this mean field description , it is necessary to include properly spatial coupling and fluctuations ( as a noise ) . *
_ spacial coupling _
+ in principle , the spatial coupling can be taken into account by introducing additional terms into eq .
however , because of the isotropic nature of the local dynamics , terms not invariant under space inversion ( as odd derivatives of the fields ) are not allowed . also , terms like @xmath103 can not appear , given the absence of surface tension in the microscopic rules @xcite .
therefore , only terms as @xmath104 and higher powers of them are allowed .
we introduce in eq .
[ det3 ] the only relevant term in the renormalization group optics @xcite , namely a diffusive coupling @xmath105 , that typically appears in continuous descriptions of interacting particle systems .
it will be checked _ a posteriori _ that , as a matter of fact , the omitted terms are irrelevant at criticality . * _ noise _ + in order to introduce properly the noise term , let us consider a small region in which there are @xmath106 `` fresh '' surface sites in contact with the solution of etching power @xmath4 .
since these fresh sites have random independent resistances , the average number of dissolved sites will be @xmath107 and the fluctuation from this average number will be poissonian , i.e. of order @xmath108 .
this implies that in the continuous description fluctuations of @xmath109 are proportional to its square - root .
consequently , a term @xmath110 has to be added to eq .
[ det3 ] , with @xmath11 being a gaussian white noise with zero mean and no spatio - temporal correlations : @xmath111 .
deviations from gaussianity and higher order corrections can be easily argued to be irrelevant in the renormalization group optics , and therefore are not taken into account . this type of noise , with amplitude proportional to the square - root of the activity field , is characteristic of systems exhibiting a transition from an active to an absorbing phase @xcite : let us emphasize that wherever the activity field is zero , the dynamics is stopped @xcite . introducing these two new ingredients in eq.[det3 ]
one obtains finally @xmath102 \rho(\xv , t ) \nonumber \\ & - & \rho(\xv , t ) \left [ \int_0^t dt ' ( 1-p(t ' ) ) \rho(\xv , t')\right ] \nonumber \\ & + & \nabla^2 \rho(\xv , t ) + \sqrt{\rho(\xv , t ) } \eta(\xv , t ) \label{ext}\end{aligned}\ ] ] up to higher order , irrelevant , terms .
it is worth stressing that even though the microscopic model has originally quenched disorder , it has been possible to describe it in terms of a stochastic equation with annealed noise .
this simplification owes to the fact that every site is tested for corrosion at most once .
if it survives , it will stay un - corroded indefinitely , as previously explained . in this way ,
as every random number is used at most once it has not to be stored , there are no time correlations in the noise and , consequently , a stochastic process with no quenched disorder can be used to cast the discrete model .
it is also important to underline that wherever @xmath10 vanishes , all activity , including fluctuations , ceases in eq.[ext ] . in other words
, @xmath12 defines an absorbing state @xcite .
this is just another way of saying that in the microscopic model , whenever there is no contact between the etchable solid and the solution , i.e. , when they are separated by an interface of passivated ( immune ) solid sites , the dynamics is arrested .
continuous descriptions of systems with absorbing states are based on equations for the activity . in the neighborhood of any absorbing state phase transition the activity is small , and series expansions on the activity density ( as the ones we have used to arrive at eq.[ext ] ) are justified @xcite . observe that , apart from the time dependence , of @xmath46 , eq . [ ext ] is identical to the langevin equation describing _ dynamical percolation _
( see eq .
[ dyp ] ) @xcite .
as described above , dynamical percolation is a rather well known dynamical process generating percolation clusters with a characteristic size determined by @xmath8 ( see eq .
[ dyp ] ) .
let us analyze the differences between eq .
[ ext ] and eq .
particular attention must be paid to the exponential factor in the expression for @xmath46 ( see eq .
[ p3 ] ) absent in dynamical percolation ; due to it some of the coefficients in eq .
[ ext ] are time dependent , while their counterparts in eq .
[ dyp ] are constants .
first we discuss whether this extra time - dependence may affect the critical properties .
note that @xmath46 does not fluctuate ( as verified in simulations in @xcite ) because it is a smooth function of the integral of the field over all the past time and the whole space .
therefore , it is a deterministically decreasing time - dependent term ; or in other words it depends on spatio - temporal integrals of the activity field and not on the local activity field itself . hence , this term has not critical fluctuations , and does not affect the system critical properties .
however , it is crucial in order to characterize the temporal crossover from the active to the absorbing phase . indeed , as the linear term coefficient in eq .
[ ext ] includes a dependence on @xmath46 , for early times it is positive , corresponding to the fact that for early times the system is in the supercritical regime . as the etching mechanism proceeds , the argument of the exponential in eq .
[ p3 ] grows in modulus , and there is a finite time , @xmath72 , at which the linear term coefficient takes its critical value .
immediately after , the process becomes subcritical @xcite , i.e. it reaches the absorbing phase and , the dynamics tends to be stopped with an exponentially fast rate ( i.e. the number of active sites decreases exponentially ) .
furthermore , as more and more sites are etched , the linear coefficient in eq . [ ext ] becomes smaller and smaller than its critical value , and the exponential stopping rate is accelerated till a final time @xmath56 , at which an absorbing ( blocking ) configuration is reached .
therefore , the main effect of the time - dependent linear term coefficient is that , by continuously diminishing , it drives the system to the neighborhood of the dynamical - percolation - field - theory critical point , but it does not add any new relevant operator that could eventually change the universality class . at this point , it is a matter of simple algebra to verify that all the terms omitted in our derivation are indeed irrelevant at the dynamical percolation , eq .
[ dyp ] , renormalization group fixed point .
given the previous discussion , one is allowed to say that eqs .
[ p3 ] and [ ext ] define a _
self - organized dynamical percolation langevin equation _ : without tuning any parameter the dynamics is arrested in the neighborhood of the percolation critical point , and critical ( fractal ) properties can be measured up to certain length scale determined solely by the parameter @xmath0 ( which controls the decrease rate of @xmath46 ) . in the limit @xmath1 the upper cut - off goes to infinity ;
i.e. , for sufficiently large values of @xmath0 the point at which the dynamics is stopped occurs at values of @xmath4 arbitrarily closed to its critical value in eq .
[ dyp ] . observe that the microscopic process is also self - organized : initially @xmath46 is taken to be super - critical , but it decreases monotonously till it reaches a critical value , and as soon as the sub - critical regime is reached the process is stopped in an exponential way .
therefore , our continuous description reproduces the essential features of the microscopic model .
the long range correlations ( generating fractal behavior ) in the dynamical etching process are generated in the regime in which the linear coefficient takes values around its critical value . as a consequence
, it is inferred that the fractal properties of the final frozen configuration are related to the standard dynamical percolation renormalization group fixed point in any dimension .
up to higher order terms , @xmath85 can be written as @xmath112 .
this variable , the integral over the past history of the activity field , represents the statistics of immunized sites as described by janssen @xcite , and therefore , in our problem the statistics of `` surviving '' solid clusters in @xcite is also related to percolation properties .
observe that regions of the cluster of corroded sites far from the final blocking corrosion front have been corroded with a value of @xmath4 larger than its critical value and therefore are not critical .
for the same reason the final corrosion front can be seen as the external perimeter of an invading percolation cluster ( the etchant solution ) with @xmath113 .
this explains the value of the fractal dimension @xmath58 found for this final corrosion front , which is nothing but the hull exponent of percolation in the lattice geometry under consideration @xcite .
finally , we can perform a finite size scaling analysis of our equation in order to determine how different magnitudes scale as a function of the only free parameter , @xmath0 ; in particular , we can evaluate the distance from the critical point at which the process will be finally stopped .
the linear coefficient in eq .
[ ext ] is @xmath114 \nonumber \\ & - & c(\xv,0 ) .
\label{masa}\end{aligned}\ ] ] for the points where the dynamics is arrested , at @xmath56 , it is clear that @xmath115 and @xmath116 ; i.e. at @xmath3 they belong to the solid bulk .
hence , for the bulk we have @xmath117 $ ] independent of @xmath118 .
as usual @xmath72 is the time at which the process is critical @xmath119 , and @xmath56 the time at which the process is actually arrested . for large values of @xmath0
the argument of the exponential in eq .
[ masa ] can be expanded in power series , and we have , up to the leading order : @xmath120 now we can use the scaling analysis presented in the first part of the paper . in the time interval between @xmath72 and @xmath56
the solution erodes a region of the solid with , roughly speaking the shape of a circular hole limited by a crown of radius @xmath121 and width @xmath28 . as argued before @xmath122 .
moreover , since the quantity @xmath123 gives the amount of material tested by the etching solution between @xmath72 and @xmath56 , we can write @xmath124 . therefore @xmath125 .
writing @xmath126 one obtains @xmath127 the fact that @xmath0 is finite then implies that @xmath28 is finite and @xmath128 .
it is only in the large @xmath0 limit that the process is stopped exactly at the dynamical percolation critical point .
as discussed in section 3 , one has @xmath129 @xcite , and therefore the distance from the final mass to the critical one scales as @xmath130 , that is as the excursion of the occupancy probabilities along the separation interface of gp .
therefore we have determined not only the universality class , but also established how finite size effect operate in the dem . in conclusion ,
all the critical ( fractal ) properties of the microscopic model can be shown to be related to ( dynamical ) percolation in any space dimension , by using the above presented continuous ( field theoretical ) representation , and finite size corrections can be evaluated .
summing up , we have found that the dynamical etching model proposed by sapoval et al .
, is a self - organized process describable by a the continuous langevin equation similar to that of dynamical percolation .
this langevin equation includes a linear term coefficient that decreases monotonously as the etching process goes on . in this way
as soon as it takes it critical value and enters the `` absorbing '' regime the etching process is stopped .
consequently , the fractal ( scale invariant ) properties of the interface in the etching process are shown to be related to the ( dynamical ) percolation renormalization group fixed point .
this result is valid in any space dimension .
in particular , our analysis permits us to conclude that the etching upper critical dimension is @xmath14 , as in ( dynamical ) percolation . on the other hand ,
we have also evaluated the role of finite size corrections that are essentially different from those of standard dynamical percolation .
an interesting aspect from a theoretical perspective is that the field theory ( langevin equation ) describing the process is self - organized , in the sense that , without any parameter fine - tuning , fractal , scale - invariant properties are generated .
however , it is only in the limit @xmath1 ( @xmath131 ) , that the upper cut - off for scaling diverges .
this is in clear analogy with what occurs in other models of self - organization , as sandpiles @xcite for which critical behavior is observed in the limit of dissipation and driving going to zero @xcite .
the mechanism discussed in this paper constitutes a new `` path '' to self - organized criticality @xcite , in which the control parameter decreases monotonously until it reaches the neighborhood of the absorbing - state phase transition at which the dynamics is arrested in an exponential way .
this same mechanism will be investigated in the context of different types of absorbing - state phase transitions ( as directed percolation @xcite ) in a future work .
observe that this scenario has the great advantage of being related in a clear - cut way to real physical systems , making therefore the observation of self - organized criticality much more accessible to experiments .
we acknowledge useful discussion with a. baldassarri , u. marini bettolo marconi and p. hurtado .
we acknowledge partial support from the european network contract erbfmrxct980183,and dgesic ( spain ) project pb97 - 0842 .
we thank it .
ffss for giving us the opportunity to discuss this problem . m.a.m . acknowledges the kind hospitality at the ecole polytechnique where this work was started .
u. r. evans , _ the corrosion and oxidation of metals : scientific principles and practical applications _ , arnold , ( london ) , 1960 ) .
h. h. uhlig , _ corrosion and corrosion control _ , ( wiley , new york ) , 1963 . in the language of excitable media ( autocatalytic chemical reactions , propagation of electrical activity in neurons , spreading of diseases )
there are , in general three types of states called : quiescent ( healthy ) , excited ( infected ) , and refractory ( immune ) .
see , g. grinstein and m. a. muoz , _ the statistical mechanics of systems with absorbing states _ , in fourth granada lectures in computational physics , edited by p. garrido and j. marro , lecture notes in physics , vol .
493 ( springer , berlin 1997 ) , p. 223 , and
references therein .
see also , h. hinrichsen , adv .
phys . * 49 * , 815 ( 2000 ) .
d. j. amit , _ field theory , the renormalization group and critical phenomena _ , ( world scientific , singapore , 1992 ) .
j. zinn - justin , _ quantum field theory and critical phenomena _ , ( clarendon press , oxford , 1990 ) .
b. sapoval , m. rosso , and j. f. gouyet , j. phys .
( paris ) , * 46 * , l149 ( 1985 ) ; b. sapoval , m. rosso , and j. f. gouyet , in _ the fractal approach to heterogeneous chemistry _ , edited by d. avnir ( john wiley and sons ltd . , new york , 1989 ) .
m. rosso , j. f. gouyet , and b. sapoval , phys .
b * 32 * , 6053 ( 1985 ) .
r. ziff and b. sapoval , j. phys .
a : math . gen .
19 , l1169 - 1172 ( 1986 ) .
the connectivity criterion is a relevant ingredient to be taken in consideration .
for example , in a square lattice , the connection between occupied sites is to first neighbor while the connection between empty ( etched ) sites is to first and second nearest neighbors .
see the discussion in @xcite .
observe that the critical value of the linear term coefficient is equal to zero only at mean field ( tree ) level .
as soon as fluctuations , i.e. diagrammatic corrections , are taken into account the critical mass is shifted to a non - vanishing value @xcite . | we propose a phenomenological field theoretical approach to the chemical etching of a disordered - solid .
the theory is based on a recently proposed dynamical etching model . through the introduction of a set of langevin equations for the model evolution ,
we are able to map the problem into a field theory related to isotropic percolation . to the best of the authors knowledge
, it constitutes the first application of field theory to a problem of chemical dynamics . by using this mapping , many of the etching process critical properties are seen to be describable in terms of the percolation renormalization group fixed point .
the emerging field theory has the peculiarity of being `` _ self - organized _ '' , in the sense that without any parameter fine - tuning , the system develops fractal properties up to certain scale controlled solely by the volume , @xmath0 , of the etching solution . in the limit @xmath1 the upper cut - off goes to infinity and the system becomes scale invariant .
we present also a finite size scaling analysis and discuss the relation of this particular etching mechanism with gradient percolation .
finally , the possibility of considering this mechanism as a new generic path to self - organized criticality is analyzed , with the characteristics of being closely related to a real physical system and therefore more directly accessible to experiments . 2 |
the shift in critical temperature of the homogeneous bose gas has been the subject of numerous investigations over the past fifty years .
as the density of this idealized system is constant , the shift due to the mean - field is zero , and the first order shift is due to long - wavelength critical fluctuations .
the first estimates were due to lee and yang @xcite , who gave two different results for the dependence on the s - wave scattering length @xmath0 . in 1999
_ et al.__@xcite determined that the result should be @xmath1 where @xmath2 is the particle number density , and @xmath3 is a constant of order unity .
several authors have attempted to calculate this constant , and a wide range of results have been obtained , as summarised in fig . 1 of @xcite .
however recent monte carlo calculations have apparently settled this matter , giving a combined result of @xmath4 @xcite . a useful summary and discussion of this topic is provided by andersen @xcite and holzmann
_ et al_. @xcite .
previously we have performed numerical simulations of an equation known as the projected gross - pitaevskii equation ( pgpe ) , which can be used to represent the highly occupied modes of bose condensed gases at finite temperature @xcite .
this equation is observed to evolve randomised initial conditions to equilibrium for which it is possible to measure a temperature @xcite .
the pgpe is non - perturbative , and hence includes the effect of critical fluctuations .
the only approximation made is that the modes of the gas are sufficiently highly occupied as to be well described by a classical rather than a quantum field .
the occupation condition is that mode @xmath5 must satisfy @xmath6 ; however for practical simulations we may choose , for example , @xmath7 @xcite .
this method is suitable for investigating many problems of current interest in ultra - cold bose gases , such as the shift in critical temperature due to non - perturbative effects @xcite .
the pgpe describes a microcanonical system , with the classical field restricted to a finite number of modes for which the occupation number condition is met . in order to study the problem of the shift in @xmath8 it is necessary to accurately determine thermodynamic quantities defined as derivatives of the entropy such as the temperature and chemical potential . in 1997
rugh developed a new approach to statistical mechanics where he derived an expression from the hamiltonian of a system for which the ensemble average gives the temperature within the microcanonical ensemble @xcite .
however , if the system is known to be ergodic then the equilibrium temperature can be determined from the system dynamics over a sufficiently long period of time .
we have applied an extension of rugh s method to the pgpe hamiltonian , and the appropriate expression to determine the temperature is given as eq .
( 22 ) of @xcite .
this method was found to agree with the less rigorous methods described in @xcite . in @xcite
we made use of this method to calculate the shift in the critical temperature of the homogeneous bose gas . despite the calculation being performed with limited statistics and suffering from finite size effects
, it gave a result of @xmath9 in agreement with the monte carlo results @xcite .
in @xcite we applied this method to the experiment of gerbier _ et al .
_ @xcite who measured the shift in critical temperature of a trapped bose gas , and found good agreement with experiment . in this paper
we give the details of our implementation of rugh s method for a general mode basis for the pgpe .
we consider a classical system with @xmath10 independent modes .
the hamiltonian can be written as @xmath11 , where @xmath12 is a vector of length @xmath13 consisting of the canonical position and momentum co - ordinates .
we define the gradient operator @xmath14 in terms of its components @xmath15 . in the notation of rugh @xcite
, the hamiltonian @xmath16 may have a number of independent first integrals , labelled by @xmath17 that are invariant under the dynamics of @xmath16 .
a particular macro - state of such a system can be specified by the values of the conserved quantities , labelled as @xmath18 .
the usual expression for the temperature of a system in the microcanonical ensemble is given by @xmath19 where all other constants of motion are held fixed , and where the entropy can be written @xmath20 \ ; \prod_i \delta[{i_i - f_i(\mathbf{\gamma})}].\ ] ] using rugh s methods , the temperature of the system can be written as @xmath21 where the angle brackets correspond to an ensemble average , and the components of the vector operator @xmath22 are @xmath23 where @xmath24 can be chosen to be any scalar value , including zero .
the vector field @xmath25 can also be chosen freely within the constraints @xmath26 geometrically this means that the vector field @xmath25 has a non - zero component transverse to the @xmath27 energy surface , and is parallel to the surfaces @xmath28 .
the expectation value in eq .
( [ eqn : temp_eqn ] ) is over all possible states in the microcanonical ensemble ; however if the ergodic theorem is applicable then it can equally well be interpreted as a time - average . for further details on the origin of this expression
we refer the reader to rugh s original papers @xcite , as well as derivations found in giardin and levi @xcite , jepps _
et al._@xcite and rickayzen and powles @xcite .
the classical hamiltonian for the dilute bose gas in dimensionless form is @xmath29 where @xmath30 , @xmath31 is the number of particles in the system , @xmath32 , @xmath33 is the unit of length , @xmath34 is the unit of energy , and @xmath35 is the mass of the particles .
the dimensionless classical bose field operator @xmath36 is here normalized to one , @xmath37 , and @xmath38 is the nonlinear constant defined as @xmath39 where @xmath40 .
@xmath41 is the single particle hamiltonian with eigenbasis @xmath42 , and @xmath43 is the dimensionless version of any potential _
additional _ to that included in the single - particle hamiltonian .
we restrict our field @xmath36 to be formed from the modes of the classical region @xmath44 defined by a high energy cutoff in the single - particle basis , such that we can write @xmath45 the equation of motion for this restricted system is the projected gross - pitaevskii equation @xmath46 the projection operator @xmath47 projects function @xmath48 onto the classical modes of the single particle hamiltonian @xmath41 via @xmath49 it eliminates collisional processes which would cause the scattering of particles into higher energy single particle modes not represented by the classical field . in the single - particle basis of @xmath41
the hamiltonian ( [ eqn : qh ] ) can be written @xmath50 where we have defined the matrix elements @xmath51 we make use of the real , canonically - conjugate coordinates @xmath52 and @xmath53 defined by @xmath54 with the inverse transformation @xmath55 in order to satisfy the conditions ( [ eqn : x_conditions ] ) we can choose a vector field of the form @xmath57 where the @xmath58 coefficients @xmath59 are determined by the @xmath58 simultaneous equations in eq .
( [ eqn : x_conditions ] ) . due to the freedom in the choice of the vector operator
@xmath22 we can set any number of components of the length @xmath13 vector @xmath25 to zero .
this turns out to be useful as the components corresponding to the momentum and position variables can be different orders of magnitude .
two particular choices we make use of later are @xmath60 with @xmath61 and @xmath62 with @xmath63 these lead to two different calculations for the temperature that only agree in general if the system is in thermal equilibrium .
this provides a useful check that the numerical simulations have indeed reached thermal equilibrium , as well as giving two independent values for the temperature . for the classical bose gas hamiltonian ( [ eqn : qh ] ) there can be several constants of motion that must be taken into account .
these depend on the details of the single - particle hamiltonian @xmath41 examples include components of the angular and/or linear momentum .
the effect of including these additional first integrals in the definition of the vector field @xmath25 is to account for the energy that is associated with a conserved quantity and hence is unavailable for thermalization .
this ensures that only the appropriate free energy is used to calculate the temperature .
we conjecture that the same result can be achieved by first transforming to a co - ordinate system where the total angular and linear momenta , etc , are all zero and therefore do not contribute to the energy of the system .
rugh demonstrated this explicitly in @xcite for a system of particles with a conserved centre - of - mass motion , and our numerical results support this conjecture . an exception to this , however , is the conservation of normalization @xmath64 .
this must be considered explicitly because there is no co - ordinate system in which it can be made to vanish .
the constraint on @xmath65 means that the ground state of the system will , in general , have a finite energy that is not accessible for thermalization .
we note that the effect of this constraint is in general more complicated than a simple subtraction of the ground state energy ( which could be achieved by hand ) and depends on the definition of the operator @xmath22 used to calculate the temperature , as shown below .
we need to choose a vector field @xmath25 which satisfies eqs .
( [ eqn : x_conditions ] ) with @xmath66 .
the result is @xmath67 where the parameter @xmath68 looks similar to a chemical potential . substituting eq .
( [ eqn : xvec ] ) into eq .
( [ eqn : temp_eqn ] ) we find that our full expression for the dimensionless temperature @xmath69 is @xmath70 } { [ { |{\mathbf{\mathcal{d}}}h|^2 - \lambda_n ( { \mathbf{\mathcal{d}}}h \cdot { \mathbf{\mathcal{d}}}n)}]^2 } \right\rangle , \label{eqn : t}\end{aligned}\ ] ] where @xmath71 the chemical potential of the system can be determined in a similar manner using a different choice of vector field .
in particular we wish to calculate @xmath72 where the conditions on the vector field are @xmath73 this results in expressions identical to the rhs of eqs .
( [ eqn : xvec],[eqn : x_with_n],[eqn : t ] ) but with @xmath16 and @xmath65 interchanged , and so we will not write them out in full .
the technique outlined above has been described previously by one of us in ref .
@xcite , and in particular the result of eq .
( [ eqn : t ] ) was given in that work .
the numerical method has also been applied by the current authors in refs .
the purpose of this paper is to outline in detail the procedure for calculating the terms in eqs .
( [ eqn : t ] ) and ( [ eqn : dl ] ) .
this is not a trivial exercise , and the proliferation of terms can make it difficult to avoid errors in both the analytical and numerical implementation . by discussing our approach to evaluating eqs .
( [ eqn : t ] ) and ( [ eqn : dl ] ) we hope to facilitate other researchers wanting to make use of the method .
the basic numerical requirement is an efficient and accurate numerical transform from real space to basis space and vice versa .
this may be provided , for example , by fast fourier transforms using a basis of plane waves appropriate for a homogeneous system @xcite , or an efficient quadrature for trapped systems @xcite .
in this section we give explicit analytic expressions for all quantities required in the evaluation of eq .
( [ eqn : t ] ) and ( [ eqn : dl ] ) . as we are using the canonical co -
ordinates defined in eq . ( [ def : xp ] ) it is easiest to evaluate the derivatives of the hamiltonian using the mode expansion of the hamiltonian ( [ eqn : hbasis ] ) .
however , this leaves us with expressions involving inner products of vector quantities with summations that would be computationally expensive to evaluate directly for any sizeable basis .
we show how to define auxiliary field functions that can be used to simplify these terms for implementation with an efficient numerical transform .
we begin by writing out the necessary derivatives of the hamiltonian using the two choices of vector operator defined earlier , @xmath74 and @xmath75 . in component form
the first derivatives of the hamiltonian are @xmath76 + { c_{\rm nl}}\sqrt{\frac { \epsilon_i}{2 } } \sum_{pqm } \left [ c^*_p c^*_q c_m \langle pq|mi\rangle + c^*_m c_p c_q \langle mi|pq\rangle\right ] , \label{dqh1 } \\ ( { \mathbf{\mathcal{d}}}_p h)_i & = & p_i + \frac{i}{\sqrt{2 \epsilon_i}}\sum_n \left[v_{ni } c_n^ * - v_{in } c_n\right ] + i \frac { { c_{\rm nl}}}{\sqrt{2 \epsilon_i } } \sum_{pqm } \left [ c^*_p c^*_q c_m \langle pq|mi\rangle - c^*_m c_p c_q \langle mi|pq\rangle\right ] , \label{dph1}\end{aligned}\ ] ] and
the second derivatives are @xmath77 + \frac{{c_{\rm nl}}}{2 } \sqrt{\epsilon_i \epsilon_j } \sum_{pq } \left [ 4 c^*_p c_q \langle pi|qj\rangle + c_p c_q \langle ij|pq\rangle + c_p^ * c_q^ * \langle pq|ij\rangle \right ] , \label{dqh2 } \\ ( { \mathbf{\mathcal{d}}}_p^2 h)_{ij } & = & \delta_{ij } + \frac{1}{2\sqrt{\epsilon_i \epsilon_j}}[v_{ij } + v_{ji } ] + \frac{{c_{\rm nl}}}{2\sqrt{\epsilon_i \epsilon_j } } \sum_{pq } \left [ 4c^*_p c_q \langle pi|qj\rangle - c_p c_q \langle ij |pq\rangle - c_p^ * c_q^ * \langle pq|ij\rangle\right ] .
\label{dph2}\end{aligned}\ ] ] we also need the derivatives of the other first integral of the hamiltonian the the normalisation given by @xmath78 .
these are @xmath79 from eq .
( [ eqn : t ] ) we can see that we need the terms @xmath80 @xmath81 .
\label{eqn : list2}\ ] ] some of these expressions are simple to calculate e.g. @xmath82 @xmath83 @xmath84 @xmath85 however , the remainder are more complicated . to calculate them efficiently we make the following vector definitions @xmath86 with the corresponding auxiliary field functions @xmath87 making use of these definitions we find @xmath88 + \frac{{c_{\rm nl}}}{2 } \sqrt{\epsilon_i \epsilon_j } \sum_{pq } \left [ 4 c^*_p c_q \langle pi|qj\rangle + c_p c_q \langle ij|pq\rangle + c_p^ * c_q^
* \langle pq|ij\rangle \right]\right ) , \nonumber\\ & = & 2 a_i \epsilon_i^{3/2 } + \sqrt{\epsilon_i}\sum_j a_j[v_{ij } + v_{ji } ] + { c_{\rm nl}}\sqrt{\epsilon_i } \sum_{pqj}a_j \left [ 4 c^*_p c_q \langle pi|qj\rangle + c_p c_q \langle ij|pq\rangle + c_p^ * c_q^ * \langle pq|ij\rangle \right ] , \label{dqdqh2}\end{aligned}\ ] ] @xmath89 + \frac{{c_{\rm nl}}}{2\sqrt{\epsilon_i \epsilon_j } } \sum_{pq } \left [ 4c^*_p c_q \langle pi|qj\rangle - c_p c_q \langle ij |pq\rangle - c_p^ * c_q^ * \langle pq|ij\rangle\right ] \right),\nonumber\\ & = & 2\sqrt{\epsilon_i}b_i + \frac{1}{\sqrt{\epsilon_i } } \sum_j b_j[v_{ij } + v_{ji } ] + \frac{{c_{\rm nl}}}{\sqrt{\epsilon_i } } \sum_{pqj}b_j \left [ 4c^*_p c_q \langle pi|qj\rangle - c_p c_q \langle ij |pq\rangle - c_p^ * c_q^ * \langle pq|ij\rangle\right ] , \label{dpdph2}\end{aligned}\ ] ] and @xmath90 + \frac{{c_{\rm nl}}}{2 } \sqrt{\epsilon_i \epsilon_j } \sum_{pq } \left [ 4 c^*_p c_q \langle pi|qj\rangle + c_p c_q \langle ij|pq\rangle + c_p^ * c_q^ * \langle pq|ij\rangle \right ] \right),\nonumber\\ & = & f_i \epsilon_i^{3/2 } + \frac{\sqrt{\epsilon_i}}{2}\sum_j f_j[v_{ij } + v_{ji } ] + \frac{{c_{\rm nl}}}{2 } \sqrt{\epsilon_i } \sum_{pqj}f_j \left [ 4 c^*_p c_q \langle pi|qj\rangle + c_p c_q \langle ij|pq\rangle + c_p^ * c_q^ * \langle pq|ij\rangle \right ] , \label{dqnd2qh}\end{aligned}\ ] ] @xmath91 + \frac{{c_{\rm nl}}}{2\sqrt{\epsilon_i \epsilon_j } } \sum_{pq } \left [ 4c^*_p c_q \langle pi|qj\rangle - c_p c_q \langle ij |pq\rangle - c_p^ * c_q^ * \langle pq|ij\rangle\right ] \right),\nonumber\\ & = & \sqrt{\epsilon_i}g_i + \frac{1 } { 2 \sqrt{\epsilon_i } } \sum_j g_j[v_{ij } + v_{ji } ] + \frac{{c_{\rm nl}}}{2\sqrt{\epsilon_i } } \sum_{pqj}g_j \left [ 4c^*_p c_q \langle pi|qj\rangle - c_p c_q \langle ij |pq\rangle - c_p^ * c_q^ * \langle pq|ij\rangle\right ] .
\label{dpnd2ph}\end{aligned}\ ] ] the above expressions are written in a form that they can be easily calculated using efficient quadratures in real space .
we outline in detail how to calculate all of these terms , and write them in the most efficient form we have found .
we make the definitions @xmath92 where we have assumed that the potential @xmath43 is real .
we can therefore write the appropriate parts of eqs .
( [ dqh1 ] ) and ( [ dph1 ] ) as @xmath93 & = & 2 \mbox{re}(w_i),\\ \sum_n \left[v_{ni } c_n^ * - v_{in } c_n\right ] & = & -2 i \mbox{im}(w_i).\end{aligned}\ ] ] the next set of terms are @xmath94 hence , parts of eqs .
( [ dqh1 ] ) and ( [ dph1 ] ) can be written @xmath95 & = & 2 \mbox{re}(k_i),\\ \sum_{pqm } \left [ c^*_p c^*_q c_m \langle pq|mi\rangle - c^*_m c_p c_q \langle mi|pq\rangle\right ] & = & -2i \mbox{im}(k_i).\end{aligned}\ ] ] for the terms involving the auxiliary fields we have expressions like @xmath96 where we have made use of the fact that @xmath97 .
we have similar expression for all the other auxiliary vector fields @xmath98 , and @xmath99 and will refrain from writing these all out .
we can write parts of eqs .
( [ dqdqh2 ] ) , ( [ dpdph2 ] ) , ( [ dqnd2qh ] ) and ( [ dpnd2ph ] ) as @xmath100 & = & 4(a_1)_i + 2\mbox{re}(a_2)_i,\\ \sum_{pqj}b_j \left [ 4c^*_p c_q \langle pi|qj\rangle - c_p c_q \langle ij |pq\rangle - c_p^ * c_q^ * \langle pq|ij\rangle\right ] & = & 4(b_1)_i - 2 \mbox{re}(b_2)_i,\\ \sum_{pqj}f_j \left [ 4 c^*_p c_q \langle pi|qj\rangle + c_p c_q \langle ij|pq\rangle + c_p^ * c_q^ * \langle pq|ij\rangle \right ] & = & 4(f_1)_i + 2 \mbox{re}(f_2)_i,\\ \sum_{pqj}g_j \left [ 4c^*_p c_q \langle pi|qj\rangle - c_p c_q
\langle ij |pq\rangle - c_p^ * c_q^ * \langle pq|ij\rangle\right ] & = & 4(g_1)_i - 2 \mbox{re}(g_2)_i,\end{aligned}\ ] ] @xmath101 & = & 2\mbox{re}(a_3)_i,\qquad \sum_j b_j[v_{ij } + v_{ji } ] \ , = \ , 2 \mbox{re}(b_3)_i,\\ \sum_j f_j[v_{ij } + v_{ji } ] & = & 2\mbox{re}(f_3)_i,\qquad \sum_j g_j[v_{ij } + v_{ji } ] \ , = \ , 2 \mbox{re}(g_3)_i.\end{aligned}\ ] ] the remaining terms are diagonal second derivatives of the form @xmath102 , in which we have matrix elements @xmath103 and thus the appropriate parts of eqs .
( [ dqh2 ] ) and ( [ dph2 ] ) are @xmath104 & = & 4(m_1)_i + 2\mbox{re}(m_2)_i,\\ \sum_{pq } \left [ 4 c^*_p c_q \langle pi|qi\rangle - c_p c_q \langle ii|pq\rangle - c_p^ * c_q^ * \langle pq|ii\rangle \right ] & = & 4(m_1)_i - 2 \mbox{re}(m_2)_i.\end{aligned}\ ] ] we have now explicitly outlined the calculation of all terms necessary to calculate the temperature and chemical potential of the equilibrium dynamical evolution of a microcanonical pgpe system
. here we summarise all the components of the vector quantities required for the evaluation of the temperature according to eq .
( [ eqn : t ] ) making use of the definitions made in the previous section for both choices of vector operator @xmath74 and @xmath75 .
the inner products between these vectors found in eq .
( [ eqn : t ] ) can be easily evaluated numerically .
in addition , these are all the terms required for calculating the chemical potential as well .
@xmath105 @xmath106,\\ ( { \mathbf{\mathcal{d}}}_p^2 h)_{ii } & = & 1 + \frac{v_{ii}}{\epsilon_i } + \frac{{c_{\rm nl}}}{\epsilon_i } [ 2(m_1)_i - \mbox{re}(m_2)_i].\end{aligned}\ ] ] @xmath107 @xmath108,\\ { \mathbf{\mathcal{d}}}_p ( |{\mathbf{\mathcal{d}}}_p h|^2)_i & = & 2\sqrt{\epsilon_i}b_i + \frac{2}{\sqrt{\epsilon_i } } \mbox{re}(b_3)_i + \frac{{c_{\rm nl}}}{\sqrt{\epsilon_i } } [ 4(b_1)_i - 2 \mbox{re}(b_2)_i],\end{aligned}\ ] ] @xmath109 ,
\\ \sum_j ( { \mathbf{\mathcal{d}}}_p n)_j ( { \mathbf{\mathcal{d}}}_p^2 h)_{ij } & = & \sqrt{\epsilon_i}g_i + \frac{1 } { \sqrt{\epsilon_i } } \mbox{re}(g_3)_i + \frac{{c_{\rm nl}}}{\sqrt{\epsilon_i } } [ 2(g_1)_i - \mbox{re}(g_2)_i ] . \label{eqn : last } \end{aligned}\ ] ]
in this section we outline the algorithm for applying the formulae of the previous section to simulation results , and give examples of the results obtained for the pgpe for trapped bose gas systems .
we have previously demonstrated in @xcite that the pgpe will evolve any randomised initial field function to an equilibrium that can be characterised by a temperature , and have made use of this in the beyond mean - field calculation of the critical temperature of a trapped bose gas in @xcite . the basic requirement to determine the temperature in such calculations is to evaluate eq .
( [ eqn : t ] ) over a sufficiently long period of simulation time for the ergodic averaging to be effective .
our procedure is as follows .
beginning with an initial random field configuration with a predetermined energy , we run our simulation for sufficiently long such that we can be reasonably sure that equilibrium has been reached ( this can be checked as described below ) .
this is typically for one hundred or more trap periods for simulations of bose gases in a harmonic potential .
we usually save one thousand or more field configurations at equally spaced times throughout the evolution , which are sufficiently separated in time that there is little correlation between samples .
the temperature analysis is performed after the simulation is completed .
the field configurations are stored as vectors of basis coefficients , and we make use of efficient computational routines that avoid numerical aliasing to transform to and from a real space representation . for each saved field configuration
we calculate the vector components @xmath110 and @xmath111 , the derivatives in eqs .
( [ eqn : list1][eqn : list2 ] ) , the auxillary fields given by eqs .
( [ eqn : aux1],[eqn : aux2 ] ) , and then their overlaps with the various functions of the field @xmath36 as in eqs .
( [ eqn : a1][eqn : a5 ] ) .
this allows us to calculate the vector quantities defined in eqs.([eqn : first][eqn : last ] ) that appear in eq .
( [ eqn : t ] ) .
we then form the appropriate dot products of these quantities to give a single sample of @xmath112 and @xmath113 .
these values are calculated for every saved field configuration .
these can then be plotted versus time , and any initial transients before the field has settled into equilibrium can be identified ( typically this a very small fraction of the total simulation time . ) the quantities @xmath112 and @xmath114 are numerically distinct from one another before equilibrium is reached , however they can be quite noisy and so sometimes it is difficult to tell when they are in agreement .
we exclude the transients from the final averaging to determine the temperature typically we discard at least the first quarter of the time evolution to ensure accuracy .
we now illustrate with a set of typical numerical results from applying the method described in this paper to the pgpe for trapped bose gas systems .
we solve the pgpe ( [ eq : gpe1 ] ) for a harmonic trapping potential @xmath115 and no additional potential @xmath43 .
the unit of length is @xmath116 and energy @xmath117 .
we have a dimensionless energy cutoff of @xmath118 such that there are 1739 classical modes , and the data shown is for @xmath119 .
we begin individual simulations with a randomised initial condition with fixed energy , and evolve in dimensionless time for until @xmath120 , which is slightly more than 190 radial trap periods .
we find that these simulations equilibrate very quickly , and use 1000 field samples over the last two - thirds of the evolution for the ergodic averaging .
the results are shown in fig .
[ fig1 ] , some of which have previously been reported in ref .
@xcite . on the rhs of eq .
( [ eqn : temp_eqn ] ) for the simulations with @xmath121 and @xmath122 .
the bars are for the operator @xmath74 , and the dots connected by solid lines for @xmath75 .
( c ) temperature versus energy .
plusses : @xmath123 , crosses : @xmath124 .
( d ) chemical potential versus energy .
plusses : @xmath125 , crosses : @xmath126 .
all quantities are in dimensionless units as defined in the text .
, width=453 ] figure [ fig1](a ) shows the condensate fraction versus energy for the pgpe system ( we remind the reader that energy given by eq .
( [ eqn : hbasis ] ) is a conserved quantity in these calculations ) .
the condensate fraction is determined by the penrose - onsager criterion : it is the largest eigenvalue found by diagonalizing the single - particle density matrix formed by ergodic averaging as described in @xcite .
it looks somewhat different to the result for the full three - dimensional system due to the basis cutoff that is present in these simulations .
figure [ fig1](b ) shows the distribution of the function @xmath127 for two different simulations , using the formulae for both @xmath74 and @xmath75 .
the temperature is found by the inverse of the mean value of these distributions , and is shown for all simulations in fig . [ fig1](c ) .
we can see that the results of both calculations agree very well , and so we can be confident that firstly the simulations have reached equilibrium , and secondly that the numerical implementation of these calculations are free from errors .
figure [ fig1](d ) shows the similar calculation for the chemical potential .
one interesting point is that the width of the distributions in fig .
[ fig1](b ) calculated using the operator @xmath75 are slightly narrower than those for @xmath74 . while this is of no consequence conceptually , in practice the narrower the distribution the fewer samples are required for an estimate of the mean to a required accuracy . given the large amount of freedom in the choice of the operator @xmath22 , it seems quite possible that for particular problems that some choices will be better than others .
we have found situations where one of these distributions is signficantly narrower than the other .
this has also been pointed out by rugh @xcite . however , with no a priori way to estimate the width of the distributions , finding the most accurate method of determing the temperature is a matter of trial and error .
to investigate the effect of critical fluctuations in bose gases using the pgpe in the microcanonical ensemble it is necessary to have accurate methods of determining the thermodynamic temperature and chemical potential . in this paper
we have explicitly outlined a method of how to do so using the assumption of ergodicity and the dynamical evolution of the pgpe .
the method could potentially be applied to other nonlinear classical field hamiltonians . | the ergodic hypothesis asserts that a classical mechanical system will in time visit every available configuration in phase space .
thus , for an ergodic system , an ensemble average of a thermodynamic quantity can equally well be calculated by a time average over a sufficiently long period of dynamical evolution . in this paper
we describe in detail how to calculate the temperature and chemical potential from the dynamics of a microcanonical classical field , using the particular example of the classical modes of a bose - condensed gas .
the accurate determination of these thermodynamics quantities is essential in measuring the shift of the critical temperature of a bose gas due to non - perturbative many - body effects . |
let @xmath2 be a measurable space , @xmath3 , @xmath4 and let @xmath5 be nonatomic finite measures defined on the same @xmath6algebra @xmath7 . let @xmath8 stand for the set of all measurable partitions @xmath9 of @xmath10 ( @xmath11 for all @xmath12 , @xmath13 , @xmath14 for all @xmath15 ) .
let @xmath16 denote the @xmath17-dimensional simplex .
for this definition and the many others taken from convex analysis , we refer to @xcite .
a partition @xmath18 is said to be @xmath0optimal , for @xmath19 , if @xmath20 this problem has a consolidated interpretation in economics .
@xmath10 is a non - homogeneous , infinitely divisible good to be distributed among @xmath1 agents with idiosyncratic preferences , represented by the measures .
a partition @xmath21 describes a possible division of the cake , with slice @xmath22 given to agent @xmath12 . a satisfactory compromise between the conflicting interests of the agents , each having a relative claim @xmath23 , @xmath12 , over the cake , is given by the @xmath0optimal partition .
it can be shown that the proposed solution coincides with the kalai - smorodinski solution for bargaining problems ( see kalai and smorodinski @xcite and kalai @xcite ) . when @xmath5 are all probability measures , i.e .. @xmath24 for all @xmath12 ,
the claim vector @xmath25 describes a situation of perfect parity among agents .
the necessity to consider finite measures stems from game theoretic extensions of the models , such as the one given in dallaglio et al .
@xcite . when all the @xmath26 are probability measures , dubins and spanier @xcite showed that if @xmath27 for some @xmath28 , then @xmath29 .
this bound was improved , together with the definition of an upper bound by elton et al .
a further improvement for the lower bound was given by legut @xcite .
the aim of the present work is to provide further refinements for both bounds .
we consider the same geometrical setting employed by legut @xcite , i.e. the partition range , also known as individual pieces set ( ips ) ( see barbanel @xcite for a thorough review of its properties ) , defined as @xmath30 let us consider some of its features .
the set @xmath31 is compact and convex ( see lyapunov @xcite ) .
the supremum in is therefore attained .
moreover @xmath32 so , the vector @xmath33 is the intersection between the pareto frontier of @xmath31 and the ray @xmath34 . to find both bounds
, legut locates the solution of the maxsum problem @xmath35 on the partition range .
then , he finds the convex hull of this point with the corner points @xmath36 ( @xmath37 is placed on the @xmath38-th coordinate ) to find a lower bound , and uses a separating hyperplane argument to find the upper bound .
we keep the same framework , but consider the solutions of several maxsum problems with weighted coordinates to find better approximations .
fix @xmath39 and consider @xmath40 let @xmath41 be a non - negative finite - valued measure with respect to which each @xmath26 is absolutely continuous ( for instance we may consider @xmath42 ) .
then , by the radon - nikodym theorem for each @xmath43 @xmath44 where @xmath45 is the radon - nikodym derivative of @xmath26 with respect to @xmath46 finding a solution for is rather straightforward : ( see ( * ? ? ?
* theorem 2 ) , ( * ? ? ?
* theorem 2 ) ( * ? ? ?
* proposition 4.3 ) ) let @xmath47 and let @xmath48 be an @xmath49partition of @xmath10 . if @xmath50 then @xmath51 is optimal for . given @xmath47 , an _ efficient value vector ( evv ) _ with respect to @xmath52 , @xmath53 is defined by @xmath54 the evv @xmath55 is a point where the hyperplane @xmath56 touches the partition range @xmath57 so @xmath55 lies on the pareto border of @xmath58
as we will see later only one evv is enough to assure a lower bound , we give a general result for the case where several evvs have already been computed .
we derive this approximation result through a convex combination of these easily computable points in @xmath57 which lie close to @xmath59 [ teo_main ] consider @xmath60 linearly independent vectors @xmath61 where @xmath62 , @xmath63 is the evv associated to @xmath64 , @xmath65 .
let @xmath66 be an @xmath67 matrix and denote as @xmath68 an @xmath69 submatrix of @xmath70 such that @xmath71 a. @xmath72 + if and only if @xmath73 where @xmath74 is the @xmath69 matrix obtained by replacing the @xmath75th column of @xmath76 with @xmath77 , obtained from @xmath78 by selecting the elements corresponding to the rows in @xmath76 .
moreover , @xmath79 if and only if @xmath80 b. for any choice of @xmath81 , @xmath82 moreover , if holds , then @xmath83_{ij } } \leq v^{\alpha}\ ] ] where @xmath84_{ij}$ ] is the @xmath85-th element of @xmath86 . to prove @xmath87 ,
suppose holds .
we show that @xmath88 , and therefore that holds , by verifying that the following system of linear equations in the variables @xmath89 , @xmath90 @xmath91 has a unique solution @xmath92 with @xmath93 , for @xmath94 .
first of all , @xmath95 implies @xmath96 for at least an @xmath97 , otherwise all the evvs would lie on the same hyperplane , contradicting the linear independence of such vectors . this fact and imply that the coefficient matrix has rank @xmath98 and its unique solution can be obtained by deleting the @xmath99 equations corresponding to the rows not in @xmath76 .
denote each column of @xmath76 as @xmath100 , @xmath94 and denote as @xmath101 , the vector obtained from @xmath78 by selecting the same components as each @xmath102 by cramer s rule we have for each @xmath94 , @xmath103 @xmath104 since by either a determinant is null or it has the same sign of the other determinants .
if holds , then @xmath105 for every @xmath94 and holds .
conversely , each row of @xmath70 not in @xmath76 is a linear combination of the rows in @xmath76 .
therefore , each point of @xmath106 is identified by a vector @xmath107 whose components correspond to the rows in @xmath76 , while the other components are obtained by means of the same linear combinations that yield the rows of @xmath70 outside @xmath76 .
let @xmath108 denote the @xmath109 matrix obtained from the matrix @xmath76 without @xmath110 , @xmath111 .
consider a hyperplane @xmath112 in @xmath106 through the origin and @xmath113 evvs @xmath114 where @xmath115 . if @xmath116 , when we separate the subspace through @xmath112 , for all @xmath115 either the ray @xmath117 is coplanar to @xmath112 , i.e. , @xmath118 or @xmath119 and @xmath110 lie in the same half - space , i.e. , @xmath120 moving the first column to the @xmath121-th position in all the matrices above , we get . in case
holds , only the inequalities in are feasible and holds . to prove @xmath122 , consider , for any @xmath12 , the hyperplane that intersects the ray @xmath117 at the point @xmath123 , with @xmath124 since @xmath31 is convex , the intersection point is not internal to @xmath31 .
so , @xmath125 for @xmath126 , and , therefore , @xmath127 .
we get the lower bound for @xmath128 as solution in @xmath89 of the system . by cramer
s rule , @xmath129_{ij } } , \end{aligned}\ ] ] where @xmath130 is the @xmath85-th minor of @xmath76 .
the second equality derives by suitable exchanges of rows and columns in the denominator matrix : in fact , swapping the first rows and columns of the matrix leaves the determinant unaltered .
the last equality derives by dividing each row @xmath38 of @xmath76 by @xmath23 , @xmath94 .
finally , by we have @xmath131 .
the above result shows that whenever @xmath132 , then @xmath133 therefore , the corresponding evv @xmath134 is irrelevant to the formulation of the lower bound and can be discarded .
we will therefore keep only those evv that satisfy and will denote them as the _ supporting evvs _ for the lower bound . in the case
@xmath135 we have [ cor_mequalsn ] suppose that there are @xmath1 vectors @xmath136 where @xmath137 , @xmath12 , is the evv associated to @xmath64 , @xmath65 . if @xmath138 , @xmath139 and , for all @xmath12 @xmath140 where @xmath141 is the @xmath142 matrix obtained by replacing @xmath134 with @xmath78 in @xmath70 , then @xmath143_{ij } } \leq v^{\alpha } \leq \min_{i \in n } \cfrac{\sum_{j \in n } \beta_{ij } u_{ij}}{\sum_{j \in n } \beta_{ij } \alpha_j}\ ] ] where @xmath144_{ij}$ ] is the @xmath85-th element of @xmath145 .
we next consider two further corollaries that provide bounds in case only one evv is available .
the first one works with an evv associated to an arbitrary vector @xmath146 [ oneevv ] ( ( * ? ? ?
* proposition 3.4 ) ) let @xmath147 be finite measures and let @xmath148 be the evv corresponding to @xmath47 such that @xmath149 then , @xmath150 } \leq v^{\alpha } \leq \cfrac{\sum_{i \in n } \beta_i u_i}{\sum_{i \in n } \alpha_i u_i}.\ ] ] consider the corner points of the partition range @xmath151 where @xmath37 is placed on the @xmath38-th coordinate ( @xmath12 ) , and the matrix @xmath152 , where @xmath153 occupies the @xmath121-th position .
now @xmath154 and , for all @xmath155 , @xmath156 which is positive by .
therefore , @xmath70 satisfies the hypotheses of corollary [ cor_mequalsn ] .
since @xmath70 has inverse @xmath157 the following lower bound is guaranteed for @xmath128 : @xmath158 } .\ ] ] the upper bound is a direct consequence of theorem [ teo_main ] . in case all
measures @xmath26 , @xmath159 are normalized to one and the only evv considered is the one corresponding to @xmath160 , we obtain legut s result .
[ legut_result ] ( ( * ? ? ?
* theorem 3 ) ) let @xmath147 be probability measures and let @xmath148 be the evv corresponding to @xmath160 .
let @xmath115 be such that holds .
then , @xmath161 where @xmath162
. simply apply corollary [ oneevv ] with @xmath24 , for all @xmath12 and @xmath160 .
then @xmath163 where @xmath164 .
finally , by theorem [ teo_main ] we have @xmath165 it is important to notice that the lower bound provided by theorem [ teo_main ] certainly improves on legut s lower bound only when one of the evvs forming the matrix @xmath70 is the one associated to @xmath166 we consider a @xmath167 $ ] good that has to be divided among three agents with equal claims , @xmath168 and preferences given as density functions of probability measures @xmath169 \ ; , \ ] ] @xmath170 being the density function of a @xmath171 distribution .
the preferences of the players are not concentrated ( following definition 12.9 in barbanel @xcite ) and therefore there is only one evv associated to each @xmath172 ( cfr .
@xcite , theorem 12.12 ) .
agent 1 : tiny dashing ; agent 2 : large dashing ; agent 3 : continuous line.,scaledwidth=60.0% ] the evv corresponding to @xmath173 is @xmath174 . consequently , the bounds provided by legut are @xmath175 consider now two other evvs @xmath176 corresponding to @xmath177 and @xmath178 , respectively .
the matrix @xmath179 satisfies the hypotheses of theorem [ teo_main ] and the improved bounds are @xmath180
the bounds for @xmath128 depend on the choice of the evvs that satisfy the hypotheses of theorem [ teo_main ] . any new evv yields a new term in the upper bound .
since we consider the minimum of these terms , this addition is never harmful .
improving the lower bound is a more delicate task , since we should modify the set of supporting evvs for the lower bound , i.e. those evvs that include the ray @xmath117 in their convex hull . when we examine a new evv we should verify whether replacing an evv in the old set will bring to an improvement .
a brute force method would require us to verify whether conditions are verified with the new evv in place of @xmath181 only in this case we have a guarantee that the new evv will not make the bound worst .
then , we should verify again , with the new evv replacing one of the @xmath182 evvs , in order to find the evv from the old set to replace .
however , in the following proposition we propose a more efficient condition for improving the bounds , by which we simultaneously verify that the new evv belongs to the convex hull of the @xmath182 evvs and detect the vector to replace .
for any couple @xmath183 denote as @xmath184 the @xmath109 matrix obtained from @xmath76 by replacing column @xmath121 with @xmath185 and by deleting column @xmath186 .
[ teo_imp ] let @xmath187 be @xmath188 evvs with the last @xmath182 vectors satisfying conditions and .
if there exists @xmath189 such that @xmath190 then @xmath191 moreover , if all the inequalities in are strict then in both and the vectors belong to the relative interior of the respective cones .
before proving the individual statements , we sketch a geometric interpretation for condition . as in theorem [ teo_main ]
we restrict our analysis to the subspace @xmath106 . for any @xmath192 , the hyperplanes @xmath193 should separate @xmath194 and @xmath78 ( strictly if all the inequalities in are strict ) in the subspace @xmath106 . to prove ,
argue by contradiction and suppose @xmath195 then , for any @xmath196 there must exist a @xmath197 such that the hyperplane @xmath198 passes through all the evv ( including @xmath185 ) but @xmath194 and @xmath199 , and supports @xmath200 therefore , @xmath78 and @xmath194 belong to the same strict halfspace defined by @xmath198 , contradicting . to show the existence of such hyperplane consider the hyperplane @xmath201 in @xmath106 passing through @xmath81 and denote with @xmath202 the intersection of @xmath203 with such hyperplane .
restricting our attention to the points in @xmath201 , the vectors @xmath81 form a simplicial polyhedron with @xmath204 .
there must thus exist a @xmath197 such that the @xmath205-dimensional hyperplane @xmath206 in @xmath207 passing through @xmath208 and @xmath209 , supports @xmath210 and contains @xmath194 ( and @xmath78 ) in one of its strict halfspaces ( see appendix . )
if we now consider the hyperplane in @xmath106 passing through the origin and @xmath206 we obtain the required hyperplane @xmath211 to prove we need some preliminary results .
first of all , under , @xmath212 otherwise , @xmath185 would be coplanar to @xmath112 and any hyperplane @xmath198 , @xmath197 , would coincide with it . in such case
the separating conditions would not hold . moreover
, there must exist some other @xmath213 for which @xmath214 otherwise , @xmath185 would coincide with @xmath194 , making the result trivial .
we also derive an equivalent condition for .
let @xmath215 be the @xmath69 matrix obtained from @xmath76 by replacing vectors @xmath194 and @xmath199 by some other vectors , say @xmath216 and @xmath217 , respectively .
if we move the first column to the @xmath186-th position , becomes @xmath218 switching positions @xmath121 and @xmath186 in the second matrix we get @xmath219 and therefore @xmath220 from , and from part @xmath87 of the present theorem , we derive @xmath221 and therefore , yields @xmath222 condition and theorem [ teo_main ] allow us to conclude that @xmath223 . regarding the last statement of the theorem ,
we have already shown . moreover , if hold with a strict inequality sign for any @xmath197 , then @xmath224 for the same @xmath186 and @xmath225 .
similarly , would hold with strict inequality signs and @xmath226 .
if holds , we get not only that @xmath117 intersects the convex hull of the @xmath182 evvs @xmath227 , but also that the ray @xmath203 intersects the convex hull of the @xmath182 evvs @xmath228
. we can therefore replace @xmath194 with @xmath185 in the set of supporting evvs for the lower bound .
if the test fails for each @xmath189 , we discard @xmath185 we keep the current lower bound ( with its supporting evvs ) . in case holds with an equality sign for some @xmath186 , conditions and together imply @xmath229 .
therefore , we could discard @xmath199 from the set of supporting evvs for the lower bound .
we consider a list of 1000 random vectors in @xmath230 and , starting from the identity matrix , we iteratively pick each vector in the list . if this satisfies condition , then the matrix @xmath70 is updated .
the update occurs 9 times and the resulting evvs which generate the matrix @xmath70 are @xmath231 corresponding , respectively , to @xmath232 correspondingly , the bounds shrink to @xmath233 the previous example shows that updating the matrix @xmath70 of evvs through a random selection of the new candidates is rather inefficient , since it takes more than 100 new random vectors , on average , to find a valid replacement for vectors in @xmath70 .
a more efficient way method picks the candidate evvs through some accurate choice of the corresponding values of @xmath52 . in @xcite
a subgradient method is considered to find the value of @xmath128 up to any specified level of precision . in that algorithm ,
legut s lower bound is used , but this can be replaced by the lower bound suggested by theorem [ teo_main ] . considering the improved subgradient algorithm , we obtain the following sharper bounds @xmath234 after 27 iterations of the algorithm in which , at each repetion
, a new evv is considered .
the authors would like to thank vincenzo acciaro and paola cellini for their precious help .
the proof of in theorem [ teo_imp ] is based on the following lemma .
this is probably known and too trivial to appear in a published version of the present work .
however , we could not find an explicit reference to cite it .
therefore , we state and prove the result in this appendix consider @xmath1 affinely independent points @xmath235 in @xmath236 and @xmath237
. for each @xmath115 there must exist a @xmath197 such that the hyperplane passing through @xmath185 and @xmath209 supports @xmath210 and has @xmath194 in one of its strict halfspaces .
fix now @xmath115 and consider @xmath112 , the hyperplane passing through the points @xmath242
. also denote as @xmath243 the intersection between @xmath112 and the line joining @xmath185 and @xmath194 .
clearly @xmath244 . therefore , for any @xmath245
, the hyperplane @xmath246 in @xmath112 passing through @xmath247 will strictly separate @xmath199 and @xmath243 .
consequently , @xmath243 should simultaneously lie in the halfspace of @xmath246 not containing @xmath248 , and in the cone formed by the other hyperplanes @xmath249 , @xmath250 and not containing @xmath248 . a contradiction . | we provide a two - sided inequality for the @xmath0optimal partition value of a measurable space according to @xmath1 nonatomic finite measures .
the result extends and often improves legut ( 1988 ) since the bounds are obtained considering several partitions that maximize the weighted sum of the partition values with varying weights , instead of a single one . |
51 ophiuchi ( 51 oph ) , a rapidly rotating b9.5ve star located at @xmath3 pc @xcite , shows an infrared ( ir ) excess ( @xmath4% ) in its spectral energy distribution ( sed ) due to the presence of silicate grains @xcite .
this system probably represents a rare nearby example of a young debris disk with gas , a planetary system just entering the late stages of formation , after the primordial gas has substantially dissipated , but before terrestrial planets have finished accreting .
its nearly edge - on disk of gas and dust and its variable absorption features @xcite suggests that 51 oph may be an analog of @xmath2 pictoris ( @xmath2 pic ) .
several spectroscopic observations support the presence of an edge - on gaseous disk around 51 oph .
double - peaked h@xmath5
emission marks the presence of ionized hydrogen gas in keplerian orbit @xcite .
sharp co bandhead emission features rule out a spherically symmetric wind and reveal a gaseous disk with temperature ranging from 2000 4000 k interior to the dust sublimation radius @xcite .
the co bandhead observations also point to a disk with inclination @xmath6 @xcite or @xmath7 @xcite . a spectral line analysis performed by @xcite revealed a large projected rotational velocity for the star of @xmath8 km s@xmath9 .
gas absorption features observed by @xcite and @xcite are also consistent with an edge - on disk .
the spatial structure of the 51 oph dust disk remains puzzling .
an hst acs v - band non - detection @xcite and a keck 18 @xmath10 non - detection @xcite suggest that the disk is relatively compact . however , far - ir photometry reveals cold dust grains extending out to thousands of au @xcite .
@xcite obtained the first spatially - resolved measurements of the 51 oph disk with the mid - ir interferometric ( midi ) instrument on the very large telescope interferometer ( vlti ) .
the large visibility values they measured ( @xmath11 ) imply that the 51 oph disk is relatively compact along the vlti - midi projected baseline ( 101.2 m , 23@xmath12 e of n ) .
@xcite fit the vlti - midi visibility with a gaussian disk and found the fwhm to be 7 mas , or 0.9 au , at 12.5 @xmath10 . here
we present new spatially - resolved observations of 51 oph using the keck interferometer nuller that help to constrain the geometry of the dust disk .
we compare these to the vlti - midi observations @xcite and _ spitzer _ irs observations @xcite .
we simultaneously model all three data sets using two simple , edge - on disk models : an optically - thin model based on our zodiacal cloud and a two - layer model based on the @xcite disk model .
observations of 51 oph were conducted using the twin keck telescopes atop mauna kea , hawaii , operated in nulling interferometric mode on 2 apr 2007 .
51 oph was observed twice in the n - band ( 8 - 13 @xmath10 ) at an hour angle of @xmath13 , with a projected baseline of 66.2 m at a position angle of @xmath14 .
a calibrator star , epsilon ophiuchi ( hd146791 ) , was observed once following the target observations .
table [ observationstable ] lists the details of our observations .
the keck nuller operates with each keck telescope aperture divided into two sub - apertures for a total of four apertures ( see * ? ? ?
* ; * ? ? ?
* for details ) .
long - baseline fringes formed by combining the light from opposite telescopes are used to null the light from the central star and observe any spatially resolved structures .
short - baseline fringes formed by combining the light from two neighboring sub - apertures are used to remove the thermal background .
the observable quantity is the null leakage , " the ratio of the amplitude of the short baseline fringe with the long baseline null fringe on target to the amplitude of the short baseline fringe with the long baseline constructive fringe on target ( see * ? ? ?
* ; * ? ? ?
* for details ) .
we estimated and subtracted the systematic null leakage by comparing the measured null leakage of the calibration star , @xmath15 oph , with the expected null leakage for a limb - darkened star with the same diameter .
we estimated the diameter of @xmath15 oph as @xmath16 mas and adopted 1.5 mas error bars
much larger than the true size error bars as a simple means of enforcing a systematic noise floor based on our estimate of the instrument performance .
figure [ kecknulldatafig ] presents the calibrated null leakage for 51 oph .
we combined the data from both observations , which had nearly identical projected baselines .
we limited our analyses to the 8 11.6 @xmath10 range since noise from decreased instrument throughput rendered data beyond 11.6 @xmath10 unusable . for wavelengths less than 11.6 @xmath10 , the null leakage remains relatively flat with a @xmath171@xmath18 rise near 9.8 @xmath10 .
we first modeled the null leakage at two different wavelengths with uniform disk and gaussian surface brightness profiles .
we found angular diameters of @xmath19 mas and @xmath20 mas at 8 and 10 @xmath10 , respectively , for the uniform disk profile .
for a gaussian surface brightness profile , we found fwhm of @xmath21 mas and @xmath22 mas at 8 and 10 @xmath10 , respectively . at a distance of 131 pc , 10
mas corresponds to a transverse distance of 1.3 au , suggesting that the disk is either truncated at a close circumstellar distance , or the axis of rotation of the near edge - on disk is oriented within a few tens of degrees of the projected keck baseline . to better understand the geometry of the 51 oph system
, we examined our keck nuller observations together with the observations of 51 oph made with vlti - midi and _ spitzer_. figure [ bestfit2comp_noshortlambda_figure ] shows a collection of three independent data sets from observations of 51 oph : the _ spitzer _ irs spectrum @xcite in the top panel , our n - band keck nuller null leakage in the middle panel , and the n - band vlti - midi visibility data @xcite in the bottom panel .
the vlti - midi visibility was obtained on 15 june 2003 with a projected baseline of 101.2 m at a position angle of @xmath23 @xcite .
figure [ uvcoverage_figure ] shows how incorporating this data set improves the @xmath24 coverage of our analysis .
although the vlti - midi baseline was oriented within @xmath25 of the keck baseline , modeled uniform disk and gaussian surface - brightness profile sizes are approximately 35% smaller for vlti - midi measurements than for keck nuller measurements .
when we modeled the vlti - midi measurements using a uniform surface - brightness disk , we found best - fit angular diameters of @xmath26 mas at 8 @xmath10 and @xmath27 mas at 10 @xmath10 , consistent with @xcite .
for a gaussian model , we found a fwhm of @xmath28 mas at 8 @xmath10 and a fwhm of @xmath29 mas at 10 @xmath10 .
the middle panel in figure [ bestfit2comp_noshortlambda_figure ] shows a 1@xmath18 rise at 9.8 @xmath10 in the keck null leakage .
the vlti - midi visibility contains a 1@xmath18 dip at 9.7 @xmath10 .
these features mirror the 10 @xmath10 silicate emission feature shown in the _ spitzer _ irs spectrum ( section [ spitzer_data ] ) and suggest that 51 oph is more extended near 10 @xmath10 . the mid - infrared spectrum , shown in the top panel of figure [ bestfit2comp_noshortlambda_figure ] , was obtained on 22 march 2004 using the infrared spectrograph ( irs ) on the _ spitzer space telescope _ @xcite .
_ spitzer _ observed 51 oph in staring mode from 5 to 36 @xmath10 using the short - low ( sl ) module from 5 to 14 @xmath10 , the short - high ( sh ) module from 10 to 19 @xmath10 , and the long - high ( lh ) module from 19 to 36 @xmath10 .
sl has a resolving power of r@xmath30128 , while sh and lh have r@xmath17600 .
51 oph was observed using three slit positions stepped across the nominal position of the source .
the spectral extraction and calibration methods are described in @xcite .
figure [ bestfit2comp_noshortlambda_figure ] shows that the spectrum exhibits a pronounced 10 @xmath10 silicate feature and a small 18 @xmath10 silicate feature .
emission from polycyclic aromatic hydrocarbon ( pah ) molecules contributes to the mid - ir spectra of many herbig ae / be stars .
we considered the possibility that this emission could contribute to our data on 51 oph .
@xcite included 51 oph in a study that looked for correlations between disk structure and mid - ir pah emission from intermediate - mass young stars .
they found no measurable emission from pah features in the 613 @xmath31 m range in the mid - ir spectrum of 51 oph .
there have been several previous attempts to model the 51 oph ir excess .
@xcite fit iras photometric data with an optically thin , spherically symmetric dust shell model and found a best fit model with a dust density proportional to @xmath32 and dust temperatures ranging from @xmath33 k to @xmath34 k. @xcite modeled photometric data from the infrared telescope facility ( irtf ) and estimated that the ir excess could be attributed to astronomical silicates smaller than 5 @xmath10 , ranging in temperature from 400 k to 1000 k. @xcite compared an optically - thick herbig ae disk model with a puffed inner rim @xcite to the 51 oph midi visibility in the 8 14 @xmath10 range and ir iso spectra and found that such a model fit poorly .
we developed new models for 51 oph to incorporate the new data from _
spitzer _ and the keck nuller .
we do not model the detailed mineralogy of the 51 oph disk and do not intend for our models to explain all of the observed spectral features to high numerical accuracy .
we focus on the 3-d dust density distribution and disk models that qualitatively describe all three data sets .
we adopted the following fitting procedure to simultaneously fit all three data sets : 1 .
model the disk contribution to the _ spitzer _ irs spectrum to obtain the radial distribution of grains , the grain size , and the surface density for each model component .
2 . using the parameters determined by the best - fit _ spitzer _ irs spectrum , fit the interferometric data using 3-d models of the dust distribution to obtain the scale height of each component and the disk position angle . to model the disk contribution to the _ spitzer _ irs spectrum , we first calculated the ir excess .
we modeled the stellar contribution to the _ spitzer _ irs spectrum as an ideal blackbody with an effective temperature of 10,000 k and a luminosity of 260 @xmath35 at 131 pc @xcite .
the stellar continuum contributes on the order of a few percent in the n - band , so any uncertainties in the stellar luminosity or temperature are too small to significantly impact the interferometric or spectral responses of our models .
for example , a @xmath36 increase in the assumed luminosity would only raise the stellar fractional flux contribution from @xmath37 to @xmath38 at 10 @xmath10 .
the 51 oph ir excess exhibits a sharp increase in flux at wavelengths shorter than 7.5 @xmath10 that resembles a continuum source
. however , this feature may not be continuum .
using iso - sws spectra with a spectral resolution of @xmath39 , @xcite showed this region exhibits emission features from hot circumstellar gas species , including @xmath40 , no , co , and @xmath41 . in light of this contamination , we chose to ignore the spectrum at wavelengths shorter than 7.5 @xmath10 while we were fitting our dust cloud models .
we also resampled the _
irs data to a resolution of @xmath42 to conserve computer time . to create 3-d optically - thin disk models
, we used the zodipic software package @xcite .
this software ( available online at http://asd.gsfc.nasa.gov/marc.kuchner/home.html ) synthesizes images of optically thin dust disks based on the @xcite empirical model of the solar zodiacal cloud as observed by cobe dirbe .
the grain size and inner and outer radius of each modeled disk component were determined by the best fit to the _ spitzer _ irs spectrum . for our first models we used a single component of silicate dust grains of size @xmath43 distributed from an inner radius , @xmath44 , to an outer radius , @xmath45 .
we assumed a density distribution similar to that of the zodiacal cloud @xcite ; the surface density @xmath46 .
we numerically calculated the temperature as a function of circumstellar distance based on the stellar spectrum and dust optical constants , accounting for the heating of small grains above the local blackbody temperature .
we examined 100 grain sizes , @xmath43 , ranging from 0.05 2.5 @xmath10 and used the astronomical silicate emissivities from @xcite .
we used a non - linear least squares fitting routine to determine the best fit inner and outer disk radii for each grain size .
the best - fit single - component disk model , shown in figure [ bestfit_singlecomponent_figure ] , consists of 1.0 @xmath10 grains distributed from the grain sublimation radius ( @xmath47 au ) to 189 au .
as you might imagine , this single component does a poor job of fitting the complexities of the _ spitzer _ irs spectrum .
we found that a single - component model can not adequately reproduce the relative 10 @xmath31 m to 18 @xmath10 flux ratio , the width of the 10 @xmath10 feature , and the flux in the 13 15 @xmath10 range .
an additional disk component , either of different composition , grain size , or location , appears necessary to fit the _
irs spectrum .
therefore , we examined several models consisting of two optically - thin components : an outer disk of small dust grains ( @xmath48 ) , which contribute the 10 @xmath10 silicate emission feature , and an inner disk of large blackbody grains .
our optically - thin models are defined by 10 adjustable parameters : inner and outer radii of the outer disk ( @xmath49 , @xmath50 ) , inner and outer radii of the inner disk ( @xmath51 , @xmath52 ) , grain size in the outer disk ( @xmath53 ) , scale height of the outer disk ( @xmath54 ) , scale height of the inner disk ( @xmath55 ) , surface density scaling factor of the outer disk ( @xmath56 ) , surface density scaling factor of the inner disk ( @xmath57 ) , and a common position angle for both disk components ( pa ) .
we assumed a fixed inner disk grain size , @xmath58 , a fixed dust density radial distribution , @xmath59 ( see below ) , and a fixed disk inclination , @xmath60 .
first , we fit our two - component model to our resampled _ spitzer _
irs spectrum .
we examined 20 values for @xmath61 ranging from @xmath62 @xmath63 and explored a wide range of inner and outer disk radii ranging from the dust sublimation radius for silicate grains ( 0.65 au for 1 @xmath10 grains ) to thousands of au . generally , for the models that best fit the spectrum , the large grain component stretched from @xmath64 au out to @xmath65 au and the small grain component stretched from a few au out to @xmath34 au .
the top panel in figure [ bestfit2comp_noshortlambda_figure ] shows the total flux of the best fit model , which we refer to as model a. " it also shows the contribution of each component of this model to the total flux .
as shown in figure [ bestfit2comp_noshortlambda_figure ] , model a qualitatively reproduces all of the major features in the _ spitzer _ irs spectrum .
we used the parameters that best fit the _ spitzer _
irs spectrum to create 3-d optically - thin disk models using zodipic .
we assumed a fixed disk inclination of @xmath66 .
we examined 237 values of disk scale heights ranging from @xmath67 to @xmath68 for both the inner and outer disks .
we examined position angles from 0 to 180@xmath12 in 1@xmath12 increments .
we calculated the keck nuller null leakage and vlti - midi visibility for all combinations of disk scale heights and position angle to find the best - fit model a , the parameters for which are listed in table [ bestfitstable ] . to calculate the keck nuller null leakage and vlti - midi visibility , we used a software suite that we designed to model these instruments
( see * ? ? ?
we confirmed that the results from our software suite agreed with the visibility modeling tool , a tool developed by the nexsci for simulating kin data ( http://nexsciweb.ipac.caltech.edu/vmt/vmtweb/ ) .
the interferometric responses of model a are shown in the middle and bottom panels of figure [ bestfit2comp_noshortlambda_figure ] .
the best simultaneous fit to the keck nuller null leakage and vlti - midi visibility , shown in black , illustrates that model a does not satisfactorily reproduce both interferometric responses simultaneously .
the keck null leakage is well - fit , but the vlti - midi visibility is underestimated by a factor of @xmath69 , indicating that this dust disk model is too extended along the vlti - midi baseline .
figure [ bestfit2comp_noshortlambda_figure ] also shows in gray the response corresponding to the best fit to the midi data alone , which does not produce a satisfactory keck null leakage .
figure [ response_vs_pa_figure ] explains the source of the problem .
the top panel in figure [ response_vs_pa_figure ] shows the keck and vlti - midi 10 @xmath10 responses for model a as a function of disk position angle .
the vlti - midi response of model a ( gray diamonds ) crosses the measured vlti - midi value ( gray line ) at a position angle of @xmath70 .
the keck nuller response of model a ( black triangles ) crosses the measured keck null leakage ( black line ) at a position angle of @xmath71 . to fit both data sets , the 10 @xmath10 responses for the model would need to intersect their respective measured values at a single position angle .
the top panel in this figure clearly shows that this does not happen ; there is no single position angle for the model that works for both the keck and vlti - midi data .
figure [ response_vs_pa_figure ] also reveals how a new model can remedy the problem .
the position angles of the maximum in the vlti - midi response and the minimum in the keck response are fixed and correspond to the alignment of the instrument s projected baseline with the disk axis of rotation .
so to improve our fit to the interferometric data , we must change the model so that we broaden the widths of the maximum in the vlti - midi response and minimum in the keck response ; we must make the model more azimuthally symmetric .
we attempted to accomplish this improvement manually by increasing the scale height of the disk models .
however , making this change alone reduces the maximum vlti - midi visibility of the model until it no longer reaches the measured value .
so to compensate , we also reduced the outer radius of the inner disk ; we forced the inner and outer radii of the inner disk to 0.54 au and 4 au , respectively , and re - ran the fitting procedure .
we call the resulting best - fit model b. " the bottom panel in figure [ response_vs_pa_figure ] shows that the 10 @xmath10 interferometric responses of model b cross the measured values approximately simultaneously at a single position angle of 122@xmath12 , indicated with a dotted vertical line in the figure . although model b does not fit the _ spitzer _
irs spectrum as well as model a , it still qualitatively reproduces the spectrum s major features , as shown in the top panel in figure [ bestfit2comp_noshortlambda_figure_7 ] .
the middle and bottom panels of figure [ bestfit2comp_noshortlambda_figure_7 ] show that these changes significantly improve the fit to the interferometric data ; model b is consistent with both the keck null leakage and vlti - midi visibility .
figure [ image_figure ] shows a simulated image of our model b circumstellar disk at 10 @xmath10 with 1 mas pixel resolution . the inner disk of large dust grains , truncated at 30.5 mas ( 4 au ) , is the brightest feature of our model .
the outer disk extends from 53 mas to @xmath179200 mas , beyond the range of the figure .
our models imply that 51 oph hosts a cloud of small grains located at hundreds to thousands of au from the star , supporting the models of @xcite and @xcite .
our models are consistent with limits placed on the disk flux by previous non - detections .
an hst acs non - detection at v - band using the @xmath72 occulting spot limits the disk surface brightness to @xmath73 jy arcsec@xmath74 between @xmath75 and @xmath76 @xcite and a keck 18 @xmath10 nondetection places an upper limit on the surface brightness at @xmath77 of 2% of the peak flux @xcite . our model
a , which extends to @xmath1 au , has a mean v - band surface brightness of @xmath78 jy arcsec@xmath74 between @xmath75 and @xmath79 and a mean 18 @xmath10 surface brightness at @xmath77 of 0.2% of the peak flux , well within the acs and keck non - detection limits .
our model b has a mean v - band surface brightness of @xmath80 jy arcsec@xmath74 between @xmath75 and @xmath76 and a mean 18 @xmath10 surface brightness at @xmath77 of 0.4% of the peak flux , also well within the non - detection limits .
we also compared a 10.6 @xmath10 model image against recent diffraction - limited keck imaging using segment tilting interferometry on a single telescope @xcite .
our model appears consistent with these data , which indicate the observed characteristic emission scale to be @xmath8130 mas , with @xmath8295% of emission arising within an aperture of 1.5 " .
@xcite observed the inner portions of the 51 oph disk at k - band with vlti - amber .
@xcite found the best fit to the 2.2 micron continuum visibility measurements using a narrow ring 0.24 au in radius , well within the dust sublimation radius , and noted that this continuum may result from hot dust interior to the sublimation radius which is shielded from stellar radiation or from free - free emission from an inner gas disk @xcite .
we calculated the k - band visibility of our model b for the three baseline orientations used by @xcite and checked these calculations with the visibility modeling tool provided by nexsci .
we calculated k - band visibilities of 0.99 , 0.74 , and 0.74 for baselines of 55 m at 34@xmath12 , 82 m at 91@xmath12 , and 121 m at 69@xmath12 , respectively .
these values are close to the measured visibilities of [email protected] , [email protected] , and [email protected] ; the latter two modeled visibilities fall just below the lower limit of the measured visibilities given the reported uncertainties .
because the keck nuller and vlti - midi do not resolve these extreme inner portions of the disk , the k - band observations may probe a different structure than what the n - band observations probe .
@xcite use photometric data to estimate that the excess continuum contributes 25% of the total flux at 2.2 microns . by estimating the stellar flux at 2.2 microns and modeling the continuum flux source as a blackbody at 1500 k ( dust at the sublimation temperature )
, we calculate that the continuum source s contribution at n - band would be no larger than 1.9 jy .
this contribution is a factor of @xmath175 less than the flux from the blackbody component of our best fit two - component optically thin disk models .
this disparity suggests that an additional blackbody component is necessary beyond what @xcite model . additionally ,
our model a and model b blackbody components contribute only 1.4 jy and 3.4 jy of flux at 2.2 microns , respectively , less than the 3.6 jy of excess continuum flux calculated by @xcite .
perhaps a three - component model , with an outer disk of small silicate grains , an inner disk of large grains , and a continuum source interior to the dust sublimation radius ( shielded hot dust or free - free emission from a gas disk ) , will be necessary to explain all of the interferometric and spectral data sets .
motivated by the @xcite and @xcite circumstellar disk models , we explored an alternative edge - on disk geometry consisting of a blackbody midplane layer sandwiched between two identical optically - thin surface layers of small dust grains .
the surface layers are thin ; they have a very small scale height and are unresolved in the direction perpendicular to the midplane .
our models were completely defined by 6 parameters : the disk position angle , the temperature and scale height of the middle layer , and the temperature , grain size , and line density of the surface layers .
we investigated 100 middle - layer and 100 surface - layer temperatures ranging from 300 1300 k , and 6 surface layer dust sizes from 0.1 1.0 @xmath10 .
we used the surface - layer temperature , which represents the temperature of the surface layer at the outer truncation radius of the disk , to calculate the outer radius of the disk . because our disk is edge - on
, we ignore any contribution by a hot inner wall .
we note that @xcite fit the vlti - midi data with a @xcite flared disk model which included a hot inner wall and found this model to fit poorly .
we first derived the total emitting area of each component , and therefore the line density of the surface layers and scale height of the middle layer , by fitting the flux of the two disk components to the _ spitzer _ observations .
for the 500 best fits to the _ spitzer _ irs spectrum , we calculated the keck null leakage and vlti - midi visibility as a function of position angle in @xmath84 increments .
figure [ bestfit3layer_noshortlambda_figure ] shows the single best fit to the three data sets .
the best fit two - layer model qualitatively reproduces the major features of the _ spitzer _ irs spectrum , shown in the top panel of figure [ bestfit3layer_noshortlambda_figure ] , although it underestimates the flux near 8 @xmath10 and overestimates the flux from 1018 @xmath10 .
the lower two panels of figure [ bestfit3layer_noshortlambda_figure ] shows this model can reproduce the mean keck null leakage , but can not simultaneously reproduce the vlti - midi visibility .
the interferometric data seems to rule out this disk geometry .
our modeling efforts have yielded an optically - thin disk model that can qualitatively reproduce our 51 oph observations with the keck nuller and also the vlti - midi and _ spitzer _ observations of this system .
this model , model b , is composed of an inner ring of large grains distributed from their sublimation radius ( @xmath85 au ) out to @xmath0 au , and an outer ring of small grains ( @xmath86 ) distributed from an inner radius of @xmath177 au to an outer radius of @xmath171200 au .
we also experimented with optically - thick models , but found them unsatisfactory .
our preferred optically - thin disk model could be interpreted as an inner birth " disk of material producing small dust grains through collisions .
grains produced in the inner ring that are larger than the blowout size either spiral inward under poynting - robertson drag or are collisionally fragmented .
grains smaller than the blowout size exit the system on hyperbolic orbits in a dynamical time ; the ejected grains and marginally ejected grains correspond to the outer disk component .
the blowout size for blackbody dust grains in a disk without gas is given by @xmath87 where @xmath88 and @xmath89 are the luminosity and mass of the star , respectively , @xmath90 is the gravitational constant , @xmath91 is the speed of light , and @xmath92 is the dust density @xcite .
assuming @xmath93 , @xmath94 , and @xmath95 gm @xmath96 , the blowout size for 51 oph is approximately 50 @xmath10 , consistent with our preferred model . a similar pattern ,
a central population of large grains and an outer ring of small grains , has been seen in observations of the debris disks around @xmath2 pictoris @xcite , vega @xcite , and au microscopii @xcite .
of course the disk around 51 oph is not gas free ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , so our scenario of an inner birth ring and outer disk of ejected grains requires that the gas disk is sparse enough to allow the small grains to be ejected from the system .
here we estimate the maximum tolerable gas density for which small grains are unbound in the 51 oph system .
the stopping time of a dust grain undergoing epstein drag forces is given by @xmath97 where @xmath98 is the gas density and @xmath99 is the sound speed ( see , e.g. * ? ? ?
the dynamical timescale is given by @xmath100 where @xmath101 is the semi - major axis of the grain .
for the dust grains to be significantly affected by gas drag , @xmath102 . substituting @xmath103 , where @xmath104 is the boltzmann constant , @xmath105 is the gas temperature , and @xmath106 is the mass of an @xmath107 molecule , we find that entraining a 1 @xmath10 grain in the gas would require @xmath108 assuming a gas temperature of 1400 k at a circumstellar distance of 5 au , near the vicinity of the inner disk , the gas density required to stop a 1 @xmath10 grain is @xmath109 @xmath96 .
we conclude that if the disk around 51 oph is comprised of an inner birth ring and an outer disk of ejected grains , then the gas density in the 51 oph disk must be less than @xmath17100 times the gas density of @xmath2 pic @xcite at a circumstellar distance of @xmath175 au .
the outer disk in model b has a scale height at least @xmath174 times larger than the scale height of the inner disk .
so the small grains at high orbital inclinations in the outer disk seem unlikely to have originated from the thin inner disk in our simple model .
such small grains can not easily be perturbed to high - inclination orbits by hidden planets either , because of their short lifetimes compared to secular time scales .
the difference in scale heights between the two components of our model suggests to us that 51 oph may have an unobserved sub - disk of larger bodies on inclined orbits in the inner regions of the disk .
the limited @xmath110 coverage and resolution of the keck and vlti - midi observations could easily keep such a sub - disk hidden from our observations .
this sub - disk might be analogous to the x - pattern , or warp " imaged by @xcite in the @xmath2 pictoris disk .
the large bodies in this sub - disk could launch the small grains onto inclined orbits , just as @xcite suggested occurs in the @xmath2 pictoris disk . here
we discuss two possibilities that might complicate our interpretation of the mid - ir interferometry of 51 oph : a ) the disk composition or structure is more complex than can be represented by our models , or b ) the disk has changed during the four year interval between vlti - midi and keck observations .
the chemical composition of the 51 oph dust disk is not well - known .
our models assumed astronomical silicates only , and ignored the possibility of a more complex composition .
a detailed model of the 51 oph disk composition , along the lines of @xcite for example , which we leave for future work , would help to further refine our models .
the disk geometry could also be more complex than our models can capture .
as previously discussed , there are a number of observations which suggest that an inner circumstellar disk is near edge - on in the 51 oph system .
our optically - thin models of the dust disk assume an outer disk which is coplanar , but the 51 oph disk inclination may be more complex .
some debris disks show warps or sub - disks at different inclinations ( e.g. * ? ? ?
it is possible that our assumption of coplanar disks does not allow model a , which best fits the _ spitzer _ irs spectrum , to simultaneously fit the vlti - midi visibility and keck null leakage .
the 51 oph disk geometry may also have changed between observations .
the blowout time for small grains originating from a belt of material at 1 au is on the order of a single dynamical time ( only a few years ) .
so the outer disk could evolve significantly within the four years between keck and vlti - midi observations via a recent collisional event ( e.g. * ? ? ? * ) .
additionally , 51 oph s inner disk could feature complex resonant structures , such as clumps or rings due to the presence of planets ( e.g. * ? ? ?
* ) , which rotate in and out of view on a dynamical timescale .
a comparison of the _ spitzer _ irs spectrum shown in figure [ bestfit2comp_noshortlambda_figure ] and the iso spectrum published in @xcite reveals that the 51 oph disk may have changed significantly in the 8 years between spectral observations . the silicate emission feature in the _ spitzer _ irs spectrum peaks at 10 @xmath10 , whereas the iso spectrum peaks at 10.5 to 11 @xmath10 .
the slope of the right side of the silicate emission feature is also noticeably steeper in the _ spitzer _ irs spectrum ; estimated slopes of the right side of the 10 @xmath10 silicate emission feature are @xmath111 jy @xmath112 and @xmath113 jy @xmath112 for iso and _ spitzer _ observations , respectively .
these differences indicate that the 51 oph disk may have had larger grains at the time of the iso observations .
finally , the flux longward of 20 @xmath10 in the iso spectrum is @xmath69 jy less than the spitzer irs spectrum , possibly because there were fewer cold grains at the time of the iso observations .
we observed the 51 oph disk at n - band using the keck interferometer operating in nulling mode . we combined the observed keck null leakage with vlti - midi visibility data and the _ spitzer _ irs spectrum and simultaneously modeled all three data sets .
we experimented with a variety of optically - thin dust cloud models and also an edge - on optically - thick disk model .
the _ spitzer _
irs spectrum ruled out the single - component optically - thin model , while the interferometric data ruled out our optically - thick model .
our preferred model consists of two separate populations of large and small grains .
the three data sets are best simultaneously fit by our model b ( table [ bestfitstable ] ) .
this model , shown in figure [ bestfit2comp_noshortlambda_figure_7 ] , contains a disk of larger grains that extends from the grain sublimation radius out to @xmath0 au and a disk of @xmath114 grains that extends from @xmath115 au to @xmath1 au .
this model may be consistent with an inner birth " disk of continually colliding parent bodies .
the large grains ( @xmath116 50 @xmath10 ) produced by the parent bodies make up the inner disk , while the small grains ( @xmath117 50 @xmath10 ) are blown outward and eventually ejected from the system by radiation pressure .
the large scale height of the outer disk compared to the inner disk suggests that the small grains which compose the outer disk may originate from an unseen inclined sub - disk or from a population of inclined cometary bodies .
although the 51 oph disk seems puzzling at first , perhaps it is not so strange after all .
the distribution of grain sizes in our models is not unique to the 51 oph system , but has been observed in the @xmath2 pictoris @xcite , vega @xcite , and au microscopii @xcite disks .
our models suggest that the 51 oph disk may be another member of a class of debris disks which exhibit similar dust distributions .
our models also indicate that there may be two sources of dust at different inclinations around 51 oph .
these models , together with previous observations of variable absorption features @xcite which may be due to transient infalling bodies suggest that the 51 oph dust disk may well be an example of a @xmath2 pictoris - like system .
we thank the national aeronautics and space administration and goddard space flight center for support of this research through funding from the graduate student researchers program , and the harvard - smithsonian center for astrophysics and nasa s navigator program for their financial support via the keck interferometer nuller shared risk science program .
part of this research was carried out at the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
the authors also thank rachel akeson and rafael millan - gabet for their help in making these observations possible . lccccccccccc model
a & @xmath118 & @xmath119 & @xmath120 & 0.03 & 1.15@xmath12110@xmath122 & 100 & @xmath123 & @xmath124 & @xmath125 & 2.49@xmath12610@xmath122 & @xmath127 + model b & 0.1 & @xmath128 & @xmath120 & 0.19 & 2.05@xmath12910@xmath122 & 100 & 0.54 & 4.0 & 0.04 & 4.3@xmath13010@xmath122 & @xmath131 + .
the best fit to the vlti - midi data alone is shown in gray and corresponds to a disk with a position angle of @xmath132 , an outer disk scale height of @xmath133 , and an inner disk scale height of @xmath134.[bestfit2comp_noshortlambda_figure ] ] coverage of the observations discussed in this paper . the outer boundary of the _ spitzer _ @xmath24 coverage is shown as a circle , corresponding to the _ spitzer _ mirror diameter at 8 microns.[uvcoverage_figure ] ] as a function of position angle for the keck nuller ( black triangles ) and vlti - midi ( gray diamonds ) for model a. the solid black and gray lines show the measured 10 @xmath10 responses for keck and vlti - midi , respectively , along with shaded regions representing their respective measurement uncertainties .
the modeled keck and vlti - midi responses do not cross their measured values together at any one position angle .
_ bottom panel : _ modeled interferometric response at 10 @xmath10 for model b. the modeled responses cross the corresponding measured values at approximately 122@xmath12.[response_vs_pa_figure ] ] image of model b in units of flux per pixel , with a pixel size of 1 mas .
the disk midplane is oriented at 122@xmath12 east of north .
the parameters of model b are listed in table [ bestfitstable ] .
[ image_figure ] ] and truncated at outer radius 3.4 au .
the disk features a middle layer of 360 k blackbody grains with a disk height of 0.163 au from the midplane , and 2 surface layers of 0.7 @xmath10 grains heated to 835 k , each with line densities of @xmath135 .
[ bestfit3layer_noshortlambda_figure ] ] | we present observations of the 51 ophiuchi circumstellar disk made with the keck interferometer operating in nulling mode at n - band .
we model these data simultaneously with vlti - midi visibility data and a _ spitzer _
irs spectrum using a variety of optically - thin dust cloud models and an edge - on optically - thick disk model .
we find that single - component optically - thin disk models and optically - thick disk models are inadequate to reproduce the observations , but an optically - thin two - component disk model can reproduce all of the major spectral and interferometric features .
our preferred disk model consists of an inner disk of blackbody grains extending to @xmath0 au and an outer disk of small silicate grains extending out to @xmath1 au .
our model is consistent with an inner birth " disk of continually colliding parent bodies producing an extended envelope of ejected small grains .
this picture resembles the disks around vega , au microscopii , and @xmath2 pictoris , supporting the idea that 51 ophiuchius may be a @xmath2 pictoris analog . |
double parton scattering ( dps ) means that two short - distance subprocesses occur in a given hadronic interaction , with two initial partons being active from each of the incident protons in a collision at the large hadron collider ( lhc ) .
the concept is shown for illustrative purposes in fig . [
fig : feyn - diag ] , and it may be contrasted with conventional single parton scattering ( sps ) in which one short - distance subprocess occurs , with one parton active from each initial hadron .
since the probability of single parton scattering is itself small , it is often expected that the chances are considerably suppressed for two or more short - distance interactions in a given collision .
however , expectations such as these bear quantitative re - examination at the lhc where the high overall center - of - mass energy provides access to very small values of the fractional momentum @xmath2 carried by partons , a region in which parton densities grow rapidly .
a large contribution from double parton scattering could result in a larger than otherwise anticipated rate for multi - jet production and produce significant backgrounds in searches for signals of new phenomena .
the high energy of the lhc also provides an increased dynamic range of available phase space for detailed investigations of dps . and
@xmath3 from one proton and @xmath4 and @xmath5 from the second proton .
the two hard scattering subprocess are @xmath6 and @xmath7 .
[ fig : feyn - diag ] ] investigations of double parton scattering have a long history theoretically @xcite , and there is evidence for their presence in collider data from the cern intersecting storage rings @xcite and fermilab tevatron @xcite . a significantly greater role for double - parton processes
may be expected at the lhc where higher luminosities are anticipated along with the higher collision energies .
of substantial importance is to know empirically how large the double parton contribution may be and its dependence on relevant kinematic variables .
our aim is to calculate characteristic final states at lhc energies in which it may be straightforward to discern a double parton signal .
we show in this paper that double parton scattering produces an enhancement of events in regions of phase space in which the `` background '' from single parton scattering is relatively small .
if such enhancements are observed experimentally , with the kinematic dependence we predict , then we will have a direct empirical means to measure the size of the double parton contribution .
in addition to its role in general lhc phenomenology , this measurement will have an impact on the development of partonic models of hadrons , since the effective cross section for double parton scattering measures the size in impact parameter space of the incident hadron s partonic hard core . from the perspective of sensible rates and experimental tagging ,
a good process to examine should be the 4 parton final state in which there are @xmath8 hadronic jets plus a @xmath9 quark and a @xmath10 antiquark , _ viz . _ @xmath11 .
if the final state arises from double parton scattering , then it is plausible that one subprocess produces the @xmath12 system and another subprocess produces the two jets .
there are , of course , many single parton scattering ( 2 to 4 parton ) subprocesses that can result in the @xmath11 final state , and we look for kinematic distributions that show notable separations of the two contributions .
as we show , the correlations in the final state are predicted to be quite different between the double parton and the single parton subprocesses .
for example , the plane in which the @xmath12 pair resides is uncorrelated with the @xmath13 plane in double parton scattering , but not in the single parton case .
the state - of - the - art of calculations of single parton scattering is well developed whereas the phenomenology of double parton scattering is as yet much less advanced . in the remainder of this introduction ,
we first describe the approach we adopt for the calculation of double parton scattering , specializing to the proton - proton situation of the lhc .
then we outline the paper and summarize our main results .
our calculations are done at leading - order in perturbative qcd , adequate for the points we are trying to make . making the usual factorization assumption
, we express the single - parton hard - scattering differential cross section for @xmath14 as @xmath15 indices @xmath16 and @xmath4 run over the different parton species in each of the incident protons .
the parton - level subprocess cross sections @xmath17 are functions of the fractional partonic longitudinal momenta @xmath18 and @xmath19 from each of the incident hadrons and of the partonic factorization / renormalization scale @xmath20 .
the parton distribution functions @xmath21 express the probability that parton @xmath16 is found with fractional longitudinal momentum @xmath18 at scale @xmath20 in the proton ; they are integrated over the intrinsic transverse momentum ( equivalently , impact parameter ) carried by the parton in the parent hadron .
a formal theoretical treatment of double parton scattering would begin with a discussion of the hadronic matrix element of four field operators and an explicit operator definition of two - parton correlation functions .
this procedure would lead to a decomposition of the hadronic matrix element into non - perturbative two - parton distribution functions and the corresponding hard partonic cross sections for @xmath22 .
an operator definition of two - parton correlation functions may be found in ref .
@xcite where the two - parton correlation function is reduced to a product of single parton distributions .
an explicit operator definition of two - parton distributions with different values of the two fractional momenta @xmath18 and @xmath23 is presented in ref .
@xcite , along with a model for the two - parton distributions in terms of normal parton distributions . in this paper
, we follow a phenomenological approach along lines similar to refs .
@xcite . in a double parton process ,
partons @xmath16 and @xmath3 are both active in a given incident proton .
we require the joint probability that parton @xmath3 carries fractional momentum @xmath23 , given that parton @xmath16 carries fractional momentum @xmath18 .
in general , this joint probability @xmath24 should also depend on the intrinsic transverse momenta @xmath25 and @xmath26 of the two partons ( or , equivalently , their impact parameters ) .
the hard scales @xmath27 and @xmath28 are characteristic of the two hard subprocesses in which partons @xmath16 and @xmath3 participate . in the sections below , we discuss the choice we make of the hard - scale and do not explore in this paper theoretical uncertainties associated with higher - order perturbative contributions .
in contrast to single parton distributions functions @xmath21 for which global analyses have produced detailed information , very little is known phenomenologically about the magnitude and functional dependences of joint probabilities @xmath24 . a common assumption made in estimates of double parton rates is to ignore possibly strong correlations in longitudinal momentum and to use the approximation @xmath29 for reasons of energy - momentum conservation , if not dynamics , the simple factorized form of eq .
( [ eq : approx1 ] ) can not be true for all values of the fractional momenta @xmath2 .
the values of @xmath23 available to the second interaction are always limited by the values of @xmath18 in the initial interaction since @xmath30 .
the approximation certainly fails even at the kinematic level if both partons carry a substantial fraction of the momentum of the parent hadron .
however , it may be adequate for applications in which the values of @xmath18 and @xmath23 are small .
we remark that the momentum integral @xmath31 as long as we can run the upper limits of the @xmath18 and @xmath23 integrations to @xmath32 , independently .
the large phase space at the lhc may make it possible to explore dynamic correlations that break eq .
( [ eq : approx1 ] ) . in fig .
[ fig : partonx ] , for the region of phase space of interest to us , we show the contributions to the @xmath33 cross section as a function of @xmath2 from both dps and sps , after minimal acceptance cuts are imposed ( sec . [
sect : calc ] ) .
the center - of - mass energy is @xmath34 tev .
it is evident that the majority of dps events are associated with low @xmath2 values , in essence never exceeding @xmath35 . the momentum carried off by the beam remnant is @xmath36 in dps and @xmath37 in sps .
the results in fig . [ fig : partonx ] show that this remnant momentum is not too different in dps and sps .
thus , the use of eq .
( [ eq : approx1 ] ) in calculations of event rates at the lhc appears adequate as a good first approximation .
while available tevatron data on double parton scattering @xcite are insensitive to possible correlations in @xmath2 , the greater dynamic range at the lhc may make it possible to observe them . )
holds at one hard scale , evolution of the parton densities with @xmath20 will induce violations at larger scales . ] in the dps and sps events .
most dps events have low @xmath2 values .
the events used for this plot include the requirements @xmath38 , @xmath39 , and the threshold cuts discussed in sec .
ii . ] assuming next that the two subprocesses @xmath6 and @xmath7 are dynamically uncorrelated , we express the double parton scattering differential cross section as : @xmath40 the symmetry factor @xmath41 is @xmath32 if the two hard - scattering subprocesses are identical and is @xmath8 otherwise . in the denominator
, there is a factor @xmath42 with the dimensions of a cross section . given that one hard - scatter has taken place , @xmath43 measures the size of the partonic core in which the flux of accompanying short - distance partons is confined .
it should be at most proportional to the transverse size of a proton .
for the first process of interest in this paper , @xmath44 , eq .
( [ eq : doubpartcross ] ) reduces to @xmath45 tevatron collider data @xcite yield values in the range @xmath46 mb .
we use this value for the estimates we make , but we emphasize that the goal should be to make an empirical determination of its value at lhc energies . in sec .
[ sect : calc ] , we present our calculation of the double parton and the single parton contributions to @xmath47 . we identify variables that discriminate the two contributions quite well . in sec .
[ sect:4jets ] , we treat the double parton and the single parton contributions to @xmath48 jet production , again finding that good separation is possible despite the combinatorial uncertainty in the pairing of jets .
we show in both cases that the double parton contribution falls off significantly more rapidly with @xmath49 , the transverse momentum of the leading jet . for the value of @xmath46 mb and
the cuts that we use , we find that , in the region in which it is most identifiable , double parton scattering is dominant for @xmath50 gev in @xmath11 at lhc energies , and @xmath51 gev in @xmath48 jet production .
our conclusions are found in sec .
[ sect : conclusions ] .
in this section , we describe the calculation of the dps and sps event rates for @xmath33 production at the lhc . for our purposes , light jets ( denoted by @xmath4 )
are assumed to originate only from gluons or one of the four lighter quarks ( @xmath52 or @xmath53 ) and , as stated above , we perform all calculations for the lhc with a center - of - mass energy of @xmath34 tev .
event rates are quoted for 10 pb@xmath54 of data .
the prediction for the dps event rate is based on the assumption that the two partonic interactions which produce the @xmath55 and @xmath56 systems occur independently ( as expressed in eq .
( [ eq : doubpartcross ] ) ) . at leading order ,
the only contribution is : @xmath57 where the symbol @xmath58 denotes the combination of one event each from the @xmath55 and the @xmath56 final states . in an attempt to model some of the effects expected from initial- and final - state radiation
, we also account for the possibility of an additional jet which is undetected because it is either too soft or outside of the accepted rapidity range .
thus , we include several other contributions to the dps event : @xmath59 where the parentheses surrounding a jet indicate that it is undetected .
we compute processes such as @xmath60 and @xmath61 at lo as 3 parton final - state processes .
the 2 to 3 parton amplitudes for @xmath61 [ and @xmath60 ] diverge as the undetected jet @xmath63 becomes soft or collinear to one of the other final state partons or to an initial parton .
the divergences are removed in a full next - to - leading order ( nlo ) treatment , in which real emission and virtual ( loop ) contributions are incorporated , and the finite @xmath64 , @xmath61 , and @xmath65 contributions are present with proper relative normalization . in the lo parton level simulations done in this paper , we employ a cut at the generator level to remove the divergences .
all the final state objects in the processes listed above are required to have transverse momentum @xmath66 gev . in this fashion , we model some aspects of the expected momentum imbalance between the @xmath9 and @xmath10 arising from the 2 to 3 process @xmath67 , but we can not claim to include the relative normalization between the @xmath64 and @xmath65 contributions that would result from a full nlo treatment . we leave a complete nlo analysis for future work . the sps cross section is computed according to eq .
( [ eq : singlescat ] ) .
it receives contributions at lowest order from the 2 parton to 4 jet final state process : @xmath68 and , in the case where a jet is undetected , from the 5-jet final states ( computed at lo ) : @xmath69 we also investigate the possibility of @xmath70 and @xmath71 final state contributions to the sps cross section where two of the jets `` fake '' @xmath9 jets .
we find that the effects from these final states are subdominant compared to the processes listed in eqs .
( [ eq : sps - loprocess ] ) and ( [ eq : sps - nloprocesses ] ) . in our numerical analysis , we use the leading - order cteq6l1 parton distribution functions ( pdfs ) @xcite to compute both dps and sps cross sections , and we evaluate all cross sections using one - loop evolution of @xmath72 . for the renormalization and factorization scales ,
we choose the dynamic scale : @xmath73 where @xmath74 is the transverse momentum of the @xmath75 jet and @xmath76 ( @xmath77 gev ) for light ( bottom ) jets . in the case of roughly equal values of the transverse momenta @xmath74 , eq .
( [ eq : mu ] ) yields @xmath78 in sps and @xmath79 in dps . at lo
there is no obviously `` right '' hard scale , and the choice in eq .
( [ eq : mu ] ) seems as good as any other .
the dps events are generated as two separate sets of events with madgraph / madevent @xcite and then combined as described above .
for example , at leading order , we generate events separately for @xmath80 and @xmath81 , and these events are then combined as indicated in eq .
( [ eq : dps - loprocess ] ) . to increase the speed of the simulations ,
the sps events are generated with alpgen @xcite since the sps processes of interest are hard - coded in alpgen , which contains more compact expressions for the squared - matrix - elements than madgraph .
the events accepted after generation are required to have 4 jets @xmath38 with 2 of these tagged as @xmath9 s @xmath39 . at the generator level , all the final state objects in the processes listed in eq .
( [ eq : dps - loprocess ] ) through eq .
( [ eq : sps - nloprocesses ] ) must have transverse momentum @xmath66 gev , as mentioned above .
furthermore , at the analysis level , all events ( dps and sps ) are required to pass the following acceptance cuts : @xmath82 where @xmath83 is the jet s pseudorapidity , and @xmath84 is the separation in the azimuthal angle ( @xmath85 ) - pseudorapidity plane between jets @xmath16 and @xmath4 : @xmath86 we model detector resolution effects by smearing the final state energy according to : @xmath87 where we take @xmath88 and @xmath89 for jets . to account for @xmath9 jet tagging efficiencies ,
we assume a @xmath9-tagging rate of 60% for @xmath9-quarks with @xmath90 and @xmath91 .
we also apply a mistagging rate for charm - quarks as : @xmath92 while the mistagging rate for a light quark is : @xmath93 over the range @xmath94 , we linearly interpolate the fake rates given above @xcite .
having detailed the calculation of the @xmath33 event rates from dps and sps , we now discuss some of the distinguishing characteristics of the two contributions .
first , however , it is important to check that our simulations of dps events are not introducing an _ artificial _ correlation between the @xmath55 and @xmath56 final states .
we do this by inspecting the angle @xmath96 between the plane defined by the @xmath55 system and the plane defined by the @xmath56 system . if the two scattering processes @xmath97 and @xmath98 which produce the dps final state are truly independent , one would expect to see a flat distribution in the angle @xmath96 .
by contrast , many diagrams , including some with non - trivial spin correlations , contribute to the 2 parton to 4 parton final state in sps , and naively one would expect some correlation between the two planes . to avoid possible effects from boosting to the lab frame
, we define the two planes in the partonic center - of - mass frame . we specify the planes by using the three - momenta of the outgoing jets .
then , the angle between the two planes defined by the @xmath56 and @xmath55 systems is : @xmath99 where @xmath100 is the unit three - vector normal to the plane defined by the @xmath101 system .
the normal is undefined when @xmath102 and @xmath103 are back - to - back or @xmath104 and @xmath105 are back - to - back , as occurs in a large fraction of the dps events .
therefore , when @xmath106 , we use a different procedure .
we use the three - momentum of one of the incoming partons along with the three - momentum of one of the outgoing @xmath9 quarks to define the @xmath55 plane .
let @xmath107 be the three - momentum of an incoming parton , and @xmath108 be the three - momentum of the final - state @xmath9 ( or @xmath10 ) quark .
we then define @xmath109 to be the azimuthal angle of the three - vector normal to the @xmath110 plane .
note that we use @xmath85 here since the normal to any three - vector and the beam - line will be transverse to the beam - line ( not the case in the sps process ) . in this way
, the jet which is not used to define the plane is guaranteed to lie in the plane .
the plane for the @xmath56 system is defined in an analogous manner . finally , the angle between the planes is then : @xmath111 in fig .
[ fig : phi - planes ] , we display the number of events as a function of the angle between the two planes .
there is an evident correlation between the two planes in sps , while the distribution is flat in dps , indicative that the two planes are uncorrelated . and @xmath56 systems .
in sps events , there is a correlation among the planes which is absent for dps events .
, scaledwidth=59.0% ] another interesting difference between dps and sps is the behavior of event rates as a function of transverse momentum .
as an example of this , in fig .
[ fig : ptj_1 ] , we show the transverse momentum distribution for the leading jet ( either a @xmath9 or light @xmath4 ) for both dps and sps .
several characteristics are evident .
first , sps produces a relatively hard spectrum , and for the value of @xmath43 and the cuts that we use , we see that sps tends to dominate over the full range of transverse momentum considered . on the other hand
, dps produces a much softer spectrum which ( up to issues of normalization in the form of @xmath43 ) can dominate at small values of transverse momentum .
the cross - over between the two contributions to the total event rate is @xmath112 gev for the acceptance cuts considered here .
a smaller ( larger ) value of @xmath43 would move the cross - over to a larger ( smaller ) value of the transverse momentum @xmath1 of the leading jet .
distribution of the leading jet in @xmath113 after minimal cuts .
, scaledwidth=59.0% ] we turn next to the search for variables that may allow for a clear separation of the dps and sps contributions . since the topology of the dps events includes two @xmath114 hard scattering events , the two pairs of jet objects are roughly back - to - back .
we expect the azimuthal angle between the pairs of jets corresponding to each hard scattering event to be strongly peaked near @xmath115 .
real radiation of an additional jet , where the extra jet is missed because it fails the threshold or acceptance cuts , allows smaller values of @xmath116 .
the relevant distribution is shown for light jets ( non @xmath9-tagged ) in fig .
[ fig : delphi]a .
there is a clear peak near @xmath117 for dps events , while the events are more broadly distributed in sps events .
the secondary peak near small @xmath118 arises from gluon splitting which typically produces nearly collinear jets .
the suppression at still lower @xmath118 comes from the isolation cut @xmath119 . in the azimuthal angles of light jet pairs for dps and both sps+dps events
the dijet pairs are back - to - back in dps events .
( b ) the variable @xmath120 for dps and sps+dps events provides a stronger separation of the underlying dps events from the total sample when compared to @xmath121 for any pair.,title="fig:",scaledwidth=49.0% ] in the azimuthal angles of light jet pairs for dps and both sps+dps events .
the dijet pairs are back - to - back in dps events .
( b ) the variable @xmath120 for dps and sps+dps events provides a stronger separation of the underlying dps events from the total sample when compared to @xmath121 for any pair.,title="fig:",scaledwidth=49.0% ] the separation of dps events from sps events becomes more pronounced if information is used from both the @xmath55 and @xmath56 systems . as an example , we consider the distribution built from a combination of the azimuthal angle separations of both @xmath56 and @xmath122 pairs , using a variable adopted from ref .
@xcite : @xmath123 in fig .
[ fig : delphi]b , we present a distribution in @xmath124 for both dps and sps+dps events . again , as in the case of the @xmath125 distribution , we see that the sps events are broadly distributed across the allowed range of @xmath120 .
however , the combined information from both the @xmath55 and @xmath56 systems shows that the dps events produce a sharp and substantial peak near @xmath126 which is well - separated from the total sample .
the narrow peaks near @xmath117 in fig .
[ fig : delphi]a and near @xmath127 in fig .
[ fig : delphi]b will be smeared somewhat once soft qcd radiation and other higher - order terms are included in the calculation .
another possibility for discerning dps is the use of the total transverse momentum of both the @xmath55 and @xmath56 systems . at lowest order for a @xmath128 process , the vector sum of the transverse momenta of the final state pair vanishes . in reality ,
radiation and momentum mismeasurement smear the expected peak near zero .
nevertheless , we still expect dps events to show a distribution in the transverse momenta of the jet pairs that is reasonably well - balanced . to encapsulate this expectation for both light jet pairs and @xmath9-tagged pairs , we use the variable @xcite : @xmath129 here @xmath130 is the vector sum of the transverse momenta of the two final state @xmath9 jets , and @xmath131 is the vector sum of the transverse momenta of the two ( non @xmath9 ) jets . the distribution in @xmath132 is shown in fig .
[ fig : sptprimecut ] .
as expected , we observe that the dps events are peaked near @xmath133 and are well - separated from the total sample . the sps events , on the other hand , tend to be far from a back - to - back configuration and , in fact , are peaked near @xmath134 .
this behavior of the sps events is presumably related to the fact that a large number of the @xmath55 or @xmath56 pairs arise from gluon splitting which yields a large @xmath135 imbalance and , thus , larger values of @xmath132 . for the dps and sps samples .
due to the back - to - back nature of the @xmath114 events in dps scattering , the transverse momenta of the jet pair and of the @xmath9-tagged jet pair are small , resulting in a small value of @xmath132 .
in ( a ) we show the @xmath132 distribution for our standard cuts , and in ( b ) we increase the cut on the transverse momentum of the leading jet , @xmath136 gev . the fraction of dps events in the whole sample decreases with increasing @xmath1.,title="fig:",scaledwidth=49.0% ] for the dps and sps samples . due to the back - to - back nature of the @xmath114 events in dps scattering , the transverse momenta of the jet pair and of the @xmath9-tagged jet pair are small , resulting in a small value of @xmath132 . in ( a )
we show the @xmath132 distribution for our standard cuts , and in ( b ) we increase the cut on the transverse momentum of the leading jet , @xmath136 gev .
the fraction of dps events in the whole sample decreases with increasing @xmath1.,title="fig:",scaledwidth=49.0% ] in this subsection , we find that extraction of the dps `` signal '' for @xmath33 production from the sps `` background '' can be enhanced by combining information from both @xmath55 and @xmath56 systems .
our simulations suggest that the variable @xmath132 may be a more effective discriminator than @xmath124 .
however , given the leading order nature of our calculations and the absence of smearing associated with initial state soft radiation , this picture may change and a variable such as @xmath120 ( or some other variable ) may become a clearer signal of dps at the lhc .
realistically , it would be valuable to study both distributions once lhc data are available in order to determine which is more instructive . in the following ,
we use the clear separation shown in fig .
[ fig : sptprimecut ] in our exploration of the distinct properties of dps and sps events . the evidence in fig . [ fig : delphi ] and fig . [
fig : sptprimecut ] for distinct regions of dps dominance prompts the search for greater discrimination in a plane represented by a two dimensional distribution of one variable against another .
we examined scatter plots involving the inter - plane angle @xmath96 , the jet - jet azimuthal angle difference @xmath116 , @xmath124 , and @xmath137 .
strong kinematic correlations are evident in the plot of @xmath124 _ vs. _ @xmath137 at the level of our leading order calculation , and we observe no additional separation of dps and sps beyond that evident in figs .
[ fig : delphi ] and [ fig : sptprimecut ] .
likewise , there are strong correlations between @xmath116 and @xmath124 .
one scatter plot with interesting features is displayed in fig .
[ fig : scatter1 ] .
the dps events are seen to be clustered near @xmath138 and are uniformly distributed in @xmath96 .
the sps events peak toward @xmath139 and show a roughly @xmath140 character . while already evident in figs .
[ fig : phi - planes ] and [ fig : sptprimecut ] , these two features are more apparent in the scatter plot fig . [
fig : scatter1 ] .
moreover , the scatter plot shows a valley of relatively low density between @xmath141 and @xmath142 . in an experimental one - dimensional @xmath96 distribution such as fig .
[ fig : phi - planes ] , one would see the sum of the dps and sps contributions . if structure is seen in data similar to that shown in the scatter plot fig .
[ fig : scatter1 ] , one could make a cut at @xmath143 or @xmath35 and verify whether the experimental distribution in @xmath96 is flat as expected for dps events . and @xmath132 for the dps and sps samples.,scaledwidth=99.0% ] in fig .
[ fig : ptj_1 ] , we show that dps produces a softer transverse momentum distribution for the leading jet ( either a @xmath9 or light @xmath4 ) . in data
one would see only the sum of the dps and sps components in a plot like fig .
[ fig : ptj_1 ] . a scatter plot of @xmath132 _ vs. _ the transverse momentum of the leading jet motivates an empirical separation of the two components . in figs .
[ fig : sptprimecut](a ) and [ fig : sptprimecut](b ) we compare the @xmath132 distributions for two different selections on the transverse momentum @xmath1 of the leading jet in the @xmath144 sample .
this comparison of the distributions confirms that events in the dps region , defined empirically by the region @xmath145 or @xmath35 , fall off more steeply with @xmath1 than the rest of the sample .
it will be important and interesting to see whether the selection @xmath145 or @xmath35 in lhc data also produces events that show a more rapid decrease with @xmath1 .
the leading - jet transverse momentum distributions are shown in figs .
[ fig : ptjcrossover](a ) and [ fig : ptjcrossover](b ) for two different cuts on @xmath132 . in both cases , we see that the sps sample has a broader distribution in @xmath1 and that the dps sample dominates for small enough values of @xmath1 . for our chosen value of @xmath46 mb , and for cuts we employ ,
the crossover points are roughly @xmath146 gev for @xmath147 and @xmath148 gev for @xmath149 . for ( a ) @xmath147 and ( b ) @xmath150 .
as the signal region becomes more dominated by sps events ( i.e. moving from ( a ) to ( b ) ) , the resulting distribution becomes harder and shifts the sps - dps cross - over from @xmath151 gev to @xmath152 gev .
, title="fig:",scaledwidth=49.0% ] for ( a ) @xmath147 and ( b ) @xmath150 . as
the signal region becomes more dominated by sps events ( i.e. moving from ( a ) to ( b ) ) , the resulting distribution becomes harder and shifts the sps - dps cross - over from @xmath151 gev to @xmath152 gev .
, title="fig:",scaledwidth=49.0% ]
in addition to @xmath153 , we can also ask how important dps can be for a generic @xmath154 final - state , where none of the jets are @xmath9-tagged . in this section , we describe our calculation of the double parton scattering and the single parton scattering contributions to the production of a @xmath154 final state , for which the cross section is larger . our exposition can be brief since we repeat the procedure described in some detail in sec . [
sect : calc ] .
the dps process for @xmath154 production is topologically equivalent to @xmath155 .
however , in the @xmath154 system , we lose the @xmath9-tagging ability that reduces the combinatorial background in @xmath153 , and the prospects for isolating and measuring dps over the sps background may appear less promising .
fortunately , in going from the @xmath122 subprocess to the @xmath56 subprocess , a much larger dps rate is possible due to the much larger cross section for @xmath56 production . as we show below
, we find that the dps signature can be extracted in this @xmath154 mode as well . the dps cross section for @xmath154 production receives contributions from the following sub - processes at the lowest order : @xmath156 where both @xmath9-quarks fail the @xmath9-tag .
we do not include the @xmath157 process due to its relatively small rate ( @xmath158 nb ) .
this rate is further reduced by requiring no @xmath9-tags , yielding roughly 40 events in the 10 pb@xmath54 of luminosity assumed here . following sec .
[ sect : calc ] , we account for the possibility of an additional jet which is undetected because it is too soft or outside of the accepted rapidity range . thus , we include several other contributions to the dps cross section : @xmath159 where the parentheses surrounding a jet signify that it is not detected .
the sps cross section receives contributions at lowest order from the final state : @xmath160 where both @xmath9-quarks fail the @xmath9-tag , and , in the case where a jet is not detected , from the final states : @xmath161 we refer to sec .
[ sect : calc ] for the specification of acceptance cuts and detector resolution , and for our treatment of the potential divergences present in the amplitudes for the processes in eqs .
( [ eq : dps - loprocess-4j])-([eq : sps - nloprocesses-4j ] ) .
similar to the @xmath153 process , the leading jet in the @xmath154 dps sample is typically softer than in the sps channels ( see fig . [
fig : ptj_1 - 4j ] ) . in this case , again using @xmath162 mb , we find that the cross - over between dps and sps dominance occurs near @xmath163 gev , higher than in the @xmath144 case shown in fig .
[ fig : ptj_1 ] . , but for @xmath154 events .
similar to the @xmath164 sample , the sps sample exhibits a harder @xmath135 spectrum.,scaledwidth=59.0% ] improvement in the separation between dps and sps in the @xmath154 case can be achieved with an analogous version of the @xmath132 variable introduced in eq .
( [ eq : sptprime ] ) : @xmath165 here @xmath166 is the vector sum of the transverse momenta of two final state jets , @xmath167 and @xmath9 , chosen among the four .
the remaining @xmath53 and @xmath168 jets are then fixed .
this choice is unique if a separation of the two hard interactions is possible . in the @xmath153 system , the separation into the @xmath122 and @xmath56 subsystems via @xmath9-tagging removed most of the degeneracy ( some degeneracy still remained via tagging efficiencies or light jet mistagging ) . in the @xmath154 system ,
the degeneracy can at first glance be problematic as there are 3 possible pairings of the four jets .
distribution for @xmath154 events shows much more combinatorial background than in the @xmath155 events . even after accepting two mis - matched jet pairs
, we see that the dps and sps samples can still be separated well.,scaledwidth=59.0% ] one might be tempted to take the pairing of jets which minimizes the value of @xmath132 .
unfortunately , this choice places a bias on the distribution that makes it potentially problematic to trust the discrimination . instead , to construct @xmath132
we take all three combinations of pairings , which includes one `` correct '' pairing and two incorrect pairings in the dps process .
this `` democratic '' @xmath132 distribution is shown in fig .
[ fig : sptprimedem ] and is re - weighted by 1/3 for proper normalization . as in the @xmath153 case , we see that the dps distribution peaks near @xmath133 , indicative that two back - to - back hard interactions are present .
in addition to this expected feature , we also see a continuum that extends above @xmath169 , associated with the wrong combination taken in the democratic approach . in fig .
[ fig : sptprimedem ] we see that dps produces a secondary peak at @xmath134 , not present in the @xmath144 case in fig .
[ fig : sptprimecut ] .
it appears to arise from the wrong pairings of jets associated with the combinatorial background . in these instances , the wrong combination of two jets that are close together in @xmath170 , meaning that their momenta are aligned , can maximize the value of @xmath132 .
overall , we see that the dps peak near @xmath171 provides a good means to separate dps events from sps events . , but for @xmath154 events with ( a ) democratic @xmath147 and ( b ) democratic @xmath150 . as in @xmath172 events ,
as one increases the cut on @xmath132 , the sps fraction increases and the total distribution is harder . ,
title="fig:",scaledwidth=49.0% ] , but for @xmath154 events with ( a ) democratic @xmath147 and ( b ) democratic @xmath150 .
as in @xmath172 events , as one increases the cut on @xmath132 , the sps fraction increases and the total distribution is harder . , title="fig:",scaledwidth=49.0% ] as in the @xmath144 case , we inspect the distribution in the @xmath135 of the leading jet after cuts on the @xmath132 variable . since there are three jet pairings per event , we now require that at least one of the three pairings has @xmath132 in the given window . due to this softer constraint , the hardening of the @xmath135 spectrum of the leading jet is less dramatic than in the @xmath153 case ( e.g. compare figs . [
fig : ptjcrossover ] and [ fig : ptj_1 - 4j - crossover ] ) .
the crossover of the sps and dps contributions occurs near @xmath146 gev for @xmath147 and near @xmath173 gev for @xmath149
our goal is to develop a method to search for a double parton scattering contribution in the @xmath95 and 4 jet final states at lhc energies and to measure the magnitude of its contribution relative to the single parton contribution to the same final states .
based on our parton level simulations , we find that variables such as @xmath132 and @xmath124 that take into account information from the entire final state , thereby including both of the hard subprocesses in dps , are more effective at discrimination than variables such as @xmath116 that reflect only a subset of the final - state .
the enhancement at low values of @xmath132 shown in figs . [ fig : sptprimecut ] , [ fig : scatter1 ] and [ fig : sptprimedem ] provides a good signature for the presence of double parton scattering .
we urge experimenters to search for such a concentration of events in data at the lhc .
having found this enhancement , we then suggest that the magnitude of this peak be examined as a function of the transverse momentum @xmath1 of the leading jet in the event sample .
the double parton scattering contribution in the peak region should fall off more rapidly with @xmath1 than the rest of the sample .
the distribution of events in the region of small values of @xmath132 should also be examined as a function of the inter - plane angle @xmath96 to see whether the flat behavior is seen , as expected for two independent production processes .
once these characteristics of double parton scattering are established , the data can be used to determine the effective normalization @xmath43 , defined and discussed in the introduction .
it will be interesting to see whether the values extracted for @xmath43 are about the same in the @xmath95 and 4 jet final states and how they compare with values measured at the fermilab tevatron .
once double parton scattering is established in data , and @xmath43 is determined , in a relatively clean process such as @xmath144 , double parton contributions to a wide range of other processes can be computed with more certainty about their expected rates at lhc energies . to be sure , given the approximations described in the introduction , some variation in the values of @xmath43 might be expected and appropriate for different processes and in different kinematic regions .
the connection of @xmath43 with the effective size of the hard - scattering core of the proton may mean that @xmath43 will have different values for @xmath174 , @xmath175 , and @xmath176 scattering .
there are several avenues for future work
. of great importance is the proper inclusion of next - to - leading order contributions @xcite .
they are needed to make more robust predictions of the relative normalization of the dps and sps contributions , of the shape of the @xmath135 distribution of the leading jet , and for proper softening of the sharp peaks seen near @xmath177 in figs .
[ fig : sptprimecut ] and [ fig : sptprimedem ] , and near @xmath178 in fig .
[ fig : delphi]b
. it will also be important to develop joint probabilities @xmath24 that are more sophisticated theoretically than the first approximation represented by eq .
( [ eq : approx1 ] ) in which parton - parton correlations are absent .
a valuable development in this direction are the studies presented in refs .
@xcite .
double parton contributions are potentially relevant for a wide range of standard model processes , many already considered in the literature @xcite , and they may also feed pertinent standard model backgrounds to new physics processes @xcite . they could be an issue in studies of higgs boson production in weak - boson - fusion since the `` forward '' jets could come from a second hard subprocess .
we benefited greatly from discussions with dr .
thomas lecompte and from communications with dr .
john campbell during the early development of this project .
we also thank tom , john , and professor jianwei qiu for valuable comments and suggestions on an earlier draft of this paper .
research in the high energy physics division at argonne is supported by the u. s. department of energy under contract no .
de - ac02 - 06ch11357 .
the research of gs at northwestern is supported by the u. s. department of energy under contract no .
de - fg02 - 91er40684 . c. goebel , f. halzen and d. m. scott , phys .
d * 22 * , 2789 ( 1980 ) .
n. paver and d. treleani , nuovo cim .
a * 70 * , 215 ( 1982 ) ; phys .
b * 146 * , 252 ( 1984 ) ; z. phys .
c * 28 * , 187 ( 1985 ) .
b. humpert , phys .
b * 131 * , 461 ( 1983 ) .
m. mekhfi , phys .
d * 32 * , 2371 ( 1985 ) ; phys .
d * 32 * , 2380 ( 1985 ) .
b. humpert and r. odorico , phys .
b * 154 * , 211 ( 1985 ) .
l. ametller , n. paver and d. treleani , phys .
b * 169 * , 289 ( 1986 ) .
f. halzen , p. hoyer and w. j. stirling , phys .
b * 188 * , 375 ( 1987 ) .
m. l. mangano , z. phys .
c * 42 * , 331 ( 1989 ) .
r. m. godbole , s. gupta and j. lindfors , z. phys .
c * 47 * , 69 ( 1990 ) .
m. drees and t. han , phys .
lett . * 77 * , 4142 ( 1996 ) [ arxiv : hep - ph/9605430 ] .
o. j. p. eboli , f. halzen and j. k. mizukoshi , phys . rev .
d * 57 * , 1730 ( 1998 ) [ arxiv : hep - ph/9710443 ] .
f. yuan and k. t. chao , j. phys .
g * 24 * , 1105 ( 1998 ) [ arxiv : hep - ph/9706293 ] .
g. calucci and d. treleani , nucl .
suppl . *
71 * , 392 ( 1999 ) [ arxiv : hep - ph/9711225 ] ; phys .
d * 80 * , 054025 ( 2009 ) [ arxiv:0907.4772 [ hep - ph ] ] ; phys .
d * 79 * , 074013 ( 2009 ) [ arxiv:0901.3089 [ hep - ph ] ] .
a. del fabbro and d. treleani , phys .
d * 61 * , 077502 ( 2000 ) [ arxiv : hep - ph/9911358 ] ; phys . rev .
d * 66 * , 074012 ( 2002 ) [ arxiv : hep - ph/0207311 ] .
a. kulesza and w. j. stirling , phys .
b * 475 * , 168 ( 2000 ) [ arxiv : hep - ph/9912232 ] . v. l. korotkikh and a. m. snigirev , phys .
b * 594 * , 171 ( 2004 ) [ arxiv : hep - ph/0404155 ] .
e. cattaruzza , a. del fabbro and d. treleani , phys . rev .
d * 72 * , 034022 ( 2005 ) [ arxiv : hep - ph/0507052 ] .
m. y. hussein , nucl .
. suppl . * 174 * , 55 ( 2007 ) [ arxiv : hep - ph/0610207 ] ; nucl .
suppl . * 152 * , 296 ( 2006 ) ; arxiv:0710.0203 [ hep - ph ] . e. maina , jhep * 0909 * , 081 ( 2009 ) [ arxiv:0909.1586 [ hep - ph ] ] ; jhep * 0904 * , 098 ( 2009 ) [ arxiv:0904.2682 [ hep - ph ] ] .
s. domdey , h. j. pirner and u. a. wiedemann , arxiv:0906.4335 [ hep - ph ] .
d. denterria , g. k. eyyubova , v. l. korotkikh , i. p. lokhtin , s. v. petrushanko , l. i. sarycheva and a. m. snigirev , arxiv:0910.3029 [ hep - ph ] .
j. r. gaunt and w. j. stirling , arxiv:0910.4347 [ hep - ph ] .
t. akesson _ et al . _
[ axial field spectrometer collaboration ] , z. phys .
c * 34 * , 163 ( 1987 ) .
f. abe _ et al . _
[ cdf collaboration ] , phys .
d * 56 * , 3811 ( 1997 ) ; phys .
lett . * 79 * , 584 ( 1997 ) .
d0 collaboration , `` double parton interactions in @xmath179 + 3 jet events in @xmath180 collisions at @xmath181 tev in d0 '' , d0 conference note 5910-conf ( april 24 , 2009 ) ; http://www-d0.fnal.gov/run2physics/www/results/prelim/qcd/q13/q13.pdf .
a. h. mueller and j. w. qiu , nucl . phys .
b * 268 * , 427 ( 1986 ) .
x. f. guo , phys .
d * 58 * , 036001 ( 1998 ) [ arxiv : hep - ph/9711453 ] ; nucl .
a * 638 * , 539c ( 1998 ) [ arxiv : hep - ph/9801450 ] .
j. pumplin , d. r. stump , j. huston , h. l. lai , p. m. nadolsky and w. k. tung , jhep * 0207 * , 012 ( 2002 ) [ arxiv : hep - ph/0201195 ] .
f. maltoni and t. stelzer , jhep * 0302 * , 027 ( 2003 ) [ arxiv : hep - ph/0208156 ] .
m. l. mangano , m. moretti , f. piccinini , r. pittau and a. d. polosa , jhep * 0307 * , 001 ( 2003 ) [ arxiv : hep - ph/0206293 ] .
h. baer , v. barger , g. shaughnessy , h. summy and l. t. wang , phys .
d * 75 * , 095010 ( 2007 ) [ arxiv : hep - ph/0703289 ] .
many next - to - leading order results can be obtained from the monte carlo program mcfm , j. m. campbell and r. k. ellis , phys .
d * 60 * , 113006 ( 1999 ) .
other nlo publications include l. j. dixon , z. kunszt , and a. signer , phys .
d * 60 * , 114037 ( 1999 ) and nucl .
b * 531 * , 3 ( 1998 ) .
z. sullivan and e. l. berger , phys .
d * 78 * , 034030 ( 2008 ) [ arxiv:0805.3720 [ hep - ph ] ] . | we evaluate the kinematic distributions in phase space of 4-parton final - state subprocesses produced by double parton scattering , and we contrast these with the final - state distributions that originate from conventional single parton scattering . our goal is to establish the distinct topologies of events that arise from these two sources and to provide a methodology for experimental determination of the relative magnitude of the double parton and single parton contributions at large hadron collider energies .
we examine two cases in detail , the @xmath0 and the 4 jet final states . after full parton - level simulations ,
we identify a few variables that separate the two contributions remarkably well , and we suggest their use experimentally for an empirical measurement of the relative cross section . we show that the double parton contribution falls off significantly more rapidly with the transverse momentum @xmath1 of the leading jet , but , up to issues of the relative normalization , may be dominant at modest values of @xmath1 . |
wireless sensor networks ( wsns ) can be utilized as target tracking systems that detect a moving target , localize it and report its location to the sink .
so far , the wsn - based tracking systems have found various applications , such as battlefield monitoring , wildlife monitoring , intruder detection , and traffic control @xcite .
this paper deals with the problem of target tracking by a mobile sink which uses information collected from sensor nodes to catch the target .
main objective of the considered system is to minimize time to catch , i.e. , the number of time steps in which the sink reaches the moving target .
moreover , due to the limited energy resources of wsn , also the minimization of data communication cost ( hop count ) is taken into consideration .
it is assumed in this study that the communication between sensor nodes and the sink involves multi - hop data transfers .
most of the state - of - the - art data collection methods assume that the current location of the target has to be reported to sink continuously with a predetermined precision .
these continuous data collection approaches are not suitable for developing the wsn - based target tracking applications because the periodical transmissions of target location to the sink would consume energy of the sensor nodes in a short time .
therefore , the target tracking task requires dedicated algorithms to ensure the amount of data transmitted in wsn is as low as possible .
intuitively , there is a trade - off between the time to catch minimization and the minimization of data communication cost . in this study
two algorithms are proposed that enable substantial reduction of the data collection cost without significant increase in time to catch .
the introduced communication - aware algorithms optimize utilization of the sensor node energy by selecting necessary data readings ( target locations ) that have to be transmitted to the mobile sink .
simulation experiments were conducted to evaluate the proposed algorithms against state - of - the - art methods .
the experimental results show that the presented algorithms outperform the existing solutions .
the paper is organized as follows .
related works are discussed in section 2 .
section 3 contains a detailed description of the proposed target tracking methods .
the experimental setting , compared algorithms and simulation results are presented in section 4 .
finally , conclusion is given in section 5 .
in the literature , there is a variety of approaches available that address the problem of target tracking in wsns .
however , only few publications report the use of wsn for chasing the target by a mobile sink . most of the previous works have focused on delivering the real - time information about trajectory of a tracked target to a stationary sink .
this section gives references to the wsn - based tracking methods reported in the literature that deal explicitly with the problem of target chasing by a mobile sink .
a thorough survey of the literature on wsn - based object tracking methods can be found in references @xcite .
kosut et al .
@xcite have formulated the target chasing problem , which assumes that the target performs a simple random walk in a two - dimensional lattice , moving to one of the four neighbouring lattice points with equal probability at each time step .
the target chasing method presented in @xcite was intended for a system composed of static sensors that can detect the target , with no data transmission between them .
each static sensor is able to deliver the information about the time of the last target detection to the mobile sink only when the sink arrives at the lattice point where the sensor is located .
a more complex model of the wsn - based target tracking system was introduced by tsai et al .
this model was used to develop the dynamical object tracking protocol ( dot ) which allows the wsn to detect the target and collect the information on target track .
the target position data are transferred from sensor nodes to a beacon node , which guides the mobile sink towards the target .
a similar method was proposed in @xcite , where the target tracking wsn with monitor and backup sensors additionally takes into account variable velocity and direction of the target . in this paper two target tracking methods are proposed that contribute to performance improvement of the above - mentioned target tracking approaches by reducing both the time to catch ( i.e. , the time in which mobile sink can reach the target ) and the data communication costs in wsn . in this study ,
the total hop count is analysed to evaluate the overall cost of communications , however it should be noted that different metrics can also be also used , e.g. , number of data transfers to sink , number of queries , number of transmitted packets , and energy consumption in sensor nodes .
the introduced algorithms provide decision rules to optimize the amount of data transfers from sensor nodes to sink during target chasing .
the research reported in this paper is a continuation of previous works on target tracking in wsn , where the data collection was optimized by using heuristic rules @xcite and the uncertainty - based approach @xcite .
the algorithms proposed in that works have to be executed by the mobile sink . in the present study
the data collection operations are managed by distributed sensor nodes . to reduce the number of active sensor nodes the proposed algorithms adopt the prediction - based tracking method @xcite . according to this method
a prediction model is applied , which forecasts the possible future positions of the target . on this basis
only the sensor nodes expected to detect the target are activated at each time step .
in this section two methods are proposed that enable reduction of data transfers in wsn during target tracking .
the wsn - based target tracking procedure is executed in discrete time steps . at each time step both the target and the sink move in one of the four directions : north , west , south or east .
their maximum velocities ( in segments per time step ) are assumed to be known .
movement direction of the target is random . for sink the direction
is decided on the basis of information delivered from wsn . during one time
step the sink can reach the nearest segments @xmath0 that satisfy the maximum velocity constraint : @xmath1 , where coordinates @xmath2 describe previous position of the sink .
sink moves into segment @xmath0 for which the euclidean distance @xmath3 $ ] takes minimal value .
note that @xmath4 are the coordinates of target that were lately reported to the sink .
let @xmath5 denote coordinates of the segment where the target is currently detected .
the sensor node that detects the target will be referred to as the target node . according to the proposed methods the information about target position is transmitted from the target node to the sink only at selected time steps .
if this information is transmitted then the destination coordinates at sink @xmath4 are updated , i.e , @xmath6 .
it means that the current position of the target is available for sink only at selected time steps . in remaining time periods the sink moves toward the last reported target position , which is determined by coordinates @xmath4 .
hereinafter , symbol @xmath7 will be used to denote the direction chosen by sink when moving toward segment @xmath8 . at each time step , the coordinates @xmath4 and @xmath5 are known for the target node . therefore , the target node can determine the direction which will be chosen by the sink in both cases : if the current target position is transmitted to the sink and if the data transfer is skipped . according to the first proposed method ,
the coordinates @xmath5 are transmitted to the sink only if @xmath9 , i.e. , if the direction chosen on the basis of coordinates @xmath4 is different than the one selected by taking into account the current position @xmath5 . in the second proposed method ,
the target node evaluates probability @xmath10 $ ] that the move of sink in direction @xmath11 will minimize its distance to the segment in which the target will be caught .
the target coordinates @xmath5 are transferred to the sink only if the difference @xmath12-p[dir(x_d , y_d)]$ ] is above a predetermined threshold . to evaluate probabilities @xmath10 $ ]
, the target node determines an area where the target can be caught .
this area is defined as a set of segments : @xmath13 where @xmath14 and @xmath15 are the minimum times required for target and sink to reach segment @xmath8 .
let @xmath16 and @xmath17 denote the segments into which the sink will enter at the next time step if it will move in directions @xmath18 and @xmath19 respectively .
in area @xmath20 two subsets of segments are distinguished : subset @xmath21 that consists of segments that are closer to @xmath16 than to @xmath17 and subset @xmath22 of segments that are closer to @xmath17 than to @xmath16 : @xmath23
< d[(x , y),(x_s , y_s)_d]\},\ ] ] @xmath24 < d[(x , y),(x_s , y_s)_c]\},\ ] ] on this basis the probabilities @xmath10 $ ] are calculated as follows : @xmath25= \frac{|a_c|}{|a| } , \ ; p[dir(x_d , y_d)]= \frac{|a_d|}{|a|},\ ] ] where @xmath26 denotes cardinality of the set .
$ ] calculations , width=226 ] the operations discussed above are illustrated by the example in fig .
1 , where the positions of target and sink are indicated by symbols @xmath27 and @xmath28 respectively .
velocity of the target is 1 segment per time step . for sink the velocity equals 2 segments per time step .
gray color indicates the area @xmath20 in which the sink will be able to catch the target .
the direction @xmath18 is shown by the arrow with number 1 and @xmath19 is indicated by the arrow with number 2 , thus @xmath29 and @xmath30 .
subset @xmath21 includes gray segments that are denoted by 1 .
the segments with label 2 belong to @xmath22 . in the analyzed example
@xmath31 , @xmath32 , and @xmath33 .
according to eq .
( 4 ) @xmath12 = 0.54 $ ] , @xmath34 = 0.38 $ ] and the difference of these probabilities equals to 0.16 .
if the first proposed method is applied for the analysed example then the data transfer to sink will be executed , since @xmath9 , as shown by the arrows in fig , 1 . in case of the second method
, the target node will send the coordinates @xmath5 to the sink provided that the difference of probability ( 0.16 ) is higher than a predetermined threshold .
the threshold value should be interpreted as a minimum required increase in the probability of selecting the optimal movement direction , which is expected to be obtained after transferring the target position data .
experiments were performed in a simulation environment to compare performance of the proposed methods against state - of - the - art approaches .
the comparison was made by taking into account two criteria : time to catch and hop count .
the time to catch is defined as the number of time steps in which the sink reaches the moving target .
hop count is used to evaluate the cost of data communication in wsn . in the experiments , it was assumed that the monitored area is a square of 200 x 200 segments .
each segment is equipped with a sensor node that detects presence of the target .
thus , the number of sensor nodes in the analysed wsn equals 40 000 .
communication range of each node covers the eight nearest segments .
maximum velocity equals 1 segment per time step for the target , and 2 segments per time step for the sink .
experiments were performed using simulation software that was developed for this study .
the results presented in sect .
4.3 were registered for 10 random tracks of the target ( fig .
each simulation run starts with the same location of both the sink ( 5 , 5 ) and the target ( 100 , 100 ) . during simulation
the hop counts are calculated assuming that the shortest path is used for each data transfer to sink , the time to catch is measured in time steps of the control procedure .
the simulation stops when target is caught by the sink . in the present study ,
the performance is analysed of four wsn - based target tracking algorithms .
algorithms 1 and 2 are based on the approaches that are available in literature , i.e. the prediction - based tracking and the dynamical object tracking .
these algorithms were selected as representative for the state - of - the - art solutions in the wsn - based systems that control the movement of a mobile sink which has to reach a moving target .
the new proposed methods are implemented in algorithms 3 and 4 .
the pseudocode in tab .
1 shows the operations that are common for all the examined algorithms .
each algorithm uses different condition to decide if current position of the target will be transmitted to the sink ( line 6 in the pseudocode ) .
these conditions are specified in tab .
2 . for all considered algorithms ,
the prediction - based approach is used to select the sensor nodes that have to be activated at a given time step @xmath35 .
prediction of the possible target locations is based on a simple movement model , which takes into account the assumptions on target movement directions and its maximum velocity .
if for previous time step @xmath36 the target was detected in segment @xmath37 , then at time step @xmath38 the set of possible target locations @xmath39 can be determined as follows : @xmath40 where @xmath41 is the maximum velocity of target in segments per time step .
.pseudocode for wsn - based target tracking algorithms [ cols= " < , < " , ] algorithm 2 is based on the tracking method which was proposed for the dynamical object tracking protocol . according to this approach
sink moves toward location of so - called beacon node @xmath4 .
a new beacon node is set if the sink enters segment @xmath4 . in such case , the sensor node which currently detects the target in segment @xmath5 , becomes new beacon node and its location is communicated to the sink .
when using this approach , the cost of data communication in wsn can be reduced because the data transfers to sink are executed less frequently than for the prediction - based tracking method .
the proposed communication - aware tracking methods are applied in algorithm 3 and algorithm 4 ( see tab .
details of these methods were discussed in sect .
simulation experiments were carried out in order to determine time to catch values and hop counts for the compared algorithms . as it was mentioned in sect .
3 , the simulations were performed by taking into account ten different tracks of the target .
average results of these simulations are shown in fig .
it is evident that the best results were obtained for algorithm 4 , since the objective is to minimise both the time to catch and the hop count .
it should be noted that fig .
3 . presents the results of algorithm 4 for different threshold values .
the relevant threshold values between 0.0 and 0.9 are indicated in the chart by the decimal numbers . according to these results , the average time to catch increases when the threshold is above 0.2 . for the threshold equal to or lower than 0.2
the time to catch takes a constant minimal value . the same minimal time to catch
is obtained when using algorithm 3 , however in that case the hop count is higher than for algorithm 4 . in comparison with algorithm 1 both proposed methods enables a considerable reduction of the data communication cost .
the average hop count is reduced by 47% for algorithm 3 and by 87% for algorithm 4 with threshold 0.2 .
algorithm 2 also reduces the hop count by about 87% but it requires much longer time to catch the target .
the average time to catch for algorithm 2 is increased by 52% .
detailed simulation results are presented in fig .
these results demonstrate the performance of the four examined algorithms when applied to ten different tracks of the target .
the threshold value in algorithm 4 was set to 0.2 .
the shortest time to catch was obtained by algorithms 1 , 3 and 4 for all tracks except the 5th one . in case of track 5 , when using algorithm 4 slightly longer time was needed to catch the target . for the remaining tracks
the three above - mentioned algorithms have resulted in equal values of the time to catch . in comparison with algorithm 1 ,
the proposed algorithms ( algorithm 3 and algorithm 4 ) significantly reduce the data communication cost ( hop count ) for all analysed cases . for each considered track algorithm 2
needs significantly longer time to reach the moving target than the other algorithms .
the hop counts for algorithm 2 are close to those observed in case of algorithm 4 . according to the presented results
, it could be concluded that algorithm 4 , which is based on the proposed method , outperforms the compared algorithms .
it enables a significant reduction of the data communication cost .
this reduction is similar to that obtained for algorithm 2 .
moreover , the time to catch for algorithm 4 is as short as in case of algorithm 1 , wherein the target position is communicated to the sink at each time step .
the cost of data communication in wsns has to be taken into account when designing algorithms for wsn - based systems due to the finite energy resources and the bandwidth - limited communication medium . in order to reduce the utilization of wsn resources ,
only necessary data shall be transmitted to the sink .
this paper is devoted to the problem of transferring target coordinates from sensor nodes to a mobile sink which has to track and catch a moving target .
the presented algorithms allow the sensor nodes to decide when data transfers to the sink are necessary for achieving the tracking objective . according to the proposed algorithms ,
only selected data are transmitted that can be potentially useful for reducing the time in which the target will be reached by the sink .
performance of the proposed algorithms was compared against state - of - the - art approaches , i.e. , the prediction - based tracking and the dynamical object tracking .
the simulation results show that the introduced algorithms outperforms the existing solutions and enable substantial reduction in the data collection cost ( hop count ) without significant decrease in the tracking performance , which was measured as the time to catch .
the present study considers an idealistic wsn model , where the information about current position of target @xmath5 is always successfully delivered through multi - hop paths to the sink and the transmission time is negligible . in order to take into account uncertainty of the delivered information ,
the precise target coordinates @xmath5 should be replaced by a ( fuzzy ) set .
relevant modifications of the presented algorithms will be considered in future experiments .
although the proposed methods consider a simple case with a single sink and a single target , they can be also useful for the compound tracking tasks with multiple targets and multiple sinks @xcite .
such tasks need an additional higher - level procedure for coordination of the sinks , which has to be implemented at a designated control node , e.g. , a base station or one of the sinks .
the extension of the presented approach to tracking of multiple targets in complex environments is an interesting direction for future works .
zheng , j. , yu , h. , zheng , m. , liang , w. , zeng , p. : coordination of multiple mobile robots with limited communication range in pursuit of single mobile target in cluttered environment .
journal of control theory and applications , vol .
441 - 446 ( 2010 ) | this paper introduces algorithms for target tracking in wireless sensor networks ( wsns ) that enable reduction of data communication cost .
the objective of the considered problem is to control movement of a mobile sink which has to reach a moving target in the shortest possible time .
consumption of the wsn energy resources is reduced by transferring only necessary data readings ( target positions ) to the mobile sink .
simulations were performed to evaluate the proposed algorithms against existing methods .
the experimental results confirm that the introduced tracking algorithms allow the data communication cost to be considerably reduced without significant increase in the amount of time that the sink needs to catch the target . |
physical , chemical , and biological systems are investigated in many laboratories using single molecule spectroscopy @xcite .
the investigation of the distribution of the number of photons emitted from a single molecule source is the topic of extensive theoretical research e.g. @xcite and @xcite for a review . since optical properties of single molecules
are usually very sensitive to dynamics and statics of their environment , and since the technique removes the many particle averaging found in conventional measurement techniques , single molecule spectroscopy reveals interesting fluctuation phenomena .
an important mechanism responsible for the fluctuations in the number of photons emitted from a single molecule source is spectral diffusion e.g. @xcite . in many cases the absorption frequency of the molecule
will randomly change due to different types of interactions between the molecule and its environment ( e.g. @xcite and ref .
therein ) .
for example for single molecules embedded in low temperature glasses , flipping two level systems embedded in the glassy environment , induce stochastic spectral jumps in the absorption frequency of the single molecule under investigation @xcite . in this way
the molecule may come in and out of resonance with the continuous wave laser field with which it is interacting .
obviously a second mechanism responsible for fluctuations of photon counts is the quantum behavior of the spontaneous emission process @xcite . in his fundamental work mandel
@xcite showed that a single atom in the process of resonance fluorescence , _ in the absence of spectral diffusion _
, exhibits sub - poissonian photon statistics @xcite .
photon statistics is characterized by mandel s @xmath0 parameter @xmath1 where @xmath2 is the number of emitted photons within a certain time interval .
the case @xmath3 is called sub - poissonian behavior , while @xmath4 is called super - poissonian .
sub - poissonian statistics has no classical analog @xcite .
briefly , the effect is related to anti - bunching of photons emitted from a single source and to rabi - oscillations of the excited state population which favors an emission process with some periodicity in time ( see details below ) .
sub - poissonian photon statistics and photon anti - bunching were measured in several single molecule , and single quantum dots experiments @xcite . while sub - poissonian statistics is well understood in the context of resonance fluorescence of an isolated electronic transition of a simple atom in the gas phase , our theoretical understanding of sub - poissonian statistics for a molecule embedded in a fluctuating condensed phase environment is still in its infant stages . in this paper
we obtain an exact analytical expression for the @xmath0 parameter in the long time limit , for a single molecule undergoing a stochastic spectral diffusion process . to obtain the exact solution we use the zheng - brown generating function method for single molecule photon statistics @xcite . for the spectral diffusion we use a simple stochastic approach , in the spirit of the kubo
andersen line shape theory @xcite .
the model exhibits generic behaviors of line shapes of molecules embedded in a condensed phase environment , e.g. motional narrowing when the stochastic fluctuations are fast , power broadening etc .
we show that the @xmath0 parameter exhibits rich types of behaviors , in particular it reveals the quantum nature of the emission process in the sub - poissonian regime , while the corresponding model line - shape exhibits a classical behavior
. a brief summary of our results was recently published @xcite .
our analytical expressions for @xmath0 classify the transitions between sub and super poissonian statistics .
they give the conditions on the spectral diffusion time scale for sub - poissonian behavior .
motional narrowing type of effect is revealed also for the @xmath0 parameter .
our exact result is valid for weak and strong excitation ( i.e arbitrary rabi frequency ) .
it yields the lower bound on @xmath0 .
the solution shows how we may choose the rabi frequency so that the quantum nature of the photon emission process becomes larger , namely how to minimize @xmath0 in the sub - poissonian regime .
this is important for the efficient detection of quantum effects in single molecule spectroscopy , since choosing too small or too large values of the rabi frequency results in very small and hence undetectable values of @xmath0 .
finally our exact result is used to test the domain of validity of the generalized wiener khintchine which yields @xmath0 in terms of a fourier transform of a three time dipole correlation ( as well known the wiener khintchine theorem yields the line shape in terms of a one time dipole correlation function ) .
the theorem @xcite is based on the semi - classical theory of interaction of light with matter , and on linear response theory ( i.e. , weak rabi frequency ) , it yields @xmath4 . as pointed out in @xcite
such a behavior is expected to be valid only for slow enough spectral diffusion processes .
we briefly explain some of the main ideas of sub - poissonian statistics .
the general idea is that the photons emitted from a single particle , e.g. a molecule , a nano - crystal or atom are correlated in time .
consider first a hypothetical molecule , interacting with an exciting laser field , which emits photons with a constant time interval @xmath5 between successive emission events .
then @xmath6 , @xmath7 , and hence @xmath8 .
due to quantum uncertainty the photon emission process is always random and therefore @xmath9 .
sub - poissonian behavior where @xmath10 implies that the stream of photons emitted from a single source maintain correlations in their arrival times to a detector .
usually when many molecules interact with a continuous wave laser the emission events are not correlated , and the fluorescence exhibits poissonian statistics @xmath11 .
in contrast a single molecule , once it emits a photon , is collapsed to its ground state .
hence immediately after an emission event the molecule can not emit a second time ( it has to be re - excited by the laser ) .
hence successive photons emitted from a single molecule , seem to repel each other on the time axis , a non - poissonian behavior .
this well known effect is called anti - bunching @xcite which is related to sub - poissonian statistics . a second effect related to sub - poissonian behavior are rabi - oscillations .
consider a simple atom in the process of resonance fluorescence .
when the electronic transition ( frequency @xmath12 ) is in resonance with a continuous wave laser field ( frequency @xmath13 ) the electronic transition can be approximated by a two level system .
first let us mentally switch off the spontaneous emission , i.e. set the inverse life time of the transition @xmath14 . for zero detuning @xmath15 the transition will exhibit well known rabi oscillations : the population of the excited state will behave like @xmath16 . since the population in the excited state attains its maximum ( minimum ) periodically , also
the emission times of successive photons maintain certain degree of periodicity in time , which implies sub - poissonian statistics .
mandel showed @xcite , that for a two level atom in the process of resonance fluorescence @xmath17 when @xmath18 we have @xmath19 since then successive photon emission times are not correlated , because the time between successive emissions becomes very large .
while when @xmath20 the excited state is populated swiftly , and only the finite spontaneous emission rate delays the emission , hence @xmath19 also in this case .
( [ eqmold ] ) the lower bound @xmath21 is easily obtained , and the minimum @xmath22 is obtained when @xmath23 .
let @xmath2 be the random number of photons emitted by a single molecule source in a time interval @xmath24 , and @xmath25 is the probability of @xmath2 emission events .
the information about the photon statistics is contained in the moment generating function @xcite @xmath26 which yields the moments of @xmath2 @xmath27 with which the @xmath0 parameter can in principle be obtained . in eq .
( [ eqa02 ] ) , and in what follows , we use the notation @xmath28 and similarly for second order derivatives with respect to @xmath29 .
the over - line in eq .
( [ eqa02 ] ) describes an average over the process of photon emission , later we will consider a second type of average with respect to the spectral diffusion process , which we will denote with @xmath30 .
the equations of motion for the generating function was given in @xcite and are called generalized optical bloch equation . for a chromophore with single excited and ground states , and interacting with a continuous wave laser field @xmath31 these equations are exact within the rotating wave approximation and optical bloch equation formalism .
they yield the same type of information on photon statistics contained in the quantum jump approach to quantum optics which is used in quantum monte carlo simulations @xcite . in eq .
( [ eqa01 ] ) @xmath32 is the spontaneous emission rate of the electronic transition and @xmath33 is the rabi frequency .
the time evolving detuning is @xmath34 where @xmath13 @xmath35 is the laser frequency ( the molecule s bare frequency ) , and @xmath36 is the stochastic spectral diffusion process . in eq .
( [ eqa01 ] ) it is assumed that the molecule in its excited and ground state have no permanent dipole moments , hence the system is described only by the transition dipole moment via the rabi frequency .
the physical meaning of @xmath37 , @xmath38 , and @xmath39 and their relation to the standard bloch equation was given in @xcite , some discussion on this issue will follow eq .
( [ eqmt ] ) . for related work on the foundations of these equations
see @xcite and ref .
therein . note that when @xmath40 the damping terms in eq .
( [ eqa01 ] ) become small [ i.e. the @xmath41 terms ] , hence relaxation of the generalized bloch equations in the important limit of @xmath40 is slow . in what follows
we will consider the moments @xmath42 , @xmath43 .
for this aim it is useful to derive equations of motion for the vector @xmath44 @xmath45 taking the first and the second derivative of eq .
( [ eqa01 ] ) with respect to @xmath29 and setting @xmath46 , we find @xmath47 where @xmath48 is a @xmath49 matrix @xmath50 @xmath51 the first three lines of @xmath48 describe the evolution of @xmath52 , these are the standard optical bloch equations in the rotating wave approximation @xcite . these equations yield @xmath53 which in turn gives the mean number of photons using eq .
( [ eqa02 ] ) @xmath54,\ ] ] and us - usual @xmath55 is the population in the excited state .
the fourth line of @xmath48 is zero , it yields @xmath56 , this equation describes the normalization condition of the problem namely @xmath57 for all times @xmath58 [ to see this use eq .
( [ eq02 ] ) and @xmath59 .
the evolution of the other terms @xmath60 are of current interest since they describe the fluctuation of the photon emission process . in particular using eqs .
( [ eqa02],[eqmt ] ) @xmath61 } = \gamma\left [ \overline{n\left(t\right ) } + 2 { \cal w}'(1 ) \right].\ ] ] solutions of time dependent equations like eq .
( [ eqmtz ] ) are generally difficult to obtain , a formal solution is given in terms of the time ordering operator @xmath62 , @xmath63 z(0)$ ] .
( [ eqmtz ] ) yields a general method for the calculation of @xmath0 for a single molecule undergoing a spectral diffusion process .
the aim of this paper is to obtain an exact solution of the problem for an important stochastic process used by kubo and anderson @xcite to investigate characteristic behaviors of line shapes .
we assume @xmath64 where @xmath65 describe frequency shifts , and @xmath66 describes a random telegraph process : @xmath67 or @xmath68 .
the transition rate between state up ( + ) and state down ( - ) and vice versa is @xmath69 .
this dichotomic process is sometimes called the kubo - anderson process @xcite .
it was used to describe generic behaviors of line shapes @xcite , here our aim is to calculate @xmath0 describing the line shape fluctuations .
we use burshtein s method @xcite of marginal averages , to solve the stochastic differential matrix equation ( [ eqmtz ] ) .
the method yields the average @xmath70 with respect to the stochastic process .
we will calculate @xmath70 in the limit of long times , and then obtain the steady state behavior of the line shape @xmath71 and the @xmath0 parameter .
let @xmath72 be the average of @xmath73 under the condition that at time @xmath58 the value of @xmath74 respectively .
@xmath72 are called marginal averages , the complete average is @xmath75 .
the equation of motion for the marginal averages is an @xmath76 matrix equation @xmath77 in eq .
( [ eqmat18 ] ) the matrix @xmath78 is identical to matrix @xmath79 in eq .
( [ eqmt ] ) when @xmath80 is replaced by @xmath81 , and @xmath82 is a @xmath49 identity matrix . in the next subsection
we obtain the long time solution of eq .
( [ eqmat18 ] ) , the reader not interested in the mathematical details may skip to subsection [ secexact ] , where the solution for the line shape and @xmath0 is presented .
in three main steps , we now find the long time behavior of the marginal averages , with which the line and @xmath0 are then obtained .
+ @xmath83 as mentioned , from normalization condition we have @xmath57 for all times .
( [ eqmat18 ] ) yields the marginal averages @xmath84 , in the steady state . inserting these identities in eq .
( [ eqmat18 ] ) we obtain an equation of motion for the vector @xmath85 @xmath86 where @xmath87 @xmath88 and @xmath89 . in the long time
limit the solution of eq . ( [ eqstep1 ] ) reaches a steady state ( ss ) given by @xmath90 and @xmath91 is the inverse of @xmath92 . from eq .
( [ eqstep11 ] ) we find @xmath93 @xmath94 where @xmath95 , and @xmath96 is the @xmath97 matrix element of @xmath98 , @xmath99 .
we note that @xmath100 yields the steady state marginal averages of the population difference between the excited and ground state . from eq .
( [ eqa02 ] ) we see that we need @xmath101 to obtain the average number of photon emissions @xmath102 .
we use eq .
( [ eqmat18 ] ) and show @xmath103 @xmath104 and @xmath105 the average number of emitted photon , in the long time limit is ( [ eqa02 ] , [ eqyd ] ) @xmath106 namely @xmath107 is proportional to the steady state occupation in the excited state .
the line shape is @xmath108 where @xmath109.\ ] ] @xmath110 we use the solutions obtained in previous step to obtain inhomogeneous equations for @xmath111 , @xmath112 where and @xmath113 with @xmath114 \right\ } \pm { \gamma^2 \over 8 r } \left [ \langle { \cal w}^{ss}(1)\rangle_{- } - \langle{\cal w}^{ss}(1)\rangle_{+ } \right ] .
\label{eqbb}\ ] ] in the long time limit we obtain @xmath115 where @xmath116 and @xmath117 are column vectors @xmath118 with @xmath119 \mp { \gamma^2 \over 8 r } \left [ \langle { \cal w}^{ss } \left(1
\right ) \rangle_{- } - \langle { \cal w}^{ss } \left(1
\right ) \rangle_{+ } \right]\ ] ] we therefore obtain @xmath120\ ] ] @xmath121.\ ] ] @xmath122 from eq .
( [ eqa02 ] ) , @xmath123 . in steady state
we have @xmath124 from eq .
( [ eqmat18 ] ) one can show that in the long time limit @xmath125 t + \ ] ] @xmath126 t\ ] ] @xmath127t^2 .
\label{eqydd}\ ] ] finally we obtain the @xmath0 parameter using : @xmath128 using eqs .
( [ eqyd],[eqydd],[eqdef ] ) we obtain the main result of this manuscript @xmath129 ^ 2 + { \gamma \over i\left ( \omega_l \right ) } \sum_{k=\pm } \left\ { \langle { \cal w}^{ss}(1)\rangle_{k } \left [ 1 + 4 \langle { \cal w}^{ss}(1)\rangle_{k } \right ] \right\ } , \label{eqmain}\ ] ] which is valid when measurement time @xmath130 . the @xmath0 parameter in eq .
( [ eqmain ] ) is expressed in terms of @xmath91 . to obtain the solution in terms of the original parameters of the problem @xmath131
, we found analytical expressions for @xmath91 using mathematica .
the formula for the @xmath0 parameter is given in the following subsection .
without loss of generality we set @xmath132 , hence @xmath13 is the detuning .
we find @xmath133 \over \mbox{denominator}\left[q\right ] } , \label{eqq}\ ] ] @xmath134 = \ ] ] @xmath135 @xmath136 = \ ] ] @xmath137 + the line shape is @xmath138 \over \mbox{denominator}[i\left(\omega_l \right ) ] } , \label{eqls}\ ] ] @xmath139 = \ ] ] @xmath140 @xmath141 = \ ] ] @xmath142 when the detuning is zero we find @xmath143 \over \left [ 4 \gamma \nu^2 + \left ( \gamma^2 + 2 \omega^2 \right ) \left ( \gamma + 4 r \right ) \right]^2 } .
\label{eqzt}\ ] ] using eq .
( [ eqzt ] ) the lower bound @xmath144 is obtained .
the absolute minimum of @xmath0 , i.e. @xmath145 is found when @xmath146 , or when @xmath147 .
namely the absolute minimum is found for a molecule whose absorption frequency is fixed , or for a very fast spectral modulation . * remark : * eq .
( [ eqzt ] ) indicates a transition from sub - poissonian statistics @xmath148 to super - poissonian statistics @xmath149 when @xmath150 .
if @xmath151 , i.e. if the bath is fast compared with the radiative life time , we find sub - poissonian behavior for all values of @xmath65 and @xmath33 .
the @xmath0 parameter is a function of two control parameters @xmath152 and @xmath33 , and three model parameters @xmath153 .
the classification of different types of physical behaviors , based on the relative magnitude of the parameters is investigated in this section .
the limiting behaviors of @xmath0 are obtained from the exact solution using mathematica . in the slow modulation regime ,
the bath fluctuation process ( r ) is very slow compared with the radiative decay rate ( @xmath32 ) , frequency fluctuation amplitude ( @xmath65 ) and the rabi frequency ( @xmath33 ) .
this case is similar to situations in many single molecule experiments , for example single molecules in low temperature glasses .
the exact solution can be simplified in the limit @xmath155 , using eq .
( [ eqls ] ) @xmath156 and @xmath157 the line is a sum of two lorentzians centered on @xmath158 , namely it exhibits splitting behavior when @xmath159 . using eq .
( [ eqq ] ) we find a simple super - poissonian behavior for @xmath0 in the slow modulation limit @xmath160 @xmath161 and since @xmath160 , @xmath0 may become very large ( e.g. @xmath162 in fig .
[ fig1 ] ) .
a simple picture can be used to understand these results . in the slow modulation limit
the molecule jumps between two states @xmath163 and @xmath164 , the time between successive jumps is very long , in such a way that many photons are emitted between jump events . in each of these two states
the molecule emits photons at a rate @xmath165 , eq .
( [ eqlspm1 ] ) .
these rates are determined by the familiar steady state solutions of the optical bloch equation , for a two level atom with the absorption frequency @xmath166 fixed in time @xcite . in this slow limit
the random number of emitted photons , in time interval @xmath24 , is @xmath167 , and @xmath168 is a stochastic intensity that jumps between two states @xmath169 with the rate @xmath69 . using this simple random walk picture
it is straightforward to derive eqs .
( [ eqsl01 ] , [ eqq1 ] ) . for mathematical details
@xcite who considered a similar slow modulation limit which is valid only for weak rabi frequency . within the slow modulation limit we distinguish between two cases .
the case @xmath171 is called the slow modulation strong coupling limit . in this case the line and @xmath0 have two well separated peaks and the broadening of the two peaks due to the finite life time , and power of the laser field is small compared with the frequency shifts ( see fig . [ fig1 ] ) . from fig .
[ fig1 ] , and as expected from eqs .
( [ eqsl01 ] , [ eqq1 ] ) , @xmath0 decreases when @xmath69 is increased , while @xmath172 is independent of @xmath69 . thus it is @xmath0 not @xmath173 that yields information on the dynamics . in fig .
[ fig1 ] the agreement between the exact solution and the approximation eq .
( [ eqq1 ] ) is good .
+ the limit @xmath175 is called the weak coupling slow modulation limit . in this case
the two peaks of the line , discussed in the previous sub - section , are overlapping and the line is approximated by @xmath176 this result is exact when @xmath177 , for arbitrary @xmath69 , the behavior of @xmath0 is demonstrated in fig .
[ fig2 ] , where we observe both super - poissonian and sub - poissonian behaviors . in the slow modulation weak coupling limit
, we must distinguish between the cases of large and small detuning .
first note that according to eq .
( [ eqq1 ] ) when the detuning is zero we find @xmath178 , namely the leading @xmath179 term in our asymptotic expansion vanishes .
we must therefore consider the higher order terms in our asymptotic expansion of eq .
( [ eqq ] ) and we find @xmath180 where @xmath181 eq .
( [ eqqapp ] ) has a simple meaning , the first term is a contribution to @xmath0 from spectral diffusion , which is identical to eq .
( [ eqq1 ] ) .
the second term @xmath182 is identical to the result obtained by mandel , for the @xmath0 parameter in the absence of spectral diffusion @xcite , and @xmath183 provided that the detuning is not too large .
the second term is dominating over the first when the detuning is small , and for zero detuning we obtain in eq .
( [ eqqapp ] ) sub - poissonian statistics .
more explicitly , we taylor expand ( [ eqlspm1 ] ) using @xmath65 as a small parameter , and obtain for the slow modulation weak coupling limit @xmath184 one may say that for zero detuning , the molecule behaves as if its absorption frequency is fixed . to conclude , we see that in the slow modulation limit @xmath0 is a sum of two additive contributions : i ) a part related to spectral diffusion @xmath185 and ii ) and a part related to quantum fluctuations , i.e. @xmath186 . such a behavior was very recently discussed in @xcite ( see also @xcite for related discussion ) .
the quantum fluctuations are however bounded from above and below @xmath187 , while the contribution from spectral diffusion is not @xmath188 .
hence detection of the quantum fluctuations is possible only when @xmath189 is small , which for our case implies zero detuning and weak coupling limit .
we now consider the fast modulation limit .
if we take @xmath191 keeping @xmath33 , @xmath32 , and @xmath65 fixed we find from eq .
( [ eqq ] ) @xmath192 given in eq .
( [ eqmandel ] ) .
hence in this limit we find a sub - poissonian behavior , provided that the detuning @xmath13 is not too large .
the line shape is identical to the expression on the right hand side of eq .
( [ eqs02 ] ) .
this behavior is expected , when the spectral diffusion is very fast the emitting single molecule can not respond to the stochastic fluctuations .
a more interesting case is to let @xmath191 and @xmath193 but keep , @xmath194 finite .
we call this limit the fast modulation limit , using eq .
( [ eqls ] ) the line shape is @xmath195 and when the rabi frequency is small @xmath196 in this limit we have the well known effect of motional narrowing : the width of the line is determine by @xmath197 and as the process becomes faster the line becomes narrower , namely @xmath198 decreases when @xmath69 is increased . the @xmath0 parameter is obtained from eq .
( [ eqq ] ) @xmath199 @xmath200 \over \left [ \gamma^3 + \gamma \gamma_{{\rm sd}}\ ^2 + 2 \gamma \omega^2 + 2 \gamma_{{\rm sd } } \left ( \gamma^2 + \omega^2 \right ) + 4 \gamma \omega_l ^2 \right]^2 } .
\label{eqq3}\ ] ] thus in the fast modulation limit the photon statistics is sub - poissonian provided that the detuning is not too large .
when @xmath201 the result for @xmath0 reduces to mandel s result eq .
( [ eqmandel ] ) . in fig .
[ fig3 ] , we show the line shape and the @xmath0 parameter , for three values of the jump rate ( r ) in the fast modulation regime while @xmath202 are kept fix .
we see that as the stochastic spectral diffusion process gets faster , both the line shape and the @xmath0 parameter become narrow .
thus both @xmath172 and @xmath0 exhibit a motional narrowing effect .
also , as the stochastic process gets faster , a stronger quantum behavior is obtained , in the sense that the minimum of @xmath0 decreases . to investigate the strong coupling limit we consider the value of @xmath0 for @xmath204 , and @xmath205 . from eq .
( [ eqq ] ) we obtain @xmath206 \over 2 r \left [ \gamma^3 + 4 \gamma r^2 + 2 \gamma \omega^2 + 2 r \left ( 2\gamma^2 + \omega^2 \right ) \right]^2 } .
\label{eqwwww}\ ] ] this equation exhibits both sub - poissonian and super poissonian behaviors . when the process is very _ slow _ , namely @xmath160 , we obtain @xmath207 a super - poissonian behavior . in the _ intermediate modulation limit _
, when @xmath208 , we obtain @xmath209 a sub - poissonian behavior .
when @xmath191 we find that @xmath0 is small @xmath210 eqs .
( [ eqnu1]-[eqnu3 ] ) are valid only in the limit of @xmath193
. however behaviors similar to the predictions of these equations are found also for finite values of @xmath65 .
on @xmath211 we have three typical behaviors : @xmath212 @xmath213 when the process is slow , see fig .
[ fig1 ] ) .
@xmath214 @xmath215 when @xmath216 , see fig .
[ fig4 ] for @xmath217 , and @xmath218 when @xmath191 we find @xmath219 , fig .
[ fig3 ] .
when @xmath221 we obtain interesting behaviors for @xmath0 . in fig .
[ fig4 ] the @xmath0 parameter shows a transition from sub - poissonian to super poissonian , photon statistics . in this regime of parameters , the shape of @xmath0 when plotted as a function of @xmath13 is very sensitive to the value of the control parameters e.g. in fig .
[ fig4 ] we change @xmath69 only moderately still we see very different types of behaviors for @xmath0 . for certain values of parameters
@xmath0 attains more than two peaks ( see fig .
[ fig4 ] for @xmath222 ) .
in contrast @xmath172 exhibits a simple splitting behavior with two peaks on @xmath158 , which is similar to the slow modulation case . besides the transition from sub to super poissonian behavior ,
a second type of transition is observed as @xmath69 is increased . in our problem
we have two types of sub - poissonian behavior .
we noticed already that when the stochastic modulation becomes very fast , @xmath0 has one minimum on zero detuning ( see fig . [ fig2 ] ) , while when @xmath220 , @xmath0 has two minima on @xmath223 ( see fig .
[ fig3 ] and @xmath224 .
the transition between these two types of sub - poissonian behaviors is shown in fig .
we now investigate the dependence of @xmath0 on the excitation field . in fig .
[ fig6 ] we consider an example line shape and @xmath0 parameter , where we fix the model parameters @xmath225 and vary the rabi frequency . for the line we see well known power broadening : as the rabi frequency is increased the line becomes wider , and as expected the photon emission rate @xmath172 increases monotonically when @xmath33 is increased . for the @xmath0 parameter
we have a turn - over behavior , as we increase @xmath33 the value of @xmath0 on zero detuning decreases then increases .
generally this type of turn - over is expected , since as discussed in sec .
[ secintrosub ] , @xmath11 when @xmath226 or @xmath227 .
thus there exists an optimal rabi frequency which yields an extremum of @xmath0 .
obviously it is important to obtain the values of @xmath33 which yield the extremum of @xmath0 , since then the fluctuations are the largest .
the extremum can be either a minimum or a maximum , as we shall show now .
we now consider the quantum regime @xmath3 . in fig .
[ fig8](b ) we demonstrate the turn - over behavior of @xmath228 for an example where the stochastic fluctuations are fast . in this fast modulation case
@xmath3 hence @xmath228 has a minimum .
for the same parameters the photon emission rate @xmath229 saturates as @xmath33 is increased , and the emission rate is never faster than @xmath32 [ see fig . [ fig8](a ) ] .
let @xmath230 be the rabi frequency which minimizes @xmath0 in the sub - poissonian case @xmath3 , and @xmath231 the corresponding value of @xmath0 .
( [ eqzt ] ) we find for zero detuning , and for @xmath232 @xmath233 and @xmath234 eqs .
( [ eqmin1],[eqmin2 ] ) yield @xmath230 and @xmath231 in terms of @xmath65 and @xmath69 .
due to motional narrowing effect , for fast processes satisfying @xmath235 , @xmath65 and @xmath69 are not easily obtained from experiment , while the parameter @xmath198 is in principle easy to obtain from the measurement of the line width .
( [ eqq3 ] ) , we find in the fast modulation limit and for zero detuning @xmath236 @xmath237 these simple equations relate between the width of the line given in eq .
( [ eqls3a ] ) and @xmath231 and @xmath230 . from eqs .
( [ eqsim1],[eqsim2 ] ) we see that when @xmath238 , @xmath239 and @xmath22 .
when @xmath240 we find @xmath241 and @xmath242 .
* remark : * when the detuning is not zero , we find using eq .
( [ eqq3 ] ) @xmath243 similar turn - over behaviors of @xmath228 are found also in other non fast parameter regimes . in fig .
[ fig7 ] we show @xmath228 versus @xmath33 for the slow modulation weak coupling limit @xmath244 and for zero detuning .
[ fig7 ] shows that @xmath0 exhibits a minimum as function of @xmath33 , this minimum is found in the vicinity of @xmath245 .
such a behavior is understood based on eq .
( [ eqqapp ] ) , the spectral diffusion contribution for @xmath0 is not important at zero detuning , while the contribution of @xmath246 yields @xmath239 . to demonstrate that the turnover behavior of @xmath228 is generic , we consider also the intermediate modulation limit in fig .
[ fig9 ] . here
we choose the detuning according to @xmath247 , since the @xmath0 parameter on zero detuning is relatively small ( see fig . [ fig5 ] ) .
in contrast to the behaviors in the quantum regime @xmath3 , in the slow modulation limit where @xmath4 , @xmath228 obtains a maximum , whose location is easy to calculate with eq .
( [ eqq1 ] ) .
such a behavior is demonstrated in fig .
[ fig10 ] for a case when the spectral shift @xmath65 is not very large .
if @xmath248 then in the slow modulation limit @xmath249 when the detuning is equal to @xmath247 . in eq .
( [ eqyh1 ] ) the second term on the right hand side is supposed to be a correction to the first term , namely @xmath250 .
let @xmath251 be the value of @xmath33 which maximizes @xmath0 in the super - poissonian regime , and @xmath252 the corresponding maximum .
this maximum always exists since as mentioned @xmath178 when @xmath227 or @xmath226 . then using eq .
( [ eqyh1 ] ) @xmath253 which is independent of @xmath69 and @xmath254 note that when the frequency shifts are very large @xmath255 we find using eq .
( [ eqyh2 ] ) @xmath256 .
hence the value of @xmath251 may become very large and then in experiment it is impossible to reach @xmath251 ( e.g. @xmath257 ) .
if we impose the condition @xmath258 we have @xmath259 for the laser detuning @xmath211 . hence @xmath185 monotonically increases and eventually saturates , similar to the behavior of the average emission rate .
the @xmath0 parameter yields informations not contained in the line shape .
the most obvious is the transition from super ( i.e. classical ) to sub ( quantum ) poissonian behavior .
such a quantum signature of the photon emission process is not obtained from the line shape . in comparison with the @xmath0 parameter of a single atomic transition ,
the @xmath0 parameter investigated here exhibits rich behaviors .
these include splitting , both in the sub and in the super poissonian regime , a transition from a fast to a slow modulation limit , and effects related to motional narrowing .
the most non - trivial behavior is obtained in the intermediate modulation limit when @xmath260 where @xmath0 attains more then two peaks . since @xmath0 contains the new information on single molecule experiments , namely information beyond the line shape , it is important to emphasize that @xmath0 attains an extremum for a particular value of the rabi frequency . in particular in the sub - poissonian regime
@xmath228 has a minimum .
hence we optimize the rabi frequency in such a way that @xmath261 is increased , e.g. we obtain @xmath230 . in other words there
exist an `` ideal '' choice of the rabi frequency in single molecule experiments . in the quantum sub - poissonian regime this optimal rabi frequency can not be considered weak , neither strong , hence perturbative approaches to single molecule spectroscopy are not likely to yield it .
this is in complete contrast to most theories of line shapes which are based on the assumption of weak external fields , e.g. the wiener
khintchine theorem and linear response theory .
single molecule theories should be able to predict the turn - over behavior of @xmath228 based on different models , since such a behavior is not expected to be limited to the model under investigation .
of - course the exact solution presented in this manuscript is very valuable in this direction , since it predicts precisely the details of this transition for the kubo - anderson stochastic process . it would be interesting to investigate further how general are our results .
from line shape theory , we know that in the fast modulation limit , line shapes have lorentzian shapes under very general conditions . from experiment
we know that motional narrowing effect , and lorentzian behavior of lines is wide spread .
thus at - least in this limit certain general features of line shapes , which are not sensitive to model assumptions are found .
similarly , we expect , that in the fast modulation limit , some of our results are general .
for example the motional narrowing behavior of @xmath0 and its approach to mandel s behavior @xmath186 is likely to be general .
it would be interesting to check if the relation between @xmath231 and @xmath230 eq . ( [ eqsim1],[eqsim2 ] ) , and the width of the line given by @xmath262 and @xmath32 is valid for other models , both stochastic and hamiltonian .
these simple equations are important since they yield the optimal rabi frequency @xmath230 in terms of the width of the line shape , which in turn is easily determined in usual line shape measurement . in ref .
( @xcite,@xcite ) a semi - classical framework for the mathematical calculation of @xmath0 for single molecule spectroscopy was investigated .
the approach yields the @xmath0 parameter in terms of a fourier transform of a three time dipole correlation function .
as mentioned in the introduction , the approach in ( @xcite,@xcite ) is based on two main approximations ( i ) external fields are weak @xmath227 , i.e. linear response theory , and the ( ii ) semi - classical approach to photon counting statistics .
the second assumption implies that @xmath4 , and as pointed out in @xcite such an approach is expected to be valid for slow processes .
the approach is useful since most single molecule experiments report on slow fluctuations .
the results obtained in this manuscript reduce to those in ( @xcite,@xcite ) in the limit of @xmath227 , and in the slow modulation limit , as they should .
the quantum behavior of @xmath0 becomes important when @xmath263 or for faster processes .
it is left for future work to construct a general quantum linear response theory , based on the eqs . of motion
[ eqmtz ] , which would yield both super and sub - poissonian statistics . finally , also
the investigation of the time dependence of @xmath0 is timely . | we investigate the distribution of the number of photons emitted by a single molecule undergoing a spectral diffusion process and interacting with a continuous wave laser field .
the spectral diffusion is modeled based on a stochastic approach , in the spirit of the anderson - kubo line shape theory . using a generating function formalism
we solve the generalized optical bloch equations , and obtain an exact analytical formula for the line shape and mandel s q parameter .
the line shape exhibits well known behaviors , including motional narrowing when the stochastic modulation is fast , and power broadening .
the mandel parameter , describing the line shape fluctuations , exhibits a transition from a quantum sub - poissonian behavior in the fast modulation limit , to a classical super - poissonian behavior found in the slow modulation limit .
our result is applicable for weak and strong laser field , namely for arbitrary rabi frequency .
we show how to choose the rabi frequency in such a way that the quantum sub - poissonian nature of the emission process becomes strongest . a lower bound on @xmath0 is found , and simple limiting behaviors are investigated .
a non - trivial behavior is obtained in the intermediate modulation limit , when the time scales for spectral diffusion and the life time of the excited state , become similar .
a comparison is made between our results , and previous ones derived based on the semi - classical generalized wiener khintchine formula . |
one of the successful approaches in describing the two - baryon interactions is the one - boson exchange ( or the nijmegen soft - core potential ) model @xcite . in this model ,
the baryon - baryon interactions are mediated by the intermediate mesons , such as the pseudoscalar octet mesons @xmath5 , @xmath6 , @xmath7 , the vector octet mesons @xmath8 , @xmath9 , @xmath10 and the scalar octet mesons @xmath11 , @xmath12 , @xmath13 , etc .
the strong coupling constants of the meson - baryon - baryon are the fundamental parameters , they have been empirically determined ( or fitted ) to reproduce the data of the nucleon - nucleon , hyperon - nucleon and hyperon - hyperon interactions .
the strong coupling constants of the vector mesons with the octet baryons ( thereafter we will denote them by @xmath14 ) can be written in term of the @xmath0 couplings and the electric ( and magnetic ) @xmath3 ratios .
the vector meson dominance theory indicates that the electric @xmath3 ratio @xmath15 be @xmath16 via the universal coupling of the @xmath8 meson to the isospin current @xcite .
it is important to determine those fundamental quantities directly from the quantum chromodynamics .
based on the assumption of the strong couplings between the quarks and vector mesons , the @xmath14 have been calculated in the external field qcd sum rules approach , while the coupling constants of the quark - meson were determined in some phenomenological models @xcite .
the strong coupling constants of the scalar mesons with the octet baryons have also been calculated in the external field qcd sum rules @xcite . in the external field
qcd sum rules , the operator product expansion is used to expand the time - ordered currents into a series of quark condensates , gluon condensates and vacuum susceptibilities which parameterize the long distance properties of the qcd vacuum and the non - perturbative interactions of the quarks and gluons with the external field @xcite . in this article , we calculate the strong coupling constants of the @xmath0 , @xmath17 and @xmath2 in the framework of the light - cone qcd sum rules approach , and determine the electric ( and magnetic ) @xmath3 ratios @xmath15 ( and @xmath18 ) .
the strong coupling constants of the @xmath0 have been calculated with the light - cone qcd sum rules approach @xcite , we revisit this subject and obtain different predictions .
furthermore , the strong coupling constants of the pseudoscalar mesons with the octet baryons have also been calculated with the light - cone qcd sum rules @xcite .
the light - cone qcd sum rules approach carries out the operator product expansion near the light - cone @xmath19 instead of the short distance @xmath20 while the non - perturbative hadronic matrix elements are parameterized by the light - cone distribution amplitudes which classified according to their twists instead of the vacuum condensates @xcite .
the non - perturbative parameters in the light - cone distribution amplitudes are calculated by the conventional qcd sum rules and the values are universal @xcite .
the article is arranged as : in section 2 , we derive the strong coupling constants @xmath0 , @xmath1 and @xmath2 in the light - cone qcd sum rules approach ; in section 3 , the numerical result and discussion ; and in section 4 , conclusion .
in the following , we write down the two - point correlation functions @xmath21 , @xmath22 where the baryon currents @xmath23 , @xmath24 and @xmath25 interpolate the octet baryons @xmath26 , @xmath27 and @xmath28 , respectively @xcite , the external state @xmath29 has the four momentum @xmath30 with @xmath31 . the vector meson @xmath32 can couple with the vector current @xmath33 with the following lagrangian , @xmath34 where the @xmath35 denotes the coupling constant .
the form factors of the vector current between two octet baryons can be written as @xmath36 where the @xmath37 is the average value of the masses of the two octet baryons . in the limit @xmath31 , @xmath38 , the form factors @xmath39 and @xmath40
are reduced to the strong coupling constants of the phenomenological lagrangian , @xmath41 according to the basic assumption of current - hadron duality in the qcd sum rules approach @xcite , we can insert a complete series of intermediate states with the same quantum numbers as the current operators @xmath23 , @xmath24 and @xmath42 into the correlation functions @xmath43 to obtain the hadronic representation . after isolating the ground state contributions from the pole terms of the baryons @xmath26 , @xmath27 and @xmath28 , we get the following results , @xmath44\left[m_i^2-q^2\right ] } \left\{-\left[g_v+g_t\right]\frac{m_\rho^2}{2}\!\not\!{\epsilon } -g_v \epsilon \cdot q \left[2\!\not\!{q } + \!\not\!{p}\right]\right\}+\cdots \ , , \nonumber \\ &
= & \pi_i^1(p , q)\!\not\!{\epsilon}+\pi_i^2(p , q)\epsilon \cdot q \!\not\!{q } + \cdots \ , , \end{aligned}\ ] ] where the following definitions have been used , @xmath45 here we use the notation @xmath46 to represent the octet baryons @xmath26 , @xmath27 and @xmath28 . we have not shown the contributions from the single pole terms in eq.(8 ) explicitly , they can be deleted completely after the double borel transformation . in the original qcd sum rules analysis of the nucleon magnetic moments
@xcite , the interval of dimensions ( of the condensates ) for the chiral odd structures is larger than the interval of dimensions for the chiral even structures , one may expect a better accuracy of the results obtained from the sum rules with the chiral odd structures . in this article
, we choose the tensor structures @xmath47 and @xmath48 for analysis . in the following ,
we briefly outline the operator product expansion for the correlation functions @xmath49 in perturbative qcd theory .
the calculations are performed at the large space - like momentum regions @xmath50 and @xmath51 , which correspond to the small light - cone distance @xmath19 required by the validity of the operator product expansion approach .
we write down the `` full '' propagator of a massive light quark in the presence of the quark and gluon condensates firstly @xcite , @xmath52|0\rangle & = & \frac{i\delta_{ab}\!\not\!{x } } { 2\pi^2x^4 } -\frac{\delta_{ab}m_q}{4\pi^2x^2}-\frac{\delta_{ab}}{12}\langle \bar{q}q\rangle + \frac{i\delta_{ab}}{48}m_q
\langle\bar{q}q\rangle-\frac{\delta_{ab}x^2}{192}\langle \bar{q}g_s\sigma gq\rangle \nonumber\\ & & + \frac{i\delta_{ab}x^2}{1152}m_q\langle \bar{q}g_s\sigma gq\rangle \!\not\!{x}-\frac{i}{32\pi^2x^2}\frac{\lambda^a_{ab}}{2 } g^a_{\mu\nu } ( \!\not\!{x } \sigma^{\mu\nu}+\sigma^{\mu\nu } \!\not\!{x } ) \nonumber\\ & & + \cdots \ , , \end{aligned}\ ] ] then contract the quark fields in the correlation functions @xmath21 with the wick theorem , and obtain the results @xmath53 \nonumber \\ & & \gamma_5\gamma^\mu d_{ck}(-x)\gamma^\nu \gamma_5\ , , \\
\pi_\sigma(p , q)&=&2i\epsilon_{abc}\epsilon_{ijk } \int d^4x \ , e^{-i q\cdot x } \ , tr\left [ \gamma_\nu c u^t_{bi}(-x)c\gamma_\mu u_{aj}(-x)\right ] \nonumber \\ & & \gamma_5\gamma^\mu s_{ck}(-x)\gamma^\nu \gamma_5\ , , \\
\pi_\xi(p , q)&=&2i\epsilon_{abc}\epsilon_{ijk } \int d^4x \ , e^{-i q \cdot x } \ ,
tr\left [ \gamma_\nu c s^t_{bi}(-x)c\gamma_\mu s_{aj}(-x)\right ] \nonumber \\ & & \gamma_5\gamma^\mu u_{ck}(-x)\gamma^\nu \gamma_5\ , , \end{aligned}\ ] ] where the @xmath54 , @xmath55 and @xmath56 stand for the full propagators of the @xmath57 , @xmath58 and @xmath59 quarks respectively .
take the replacement @xmath60 for one of the quark propagators in the correlation functions in eqs.(11 - 13 ) , perform the following fierz re - ordering @xmath61 and substitute the hadronic matrix elements ( such as the @xmath62 , @xmath63 , @xmath64 , etc . ) with the corresponding light - cone distribution amplitudes of the @xmath8 meson , finally we obtain the spectral densities at the coordinate space . in calculation , we can encounter some terms like @xmath65 , meson to illustrate the calculation .
] there are two approaches to deal with them , @xmath66 and @xmath67 the two approaches can lead to different results , the analytical expressions of the second approach are more cumbersome , after the double borel transformation and some technical details , we can prove that the two approaches are equal . in the following
, we will present the analytical results of the first approach only .
once the spectral densities in the coordinate space are obtained , we can translate them to the momentum space with the @xmath68 dimensional fourier transformation , @xmath69 @xmath70 @xmath71\frac{\gamma(\epsilon-1)}{(-q^2)^{\epsilon-1 } } \nonumber \\ & & + \frac{1}{\pi^2}f_\rho m_\rho^3 \int_0 ^ 1 du \int^u_0 dt \int_0^t d\lambda c(\lambda ) \frac{\gamma(\epsilon)}{(-q^2)^{\epsilon } } \nonumber \\ & & + \frac{1}{16\pi^2 } f_\rho m_\rho^3 \int_0 ^ 1 du a(u)\frac{\gamma(\epsilon)}{(-q^2)^{\epsilon}}\nonumber\\ & & -\frac{1}{3}m_s \langle \bar{s}s\rangle f_\rho m_\rho \int_0 ^ 1 du \left[\phi_{\parallel}(u ) + 2g_{\perp}^{(v)}(u)\right]\frac{\gamma(1)}{(-q^2)^{1 } } \nonumber \\ & & + \frac{2}{3}\langle \bar{q}q\rangle f^t_\rho m_\rho^2 \int_0 ^ 1 du h_{\parallel}^{(s)}(u)\frac{\gamma(1)}{(-q^2)^{1}}\nonumber\\ & & -\frac{1}{72 } \langle\frac { \alpha_sgg}{\pi}\rangle f_\rho m_\rho \int_0 ^ 1 du \left[4g_{\perp}^{(v)}(u)+\phi_{\parallel}(u)\right ] \frac{\gamma(1)}{(-q^2)^{1}}\nonumber \\ & & + \frac{1}{18 } m_s \langle \bar{s}g_s\sigma gs\rangle
f_\rho m_\rho \int_0 ^ 1 du \phi_{\parallel}(u ) \frac{\gamma(2)}{(-q^2)^{2}}\nonumber \\ & & + \frac{1}{12 } m_s \langle \bar{s}s\rangle f_\rho m_\rho^3 \int_0 ^ 1 du a(u)\frac{\gamma(2)}{(-q^2)^{2 } } \nonumber \\ & & -\frac{1}{6}\langle \bar{q}g_s\sigma gq\rangle f^t_\rho m_\rho^2 \int_0 ^ 1 du h_{\parallel}^{(s)}(u)\frac{\gamma(2)}{(-q^2)^{2}}\nonumber \\ & & + \frac{4}{3 } m_s \langle \bar{s}s\rangle f_\rho m_\rho^3\int_0 ^ 1 du \int^u_0 dt \int_0^t d\lambda c(\lambda)\frac{\gamma(2)}{(-q^2)^{2}}\nonumber \\ & & + \frac{1}{12 } \langle\frac { \alpha_sgg}{\pi}\rangle f_\rho m_\rho^3 \int_0 ^ 1 du \int^u_0 dt \int_0^t d\lambda c(\lambda ) \frac{\gamma(2)}{(-q^2)^{2}}\nonumber \\ & & + \frac{1}{288}\langle\frac { \alpha_sgg}{\pi}\rangle f_\rho m_\rho^3 \int_0 ^ 1 du a(u)\frac{\gamma(2)}{(-q^2)^{2 } } \ , , \end{aligned}\ ] ] @xmath72 @xmath73\frac{\gamma(1)}{(-q^2)^{1 } } \nonumber \\ & & + \frac{1}{72}\langle\frac { \alpha_sgg}{\pi}\rangle f_\rho m_\rho \int_0 ^ 1 du \left[2g_{\perp}^{(v)}(u)-\phi_{\parallel}(u)\right]\frac{\gamma(1)}{(-q^2)^{1 } } \nonumber \\ & & + \frac{1}{9 } m_s \langle \bar{s}g_s\sigma gs\rangle
f_\rho m_\rho \int_0 ^ 1 du \phi_{\parallel}(u ) \frac{\gamma(2)}{(-q^2)^{2}}\nonumber \\ & & + \frac{1}{6 } m_s \langle \bar{s}s\rangle f_\rho m_\rho^3 \int_0 ^ 1 du a(u)\frac{\gamma(2)}{(-q^2)^{2 } } \nonumber \\ & & + \frac{1}{288}\langle\frac { \alpha_sgg}{\pi}\rangle f_\rho m_\rho^3 \int_0 ^ 1 du a(u ) \frac{\gamma(2)}{(-q^2)^{2}}\ , , \end{aligned}\ ] ] @xmath74 where @xmath75 , the @xmath76 is a small positive quantity , after taking the double borel transformation , we can take the limit @xmath77 .
the light - cone distribution amplitudes @xmath78 , @xmath79 , @xmath80 , @xmath81 and @xmath82 of the @xmath8 meson are presented in the appendix @xcite , the non - perturbative parameters in the light - cone distribution amplitudes are scale dependent , in this article , the energy scale is taken to be @xmath83 .
here we have neglected the contributions from the gluons @xmath84 , the contributions proportional to the @xmath85 can give rise to three - particle ( and four - particle ) meson distribution amplitudes with a gluon ( and quark - antiquark pair ) in addition to the two valence quarks , their corrections are usually not expected to play any significant roles , the factorizable contribution is zero and the non - factorizable contributions from the soft hadronic matrix elements are too small to accommodate the experimental data @xcite ; the net contributions from the three - valence particle light - cone distribution amplitudes to the strong coupling constant @xmath86 are rather small , about @xmath87 @xcite .
the contributions of the three - particle ( quark - antiquark - gluon ) distribution amplitudes of the mesons are always of minor importance comparing with the two - particle ( quark - antiquark ) distribution amplitudes in the light - cone qcd sum rules . in our previous work ,
we study the four form - factors @xmath88 , @xmath89 , @xmath90 and @xmath91 of the @xmath92 in the framework of the light - cone qcd sum rules approach up to twist-6 three - quark light - cone distribution amplitudes and obtain satisfactory results @xcite . in the light - cone qcd sum rules
, we can neglect the contributions from the valence gluons and make relatively rough estimations . ] .
matching the hadronic representations in eq.(8 ) with the corresponding ones in eqs.(19 - 24 ) below the threshold @xmath93 , then perform the double borel transformation with respect to the variables @xmath94 and @xmath95 respectively , subtract the contributions from the continuum states , @xmath96}{\left[u(1-u)m_\rho^2+(1-u)q_1 ^ 2+uq_2 ^ 2\right]^n } & = & \frac{m^{2(2-n)}}{m_1 ^ 2 m_2 ^ 2 } e^{-\frac{u(1-u)m_\rho^2}{m^2 } } \delta(u - u_0 ) \ , , \nonumber\\ \frac{1}{m^2}&= & \frac{1}{m^2_1}+\frac{1}{m^2_2 } \ , , \nonumber\\ u_0&= & \frac{m_1 ^ 2}{m_1 ^ 2+m_2 ^ 2}\ , , \nonumber\\ m^{2n}&\rightarrow & \frac{1}{\gamma[n]}\int_0^{s_0 } ds s^{n-1}e^{-\frac{s}{m^2 } } \ , , \end{aligned}\ ] ] finally we obtain the following six sum rules for the strong coupling constants , @xmath97 @xmath98 @xmath99 @xmath100 @xmath101\nonumber\\ & & + \frac{1}{16\pi^2}m^4 e_1(x)f_\rho m_\rho^3 a(u_0)\nonumber\\ & & + \frac{1}{\pi^2}m^4 e_1(x)f_\rho m_\rho^3 \int^{u_0}_0 dt \int_0^t d\lambda c(\lambda ) \nonumber\\ & & -\frac{1}{3}m^2e_0(x)m_s \langle \bar{s}s\rangle f_\rho
m_\rho \left[\phi_{\parallel}(u_0 ) + 2g_{\perp}^{(v)}(u_0)\right ] \nonumber \\ & & + \frac{2}{3}m^2e_0(x)\langle \bar{q}q\rangle f^t_\rho m_\rho^2 h_{\parallel}^{(s)}(u_0)\nonumber\\ & & -\frac{1}{72}m^2e_0(x ) \langle\frac { \alpha_sgg}{\pi}\rangle f_\rho m_\rho \left[4g_{\perp}^{(v)}(u_0)+\phi_{\parallel}(u_0)\right ] \nonumber \\ & & + \frac{1}{18 } m_s \langle \bar{s}g_s\sigma gs\rangle
f_\rho m_\rho \phi_{\parallel}(u_0 ) + \frac{1}{12 } m_s \langle \bar{s}s\rangle f_\rho m_\rho^3
a(u_0 ) \nonumber \\ & & -\frac{1}{6}\langle \bar{q}g_s\sigma gq\rangle f^t_\rho m_\rho^2 h_{\parallel}^{(s)}(u_0)\nonumber\\ & & + \frac{4}{3 } m_s \langle \bar{s}s\rangle f_\rho m_\rho^3 \int^{u_0}_0 dt \int_0^t d\lambda c(\lambda)\nonumber \\ & & + \frac{1}{12 } \langle\frac { \alpha_sgg}{\pi}\rangle f_\rho m_\rho^3 \int^{u_0}_0 dt \int_0^t d\lambda c(\lambda ) \nonumber\\ & & \left.+\frac{1}{288}\langle\frac { \alpha_sgg}{\pi}\rangle f_\rho m_\rho^3 a(u_0 ) \right\ } \ , , \end{aligned}\ ] ] @xmath102 \nonumber \\ & & + \frac{1}{72}m^2e_0(x)\langle\frac { \alpha_sgg}{\pi}\rangle f_\rho m_\rho \left[2g_{\perp}^{(v)}(u_0)-\phi_{\parallel}(u_0)\right ] \nonumber \\ & & + \frac{1}{9 } m_s \langle \bar{s}g_s\sigma gs\rangle
f_\rho m_\rho \phi_{\parallel}(u_0 ) + \frac{1}{6 } m_s \langle \bar{s}s\rangle f_\rho m_\rho^3 a(u_0)\nonumber
\\ & & \left.+\frac{1}{288}\langle\frac { \alpha_sgg}{\pi}\rangle f_\rho m_\rho^3 a(u_0 ) \right\ } \ , , \end{aligned}\ ] ] where @xmath103 the @xmath93 are the threshold parameters .
the input parameters are taken as @xmath104 , @xmath105 , @xmath106 , @xmath107 , @xmath108 , @xmath109 , @xmath110 , @xmath111 , @xmath112 @xcite , @xmath113 , @xmath114 , @xmath115 , @xmath116 , @xmath117 , @xmath118 , @xmath119 , @xmath120 , @xmath121 , @xmath122 , @xmath123 @xcite , @xmath124 , @xmath125 , @xmath126 , @xmath127 , @xmath128 , @xmath129 , @xmath130 , @xmath131 , @xmath132 and @xmath133 @xcite .
the values of the vacuum condensates have been updated with the experimental data for the @xmath134 decays , the qcd sum rules for the baryon masses and analysis of the charmonium spectrum @xcite , in this article , we choose the standard ( or old ) values to keep in consistent with the sum rules used in determining the non - perturbative parameters in the light - cone distribution amplitudes .
the borel parameters are chosen as @xmath135 and @xmath136 , in those regions , the values of the strong coupling constants @xmath137 , @xmath138 and @xmath139 are rather stable with the variation of the borel parameter @xmath140 from the sum rules in eqs.(26 - 28 ) , while the sum rules for the @xmath141 , @xmath142 and @xmath143 in eqs.(29 - 31 ) are not as stable as the corresponding ones for the @xmath137 , @xmath138 and @xmath139 , which are shown in the figs.(1 - 3 ) . taking into account all the uncertainties ,
finally we obtain the numerical results of the strong coupling constants @xmath137 , @xmath138 , @xmath139 , @xmath141 , @xmath142 and @xmath143 , which are shown in the figs.1 - 3 , @xmath144 the numerical values are in agreement with the existing calculations in part , for examples , the values of the external field qcd sum rules ( @xmath145 , @xmath146 @xmath147 , @xmath148 , @xmath149 , @xmath150 ) @xcite and the phenomenologically fitted values of the one - boson exchange model ( @xmath151 , @xmath152 @xmath153 , @xmath154 , @xmath155 , @xmath156 ) @xcite . in calculation , we observe that the main contributions come from the perturbative terms and the quark condensates terms ( @xmath157 and @xmath158 ) , the contributions of the mixed condensates and gluon condensates are of minor importance .
it is not un - expected . from the table.1
, we can see that the contributions from the terms of quark condensates are comparable with the perturbative terms in the region @xmath136 .
the presence of the chiral symmetry breaking quark condensates and the existence of the massive baryons are closely related to each other , @xmath159 @xcite , the contributions from the quark condensates ( of order @xmath160 or @xmath161 ) in the sum rules in eqs.(26 - 31 ) are of great importance , the subtractions of the continuum states can be implemented by the simple replacement @xmath162 for @xmath163 , we have to introduce some unknown parameters @xmath164 and @xmath165 to subtract the contributions from the continuum states of order @xmath166 with @xmath167 and @xmath168 , however , it is difficult to obtain the values of the @xmath164 and @xmath165 . in the region
@xmath136 , @xmath169 @xcite , if the radiative @xmath170 corrections to the perturbative terms are companied with large numerical factors , just like in the case of the qcd sum rules for the masses of the proton @xcite , the contributions of order @xmath171 may be large , neglecting them can impair the predictive power .
the contributions of the three - particle ( quark - quark - gluon ) light - cone distribution amplitudes may be of the same order as the mixed condensates ( @xmath172 and @xmath173 ) , neglecting them would not impair the predictive power much .
the consistent and complete light - cone qcd sum rules analysis should include the contributions from the perturbative @xmath170 corrections , the distribution amplitudes with additional valence gluons and quark - antiquark pairs , the reasonable subtractions of the continuum states of order @xmath166 , and improve the parameters which enter in the light - cone qcd sum rules , that may be our next work .
.the power counting order of the @xmath137 , @xmath138 , @xmath139 , @xmath174 , @xmath175 , @xmath143 , respectively .
the @xmath176 stands for the perturbative terms and the @xmath177 stands for contributions from the quark condensates , we have not shown the terms which are suppressed by negative powers of the @xmath140 explicitly . [ cols="^,^,^,^,^,^,^",options="header " , ] and @xmath141 with the borel parameter @xmath140 .
, title="fig:",width=264 ] and @xmath141 with the borel parameter @xmath140 .
, title="fig:",width=264 ] and @xmath142 with the borel parameter @xmath140 . ,
title="fig:",width=264 ] and @xmath142 with the borel parameter @xmath140 . ,
title="fig:",width=264 ] and @xmath143 with the borel parameter @xmath140 .
, title="fig:",width=264 ] and @xmath143 with the borel parameter @xmath140 .
, title="fig:",width=264 ] the clebsch - gordan coefficients of the octet vnn vertices indicate that @xcite @xmath178 where we use the same notation @xmath179 to present the electric and magnetic ratios @xmath15 and @xmath18 .
our numerical values @xmath180 and @xmath181 deviate greatly from the prediction of the vector meson dominance theory , @xmath16 @xcite .
the numerical values of the @xmath137 , @xmath138 , @xmath139 , @xmath182 , @xmath183 , @xmath184 do not obey the simple relation obtained from the group theory in eq.(33 ) , the @xmath4 symmetry breaking effects are very large .
in this article , we calculate the strong coupling constants @xmath137 , @xmath138 , @xmath139 , @xmath182 , @xmath183 , @xmath184 of the @xmath0 , @xmath1 and @xmath2 in the framework of the light - cone qcd sum rules approach .
the strong coupling constants of the meson - baryon - baryon are the fundamental parameters in the one - boson exchange model which describes the baryon - baryon interactions successfully .
the numerical results are in agreement with the existing calculations in part , for examples , the predictions of the external field qcd sum rules and the phenomenologically fitted values of the one - boson exchange model .
the electric and magnetic @xmath3 ratios deviate greatly from the prediction of the vector meson dominance theory , the @xmath4 symmetry breaking effects are very large
. those deviations may be due to the failure to take into account the perturbative @xmath170 corrections , the un - subtracted contributions from the continuum states of order @xmath166 , and the neglected distribution amplitudes with additional valence gluons and quark - antiquark pairs .
the consistent and complete light - cone qcd sum rules analysis should include the contributions from the perturbative @xmath170 corrections , the distribution amplitudes with additional valence gluons and quark - antiquark pairs , the reasonable subtractions of continuum states of order @xmath166 , and improve the parameters which enter in the light - cone qcd sum rules .
the light - cone distribution amplitudes of the @xmath8 meson are defined by @xcite @xmath185f_\rho m_\rho \int_0 ^ 1 du e^{-i u p \cdot x } g_{\perp}^{(v)}(u ) \nonumber\\ & & -\frac{1}{2}x_\mu \frac{\epsilon \cdot x}{(p \cdot x)^2 } f_\rho m_\rho^3 \int_0 ^ 1 du e^{-iup \cdot x}c(u ) \ , , \end{aligned}\ ] ] @xmath186 m _
\rho^2\epsilon \cdot x \int_0 ^ 1 du e^{-i u p \cdot x } h_{\parallel}^{(s)}(u ) \ , .
\end{aligned}\ ] ] the light - cone distribution amplitudes are parameterized as @xcite @xmath187 where the @xmath188 , and the @xmath189 and @xmath190 are gegenbauer polynomials @xcite .
this work is supported by national natural science foundation , grant number 10405009 , and key program foundation of ncepu .
m. m. nagels , t. a. rijken , j. j. de swart , phys . rev .
* d15 * ( 1977 ) 2547 ; m. m. nagels , t. a. rijken , j. j. de swart , phys . rev . * d20 * ( 1979 ) 1633 ; p.
m. m. maessen , t. a. rijken , j. j. de swart , phys . rev .
* c40 * ( 1989 ) 2226
. t. a. rijken , v. g. j. stoks , y. yamamoto , phys . rev .
* c59 * ( 1999 ) 21 ; v. g. j. stoks , t. a. rijken , phys . rev .
* c59 * ( 1999 ) 3009 .
j. j. sakurai , ann .
* 11 * ( 1960 ) 1 . c. y. wen , w. y. p. hwang , phys
c56 * ( 1997 ) 3346 ; g. erkol , r. g. e. timmermans , t. a. rijken , phys . rev .
* c74 * ( 2006 ) 045201 .
t. m. aliev , a. ozpineci , s. b. yakovlev , v. zamiralov , hep - ph/0609026 ; t. m. aliev , a. ozpineci , m. savci , phys .
* d64 * ( 2001 ) 034001 ; t. doi , y. kondo , m. oka , phys . rept .
* 398 * ( 2004 ) 253 ; and references therein .
i. i. balitsky , v. m. braun and a. v. kolesnichenko , nucl . phys .
* b312 * ( 1989 ) 509 ; v. l. chernyak and i. r. zhitnitsky , nucl .
b345 * ( 1990 ) 137 ; v. l. chernyak and a. r. zhitnitsky , phys .
* 112 * ( 1984 ) 173 ; v. m. braun and i. e. filyanov , z. phys .
* c44 * ( 1989 ) 157 ; v. m. braun and i. e. filyanov , z. phys .
* c48 * ( 1990 ) 239 .
m. a. shifman , a. i. vainshtein and v. i. zakharov , nucl . phys .
* b147 * ( 1979 ) 385 , 448 ; l. j. reinders , h. rubinstein and s. yazaki , phys
* 127 * ( 1985 ) 1 ; s. narison , qcd spectral sum rules , world scientific lecture notes in physics * 26 * ( 1989 ) 1 . p. ball , v. m. braun , nucl . phys .
* b543 * ( 1999 ) 201 ; p. ball , v. m. braun , hep - ph/9808229 ; p. ball , v. m. braun , phys . rev . * d54 * ( 1996 ) 2182 ; p. ball , v. m. braun , y. koike , k. tanaka , nucl . phys .
* b529 * ( 1998 ) 323 ; p. ball , g. w. jones , r. zwicky , hep - ph/0612081 . c. b. chiu , j. pasupathy , s. l. wilson , phys
* d32 * ( 1985 ) 1786 ; c. b. chiu , j. pasupathy , s. l. wilson , phys .
rev . * d36 * ( 1987 ) 1451 ; c. b. chiu , s. l. wilson , j. pasupathy , j. p. singh , phys . rev .
* d36 * ( 1987 ) 1553 . c. b. chiu , j. pasupathy , s. l. wilson , phys
* d32 * ( 1985 ) 1786 ; c. b. chiu , j. pasupathy , s. l. wilson , phys . rev .
* d36 * ( 1987 ) 1451 ; c. b. chiu , s. l. wilson , j. pasupathy , j. p. singh , phys . rev .
* d36 * ( 1987 ) 1553 .
b. l. ioffe , k. n. zyablyuk , nucl . phys .
* a687 * ( 2001 ) 437 ; b. v. geshkenbein , b. l. ioffe , k. n. zyablyuk , phys . rev .
* d64 * ( 2001 ) 093009 ; b. l. ioffe , k. n. zyablyuk , eur .
j. * c27 * ( 2003 ) 229 ; k. zyablyuk , jhep * 0301 * ( 2003 ) 081 . | in this article , we calculate the strong coupling constants of the @xmath0 , @xmath1 and @xmath2 in the framework of the light - cone qcd sum rules approach .
the strong coupling constants of the meson - baryon - baryon are the fundamental parameters in the one - boson exchange model which describes the baryon - baryon interactions successfully .
the numerical values are in agreement with the existing calculations in part .
the electric and magnetic @xmath3 ratios deviate from the prediction of the vector meson dominance theory , the @xmath4 symmetry breaking effects are very large . + zhi - gang wang + department of physics , north china electric power university , baoding 071003 , p. r. china pacs numbers : 24.85.+p ; 12.40.vv ; 13.75.ev ; 13.75.gx |
after more than two decades of investigations , black hole thermodynamics is still one of the most puzzling subjects in theoretical physics .
one approach to studying the thermodynamical aspects of a black hole involves considering the evolution of quantum matter fields propagating on a classical ( curved ) background spacetime .
this gives rise to the phenomenon of black hole radiation that was discovered by hawking in 1974 @xcite .
combining hawking s discovery of black hole radiance with the classical laws of black hole mechanics @xcite , leads to the laws of black hole thermodynamics .
the entropy of a black hole obtained from this approach may be interpreted as resulting from averaging over the matter field degrees of freedom lying either inside the black hole @xcite or , equivalently , outside the black hole @xcite , as was first anticipated by bekenstein @xcite even before hawking s discovery .
the above approach was further developed in the following years @xcite .
a second route to black hole thermodynamics involves using the path - integral approach to quantum gravity to study _ vacuum _ spacetimes ( i.e. , spacetimes without matter fields ) . in this method ,
the thermodynamical partition function is computed from the propagator in the saddle point approximation @xcite and it leads to the same laws of black hole thermodynamics as obtained by the first method .
the second approach was further developed in the following years @xcite .
the fact that the laws of black hole thermodynamics can be derived without considering matter fields , suggests that there may be a purely geometrical ( spacetime ) origin of these laws .
however , a complete geometrical understanding of black hole thermodynamics is not yet present . in general , a basic understanding of the thermodynamical properties of a system requires a specification of the system s ( dynamical ) degrees of freedom ( d.o.f . ) .
obtaining such a specification is a nontrivial matter in quantum gravity . in the path - integral approach
one avoids the discussion of the dynamical d.o.f .. there , the dominant contribution to the partition function comes from a saddle point , which is a classical euclidean solution @xcite . calculating the contribution of such a solution to the partition function
does not require an identification of what the dynamical d.o.f.s of this solution are .
though providing us with an elegant way of getting the laws of black hole thermodynamics , the path - integral approach does not give us the basic ( dynamical ) d.o.f .
from which we can have a better geometrical understanding of the origin of black hole thermodynamics .
it was only recently that the dynamical geometric d.o.f . for a spherically symmetric vacuum schwarzschild black hole
were found @xcite under certain boundary conditions .
in particular , by considering general foliations of the complete kruskal extension of the schrawzschild spacetime , kucha @xcite finds a reduced system of _ one _ pair of canonical variables that can be viewed as global geometric d.o.f .. one of these is the schwarzschild mass , while the other one , its conjugate momentum , is the difference between the parametrization times at right and left spatial infinities . using the approach of kucha ,
recently louko and whiting @xcite ( henceforth referred to as lw ) studied black hole thermodynamics in the hamiltonian formulation .
as shown in fig .
2 , they considered a foliation in which the spatial hypersurfaces are restricted to lie in the right exterior region of the kruskal diagram and found the corresponding reduced phase space system .
this enabled them to find the unconstrained hamiltonian ( which evolves these spatial hypersurfaces ) and canonically quantize this reduced theory .
they then obtain the schrdinger time - evolution operator in terms of the reduced hamiltonian . the partition function @xmath0 is defined as the trace of the euclideanised time - evolution operator @xmath1 , namely , @xmath2 , where the hat denotes a quantum operator .
this partition function has the same expression as the one obtained from the path - integral approach and expectedly yields the laws of black hole thermodynamics . in a standard thermodynamical system
it is not essential to consider _
euclidean_-time action in order to study the thermodynamics .
if @xmath3 is the lorentzian time - independent hamiltonian of the system , then the partition function is defined as @xmath4 where @xmath5 is the inverse temperature of the system in equilibrium .
however , in many cases ( especially , in time- independent systems ) the euclidean time - evolution operator turns out to be the same as @xmath6 .
nevertheless , there are cases where , as we will see in section [ subsec : lwham ] , the euclidean time - evolution operator is not the same as @xmath7 .
this is the case for example in the lw approach , i.e. , @xmath8 , where @xmath9 is the reduced hamiltonian of the quantized lw system .
there is a geometrical reason for this inequality and in this work we discuss it in detail . in this paper , we ask if there exists a hamiltonian @xmath3 ( which is associated with certain foliations of the schwarzschild spacetime ) appropriate for finding the partition function of a schwarzschild black hole enclosed inside a finite - sized box using ( [ partition - trace ] )
. such a procedure will not resort to euclideanization . in our quest to obtain the hamiltonian that is appropriate for defining the partition function for ( [ partition - trace ] )
, we also clarify the physical significance of the lw hamiltonian . by doing
so we hope to achieve a better understanding of the geometrical origin of the thermodynamical aspects of a black hole spacetime . in a previous work @xcite , brown and york ( henceforth referred to as by ) found a general expression for the quasilocal energy on a timelike two - surface that bounds a spatial three - surface located in a spacetime region that can be decomposed as a product of a spatial three - surface and a real line interval representing time . from this expression
they obtained the quasilocal energy inside a spherical box centered at the origin of a four - dimensional spherically symmetric spacetime .
they argued that this expression also gives the correct quasilocal energy on a box in the schwarzschild spacetime . in this paper
we show that , although their expression for the quasilocal energy on a box in the schwarzschild spacetime is correct , the analysis they use to obtain it requires to be extended when applied to the case of schwarzschild spacetime . in this case
, one needs to impose extra boundary conditions at the timelike boundary inside the hole ( see fig .
3 ) . as mentioned above , in principle
, one can use the hamiltonian @xmath10 so obtained to evaluate the partition function , @xmath11 .
this partition function corresponds to the canonical ensemble and describes the thermodynamics of a system whose volume and temperature are fixed but whose energy content is permitted to vary .
such a hamiltonian , @xmath10 would then lead to a description of black hole thermodynamics without any sort of euclideanisation .
the only obstacle to this route to the partition function is that the trace can be evaluated only if one knows the density of the energy eigenstates .
unfortunately , without knowing what the thermodynamical entropy of the system is , it is not clear how to find this density in terms of the reduced phase - space variables of kucha @xcite .
so how can one derive the thermodynamical laws of the schwarzschild black hole using a lorentzian hamiltonian without knowing the density of states ?
based on an observation that identifies the thermodynamical roles of the by and the lw hamiltonians we succeed in studying black hole thermodynamics within the hamiltonian formulation but without euclideanization . in section [ sec : thermoconsi ] we describe the thermodynamical roles of the by and the lw hamiltonians .
identifying these roles allows us to immediately calculate the partition function and recover the thermodynamical properties of the schwarzschild black hole . in section [ sec : geoconsi ] we study the geometrical significance of these hamiltonians .
in particular , we extend the work of brown and york @xcite to find the nature of the spatial slices that are evolved by the by hamiltonian in the full kruskal extension of the schwarzschild spacetime . in section [ geothermo ]
we use the observations made in sections [ sec : thermoconsi ] and [ sec : geoconsi ] to ascribe geometrical basis to the thermodynamical parameters of the system , thus gaining insight into the geometrical nature of black hole thermodynamics .
we conclude the paper in section [ sec : conclu ] by summarising our results and discussing the connection between the foliation geometry and equilibrium black hole thermodynamics . in appendix
a we extend our results to the case of two - dimensional dilatonic black holes . in appendix
b we discuss an alternative foliation ( see fig .
4 ) , in which the spatial slices are again evolved by the by hamiltonian @xmath12 . this illustrates the non - uniqueness of the foliation associated with the by hamiltonian .
we shall work throughout in `` geometrized - units '' in which @xmath13 .
it was shown by brown and york @xcite that in 4d spherically symmetric einstein gravity , the quasilocal energy of a system that is enclosed inside a spherical box of finite surface area and which can be embedded in an asymptotically flat space is @xcite @xmath14 where @xmath15 is the adm mass of the spacetime and @xmath16 is the fixed curvature radius of the box with its origin at the center of symmetry .
we will call @xmath12 the brown - york hamiltonian .
the brown and york derivation of the quasilocal energy can be summarised as follows .
the system they consider is a spatial three - surface @xmath17 bounded by a two - surface @xmath18 in a spacetime region that can be decomposed as a product of a spatial three - surface and a real line interval representing time ( see fig .
the time evolution of the two - surface boundary @xmath18 is the timelike three - surface boundary @xmath19 .
they then obtain a surface stress - tensor on the boundary by taking the functional derivative of the action with respect to the three - metric on @xmath19 .
the energy surface density is the projection of the surface stress tensor normal to a family of spacelike two - surfaces like @xmath18 that foliate @xmath19 .
the integral of the energy surface density over such a two - surface @xmath18 is the quasilocal energy associated with a spacelike three - surface @xmath17 whose _ orthogonal _ intersection with @xmath19 is the two - boundary @xmath18 . as argued by by , eq .
( [ quasih ] ) also describes the total energy content of a box enclosing a schwarzschild hole .
one would thus expect to obtain the corresponding partition function from it by the prescription @xmath11 . as mentioned above
, the only obstacle to this calculation is the lack of knowledge about the density of states of the system , which is needed to evaluate the trace .
however , as we discuss in the next subsection , there is another hamiltonian associated with the schwarzschild spacetime that allows us to obtain the relevant partition function without euclideanization .
this is the louko - whiting hamiltonian . in their quest to obtain the partition function for the schwarzschild black hole in the hamiltonian formulation
, lw found a hamiltonian that time - evolves spatial hypersurfaces in a schwarzschild spacetime of mass @xmath15 such that the hypersurfaces extend from the bifurcation 2-sphere to a timelike box - trajectory placed at a constant curvature radius of @xmath20 ( see fig .
as we will show in the next subsection , the lw hamiltonian describes the correct free energy of a schwarzschild black hole enclosed inside a box in the thermodynamical picture .
the lw hamiltonian is @xmath21 where generically @xmath22 and @xmath23 are functions of time @xmath24 , which labels the spatial hypersurfaces .
physically , @xmath25 , where @xmath26 is the time - time component of the spacetime metric on the box . on the other hand , the physical meaning of @xmath23 is as follows . on a classical solution ,
consider the future timelike unit normal @xmath27 to a constant @xmath24 hypersurface at the bifurcation two - sphere ( see fig .
then @xmath23 is the rate at which the constant @xmath24 hypersurfaces are boosted at the bifurcation 2-sphere : @xmath28 where @xmath29 is the initial hypersurface and @xmath30 is the boosted hypersurface .
if one is restricted to a foliation in which the spatial hypersurfaces approach the box along surfaces of constant proper time on the box , then @xmath31 .
on classical solutions , the spatial hypersurfaces approach the bifurcation 2-sphere along constant killing - time hypersurfaces ( see the paragraph containing eqs .
( [ ls - r ] ) in section [ subsec : lwgeoham ] ) . the lw fall - off conditions ( [ ls - r ] ) , which are imposed on the adm variables at the bifurcation 2-sphere , can be used to show that on solutions , @xmath32 , where @xmath33 is the killing time and @xmath34 is the surface gravity of a schwarzschild black hole . in the particular case where the label time @xmath24 is taken to be the proper time on the box , we have @xmath35^{-1 } \ \ .\ ] ] with such an identification of the label time @xmath24 , eq .
( [ lwht ] ) shows that on classical solutions the time - evolution of these spatial hypersurfaces is given by the hamiltonian @xmath36 where @xmath23 is given by ( [ n0sol ] ) .
one may now ask if one can use the lw hamiltonian ( [ lwh ] ) to obtain a partition function for the system and study its thermodynamical properties .
unfortunately , one can not do so in a straightforward manner .
first , one can not replace @xmath3 in ( [ partition - trace ] ) by @xmath9 , the quantum counterpart of ( [ lwh ] ) , to obtain the partition function .
the reason is that classically @xmath37 does not give the correct energy of the system ; the by hamiltonian @xmath12 of ( [ quasih ] ) does . to avoid this problem , lw first construct the schrdinger time - evolution operator @xmath38 .
they then euclideanize this operator and use it to obtain the partition function @xmath39 . the partition function so obtained does not equal @xmath40 , but rather it turns out to be the same as that obtained via the path integral approach of gibbons and hawking @xcite .
however , apart from this end result , a justification at some fundamental level has been lacking as to why the lw hamiltonian ( [ lwh ] ) and not any other hamiltonian ( eg . , ( [ quasih ] ) ) should be used to obtain the partition function using the lw procedure . in the next subsection , we will find the thermodynamical roles played by the by and lw hamiltonians .
we will also show how this helps us in obtaining the partition function without euclideanization . this way we will avoid the ambiguity mentioned above that arises in the lw - method of constructing the partition function .
as argued by brown and york @xcite , on solutions , the by hamiltonian @xmath12 in eq .
( [ quasih ] ) denotes the internal energy @xmath41 residing within the box : @xmath42 in fact eq .
( [ e ] ) can be shown to yield the first law of black hole thermodynamics @xmath43 where @xmath44 is the surface pressure on the box - wall @xcite @xmath45 the first term on the rhs of ( [ de ] ) is negative of the amount of work done by the system on its surroundings and , with hindsight , the second term is the product @xmath46 , where @xmath47 is the temperature and @xmath48 is the entropy of the system .
we will not assume the latter in the following analysis ; rather we will deduce the form of @xmath47 and @xmath48 from first principles .
we next show that the lw hamiltonian @xmath37 of eq .
( [ lwh ] ) plays the role of helmholtz free energy of the system .
recall that the helmholtz free energy @xmath49 is defined as @xmath50 where @xmath41 is the internal energy .
thus in an isothermal and reversible process , the first law of thermodynamics implies that the amount of mechanical work done by a system , @xmath51 , is equal to the decrease in its free energy , i.e. , @xmath52 as a corollary to this statement it follows that for a mechanically isolated system at a constant temperature , the state of equilibrium is the state of minimum free energy .
we now show that under certain conditions on the foliation of the spacetime with spatial hypersurfaces , the lw hamiltonian @xmath37 in eq .
( [ lwh ] ) plays the role of _
free energy_. we choose a foliation such that on solutions the lapse @xmath23 obeys ( [ n0sol ] ) . using the expression ( [ e ] ) for @xmath41 , the hamiltonian @xmath37 in eq . ( [ lwh ] ) can be rewritten as @xmath53 now let us perturb @xmath37 about a solution by perturbing @xmath15 and @xmath16 such that @xmath23 itself is held fixed .
then @xmath54 note that keeping @xmath23 fixed , i.e. , @xmath55 , does not necessarily imply through ( [ n0sol ] ) that @xmath56 and @xmath57 are not independent perturbations .
this is because , in general , the perturbed @xmath37 may not correspond to a solution and hence the perturbed @xmath23 need not have the form ( [ n0sol ] ) .
however , here we will assume that the perturbations do not take us off the space of static solutions and , therefore , the perturbed @xmath23 has the form ( [ n0sol ] ) .
hence in our case @xmath58 and @xmath57 are not independent perturbations . using ( [ de ] ) and ( [ n0sol ] ) in ( [ dhitode ] ) yields
@xmath59 finally , from ( [ dhdw ] ) and ( [ fw ] ) we get @xmath60 where @xmath61 is a constant independent of @xmath15 .
to find @xmath61 , we take the limit @xmath62 . in this limit
both @xmath37 and @xmath49 vanish and , therefore , @xmath61 has to be zero .
another way to see that @xmath61 should vanish is to identify the geometric quantity @xmath23 with the temperature , @xmath47 , of the system ( up to a multiplicative constant )
. then the perturbation ( [ dhitode ] ) in @xmath37 , keeping @xmath23 ( and , therefore , @xmath47 ) fixed , describes an isothermal process .
but eq . ( [ hfc ] ) shows that @xmath61 has to be an extensive function of thermodynamic invariants of the isothermal process since @xmath37 and @xmath49 are both extensive .
the only thermodynamic quantity that we assume to be invariant in this isothermal process is the temperature @xmath47 .
but since @xmath47 is not extensive , @xmath61 has to be zero .
the fact that @xmath23 indeed determines the temperature of the system will be discussed in detail in a later section .
the above proof of the lw hamiltonian being the helmholtz free energy immediately allows us to calculate the partition function for a canonical ensemble of such systems , @xmath63 by simply putting @xmath64 . in this way we recover the thermodynamical properties of a schwarzschild black hole without euclideanization .
we will do so in detail in section [ geothermo ] but first we establish the geometrical significance of by and lw hamiltonians in the next section .
a study of the geometrical roles of the by and lw hamiltonians provides the geometrical basis for the thermodynamical parameters associated with a black hole that were discussed in the preceeding section . in this section
we begin by setting up the hamiltonian formulation appropriate for the two sets of boundary conditions that lead to the by and lw hamiltonians as being the unconstrained hamiltonians that generate time - evolution of foliations in schwarzschild spacetime .
the notation follows that of kucha @xcite and lw . a general spherically symmetric spacetime metric on the manifold @xmath65 can be written in the adm form as @xmath66 where @xmath67 , @xmath68 , @xmath69 and @xmath70 are functions of @xmath24 and @xmath71 only , and @xmath72 is the metric on the unit two - sphere .
we will choose our boundary conditions in such a way that the radial proper distance @xmath73 on the constant @xmath24 surfaces is finite .
this implies that the radial coordinate @xmath71 have a finite range , which we take to be @xmath74 $ ] , without any loss of generality .
the spatial metric and the spacetime metric will be assumed to be nondegenerate , in particular , @xmath69 , @xmath70 , and @xmath67 are taken to be positive . for the metric ( [ 4-metric ] ) , the einstein - hilbert action is @xmath75 & & \nonumber \\
= \int dt \int_0 ^ 1 d r \ , \bigg [ & & -n^{-1 } \left ( r \bigl ( - { \dot \lambda } + ( \lambda n^r ) ' \bigr ) ( - { \dot r } + r ' n^r ) + \case{1}{2 } \lambda { ( - { \dot r } + r ' n^r ) } ^2
\right ) \nonumber \\ & & + n \left ( - \lambda^{-1 } r r '' + \lambda^{-2 } r r ' \lambda ' - \case{1}{2 } \lambda^{-1 } { r'}^{2 } + \case{1}{2 } \lambda \right ) \bigg ] \ \ , \label{s - lag}\end{aligned}\ ] ] where the subscript @xmath17 denotes that @xmath76 is a hypersurface action that is defined only up to the possible addition of boundary terms .
above , the overdot and the prime denote @xmath77 and @xmath78 , respectively .
the equations of motion derived from ( [ s - lag ] ) are the full einstein equations for the metric ( [ 4-metric ] ) , and they imply that every classical solution is part of a maximally extended schwarzschild spacetime , where the value of the schwarzschild mass @xmath79 may be positive , negative , or zero . in what follows , we will choose our boundary conditions such that @xmath80 .
we shall discuss the boundary conditions and the boundary terms after passing to the hamiltonian formulation .
the momenta conjugate to @xmath69 and @xmath70 are found from the lagrangian action ( [ s - lag ] ) to be @xmath81 a dual - legendre transformation then yields the hamiltonian action @xmath82 = \int dt \int_0 ^ 1 dr \left ( p_\lambda { \dot \lambda } + p_r { \dot r } - nh - n^r h_r \right ) \ \ , \label{s - ham}\ ] ] where the super - hamiltonian @xmath83 and the radial supermomentum @xmath84 are @xmath85 it can be verified that the poisson brackets of the constraints close according to the radial version of the dirac algebra @xcite .
we next consider the boundary terms that must be added to the hypersurface action ( [ canhypaction ] ) for the total action to yield , upon variation , only a volume term corresponding to the equations of motion .
however , the boundary terms depend intricately on the choice of the spacetime foliation .
different foliations require different boundary conditions on the geometric variables in the variational analysis , thus requiring the addition of different boundary terms to ( [ canhypaction ] ) .
as is well known in general relativity , it is these boundary terms that determine the true hamiltonian of the system .
hence , as we show below explicitly , this implies that different foliations correspond to different hamiltonians , which are the generators of time - evolution of the spatial slices in the foliations . in general , the analysis of brown and york ( see subsection [ subsec : byham ] ) breaks down in cases where spacetime regions of non - trivial topologies are enclosed inside the spherical box , particularly so in the case of a schwarzschild black hole enclosed inside the box .
as we show below , in this case one is forced to introduce an `` inner''-boundary where the spatial hypersurfaces of brown and york must extend to .
this fact becomes transparent when one looks at the full kruskal extension of the schwarzschild metric .
there , one begins by performing a @xmath86 decomposition of the spacetime in terms of a one - parameter family of spatial hypersurfaces . the kruskal diagram ( see figs
. 2 and 3 ) shows that any such foliation would necessarily require two boundaries : an outer boundary and an inner boundary .
the hamiltonian that evolves these spatial hypersurfaces in time will in general depend on the boundary conditions specified on these 2-boundaries . in this section
we show that the spatial hypersurfaces that are evolved by the hamiltonian corresponding to the quasilocal energy , given in ( [ quasih ] ) , are ones that extend from the box ( the outer timelike boundary ) , on the right end , to an inner timelike boundary located completely inside the dynamical region of the kruskal diagram , on the left end .
we first find the boundary conditions and the foliation that correspond to time evolution by the by hamiltonian and later we compare these with those corresponding to the lw hamiltonian . at @xmath87 ,
we prescribe the following fall - off conditions [ s - r ] @xmath88 where @xmath89 and @xmath90 are positive .
this ensures that on classical solutions @xmath79 is positive .
also @xmath91 .
here @xmath92 stands for a term whose magnitude at @xmath93 is bounded by @xmath94 times a constant , and whose @xmath95th derivative at @xmath93 is similarly bounded by @xmath96 times a constant for @xmath97 .
the fact that the shift @xmath98 vanishes as @xmath99 implies that on solutions , the inner timelike boundary lies along a constant killing - time surface located completely in the past and the future dynamical regions and cutting across the bifurcation two - sphere . also , on solutions , the variable @xmath90 corresponds to the throat - radius .
it is straightforward to verify that the conditions ( [ s - r ] ) are consistent with the equations of motion : provided that the constraints obey @xmath100 and the fall - off conditions ( [ s - r - lambda])([s - r - pr ] ) hold for the initial data , and provided that the lapse and shift satisfy ( [ s - r - n ] ) and ( [ s - r - nr ] ) , it then follows that the fall - off conditions ( [ s - r - lambda])([s - r - pr ] ) are preserved in time by the time - evolution equations . on the other hand , at @xmath101 , the boundary conditions are as follows : we fix @xmath70 and @xmath102 to be prescribed positive - valued functions of @xmath24 .
this means fixing the metric on the three - surface @xmath101 to be timelike . in the classical solutions ,
the surface @xmath101 is located in the right exterior region of the kruskal extension of the schwarzschild spacetime .
we now give an action principle appropriate for these boundary conditions .
to begin , note that the surface action @xmath103 $ ] in ( [ s - ham ] ) is well defined under the above conditions .
consider the total action @xmath104 = s_\sigma [ \lambda , r , p_\lambda , p_r ; n , n^r ] + s_{\partial\sigma } [ \lambda , r , p_\lambda , p_r ; n , n^r ] \ \ , \label{s - total}\ ] ] where the boundary action is given by @xmath105 \nonumber \\ & & = \int dt { \biggl [ n r r ' \lambda^{-1 } - n^r \lambda p_\lambda - \case{1}{2 } r { \dot r } \ln \left| { n + \lambda n^r \over n - \lambda n^r } \right| \biggr]}_{r=1 } \ \ , \label{s - boundary}\end{aligned}\ ] ] where @xmath106_a$ ] is value of the _ term _ evaluated at @xmath107 .
the variation of the total action ( [ s - total ] ) can be written as a sum of a volume term proportional to the equations of motion , boundary terms from the initial and final spatial surfaces , and boundary terms from @xmath87 and @xmath101 .
the boundary terms from the initial and final spatial surfaces take the usual form @xmath108 with the upper ( lower ) sign corresponding to the final ( initial ) surface .
these terms vanish provided we fix the initial and final three - metrics . the boundary term from @xmath87 vanishes under the fall - off conditions specified in ( [ s - r ] ) . as will be shown in subsection [ subsec : reduction ] ,
this is crucial in obtaining a reduced hamiltonian that corresponds to the correct quasilocal energy .
the boundary term from @xmath101 reads @xmath109}_{r=1 } \ \ . \label{bt-1}\end{aligned}\ ] ] since @xmath70 and @xmath110 are fixed at @xmath101 , the first three terms in ( [ bt-1 ] ) vanish .
the integrand in the last term in ( [ bt-1 ] ) is proportional to the equation of motion ( [ plambda ] ) , which is classically enforced for @xmath111 by the volume term in the variation of the action .
therefore , for classical solutions , also the last term in ( [ bt-1 ] ) will vanish by continuity .
thus the action ( [ s - total ] ) is appropriate for a variational principle which fixes the initial and final three - metrics , and the three - metric on the timelike boundary at @xmath101 .
each classical solution belongs to that region of a kruskal diagram that lies within two timelike boundaries such that the inner boundary lies along a constant killing - time surface located completely in the dynamical regions and the outer boundary is a timelike surface located in the right exterior region ( see fig .
the constant @xmath24 slices are spacelike everywhere between the two timelike boundaries . to obtain the reduced action and extract the unconstrained hamiltonian system one needs to first solve the constraints ( [ superham ] ) and ( [ supermom ] ) . in the following , we will follow kucha s way of handling the constraints @xcite .
it was shown by kucha that in the context of a vacuum schwarzschild spacetime ( in the absence of timelike boundaries ) there exists a set of new variables , which are related to the adm variables through a canonical transformation , such that in terms of the new variables the constraints are remarkably simple and solvable .
this allows one to perform a hamiltonian reduction . in this section
we show that the canonical transformation given by kucha from the adm variables @xmath112 to the new variables @xmath113 is readily adapted to our boundary conditions . as mentioned earlier
, the boundary conditions ensure that @xmath80 .
recall from @xcite that the new variables @xmath113 are defined by [ trans ] @xmath114 where @xmath115 in the classical solution , @xmath79 is the value of the schwarzschild mass and @xmath116 is the derivative of the killing time with respect to @xmath71 .
a pair of quantities which will become new lagrange multipliers are defined by [ n - def ] @xmath117 using arguments similar to kucha and lw , it can be shown that under our boundary conditions the transformation ( [ trans ] ) is a canonical transformation , which is also invertible @xcite . the hamiltonian action ( [ s - ham ] )
can now be written in terms of the new variables .
using eqs .
( [ n - def ] ) , one sees that the constraint terms @xmath118 in the old surface action ( [ s - ham ] ) take the form @xmath119 .
thus the new surface action is @xmath120 = \int dt \int _ { 0}^{1 } dr \left ( p_m { \dot m } + p_{\sf r } \dot{\sf r } + 4 { \sf n } m m ' -n^{\sf r } p_{\sf r } \right ) \
\ , \label{s2-ham}\ ] ] where the quantities to be varied independently are @xmath79 , @xmath121 , @xmath122 , @xmath123 , @xmath124 , and @xmath125 .
the complete set of equations of motion is [ eom2 ] @xmath126 we now turn to the boundary conditions and boundary terms . as a preparation for this ,
let us denote by @xmath127 the quantity @xmath128 when expressed as a function of the new canonical variables and lagrange multipliers .
a short calculation using ( [ trans])([n - def ] ) yields @xmath129 in general , @xmath127 need not be positive for all values of @xmath71 , even for classical solutions .
however , as in subsection [ subsec : bygeo ] , we shall introduce boundary conditions that fix the intrinsic metric on the three - surface @xmath101 to be timelike , and under such boundary conditions @xmath127 is positive at @xmath101 .
consider now the total action @xmath130 = s_\sigma [ m , { \sf r } , p_m , p_{\sf r } ; { \sf n } , n^{\sf r } ] + s_{\partial\sigma } [ m , { \sf r } , p_m , p_{\sf r } ; { \sf n } , n^{\sf r } ] \ \ , \label{s2-total}\ ] ] where the boundary action is given by @xmath131 \nonumber \\
& = & \int dt { \left [ { \sf r } \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } + \case{1}{2 } { \sf r } \dot{\sf r } \ln \left ( { \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } - \dot{\sf r } \over \sqrt{{\sf
f}q^2 + { \dot{\sf r}}^2 } + \dot{\sf r } } \right ) \vphantom { { \left| { \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } - \dot{\sf r } \over \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } + \dot{\sf r } } \right|}^q_q } \right]}_{r=1 } \label{s2-boundary}\end{aligned}\ ] ] with @xmath132 .
note that the argument of the logarithm in ( [ s2-boundary ] ) is always positive .
the variation of ( [ s2-total ] ) contains a volume term proportional to the equations of motion , as well as several boundary terms . from the initial and final spatial surfaces one gets the usual boundary terms @xmath133
which vanish provided we fix @xmath79 and @xmath121 on these surfaces .
similarly one can show that with our choice of boundary conditions ( given in section [ subsec : bygeo ] ) the remaining boundary terms from the timelike surfaces at @xmath87 and @xmath101 also vanish .
we now concentrate on the variational principle associated with the action @xmath134 $ ] ( [ s2-total ] ) .
we shall reduce the action to the true dynamical degrees of freedom by solving the constraints .
the constraint @xmath135 ( [ eom2-mm ] ) implies that @xmath79 is independent of @xmath71 .
we can therefore write @xmath136 substituting this and the constraint @xmath137 ( [ eom2-psfr ] ) back into ( [ s2-total ] ) yields the true hamiltonian action @xmath138 = \int dt \left ( { \bf p } { \dot { \bf m } } - { \bf h } \right ) \ \ , \label{s - red}\ ] ] where @xmath139 the reduced hamiltonian @xmath140 in ( [ s - red ] ) takes the form @xmath141 here @xmath142 and @xmath143 are the values of @xmath121 and @xmath127 at the timelike boundary @xmath101 , and @xmath144 . as mentioned before , @xmath142 and @xmath143
are prescribed functions of time , satisfying @xmath145 and @xmath146 .
note that @xmath140 is , in general , explicitly time - dependent .
the variational principle associated with the reduced action ( [ s - red ] ) fixes the initial and final values of @xmath15 .
the equations of motion are [ red - eom ] @xmath147 equation ( [ red - eom1 ] ) is readily understood in terms of the statement that @xmath15 is classically equal to the time - independent value of the schwarzschild mass . to interpret equation ( [ red - eom2 ] ) , recall from sec .
[ subsec : transformation ] that @xmath116 equals classically the derivative of the killing time with respect to @xmath71 , and @xmath148 therefore equals by ( [ bfp ] ) the difference of the killing times at the left and right ends of the constant @xmath24 surface .
as the constant @xmath24 surface evolves in the schwarzschild spacetime , ( [ red - eom2 ] ) gives the negative of the evolution rate of the killing time at the right end of the spatial surface where it terminates at the outer timelike boundary at @xmath101 .
note that @xmath149 gets no contribution from the inner timelike boundary located at @xmath87 in the dynamical region .
this is a consequence of the fall - off conditions ( [ s - r ] ) which ensure that on solutions , the rate of evolution of the killing time at @xmath87 is zero .
the case of interest is when the radius of the ` outer ' boundary two - sphere does not change in time , i.e. , @xmath150 . in that case
the second term in @xmath151 ( [ bfhb ] ) vanishes , and @xmath149 in ( [ red - eom2 ] ) is readily understood in terms of the killing time of a static schwarzschild observer , expressed as a function of the proper time @xmath152 and the blueshift factor @xmath153 .
the reduced hamiltonian is given by @xmath154 where @xmath16 is the time - independent value of @xmath142 .
unfortunately , the above hamiltonian does not vanish as @xmath15 goes to zero .
the situation is remedied by adding the @xmath155 term of gibbons and hawking @xcite to @xmath140 .
physically , this added term arises from the extrinsic curvature of the ` outer ' boundary two - sphere when embedded in flat spacetime . with the added term ,
the hamiltonian becomes @xmath156 this is the quasilocal energy of brown and york when @xmath31 .
the choice of @xmath22 determines the choice of time in the above hamiltonian .
setting @xmath157 geometrically means choosing a spacetime foliation in which the rate of evolution of the spatial hypersurface on the box is the same as that of the proper time .
then the new hamiltonian is @xmath158 namely , the quasilocal energy ( [ quasih ] ) in schwarzschild spacetime . in the next section ,
we discuss the geometric relevance of the lw hamiltonian that , as we showed earlier , yields the correct free energy of the system .
we now summarize the lw choice of the foliation of the schwarzschild spacetime , state the corresponding boundary conditions they imposed , and briefly mention how they obtain their reduced hamiltonian .
the main purpose of this section is to facilitate a comparison between the lw boundary conditions and our choice of the boundary conditions ( as discussed in the preceeding subsections ) that yield the by hamiltonian . as shown in fig .
2 , lw considered a foliation in which the spatial hypersurfaces are restricted to lie in the right exterior region of the kruskal diagram .
each spatial hypersurface in this region extends from the box at the right end up to the bifurcation 2-sphere on the left end .
the boundary conditions imposed by lw are as follows . at @xmath93
, they adopt the fall - off conditions [ ls - r ] @xmath159 where @xmath89 and @xmath90 are positive , and @xmath91
. equations ( [ ls - r - lambda ] ) and ( [ ls - r - r ] ) imply that the classical solutions have a positive value of the schwarzschild mass , and that the constant @xmath24 slices at @xmath93 are asymptotic to surfaces of constant killing time in the right hand side exterior region in the kruskal diagram , all approaching the bifurcation two - sphere as @xmath93 .
the spacetime metric has thus a coordinate singularity at @xmath93 , but this singularity is quite precisely controlled . in particular , on a classical solution
the future unit normal to a constant @xmath24 surface defines at @xmath93 a future timelike unit vector @xmath160 at the bifurcation two - sphere of the schwarzschild spacetime , and the evolution of the constant @xmath24 surfaces boosts this vector at the rate given by @xmath161 at @xmath101 , we fix @xmath70 and @xmath102 to be prescribed positive - valued functions of @xmath24 .
this means fixing the metric on the three - surface @xmath101 , and in particular fixing this metric to be timelike . in the classical solutions ,
the surface @xmath101 is located in the right hand side exterior region of the kruskal diagram . to obtain an action principle appropriate for these boundary conditions ,
consider the total action @xmath104 = s_\sigma [ \lambda , r , p_\lambda , p_r ; n , n^r ] + s_{\partial\sigma } [ \lambda , r , p_\lambda , p_r ; n , n^r ] \ \ , \label{ls - total}\ ] ] where the boundary action is given by @xmath105 \nonumber \\ & & = \case{1}{2 } \int dt \
, { \left [ r^2 n ' \lambda^{-1 } \right]}_{r=0 } \ ; \ ; + \int dt { \biggl [ n r r ' \lambda^{-1 } - n^r \lambda p_\lambda - \case{1}{2 } r { \dot r } \ln \left| { n + \lambda n^r \over n - \lambda n^r } \right| \biggr]}_{r=1 } \ \ .
\label{ls - boundary}\end{aligned}\ ] ] the variation of the total action ( [ ls - total ] ) can be written as a sum of a volume term proportional to the equations of motion , boundary terms from the initial and final spatial surfaces , and boundary terms from @xmath87 and @xmath101 . to make the action ( [ ls - total ] ) appropriate for a variational principle ,
one fixes the initial and final three - metrics , the box - radius @xmath70 , and the three - metric on the timelike boundary at @xmath101 .
these are similar to the boundary conditions that we imposed to obtain the by hamiltonian . however , for the lw boundary conditions , one has to also fix the quantity @xmath162 at the bifurcation 2-sphere .
each classical solution is part of the right hand exterior region of a kruskal diagram , with the constant @xmath24 slices approaching the bifurcation two - sphere as @xmath93 , and @xmath163 giving via ( [ n - boost ] ) the rate of change of the unit normal to the constant @xmath24 surfaces at the bifurcation two - sphere .
although we are here using geometrized units , the argument of the @xmath164 in ( [ n - boost ] ) is a truly dimensionless boost parameter " even in physical units . to obtain the ( reduced ) lw hamiltonian , one needs to solve the super - hamiltonian and the supermomentum constraints . just as in the case of the by hamiltonian ( see subsection [ subsec : transformation ] ) , it helps to first make a canonical transformation to the kucha variables ( see lw for details ) . after solving the constraints , one obtains the following reduced action @xmath165 = \int dt \left ( { \bf p } { \dot { \bf m } } - { \bf h } \right ) \ \ , \label{ls - red}\ ] ] where @xmath166 and the reduced hamiltonian @xmath167 in ( [ ls - red ] ) is @xmath168 which is the same as the one given in eq .
( [ lwht ] ) . in obtaining the above reduced form @xmath169
, we have assumed that the box - radius is constant in time , @xmath170 , just as we did in obtaining the by hamiltonian ( [ byhq ] ) .
having established the geometrical significance of the by and lw hamiltonians , the basis for their thermodynamical roles becomes apparent .
we showed that the by hamiltonian evolves spatial hypersurfaces in such a way that they span the spacetime region both inside and outside the event horizon .
this is what one would expect from the fact that it corresponds to the quasilocal energy of the complete spacetime region enclosed inside the box . on the other hand ,
the lw hamiltonian evolves spatial slices that are restricted to lie outside the event horizon . with our choice of the boost parameter @xmath23
, this corresponds to the helmholtz free energy of the system , which is less than the quasilocal energy : this is expected since the lw slices span a smaller region of the spacetime compared to the by slices .
also , the fact that the lw slices are limited to lie outside the event horizon implies that the energy on these slices can be harnessed by an observer located outside the box .
this is consistent with the fact that it corresponds to the helmholtz free energy of the system which is the amount of energy in a system that is available for doing work by the system on its surroundings . using the thermodynamical roles played by the lw hamiltonian and the by hamiltonian ( see section [ sec : thermoconsi ] ) ,
we now derive , at the classical level , many of the thermodynamical quantities associated with the schwarzschild black hole enclosed inside a box .
we begin by finding the temperature on the box . from eq .
( [ lwh ] ) , the helmholtz free energy is @xmath171 the above equation , along with eqs .
( [ f ] ) and ( [ e ] ) , implies that @xmath172 or , @xmath173 where @xmath174 .
equation ( [ s1 ] ) gives an expression for the entropy in terms of the geometrical quantity @xmath23 . on the other hand one can find @xmath48 also from the thermodynamic identity @xmath175 where @xmath0 is the partition function defined by eq .
( [ partitionh ] ) and eq .
( [ lwhf1 ] ) .
( [ s2 ] ) gives the entropy to be @xmath176 the above equation gives another expression for the entropy , now in terms of the derivative of @xmath23 .
comparing eqs .
( [ s1 ] ) and ( [ s3 ] ) we find @xmath177 where @xmath178 is some undetermined quantity that is independent of @xmath5
. the exact form of @xmath47 as a function of @xmath15 is found by noting that the free energy @xmath49 should be a minimum at equilibrium . since in a canonical ensemble the box - radius @xmath70 and the temperature @xmath47 ( which is proportional to @xmath23 )
are fixed , the only quantity in @xmath49 that can vary is @xmath15 . thus the question we ask is the following : for a fixed value of the curvature radius @xmath20 and the boost parameter @xmath23 , what is the value of @xmath15 that minimizes @xmath49 ?
> from the expression for @xmath49 in eq .
( [ lwhf1 ] ) one finds this value of @xmath15 , to be a function of @xmath23 .
inverting this relation gives @xmath179 > from ( [ n0beta ] ) and ( [ n0 m ] ) we find that the equilibrium temperature on a box of radius @xmath16 enclosing a black hole of mass @xmath15 obeys @xmath180 in agreement with known results .
significantly , eq . ( [ n0beta ] ) shows that the equilibrium temperature geometrically corresponds to a particular value of the boost parameter .
> from eqs .
( [ s1 ] ) and ( [ n0beta ] ) we find that the entropy of a schwarzschild black hole is quadratic in its mass .
unfortunately , in this formalism one can not determine the correct constants of proportionality in @xmath48 and @xmath47 .
however , notice that our derivation is purely classical .
although simple mathematically , this derivation is incomplete due to the lack of the constant of proportionality @xmath178 in eq .
( [ n0beta ] ) . the correct value for this constant , @xmath181 , can be obtained only from a quantum treatment .
finally , we note that for the spatial slices that obey @xmath182 , the free energy can be obtained from ( [ lwh ] ) to be @xmath183 the above equation shows that if the radius of the box @xmath16 is kept fixed , then the free energy of the system is minimum for the configuration with a black hole of mass @xmath184 .
in this work our goal was to seek a geometrical basis for the thermodynamical aspects of a black hole .
we find that the value of the brown - york hamiltonian can be interpreted as the internal energy of a black hole inside a box . whereas the value of the louko - whiting hamiltonian gives the helmholtz free energy of the system . after finding these thermodynamical roles played by the by and lw hamiltonians
, we ask what the geometrical significance of these hamiltonians is . in this regard the geometrical role of the lw hamiltonian was already known .
it was recently shown by lw that their hamiltonian evolves spatial hypersurfaces in a special foliation of the kruskal diagram .
the characteristic feature of this foliation is that it is limited to only the right exterior region of this spacetime ( see fig .
2 ) and the spatial hypersurfaces are required to converge onto the bifurcation 2-sphere , which acts as their inner boundary ( the box itself being the outer boundary ) . on the other hand , the geometrical significance of the by hamiltonian as applied to the black hole case was not fully known , although it had been argued that its value is the energy of the schwarzschild spacetime region that is enclosed inside a spherical box . in this work
we establish the geometric role of the by hamiltonian by showing that it is the generator of time - evolution of spatial hypersurfaces in certain foliations of the schwarzschild spacetime . establishing the thermodynamic connection of the by and lw hamiltonians allowed us to obtain a geometrical interpretation for the equilibrium temperature of a black hole enclosed inside a box , i.e. , as measured by a stationary observer on the box .
geometrically , the temperature turns out to be the rate at which the lw spatial hypersurfaces are boosted at the bifurcation 2-sphere .
one could however ask what happens if the lw hypersurfaces are evolved at a different rate , i.e. , if the label time @xmath24 is chosen to be boosted with respect to the proper time of a stationary observer on the box . in that case , it can be shown that the by hamiltonian and the rate at which the lw hypersurfaces are evolved at the bifurcation 2-sphere get `` blue - shifted '' by the appropriate boost - factor . on the other hand , the entropy of the system can still be interpreted as the change in free energy per unit change in the temperature of the system .
we thank abhay ashtekar , viqar husain , eric martinez , jorg pullin , lee smolin , and jim york for helpful discussions .
we would especially like to thank jorma louko for critically reading the manuscript and making valuable comments .
financial support from iucaa is gratefully acknowledged by one of us ( sb ) .
this work was supported in part by nsf grant no .
phy-95 - 07740 .
the approach we describe above in studying the thermodynamics of 4d spherically symmetric einstein gravity can also be extended to the case of the 2d vacuum dilatonic black hole @xcite in an analogous fashion . in the case of a 2d black hole ,
the event horizon is located at a curvature radius @xmath185 , where @xmath186 is a positive constant that sets the length - scale in the 2d models .
the quasilocal energy of a system comprising of such a black hole in the presence of a timelike boundary situated at a curvature radius @xmath16 can be shown to be @xmath187 which strongly resembles the 4d counterpart in ( [ quasih ] ) .
@xmath188 evolves constant @xmath24 spatial hypersurfaces that extend from an inner timelike boundary lying on a constant killing - time surface in the dynamical region up to a timelike boundary ( the box ) placed in the right exterior region ( see fig .
3 ) . the hamiltonian that evolves the two - dimensional counterpart of the louko and whiting spatial slices that extend from the bifurcation point up to the box ( see fig .
2 ) is @xmath189 where in general @xmath22 and @xmath190 are functions of time .
the above hamiltonian @xmath191 was found in ref .
there it was found that @xmath192 is the rate at which the spatial hypersurface are boosted at the bifurcation point . on the other hand , @xmath193 , @xmath26 being the time - time component of the spacetime metric on the box .
if one restricts the spatial hypersurfaces to approach the box along constant proper - time hypersurfaces , then , as in 4d , @xmath31 . using fall - off conditions on the adm variables at the bifurcation point analogous to the lw fall - off conditions ( [ ls - r ] ) , it can be shown that on solutions @xcite @xmath194 where @xmath33 is the killing time , and @xmath195 is the surface gravity of a witten black hole .
the time - evolution of these restricted spatial hypersurfaces is given by the hamiltonian @xmath196 like the 4d case , here too it can be shown that @xmath188 is analogous to the internal energy , whereas @xmath197 denotes the helmholtz free energy of the 2d system .
a similar analysis also shows that @xmath198 and @xmath199 which is inverse of the blue - shifted temperature on the box .
the temperature of a 2d black hole at infinity on the other hand is @xmath200 , which is independent of the black hole mass .
in section [ sec : thermoconsi ] , we found a choice of spatial hypersurfaces that were evolved by the by hamiltonian under a specific set of boundary conditions . in this appendix
we find a different choice of spatial hypersurfaces , i.e. , with a different inner boundary , that is evolved by the by hamiltonian under a different set of boundary conditions .
we begin by stating the boundary conditions and specifying the spacetime foliation they define . at the inner boundary @xmath87 ,
we fix @xmath70 and @xmath201 to be prescribed positive - valued functions of @xmath24 .
this means fixing the metric on the three - surface @xmath87 , and in particular fixing this metric to be spacelike there . on the other hand , at @xmath101 , we fix @xmath70 and @xmath102 to be prescribed positive - valued functions of @xmath24 .
this means fixing the metric on the three - surface @xmath101 to be timelike . in the classical solutions ,
the surface @xmath101 is located in the right exterior region of the kruskal diagram .
we now wish to give an action principle appropriate for these boundary conditions .
note that the surface action @xmath103 $ ] given by eq .
( [ s - ham ] ) is well defined under the above conditions .
consider the total action @xmath104 = s_\sigma [ \lambda , r , p_\lambda , p_r ; n , n^r ] + s_{\partial\sigma } [ \lambda , r , p_\lambda , p_r ; n , n^r ] \ \ , \label{app : s - total}\ ] ] where the boundary action is given by @xmath105 \nonumber \\ & & = \int dt { \biggl [ n r r ' \lambda^{-1 } - n^r \lambda p_\lambda - \case{1}{2 } r { \dot r } \ln \left| { n + \lambda n^r \over n - \lambda n^r } \right| \biggr]}_{r=0}^{r=1 } \ \ , \label{app : s - boundary}\end{aligned}\ ] ] where @xmath106_a^b$ ] implies the difference in the values of the _ term _ evaluated at @xmath202 and at @xmath107 .
the variation of the total action ( [ app : s - total ] ) can be written as a sum of a volume term proportional to the equations of motion , boundary terms from the initial and final spatial surfaces , and boundary terms from @xmath87 and @xmath101 .
the boundary terms from the initial and final spatial surfaces take the usual form @xmath203 with the upper ( lower ) sign corresponding to the final ( initial ) surface .
these terms vanish provided we fix the initial and final three - metrics .
the boundary term from @xmath87 and @xmath101 read @xmath109}_{r=0}^{r=1 } \ \ , \label{app : bt-1}\end{aligned}\ ] ] where @xmath106^b_a$ ] implies the difference between the values of the _ term _ evaluated at @xmath202 and at @xmath107 . as @xmath70 and @xmath110 are fixed at @xmath87 and @xmath101 ,
the first three terms in ( [ app : bt-1 ] ) vanish .
the integrand in the last term in ( [ app : bt-1 ] ) is proportional to the equation of motion ( [ plambda ] ) , which is classically enforced for @xmath111 by the volume term in the variation of the action .
therefore , for classical solutions , also the last term in ( [ app : bt-1 ] ) will vanish by continuity .
we thus conclude that the action ( [ app : s - total ] ) is appropriate for a variational principle which fixes the initial and final three - metrics , the three - metric on the spacelike boundary at @xmath87 and the three - metric on the timelike boundary at @xmath101 .
each classical solution belongs to that region of a kruskal diagram that lies to the future of the null line at killing time @xmath204 .
the constant @xmath24 slices are spacelike everywhere between the spacelike boundary at @xmath87 , and the timelike boundary at @xmath101 . the canonical transformation given in kucha @xcite from the variables @xmath112 to the new variables @xmath113 is readily adapted to our boundary conditions .
as mentioned earlier , we shall assume that @xmath80 .
recall that the new variables @xmath113 have been defined in subsection [ subsec : transformation ] by equations ( [ trans ] ) and ( [ f - def ] ) .
the new lagrange multipliers are defined in eq .
( [ n - def ] ) .
it can be shown that the transformation ( [ trans ] ) is a canonical transformation also under the new boundary conditions being considered in this appendix .
we wish to write an action in terms of the new variables .
using eqs .
( [ n - def ] ) , one finds that the constraint terms @xmath118 in the old surface action ( [ s - ham ] ) take the form @xmath119 and the new surface action is the same as that given in ( [ s2-ham ] ) .
therefore , the equations of motion remain unchanged and are given by ( [ eom2 ] ) .
we now turn to the boundary conditions and boundary terms . as
before , we define @xmath205 in general , @xmath127 need not be positive for all values of @xmath71 , even for classical solutions . as in the preceeding section ,
we shall introduce boundary conditions that fix the intrinsic metric of the three - surfaces @xmath87 and @xmath101 to be spacelike and timelike , respectively , and under such boundary conditions @xmath127 is negative at @xmath87 but positive at @xmath101 . from ( [ app : q2 ] ) it is then seen that @xmath124 is nonzero at @xmath87 and @xmath101 .
recalling that we are assuming @xmath206 , eq .
( [ sfn - def ] ) shows that @xmath124 is positive at @xmath101 for classical solutions with the schwarzschild slicing , since in this slicing one has @xmath207 .
continuity then implies that @xmath124 must be positive at @xmath101 for all classical solutions compatible with our boundary conditions . on the other hand , at @xmath87 we now put the additional condition that @xmath208 ( or , equivalently , @xmath209 ) . on classical solutions ,
this extra condition restricts the surface @xmath87 to lie either in the past or the future dynamical region of the schwarzschild spacetime .
although the final expression for the quasilocal energy is independent of this choice , for definiteness we will choose the spatial boundary at @xmath87 to lie in the future dynamical region ( see fig .
4 ) . then at @xmath87 , eq .
( [ sfn - def ] ) shows that @xmath124 has to be negative because @xmath210 there .
we can therefore , without loss of generality , choose to work in a neighborhood of the classical solutions such that @xmath124 is positive at @xmath101 whereas @xmath124 is negative at @xmath87 .
consider now the total action @xmath130 = s_\sigma [ m , { \sf r } , p_m , p_{\sf r } ; { \sf n } , n^{\sf r } ] + s_{\partial\sigma } [ m , { \sf r } , p_m , p_{\sf r } ; { \sf n } , n^{\sf r } ] \ \ , \label{app : s2-total}\ ] ] where the boundary action is given by @xmath131 \nonumber \\ & = & \int dt { \left [ { \sf r } \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } + \case{1}{2 } { \sf r } \dot{\sf r } \ln \left ( { \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } - \dot{\sf r } \over \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } + \dot{\sf r } } \right ) \vphantom { { \left| { \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } - \dot{\sf r } \over \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } + \dot{\sf r } } \right|}^q_q } \right]}_{r=1 } \nonumber \\ & -&\int dt { \left [ { \sf r } \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } + \case{1}{2 } { \sf r } \dot{\sf r } \ln \left ( { \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } - \dot{\sf r } \over \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } + \dot{\sf r } } \right ) \vphantom { { \left| { \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } - \dot{\sf r } \over \sqrt{{\sf f}q^2 + { \dot{\sf r}}^2 } + \dot{\sf r } } \right|}^q_q } \right]}_{r=0 } \label{app : s2-boundary}\end{aligned}\ ] ] with @xmath132 .
note that the argument of the logarithm in ( [ app : s2-boundary ] ) is always positive .
the variation of ( [ app : s2-total ] ) contains a volume term proportional to the equations of motion , as well as several boundary terms .
these boundary terms vanish if on the initial and final three - surfaces we fix the new canonical coordinates @xmath79 and @xmath121 , and at @xmath87 and @xmath101 we fix @xmath121 and the intrinsic metric on these three - surfaces .
we now reduce the action @xmath134 $ ] ( [ s2-total ] ) to the true dynamical degrees of freedom by solving the constraints ( [ eom2-mm ] ) and ( [ eom2-psfr ] ) as before .
this gives the true hamiltonian action to be @xmath138 = \int dt \left ( { \bf p } { \dot { \bf m } } - { \bf h } \right ) \ \ , \label{app : s - red}\ ] ] where @xmath148 and @xmath15 are defined as in section [ subsec : reduction ] . the reduced hamiltonian @xmath167 in ( [ app : s - red ] ) takes the form @xmath211 with here @xmath142 ( @xmath213 ) and @xmath143 ( @xmath214 ) are the values of @xmath121 and @xmath127 at the timelike ( spacelike ) boundary @xmath101 ( @xmath87 ) , and @xmath144 .
@xmath142 , @xmath143 , @xmath213 , and @xmath214 are considered to be prescribed functions of time , satisfying @xmath215 and @xmath216 .
note that @xmath140 is , in general , explicitly time - dependent .
the interpretation of ( [ app : red - eom1 ] ) remains unchanged .
to interpret equation ( [ app : red - eom2 ] ) , note that @xmath148 equals by ( [ bfp ] ) the difference of the killing times at the left and right ends of the constant @xmath24 surface .
as the constant @xmath24 surface evolves in the schwarzschild spacetime , the first term in ( [ app : red - eom2 ] ) gives the evolution rate of the killing time at the left end of the hypersurface , where the hypersurface terminates at a spacelike surface located completely in the future dynamical region ( see fig .
4 ) . the second term in ( [ app : red - eom2 ] )
gives the negative of the evolution rate of the killing time at the right end of the surface , where the surface terminates at the timelike boundary .
the two terms are generated respectively by @xmath218 ( [ app : bfhs ] ) and @xmath219 ( [ app : bfhb ] ) .
the case of interest is when the ` inner ' spacelike boundary lies on the schwarzschild singularity , i.e. , @xmath220 , and when the radius of the ` outer ' boundary two - sphere does not change in time , @xmath150 . in that case @xmath218
( [ app : bfhs ] ) and the second term in @xmath151 ( [ app : bfhb ] ) vanish .
one can also make the first term in @xmath221 ( [ app : red - eom2 ] ) vanish provided one restricts the slices to approach the surface at @xmath87 in such a way that @xmath222 vanishes faster than @xmath223 there . the second term in ( [ app : red - eom2 ] )
is readily understood in terms of the killing time of a static schwarzschild observer , expressed as a function of the proper time @xmath152 and the blueshift factor @xmath153 .
the reduced hamiltonian is given by @xmath154 where @xmath16 is the time - independent value of @xmath142 .
following the same arguments as given in sec .
[ subsec : reduction ] , we find that the appropriate hamiltonian under the new boundary conditions of this appendix is @xmath224 which is the by hamiltonian . similarly , from the time - reversal symmetry of the kruskal extension of the schwarzschild spacetime , the by hamiltonian could also be interpreted to evolve spatial slices that extend from the box upto an inner boundary that is the past white hole spacelike singularity .
m. srednicki , phys .
lett . * 71 * , 666 ( 1993 ) ; + g. t hooft , nucl .
phys . * b256 * , 727 ( 1985 ) ; + l. susskind and j. uglum , phys .
d * 50 * , 2700 ( 1994 ) ; + m. maggiore , nucl .
* b429 * , 205 ( 1994 ) ; + c. callan and f. wilczek , phys .
* b333 * , 55 ( 1994 ) ; + s. carlip and c. teitelboim , phys .
d * 51 * , 622 ( 1995 ) ; + s. carlip , phys .
d * 51 * , 632 ( 1995 ) ; + y. peleg , `` quantum dust black holes '' , brandeis university report no.brx-th-350 , hep - th/93077057 ( 1993 ) .
figure 1 : a bounded spacetime region with boundary consisting of initial and final spatial hypersurfaces @xmath29 and @xmath30 and a timelike three - surface @xmath19 .
@xmath19 itself is the time - evolution of the two - surface @xmath18 that is the boundary of an arbitrary spatial slice @xmath17 .
figure 2 : the louko - whiting choice of a foliation of the schwarzschild spacetime .
the spatial slices of this foliation extend from the bifurcation two - sphere to the box .
the initial and final spatial hypersurfaces have label time @xmath225 and @xmath226 , respectively .
figure 3 : a choice of foliating the schwarzschild spacetime that is different from the louko - whiting choice . here
the spatial slices extend from the box to a timelike inner boundary that is located completely inside the hole . | in this work , we extend the analysis of brown and york to find the quasilocal energy in a spherical box in the schwarzschild spacetime .
quasilocal energy is the value of the hamiltonian that generates unit magnitude proper - time translations on the box orthogonal to the spatial hypersurfaces foliating the schwarzschild spacetime .
we call this hamiltonian the brown - york hamiltonian .
we find different classes of foliations that correspond to time - evolution by the brown - york hamiltonian .
we show that although the brown - york expression for the quasilocal energy is correct , one needs to supplement their derivation with an extra set of boundary conditions on the interior end of the spatial hypersurfaces inside the hole in order to obtain it from an action principle . replacing this set of boundary conditions with another set yields the louko - whiting hamiltonian , which corresponds to time - evolution of spatial hypersurfaces in a different foliation of the schwarzschild spacetime .
we argue that in the thermodynamical picture , the brown - york hamiltonian corresponds to the _ internal energy _ whereas the louko - whiting hamiltonian corresponds to the _ helmholtz free energy _ of the system .
unlike what has been the usual route to black hole thermodynamics in the past , this observation immediately allows us to obtain the partition function of such a system without resorting to any kind of euclideanization of either the hamiltonian or the action . in the process , we obtain some interesting insights into the geometrical nature of black hole thermodynamics .
= 10000 epsf |
a big achievement in the 70s-80 s was to show that , inside the axiomatic formulation of quantum mechanics , based on _ positive operator valued measures _ and _ instruments_,@xcite a consistent formulation of the theory of measurements continuous in time ( _ quantum continual measurements _ ) was possible.@xcite the main applications of quantum continual measurements are in the photon detection theory in quantum optics ( _ direct , heterodyne , homodyne detection_).@xcite a very flexible and powerful formulation of continual measurement theory was based on stochastic differential equations , of classical type ( commuting noises , it calculus ) and of quantum type ( non commuting noises , hudson - parthasarathy equation).@xcite in this paper we start by giving a short presentation of continual measurement theory based on quantum sde s .
we consider only the type of observables relevant for the description of homodyne detection and we make the mathematical simplification of introducing only bounded operators on the hilbert space of the quantum system and a finite number of noises .
then , we introduce the spectrum of the classical stochastic process which represents the output and we study the general properties of the spectra of such classical processes by proving characteristic bounds due to the heisenberg uncertainty principle . finally , we present the case of a two - level atom , where the spectral analysis of the output can reveal the phenomenon of squeezing of the fluorescence light , a phenomenon related to the heisenberg uncertainty relations .
let @xmath0 be the _ system space _
, the complex separable hilbert space associated to the observed quantum system , which we call system @xmath1 . quantum stochastic calculus and
the hudson - parthasarathy equation@xcite allow to represent the continual measurement process as an interaction of system @xmath1 with some quantum fields combined with an observation in continuous time of these fields .
let us start by introducing such fields .
we denote by @xmath2 the hilbert space associated with @xmath3 boson fields , that is the symmetric _ fock space _ over the `` one particle space '' @xmath4 , and we denote by @xmath5 , @xmath6 , the _ coherent vectors _ , whose components in the @xmath7 particle spaces are @xmath8 .
let @xmath9 be the canonical basis in @xmath10 and for any @xmath11 let us set @xmath12 .
we denote by @xmath13 , @xmath14 , @xmath15 the _ annihilation , creation and conservation processes _ : @xmath16 the annihilation and creation processes satisfy the _ canonical commutation rules _ ( ccr ) ; formally , @xmath17=t\wedge s$ ] , @xmath18=0 $ ] , @xmath19=0 $ ] .
let @xmath20 , @xmath21 , @xmath22 , @xmath23 , be bounded operators on @xmath0 such that @xmath24 and @xmath25 .
we set also @xmath26 .
then , the quantum stochastic differential equation @xcite @xmath27 with the initial condition @xmath28 , has a unique solution , which is a strongly continuous family of unitary operators on @xmath29 , representing the system - field dynamics in the interaction picture with respect to the free field evolution .
the states of a quantum system are represented by statistical operators , positive trace - class operators with trace one ; let us denote by @xmath30 the set of statistical operators on @xmath0 .
as initial state of the composed system `` system @xmath1 plus fields '' we take @xmath31 , where @xmath32 is generic and @xmath33 is a coherent state , @xmath34
. one of the main properties of the hudson - parthasarathy equation is that , with such an initial state , the reduced dynamics of system @xmath1 obeys a quantum master equation.@xcite indeed , we get @xmath35 , \qquad \eta_t:={\operatorname{tr}}_\gamma \left\ { u(t)\bigl ( \rho \otimes \varrho_\gamma(f)\bigr)u(t)^*\right\},\ ] ] where the liouville operator @xmath36 turns out to be given by @xmath37= \left(k-\sum_{kl}r_k^ * s_{kl}f_l(t)\right)\rho + \rho \left(k^*-\sum_{kj}\overline{f_j(t)}s_{kj}^{\;*}r_k\right ) \\ { } + \sum_k \left(r_k-\sum_l s_{kl}f_l(t)\right)\rho\left(r_k^*- s_{kl}^{\;*}\overline{f_l(t)}\right ) - { \left\vertf(t)\right\vert}^2\rho.\end{gathered}\ ] ] a particularly important case is @xmath38 , when @xmath36 reduces to @xmath37= -{\mathrm{i}}\left [ h-{\mathrm{i}}\sum_k f_k(t)r_k^*+ { \mathrm{i}}\sum_k\overline{f_k(t)}r_k,\,\rho \right ] \\ { } + \sum_k \left(r_k\rho r_k^*-\frac 1 2 r_k^*r_k\rho -\frac 1 2\rho r_k^*r_k\right).\end{gathered}\ ] ]
it is useful to introduce also the evolution operator from @xmath39 to @xmath40 by @xmath41 with this notation we have @xmath42 $ ] .
the key point of the theory of continual measurements is to consider field observables represented by time dependent , commuting families of selfadjoint operators in the heisenberg picture.@xcite being commuting at different times , these observables represent outputs produced at different times which can be obtained in the same experiment . here
we present a very special case of family of observables , a field quadrature .
let us start by introducing the operators @xmath43 @xmath44 $ ] and @xmath45 are fixed .
the operators @xmath46 are selfadjoint ( they are essentially selfadjoint on the linear span of the exponential vectors ) . by using ccr s , one can check that they commute : @xmath47=0 $ ] ( better : the unitary groups generated by @xmath46 and @xmath48 commute ) .
the operators have to be interpreted as linear combinations of the formal increments @xmath49 , @xmath50 which represent field operators evolving with the free - field dynamics ; therefore , they have to be intended as operators in the interaction picture .
the important point is that these operators commute for different times also in the heisenberg picture , because @xmath51 this is due to the factorization properties of the fock space and to the properties of the solution of the hudson - parthasarathy equation .
these `` output '' quadratures are our observables .
they regard those bosons in `` field 1 '' which eventually have interacted with @xmath1 between time @xmath52 and time @xmath40 .
commuting selfadjoint operators can be jointly diagonalized and the usual postulates of quantum mechanics give the probabilities for the joint measurement of the observables represented by the selfadjoint operators @xmath53 , @xmath54 . let us stress that operators of type with different angles and frequencies represent incompatible observables , because they do not commute but satisfy @xmath55=\frac{4{\mathrm{i}}\sin\left(\frac{t\wedge s}2 \left(\nu-\mu\right)\right)\sin \left(\theta-\phi+\frac{t\wedge s}2 \left(\nu-\mu\right)\right)}{\nu-\mu}\ , .\ ] ] when `` field 1 '' represents the electromagnetic field , a physical realization of a measurement of the observables is implemented by what is called an heterodyne / homodyne scheme .
the light emitted by the system in the `` channel 1 '' interferes with an intense coherent monochromatic laser beam of frequency @xmath56 . the mathematical description of the apparatus is given in section 3.5 of ref .
the commuting selfadjoint operators have a joint pvm ( projection valued measure ) which gives the distribution of probability for the measurement . anyway
, at least the finite - dimensional distributions of the output can be obtained via an explicit and easier object , the _ characteristic operator _
@xmath57 , a kind of fourier transform of this pvm . for any real test function @xmath58 and any time @xmath59
we define the unitary weyl operator @xmath60 then , there exists a measurable space @xmath61 , a pvm @xmath62 ( acting on @xmath2 ) with value space @xmath61 , a family of real valued measurable functions @xmath63 on @xmath64 , such that @xmath65 , and , for any choice of @xmath66 , @xmath67 , @xmath68 , @xmath69\biggr\ } \\ { } = \int_\omega \exp\biggl\ { { \mathrm{i}}\sum_{j=1}^n \kappa_j \big [ x(t_j;\omega ) - x(t_{j-1};\omega)\big]\biggr\ } \xi_\vartheta^\nu({\mathrm{d}}\omega)\,,\end{gathered}\ ] ] where @xmath70 . let us stress that the pvm depends on the observables and , so , on the parameters @xmath71 and @xmath56 , while the choices of the trajectory space ( the measurable space @xmath61 ) and of the process @xmath72 are independent of @xmath71 and @xmath56 .
then , we introduce the characteristic functional @xmath73 all the finite - dimensional probabilities of the increments of the process @xmath72 are determined by @xmath74\,,\end{gathered}\ ] ] where we have introduced the test function @xmath75 , with @xmath67 , @xmath68 .
the fact that the theory gives in a simple direct way the distribution for the increments of the process @xmath72 , rather than its finite - dimensional distributions , is related also to the interpretation : the output @xmath72 actually is obtained by a continuous observation of the generalized process @xmath76 followed by post - measurement processing .
starting from the characteristic functional it is possible to obtain the moments of the output process @xmath77 and to express them by means of quantities concerning only system @xmath1.@xcite let us denote by @xmath78 the expectation with respect to @xmath79 ; for the first two moments we obtain the expressions [ moments ] @xmath80={\operatorname{tr}}\left\{\left(z(t)+z(t)^*\right)\eta_t\right\},\ ] ] @xmath81 = \delta(t - s ) \\ { }
+ { \operatorname{tr}}\left\{\left(z(t_2)+z(t_2)^*\right)\upsilon(t_2,t_1)\left[z(t_1)\eta_{t_1}+\eta_{t_1 } z(t_1)^*\right ] \right\},\end{gathered}\ ] ] where @xmath82 , @xmath83 and @xmath84
in the classical theory of stochastic processes , the spectrum is related to the fourier transform of the autocorrelation function .
let @xmath85 be a stationary real stochastic process with finite moments ; then , the mean is independent of time @xmath86={\operatorname{\mathbb{e}}}[y(0)]=:m_y$ ] , @xmath87 , and the second moment is invariant under time translations @xmath88={\operatorname{\mathbb{e}}}[y(t - s)y(0)]=:r_y(t - s ) , \qquad \forall t , s\in { \mathbb{r}}\,.\ ] ] the function @xmath89 , @xmath90 , is called the _ autocorrelation function _ of the process . obviously , we have @xmath91=r_y(t - s)-m_y^{\,2}$ ] .
the _ spectrum _ of the stationary stochastic process @xmath85 is the fourier transform of its autocorrelation function : @xmath92 this formula has to be intended in the sense of distributions .
for instance , if @xmath93\in l^1({\mathbb{r}})$ ] , we can write @xmath94 { \mathrm{d}}\tau\,.\ ] ] by the properties of the covariance , the function @xmath93 $ ] is positive definite and , by the properties of positive definite functions , this implies @xmath95 { \mathrm{d}}\tau\geq 0;\ ] ] then , also @xmath96 . by using the stationarity and some tricks on multiple integrals
, one can check that an alternative expression of the spectrum is @xmath97.\ ] ]
the advantage now is that positivity appears explicitly and only positive times are involved .
expression can be generalized also to processes which are stationary only in some asymptotic sense and to singular processes as our @xmath77 .
let us consider our output @xmath76 under the physical probability @xmath79 .
we call `` spectrum up to time @xmath98 '' of @xmath77 the quantity @xmath99.\ ] ] when the limit @xmath100 exists , we can speak of _ spectrum of the output _ , but this existence depends on the specific properties of the concrete model . by writing the second moment
defining the spectrum as the square of the mean plus the variance , the spectrum splits in an elastic or coherent part and in an inelastic or incoherent one : @xmath101 @xmath102 \right\vert}^2,\ ] ] @xmath103 \\ { } + \frac 1 t{\operatorname{var}}_\rho^{\vartheta,\nu}\left[\int_0^t \sin\mu t\,{\mathrm{d}}x(t ) \right ] , \end{gathered}\ ] ] let us note that @xmath104 by using the expressions for the first two moments we get the spectrum in a form which involves only system operators : [ spectrums ] @xmath105 @xmath106 \right\},\end{gathered}\ ] ] @xmath107 equations give the spectrum in terms of the reduced description of system @xmath1 ( the fields are traced out ) ; this is useful for concrete computations .
but the general properties of the spectrum are more easily obtained by working with the fields ; so , here we trace out first system @xmath1 .
let us define the reduced field state @xmath108 and the field operators @xmath109 let us stress that @xmath110 commutes with its adjoint and that @xmath111 . by using eqs . and and
taking first the trace over @xmath0 , we get [ eq : sp2i(t ) ] @xmath112 to elaborate the previous expressions it is useful to introduce annihilation and creation operators for bosonic modes , which are only approximately orthogonal for finite @xmath98 : @xmath113 @xmath114= [ a_t^\dagger(\omega ) , a_t^\dagger(\omega^\prime)]=0 , \\ [ a_t(\omega ) , a_t^\dagger(\omega^\prime)]= \begin{cases } 1 & \text{for } \omega^\prime=\omega , \\ \frac{{\mathrm{e}}^{{\mathrm{i}}(\omega -\omega^\prime)t}-1 } { { \mathrm{i}}(\omega -\omega^\prime)t } & \text{for } \omega^\prime\neq \omega . \end{cases}\end{gathered}\ ] ] then , we have easily @xmath115 [ thetadependence ]
@xmath116 @xmath117 independently of the system state @xmath118 , of the field state @xmath33 and of the hudson - parthasarathy evolution @xmath119 , for every @xmath71 and @xmath56 we have the two bounds @xmath120 the first bound comes easily from @xmath121 and eqs . .
to prove the second bound , let us introduce the operator @xmath122 which satisfy the ccr @xmath123= 1 , \qquad [ b_t(\omega ) , b_t(\omega)]= [ b_t^\dagger(\omega ) , b_t^\dagger(\omega)]=0.\ ] ] then , we can write @xmath124 the usual tricks to derive the heisenberg - scrdinger - robertson uncertainty relations can be generalized also to non - selfadjoint operators.@xcite for any choice of the state @xmath125 and of the operators @xmath126 , @xmath127 ( with finite second moments with respect to @xmath125 ) the @xmath128 matrix with elements @xmath129 is positive definite and , in particular , its determinant is not negative
. then , we have @xmath130 by taking @xmath131 , @xmath132 , + @xmath133 , we get @xmath134 but we can change @xmath135 in @xmath136 and we have also @xmath137 the two inequalities together give @xmath138 which is what we wanted . ref . introduces a class of operators for the electromagnetic field , called _ two - mode quadrature - phase amplitudes _
, which have the structure of our operators @xmath110 .
anyway only two modes are involved , as if we fixed @xmath139 and @xmath56 .
let us denote here those operators by @xmath140 .
the paper explicitly constructs a class of quasi - free ( or gaussian ) field states @xmath141 for which @xmath142 .
such states are called _ two - mode squeezed states_. more generally , one speaks of _ squeezed light _ if , at least in a region of the @xmath135 line , for some @xmath71 one has @xmath143 .
if this happens , the heisenberg - type relation says that necessarily @xmath144 in such a way that the product is bigger than one .
let us take as system @xmath1 a two - level atom , which means @xmath145 , @xmath146 ; @xmath147 is the _ resonance frequency _ of the atom .
we denote by @xmath148 and @xmath149 the lowering and rising operators and by @xmath150 , @xmath151 , @xmath152 the pauli matrices ; we set also @xmath153 .
we stimulate the atom with a coherent monochromatic laser and consider homodyne detection of the fluorescence light .
the quantum fields @xmath2 model the whole environment .
the electromagnetic field is split in two fields , according to the direction of propagation : one field for the photons in the forward direction ( @xmath154 ) , that of the stimulating laser and of the lost light , one field for the photons collected to the detector ( @xmath155 ) .
assume that the interaction with the atom is dominated by absorption / emission and that the direct scattering is negligible : @xmath156 the coefficient @xmath157 is the natural _ line - width _ of the atom , @xmath158 is the fraction of fluorescence light which reaches the detector and @xmath159 is the fraction of lost light ( @xmath160).@xcite we introduce also the interaction with a thermal bath , @xmath161 and a term responsible of _ dephasing _ ( or decoherence ) , @xmath162 to represent a coherent monochromatic laser of frequency @xmath163 , we take @xmath164}(t)$ ] ; @xmath98 is a time larger than any other time in the theory and the limit @xmath100 is taken in all the physical quantities . the quantity @xmath165 is called _ rabi frequency _ and @xmath166 is called _ detuning_. the squeezing in the fluorescence light is revealed by homodyne detection , which needs to maintain phase coherence between the laser stimulating the atom and the laser in the detection apparatus which determines the observables @xmath46 ; this in particular means that necessarily we must take @xmath167 .
the limit @xmath100 can be taken in eqs . and
it is independent of the atomic initial state.@xcite the result is @xmath168 @xmath169 where @xmath170\,\rho_\textrm{eq}\big)\,\vec\sigma\big],\qquad \vec{s}=\begin{pmatrix}\cos\vartheta\\ \sin\vartheta\\0\end{pmatrix } , \\
\rho_\textrm{eq}=\frac{1}{2}\left(1+\vec{x}_\textrm{eq}\cdot\vec\sigma\right ) , \qquad \vec{x}_\textrm{eq } = -\gamma a^{-1}\ , \begin{pmatrix}0\\0\\1\end{pmatrix } , \\
a=\begin{pmatrix}\gamma\left(\frac{1}{2}+\overline{n}+2k_{\mathrm{d}}\right ) & \delta \omega&0\\ -\delta \omega & \gamma\left(\frac{1}{2}+\overline{n}+2k_{\mathrm{d}}\right)&\omega\\ 0&-\omega&\gamma\left(1 + 2\overline{n } \right)\end{pmatrix}.\end{gathered}\ ] ] examples of inelastic spectra are plotted for @xmath171 , @xmath172 , @xmath173 , and two different values of @xmath174 .
the rabi frequency @xmath64 and @xmath175 are chosen in both cases to get good visible minima of @xmath176 below 1 .
thus in this case the analysis of the homodyne spectrum reveals the squeezing of the detected light .
also complementary spectra are shown to verify theorem 2.1 .
one could also compare the homodyne spectrum with and without @xmath177 and @xmath178 , thus verifying that the squeezing is very sensitive to any small perturbation .
a. barchielli , g. lupieri , _ quantum stochastic calculus , operation valued stochastic processes and continual measurements in quantum mechanics _ , j. math .
* 26 * ( 1985 ) 22222230 .
a. barchielli , _ measurement theory and stochastic differential equations in quantum mechanics _ , phys .
a * 34 * ( 1986 ) 16421649 .
v. p. belavkin , _ nondemolition measurements , nonlinear filtering and dynamic programming of quantum stochastic processes_. in a. blaquire ( ed . ) , _ modelling and control of systems _ , lecture notes in control and information sciences * 121 * ( springer , berlin , 1988 ) pp .
245265 .
a. barchielli , _ direct and heterodyne detection and other applications of quantum stochastic calculus to quantum optics _ , quantum opt . * 2 * ( 1990 ) 423441 .
h. j. carmichael , _ an open system approach to quantum optics _
notes phys . * m 18 * ( springer , berlin , 1993 ) . h. m. wiseman , g. j. milburn , _ interpretation of quantum jump and diffusion processes illustrated on the bloch sphere _
a * 47 * ( 1993 ) 16521666 .
h. m. wiseman , g. j. milburn , _ quantum theory of optical feedback via homodyne detection _
70 * ( 1993 ) 548551 .
a. barchielli , _ continual measurements in quantum mechanics and quantum stochastic calculus_. in s. attal , a. joye , c .- a .
pillet ( eds . ) , _ open quantum systems iii _ , lecture notes in mathematics * 1882 * ( springer , berlin , 2006 ) , pp . 207291 . | when a quantum system is monitored in continuous time , the result of the measurement is a stochastic process .
when the output process is stationary , at least in the long run , the spectrum of the process can be introduced and its properties studied .
a typical continual measurement for quantum optical systems is the so called homodyne detection . in this paper
we show how the heisenberg uncertainty relations give rise to characteristic bounds on the possible homodyne spectra and we discuss how this is related to the typical quantum phenomenon of squeezing . |
the distribution of wealth , income or company size is one of important issues not only in economics but also in econophysics @xcite . in these distributions ,
a cumulative number @xmath0 obeys a power - law for @xmath1 which is larger than a certain threshold @xmath2 : @xmath3 this power - law and the exponent @xmath4 are called pareto s law and pareto index , respectively @xcite . here
@xmath1 is wealth , income , profits , assets , sales , the number of employees and etc .
the study of power - law distributions in the high range is quite significant . because a large part of the total wealth , income or profits is occupied by persons or companies in the high region ,
although the number of them is a few percent .
they have the possibility to influence economics .
power - law distributions are observed in fractal systems which have self - similarity .
this means that there is self - similarity in economic systems .
the power - law distribution in the high region is well investigated by using various models in econophysics .
recently , fujiwara et al .
@xcite find that pareto s law ( and the reflection law ) can be derived kinematically form the law of detailed balance and gibrat s law @xcite which are observed in high region @xmath5 . in the proof , they assume no model and only use these two underlying laws in empirical data .
the detailed balance is time - reversal symmetry : @xmath6 here @xmath7 and @xmath8 are two successive incomes , profits , assets , sales , etc , and @xmath9 is a joint probability distribution function ( pdf ) .
gibrat s law states that the conditional probability distribution of growth rate @xmath10 is independent of the initial value @xmath7 : @xmath11 here growth rate @xmath12 is defined as the ratio @xmath13 and @xmath10 is defined by using the pdf @xmath14 and the joint pdf @xmath15 as @xmath16 in ref .
@xcite , by using profits data of japanese companies in 2002 ( @xmath7 ) and 2003 ( @xmath8 ) , it is confirmed that the law of detailed balance ( [ detailed balance ] ) holds in all regions @xmath17 and @xmath18 ( fig .
[ profit2002vsprofit2003 ] ) .
we also find that gibrat s law still holds in the extended region @xmath19 and @xmath18 . from these observations
, we have shown that pareto index is also induced from the growth rate distribution empirically . on the other hand , it is also well known that the power - law is not observed below the threshold @xmath2 @xcite .
for instance , we show the profits distributions of japanese companies in 2002 and 2003 ( fig .
[ profitdistribution ] ) .
we find that pareto s law holds in the high profits region and it fails in the middle one .
the study of distributions in the middle region is as important as the study of power - law those . because a large number of persons or companies is included in the middle region .
furthermore , it is interesting to study the breaking of fractal . in order to obtain the distribution in the middle region ,
we examine data below the threshold @xmath2 .
it is reported that gibrat s law is valid only in the high region ( @xcite for instance ) .
the recent works about the breakdown of gibrat s law is done by stanley s group @xcite .
aoyama et al .
also report that gibrat s law does not hold in the middle region by using data of japanese companies @xcite . in the analysis of gibrat s law in ref .
@xcite , we concentrate our attention to the region @xmath19 and @xmath18 . in this paper , we examine the growth rate distribution in all regions @xmath17 and @xmath18 and identify the law which is the extension of gibrat s law . and @xmath18 non - gibrat s law . ]
we employ profits data of japanese companies in 2002 and 2003 which are available on the database `` cd eyes '' published by tokyo shoko research , ltd . @xcite . by using the extended gibrat s law
, we derive the distribution function in the high and middle profits region under the law of detailed balance .
it explains empirical data with high accuracy .
notice that the distribution function has no fitting parameter .
the parameters of the function are already decided in the extended gibrat s law .
in this section , we reconfirm gibrat s law in the high profits region and identify non - gibrat s law in the middle one .
we divide the range of @xmath7 into logarithmically equal bins as @xmath20 $ ] thousand yen with @xmath21 .. ] in fig .
[ profitgrowthratel ] , [ profitgrowthratem ] and [ profitgrowthrateh ] , the probability densities for @xmath22 are expressed in the case of @xmath23 , @xmath24 and @xmath25 , respectively . the number of the companies in fig
. [ profitgrowthratel ] , [ profitgrowthratem ] and [ profitgrowthrateh ] is @xmath26 " , @xmath27 " and @xmath28 " , respectively .
here we use the log profits growth rate @xmath29 . the probability density for @xmath22 defined by @xmath30
is related to that for @xmath12 by @xmath31 from fig .
[ profitgrowthratel ] , [ profitgrowthratem ] and [ profitgrowthrateh ] , we express the relation between @xmath22 and @xmath30 as follows : @xmath32 by the use of eq .
( [ qandq ] ) , these relations are rewritten in terms of @xmath12 as @xmath33 with @xmath34 . by applying the expressions ( [ approximation1 ] ) and ( [ approximation2 ] ) to data in fig .
[ profitgrowthratel ] , [ profitgrowthratem ] and [ profitgrowthrateh ] , the relation between @xmath7 and @xmath35 is obtained ( fig .
[ x1vst ] ) . in fig .
[ x1vst ] , @xmath36 hardly responds to @xmath7 for @xmath37 .
this means that gibrat s law holds in the high profits region .
on the other hand , @xmath38 ( @xmath39 ) increases ( decreases ) linearly with @xmath40 for @xmath41 . in the middle profits region , not gibrat s law but the other law
( non - gibrat s law ) holds as follows : , we examine another type of non - gibrat s law . ]
@xmath42 in fig .
[ x1vst ] , @xmath43 and @xmath44 are estimated as @xmath45 where @xmath46 thousand yen ( @xmath47 million yen ) and @xmath48 thousand yen . in this paper
we call the combination of gibrat s law ( ( [ t+ ] ) , ( [ t- ] ) and ( [ alphah ] ) ) and non - gibrat s law ( ( [ t+ ] ) , ( [ t- ] ) and ( [ alpham ] ) ) extended gibrat s law .
in refs . @xcite , pareto s law ( [ pareto ] ) and the pareto index @xmath4 can be derived from the detailed balance ( [ detailed balance ] ) and gibrat s law ( [ gibrat ] ) in the high profits region . in this section ,
we derive profits distribution not only in the high profits region but also in the middle one by using the detailed balance and the extended gibrat s law . due to the relation of @xmath49 under the change of variables from @xmath50 to @xmath51 ,
these two joint pdfs are related to each other , @xmath52 by the use of this relation , the detailed balance ( [ detailed balance ] ) is rewritten in terms of @xmath15 as follows : @xmath53 substituting the joint pdf @xmath15 for the conditional probability @xmath10 defined in eq . ( [ conditional ] ) ,
the detailed balance is expressed as @xmath54 in the preceding section , the conditional probability @xmath10 is identified as ( [ rline1 ] ) or ( [ rline2 ] ) . under the change of variables @xmath55 ,
the conditional probability @xmath56 is expressed as @xmath57 from the detailed balance and the extended gibrat s law , one finds the following : @xmath58 for @xmath59 . here
we assume that the @xmath1 dependence of @xmath60 is negligible in the derivation , the validity of which should be checked against the results . by expanding eq .
( [ de0 ] ) around @xmath61 , the following differential equation is obtained @xmath62 p(x ) + x~ p'(x ) = 0,\end{aligned}\ ] ] where @xmath1 denotes @xmath7 .
the same differential equation is obtained for @xmath63 .
the solution is given by @xmath64 with @xmath65 .
here we use the relation @xmath66 which is confirmed in ref .
in the previous section , we derive the profits distribution function ( [ handm ] ) in the high and middle profits region from the detailed balance and the extended gibrat s law . in this section
, we directly examine whether it fits with profits distribution data . in order to average the scattering of data points , we employ the cumulative number of companies ( fig .
[ profitdistribution ] ) . for @xmath19 , by using eqs .
( [ alphah ] ) and ( [ handm ] ) the cumulative distribution in the high profits region is expressed as @xmath67 on the other hand , by the use of eqs .
( [ alpham ] ) and ( [ handm ] ) the cumulative distribution in the middle profits region @xmath68 is given by @xmath69 here @xmath70 is error function defined by @xmath71 . in fig .
[ profitdistributionfit ] , the distribution functions ( [ cdh ] ) in the high profits region ( @xmath19 ) and ( [ cdm ] ) in the middle one ( @xmath68 ) explain empirical data in 2003 with high accuracy .
this guarantees the validity of the assumption in the previous section .
notice that there is no ambiguity in parameter fitting , because indices @xmath4 , @xmath72 and the bounds @xmath2 , @xmath73 is already given in the extended gibrat s law .
is especially called zipf s law @xcite . ]
in section [ sec - gibrat s law and non - gibrat s law ] , we present the non - gibrat s law ( [ t+ ] ) , ( [ t- ] ) and ( [ alpham ] ) as a linear approximation of @xmath35 in fig .
[ x1vst ] . in this section ,
we examine another linear approximation in fig .
[ x1vst2 ] , the vertical axis of which is the logarithm . in fig .
[ x1vst2 ] , @xmath74 ( @xmath75 ) increases ( decreases ) linearly with @xmath40 for @xmath41 .
this relation is expressed as @xmath76 where @xmath77 here @xmath2 and @xmath78 are same values in section [ sec - gibrat s law and non - gibrat s law ] . from the detailed balance and this extended gibrat s law
, one finds @xmath79 for @xmath80 . by expanding this equation around @xmath61 , the following differential equation
is obtained @xmath81 p(x ) + x~ p'(x ) = 0~ , \label{de } \end{aligned}\ ] ] where @xmath1 denotes @xmath7 .
the same differential equation is obtained for @xmath63 . in order to take @xmath82 limit into account ,
we rewrite this as follows : @xmath83 p(x ) + x~ p'(x ) = 0~.\end{aligned}\ ] ] the solution is given by @xmath84 with @xmath85 . here
we use the expansion @xmath86 . if we neglect @xmath87 terms in eq .
( [ handmanother ] ) , the profits distribution function ( [ handmanother ] ) can be identified with ( [ handm ] ) , because @xmath88 numerically . in other words
, there is no essential difference between two expressions ( [ t+ ] ) , ( [ t- ] ) and ( [ t+another ] ) , ( [ t - another ] ) in the middle region .
in this paper , we have kinematically derived the profits distribution function in the high and middle region from the law of detailed balance and the extended gibrat s law by employing profits data of japanese companies in 2002 and 2003 .
firstly , we have reconfirmed gibrat s law in the high profits region ( @xmath5 ) .
the value of @xmath2 is estimated to be about @xmath89 million yen in fig .
[ x1vst ] or [ x1vst2 ] . at the same time
, we have found that gibrat s law fails and another law holds in the middle profits region ( @xmath90 ) .
we have identified the non - gibrat s law and the value of @xmath73 is estimated to be about @xmath91 thousand yen in fig .
[ x1vst ] or [ x1vst2 ] .
we have called the combination of gibrat s law and the non - gibrat s law the extended gibrat s law .
secondly , we have derived not only pareto s law in the high profits region but also the distribution in the middle one from the detailed balance and the extended gibrat s law .
the derivation has been described uniformly in terms of the extended gibrat s law .
the profits distribution in the middle region is very similar to log - normal one : @xmath92~ , \label{log - normal}\end{aligned}\ ] ] where @xmath93 is mean value and @xmath94 is variance .
the difference between two distributions ( [ handm ] ) and ( [ log - normal ] ) is only the power of @xmath1 .
it brings a translation along the horizontal axis in the log - log plot . in this sense ,
the distribution in the middle profits region obtained in this paper is essentially equivalent to the log - normal one .
notice that it has no fitting parameter that the log - normal distribution , prepared only for data fitting , has .
indies @xmath4 , @xmath72 and the bounds @xmath2 , @xmath73 is already given in the extended gibrat s law .
in the derivation of the profits distribution , we have used the detailed balance and the extended gibrat s law .
the detailed balance is observed in a relatively stable period in economy @xcite .
the extended gibrat s law is interpreted as follows .
companies , classified in small - scale profits category , have more ( less ) possibilities of increasing ( decreasing ) their profits than companies classified in large - scale one .
in other words , it is probably difficult that companies gain large - scale profits in two successive years
. this is the non - gibrat s law and it leads the distribution in the middle profits region .
this phenomenon is not observed above the threshold @xmath2 . for @xmath5 ,
companies , classified in small - scale profits category , have same possibilities of increasing ( decreasing ) their profits with companies classified in large - scale one .
this is the gibrat s law and it leads the power distribution in the high profits region . in this paper
, we can not mention the distribution in the low profits region @xmath95 , because no law is observed in the region ( fig .
[ x1vst ] or [ x1vst2 ] ) .
this may be caused by insufficient data in the low region .
lastly , we speculate possible distributions in the region . if we do not take @xmath82 limit , we should solve the differential equation ( [ de ] ) .
the solution is given by @xmath96~. \label{profitdistributionl}\end{aligned}\ ] ] for the case @xmath97 and @xmath98 , the distribution ( [ profitdistributionl ] ) takes the exponential form in refs .
@xcite . on the other hand , for the case
@xmath99 and @xmath100 , the distribution ( [ profitdistributionl ] ) takes weibull form in ref . @xcite
we can decide the distribution in the low profits region if sufficient data in the region are provided .
we have showed that a distribution function is decided by underlying kinematics . for profits data we used ,
the distribution is power in the high region and log - normal type in the middle one .
this does not claim that all the distributions in the middle region are log - normal types .
for instance , the personal income distribution may take a different form , if the extended gibrat s law changes . even in the case
, the other extended gibrat s law will decide the other distribution by the use of the method in this paper .
the author is grateful to professor h. aoyama for useful discussions about his lecture . 99 r.n .
mategna and h.e .
stanley , an introduction to econophysics , cambridge university press , uk , 2000 . v. pareto , cours deconomique politique , macmillan , london , 1897
. h. aoyama , w. souma , y. nagahara , h.p .
okazaki , h. takayasu and m. takayasu , cond - mat/0006038 , fractals 8 ( 2000 ) 293 ; + w. souma , cond - mat/0011373 , fractals 9 ( 2001 ) 463 .
a. dr@xmath101gulescu and v.m .
yakovenko , cond - mat/0103544 , physica a299 ( 2001 ) 213 .
y. fujiwara , w. souma , h. aoyama , t. kaizoji and m. aoki , cond - mat/0208398 , physica a321 ( 2003 ) 598 ; + h. aoyama , w. souma and y. fujiwara , physica a324 ( 2003 ) 352 ; + y. fujiwara , c.d .
guilmi , h. aoyama , m. gallegati and w. souma , cond - mat/0310061 , physica a335 ( 2004 ) 197 ; + y. fujiwara , h. aoyama , c.d .
guilmi , w. souma and m. gallegati , physica a344 ( 2004 ) 112 ; + h. aoyama , y. fujiwara and w. souma , physica a344 ( 2004 ) 117 .
r. gibrat , les inegalites economiques , paris , sirey , 1932 .
a. ishikawa , pareto index induced from the scale of companies , physics/0506066 . w.w .
badger , in : b.j .
west ( ed . ) , mathematical models as a tool for the social science , gordon and breach , new york , 1980 , p. 87 ; + e.w .
montrll and m.f .
shlesinger , j. stat .
32 ( 1983 ) 209 .
k. okuyama , m. takayasu and h. takayasu , physica a269 ( 1999 ) 125 .
stanley , l.a.n .
amaral , s.v .
buldyrev , s. havlin , h. leschhorn , p. maass , m.a . salinger and h.e .
stanley , nature 379 ( 1996 ) 804 ; + l.a.n .
amaral , s.v .
buldyrev , s. havlin , h. leschhorn , p. maass , m.a .
salinger , h.e .
stanley and m.h.r .
stanley , j. phys .
( france ) i7 ( 1997 ) 621 ; + s.v .
buldyrev , l.a.n .
amaral , s. havlin , h. leschhorn , p. maass , m.a .
salinger , h.e .
stanley and m.h.r .
stanley , j. phys .
( france ) i7 ( 1997 ) 635 ; + l.a.n .
amaral , s.v .
buldyrev , s. havlin , m.a . salinger and h.e .
stanley , phys .
80 ( 1998 ) 1385 ; + y. lee , l.a.n .
amaral , d. canning , m. meyer and h.e .
stanley , phys .
81 ( 1998 ) 3275 ; + d. canning , l.a.n .
amaral , y. lee , m. meyer and h.e .
stanley , economics lett .
60 ( 1998 ) 335 .
9th annual workshop on economic heterogeneous interacting agents ( wehia 2004 ) ; + the physical society of japan 2004 autumn meeting .
tokyo shoko research , ltd .
, http://www.tsr - net.co.jp/. g.k .
gipf , human behavior and the principle of least effort , addison - wesley , cambridge , 1949 .
m. nirei and w. souma , sfi/0410029 .
m. anazawa , a. ishikawa , t. suzuki and m. tomoyose , cond - mat/0307116 , physica a335 ( 2004 ) 616 ; + a. ishikawa and t. suzuki , cond - mat/0403070 , physica a343 ( 2004 ) 376 ; + a. ishikawa , cond - mat/0409145 , physica a349 ( 2005 ) 597 . | employing profits data of japanese companies in 2002 and 2003 , we identify the non - gibrat s law which holds in the middle profits region . from the law of detailed balance in all regions , gibrat s law in the high region and the non - gibrat s law in the middle region , we kinematically derive the profits distribution function in the high and middle range uniformly .
the distribution function accurately fits with empirical data without any fitting parameter .
pacs code : 04.60.nc + keywords : econophysics ; pareto law ; gibrat law ; detailed balance |
the gravitational deflection of photons can be used as a probe of the matter distribution along the line of sight to the sources .
the latter may be at the last scattering surface ( @xmath1 ) , in the case of the cosmic microwave background ( seljak 1996 ; martnez - gonzlez et al .
1997 ; bernardeau 1997 ) , or high@xmath2 objects as qsos or galaxies ( blanford et al . 1991 ; kaiser 1992 ; kaiser & squires 1993 ; bartelmann 1995 ; villumsen 1995b ; villumsen 1996 ; bernardeau , van waerbeke & mellier 1996 ; kaiser 1996 ; jain & seljak 1996 ) .
information about the matter fluctuations can be obtained on different scales ranging from galaxy haloes to the large - scale structure of the universe .
many of the theoretical studies on this subject have dealt with the polarization or ellipticity produced on background galaxies by the large - scale structure of the universe , and there are currently several ongoing observational projects trying to detect and quantify this effect .
nevertheless , measuring shear amplitudes as the ones predicted by the above mentioned calculations is very difficult from a technical point of view ( although see mould et al .
1991 ; villumsen 1995a ) , and it is not totally clear if such a precision would be routinely achievable in the near future ( ringberg workshop 1997 ) . however , there is another observable phenomenon produced by gravitational lensing of background sources by foreground mass distributions which may have been already detected : qso - galaxy associations due to the magnification bias effect @xcite .
the surface density of a sample of flux - limited background sources behind a lens which magnifies them by a factor @xmath3 is changed in the form @xmath4 , where @xmath5 is the unperturbed background source density .
if @xmath6 ( or @xmath7 ) , the change in the density can be characterized by the factor @xmath8 .
thus , depending on the slope @xmath9 there may be an excess of background sources ( @xmath10 ) , a depletion ( @xmath11 ) , or the density may remain unchanged ( @xmath12 ) .
if we properly choose the background source population , so that it has a slope @xmath9 considerably different from 1 , there would be a correlation ( or anticorrelation ) between the position of the matter overdensities acting as lenses and the background sources .
now , these matter perturbations will be traced , up to a bias factor , by galaxies and thus , there will be a correlation between these foreground galaxies ( or any other tracers of dark matter ) and the background sources .
there are several reported associations between foreground galaxies and high redshift , background agns ( see schneider , ehlers & falco ( 1992 ) ; narayan & bartelmann ( 1996 ) or wu ( 1996 ) for reviews ) , but only a few of these studies extend to relatively large scales .
bartelmann & schneider ( 1994 ) found a strong association between galaxies selected from the iras faint source catalogue and high@xmath2 agn from the 1jy catalogue . in bentez & martnez- gonzlez ( 1995 )
it was found that red apm galaxies tracing large scale structures were correlated with 1jy qsos .
another sample of radio loud qsos , extracted from the pks catalogue has also been found to be correlated with cosmos galaxies @xcite , with a correlation scale of several arcmin .
other studies considering the correlation between galaxy clusters and high - z qsos ( seitz & schneider 1995 , wu & han 1996 ) have also found positive results . in this paper , we shall study the effects of weak gravitational lensing by foreground matter fluctuations on a population of background sources at high@xmath2 .
we consider different values of @xmath0 and model the fluctuations assuming cdm with a power spectrum whose evolution in time follows a standard ansatz ( hamilton et al .
1991 , peacock & dodds 1996 , linear and non - linear contributions are considered ) .
we assume that these matter perturbations are traced , up to a global bias parameter @xmath13 by galaxies .
more specifically , we shall explore the behavior of @xmath14 , i.e. the large - scale correlation between the ellipticity of background galaxies and the position of foreground ones , which apparently has not been considered in the literature .
we shall also consider in detail other correlations ( in particular their dependence on @xmath0 ) such as @xmath15 i.e. magnification - foreground galaxies and magnification - magnification .
@xmath16 can be indirectly estimated through the galaxy - galaxy correlation function ( villumsen 1995b ) . however , measuring @xmath17 offers several advantages over @xmath16 from the observational point of view . in the first place
, @xmath17 has an amplitude several times higher than @xmath16 . besides
, if the foreground and background galaxy populations are properly selected so that there is no redshift overlap between them ( e.g high@xmath2 qsos and bright galaxies ) , one does not have to bother about intrinsic correlations : any measured effect should be caused by gravitational lensing .
section 2 develops the formalism dealing with weak gravitational lensing for a flat and open cosmological model , the concepts of magnification and polarization ( or ellipticity ) and the different correlations . in section 3
appear the main theoretical results as well as comments on different observational perspectives .
finally , in section 4 we give the conclusions of the paper .
we will consider the propagation of photons from a source at redshift @xmath18 to the observer ( @xmath19 ) , the universe being a perturbed friedmann model with vanishing pressure . for scalar perturbations ,
the metric in the conformal newtonian gauge is given in terms of the scale factor @xmath20 and a single potential @xmath21 , that satisfies the poisson equation , as follows ( martnez - gonzlez et al .
1997 ) @xmath22,\ ] ] @xmath23 we take units such that @xmath24 and @xmath25 denote the flat , open and closed friedmann background universe . assuming a perturbation scheme ( `` weak lensing '' ) , the null geodesic equation for the previous metric can be integrated in the form @xmath26 , where @xmath27 is the direction of observation and @xmath28 is the distance to the photon in the background metric , i.e. the perturbation @xmath30 can be decomposed in a term parallel to the direction of observation @xmath27 and a term , @xmath31 , ortoghonal to such a direction .
the last term is given by @xmath32 where @xmath33 is a window function @xmath34 for photons that are propagated from a source at redshift @xmath18 ( distance @xmath35 ) to the observer(@xmath36 or @xmath37 ) , the lensing vector @xmath38 is defined in the usual way : @xmath39 thus we find @xmath40 once we have obtained the expression for the trajectory of the photon in the conformal newtonian gauge , it is easy to calculate everything in the conformal synchronous - comoving gauge ( martnez - gonzlez et al .
the lensing vector in such a gauge ( that is the appropriate one from the point of view of observations ) is given by the expression ( 5 ) plus some additional terms that can be interpreted as doppler contributions at the source and observer and an acceleration term at the observer .
the last two terms can be estimated from the doppler velocity respect to the cosmic microwave background and from our local infall towards the virgo cluster ( or great attractor ) .
these extra contributions are very small , so the lensing vector @xmath38 in the synchronous - comoving gauge is approximately given by @xmath38 , as defined by equations ( 3 , 5 ) .
let us assume a population of background sources ( e.g. quasars or galaxies ) , placed at different distances @xmath35 with a distribution @xmath41 ( @xmath42 ) .
then , we can define the integrated lensing vector @xmath43 and taking into account equations ( 3 - 5 ) we obtain @xmath44 @xmath45 @xmath46 if we take the derivative along the plane orthogonal to the direction of observation @xmath27 , we get the convergence tensor @xmath47 which can be decomposed in the form @xmath48 for weak lensing , the convergence scalar @xmath49 is related to the magnification by @xmath50 .
moreover , the polarization components in the plane orthogonal to @xmath27 can be defined in the standard way @xmath51 and the complex polarization is @xmath52 .
so , we can define the scalar polarization @xmath53 .
let us consider a second population of foreground sources ( e.g. galaxies ) placed at different distances @xmath28 with a distribution @xmath54 ( @xmath55 ) .
if @xmath56 represents the density fluctuation ( that satisfies the poisson equation @xmath57 ) , we shall assume that there is a constant bias factor @xmath13 relating the number fluctuation , @xmath58 , and such an overdensity .
then , we can define the integrated overdensity @xmath59 .
taking into account equations ( 3 - 7 ) , we obtain @xmath60 @xmath61 where @xmath62 . using the limber approximation ( see appendix ) , i. e. only a small region @xmath63 is contributing with @xmath64 , the previous equation can be approximated by @xmath65\times\nonumber\\ & \times&{\int}_{\theta s}^{\infty}dr \frac{c_{\phi \delta}(\lambda ; r ) } { { ( r^2 - { \theta}^2 s^2)}^{1/2}},\end{aligned}\ ] ] where now appears the correlation at a single time and @xmath66 is given by @xmath67^{-1}$ ] . introducing the power spectrum @xmath68 of the matter density fluctuations defined by @xmath69 and the relation @xmath70 , obtained via the poisson equation for small scales , the last expression ( 11 ) is @xmath71 finally , taking another @xmath72 derivative on the equation ( 9 ) , we get the following correlations @xmath73 @xmath74 @xmath75 and @xmath76 is the bessel function of 1st kind .
these are the basic formulas to be applied when two different populations are correlated . in particular , the background - foreground correlation function is given by bartelmann ( 1995 ) . @xmath77 where @xmath78 is the slope of the background source number counts .
if the background population is concentrated at a certain redshift @xmath79 corresponding to a distance @xmath80 , having a dirac delta distribution @xmath81 , then the window @xmath82 is given by @xmath83 @xmath84 if the foreground population has also a dirac delta distribution form @xmath85 , then the window @xmath82 in equation ( 16 ) is @xmath86 and @xmath87 } { \pi\lambda_b(1-\lambda_f)^2[1-(1-\omega)\lambda_f^2]}\times\nonumber\\ & \times&\int_0^\infty dk k p(\lambda_f , k)j_i(k\theta s_f)\end{aligned}\ ] ] where @xmath88^{-1}$ ] .
let us consider a population of background sources ( e.g. galaxies ) placed at different distances @xmath28 with a distribution @xmath41 ( @xmath89 ) .
we are now interested in the gravitational lensing properties induced by the population itself .
the magnification - magnification @xmath90 , polarization - polarization @xmath91 and magnification - overdensity @xmath92 correlations are defined by @xmath93 @xmath94 @xmath95 after an straightforward calculation , one obtains @xmath96}^3}{{(1 - \lambda ) } ^4}\times\nonumber\\ & \times&\int_0 ^ 1dkk^{-3}p(\lambda , k)j_0(ks\theta ) , \end{aligned}\ ] ] where the limber approximation has been assumed and @xmath82 is given by equation ( 7 ) .
now , taking into account the definitions for the correlations , one gets ( see also kaiser 1992 , 1996 ; jain & seljak 1996 ) @xmath97}\times\nonumber\\ & \times&\int_0^\infty dkkp(\lambda , k)j_0(ks\theta ) , \end{aligned}\ ] ] @xmath98 the last expression in agreement with villumsen ( 1995b ) . for a population with a dirac delta distribution @xmath99 , the window @xmath100 is given by equation ( 18 ) .
moreover , @xmath101 is given by equation ( 14 ) but now @xmath102 must be substituted by @xmath103 .
trivially , for a strongly peaked @xmath18-distribution such a correlation is very small ( see the dependence on @xmath104 in equation ( 19 ) ) .
it is interesting to note that the observed correlation function @xmath105 associated to a population is given by the following contributions ( villumsen 1995b ) @xmath106 @xmath107 where @xmath108 is the matter correlation function .
so , gravitational lensing modifies the intrinsic correlation approximately by the magnification term .
with the formalism presented in the previous section we have calculated the correlations @xmath17 , @xmath14 and @xmath16 .
we assume a cdm model with a primordial harrison - zeldovich spectrum , a hubble parameter h@xmath109 ( h@xmath110h km s@xmath111 mpc@xmath111 ) and flat as well as open universe models . for the power spectrum
we have used the fit given by equation ( g3 ) of bardeen et al .
( 1986 ) which is normalized to the cluster abundance : @xmath112 , following viana and liddle ( 1996 ) ( see also white , efstathiou and frenk 1993 ; eke , cole and frenk 1996 ; see kaiser 1997 for a discussion on alternative normalizations ) .
for the nonlinear evolution of the power spectrum we use the recently improved fitting formula given by peacock and dodds ( 1996 ) .
that formula is based on the hamilton et al .
( 1991 ) scaling procedure to describe the transition between linear and nonlinear regimes .
it accounts for the correction introduced by jain , mo and white ( 1995 ) for spectra with @xmath113 and applies to flat as well as open universes . for the redshift distributions of the background and foreground sources we consider a dirac delta distribution peaked at @xmath114 and @xmath115 respectively , @xmath116 and @xmath117 .
these simple distributions are very useful since they reduce the calculations and the results differ only slightly when compared to other more realistic distributions ( see below ) . @xmath17 and @xmath14 are computed using equations ( 14,15,19 ) .
@xmath118 is obtained from equation ( 25 ) where the function @xmath82 is given by equation ( 18 ) .
the crosscorrelation @xmath119 for a population of background sources peaked at @xmath120 and another of foreground lenses at @xmath121 is given in figure ( 1 ) .
the effect is maximum at zero lag and rapidly decreases for scales above a few arcmin .
the amplitude is always above a few percent for scales @xmath122 and at these scales the linear contribution is negligible compared to the nonlinear one for all the @xmath0 values .
notice that @xmath17 increases when @xmath0 decreases whereas considering only the linear contribution the situation is reversed . from the observational point of view
it is useful to calculate the average crosscorrelation ( or equivalently the mean relative excess of background sources around foreground lenses ) within a given radius @xmath123 , @xmath124 .
the variation of @xmath125 with @xmath126 for different values of @xmath79 and a @xmath127 radius is shown in figures ( 2a , b , c ) .
a maximum amplitude of a few percent is obtained at a @xmath126 in the range @xmath128 for all background populations and all cosmological models .
( at this point it is important to recall that to compare with observations of background - foreground object correlations @xmath129 , the bias factor @xmath13 of the foreground population and the slope @xmath9 of the background population enter in the calculation following equation 16 ) .
it is interesting to notice that there are already available large galaxy samples , like the apm or cosmos catalogues , which peak at a redshift within that range ( for the apm catalogue @xmath130 for a magnitude limit @xmath131 , see efstathiou 1995 ) .
moreover , the use of realistic redshift distributions to represent the foreground and background populations ( with a bell - like shape similar to that of the apm one ) , changes the results in only @xmath132 for the relevant angular scales compared to the dirac delta distribution used here . those catalogues , which except for a bias factor are assumed to follow the large scale matter distribution , are therefore very suitable to crosscorrelate with a background source population . this has already been done by bentez and martnez - gonzlez ( 1995 , 1997 ) for the 1 jansky and parkes samples of radio loud qsos as background populations , finding clear evidences of positive crosscorrelations . in figure 3 , we show @xmath119 for @xmath133 and for @xmath134 and @xmath120 , mean redshift values appropriate for the galaxy and radio qso samples considered above .
the dependence of @xmath17 with @xmath0 is more relevant at small angular scales . in figure 4
, we also represent the average crosscorrelation @xmath125 as a function of @xmath0 .
a rough comparison between its expected amplitude and the measured value for the cosmos - parkes samples , as given in figure 4 of bentez and martnez - gonzlez 1997 , shows agreement for realistic values of @xmath0 and the bias parameter .
a detailed comparison of the theoretical calculations with the observational results will be given elsewhere .
dolag and bartelmann ( 1997 ) have recently presented calculations of the qso - galaxy correlation function for such qso and galaxy populations produced by gravitational lensing due to the large scale structure , following a similar theoretical scheme .
the crosscorrelation between the polarization of a background source population peaked at @xmath79 and the matter density fluctuations peaked at @xmath126 , @xmath135 , is given in figure ( 5 ) for @xmath120 and @xmath136 .
the maximum value is in the angular range @xmath137 and this scale is smaller for low @xmath0 models . as in the case of @xmath17 ,
the linear contribution is negligible ; however , it peaks at a much larger angle of @xmath138 as a consequence of the much larger scales which contribute to the linear level . including the nonlinear evolution ,
a correlation of @xmath139 can be expected at angular scales @xmath122 for realistic models of structure formation .
the use of realistic redshift distributions to represent the foreground and background populations ( with a bell - like shape typical of magnitude limited samples ) changes the results in only @xmath132 for the relevant angular scales compared to the dirac delta distribution used here . in analogy to the previous subsection
, it is useful to calculate the average crosscorrelation within a given radius @xmath123 , @xmath140 .
the variation of @xmath141 with @xmath126 for several values of @xmath79 and @xmath0 is shown in figures ( 6a , b , c ) .
maximum amplitudes of the order of @xmath142 are found for @xmath126 within a relatively wide range @xmath143 .
the amplitude grows appreciably with @xmath79 being a factor of @xmath144 difference between @xmath145 and @xmath146 .
considering the behavior of @xmath140 from these figures , suitable populations to detect the crosscorrelation would be a foreground sample with @xmath147 and a background one peaked at @xmath148 .
in this subsection we concentrate on a single population of background sources peaked at a given @xmath79 and calculate @xmath149 .
this is done in figure ( 7 ) for @xmath120 .
the maximum effect is at zero lag and its amplitude is relatively small @xmath150 .
the nonlinear contribution clearly dominates over the linear one but now the amplitude grows with @xmath0 , contrary to the crosscorrelations @xmath119 and @xmath135 .
our results are in agreement with jain & seljak ( 1996 ) , and with kaiser ( 1997 ) for one degree scales ( see also those papers for a more detailed analysis of @xmath16 ) , where the dominant contribution is the linear one .
considering also the slope of the population of sources following equation ( 26 ) , @xmath16 could be estimated from the observed autocorrelation of faint galaxies .
nevertheless , measuring @xmath91 should be more feasible from the practical point of view , as we do not have to disentangle the contribution caused by lensing from the intrinsic correlations ; it is usually assumed ( and strongly hoped ) that the intrinsic ellipticities of background galaxies are not correlated .
we have obtained the expressions for the correlations between the magnification or polarization of background sources and the foreground matter distribution as function of the nonlinear evolution of the power spectrum .
these formuli are valid for flat and open universes . for the crosscorrelation of the background magnification and foreground matter distribution , @xmath119 ,
the maximum is at zero lag and the amplitude remains above a few percent for scales @xmath122 .
@xmath119 increases significantly when @xmath0 decreases .
the linear contribution is negligible compared to the nonlinear one for the relevant scales below a few arcmin .
varying the redshift of the foreground population , @xmath126 , a maximum amplitude of a few percent for the integrated correlation @xmath125 is obtained at @xmath151 for all background populations and all cosmological models .
the crosscorrelation of the background polarization and foreground matter distribution , @xmath135 , presents a maximum of the order of @xmath142 at a non - null angle , typically in the range @xmath152 . fixing the redshifts @xmath126 and @xmath79 of the two populations , the angular scale of the maximum decreases with @xmath0 .
@xmath135 increases significantly when @xmath0 decreases .
the linear contribution is negligible compared to the nonlinear one for the relevant scales below a few arcmin .
varying the redshift of the foreground population , @xmath126 , a maximum amplitude of @xmath153 for the integrated correlation @xmath141 is obtained within a relatively wide range @xmath154 .
this amplitude grows appreciably with @xmath79 being a factor of @xmath144 difference between @xmath145 and @xmath146 . finally , the correlation of the magnifications for a single population , @xmath149 , has a relatively small maximum amplitude @xmath155 in all cases .
the amplitude grows with @xmath0 , contrary to the crosscorrelations @xmath119 , @xmath135 .
the authors thank matthias bartelmann , klaus dolag and peter schneider for interesting and useful discussions .
we also thank jos revuelta for his help with the aesthetic aspects of this paper .
jls , emg and nbl acknowledge financial support from the spanish dges , project pb95 - 0041 .
nb acknowledges a spanish m.e.c .
ph.d . scholarship .
some equations ( e.g. equation ( 11 ) ) have been obtained assuming the limber approximation , i.e. one assumes that the maximum scale of appreciable correlation is small compared to the typical distances to the foreground and background objects in the samples .
so , if we consider a double integral with the same form as equation ( 10 ) , the contribution to the integral is appreciable only when the points are nearly at the same time @xmath156 and their separation is small , @xmath157 , when @xmath158 , the spatial separation of two points is where @xmath163 and @xmath164 , the latter being a smooth function .
the previous integral for the variable @xmath165 can be approximated by @xmath166 if one assumes that the foreground sources are placed at distances far away from the observer and hubble distance ( case of practical interest ) .
finally , changing the variable @xmath165 by @xmath63 one gets equation ( 11 ) .
we remark that equation ( 11 ) is also valid when a dirac distribution is assumed to represent the foreground redshift distribution .
the limber approximation in fourier space has been also considered by kaiser ( 1992 ) in the context of weak gravitational lensing . | correlations between the magnification or polarization of background sources , induced by gravitational lensing due to the large - scale structure , and the positions of foreground galaxies are investigated .
we found that their amplitude is enhanced with respect to correlations for a single population .
we analize the dependence of the correlations with the density parameter @xmath0 considering a nonlinear evolution of the matter power spectrum .
the contribution of the linear evolution is negligible at scales below several arcminutes .
detailed results on the dependence of the correlations on the redshift of the foreground and background populations for different cosmological models are presented .
the effect reaches its maximum amplitude for foreground populations with relatively small redshifts due to the fast increase of the nonlinear matter power spectrum at recent times .
cosmology : theory - gravitational lensing - large - scale structure |
it is widely believed that @xmath0 is the correct theory of strong interactions . on the other hand nuclear forces
are studied within different potential models .
it is not clear whether or not these phenomenological approaches can be justified from fundamental theory .
effective field theory is thought as a bridge between @xmath0 and potential models .
chiral perturbation theory serves as a low - energy effective field theory inspired by qcd .
there has been much recent interest in the eft approach to the nucleon - nucleon scattering problem ( see @xcite and citations in these papers ) .
the chiral perturbation theory approach for processes involving an arbitrary number of nucleons was formulated in @xcite . unlike purely pionic processes @xcite , for the @xmath1-nucleon problem power counting is used for the `` effective potentials '' instead of full amplitudes .
the effective potential is defined as a sum of time - ordered perturbation theory diagrams for the @xmath2-matrix excluding those with purely nucleonic intermediate states . to find the full @xmath3-matrix
one should solve a lippmann - schwinger equation ( or schr " oedinger equation ) with this effective potential in place of the interaction hamiltonian , and with _ only _
@xmath1-nucleon intermediate states @xcite .
the lagrangian of effective field theory is highly non - renormalizable in the traditional sense but it contains all possible terms which are not suppressed by the symmetries of the theory and the ultraviolet divergences are absorbed into the parameters of the lagrangian .
renormalization points are chosen of the order of external momenta @xmath4 .
after renormalization , the effective cut - off is of order @xmath4 @xcite .
power counting apparently depends on the normalisation condition ( one could choose a normalisation condition for which the power counting breaks down for extremally low energies or there is no power counting at all ) .
if one wants the power counting to be valid for high enough energies , one should use an appropriately chosen normalisation condition . while the complete expressions of physical quantities should not depend on this choice the renormalised feynman diagrams certainly do .
subtractively renormalised effective field theory encounters quite severe ( technical ) problems : if one takes the potential up to some order and tries to iterate via the lippmann - schwinger equation one will encounter divergences .
one could try to include counter - terms in the potential , but due to the non - renormalizability of the theory the inclusion of an infinite number of terms with more and more derivatives would be needed .
one could even think that weinberg s power counting breaks down because higher order counter - terms are strongly involved .
but it should be remembered that power counting ( for both amplitudes and the potentials ) is valid after renormalization when the contributions of counter terms are already taken into account @xcite .
as was explained in details in @xcite and @xcite this involvement of higher order counter - terms into low order calculations do not invalidate weinberg s power counting arguments .
so , one has either to exactly solve ( formally ) the equation and after subtract divergences explicitly , or otherwise one should draw all relevant diagrams , subtract them and then sum these renormalised diagrams up . in recent papers @xcite kaplan et .
al suggested a systematic method of summation of an infinite number of diagrams using dimensional regularization and the power divergent subtraction scheme .
but , as was mentioned in the above cited papers , in the theory with explicitly included pions for the external momenta exceeding 100 mev it is difficult to justify suggested approximations ( in particular the perturbative inclusion of pions ) .
so for higher energies the problem of summation of renormalized diagrams remains open .
fortunately these problems can be overcome using cut - off theory .
one can calculate up to any desired order , but there is a very crucial question : what is the relation between subtractively renormalised and cut - off theories ?
this question is addressed in a number of papers @xcite , but as yet the complete answer has not been determined .
moreover some authors question the validity and systematic character of cut - off theory calculations ( see for example @xcite ) .
this work is an attempt to investigate some details about the equivalence of subtractively renormalized and cut - off theories .
if one applies weinberg s power counting directly to the amplitude of @xmath5 scattering one gets a series : @xmath6 where @xmath7 is suppressed by @xmath8-th order of a small parameter . each term in this infinite series is a sum of an infinite number of diagrams itself .
these diagrams are of the same order .
if translated into the language of the potential , @xmath9 contains all diagrams which are obtained when leading order potential is iterated via the lippmann - schwinger equation .
@xmath10 contains diagrams with one insertion of the first order potential and arbitrary number of the leading order potential .
@xmath11 contains all diagrams with one insertion of second order diagram or two insertions of first order potential etc .
( note that for a theory with just nucleons , which is considered in this paper , all @xmath12 vanish for odd @xmath13 . )
the expansion parameter in ( [ nuexpampl ] ) is @xmath14 where @xmath15 is of the order of external momenta and @xmath16 is expected to be of the order of the mass of lightest particle which was integrated out . if @xmath17 then the first few terms of ( [ nuexpampl ] ) should approximate the whole amplitude well .
if applied to the complete theory the equivalence of subtractively renormalised and cut - off theories would require that they produce identical series for scattering amplitudes provided that the same normalisation condition is used . in actual calculations
one takes a potential up to some order and solves lippman - schwinger equation .
scattering amplitude determined from lippman - schwinger equation contains all contributions up to given order ( the order of the potential ) and also some parts of higher order contributions i.e. some diagrams contributing to @xmath12 with large @xmath13 are included and others are not .
as far as these contributions are small it is not inconsistent to include part of higher order contributions while other parts are not included ; the error is of the order of neglected terms . beyond the range of validity of power counting high order contributions become large and make the complete expression unreliable . as for
the equivalence of subtractively renormalised and cut - off theories one would expect that they will give identical results up to the order of considered approximation .
the difference between two results should be small , of the order of neglected terms . of course one would not expect that this difference is small beyond the range of validity of power counting .
( [ nuexpampl ] ) generates an expansion in the same small parameter for the inverse amplitude @xmath18 if two expressions for the amplitude generated by
subtractively renormalised and cut - off theories are equal up to some order and the difference between them is small then the same is true for inverse amplitudes and vice - versa . below the simple example of contact interaction of nucleons in @xmath19 wave
is considered .
the amplitude is renormalized by subtracting divergent integrals at some normalisation point and its relation to the amplitude obtained from cut - off theory is studied .
the numerical values of phase shifts obtained from subtractively renormalised and cut - off theories ( without removing cut - off ) are compared .
for the very low energy nucleon - nucleon scattering processes the pions can be integrated out and the effective non - relativistic lagrangian takes the following form @xcite : @xmath21 @xmath22 where the nucleonic field @xmath23 is a two - spinor in spin space and a two - spinor in isotopic spin space and @xmath24 are the pauli matrices acting on spin indices .
@xmath25 is the mass of nucleon and the ellipses refer to additional 4-nucleon operators involving two or more derivatives , as well as relativistic corrections to the propagator . @xmath26 and @xmath27 are couplings introduced by weinberg @xcite , they are of dimension @xmath28 and @xmath29 is of the order @xmath30 .
the leading order contribution to the 2-nucleon potential is @xmath31 in the @xmath32 wave it gives : @xmath33 where @xmath34 .
the next to leading order contribution to the 2-nucleon potential in the @xmath32 channel takes the form : @xmath35 the formal iteration of the potential @xmath36 using the lippmann - schwinger equation gives for on - shell ( @xmath37 ) @xmath38-wave @xmath2-matrix @xcite : @xmath39 @xmath40 where @xmath4 is the on - shell momentum and @xmath41 , @xmath42 and @xmath43 are divergent integrals . to renormalize ( [ 2 ] ) it is necessary to include contributions of an infinite number of counter - terms with higher and higher ( up to infinity ) derivatives @xcite .
while it is impossible to write down all these contributions explicitly it is quite straightforward to renormalize ( [ 2 ] ) by just subtracting divergent integrals . before implementing this scheme it would be useful to write down some of the leading and @xmath20 order counter - terms .
one can write down the chiral expansion for @xmath2-matrix @xcite ( it is equivalent to an expansion of @xmath2 obtained from ( [ 2 ] ) in powers of @xmath29 ) : @xmath44 and @xmath45 the final goal is to absorb divergences in ( [ chtmatrix ] ) into @xmath46 and @xmath29 which are to be given by @xmath47 where @xmath48 and @xmath49 are renormalized coupling constants and @xmath50 , @xmath51 , @xmath52 ... are functions of @xmath48 . to determine @xmath46 and @xmath29 in terms of @xmath48 and @xmath49 it is simpler to work with ( [ invchtm ] ) and require :
@xmath53 where @xmath54 and @xmath55 are divergent parts of @xmath56 and @xmath42 integrals ( with arbitrary finite contributions ) .
equating coefficients of different powers of @xmath4 one gets from ( [ npct ] ) : @xmath57 and from ( [ npct1 ] ) : @xmath58 ^ 2\right)^{1\over 2}\over 4c_2^ri_3 } \label{npct2}\ ] ] in ordinary perturbation theory ( expansion in terms of coupling constants ) one has @xmath59 . the non - perturbative expression ( [ npct2 ] ) respects this perturbative expansion if the @xmath60 sign is taken .
expanding the chosen solution in powers of @xmath61 and keeping only terms of first order one obtains : @xmath62 and @xmath63 substituting ( [ npct3 ] ) and ( [ npct4 ] ) into ( [ chtmatrix ] ) one gets a finite renormalized expression : @xmath64}+{2p^2c_2^r+ 2c_rc_2^r\left ( i_3-\delta_3\right)\over \left ( 1-c_r\left [ i(p)- \delta\right]\right)^2}+ ...
\label{renchtmatrix}\ ] ] so , to get rid off divergences from ( [ chtmatrix ] ) one has just to express bare couplings in terms of renormalised ones using ( [ npct3 ] ) and ( [ npct4 ] ) and substitute into ( [ chtmatrix ] ) .
one gets an expression with subtracted integrals and renormalised coupling constants .
switching back to ( [ 2 ] ) one can apply the subtraction scheme analogous to the one originally used by weinberg @xcite and subtract divergent integrals at @xmath65 .
integrals are divided into two parts : @xmath66= i_n^d+i_n^r , \
n=3,5\ ] ] @xmath67= i^d+i^r(p)\ ] ] where @xmath68 and @xmath69 are divergent parts and are to be cancelled by contributions of counter - terms . to absorb all contributions of @xmath68 and @xmath69 in ( [ 2 ] )
one needs to include contributions of an infinite number of counter - terms with higher and higher order ( up to infinity ) derivatives @xcite .
while it is impossible to write down these counter terms explicitly , one can take their contributions into account by just neglecting @xmath68 and @xmath69 terms and replacing @xmath46 and @xmath29 by renormalized couplings .
finally the amplitude is left with finite parts of integrals @xmath70 and @xmath71 ( note that @xmath72 ) : @xmath73 where @xmath74 note that the subtraction scheme used here is just one among an infinite number of possibilities . any scheme which puts effective cut - off of the order of external momenta should be equally good .
( [ rentmp2 ] ) is not obtained from ( [ 2 ] ) by just expressing @xmath46 and @xmath29 in terms of renormalised coupling constants .
contributions of infinite number of counter - terms have been taken into account . while the imaginary part of ( [ 2 ] ) can not be altered by adding contributions of counter - terms ,
the real part depends on finite parts of those counter - terms . hence starting from the same ( inverse )
amplitude in terms of bare parameters one can get quite different expressions for renormalised ( inverse)amplitude .
all these amplitudes are equivalent up to the order one is working with provided that chosen normalisation conditions respect power counting .
note that although the expression ( [ rentmp2 ] ) was obtained in @xcite using power divergent subtractions , that scheme is completely different from the one applied in this work . the difference is clearly seen when pions are included explicitly . matching ( [ rentmp2 ] ) to the effective range expansion @xmath75 gives @xmath76 and from ( [ rentmp2 ] ) @xmath77 the result given in ( [ rentmp2ere ] ) does not depend on the regularization scheme .
below a cut - off version of the above effective theory is considered and it is demonstrated that these two approaches are equivalent up to ( including ) @xmath20 order .
effective potential with sharp cut - off has the following form : @xmath78 here @xmath79 is the cut - off parameter and @xmath80 and @xmath81 depend on this parameter .
@xmath79 should be of the order of the mass of lightest particle which was integrated out @xcite .
it is not difficult to write down the solution of the lippmann - schwinger equation explicitly ( see @xcite ) : @xmath82 where @xmath83 @xmath84-{imp\over 4\pi } \label{in1}\ ] ] matching ( [ t2 ] ) to the effective range expansion ( [ effre ] ) leads to : @xmath85 and the solution of these equations for @xmath80 and @xmath81 gives : @xmath86 \label{c2cut } \\
\bar c={1\over x+yi_3^l}-\bar c_2 ^ 2i_5^l \label{ccut}\end{aligned}\ ] ] @xmath81 was obtained by solving quadratic equation .
analogously to ( [ npct2 ] ) the sign in solution was fixed respecting the structure of ordinary perturbation theory ( expansion in coupling constants ) . substituting ( [ c2cut ] ) and ( [ ccut ] ) into ( [ t2 ] )
one gets : @xmath87 note that higher - order corrections to the cut - off expression are suppressed by powers of @xmath88 and hence are small for momenta well below the cut - off . the solution of @xmath89 and @xmath90 from ( [ matchingc ] ) and ( [ matchingc2 ] ) ( for some value of @xmath91 ) and substitution into ( [ c2cut ] ) and ( [ ccut ] ) leads to a lengthy but simple relation between @xmath80 , @xmath81 and @xmath48 , @xmath49 the fulfilment of which guarantees that the cut - off and subtractively renormalised inverse @xmath2-matrices are equal up to ( including ) @xmath20 order .
this equality is manifested by ( [ rentmp2ere ] ) and ( [ t2efe ] ) .
consequently in terms of @xmath92 expansion of the feynman amplitude given in @xcite the two amplitudes are equal up to ( including ) sub - leading order .
higher order corrections to the cut - off expression are suppressed by powers of cut - off parameter @xmath79 which should be taken of the order of lightest integrated particle , so they are of the order of terms which are neglected by the approximation taken in renormalized theory . substituting actual values for scattering length and effective range @xmath93 and @xmath94 into ( [ rentmp2ere ] ) and ( [ t2efe ] )
one can calculate the phase shifts .
the results for @xmath95 and @xmath96 ( note that @xmath97 does not violate the power counting at least for the present problem ) are plotted in fig.1 . as is seen from this graph
if one takes @xmath98 the cut - off phase shifts are in good agreement with the ones of subtractively renormalised theory for the momenta up to @xmath99 . if smaller value of @xmath91 is taken then the range of the validity of power counting decreases .
so does the range of the momenta for which the results of two approaches are in good agreement . as was mentioned above the power counting
depends on the chosen normalisation condition .
one does not expect good agreement between results of subtractively renormalised and cut - off theories unless good normalisation condition is chosen .
the @xmath100 graph shows the failure of @xmath101 renormalised theory encountered in @xcite .
one can also calculate the numerical values of cut - off coupling constants @xmath102 and @xmath103 . to study the dependence of phase shifts on cut - off parameter the phase shifts for different values of this parameter are plotted in fig.2 .
it is seen that for cut - offs @xmath104 phase shifts do not depend on cut - off up to momenta @xmath105 .
note that figures quite analogous to fig.1 and fig.2 but in different context and with different subsequent conclusions are given in @xcite .
to estimate the corrections from the next orders let us consider the @xmath106 order potential : @xmath107 this potential can be written as separable one : @xmath108 where @xmath109 the relativistic corrections are suppressed by the mass of the nucleon and hence they are not included .
a straightforward generalisation of calculations with @xmath20-order potential given in @xcite leads to the following expression : @xmath110 where @xmath111 @xmath112 @xmath113 and @xmath114 @xmath115 + i_7\left ( { 2p^2bb_1}+{2p^2b^2}+{2c_2b}\right)+b^2i_9\ ] ] @xmath116 analogously to the @xmath20 order case one can renormalize ( [ tmp4 ] ) by subtracting divergent integrals at @xmath65 and get : @xmath117 where @xmath48 , @xmath61 , @xmath118 and @xmath119 are renormalised coupling constants and @xmath120 comparing ( [ rentmp4 ] ) to the effective range expansion @xmath121 gives @xmath122 introducing a sharp cut - off ( factor of @xmath123 ) into the potential ( [ potmp4 ] ) and solving lippmann - schwinger equation for the @xmath2-matrix one gets the expressions ( [ tmp4 ] ) , ( [ t3num ] ) , ( [ t3denum ] ) with @xmath124 replaced by @xmath125 , @xmath56 replaced by @xmath126 and cut - off dependent couplings @xmath80 , @xmath127 , @xmath128 and @xmath129 . for the purposes of this work one can take @xmath130 . while simplifying calculations significantly this value is quite satisfactory as far as adjusting remaining parameters one can satisfy the equivalence criteria . substituting this value
one gets : @xmath131 comparing ( [ b0t3 ] ) to the effective range expansion ( [ effrep4 ] ) , after a lengthy but straightforward calculation one obtains : @xmath132^{1\over 2}\right\ } \label{matchingcutpar2 } \\ \bar c=-{2i_5\bar c_2\over i_3}+{i_3+zi_5\over i_3x+i_3 ^ 2z } \label{matchingcutpar3}\end{aligned}\ ] ] where @xmath133 and @xmath134 solving for @xmath89 , @xmath90 and @xmath135 from ( [ a4r])-([matching ] ) and substituting into ( [ matchingcutpar1])-([a3 ] ) one obtains lengthy algebraic relations between @xmath80 , @xmath81 , @xmath129 and @xmath48 , @xmath49 , @xmath118 and @xmath136 , the fulfilment of which along the condition @xmath130 guarantees the equality of cut - off and subtractively renormalized inverse amplitudes up to ( including ) @xmath106 order , and consequently the @xmath2 matrices of two approaches are equal up to ( including ) @xmath137 order .
the higher order corrections to the cut - off expression are again suppressed by powers of @xmath88 and hence are small for momenta well below the cut - off . substituting the numerical values for @xmath93 , @xmath138 , @xmath139 (
the first two terms in effective range expansion describe experimental data quite well so there is no need to determine @xmath135 from data at least for the purposes of this paper ) and @xmath95 one can calculate coupling constants @xmath140 , @xmath141 ( the sign `` - '' in ( [ matchingcutpar2 ] ) is again chosen respecting the structure of ordinary perturbation theory ) , @xmath142 . using these values
one calculates phase shifts from ( [ b0t3 ] ) .
these phase shifts are compared with results of effective range expansion and also of ( [ rentmp4 ] ) in fig .
3 . the phase shifts of cut - off theory agree well with ones obtained from subtractively renormalised theory for all momenta for which the second approach describes the data well .
one did not expect the results of two approaches agree well beyond this range .
in fig.4 the phase shifts for different values of the cut - off parameter are plotted .
it can be seen that for cut - offs @xmath104 phase shifts are cut - off independent up to @xmath143 .
in the simple example of low energy effective field theory for nucleon - nucleon scattering it was demonstrated that the cut - off theory can reproduce the scattering amplitude calculated using subtractively renormalised ( using off mass - shell subtractions ) theory up to the order of accuracy of the considered approximation , the difference between two amplitudes being of higher order ( in small expansion parameter ) .
this simple example serves as a demonstration of some more general considerations about cut - off field theories formulated below .
using chiral power counting originally developed by weinberg one can find the potential up to any desired order . then to remove divergences one can impose cut - off regularization .
the cut - off regularization destroys chiral and gauge symmetries and to restore them it is necessary to include additional terms into the lagrangian ( and consequently into the potential ) . cut - off dependence of the physical quantities can be removed systematically by including additional terms in the lagrangian @xcite .
the power - law divergences , which caused higher order operators to be involved in the renormalization of the diagrams obtained by iterating the low order potential , now emerge as powers of the cut - off parameter .
as far as cut - off should be taken of the order of masses of particles which were integrated out , it should be clear that cut - off regularization does not respect power counting and it seems that imposing this regularization will destroy the whole machinery . ( the problem can not be solved by imposing a small cut - off as cut - off regularized
integrals contain inverse powers of cut - off parameters as well ) .
however the large factors which seem to threaten power counting can be absorbed by redefining the couplings already included into the potential @xcite .
fitting the parameters of the cut - off theory one can reproduce the results of the subtractively renormalised theory up to the order of accuracy determined by approximation made in the potential .
the results of cut - off theory are as reliable as the ones of subtractively renormalised theory , the error being of the order of terms neglected in the potential .
as far as the cut - off is of the order of the mass of lightest particle which was integrated out the higher order ( in momenta ) cut - off dependent corrections are suppressed by powers of this parameter . by increasing the cut - off parameter one could make the mentioned corrections smaller but for large cut - off it would be problematic to include the positive powers of the cut - off parameter in to a redefinition of the coupling constants .
so , the equivalence between subtractively renormalized and cut - off theories can be achieved only for the cut - off of the order of the mass of lightest particle which was integrated out .
the results of two approaches are expected to deviate significantly beyond the range of validity of the power counting .
although the difference between two expressions is of higher order it becomes large for the momenta for which the power counting breaks down in subtractively renormalised theory .
the range of validity of power counting depends on normalisation condition .
hence the range of the equivalence i.e. the range of the momenta for which the difference between results of two approaches is small is determined by the choice of normalisation condition in subtractively renormalised theory as well as by the choice of the value of cut - off parameter in cut - off theory .
choosing optimal conditions one can claim that two approaches lead to equivalent results for the range of the momenta ( energy ) for which the power counting is at work .
one should conclude that the doubts about consistency and systematic character of cut - off theories @xcite are ungrounded .
so the reasonable success of the cut - off chiral perturbation theory originally started with work @xcite should not be a surprise . although the cut - off theory is technically a little complicated it has a great advantage in that one can determine amplitudes from equations .
note that there are no self - contained equations for subtractively renormalised amplitudes in these non - renormalizable ( in the traditional sense ) effective field theories and one instead has to sum up renormalised diagrams .
this work was carried out whilst the author was a recipient of an overseas postgraduate research scholarship and a flinders university research scholarship at the flinders university of south australia .
s.weinberg , _ physica _ * 96 a * ( 1979 ) 327 ; s.weinberg , phys.lett . * b251 * , 288 ( 1990 ) .
s.weinberg , nucl.phys .
* b363 * , 3 ( 1991 ) .
c.ordonez , l.ray and u. van kolck , phys.rev .
c * 53 * , 2086 ( 1996 ) .
t.d.cohen , phys.rev .
c * 55 * , 67 ( 1997 ) .
d.r.phillips and t.d.cohen , phys.lett . *
b390 * , 7 ( 1997 ) , nucl - th/9607048 .
k.a.scaldeferri , d.r.phillips , c .- w.kao , and t.d.cohen , phys.rev.c56:679-688 , 1997 , nucl - th/9610049 .
d.r.phillips , nucl - th/9804040 ; d.r.phillips , s.r.beane , and t.d.cohen , hep - th/9706070 ; s.r.beane , t.d.cohen and d.r.phillips , nucl - th/9709062 ; u. van kolck , hep - ph/9711222 ; u. van kolck , nucl - th/9808007 ; s.k .
adhikari , and a. ghosh , j. phys .
a * 30 * , 6553 ( 1997 ) ; c.f .
de araujo , jr .
, l. tomio , and s.k .
adhikari , ibid . , 4687 ; s.k .
adhikari , t. frederico , and i.d .
goldman , phys .
lett . * 74 * , 487 ( 1995 ) .
d.b.kaplan , m.j.savage , and m.b.wise , nucl.phys . * b478*,629 ( 1996 ) ; d.b.kaplan , m.j.savage , and m.b.wise , nucl - th/9801034 ; d.b.kaplan , m.j.savage , and m.b.wise , nucl - th/9802075 ; j.gegelia , phys.lett .
* b429 * , 227 ( 1998 ) , nucl - th/9802038 ; j.gegelia , nucl - th/9806028 ; t .- s.park , k.kubodera , d .- p.min and m.rho , nucl - th/9807054 ; g.p.lepage , nucl - th/9706029 ; g.p.lepage , _ what is renormalization ? , _ in _ from actions to answers , _ edited by t.degrand and d.toussaint , world scientific press ( singapore,1990 ) .
e.epelbaoum , w.gl " ockle , ulf - g .
meissner nucl - th/9801064 ; e.epelbaoum , w.gl " ockle , ulf - g .
meissner nucl - th/9804005 ; | the equivalence of subtractively renormalized and cut - off effective field theories is demonstrated for the example of very low energy effective field theory for the nucleon - nucleon interaction . |
during the expansion of the outcoupled atoms , optical levitation is performed with a blue - detuned @xmath67 nm laser beam to compensate for gravity and a radio frequency dressing @xcite is used to keep the out - coupled fraction confined and clearly detectable after the @xmath9 ms expansion . in particular , the rf field is such to produce a mexican - hat potential which limits the radial expansion to about @xmath68 @xmath25 m , whereas the slower axial expansion is barely perturbed .
@xmath25 m .
the in - situ value can be obtained considering a scale factor of @xmath69 , given by the ratio between the in - situ and expanded tf radius at @xmath46 ms ; this because the assumption of a constant @xmath70 during the expansion .
this gives a mean @xmath71 of @xmath72 with a standard deviation of @xmath73 .
there is no statistical difference between the single - vortex distribution and the double - vortex one . ]
a precise statistical analysis is not possible here because information on the phase shift can be extracted only in the data subset where the crossing point occurs at about half of the inspected time evolution ( @xmath74 of the cases ) .
clear phase shifts are present in about half of this subset . | we study the real - time dynamics of vortices in a large elongated bose - einstein condensate ( bec ) of sodium atoms using a stroboscopic technique .
vortices are produced via the kibble - zurek mechanism in a quench across the bec transition and they slowly precess keeping their orientation perpendicular to the long axis of the trap as expected for solitonic vortices in a highly anisotropic condensate .
good agreement with theoretical predictions is found for the precession period as a function of the orbit amplitude and the number of condensed atoms . in configurations with two or more vortices ,
we see signatures of vortex - vortex interaction in the shape and visibility of the orbits .
in addition , when more than two vortices are present , their decay is faster than the thermal decay observed for one or two vortices .
the possible role of vortex reconnection processes is discussed .
vortex dynamics is an essential feature of quantum fluids @xcite and plays a key role in superfluid helium @xcite , superconductors @xcite , neutron stars @xcite and magnetohydrodynamics @xcite .
the interaction between vortices is crucial for understanding the formation of vortex lattices in rotating superfluids and is the basic mechanism leading to quantum turbulence _ via _ vortex reconnection @xcite .
vortices have been extensively investigated in atomic gases @xcite , where a variety of techniques permits the observation of single ones up to a few hundreds , interacting in a clean environment and on a spatial scale ranging from the healing length ( core size ) @xmath0 to a few tens of @xmath0 .
the fact that atoms are confined by external fields of tunable geometry makes them suitable to explore the physics of reconnection and dissipation in inhomogeneous systems and in the presence of boundaries .
seminal experiments were performed in rotating bose - einstein condensates ( becs ) , where the effect of rotation and long - range interaction favors vortex alignment and the formation of vortex lattices @xcite and hence crossing and reconnection mechanisms are inhibited .
interacting vortices have been observed in nonrotating oblate becs , where vortex lines are short and either parallel or antiparallel , thus behaving as pointlike particles dominated by their long - range interaction in a quasi-2d background @xcite .
in our experiment we use a cigar - shaped bec which is particularly suitable for studying the dynamics of vortex lines in 3d . because of the boundary conditions imposed by the tight radial confinement each vortex line lies in a plane perpendicular to the long axis @xmath1 of the trap , such to minimize its length and therefore its energy , as in the solitonic vortex configuration predicted in refs . @xcite and recently observed both in a bec @xcite and in a superfluid fermi gas @xcite .
the line is randomly oriented in the plane and away from it , at distances of the order of the system transverse size , the superfluid flow quickly vanishes and the long - range part of the vortex - vortex interaction is suppressed .
hence , vortices can move almost independently along elliptic orbits except when they approach each other and may collide with a random relative angle . at the scale of the healing length , where reconnection can take place , the system is still equivalent to a uniform superfluid , like liquid he , but with the advantage that vortex filaments collide at measurable relative velocities .
the experimental apparatus is described in ref .
@xcite .
we evaporate sodium atoms in a magnetic harmonic trap with frequencies @xmath2 hz .
vortices with random position and velocity spontaneously originate _ via _ the kibble - zurek mechanism @xcite from phase defects in the condensate when crossing the bec transition and their average number scales as a power law with the evaporation rate . at the end of the evaporation we have an almost pure prolate bec with about @xmath3 atoms at @xmath4 nk in the state @xmath5 . in refs .
@xcite we counted and characterized defects using destructive absorption imaging . here
we apply a stroboscopic technique , similar to that in refs .
@xcite , which allows us to observe the real - time dynamics .
starting from an initial number of atoms @xmath6 , we remove a small fraction @xmath7 by outcoupling them to the antitrapped state @xmath8 _ via _ a microwave pulse , short enough to provide a resonance condition throughout the whole sample .
outcoupled atoms are imaged along a radial direction after a @xmath9 ms expansion @xcite without affecting the trapped ones .
the extraction mechanism is repeated @xmath10 times with time steps @xmath11 , keeping @xmath12 fixed .
raw images are fitted to a thomas - fermi ( tf ) profile @xcite and the residuals are calculated .
because of the peculiar structure of the superfluid flow of solitonic vortices @xcite , after expansion the whole radial plane containing a vortex exhibits a density depletion and vortices are seen as dark stripes independently of their in - plane orientation . during the extraction sequence
the remaining condensate evolves in trap , only weakly affected by atom number change , provided @xmath13 is sufficiently small .
we can then identify the axial position of the vortex in each image of the outcoupled atoms and analyze its oscillation as a faithful representation of the in - trap dynamics .
typical examples are shown in figs .
[ figure1](a)-[figure1](i ) .
alternatively we image the full bec along the axial direction after a long expansion with a destructive technique as in @xcite and directly see the shape and orientation of the vortex lines as in figs .
[ figure1](j)-[figure1](m ) .
images of the density distribution of the atoms extracted from three becs ; frames are taken every @xmath14 ms , each after a @xmath9 ms expansion .
( a ) static vortex .
( b)-(c ) vortices precessing with different amplitudes .
each vortex is randomly oriented in the @xmath15 plane and , after expansion , it forms a planar density depletion @xcite which is visible as a stripe .
( d)-(i ) sequences with two and three vortices , with @xmath16 ms ; here frames are not to scale and vertically squeezed to enhance visibility .
( j)-(m ) destructive absorption images of the whole bec taken along the axial direction @xmath1 after @xmath17 ms of expansion , showing ( j ) a single vortex filament crossing the condensate from side to side and ( k)-(m ) two vortices with different relative orientation and shape .
all images show the residuals after subtracting the fitting tf profile . ]
we first choose an evaporation rate of @xmath18 khz / s , yielding one vortex in each bec on average .
from the sequence of radial images we extract the axial position of each vortex @xmath19 .
frames are recorded every @xmath14 ms . figures [ figure2](a ) and [ figure2](b ) show two examples corresponding to the raw images of figs .
[ figure1](b ) and [ figure1](c ) , respectively .
the observations are consistent with a vortex precession around the trap center , as the one observed in oblate becs @xcite . in a nonrotating elongated condensate , a straight vortex line , oriented in a radial plane ,
is expected to follow an elliptic orbit in a plane orthogonal to the vortex line , corresponding to a trajectory at constant density @xcite .
the observed motion of each dark stripe in figs .
[ figure1](a)-[figure1](c ) is the axial projection of such a precession .
given @xmath20 the in - trap amplitude of the orbit normalized to the tf radii @xmath21 and @xmath22 @xcite , the precession period is predicted to be @xmath23 where @xmath24 is the axial trapping period and @xmath0 is related to the chemical potential @xmath25 by @xmath26 .
this result , which is valid to logarithmic accuracy , has been derived for a disk - shaped nonaxisymmetric condensate in refs .
@xcite within the gross - pitaevskii theory at @xmath27 and in the tf approximation , corresponding to @xmath28 ( in our case , @xmath29 ranges from @xmath30 to @xmath10 ) .
it can also be obtained by means of the superfluid hydrodynamic approach introduced in ref .
@xcite to describe the motion of vortex rings in elongated condensates , appropriately generalized to the case of solitonic vortices as in ref .
@xcite .
the quantity @xmath31 is the local chemical potential along the vortex trajectory and we assume @xmath32 to be constant during expansion , as distances are expected to scale in the same way in the slow axial expansion . in comparing the observed period with eq .
( [ eqn : period ] ) we must consider that the number of atoms is decreasing from shot to shot . since extraction is spatially homogeneous , the gradients of the density , and hence the equipotential lines for the vortex precession and the orbit amplitude remain almost unchanged .
however , @xmath33 ( hence @xmath34 ) decreases in time and so does the vortex orbital period @xmath35 , as is clearly visible in figs .
[ figure2](a ) and [ figure2](b ) .
we define an instantaneous period at time @xmath36 as the period obtained from a sinusoidal fit to the measured position in a time interval centered at @xmath36 and containing about one oscillation .
such @xmath37 is plotted in fig .
[ figure2](c ) and [ figure2](d ) and compared to eq .
( [ eqn : period ] ) , where we include the effect of the observed @xmath36 dependence on @xmath38 , shown in fig .
[ figure2](e ) , both in @xmath25 and @xmath0 .
the agreement is good , the major limitation being the experimental uncertainty in @xmath38 .
we also show the period expected for the oscillation of a dark or grey soliton , which is @xmath39 independently of @xmath38 @xcite . in fig .
[ figure2](f ) we plot the period of vortices orbiting with different amplitude @xmath32 .
the agreement with theory is again good and can be further appreciated by considering the ratio between each value of @xmath35 measured at a given @xmath32 and the theoretical value in eq .
( [ eqn : period ] ) obtained for the same @xmath32 and @xmath38 .
figure [ figure2](g ) shows the histogram of all values obtained by extracting @xmath35 and @xmath32 from a fit to the first oscillation , using @xmath40 in eq .
( [ eqn : period ] ) .
the histogram gives @xmath41 .
this remarkable agreement with theory is nontrivial since eq .
( [ eqn : period ] ) assumes @xmath42 and a rigid straight vortex line , while off - centered vortices actually bend toward the curved bec surface . for rotating condensates the bending mechanism has been discussed in refs . @xcite and observed in ref .
@xcite .
examples of straight and bent vortices in our condensate are given in figs .
[ figure1](j)-[figure1](m ) . in our elongated bec , with strong radial inhomogeneity
, this bending mechanism is expected to be more effective than in oblate becs .
our observations seem to indicate that its effect on the period is small , possibly of the same order of the logarithmic corrections to eq .
( [ eqn : period ] ) predicted for a straight vortex in a 2d geometry @xcite . this may be due to the fact that the difference in length between a bent and a straight vortex , at a comparable @xmath32 , is relatively small and the overall structure of the vortical flow is also quite similar , so that the key quantities entering the hydrodynamic description ( i.e , the force acting on a unit of length of the vortex and the momentum of the vortex , in the language of ref .
@xcite ) are almost the same in the two cases . ,
remaining in a condensate at time @xmath36 starting from configurations with @xmath43 ( circles ) , @xmath44 ( triangles ) and @xmath45 ( diamonds ) at @xmath46 .
solid lines are exponential fits . ]
vortex lifetime in nonrotating becs is limited by scattering of thermal excitations , which causes the dissipation of the vortex energy into the thermal cloud . since a vortex behaves as a particle of negative mass , dissipation causes an antidamping of the orbital motion and vortices decay at the edge of the condensate @xcite .
we can measure the lifetime @xmath47 by counting the average number of vortices @xmath48 remaining in the condensate at time @xmath36 , starting with @xmath49 .
if @xmath50 we find a clear exponential decay with @xmath51 ms ( fig .
[ figure3 ] ) , close to that measured in refs . @xcite and of the same order of the one observed in a fermionic superfluid @xcite .
( d)-[figure1](g ) , respectively .
solid and empty symbols are used to distinguish high and low density contrast , respectively . ] using a faster evaporation ramp ( @xmath52 khz / s ) , we produce more vortices and search for signatures of mutual interaction .
examples are shown in fig .
[ figure1](d)-[figure1](i ) and typical trajectories are also reported in fig .
[ figure4 ] . in some cases
, vortices perform unperturbed oscillations [ fig .
[ figure4](a ) ] ; in others , we clearly see a shift in their trajectories at the crossing point [ fig .
[ figure4](b ) ] .
the average relative velocity at the crossing in the latter case is systematically smaller ( @xmath53 mm / s ) than in the former ( @xmath54 mm / s ) @xcite .
the shift has a consequence also in the determination of the orbital period as it causes a broadening of the probability distribution of the ratio @xmath55 which now gives @xmath56 , with a standard deviation three times larger than for the single vortex [ fig .
[ figure2](g ) ] . in addition
, crossings are frequently associated with a sudden change of visibility of one or both vortices [ figs .
[ figure1](e)-[figure1](h ) ) .
finally , by analyzing the lifetime of vortices for the initial condition @xmath57 and @xmath58 we observe a lifetime @xmath59 ms for the two - vortex configuration , consistent with the one - vortex configuration .
the situation instead changes in the three - vortex configuration , where a faster decay is observed , @xmath60 ms ( fig .
[ figure3 ] ) .
the frequent observation of unperturbed orbits for multiple vortices is intriguing .
two vortex lines moving back and forth in the condensate with random radial orientations should have large probability to cross each other at some point .
if crossings occur , reconnections are expected to take place @xcite with possible drastic ( and almost temperature independent @xcite ) effects on the vortical dynamics .
the actual dynamics can strongly depend on the relative angle @xmath61 between vortex lines as well as the relative velocity @xmath62 between the planes where they lie .
when @xmath61 is close to @xmath63 ( @xmath64 ) , the vortex lines tend to align ( antialign ) , thus reducing the chance of reconnection for vortices on different orbits .
but when vortices approach with @xmath65 reconnection can be hardly avoided .
the fact that we observe the same vortex lifetime for @xmath50 and @xmath44 implies that such reconnections are either suppressed or they induce a negligible dissipation .
a possible explanation is the occurrence of double reconnection processes @xcite .
vortex reconnection corresponds to the switching of a pair of locally coplanar vortex lines , accompanied by a change of topology . in our geometry
a finite @xmath62 implies that the newly formed filaments must stretch in the condensate while the two planes separate again after reconnection .
the consequent energy cost is instead avoided if vortices perform a consecutive second reconnection when they are still at close distance .
this would preserve the vortex number , consistent with our observation of an equal vortex lifetime for @xmath50 and @xmath44 .
it is worth mentioning that a similar scenario has also been recently suggested for the collision of cosmic strings @xcite .
the occurrence of a shift in the trajectories , that apparently depends on @xmath62 , could be associated with the role of the collision time : faster vortices have less time to interact and their trajectories are marginally affected , and this scenario may be applicable both to fly - by vortices and double reconnections . also kelvin modes can be excited in the collision @xcite but , if present , they seem not to affect the lifetime , while they are likely responsible for the change of visibility of the vortices , as they can produce out - of - plane distortions and hence a change of contrast in the density distribution .
finally , the observation of a shorter lifetime in configurations with @xmath66 can be understood by considering the role of a third vortex in the collision of two other vortices , whose tendency to rotate in the radial plane is frustrated by three - body interaction , thus enhancing the probability of collisions and reconnections .
a similar role of three - body interactions in the dynamics of vortices was recently investigated in the context of 2d classical turbulence @xcite .
our experimental results demand new theoretical models .
so far , numerical simulations of vortex reconnection are usually performed with vortex lines initially at rest , at small distance , which then evolve in time @xcite , while in our case the role of the relative velocity seems to be crucial .
shedding light on this , and generally on the dynamics of few vortices in such a relatively simple configuration , can help to understand the physics of vorticity in more complex settings , like those of refs .
@xcite , in the search of a satisfactory comprehension of quantum turbulence in superfluids with boundaries .
we thank l.p .
pitaevskii , n.p .
proukakis , i - kang liu , n.g .
parker and c.f .
barenghi for insightful discussions .
we acknowledge provincia autonoma di trento for funding .
63ifxundefined [ 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 in link:\doibase 10.1016/s0079 - 6417(08)60077 - 3 [ _ _ ] , vol . , ( , ) p. @noop _ _ ( , ) p. @noop _ _ ( , ) p.
link:\doibase 10.1146/annurev.ns.25.120175.000331 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.115.025001 [ * * , ( ) ] link:\doibase 10.1146/annurev - conmatphys-062910 - 140533 [ * * , ( ) ] link:\doibase 10.1063/1.4772198 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.81,647 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.84.806 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physrevlett.89.100403 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.90.170405 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.91.100402 [ * * , ( ) ] link:\doibase 10.1038/nature07334 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.104.160401 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physreva.84.011605 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.110.225301 [ * * , ( ) ] link:\doibase 10.1103/physreva.90.063627 [ * * , ( ) ] link:\doibase 10.1103/physreva.65.043612 [ * * , ( ) ] link:\doibase 10.1103/physreva.68.043617 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.113.065302 [ * * , ( ) ] link:\doibase 10.1140/epjst / e2015 - 02389 - 7 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1016/0370 - 1573(80)90091 - 5 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) \doibase http://dx.doi.org/10.1063/1.4747163 [ * * , ( ) ] @noop , * * , ( ) link:\doibase 10.1103/revmodphys.71.463 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.85.2857 [ * * , ( ) ] link:\doibase 10.1103/physreva.70.063620 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop ( ) link:\doibase 10.1103/physrevlett.84.2298 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.93.240403 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physreva.64.043611 [ * * , ( ) ] link:\doibase 10.1103/physreva.63.041603 [ * * , ( ) ] link:\doibase 10.1103/physreva.64.053611 [ * * , ( ) ] link:\doibase 10.1140/epjd / e2003 - 00015-y [ * * , ( ) ] link:\doibase 10.1103/physrevlett.89.200403 [ * * , ( ) ] link:\doibase 10.1103/physreva.61.063612 [ * * , ( ) ] link:\doibase 10.1103/physreva.70.043624 [ * * , ( ) ] link:\doibase 10.1103/physreva.60.r1779 [ * * , ( ) ] @noop * * , ( ) \doibase http://dx.doi.org/10.1016/j.physd.2009.03.006 [ * * , ( ) ] link:\doibase 10.1103/physreva.90.013601 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physrevd.84.105036 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1073/pnas.1312536110 [ * * , ( ) ] link:\doibase 10.1103/physreve.84.056317 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.71.1375 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physreve.58.2576 [ * * , ( ) ] link:\doibase 10.1007/s10909 - 015 - 1285-y [ * * , ( ) ] link:\doibase 10.1103/physrevlett.103.045301 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1007/s10909 - 015 - 1288 - 8 [ * * , ( ) ] * supplemental material * |
there is an increasing interest in @xmath8-meson physics both experimentally and theoretically . on the experimental side
several facilities are now able to produce sufficient @xmath8 s to enable a study to be made of their interactions with other particles .
in particular , the photon machines mami@xcite and graal@xcite are supplementing the earlier hadronic machines such as saturne@xcite , celsius@xcite and cosy@xcite . the current theoretical interest stems partly from the early indications that the @xmath16 interaction is attractive and so could possibly lead to @xmath8-nucleus quasi - bound states ( e.g. refs .
@xcite , @xcite ) .
the theoretical approaches fall into two main categories . in the one ,
the various processes involving @xmath8-meson interactions are described in terms of microscopic models containing baryon resonances and the exchange of different mesons ( e.g. refs .
@xcite , @xcite ) which may be based on a chiral perturbation approach ( e.g.ref .
@xcite ) or a quark model ( e.g. ref . @xcite ) .
unfortunately , this approach requires a knowledge of the magnitudes and relative phases of many hadron - hadron couplings several of which are very poorly known .
in addition , since @xmath8 interactions in the absence of @xmath8-meson beams can only be studied as final state interactions , one has to exploit relationships between the many processes involved .
for example , in the present note , the main interest is in the reaction a ) @xmath7 .
however , this is dependent on the final state interaction b ) @xmath17 , which in turn depends on the reactions c ) @xmath5 and d ) @xmath4 .
similarly , reactions c ) and d ) are related to e ) @xmath6 . therefore ,
any model that claims to describe reaction a ) must also see its implications in reactions b ) , .. , e ) .
this , we believe , is too ambitious a program at present . at this stage
it is probably more informative to check the consistency between the data of the above five reactions and be able to relate them in terms of a few phenomenological parameters .
when this has been accomplished , it will hopefully be possible to understand these parameters in terms of more microscopic models . with this in mind , in ref .
@xcite a @xmath3-matrix model was developed by the authors to describe the reactions a ) , b ) , c ) and d ) in an energy range of about 100 mev each side of the @xmath8 threshold .
this model was expressed in the form of two coupled channels for @xmath18-wave @xmath19 and @xmath16 scattering with the effect of the two pion channel ( @xmath20 ) being included only implicitly .
the latter was achieved by first introducing the two pion process as a third channel in the @xmath3-matrix and subsequently eliminating that channel as an `` optical potential '' correction to the other two channels .
it should be emphasized that this is not an approximation but is done only for convenience , since we do not address cross sections involving explicitly two final state pions . in ref .
@xcite the @xmath8-photoproduction cross section was assumed to be proportional to the elastic @xmath16 cross section ( @xmath21 ) .
this is in line with the so - called watson approximation @xcite . in this way
each of the matrix elements in the two - by - two @xmath22-matrix of ref .
@xcite was associated with some specific experimental data
@xmath23 with the @xmath0 amplitudes of arndt _ et al . _
@xcite , @xmath24 with the @xmath8-production cross section in the review by nefkens@xcite and @xmath25 with the @xmath8-photoproduction cross section of krusche _ et al._@xcite . in this note
we now wish to treat the @xmath2 channel explicitly .
an enlargement of the @xmath3-matrix basis then permits a direct estimate of the matrix element @xmath26 , so that @xmath27 , thereby avoiding the earlier assumption that @xmath28 .
the @xmath3-matrix would now be a four - by - four matrix with the channels @xmath0 , @xmath1 , @xmath29 and @xmath2 . in principle , 10 different processes , corresponding to each matrix element , could be analysed simultaneously .
however , in practice , it is more convenient to elimate some channels by the `` optical potential '' method used already in ref .
we , therefore , describe in section 2 the above reactions in terms of three separate @xmath22-matrices . in section 3 , we give the fitting strategy and also the numerical results in terms of the 13 parameters needed to specify the @xmath3-matrices .
this section also includes expansions in terms of the @xmath8 momentum for the amplitudes of the @xmath17 and @xmath7 reactions near the @xmath8 threshold .
section 4 contains a discussion and some conclusions .
in principle , the four channels of interest @xmath0 , @xmath1 , @xmath29 and @xmath2 should be treated simultaneously . however , it is more convenient and transparent if the problem is analysed in terms of three separate @xmath22-matrices .
the first @xmath22-matrix is precisely the same as in ref .
@xcite , where only the @xmath0 and @xmath1 channels denoted by the indices @xmath30 , @xmath8 are explicit .
this can be written as @xmath31 where @xmath32 are the center - of - mass momenta of the two mesons in the two channels @xmath33 and the channel scattering lengths @xmath34 are expressed in terms of the @xmath3-matrix elements , via the solution of @xmath35 , as @xmath36 , @xmath37 @xmath38 at this stage the @xmath39 channel is incorporated as an `` optical model '' correction to the corresponding matrix element of @xmath40 and the @xmath2 channel is simply ignored since this @xmath22-matrix is used to describe only reactions b ) , c ) and d ) , where the effect of the @xmath2 channel is small being only an electromagnetic correction to these three reactions . as discussed in ref .
@xcite various features of the experimental data suggest that the @xmath3-matrix elements can be parametrized in terms of energy independent constants the background terms @xmath41 plus poles associated with the @xmath42-wave @xmath0 resonances @xmath9 and @xmath43 .
this results in @xmath44 , @xmath45 @xmath46 where @xmath47 the last terms on the rhs of eqs .
( [ ks ] ) represent the effect of the eliminated @xmath29 channel .
0.4 cm the second @xmath22-matrix involves only the two channels @xmath1 and @xmath2 denoted by the indices @xmath8,@xmath48 where now it is the @xmath29 and @xmath0 channels that are treated as optical potentials .
this @xmath22-matrix is written as @xmath49 @xmath50 here we are not interested in @xmath51 , since this would describe the @xmath52 reaction .
the forms of @xmath53 , @xmath54 , @xmath55 , @xmath56 and @xmath57 are the same as given above .
however , @xmath58 also we now need @xmath59 @xmath60 and @xmath61 where the last terms on the rhs represent the effect of the eliminated @xmath0- and @xmath39-channels .
also we need @xmath62 0.4 cm the third @xmath22-matrix involves only the two channels @xmath0 and @xmath2 denoted by the indices @xmath30,@xmath48 where now it is the @xmath1 and @xmath29 channels that are treated as optical potentials .
this @xmath22-matrix is written as @xmath63 @xmath64 as before , we are not interested in @xmath51 . the forms of @xmath65=@xmath66 , @xmath54 , @xmath55 , @xmath56 and @xmath57 are the same as given above .
however , @xmath67 also we now need @xmath68 @xmath69 where the last terms on the rhs represent the effect of the eliminated @xmath1- and @xmath39-channels .
also we need @xmath70 the definitions of all other parameters are the same as for @xmath71
. 0.4 cm
compared with ref .
@xcite there are now four new parameters @xmath72 , @xmath73 , @xmath74 and @xmath75 explicitly dependent on the index @xmath48 .
these four parameters replace the single free parameter @xmath76 that related @xmath77 and @xmath78 . in
all there are now 13 parameters that are determined by a minuit fit of upto 158 pieces of data 23 are @xmath0 amplitudes ( real and imaginary)@xcite , 11 are @xmath79 cross sections[@xmath80 @xcite and 53 are @xmath7 cross sections [ @xmath81@xcite .
in addition , from ref .
@xcite we use upto 48 @xmath82 amplitudes in the energy range @xmath83mev .
there are several reasons for choosing this upper limit : + a ) we wish to include the full effect of the @xmath9 .
+ b ) the @xmath6 and @xmath7 reactions are closely related and so attempting to fit them simultaneously over very different energy ranges could give misleading results .
therefore , we do not attempt to use the available data at higher energies .
+ c ) the values of the @xmath6 amplitudes are far from being unique as is clear when comparing the amplitudes of refs .
@xcite and @xcite .
in fact , in view of this lack of uniqueness we do not use the quoted errors of ref .
instead , we make two overall fits where , in the one case , all the errors in ref .
@xcite are increased to @xmath84 for both the real and imaginary components and , in the second case , the increase is only to @xmath85 .
these choices were made so that the resultant @xmath86/dpt for this reaction are comparable to those in the other reactions .
we realise that this procedure is throwing away information . however , the main aim in this work is to study the @xmath7 reaction with the @xmath6 playing only a secondary role as a possible stabilizing effect .
therefore , we want a @xmath3-matrix fit that is good for the well established reactions but , at the same time , also reproduces the qualitative trends in the @xmath6 reaction suggested by refs .
@xcite and @xcite .
in fact , we could even turn the argument around and say that our @xmath87 amplitudes are a _ prediction _ that is consistent with the other reactions . in practice ,
the actual @xmath8-production cross section data was used in a reduced form , from which threshold factors have been removed namely : @xmath88 in ref .
@xcite the last equation was replaced by @xmath89 where @xmath76 was treated as a free parameter in the minuit minimization . at first , because of the lack of uniqueness in the two analyses published in refs .
@xcite and @xcite , only the 32 @xmath82 amplitudes with @xmath90 mev were used , since this upper energy limit is about the same as for the @xmath7 data .
this resulted in a good fit with parameters qualitatively the same as in ref .
@xcite and also in line with the particle data group@xcite see columns a , d and pdg in table [ table1 ] . in column
a , the error bars in the @xmath82 amplitudes of ref .
@xcite have all been increased to @xmath84 for the reasons discussed earlier . in this case , the overall @xmath86/dof and the separate @xmath86/dpt are all near unity . however ,
when in column d the errors are increased to only @xmath85 , the @xmath86/dpt for the @xmath91 amplitudes become significantly larger .
columns b and c show the corresponding results when the @xmath82 data base is increased to include data with @xmath92 upto 1650mev .
the fits are now systematically worse than in column a with the overall @xmath86/dof increasing from 0.89 to 1.23 in column b. in column c the case with smaller errors and the larger data base the fit obtained was quite poor to such an extent that reasonable errors on the parameters could not be extracted .
the latter fit , when all 13 parameters were varied simultaneously , did not give from minuit a migrad result that converged to sensible parameters .
the fit displayed in column c is based on the parameters of column b , some of which are first fixed and then released and scanned by minuit .
the comparison with the data being fitted is shown in figs .
14 the main conclusions to be drawn from table [ table1 ] and these figures are : + 1 ) all four fits to the data are reasonable with cases a and b being superior .
+ 2 ) the main distinguishing feature between the four fits is the relative ability to fit the @xmath82 data , since this is the channel that contributes most to the overall @xmath86/dof with the @xmath86/dpt s from the other four channels being reasonably constant and comparable to unity in all fits .
this suggests that it will be hard for the present type of analysis to maintain these latter @xmath86/dpt s and , at the same time , achieve a good @xmath86/dpt for the @xmath82 data presented in refs .
@xcite and @xcite .
the authors , therefore , suggest that the @xmath93 amplitudes from the @xmath94 matrix model could be a more realistic set than those in refs .
@xcite and @xcite , since they are now consistent with more reactions @xmath4 , @xmath5 and @xmath7 .
+ 3 ) figure 3 shows that , beyond @xmath95mev , cases a and d give larger cross sections than b and c the difference increasing to about a factor of two by @xmath96mev . in the near future , the graal collaboration @xcite is expected to provide total cross section data upto this energy and so , hopefully , distinguish between these cases . in table
[ table1 ] the parameters @xmath97 , @xmath98 @xmath99 , @xmath100 and @xmath101 are quoted , whereas the earlier formalism is expressed in terms of @xmath102 , @xmath103 , and @xmath104 .
the two notations are related as follows : + 1 ) @xmath105 $ ] , 2 ) @xmath106 $ ] , + 3 ) @xmath107 $ ] , 4 ) @xmath108/q_{3}[e_0(r)]$ ] , + 5 ) @xmath109/q_{3}[e_1(r)]$ ] and 6 ) @xmath110\gamma_{\gamma}(0,1)$ ] . + this
now requires a choice to be made for the reference energies @xmath111 and @xmath112 , which preferably should be close to the @xmath113 in table [ table1 ] . here , we take simply @xmath114 1535 , 1650 mev respectively .
this gives @xmath115=0.945 , \ \
q_{\pi}[e_0(r)]=2.365 , \
\ q_{\pi}[e_1(r)]=2.770 , \
\ q_{3}[e_0(r)]=1.067 , \
\ q_{3}[e_1(r)]=1.245 , \ \
q_{\gamma}[e_{0}(r)]=2.436 $ ] and @xmath116=2.829 $ ] @xmath117 .
it should be added that this is not an assumption or an approximation .
it is just setting a scale that is needed when converting from one notation to the other . in table [ table2 ] , the @xmath118 are tabulated alongwith @xmath119 . in the above , we have been very explicit in describing the formalism .
therefore , in principle , the reader should be able to reconstruct all three @xmath22-matrices and so determine each of the complex amplitudes needed in the five processes @xmath4 , @xmath5 , @xmath120 , @xmath6 and @xmath7
. this formalism also enables these amplitudes to be calculated at unphysical energies . for example , in the study of possible @xmath8-nucleus quasi - bound states , the @xmath120 amplitudes are needed below the @xmath8 threshold .
this is easily achieved by simply using an @xmath8 momentum(@xmath121 ) that is purely imaginary . in spite of the model being very explicit , it is sometimes convenient to have simplified versions of some of the amplitudes .
the ones we consider are those that are expansions in terms of @xmath121 about the @xmath8 threshold in particular @xmath122 and @xmath123 . the former results in the usual @xmath120 effective range expansion of ref .
@xcite , the parameters of which are now updated in table [ table3 ] .
this shows that the scattering length(@xmath124 ) is larger than that extracted in ref .
@xcite the increase being 15% for case a and 40% for case b. however , it should be remembered that case b extrapolates the model into a region where the @xmath7 data is lacking , and it is just this reaction that is crucial in determining the scattering length .
given this expansion , then @xmath25 is readily calculated from @xmath122 as in eq .
[ kt1 ] at energies both above and below the @xmath8 threshold .
the other amplitude of interest is @xmath26 in eq .
[ kt2 ] , which is seen to depend on both @xmath122 and @xmath123 . by analogy with the expansion of @xmath25
, we express @xmath26 in the form @xmath125 the two entities @xmath126 and @xmath127 are then expanded as @xmath128 with the parameters @xmath129 being given in table [ table4 ] .
both of these expansions do very well over the energy range of ref .
for example , with case a at @xmath130mev an energy that is 50 mev above the @xmath8 threshold , the expansion of @xmath126 gives 8.4i20.0 @xmath117 compared with the exact value of 7.9i19.7 @xmath117 and the expansion of @xmath127 gives 37.12i5.4 compared with the exact value of 37.08i5.7 .
this latter agreement and the weak energy dependence of this quantity explains why , in ref .
@xcite , the replacement of @xmath131 by @xmath132 was a good approximation , since as seen from eq .
also the value of @xmath134 in table [ table1 ] is essentially given by @xmath135@xmath136 . in fig . 5 ,
the real and imaginary components of @xmath26 are shown , when these two expansions are used in eq.[intge ] , which is then inverted to give @xmath26 .
it is seen that they give a good representation over a wide energy range especially for energies below the @xmath8 threshold .
this agreement is very similar to that found in ref .
@xcite for @xmath25 .
it should be added that , if the form of @xmath26 written in eq .
[ kt2 ] is used directly with expansions of @xmath137 and @xmath122 , then the fit is much poorer as also seen in fig . 5 .
in this paper the authors have developed a simple @xmath3-matrix parametrization that gives , in an energy range of about 100 mev each side of the @xmath8 threshold , a good fit to @xmath4 , @xmath138 and @xmath7 data .
in addition , it has the same trends as the @xmath6 data , which at present is not unique over this energy range . however , this consistent fit should not be considered an end in itself , since it also results in predictions for the @xmath139 @xmath42-wave amplitude . near the @xmath8 threshold
this amplitude has been parametrized in the form of the effective range expansion
the resultant parameters being given in table [ table3 ] . since
this expansion is good over a wide energy range each side of the @xmath8 threshold , it is very useful for discussions concerning the possibility of @xmath8-nucleus quasi - bound states e.g. in ref . @xcite the effective range expansion of ref .
@xcite was used to study the production of @xmath8-nuclei , while ref .
@xcite uses such an expansion to describe @xmath8-nucleus final state interactions .
the indications from table [ table3 ] are that the @xmath140 scattering length is now larger than that extracted in ref .
this is even more favourable for the existence of @xmath8-nucleus quasi - bound states and may lead to an early onset of nuclear p - wave states , which are easier to detect in the darmstadt experiment outlined in ref .
@xcite . one result of the above fits is the extraction of the photon - nucleon-@xmath141(1535 ) coupling constant @xmath142 as indicated in table 1 , which is equivalent to the partial decay width @xmath143 for @xmath144 .
the definition of @xmath143 is not unique , however .
below , this question is elucidated on a simple soluble model of the @xmath22 matrix .
this model is also used to understand the interference of a resonant interaction described by a singularity in the @xmath3 matrix and potential interactions described by the background parameters @xmath145 .
let us , assume a separable @xmath3 matrix model with @xmath146 where , in the notation of eqs .
[ ks ] and [ ketagamma ] , @xmath147 .
this leads to a separable solution for the @xmath22-matrix @xmath148 } .\ ] ] when the background term @xmath145 vanishes , this model is equivalent to simple breit - wigner multichannel resonances of eigen - width @xmath149 .
however , when we relax this restriction a new structure is built upon the resonance .
it is determined by the energy dependent term @xmath150,$ ] which generates a zero of the cross section at @xmath151 .
now , it is the @xmath152 that sets a new energy scale , which may be independent of the scale given by the width . for a large @xmath145
one finds the resonance to be accompanied by a nearby zero , whereas for small @xmath153 this zero is moved away beyond the resonance width .
the resonance shape is thus very different from the lorentian : one reason being the strong energy dependence in @xmath154 and another being the pole - background interference .
as discussed above in connection with table [ table2 ] , it is usually natural to define the partial width of a resonance on the basis of eqs .
[ sepk ] and [ sept ] as @xmath155 . for the best fits to the data
( sets @xmath156 of table 1 ) this equation produces @xmath157 mev and @xmath158 mev respectively .
however , with complicated phenomenological @xmath22 matrices one could define @xmath159 otherwise , e.g. by moulding @xmath22 into the breit - wigner form in the proximity of @xmath160 .
thus , at @xmath161 one has @xmath162 .
inserting from table [ table2 ] the values of @xmath163 and @xmath164 gives another estimate of @xmath165 .
for example , with case @xmath166 , @xmath167 at @xmath168 mev giving @xmath169 and @xmath170 @xmath117 .
this results in @xmath171 mev . similarly , @xmath172 at @xmath173 mev giving @xmath174 and @xmath175 @xmath117 .
this results in @xmath176 mev .
the proximity of these three widths reflects the fact that the realistic situation is fairly close to the separability situation described by eqs .
[ sepk ] and [ sept ] .
it is found to hold approximately , for all the parameter sets in table 1 .
there is another , somewhat unexpected effect of the @xmath150 $ ] interference term in eq .
[ sept ] . for @xmath177
one finds that the amplitudes below the resonance are enhanced , and the amplitudes above the resonance are reduced with respect to the pure resonance term .
this effect is seen clearly for the @xmath145 and @xmath178 parameter sets , where below the resonance the @xmath179 parameter in the @xmath16 channel is the largest and the real parts of the @xmath16 scattering lengths given in table [ table3 ] are also the largest . on the other hand the @xmath180 production cross section , dominated by the final state @xmath16 interactions , become the smallest above the resonance as seen fig . 4 .
one consequence of this effect is that an extension of the @xmath180 cross section measurements to energies above the @xmath9 resonance may by instrumental in fixing more precisely the @xmath16 scattering length . as indicated in the introduction ,
the real part of this scattering length is crucial in the determination of quasibound states in @xmath8few nucleon systems . on the experimental side
there are several groups @xcite,@xcite studying the @xmath7 reaction in or near this interesting energy range .
the observation of the cross section near and above @xmath92=1540mev would be of great interest enabling a detailed study to be made of the @xmath9 and possibly leading to a better understanding of the internal structure of this object . at present
there is no definite conclusion as to whether or not this resonance structure is due to a pole in the @xmath3-matrix , as advocated here , or arising through coupling to high lying closed channels - see ref .
@xcite . in the near future , the authors of ref . @xcite are expected to extract , directly from experiment , separate values for the real and imaginary components of @xmath181
. these will be analogous to the @xmath10 data already available in ref .
@xcite and used in the above fits .
such a development will then enable the present type of @xmath3-matrix analysis to be even more constrained .
one of the authors ( s.w . )
wishes to acknowledge the hospitality of the research institute for theoretical physics , helsinki , where part of this work was carried out .
in addition he was partially supported by grant no kbn 2p03b 016 15 .
the authors also thank drs .
r. arndt and b. krusche for useful correspondence and members of the graal collaboration for useful discussions .
this line of research involving @xmath8-mesons is partially supported by the academy of finland .
a. m. green and s. wycech , phys .
c*55 * , r2167 ( 1997 ) m.l . goldberger and k.m .
watson , _ collision theory _
( wiley , new york london , 1964 ) r.arndt , j.m.ford and l.d.roper , phys.rev . d*32 * , 1085 ( 1985 ) .
. the optimised parameters from minuit defining the @xmath3-matrices : there are in column a only 32 @xmath6 data points with @xmath185 mev and error bars all 1.41 , in column b 48 data points with @xmath186 mev and error bars all 1.41 . cases d and c are the same as a and b except that the error bars are reduced to 0.70 .
in addition , the first column shows the results from ref .
@xcite and the last column the corresponding values from the particle data group @xcite .
[ cols="^,^,^,^,^,^,^",options="header " , ] figure 5 .
the a ) real and b ) imaginary parts of @xmath26 .
the solid curve is the exact value as given by the model .
the dotted curve uses expansions of @xmath126 and @xmath127 , whereas the dashed curve uses those for @xmath123 and @xmath187 . | the coupled @xmath0 , @xmath1 , @xmath2 systems are described by a @xmath3-matrix method . the parameters in this model
are adjusted to get an optimal fit to @xmath4 , @xmath5 , @xmath6 and @xmath7 data in an energy range of about 100 mev each side of the @xmath8 threshold .
+ the coupling of photons to the @xmath9 state is extracted and also an alternative to the current @xmath10 amplitudes suggested .
expansions are given for the @xmath11 and @xmath12 amplitudes in terms of the @xmath8 momentum .
effects of interference of this state with background potential interactions are discussed and experimental consequences are indicated .
epsf 2ex = hip 1999 21 / th + a.m. green@xmath13 , and s. wycech@xmath13 + @xmath14department of physics and helsinki institute of physics , p.o .
box 9 , + fin00014 university of helsinki , finland + @xmath15soltan institute for nuclear studies , warsaw , poland pacs numbers : 13.75.-n , 25.80.-e , 25.40.v |
the aim of this paper is to describe in an explicit way the spectral theory of the adjacency operator on an infinite graph @xmath0 which , outside of a finite sub - graph @xmath1 , looks like a regular tree @xmath2 of degree @xmath3 .
we mainly adapt the case of the schrdinger operators as presented in @xcite .
the proofs are often simpler here and the main results are similar .
this paper can be read as an introduction to the scattering theory for differential operators on smooth manifolds . even if we do not find our results in the literature ,
there is probably nothing really new for experts in the scattering theory of schrdinger operators , except the combinatorial part in section [ sec : combi ] .
the main result is an explicit spectral decomposition : the hilbert space @xmath4 splits into a sum of two invariant subspaces @xmath5 .
the first one is an absolutely continuous part isomorphic to a closed sub - space of that of the regular tree of degree @xmath3 , while the second one is finite dimensional and we have an upper bound on its dimension .
the absolutely continuous part of the spectral decomposition is given in terms of explicit generalized eigenfunctions whose behavior at infinity is described in terms of a scattering matrix . we first introduce the setup , then we recall the spectral decomposition of the adjacency operator @xmath6 of a regular tree @xmath2 by using the fourier - helgason transform . in section 3
, we consider a schrdinger operator @xmath7 on @xmath2 , where @xmath8 is a compactly supported non local potential .
we build the generalized eigenfunctions for @xmath9 , define a deformed fourier - helgason transform and get a spectral decomposition of @xmath9 ( theorem [ fhtil ] ) . in section 4 , we derive a similar spectral decomposition of the adjacency operator of any graph @xmath0 asymptotic to a regular tree @xmath2 by proving the following combinatorial result ( theorem 4.2 ) : any such graph @xmath0 is isomorphic to a connected component of a graph @xmath10 which is obtained from @xmath2 by a finite number of modifications .
this implies that the adjacency operator of @xmath10 is a finite rank perturbation of the adjacency operator of @xmath2 . in section 5 ,
we investigate some consequences of the scattering theory developed in section 3 : we write the point - to - point correlations of scattered waves in terms of the green s function , we define the transmission coefficients , connect them to the scattering matrix , and get an explicit expression of them in terms of a dirichlet - to - neumann operator . for the sake of clarity , this part has been postponed , since it is not necessary to prove theorem 4.2 .
let us consider a connected graph @xmath11 with @xmath12 the set of _ vertices _ and @xmath13 the set of _
edges_. we write @xmath14 for @xmath15 .
[ astl ] let @xmath16 be a fixed integer .
we say that the infinite connected graph @xmath0 is asymptotic to a regular tree of degree @xmath3 if there exists a finite sub - graph @xmath1 of @xmath0 such that @xmath17 is a disjoint union of a finite number of trees @xmath18 rooted at a vertex @xmath19 linked to @xmath1 and so that all vertices of @xmath20 different from @xmath19 are of degree @xmath3 .
the trees @xmath18 are called the ends of @xmath0 .
equivalently , @xmath21 is infinite , has a finite number of cycles and a maximal sub - tree of @xmath21 has all vertices of degree @xmath3 except a finite number of them .
[ bdry ] we define the edge boundary @xmath22 of @xmath1 as the set of edges of @xmath21 connecting a vertex of @xmath1 to a vertex of @xmath23 , namely one of the @xmath19 s .
we denote by @xmath24 the combinatorial distance of @xmath25 to @xmath1 . in particular , for @xmath26 , @xmath27 .
the space of complex - valued functions on @xmath12 is denoted @xmath28 and @xmath29 is the subspace of functions with finite support .
we define also @xmath30 it is a hilbert space when equipped with the inner product : @xmath31 let us emphasize that we take the physicist s notation , as in @xcite for example : our inner product is conjugate - linear in the first vector and linear in the second . on @xmath32 , we define the adjacency operator @xmath33 by the formula : @xmath34 the operator @xmath33 is bounded on @xmath35 if and only if the degree of the vertices of @xmath21 is bounded , which is the case here . in that case , the operator @xmath33 is self - adjoint ; otherwise , the operator @xmath33 defined on @xmath36 could have several self - adjoint extensions .
for any @xmath37 outside the spectrum of @xmath38 , we denote by @xmath39 the resolvent @xmath40 and by @xmath41 with @xmath42 the matrix of @xmath43 , also called the _ green s function_.
let @xmath44 be the regular tree of degree @xmath3 and let us choose an origin , also called a root , @xmath45 .
we denote by @xmath46 the combinatorial distance of the vertex @xmath47 to the root .
the set of points at infinity denoted @xmath48 is the set of infinite simple paths starting from @xmath45 .
we will say that a sequence @xmath49 tends to @xmath50 if , for @xmath51 large enough , @xmath52 belongs to the path @xmath53 and is going to infinity along that path .
if @xmath47 is another vertex of @xmath54 , the sets @xmath55 and @xmath56 are canonically identified by considering paths which concide far from @xmath45 and @xmath47 .
there is a canonical probability measure @xmath57 on @xmath58 : @xmath59 is the unique probability measure on @xmath55 which is invariant by the automorphisms of @xmath2 leaving @xmath45 fixed . later on
we will always denote by @xmath60 the set of points at infinity , because the root is fixed . for the tree @xmath2 , the _ busemann function _
@xmath61 associated to the point @xmath62 is defined as follows : let us denote by @xmath63 the last point lying on @xmath64 in the geodesic path joining @xmath45 to @xmath47 , ( take @xmath65 in the case where @xmath45 belongs to the geodesic from @xmath47 to @xmath64 ) , and let us set @xmath66 .
the level sets of @xmath67 are the _
horocycles _ associated to @xmath68 .
we notice that the function @xmath69 increases by one for one of @xmath47 s neighbors , namely the one of the ray from @xmath47 to @xmath64 , and decreases for the others .
thus the function @xmath69 goes to @xmath70 as @xmath47 tends to @xmath68 .
as @xmath47 tends to @xmath71 , the function @xmath69 tends to @xmath72 , whereas the quantity @xmath73 remains bounded , since it tends to @xmath74 .
let us define the riemann surface @xmath75 with @xmath76 .
we denote by @xmath77 the circle @xmath78 , and we set @xmath79 $ ] . for any @xmath80 , we set : @xmath81 the map @xmath82 @xmath83 is holomorphic from @xmath84 to @xmath85 .
it maps bijectively the physical sheet @xmath86 onto @xmath87 . by this map
the circle @xmath88 is a double covering of @xmath89 .
if @xmath90 is a subset of @xmath85 , we denote by @xmath91 the _ pre - image _ of @xmath90 by the map @xmath92 , i.e. @xmath91 is the subset of @xmath84 defined by @xmath93 .
the results of this section are classical , see for example the paper by p. cartier @xcite .
we denote by @xmath6 ( resp .
@xmath94 ) the adjacency operator ( resp .
the green s function ) on @xmath2 .
we will compute explicitly @xmath95 .
let us recall that the regular tree is 2-point regular : for any @xmath96 so that @xmath97 , there exists an automorphism @xmath90 of @xmath2 so that @xmath98 and @xmath99 .
the green s function @xmath100 satisfies @xmath101 for any automorphism @xmath90 of @xmath2 .
hence , @xmath100 is a function of the distance @xmath102 .
it is therefore enough to compute @xmath103 for an @xmath104 , that is the value @xmath105 of the @xmath106 solution of @xmath107 where @xmath105 depends only on the distance @xmath46 to the origin @xmath45 .
so we set @xmath108 if @xmath109 , @xmath110 , and rewrite equation ( [ resp ] ) as follows : \i ) @xmath111 for@xmath112
\ii ) @xmath113 \iii ) @xmath114 the last condition stands for @xmath115 to be in @xmath116 . *
if @xmath117 , the equation @xmath118 admits an unique solution @xmath119 such that @xmath120 . from i ) and iii ) , we get that @xmath121 and the constant @xmath122 is determined by ii ) : @xmath123 therefore we have @xmath124 where @xmath125 denotes the determination of @xmath126 in @xmath127 equivalent to @xmath128 as @xmath128 tends to infinity . thus using the invariance of the green s function by the group of automorphisms of the tree , we see that the green s function @xmath129 is a function of the distance @xmath102 and we have , for any @xmath130 , @xmath131 the operator of matrix @xmath132 is clearly bounded in @xmath116 and @xmath37 is not in the spectrum of @xmath6 . *
if @xmath133 , there is no @xmath106 solution of equation ( [ resp ] ) .
therefore we can not solve @xmath134 in @xmath106 , the resolvent does not exist and @xmath37 is in the spectrum of @xmath6 . using the parameter @xmath135 , we have @xmath136 [ greenarbre ] the spectrum of @xmath6 is the interval @xmath137 $ ] .
the green s function of the tree @xmath2 is given , for @xmath138 by @xmath139 as a function of @xmath140 , the green s function extends meromorphically to @xmath84 with two poles @xmath141 and @xmath142 .
moreover we have , for any @xmath104 and any @xmath143 belonging to the ray from @xmath144 to @xmath64 , @xmath145 with @xmath146 the last result comes from the definition @xmath147 . let us recall how to introduce a notion of spectral measure ( also called density of states ) on the graph @xmath0 .
for a given continuous function @xmath148 , we associate by the functional calculus an operator @xmath149 on @xmath4 , which has a matrix @xmath150(x , x')$ ] .
we consider then , for any @xmath151 , the linear form on @xmath152 @xmath153(x , x ) \ .\ ] ] @xmath154 is positive and satisfies @xmath155 , so we have @xmath156 where @xmath157 is a probability measure on @xmath158 , supported by the spectrum of @xmath33 which is called the spectral measure of @xmath0 at the vertex @xmath47 .
the density of states of @xmath2 is given by the [ cdv](see for example @xcite ) the spectral measure @xmath157 of @xmath2 is independent of the vertex @xmath47 and is given by @xmath159 for the sake of clarity , we recall the main ingredients : \1 ) an explicit computation of the diagonal entries of the green s function @xmath160 where @xmath125 denotes as previously the determination of @xmath126 in @xmath161 ( with @xmath162 $ ] ) equivalent to @xmath128 for great values of @xmath128 .
\2 ) the expression of the spectral measure via stone formula @xmath163 the previous density of states is the weak limit for the densities of a graph asymptotic to a regular tree .
more precisely we have [ dec1 ] let @xmath0 be as in definition [ astl ] . consider the adjacency operator @xmath33 defined by ( [ adja ] ) , denote @xmath164 for simplicity .
when @xmath47 tends to infinity , the densities of states @xmath165 of @xmath0 converge weakly to the density of states @xmath166 of @xmath2 defined by ( [ msa ] ) .
it is enough to compute the limits of @xmath167 for @xmath51 fixed and @xmath168 . by definition
, we have @xmath169(x , x)$ ] , and @xmath170(x , x)=\sum a_{x , x_1}a_{x_1,x_2}\cdots a_{x_{n-1},x } ~\ ] ] where the sum is on loops @xmath171 of length @xmath51 based at @xmath47 .
if we assume that @xmath172 , the loops do not meet @xmath1 and therefore @xmath173(x , x)=[a_0^n](x , x)$ ] .
let us recall the definition of the fourier - helgason transform on the tree @xmath2 with the root @xmath45 .
[ fht1 ] for any @xmath174 , the fourier - helgason transform @xmath175 is the function defined by the finite sum @xmath176 for any @xmath177 and any @xmath80 . [ ipw ] for any @xmath177 and any @xmath178 we define the `` incoming plane wave '' @xmath179 as the function @xmath180 , where @xmath181 for @xmath182 , such a plane wave is a generalized eigenfunction for the adjacency operator @xmath6 on @xmath183 in the sense that it satisfies @xmath184 but is not in @xmath106 .
if we restrict ourselves to @xmath185 , definition [ fht1 ] writes @xmath186 and the completeness of the set @xmath187 is expressed by the following inversion formula ( see [ cms ] ) : [ fhti ] for any @xmath188 , the following inverse transform holds @xmath189 where @xmath190 moreover the fourier - helgason transform extends to a unitary map from @xmath116 into @xmath191 .
the fourier - helgason transform is not surjective : its range is the subspace @xmath192 of the functions @xmath193 of @xmath191 which satisfy the symmetry condition ( see , for example , @xcite or @xcite ) @xmath194 the fourier - helgason transform provides a spectral resolution of @xmath6 : if @xmath195 is continuous , @xmath196 where @xmath197 denotes the operator of multiplication by that function on @xmath198 . from the inverse fourier - helgason transform formula ( [ fit ] )
we find back the expression of the spectral measure of @xmath199 ( see theorem [ cdv ] ) . by homogeneity of the tree @xmath2 , for any continuous function @xmath200 , @xmath201(x , x)$ ] is independent of @xmath47 . using ( [ fit ] ) , we get @xmath202(o , o ) = \int_{\go}\int_{s^0 } \phi ( \gl_s ) \overline{e_0(o,\go , s ) } e_0(o,\go , s)\ d\gs_o(\go ) d\mu(s ) \ = \int_{s^0 } \phi ( \gl_s ) \
d\mu(s ) .\ ] ] let us perform the change of variables @xmath203 using ( [ chgtit ] ) and the fact that , by the map @xmath204 , the circle @xmath88 is a double covering of @xmath89 , we write @xmath202(x , x ) \ = 2 \frac{(q+1)\log q}{\pi}\int_{i_q } \phi ( \gl ) \frac { 1- \gl^2/4q } { q+q^{-1 } + 2 -\gl^2/q}f'_q(\gl ) d\gl\ ] ] @xmath205 which actually implies formula ( [ msa ] ) .
we are concerned here with the scattering on @xmath2 between the adjacency operator @xmath206 and the schrdinger operator @xmath207 , where @xmath8 is a compactly supported non local potential .
more precisely the hermitian matrix ( also denoted @xmath8 ) associated to this potential is supported by @xmath208 where @xmath209 is a finite part of @xmath54 .
we assume in what follows that @xmath209 is chosen minimal , so that : @xmath210 let us first describe the spectral theory of @xmath9 : it follows from @xcite , sec .
xi 3 , and from the fact that @xmath9 is a finite rank perturbation of @xmath6 ( see also section [ ss : ppsp ] ) that the hilbert space @xmath116 admits an orthogonal decomposition into two subspaces invariant by @xmath9 : @xmath211 where * @xmath212 is the isometric image of @xmath116 by the _ wave operator _
+ @xmath213 we have @xmath214 , so that the corresponding part of the spectral decomposition is isomorphic to that of @xmath6 which is an absolutely continuous spectrum on the interval @xmath215 . *
the space @xmath216 is finite dimensional , admits an orthonormal basis of @xmath106 eigenfunctions associated to a finite set of eigenvalues , some of them may be embedded in the continuous spectrum @xmath89
. we will denote by @xmath217 and @xmath218 the orthogonal projections on both subspaces . in order to make the spectral decomposition more explicit
, we will introduce suitable generalized eigenfunctions of @xmath9 .
these generalized eigenfunctions are particular solutions of @xmath219 meaning not @xmath106 solutions , but only point - wise solutions . for
the adjacency operator @xmath6 , we have seen that these generalized eigenfunctions , called the `` plane waves '' are given by the @xmath220 s with @xmath221 and @xmath62 ( see definition [ ipw ] ) and give the fourier - helgason transform which is the spectral decomposition of @xmath6 ( theorem [ fhti ] ) .
we are going to prove a similar eigenfunction expansion theorem for @xmath9 , using generalized eigenfunctions of @xmath9 .
we will mainly adapt the presentation of @xcite , sec .
xi.6 , for schrdinger operators in @xmath222 ( see also @xcite ) .
our first goal is to build the generalized eigenfunctions @xmath223 also denoted @xmath224
. we will derive and solve the so - called _ lippmann - schwinger equation_. this is an integral equation that @xmath225 will satisfy .
let us proceed first in a _ formal way _ by transferring the functions @xmath220 by the wave operator : if @xmath226 is the image of @xmath220 by the wave operator @xmath227 in some sense ( they are not in @xmath106 ! ) , then we should have @xmath228 + @xmath229 $ ] + @xmath230 + @xmath231\ .$ ] + so @xmath226 should obey the following `` lippmann - schwinger - type '' equation @xmath232 let @xmath233 be a compactly supported real - valued function so that @xmath234 . for example
@xmath235 can be the characteristic function of @xmath209 .
we first introduce a modified `` lippmann - schwinger - type '' equation .
if @xmath226 obeys ( [ foitt ] ) and @xmath236 , then @xmath237 obeys @xmath238 we have the following result : [ indep ] let @xmath233 be a compactly supported real - valued function so that @xmath234 . set @xmath239 1 .
the set @xmath240 is finite and independent of the choice of @xmath235 .
2 . if @xmath241 , then ( [ mlse ] ) has a unique solution @xmath242 and the function @xmath243 is the unique solution of the lippmann - schwinger equation ( [ foitt ] ) .
the set @xmath240 is invariant by @xmath244 and consequently it is the pre - image by @xmath245 of a subset of @xmath89 which we denote by @xmath246 . we first prove 2 ) .
let @xmath247 be the finite rank operator on @xmath116 defined by @xmath248 .
the map @xmath249 extends holomorphically to @xmath250 .
equation ( [ mlse ] ) takes the form @xmath251 where @xmath252 . by the analytic fredholm theorem ( @xcite , p. 101 )
, there exists a finite subset @xmath240 of @xmath253 , defined by @xmath254 , so that equation ( [ mlse ] ) has a unique solution @xmath242 whenever @xmath255 the second assertion of 2 ) comes from the fact that @xmath256 .
let us now prove 1 ) : the `` minimal '' @xmath235 is @xmath257 . if @xmath237 is a non trivial solution of @xmath258 , and @xmath259 , @xmath237 is also solution of @xmath260 .
conversely , if @xmath260 , we have @xmath261 .
we have to prove that @xmath262 .
if @xmath263 , we would have @xmath264 and @xmath265 . to prove 3 ) it is enough to notice that for any @xmath266 , we have @xmath267 .
[ compact ] if @xmath268 with @xmath133 and @xmath269 , then @xmath270 where @xmath271 is the smallest subset of @xmath54 so that @xmath272 and all connected components of @xmath273 are infinite .
we will proceed by contradiction .
let @xmath274 be so that @xmath275 .
let us define an infinite sub - tree @xmath276 of @xmath277 as follows : let @xmath278 be the vertices of @xmath2 which satisfy @xmath279 and @xmath280 is closer to @xmath271 than @xmath47 .
then @xmath276 is the connected component of @xmath47 in the graph obtained from @xmath183 by removing the edges @xmath281 for @xmath282 .
let us consider the `` averaged '' function @xmath283 then @xmath284 satisfies the ordinary difference equation @xmath285 .
we thus get a contradiction , since this equation has no non - zero @xmath106 solution when @xmath37 is in @xmath89 and hence @xmath286 .
@xmath287 .
this holds because any eigenfunction associated to an eigenvalue in @xmath288 is supported in @xmath271 and the dimension of the vector space of functions supported in @xmath289 is @xmath290 .
if @xmath185 , @xmath291 and @xmath292 , then @xmath293 .
conversely , if @xmath294 , there exists @xmath295 so that @xmath291 and @xmath296 .
due to proposition [ compact ] , the support of such an @xmath115 is included in @xmath271 and @xmath297 .
we can apply @xmath298 to both sides of the equation , ( although @xmath299 is in the spectrum of @xmath6 ) because the functions on both sides are finitely supported . on the lefthandside we have @xmath300
: this is true for @xmath301 because @xmath89 is the spectrum of @xmath6 and hence by continuity ( @xmath298 extends holomorphically near @xmath88 ) for every @xmath302 since @xmath115 is compactly supported . hence applying @xmath298 to both sides yields @xmath303 . due to proposition [ indep ]
, we can choose for @xmath235 the characteristic function of @xmath271 , so we get @xmath304 .
we have a non trivial solution of @xmath305 , namely @xmath306 .
conversely , let us start from @xmath237 , a non trivial solution of @xmath305 and define @xmath307
. then @xmath308 and @xmath309 by analytic extension from @xmath310 .
hence , using @xmath311 , @xmath312 from the definition of @xmath115 , we get that @xmath115 is a finite linear combination of the functions @xmath313 and we can use equation ( [ rad ] ) to get the bound in @xmath47 .
the pure point spectrum @xmath314 of @xmath9 splits into 3 parts @xmath315 where @xmath316-\infty , -2\sqrt{q}[$ ] , @xmath3172\sqrt{q } , + \infty [ $ ] , and @xmath318 .
we have @xmath319 and @xmath320 .
the first estimate comes from the mini - max principle and the fact that @xmath8 is a rank @xmath321 perturbation of @xmath6 with @xmath322 . the second one is already proved .
the reader could ask if there can really be some compactly supported eigenfunctions .
they can exist as shown by the following 2 examples .
@xmath323 is a tree with root @xmath45 and @xmath324 for any @xmath325 .
all other entries of @xmath8 vanish .
then if @xmath326 , @xmath327 , we have @xmath328 .
the graph @xmath323 is the union of a cycle with 4 vertices @xmath329 and a tree whose root is attached to 2 neighboring vertices of the cycle . if @xmath330 on the cycle and @xmath331 on all other vertices , @xmath332 .
however the proof of the following result is left to the reader : if @xmath323 is an infinite tree , then @xmath333 has no compactly supported eigenfunction .
[ fht2 ] we define the deformed fourier - helgason transform @xmath334 of @xmath335 as the function @xmath336 on @xmath337 defined by @xmath338 we want to prove the following [ fhtil ] for any @xmath188 and any closed interval @xmath339 , if we denote by @xmath340 the inverse image of @xmath90 by @xmath83 , the following inverse transform holds @xmath341 moreover , @xmath342 extends to an isometry from @xmath343 onto @xmath344 . denoting , with a slight abuse of notation , for @xmath138 , by @xmath345 the operator @xmath346 and similarly by @xmath347 the operator @xmath348
, we have the resolvent equation @xmath349 for @xmath350 and @xmath140 in @xmath351 , we set @xmath352 where the right hand side is a convergent series which identifies to @xmath353-times the inverse fourier - helgason transform of @xmath354 . using the definition of @xmath94 and ( [ equ : fhti ] ) we have @xmath355= e_0(\go , \gs)\ , \ ] ] so equation ( [ resol ] ) for @xmath356
gives an integral equation for @xmath357 @xmath358 and , if @xmath359 , @xmath360 the key fact is the relation between ( [ mlsei ] ) and the modified `` lippmann - schwinger - type '' equation ( [ mlse ] ) .
if @xmath135 is fixed and @xmath361 , then the equation for @xmath362 is identical to equation ( [ mlse ] ) for @xmath363 this can be used to prove [ endd ] let us consider @xmath364 , @xmath365 and @xmath138 .
then the function @xmath366 has a holomorphic extension in @xmath367 to @xmath351 and @xmath368 we thus have related @xmath336 to the resolvent .
let @xmath369 with @xmath370 and @xmath371 , and @xmath138 ( this implies @xmath372 ) .
up to a factor of @xmath373 , @xmath374 is the inverse fourier - helgason transform of @xmath375 ; so the plancherel theorem implies ( after multiplying by @xmath376 ) that @xmath377 @xmath378 if we sum over all @xmath47 s and @xmath143 s , we obtain for the left - hand side @xmath379f \rangle\ ] ] whereas the right - hand side becomes @xmath380 we thus conclude that , for any closed sub - interval @xmath90 of @xmath89 disjoint from @xmath381 , @xmath382f \rangle d\gl = \frac { 1}{\pi } \int_{j } d\gl \int_{s^0}\int_{\go}\frac { \ge } { |\gl_{\gs}-\gl|^2 + \ge^2}| \gf(s ,
\go,\gs ) |^2d\gs_o(\go ) d\mu(\gs)\ ] ] as @xmath383 , stone s formula implies that the left - hand side approaches @xmath384 .
moreover the measures @xmath385}\ ] ] converge weakly to @xmath386 as @xmath387 .
so one has to put @xmath388 , which implies @xmath389 .
thus one gets that the right - hand side tends to latexmath:[$\int_{\hat{j}}\int_{\go } @xmath91 is the inverse image of @xmath90 by @xmath83 .
we are concerned here with the spectral theory of the adjacency matrix of a graph @xmath0 asymptotic to a regular tree of degree @xmath3 , in the sense of definition [ astl ] , which we recall here : let @xmath16 be a fixed integer .
we say that the infinite graph @xmath0 is asymptotic to a regular tree of degree @xmath3 if @xmath21 is connected and there exists a finite connected sub - graph @xmath1 of @xmath0 such that @xmath17 is a disjoint union of a finite number of trees @xmath18 rooted at a vertex @xmath19 linked to @xmath1 and so that all vertices of @xmath20 different from @xmath19 are of degree @xmath3 .
the trees @xmath18 are called the ends of @xmath0 .
we want to reduce the spectral theory of @xmath38 to the situation studied in section [ sec : schro ] . for that , we need a preliminary combinatorial study which could be of independent interest . we need the following combinatorial result : [ theo : comb ] if @xmath21 is asymptotic to a regular tree of degree @xmath3 , then @xmath21 is isomorphic to a connected component of a graph @xmath391 which can be obtained from @xmath392 by adding and removing a finite number of edges . by removing a finite number of edges , one could assume that @xmath21 is a tree .
then the result is quite elementary if the degree of all vertices of @xmath0 is @xmath393 : it is then enough to add infinite regular trees to the vertices of degrees @xmath394 in order to get the final result .
this argument , suggested by the referee , is not enough to give a complete proof . in order to prove theorem [ theo : comb ] , we first introduce an integer @xmath395 associated to the graph @xmath21 ; the integer @xmath396 is a combinatorial analogue of the regularized total curvature of a riemannian surface @xmath84 which is of constant curvature @xmath397 near infinity , namely @xmath398 .
if @xmath21 is asymptotic to a regular tree of degree @xmath3 , we define @xmath399 by @xmath400 where @xmath401 is the degree of the vertex @xmath47 and @xmath402 is the first betti number of @xmath0 or equivalently the number of edges to be removed from @xmath21 in order to get a tree .
note that , if @xmath403 is a maximal sub - tree of @xmath21 , @xmath404 .
we will need the [ lemm : nu ] if , for @xmath405 , @xmath406 , then we have @xmath407 where @xmath408 is the number of inner vertices of @xmath409 and @xmath410 the number of boundary vertices ( i.e. connected to a vertex of @xmath411 ) of @xmath409 .
each of the @xmath410 boundary vertices has @xmath412 neighbors in @xmath411 and one in @xmath409 . from euler formula
applied to the sub - graph @xmath413 which is connected by the assumption on @xmath1 , we get @xmath414 thus @xmath415 is equal to @xmath416 we will also need
the : [ lemm : counting ] let @xmath417 be a finite tree whose all vertices are of degree @xmath3 except the ends which are of degree @xmath418 .
let @xmath410 be the number of ends and @xmath419 the number of inner vertices of @xmath193 .
we have the relation @xmath420 conversely , for each choice of @xmath421 satisfying equation ( [ equ : euler ] ) , there exists such a tree @xmath193 . from euler formula
applied to @xmath193 , we get @xmath422 . moreover @xmath423 .
let us choose a root inside @xmath193 and orient the edges from that root .
then we count the edges by partitioning them with their @xmath419 possible origins ; this gives @xmath424 .
conversely , the statement is true for @xmath425 and we proceed by induction on @xmath419 by adding @xmath412 edges to a boundary vertex and the corresponding @xmath412 boundary vertices , we have @xmath426 .
[ lemm : nu=0 ] if @xmath21 is asymptotic to a regular tree of degree @xmath3 , @xmath21 can be obtained from a tree @xmath2 by removing and adding a finite number of edges if and only if @xmath427 .
all the changes will take place inside the sub - graph @xmath409 .
if we denote by @xmath410 the number of boundary vertices and @xmath419 the number of inner vertices of @xmath409 , we have , using @xmath427 and lemma [ lemm : nu ] , @xmath428 .
we replace the graph @xmath409 by a tree @xmath193 whose existence is stated in lemma [ lemm : counting ] .
the vertices of both graphs are the same and all vertices of the new graph have degree @xmath3 .
hence the new graph is a regular tree @xmath2 .
we will now make some modifications of @xmath21 in order to get a new graph @xmath429 with @xmath430 .
[ lemm : moves ] if @xmath431 is defined by adding to @xmath21 a vertex and an edge connecting that vertex to a vertex of @xmath432 , then @xmath433 .
if @xmath434 is defined by adding to @xmath21 a tree whose root @xmath47 is of degree @xmath412 and all other vertices of degree @xmath3 and connecting @xmath47 by an edge to a vertex of @xmath435 , @xmath436 is asymptotic to a regular tree of degree @xmath3 and @xmath437 .
this lemma is quite easy to check .
_ proof of theorem [ theo : comb]. _ let us now write @xmath438 with @xmath439 and @xmath440 . by performing @xmath441 times the move @xmath442 and @xmath443 times the move @xmath444
, we arrive to a graph @xmath445 with @xmath446 .
let @xmath447 be the graph obtained by removing from @xmath445 the @xmath448 edges not in @xmath449 , one of whose vertices is in @xmath450 .
the graph @xmath429 is clearly asymptotic to a regular tree of degree @xmath451 and @xmath21 is a connected component of @xmath447 .
it remains to prove that , by removing and adding a finite number of edges to @xmath447 , we get a tree @xmath2 : this is the content of lemma [ lemm : nu=0 ] . from theorem [ theo :
comb ] , we can identify the set of vertices of @xmath21 to a subset of the set of vertices of @xmath447 which is the same as the set of vertices of @xmath2 .
we deduce the existence of a hilbert space @xmath452 so that @xmath453 and this decomposition is invariant by @xmath454 .
moreover @xmath454 is a finite rank perturbation of @xmath455 .
this will allow us to describe the spectral theory of @xmath38 by using the results of section [ sec : schro ] . in order to get the spectral decomposition of @xmath38 in terms of the spectral decomposition of @xmath454 given in section [ ss : deffh ] , we will need the [ lemm : support ] let @xmath456 with @xmath457 and @xmath458 finite .
let @xmath21 be an unbounded connected component of @xmath459 and @xmath68 a point at infinity of @xmath21 .
then , for any @xmath266 , we have @xmath460 conversely , if @xmath461 is a point at infinity of @xmath429 which is not a point at infinity of @xmath21 then @xmath462 we will apply the general theorem [ fhtil ] in our combinatorial context .
let us prove the first assertion , the proof of the second is similar .
it is enough to prove it for @xmath463 close to @xmath464 and hence @xmath299 not in the spectrum of @xmath6 , because @xmath465 is meromorphic on @xmath84 .
we have then ( equation ( [ foitt ] ) ) @xmath466 from the explicit expression of @xmath467 ( see definition [ ipw ] ) , we get that the first term belongs to @xmath468 , and so does the second one , as the image of a compactly supported function by the resolvent for @xmath469 ( recall that the resolvent is continuous in @xmath106 ) .
this proves that the restriction of @xmath470 to @xmath471 is an @xmath106 eigenfunction , with eigenvalue @xmath472 , of @xmath454 .
since @xmath454 has no eigenvalue @xmath472 for @xmath473 close to @xmath464 , it follows that @xmath474 vanishes for @xmath138 close to @xmath464 and @xmath475 .
theorem [ theo : comb ] allows to consider the set @xmath60 of points at infinity of @xmath323 as a subset of the set @xmath476 of the points at infinity of @xmath477 .
the space @xmath478 splits as a direct sum @xmath479 which is preserved by the adjacency matrix .
lemma [ lemm : support ] shows that the support of the generalized eigenfunctions @xmath480 for @xmath481 is included in @xmath482 . using this
, we can state the spectral decomposition of @xmath333 as an immediate corollary of theorem [ fhtil ] .
the hilbert space @xmath483 splits into a finite dimensional part @xmath484 and an absolutely continuous part @xmath485 .
this decomposition is preserved by @xmath333 .
if @xmath486 and , for @xmath481 , @xmath487 , then the map @xmath488 extends to an isometry from @xmath485 onto @xmath489 which intertwines the action of @xmath490 with the multiplication by @xmath472 .
we are again concerned here with the scattering theory on @xmath2 between the adjacency operator @xmath206 and the schrdinger operator @xmath207 , where @xmath8 is a compactly supported non local potential .
let us recall that @xmath210 in the paper @xcite , the first author computed the point - point correlations of the plane waves for a scattering problem in @xmath491 in terms of the green s function : for a fixed spectral parameter , plane waves are viewed as random waves parametrised by the direction of their incoming part .
the motivation comes from passive imaging in seismology , a method developped by michel campillo s seismology group in grenoble , as described for example in the papers @xcite . following a similar method , we will compute the correlation of plane waves for our graphs viewed as random waves parametrised by points at infinity . from theorem [ fhtil ]
we get , for any @xmath492 such that @xmath493 , the following formula for the kernel of @xmath494 : @xmath495_{x , y } = \int_{\go}\int_{s^0 \setminus \hat{\cal e } } \phi ( \gl_s ) \overline{e_(x,\go , s ) } e_(y,\go , s)\ d\gs_o(\go ) d\mu(s ) \ ] ] taking @xmath496 , the characteristic function of some interval @xmath497\subset i_q \setminus \cal e$ ] , we get : @xmath498_i ( x , y ) = 2\int_{\go}\int_{f_q(\gl)}^{f_q(a ) } \overline{e(x,\go , s ) } e(y,\go , s)\ d\gs_o(\go ) d\mu(s ) \ ] ] where we set @xmath499 as in ( [ chge ] ) , and where we use the fact that , by the map @xmath204 , the circle @xmath88 is a double covering of @xmath89 . in particular @xmath500 . in the sequel we note @xmath501 for simplicity .
if we consider the plane wave @xmath502 for @xmath503 , as a random wave , we can define the point - to - point correlation @xmath504 of such a random wave in the usual way : for any @xmath505 , the point - to - point correlation @xmath504 of the random wave @xmath502 is given by @xmath506 denoting again by @xmath356 l the green s function of @xmath9 : @xmath507:= g(\gl , x , y)$ ] for @xmath508 we prove the for any @xmath505 and any vertices @xmath509 the point - to - point correlation can be expressed in terms of the green s function as @xmath510 taking the derivative with respect to @xmath128 in equation ( [ chart ] ) yields : @xmath511_i ( x , y ) = ... \nonumber\\ & = & -2 f'_q(\gl ) \frac{(q+1)\log q}{\pi } \frac { \sin^2 ( s(\gl ) \log q ) } { q+q^{-1 } -2 \cos ( 2s(\gl ) \log q)}\ \int_{\go } \overline{e(x,\go , s(\gl ) ) } e(y,\go , s(\gl))\ d\gs_o(\go)\nonumber\\ & = & \frac{q+1}{2\pi}\frac { \sqrt{4q-\gl^2 } } { ( q^2 + 2q + 1-\gl^2 ) } \ \int_{\go } \overline{e(x,\go , s(\gl ) ) } e(y,\go , s(\gl))\ d\gs_o(\go ) . \nonumber \end{aligned}\ ] ] thus we have @xmath512 ( x , y ) = \frac{q+1}{2\pi } \frac { \sqrt{4q-\gl^2 } } { ( q^2 + 2q + 1-\gl^2 ) } \ c_{\gl}^{sc}(x , y)\ .\ ] ] now we use the resolvent kernel of @xmath9 : @xmath507:= g(\gl , x , y)$ ] for @xmath508 and stone formula ( [ sff ] ) to write @xmath513_i ( x , y ) = -\frac{1}{\pi } \int_{a}^{\gl } \im g(t+i0,x , y ) dt\ ] ] and get the result .
the lippmann - schwinger eigenfunctions @xmath514 are especially useful to describe the so - called @xmath515matrix ( @xmath516 ) .
first we introduce the following object : [ tmat ] let @xmath517 and @xmath518 be in @xmath337
. define @xmath519 @xmath520 is called the @xmath521matrix . the goal of this section is to establish a relation between @xmath84 and @xmath403 ( theorem [ tvers ] ) . to get the result
we will need the following [ ffff ] for any @xmath188 @xmath522 suppose that we can prove @xmath523 then ( [ ome ] ) follows from @xmath524 so , by plancherel formula it is enough to prove that @xmath525 in the sequel , we set @xmath526 to simplify notations .
we have @xmath527 but @xmath528 if either @xmath115 or @xmath357 is in @xmath529 . as a result ( using definition [ fht2 ] ) @xmath530 thus@xmath531 @xmath532 @xmath533(x )
\overline { e(x,\go , s ) } \ d\gs\ ] ] @xmath534 @xmath535
g(y ) \ d\gs\ ] ] @xmath536 g(y)\ d\gs\ ] ] @xmath537 in the second line above we used the lippmann - schwinger equation and at the last step we used the isometric property of the deformed fourier helgason transform ( theorem [ fhtil ] ) .
we are ready to prove the following [ tvers ] for any @xmath115 and @xmath538 @xmath539 where we set : @xmath540 , ( see appendix a for the precise definition of the measure @xmath541 ) .
this can be written symbolically by @xmath542 from the definition of @xmath84 we get that @xmath543 @xmath544 with@xmath545 in the last step we used the isometric property of @xmath546 and the fact that @xmath547 .
moreover we have @xmath548 ( lemma [ ffff ] ) and @xmath549 @xmath550 + @xmath551 thus the expression in ( [ calcul ] ) is @xmath552 and after doing the t - integration we get ,
@xmath553 we conclude by noticing as previously that the measures @xmath554 converge weakly to @xmath555 as @xmath387 .
a consequence of the relation between @xmath403 and @xmath84 is the unitarity relation for @xmath403 : [ unitar ] suppose @xmath556 .
then for any @xmath140 and @xmath557 with @xmath558 , and for any @xmath559 @xmath560 by theorem [ tvers ] we have @xmath561 the adjoint of the map @xmath562 is clearly given by @xmath563 the relation @xmath564 , which follows from @xmath565 , implies that ( [ calc ] ) holds .
next result explicits the link between the asymptotic behavior of the generalized eigenfunctions and the coefficients of the scattering matrix .
there exist `` transmission coefficients ''
@xmath566 so that the solution of the lippmann - schwinger equation ( [ foitt ] ) writes @xmath567 for any @xmath47 close enough to @xmath568 and these coefficients are related to the scattering matrix by the following formula @xmath569 with@xmath570 from the study of the lippmann - schwinger equation ( [ foitt ] ) , we write the decomposition @xmath571 where @xmath572 where @xmath573 . + let us look at the asymptotic behaviour of @xmath574 as @xmath575 .
we have seen ( theorem [ greenarbre ] ) that the green s function @xmath576 satisfies equation ( [ mspi ] ) @xmath577 then ( [ futtt ] ) and ( [ rad ] ) imply that , if @xmath578 , @xmath579 with @xmath580 noticing that @xmath581 we get that @xmath582 and from ( [ tverse ] ) we derive formula ( [ trco ] ) .
for any @xmath583 we have @xmath584 if @xmath68 and @xmath461 belong to the same end of @xmath585 .
this implies that the function @xmath586 is in fact constant in each end of @xmath585 , so that the transmission coefficient @xmath587 can be written as a function @xmath588 .
moreover the reduced lippmann - schwinger equation depends only on the restriction of @xmath467 to @xmath209 , this implies that the function @xmath589 is also constant in each end of @xmath585 . finally , we get an @xmath590 matrix depending on @xmath140 , denoted by @xmath591 . in this section , we compute the transmission coefficients following the method of @xcite .
let us recall that @xmath210 we recall that @xmath46 denotes the combinatorial distance of the vertex @xmath47 to the root @xmath45 of @xmath2 .
let us set @xmath592 , where @xmath51 is chosen so that @xmath593 is the supremum of @xmath46 for @xmath47 in @xmath209 .
we denote by @xmath20 the ends of @xmath594 ( @xmath595 ) , by @xmath19 the root of @xmath20 and by @xmath596 the boundary of @xmath20 , which consists in the set of all geodesic rays starting from @xmath19 and staying into @xmath597 .
the set of the roots @xmath598 is the circle of radius @xmath51 and @xmath599 . from now on ,
we consider a fixed @xmath600 ( @xmath595 ) , a fixed geodesic ray @xmath64 in @xmath601 , and the associated `` incoming plane wave '' @xmath602 where @xmath221 .
we recall that such a plane wave is a generalized eigenfunction for the adjacency operator @xmath6 on @xmath183 in the sense that it satisfies @xmath603 but is not in @xmath106 .
we are looking for solutions @xmath604 of the equation @xmath605 where the scattered wave @xmath606 satisfies : @xmath607 where @xmath608 ( the so - called radiation condition ) and the coefficients @xmath609 are the transmission coefficients .
these radial waves are generalized eigenfunctions of @xmath6 in the sense defined previously .
we want to get an explicit expression of the transmission vector @xmath610 as we shall see , the transmission vector does not depend on the choice of the geodesic ray @xmath64 , it is uniquely determined by the choice of @xmath600 ; we define @xmath611 we thus recover the result of the previous section , with the following relation @xmath612 we begin with noticing that @xmath613 does not depend of @xmath614 for any @xmath615 .
we set @xmath616 and denote by @xmath617 the vector in @xmath618 having all null coordinates excepted the @xmath600-th coordinate , which is equal to @xmath418 . + we will prove the following consider the integer @xmath51 so that @xmath619 is the smallest ball containing the finite graph @xmath209 .
set @xmath620 , @xmath621 , denote by @xmath622 the restriction of @xmath623 to @xmath624 in the sense that @xmath625 , define @xmath626 in the same way , set @xmath627 and denote by @xmath628 the corresponding dirichlet - to neumann operator ( see definition [ dton ] , appendix b ) .
then @xmath628 and the transmission vector @xmath629 defined by ( [ holm]),([vector ] ) and ( [ trv ] ) exist for any @xmath630 and @xmath631\ , \ ] ] with @xmath632 , @xmath633 let us recall that we have fixed @xmath600 ( @xmath595 ) and a geodesic ray @xmath64 in @xmath601 . from now on we write @xmath634 instead of @xmath635 for any @xmath104 for simplicity .
+ equation ( [ helm ] ) splits into 3 expressions , depending on where @xmath47 is taken . *
if @xmath636 the equation is already verified , since @xmath9 coincide with @xmath6 on each end @xmath637 . *
if @xmath638 , the equation becomes the dirichlet problem @xmath639 , @xmath640 and is uniquely solvable for any @xmath140 outside @xmath641 , the prescribed values of the function @xmath115 at the boundary being given by the @xmath642-vector @xmath643 to be determined in the sequel . *
if @xmath644 , then @xmath47 is one of the roots @xmath645 , and equation ( [ helm ] ) writes : @xmath646 where @xmath647 is the unique interior neighbor of @xmath645 , ( see figure [ fig : tree ] ) , and where we used that the potential @xmath8 vanishes outside @xmath619 . according to definition [ dton ] ( appendix b )
, the dirichlet - to neumann operator @xmath628 corresponding to @xmath627 writes @xmath648 therefore , if @xmath649 , ( [ front ] ) can be rewritten as follows @xmath650 now it remains to compute @xmath651 .
we have , for @xmath652 , @xmath653 .
according to the expression of the radial function @xmath654 , we get , for any @xmath649 and @xmath655 , that @xmath656 let us write the set @xmath657 as @xmath658 , where @xmath659 belongs to the infinite path @xmath64 whereas the @xmath660 vertices of @xmath661 do not . then , using the properties of the busemann function we have @xmath662 so we get for any @xmath655 and after setting @xmath663 , @xmath664 with @xmath665 hence we have @xmath666 @xmath667 these equations can be summarised , for any @xmath655 , as @xmath668+q\gt(s,\go , l ' ) \ga^{n+1}\ ] ] so that equation ( [ frint ] ) gives , for any @xmath655 @xmath669+q\gt(s,\go , l ' ) \ga^{n+1 } = 0 \ .\ ] ] let us set @xmath670 , + and write @xmath671 ( recall that + @xmath672 and @xmath673 are @xmath674vectors having respectively @xmath675 and @xmath609 as their @xmath676coordinate ) .
+ substituting in ( [ fronte ] ) and denoting by @xmath677 the vector in @xmath618 having all null coordinates except @xmath678 , we get @xmath679+q\ga { \vec a}_l + \ga^{-n}(\ga^{-1}-\ga)\vec { e_{l } } + q\overrightarrow{\gt(s,\go ) } \ga^{n+1}= 0\,\ ] ] which yields @xmath680 using the expression of @xmath119 and @xmath681 , we have then @xmath682 since the matrix @xmath683 is real symmetric , @xmath684 is an invertible matrix for any @xmath185 so that @xmath685 , and @xmath686 we conclude the proof by noticing that , for any @xmath655 , @xmath687 .
_ acknowledgment : we thank the referee for his careful reading of our manuscript and for suggesting many improvements to our initial text . _
the goal of this appendix is to define in a precise way the meaning of the measures @xmath688 where @xmath689 is absolutely continuous w.r . to the lebesgue measure in @xmath491 and @xmath84
is a @xmath690 real valued function so that @xmath691 does not vanish on the hyper - surface @xmath692 .
the measure @xmath688 is supported by the hyper - surface @xmath693 .
we can assume that @xmath491 and the hyper - surface @xmath692 are oriented , so that we can play with differential forms instead of measures .
the proof of the following lemma is left to the reader : there exists a differential form @xmath694 defined in some neighborhood of @xmath695 so that @xmath696 .
moreover the restriction of @xmath694 to @xmath695 is uniquely defined . if @xmath396 is the measure on @xmath695 associated to the restriction of @xmath694 to @xmath695 .
we can view @xmath396 as a measure on @xmath491 denoted @xmath697 .
we can view @xmath698 as weak limits : if @xmath699 is a positive @xmath700 function of integral @xmath418 and @xmath701 the measure @xmath702 is the weak limit as @xmath703 of the measures @xmath704 ( for the proof , take local coordinates so that @xmath705 ) .
usual choices are @xmath706 the characteristic function of the interval @xmath707 $ ] and @xmath708 .
let @xmath710 be a connected finite graph and let @xmath711 be a subset of @xmath712 called the `` boundary of @xmath0 '' .
let @xmath713 be a symmetric matrix associated to @xmath0 , namely @xmath714 after setting @xmath715 , we define @xmath716 as the restriction of @xmath717 to the functions which vanish on @xmath711 .
[ dton ] assume that @xmath718 is invertible .
let @xmath719 , and @xmath193 be the unique solution of the dirichlet problem @xmath724 .
then , the dirichlet - to - neumann operator form @xmath725 is defined as follows : if @xmath726 , @xmath727 y. colin de verdire . _ a semi - classical calculus of correlations .
_ thematic issue `` imaging and monitoring with seismic noise '' of the series `` comptes rendus gosciences '' , from the acadmie des sciences * 343*:496501 ( 2011 ) . | we describe the spectral theory of the adjacency operator of a graph which is isomorphic to a regular tree at infinity . using some combinatorics ,
we reduce the problem to a scattering problem for a finite rank perturbation of the adjacency operator on a regular tree .
we develop this scattering theory using the classical recipes for schrdinger operators in euclidian spaces . |
the dynamics of magnetic flux propagation in a long josephson junction ( ljj ) has been and continues to be a subject of strong theoretical and practical interest during the last three decades @xcite .
the magnetic flux quantum in a ljj is a soliton ( also known as _ fluxon _ ) governed by the well - known sine - gordon ( sg ) equation .
the convenient way to prepare a junction with the required properties is to install various inhomogeneities into it . up to now
the substantial theoretical work has been devoted to the study of the fluxon motion in the one - dimensional ( 1d ) ljjs with point - like @xcite and spatially extended @xcite inhomogeneities .
experimental results on the fluxon scattering on impurities are reported in @xcite spatially inhomogeneous josephson systems with trapped fluxons have been discussed as prospective applications , such as fluxon - based information devices @xcite .
real ljjs are always two - dimensional ( 2d ) , or , more precisely , quasi - one - dimensional ( q1d ) , in the sense that they have a finite width in the direction perpendicular to the direction of fluxon propagation . up to now
the fluxon dynamics in q1d junctions has been scarcely investigated as compared to the pure 1d case .
most of attention has been focused on the various isotropic 2d structures like oscillons and ring kinks @xcite .
it is worth mentioning also the case of window junctions @xcite , which can be called inverse in some sense : it studies point or rectangular junctions embedded in a larger two - dimensional superconduncting sample .
quasi - one - dimensional solitons , or , in other words , solitonic fronts in infinite ( in both @xmath1 and @xmath2 directions ) samples have been studied in detail by malomed @xcite .
several interesting results in the absence of dissipation have been reported including the waves in the q1d sine - gordon equation , travelling along the soliton line @xcite and skyrmion scattering on impurities in the 2d baby scyrme model @xcite . however , in the case of the fluxon dynamics in a ljj the presence of dissipation is unavoidable .
it is of interest to investigate the q1d fluxon dynamics in the presence of spatial inhomogeneities when the junction width is finite .
we expect that the fluxon transmission in this case will be significantly enhanced comparing to the pure 1d case .
for example , in the previously studied case of the lattice acoustic soliton front interaction with mass impurities @xcite it has been shown that the front can _ round _ the point impurity while a 1d lattice soliton gets reflected from it . moreover
, the soliton front can overcome even the impurity of the infinite mass .
to our knowledge the fluxon interaction with spatial inhomogeneities has not been studied except in @xcite , however this paper deals only with the infinite sample width .
thus , in this letter we aim at studying the q1d fluxon interaction with impurities and finding out how this interaction depends on the junction width and other system parameters .
in particular , it is planned to find how the retrapping current ( e.g. , the minimal bias current for which the fluxon propagation is still possible ) depends on the junction width .
the paper is organized as follows . in the next section , the model is described .
section [ sec3 ] is devoted to the fluxon transmission through impurities . in the last section discussion and conclusions
are presented .
we consider the q1d long josephson junction ( ljj ) subjected to the external time - independent bias .
the main dynamical variable is the difference between the phases @xmath3 of the macroscopic wave functions of the superconducting layers of the junction .
the time evolution of the phase difference is governed by the perturbed sine - gordon ( sg ) equation of the form @xmath4\sin \phi=-\alpha \phi_t-\gamma,\ ] ] where @xmath5 and the indexes @xmath6 stand for the respective partial derivatives . in this dimensionless equation
the spatial variables @xmath1 and @xmath2 are normalised to the josephson penetration depth @xmath0 , the temporal variable @xmath7 is normalised to the inverse josephson plasma frequency @xmath8 @xcite .
the bias current @xmath9 is normalised to the critical josephson current of the junction and @xmath10 is the dimensionless dissipation parameter .
the function @xmath11 describes the spatial inhomogenity . in the case of point impurities in the general form it reads @xmath12 it is supposed that there are @xmath13 impurities in this junction , positioned at the points @xmath14 , @xmath15 , @xmath16 , with @xmath17 being the `` strength '' or the amplitude of the impurity .
only the microshorts ( @xmath18 ) , i.e. , a narrow regions of locally enhanced critical density of the tunnelling superconctucting current will be investigated in this article . however , the size an inhomogeneity in experimental samples is finite @xcite .
therefore , we consider also the case of the line microshort of width @xmath19 , stretched along the @xmath2 direction : @xmath20 \,.\ ] ] here @xmath21 is the heavyside function . and finally , the rectangular impurity of the finite size in both @xmath1 and @xmath2 directions @xmath22 \left [ \theta\left ( y+\frac{d_y}{2}\right ) + \theta \left ( \frac{d_y}{2}-y\right ) \right ] \ ] ] will be considered as well .
it should be noted that the impurity strength @xmath17 has different meanings in all three cases ( [ 2])-([4 ] ) .
for the point impurity ( [ 2 ] ) setting @xmath23 does not automatically yield the pure 1d case studied before in @xcite .
this case can be retained in the strip impurity case if @xmath24 .
the 1d finite - size impurity case @xcite is retained in the same way .
we choose the boundary conditions along the @xmath2 direction in the von neumann form : @xmath25 the boundary conditions along the @xmath1 axis are periodic : @xmath26 , where @xmath27 is the junction length , @xmath28 .
in order to get an idea about the character of the fluxon dynamics , the numerical integration of the q1d sg equation ( [ 1 ] ) has been performed . the josephson phase and its space derivatives
are discretized in the following way : @xmath29 , + @xmath30 , while the @xmath31-function is approximated by the kronecker @xmath31-symbol .
the resulting set of ode s with the boundary conditions were integrated using the 4th order runge - kutta scheme .
details of the fluxon propagation through the two identical point impurities ( [ 2 ] ) placed along the @xmath2 axis at @xmath32 and @xmath33 are given in fig .
it is important to mention that the dissipation in eq .
( [ 1 ] ) is crucial and the soliton interaction with impurities differs from the dissipationless case where the complex resonant behaviour occurs either for the @xmath31-like @xcite or for the finite - size @xcite obstacles . far away from the impurities the fluxon exists as only one attractor of the system with the velocity , predefined by the damping parameter and external bias .
therefore , contrary to the non - dissipative case , the transmission consists of only two possible scenarios : trapping and passage .
for the sake of better visualization the derivative @xmath34 is plotted on the @xmath35 plane for the different time moments and for three different dissipation values : @xmath36 , @xmath37 and @xmath38 . without loss of generality the topological charge
is assumed to be @xmath39 ( soliton ) throughout the paper .
the initial conditions are taken in the form of the approximate ( invariant in the @xmath2 direction ) soliton solution , placed at the beginning of the junction and having with the equilibrium velocity .
the fluxon interaction with the rectangular impurity is presented in fig .
[ fig2 ] for @xmath36 and @xmath38 . in this case , the countour plot of the function @xmath34 is plotted on the @xmath35 plane .
the fluxon retrapping current was computed numerically as well , showing its steady decay when the junction width was increased .
these results will be discussed in detail in the next subsection .
the following conclusions can be drawn from these results .
the fluxon shape experiences certain changes after passing the impurities , namely the redistribution of the josephson phase along the fluxon line in the @xmath2 direction , as well as the slight bending of the fluxon shape in the same direction .
these distortions eventually die out after some time , for @xmath37 and @xmath36 this happens quite soon after the passage through the impurity . for smaller dissipation ( @xmath38 , as shown in figs .
[ fig1]i - l ) the oscillations of the josephson phase along the @xmath2 direction seem to survive for much longer time , comparable with the fluxon propagation time along the junction .
the numerical simulations for thinner junctions , @xmath40 ( not shown in the paper ) , demonstrate that these shape distortions are much weaker and are hardly noticeable .
thus , when studying the q1d fluxon interaction with impurities one can assume with the high degree of certainty that the fluxon is an almost hard rod at least if @xmath41 .
due to finiteness of the junction in the @xmath2 direction and the boundary conditions ( [ 5 ] ) , the straight soliton front is the energetically most favourable solution and thus it is not possible to observe the arc - like solitons reported in @xcite . similarly to fluxon propagation in the 1d ljj @xcite ,
there must exist two characteristic values of the bias current , @xmath42 and @xmath43 , @xmath44 .
moreover , the current - voltage characteristics of the ljj with impurities have hysteretic nature @xcite . if @xmath45 , the pinning on the impurity is not possible and there exists only one attractor that corresponds to fluxon propagation .
this happens because the bias current is too strong for the fluxon to get trapped on the impurity . in the interval @xmath46 , at least two attractors coexist : one corresponds to fluxon pinning on the impurity ( there can be several different pinned fluxons if the impurity has finite length @xcite ) and another one to fluxon propagation . if @xmath47 , the only possible regime is fluxon pinning on the impurity .
the value of @xmath42 is defined only by the properties of the impurity , and can be obtained directly from the 1d analog .
contrary , for the retrapping current the dimensionality of the junction and its width are crucial . far from the impurity , the fluxon kinetic energy is proportional to the junction width and equals @xmath48 $ ] . here ,
@xmath49^{-1/2}$ ] is the equilibrium fluxon velocity in the spatially homogeneous ljj @xcite . in the non - relativistic fluxon case ( @xmath50 ) ,
one gets @xmath51 . by substituting the ansatz + @xmath52/\sqrt{1-v^2}\}$ ] into eq .
( [ 1 ] ) , where @xmath53 is the coordinate of the fluxon center of mass , one can obtain the newtonian equation of motion for the fluxon center . since the fluxons under consideration
are extended objects in the @xmath2 direction , the equation for the center of mass dynamics as well as the impurity potential should depend on @xmath2 .
taking into account the numerical simulation in the previous section , we consider the fuxon as an absolutely rigid rod .
we also mention that the impurity function @xmath11 can be factorised in the cases ( [ 3])-([5 ] ) , therefore its center of mass dynamics can be effectively projected on the @xmath1 axis and the respective equation of motion can be written as @xmath54 where the center of mass coordinate @xmath55 depends only on time .
the impurity potential @xmath56 now can be calculated from the respective 1d problem by simply taking away the @xmath2-dependent part in eqs .
( [ 2])-([4 ] ) .
the fluxon mass @xmath57 has to be rescaled depending on the type of the impurity . in the pure 1d case @xmath58 .
this assumption works well only if the impurity is consists of lines and/or rectangles .
if it has a more complex shape , for example , like triangle , the projection on the 1d problem becomes more complicated .
* point impurity .
* in the point impurity case ( [ 2 ] ) , we consider the line of @xmath13 identical equidistant impurities with @xmath59 , @xmath16 which are placed symmetrically with respect to the central line @xmath60 .
each impurity creates the potential @xmath61 . in order to project the problem on the 1d picture it is necessary to rescale the fluxon mass from @xmath58 to @xmath62 , thus within the kinematic approach the retrapping current can be found as a root of the energy balance equation @xmath63 , where @xmath64\simeq 4wv_\infty^2/n+{\cal o}(v_\infty^4)$ ] is the fluxon kinetic energy . in the non - relativistic case one gets @xmath65 .
the correction of the order @xmath66 can be taken into account with the help of the method , developed in @xcite .
the modification of this method for the q1d case is straightforward , therefore we describe only the main steps .
the improved energy balance relation equates the fluxon energy at @xmath67 and its losses due to the dissipation , @xmath68 , with the maximal height of the potential barrier @xmath69 : @xmath70 where @xmath71 , @xmath72 are the extrema of the potential @xmath69 , @xmath73 , @xmath74 .
equation ( [ en - bal ] ) is universal and will be used for all types of impurities .
the energy loss due to the dissipation equals + @xmath75 . inserting this correction term @xmath68 into the improved energy balance equation and keeping the terms up to the order @xmath66
, one gets the final corrected expression for the retrapping current : @xmath76 it appears that the @xmath77 relation is obtained simply by dividing the impurity strength by the factor @xmath78 , moreover , the @xmath66 correction does not depend on the junction width at all . from this expression
one can clearly see that if only the @xmath79 term is taken into account , the retrapping current disappears as @xmath80 , thus , in the infinitely wide junction a fluxon always passes the impurity . the second term in eq .
( [ 4a ] ) does not depend on @xmath81 , and , therefore it may lead to the wrong conclusion that @xmath43 does not tend to zero as @xmath80 .
however , it should be noted that this term has been derived under the assumption of @xmath81 being finite .
the numerical simulations confirm the main suggestion of this letter : the retrapping current decays with the growth of the junction width . the approximate expression ( [ 4a ] ) appears to be in good agreement with the numerical data for smaller values of @xmath81 , while for larger @xmath81 the numerical and analytic data diverge ( however , the agreement remains to be satisfactory ) .
all this is presented in fig .
[ fig1]a where the analytic results are given by solid lines and the numerical data are shown by markers . in the case of two and three impurities , placed along the line in the @xmath2 direction , the retrapping current
virtually does not depend on the distance @xmath82 between them . for instance
, the markers that correspond to @xmath83 ( @xmath84 ) , @xmath85 ( @xmath86 ) , @xmath87 ( @xmath88 ) at @xmath89 and @xmath90 are almost indistinguishable .
one may consider the excentrically placed ( i.e. , lying away from the @xmath91 axis ) impurity .
the kinematic approach does not distinguish the impurities , placed at the different positions .
the numerical simulations show that the difference between the retrapping currents in both the cases is very small .
for example , @xmath92 for the centrally placed impurity , @xmath93 , at @xmath94 ( the rest of the parameters are as in fig .
[ fig1]a ) and @xmath95 for the impurity at @xmath96 , @xmath97 .
* line impurity .
* in the case of the impurity line ( [ 3 ] ) , the same energy balance expression holds as for the point impurity but the fluxon mass is assumed to be @xmath98 .
the rest of the calculation procedure for the retrapping current is the same as for the point impuritiy . as a result
, we obtain the the final expression for the retrapping current : @xmath99 numerically computed @xmath43 appears to be in good correspondence with the approximate expression ( [ 4b ] ) as shown in fig .
similarly to the point impurity case , the discrepancy between the analytical and numerical results increases at larger @xmath81 . in the limit @xmath100 , the effective 1d picture
is restored because the impurity strip crosses the whole junction in the @xmath2 direction .
thus , the retrapping current attains the value @xmath101 for the 1d soliton case ( shown by the thick horizontal line in fig .
[ fig3]b ) .
thus , we observe that if the strip impurity consitutes , for example , about @xmath102 of the junction width , the retrapping current is about @xmath103 less than the respective 1d value .
* rectangular impurity . * finally , we consider the rectangular impurity ( [ 4 ] ) . in the point - particle description ,
the fluxon `` feels '' this impurity as the potential @xmath104 $ ] @xcite . in this case
, the energy balance reads @xmath105=4\mu d_y \tanh ( d_x/2)$ ] , from which the expression for the threshold current can be easily computed . for large @xmath106
it can be complemented by the correction @xmath66 computed for the 1d case in @xcite .
the fluxon mass should be renormalized as @xmath107 . in this limit
it is assumed that the impurity creates the step - like potential @xmath108 $ ] , which has a minimum at @xmath109 and a maximum at @xmath110 .
the energy losses due to the dissipation @xmath111+{\cal o}(1)\}$ ] should be substituted in the energy balance equation ( [ en - bal ] ) .
the final approximation reads @xmath112.\ ] ] note that at variance with the point and line impurities , the second correction depends on the both junction and the impurity width . in the 1d limit ( @xmath100 ) , the formula ( [ 9 ] ) retains the form , found in @xcite . in the limit @xmath113
, the second term decays to zero , but much slower than the decay law @xmath114 of the first term .
once again , the second term is a correct approximation only for finite values of @xmath81 . in the opposite limit , @xmath115
, one can use the following approximation : @xmath116\simeq d_x \mbox{sech}^2 x + { \cal o}(d_x^2)$ ] .
thus , the effective potential @xmath69 is approximately the same as for the line impurity up to the coefficient @xmath117 .
therefore the formula ( [ 4b ] ) can be modified accordingly and finally the retrapping current reads @xmath118.\ ] ] the results of numerical simulations given in fig .
[ fig4 ] demonstrate satisfactory correspondence with the analytical approximations ( [ 9])-([10 ] ) .
the deviations from the numerical results appear to be stronger as compared to the point and line impurities , mostly because we were unable to provide the @xmath66 correction that works equally well in both limits @xmath119 and @xmath120 and also due to higher degree of the fluxon deformation during the interation with the impurity .
it should be mentioned that even for @xmath121 one can consider the `` long '' @xmath106 limit and it works fairly well .
indeeed , the markers that correspond to @xmath122 ( @xmath123 ) , @xmath124 ( @xmath125 ) and @xmath126 ( @xmath88 ) seem to be almost indistinguishable in fig .
to summarize , we have shown that quasi - one - dimensional fluxon passage across microshorts is significantly enhanced in comparison with the purely one - dimensional case .
the retrapping threshold current decays with the junction width approximately as @xmath127 , according to the kinematic approach .
the numerical simulations support this dependence for intermediate ( several @xmath0 ) values of @xmath81 . with the increase of @xmath81 ,
the discrepancy between the numerical results and the kinematic approximation becomes more distinct .
the reason of this discrepancy lies in the fact that for @xmath128 the fluxon can not be longer considered as an completely rigid object and its deformation the @xmath2 direction should be taken into account .
one can formulate the following simple argument that explains the enhanced fluxon transmission across the obstacle of the width @xmath19 in the q1d case .
the obstacle can be described as a localized ( both in the @xmath1 and @xmath2 directions ) potential barrier . only the central part ( @xmath129 ) of the initially homogeneous in the @xmath2 direction fluxon takes part in the interaction process , while the marginal areas @xmath130 do not .
thus , if the energy in the tails is sufficient enough to overcome the barrier , the fluxon will pass .
if @xmath113 , the energy in the non - interacting part of the fluxon tends to infinity , and , consequently , it will overcome any localized obstacle .
finally , we would like to outline the future research directions . in our opinion
they lie beyond the rigid rod approximation used in this letter . in this case
the fluxon center of mass depends also on the transverse coordinate @xmath2 and the effective equations of motion are pde s and not ode s , like eq .
( [ 7 ] ) .
accounting for the dependence on @xmath2 becomes important in the following cases : ( i ) the junction width is too large and the flexural oscillations along the fluxon line appear ; ( ii ) the dissipation is rather small and/or the junction length is @xmath131 , thus the spatial distortions of the fluxon line do not have enough time to die out .
these cases are also interesting in connection with the arc - like fluxons found in the infinitely wide junction @xcite and the variety of the excitations that travel along the fluxon crest in the dissipationless 2d junctions @xcite .
the combined effect of the dissipation and the finite junction width makes the transverse - invariant fluxon front the energetically most favourable solution .
however , other attractors of the system which are non - transverse - invariant may exist as suggested by figs .
[ 1]j - l , especially when dissipation is small .
finding them and investigating how they will manifest themselves on the junction current - voltage dependence is an important problem .
authors are indebted to dffd ( project f35/544 - 2011 ) for financial support .
s. sakai , h. akoh and h. hayakawa , japan .
( 1985 ) l771 ; h. akoh , s. sakai , a. yagi and h. hayakawa , ieee trans .
( 1985 ) 737 ; i.l .
serpuchenko and a.v .
ustinov , pisma zh .
46 ( 1987 ) 435 [ sov
jetp lett .
46 ( 1987 ) 549 ] . | fluxon transmission through impurities of different shape in a quasi - one - dimensional long josephson junction is investigated .
the junction width is significantly less than its length but , at the same time , is of the order of the josephson penetration length @xmath0 or exceeds it .
the retrapping current on the impurities of the point , line and rectangular shape is computed as a function of the junction width both numerically and analytically .
good agreement between the analytic formulae and the numerical simulation results for the intermediate ( several @xmath0 ) junction width is observed . solitons , josephson junctions , fluxons , impurities |
the atlas central detector @xcite consists of an inner tracking detector ( @xmath0 ) , electromagnetic and hadronic calorimeters ( @xmath1 ) and the muon spectrometer ( @xmath2 ) .
in addition , atlas is also equipped with the lucid @xcite , zdc @xcite and alfa @xcite detectors which partially cover the forward rapidity region .
the cross - sections for elastic and diffractive production are large . at the centre - of - mass energy of 14 tev
the elastic cross - section is estimated to be 25 - 30 mb .
the cross - section for single and double diffraction are estimated 10 - 15 mb .
thus , elastic and diffractive processes account for roughly half of the @xmath3 total cross - section of @xmath4 mb .
thus only modest luminosity is required to study these processes .
this is fortuitous since event pile - up resulting from higher luminosity running will tend to destroy the rapidity gap signature of these forward physics processes .
lucid is composed of two modules located at @xmath5 m from the interaction point that provide a coverage @xmath6 for charged particles .
each lucid detector is a symmetric array of 1.5 m long polished aluminium tubes that surrounds the beam - pipe and points toward the atlas interaction point ( ip ) .
this results in a maximum of cerenkov emission from charged particles from the ip that traverse the full length of the tube .
each tube is 15 mm in diameter and filled with c4f10 gas maintained at a pressure of 1.2 - 1.4 bar giving a cerenkov threshold of 2.8 gev for pions and 10 mev for electrons .
the cerenkov light emitted by the particle traversing the tube has a half - angle of @xmath7 and is reflected an average 3 - 4 times before the light is measured by photomultiplier tubes which match the size of cerenkov tubes .
the fast timing response ( a few ns ) provides the unambiguous measurements of individual bunch - crossings .
lucid is sitting in the high radiation area that is estimated to receive a radiation dose of @xmath8 mrad per year at maximum luminosity ( @xmath9 ) .
lucid is a relative luminosity detector and during the initial period of lhc operation , the absolute calibration would come from the lhc machine parameters allowing the luminosity to be determined to a precision of @xmath10% .
after an initial period of lhc running @xmath11 boson counting can be used , as the production cross sections are known well enough to allow and absolute luminosity calibration to 5 - 8% accuracy .
qed processes such as exclusive muon pair production via two photon exchange can be calculated to be better than 1% providing another physics based calibration .
however , the rates of such processes are quite low and their experimental acceptance and detection efficiency are difficult to estimate accurately . the final absolute luminosity calibration will be determined to a precision of a few percent using elastic proton - proton scattering in the coulomb nuclear interference ( cni ) region covered by the alfa detector .
this method requires special low luminosity high beta runs and consequently it is unlikely that this source of calibration will be available in initial lhc running .
the zero degree calorimeters ( zdcs ) provide coverage of the region @xmath12 for neutral particles .
they reside in a slot in the tan ( target absorber neutral ) absorber , which would otherwise contain copper shielding .
the zdc is located at @xmath13 m from the interaction point , at a place where the straight section of the beam - pipe divides into two independent beam - pipes .
there will be four zdc modules installed per arm : one electromagnetic ( em ) module and three hadronic modules .
each em module consists of 11 tungsten plates , with their faces perpendicular to the beam direction .
the height of these plates is extended in the vertical direction with 290 mm long steel plates .
two types of quartz radiator are used : vertical quartz strips for energy measurement and horizontal quartz rods which provide position information . at present only hadronic modules
are installed .
the em module will be installed once the lhcf project has completed data taking .
the roman - pot spectrometers are located @xmath14 m away from the interaction point ( ip ) .
there will be two roman pot stations separated by four meters on either side of the ip .
the main requirements on the alfa scintillating fibre detectors that will be housed in the roman pots are : a spatial resolution of about @xmath15 m ; no significant inactive region ; minimal sensitivity to the radio frequency noise from the lhc beams ; and , ability to operate in the vacuum maintained in the roman pots . at the beginning of the run ,
the alfa detectors are in withdrawn position far from the beam .
after the beam has stabilized , the detectors are moved back to within 1.5 mm of the beam .
elastic and diffractive protons deflected from the beam pass through arrays of scintillating fibre trackers ( 20 @xmath16 64 fibres in each array ) , which measure the distance of the proton to the beam .
traditionally , the absolute luminosity at hadron colliders has been determined via elastic scattering at small angles .
atlas also pursues this approach with the alfa detector .
the extremely small angles ( @xmath17 ) needed to make these measurements are smaller than the nominal beam divergence . so special beam conditions e.g. high - beta ( @xmath18 ) optics in combination with reduced beam emittance , are required .
alfa will be used to determine the absolute luminosity via elastic scattering at small angles in the coulomb - nuclear interference region .
single diffractive ( sd ) can be tagged by identifying the rapidity gap , by requiring that the forward detector system register little hadronic activity .
the atlas forward calorimeter ( fcal ) , lucid and the zdc can be utilized as part of a rapidity gap requirement for the sd analysis .
di - jet production by sd should be measurable with @xmath19 of the data , corresponding to around 1.5 years of data acquisition at @xmath20 .
the cross - section for sd di - jet production is predicted by the pomwig event generator @xcite to be 3.6 ( 0.20)@xmath21b for @xmath22 for jet transverse energy ( @xmath23 ) greater than 20 ( 40)gev , where @xmath24 is the fractional momentum lost by proton during the interaction .
di - jet production permits a study of factorization breaking in diffractive events .
additional soft interactions and multiple parton - parton scattering during the @xmath25-interaction reduce the observed cross - section for diffractive processes at hadron colliders , with respect to the predicted cross - section obtained from diffractive parton distribution function measured at hera . the central exclusive production ( cep ) is defined as the process @xmath26 , where all of the energy lost by proton goes into the production of a hard central system , @xmath27 .
thus the final state consists of two outgoing protons , a hard central system and no other activity .
the cep allows direct access to quantum numbers of @xmath27 and has the direct relation of the scattered protons energy loss to the central mass @xmath28 with clean azimuthal correlation of the both scattered protons . at certain scenario it could provide a clean higgs discovery channel @xcite .
the di - jet cross section is predicted by the exhume event generator @xcite to be approximately 8nb for a minimum jet transverse energy of 20gev .
it should be noted that the current measurements of cep by the cdf collaboration @xcite are in good agreement with the theoretical predictions that form the physics basis of the exhume generator .
given the large pre - scale on low @xmath23 jets , one would expect approximately 100 events in 100pb@xmath29 of data .
it is necessary to reduce the level 1 ( l1 ) pre - scaling to obtain the good measurement of the cep . it may be possible to do this in atlas by exploiting the clean nature of the exclusive event by requiring a rapidity gap in the l1 trigger ( using lucid , zdc ) , in conjunction with a triggered jet .
the single diffractive ( sd ) is characterized by a centrally produced system separated by a rapidity gap , or lack of hadronic activity , from an outgoing proton . in sd exchange
the outgoing proton can be tagged and measured during special lhc runs by the alfa detectors .
however , the low luminosity means that only soft sd processes can be studied , in particular the forward proton spectrum at low @xmath24 .
the acceptance is @xmath30 50% ( 10% ) for @xmath31 0.01 ( 0.1 ) .
it is expected that at a luminosity of @xmath32 there would be 1.2 to 1.8 million events recorded in 100 hours of data acquisition .
the @xmath24 measurement resolution is approximately 8% for @xmath33 , falling to 2% for @xmath34 .
the aim of the atlas forward proton ( afp ) @xcite project is to install spectrometers at @xmath35 m and @xmath36 m from the interaction point of atlas . in exclusive central production processes where the incident protons remain intact ,
the precise measurement of fractional momentum loses ( @xmath37 ) can be used to determine the mass of the central system with great accuracy , using the relation @xmath38 where @xmath39 is square of the center of mass energy .
the @xmath24 acceptance of the afp detectors at 220 m and 420 m is @xmath40 and @xmath41 , respectively .
this implies a mass acceptance of the central system spanning the range from 80 gev to masses in excess of 1 tev .
afp project opens up a possibility of searching for new physics in cep processes , such as higgs boson production in sm , mssm and nmssm @xcite .
because of the limited available space at 420 m , the traditional roman pot technique can not be used . instead
, the afp group have opted to employ so called a hamburg movable beam pipe system to deploy the afp forward spectrometers . in order to achieve a good acceptance and mass resolution ,
3d silicon edgeless technology has been chosen for the spectrometers , where the 3d silicon sensors have rectangular pixels of dimensions 50 microns by 400 microns .
it is envisaged that the pile - up background can be handled by using ultra precise time - of - flight ( tof ) detectors to differentiate the vertex of interest from the vertices of the pile - up events by measuring , the arrival time of the two deflected protons in the tof detectors with a precision of @xmath4210ps .
there are two approaches to tof measurement currently being studied .
both of these approaches utilize cerenkov detectors readout by microchannel plate ( mcp ) photomultipliers .
one of these approaches utilizes a gas cerenkov radiator ( gastof ) , whilst the other employs a fused - silica radiator ( quartic ) .
the luminosity monitor lucid , calibrated by the alfa detector , will allow the luminosity delivered to atlas to be determined to better than 5% accuracy .
the zdc will measure forward spectators for heavy ion collisions and provide trigger and centrality measurements .
it will also provide a luminosity measurement and , measure forward particle production for mc tuning .
low luminosity forward physics topics include : elastic scattering using alfa ; sd forward proton spectrum ( alfa ) ; single diffractive di - jet and @xmath43 production and di - jets from double pomeron exchange ( dpe ) and cep ( with rapidity gap vetoes from fcal , lucid , zdc ) . at high luminosities the atlas forward proton ( afp ) project
aims to deploy proton taggers at @xmath35 m and @xmath36 m in order to obtain access to a rich new vein of cep physics , that includes sm / mssm / nmssm higgs boson studies , @xmath43 pair production , slepton production and gluino pair production , etc .
9 g. aad _ et al . _ , ( the atlas collaboration ) , jinst 3 s08003 ( 2008 ) .
et al . _ , ( the atlas collaboration ) , jinst 3 s08003 , * 206 * ( 2008 ) .
et al . _ , ( the atlas collaboration ) , jinst 3 s08003 , * 214 * ( 2008 ) .
et al . _ , ( the atlas collaboration ) , jinst 3 s08003 , * 211 * ( 2008 ) .
cox and j.r .
forshaw , comput .
144 , 104 ( 2002 ) .
khoze , a.d .
martin and m.g .
ryskin , eur .
j. c. * 34 * ( 2004 ) 327 ; j.r .
forshaw , arxiv:0901.3040 ( 2009 ) .
t. aaltonen _
et al . _ [ cdf coll .
( 2007 ) 242002 ; t. aaltonen _ et al .
_ [ cdf coll .
d 77 ( 2008 ) 052004 . j. monk and a. pilkington , comput .
communication 175 ( 2006 ) 232 .
s. heinemeyer _ et .
j. c 53 ( 2008 )
. b. cox _ et .
al . _ , jhep 0710:090 , ( 2007 ) . | in this communication i describe the atlas forward physics program and the detectors , lucid , zdc and alfa that have been designed to meet this experimental challenge .
in addition to their primary role in the determination of atlas luminosity these detectors - in conjunction with the main atlas detector - will be used to study soft qcd and diffractive physics in the initial low luminosity phase of atlas running .
finally , i will briefly describe the atlas forward proton ( afp ) project that currently represents the future of the atlas forward physics program . |
ngc891 is one of the best known and studied nearby edge - on spiral galaxies .
it is at the distance of 9.5 mpc , is classified as a sb / sbb , and it is often referred to as a galaxy very similar to the milky way @xcite . because of its very high inclination ( [email protected]@xmath3 , * ? ? ?
* ) it is very suitable for the study of the distribution and kinematics of the gas above the plane .
ngc891 has been the subject of numerous studies at different wavelengths that have led to the detection of various halo components : an extended radio halo @xcite , an extended layer of diffuse ionised gas ( dig ) ( e.g. * ? ? ?
* ) and diffuse extra - planar hot gas @xcite . also `` cold '' ism components have been detected in the halo such as hi @xcite , dust @xcite and co @xcite .
here we concentrate on the neutral gas and present results from recent third generation hi observations obtained with the westerbork synthesis radio telescope ( wsrt ) .
ngc891 was first studied in hi in the late seventies with the wsrt and the presence of neutral gas seen in projection out of the plane was reported @xcite .
subsequently , a new study with higher sensitivity showed that the extra - planar emission was very extended , up to 5 kpc from the plane .
3d modeling indicated that such emission was produced by a thick layer of neutral gas rotating more slowly than the gas in the disk @xcite .
since then , several other studies have confirmed the presence of neutral gas in the halos of spiral galaxies .
it has been detected in edge - on or nearly edge - on systems ( e.g. ugc7321 , @xcite and ngc253 , @xcite ) , as well as in galaxies viewed at different inclination angles ( e.g. ngc2403 , * ? ? ?
indications of vertical gradients in rotation velocity have been found in several galaxies also in the ionised gas ( e.g. ngc891 , @xcite , @xcite ; ngc5055 , @xcite ) . in the first part of the paper we present the new hi observations of ngc891 together with a 3d modeling of the hi layer . in the second part ( section 3 ) we study the kinematics of the extra - planar gas and , in the third one ( section 4 ) , we present results from a dynamical model of the extra - planar gas .
the new observations of ngc891 have been carried out during the second semester of 2002 with the westerbork synthesis radio telescope ( wsrt ) and a total integration time of about 200 hrs . with this long integration time
we reach a sensitivity ( r.m.s .
noise per channel = 0.22 mjy / beam at 28@xmath4 km s@xmath1resolution ) that is about a factor 4 better than the previous observations of @xcite . a complete presentation of the observations will be given in oosterloo et al .
( in preparation ) .
figure 1 ( right panel ) shows the new total hi map of ngc891 at 28@xmath5 ( @xmath61.3 kpc ) .
hi emission is detected at a projected distance of as far as 15 kpc from the plane ( see the spur in the n - w side of the disk ) .
the size of the disk itself ( in the plane ) is very similar to that reported by @xcite and @xcite suggesting that we have possibly reached the edge of the hi disk , especially on the n - e side .
the emission above the plane , instead , is significantly more extended than in previous observations and almost everywhere extends up to 10 kpc above and below the plane .
figure 2 shows the position - velocity plot along the major axis of ngc891 at 28@xmath5 resolution with the rotation curve ( white squares ) overlaid .
the rotation curve was derived with the method described in section 3 using only the n - e ( approaching ) side of the galaxy .
the kinematics of the receding side within a radius of @xmath66 kpc is very similar to that of the approaching one . at larger radii
the velocity is apparently declining .
however , we can not be sure that the gas in this extension is at the line of nodes and , therefore , the derivation of a rotation curve in that region is not possible . in the inner regions of the galaxy ( about 1 - 2 kpc ) we confirm the presence of a fast rotation disk or ring
( see also * ? ? ?
the total hi map of ngc891 ( figure 1 ) shows extended emission in the direction perpendicular to the plane .
is this emission coming from gas located in the halo of the galaxy or is it the result of projection effects ? here
we address this question using a 3d modeling technique similar to that of @xcite . in figure 3
we compare some of the observed channels maps of ngc891 ( right hand column ) with models of the hi layer . from the left hand column
they are : 1 ) a warp along the line of sight , i.e. a change of the inclination angle of the disk , from 90 to 60 degrees , in the outer parts ; 2 ) a flaring ( increasing thickness ) of the outer disk from a fwhm of 0.5 kpc up to @xmath66 kpc ; 3 ) a two - component model with thin disk + thick ( fwhm@xmath66 kpc ) disk corotating ; 4 ) and 5 ) two - component models with the thick disk rotating more slowly ( 35 km s@xmath1 ) than the disk .
the models in columns 4 and 5 differ only for the radial density distribution of the thick component : one ( 5 ) is the same ( scaled ) as that of the gas in the disk and the other ( 4 ) has a depression in the central regions . of the models reported here , only those in columns 4 and 5 give a reasonable reproduction of the data .
in particular the warp model does not reproduce the shape of channel maps at 374 and 432 km s@xmath1 , while the flare and corotating model do not reproduce the thin channels of the two top rows ( at 275.5 and 308.5 km s@xmath1 ) .
this clearly indicates that the extra - planar emission in ngc891 is produced by a thick ( fwhm@xmath66 kpc ) layer of hi rotating more slowly than the gas in the plane .
the halo is possibly relatively denser than the disk in the outer parts ( with a radial density distribution somewhat in between the models in column 4 and 5 ) .
in the previous section we have presented some simple galaxy models showing that the emission above the plane in ngc891 comes from gas which is located in the halo region and rotates more slowly ( is lagging ) with respect to the gas in the plane . here
we quantify this lag by deriving the 2d rotation velocity field ( rotation surface ) or rotation curves at different heights from the plane . a rotation curve of a spiral galaxy seen edge - on is usually derived using an envelop - tracing method i.e. by taking the highest measured velocity along the line of sight as the rotation velocity ( e.g. * ? ? ?
this method was , for instance , used to derive the rotation curve shown in figure 2 . in an ideal situation ( high s / n ,
galaxy perfectly edge - on , pure rotation , etc . ) the highest velocity measured along the line of sight is indeed produced by the gas at the line of nodes ( i.e. rotation velocity ) .
however , the real situation can be very different .
in particular it is possible that the emission from the gas at the line of nodes is below detection and , as a consequence , the rotation velocity is underestimated .
this could easily happen when observing the gas above the plane where the density is lower and especially in cases of central hidepressions or holes .
figure 4 ( left panel ) shows the rotation velocity surface ( contours with constant @xmath7 ) of ngc891 derived with the envelop - tracing method for the n - e side of the galaxy .
can this rotation velocity field be fully trusted ? in order to test this , we have constructed model galaxies with known ( flat ) rotation velocity in r and z. figure 4 ( central and right panels ) shows two rotation velocity surfaces obtained from two of these models .
these surfaces are obtained using exactly the same fit and fitting parameters as for the one of ngc891 ( left panel ) .
the two models have different radial density distributions for the extraplanar gas .
these were obtained from the data from above ( n - w side , model a ) and below ( s - e side , model b ) the disk .
the results show that the fit works very well in a large part of the galaxy giving a constant value of the velocity in r and z ( uniform lowest gray level ) .
the fit does not return the input constant rotation velocity with z only in the inner regions outlined by the dashed lines in figure 4 .
there is very little difference between the results obtained with the two different radial density distributions .
these results were used to exclude the inner regions in the 2d rotation velocity field of ngc891 ( left panel ) .
figure 5 shows the results for ngc891 plotted as rotation curves at various distances from the plane after the uncertain inner points have been removed .
the solid line shows the ( smoothed ) rotation curve in the plane while the dashed lines outline the value of the flat part of the rotation curves at the different heights .
the halo of ngc891 appears to corotate up to 1.3 kpc , then it starts to lag with respect to the disk , the lagging increasing with height from the plane .
given the limited angular resolution , the apparent corotation below 1.3 kpc may be the effect of beam - smearing .
the gradient in rotation velocity is roughly 15 km s@xmath1 kpc@xmath1 .
the west and east side of the galaxy do not show significant differences .
a detailed description of this procedure and results will be given in a forthcoming paper ( fraternali , in preparation ) .
the presence of neutral gas in the halo of spiral galaxies is still unexplained
. it can be either the result of a galactic fountain @xcite and/or of accretion from the intergalactic medium ( e.g. * ? ? ?
the processes involved are probably non - stationary and require a dynamical modeling of the medium above the plane ( see @xcite for a stationary model ) . as a first approach to the problem
we have considered a ballistic ( particle - based ) model for the gas expelled from the plane as a result of star formation activity ( see also * ? ? ?
we have considered a multicomponent potential with stellar and gaseous exponential disks , a r@xmath8 bulge and a double power - law dark matter halo with adjustable flatness ( in analogy to the galactic models of * ? ? ?
a c++ code was written to integrate the orbits of the particles and project them along the line of sight .
the particles are shot vertically with different kick velocities from 0 to a maximum v@xmath9 that can depend on r. for each @xmath10 ( time interval ) the position and velocity of the particle is projected along the line of sight and the integration is carried out until the particle falls back to the plane .
the output of the code is a model cube that can be compared with the data cubes .
these models can be applied to external galaxies viewed at any inclination angle as well as to the milky way ( internal projection ) . here
we show a preliminary application to ngc891 . details and improvements of the model to take into account also non - ballistic effects and gas accretion from the intergalactic medium
will be presented in a forthcoming paper ( fraternali & binney , in preparation ) .
figure 6 shows the total hi map and some representative channel maps for ngc891 , compared with the outputs of two ballistic models .
the two models are for a maximum and a minimum disk fit of the rotation curve with b band m / l ratios of 7 and 1 respectively .
the parameters of the models have been `` tuned '' to reproduce as well as possible the vertical distribution of the gas and the shape of the channel maps .
this figure shows that both ballistic models can qualitatively reproduce the data .
however , there seem to be problems in reproducing the thinness of the channel maps around 300 km s@xmath1 i.e. there is not enough lagging in the halo .
this is not the effect of the potential considering that little differences are visible when considering extreme choices of the potential shape .
this preliminary analysis suggests that the pure ballistic galactic fountain models fail to explain the amount of lagging in galactic halos and that other phenomena such as gas accretion may be important .
new hi observations of the edge - on spiral galaxy ngc891 show the presence of extraplanar neutral gas up to distances of 15 kpc from the plane .
we have used these data to derive , for the first time , a 2d rotation velocity field ( in r and z ) of an edge - on spiral galaxy .
we find that the extra - planar gas is corotating with the gas in the disk up to about 1.3 kpc ; beyond that it rotates more slowly by about 15 km s@xmath1 kpc@xmath1 . dynamical models of ngc891 show that a pure ballistic galactic fountain can qualitatively reproduce the data .
however , problems with missing low angular momentum material suggest that other mechanisms such as accretion may also play a role .
allen , r.j . , sancisi , r. , baldwin , j.e .
1978 , , 62 , 397 barnab m. , ciotti l. , fraternali f. , sancisi r. 2004 , this conference bregman , j.n .
1980 , , 236 , 577 bregman , j.n . , pildis , r.a .
1994 , , 420 , 570 boomsma , r. , oosterloo t. , fraternali f. , van der hulst j.m .
, sancisi r. 2004 , , accepted collins , j.a . , benjamin , r.a . , rand , r.j .
2002 , , 578 , 98 dehnen , w. , binney , j. 2001 , mnras , 294 429 dettmar , r.j .
1990 , , 232 , l15 garcia - burillo , s. , guelin , m. , cernicharo , j. , dahlem , m. 1992 , , 266 , 21 fraternali , f. , oosterloo , t. , sancisi , r. , van moorsel , g. , 2001 , , 562 , l47 fraternali , f. , oosterloo , t. , boomsma r. , swaters r. , sancisi r. 2003 , `` recycling intergalactic and interstellar matter '' , iau , symposium no .
217 , 44 howk , j.c . , savage , b.d .
1999 , , 117 , 2077 van der hulst , t. , sancisi , r. 1988 , , 95 , 1354 hummel , e. , dahlem , m .. van der hulst , j.m . , sukumar , s. 1991 , , 246 , 10 van der kruit , p.c . l. 1981 , , 99 , 298 van der kruit , p.c . ,
searle , l. 1981 , , 95 , 105 matthews , l.d . , wood , k. 2003 , , 593 , 721 oort , j.h .
1970 , , 7 , 381 pildis a.r . , bregman j.n . , schombert j.m .
1994 , , 423 , 190 rand , r.j .
1997 , , 474 , 129 rand , r.j .
2000 , , 537 , l13 rupen , m.p .
1991 , , 102 , 48 1979 , , 74 , 73 shapiro , p.r . , &
field g.b .
1976 , , 205 , 762 sofue , y. , rubin , v. 2001 , ara&a , 39 , 137 swaters , r.a . ,
sancisi , r. , & van der hulst , j.m .
1997 , , 491 , 140 wakker , b.p .
, & van woerden h. 1997 , , 35 , 217 | we present neutral hydrogen observations of the nearby edge - on spiral galaxy ngc891 which show extended extra - planar emission up to distances of 15 kpc from the plane .
3d modeling of the galaxy shows that this emission comes from halo gas rotating more slowly than the gas in the disk .
we derive the rotation curves of the gas above the plane and find a gradient in rotation velocity of @xmath015 km s@xmath1 kpc@xmath1 .
we also present preliminary results of a galactic fountain model applied to ngc891 . |
quantum optical technologies require photon states with specific spectral properties .
for example , quantum information processing using linear optics is based upon the availability of pure - state single - photon wavepackets @xcite .
single - photon wavepackets can be prepared using pair generation by means of spontaneous parametric downconversion ( pdc ) , or by spontaneous four wave mixing ( sfwm ) @xcite . in both cases individual photons
are heralded by the detection of their siblings .
the prepared photons will not normally be in pure states unless special care is taken to remove all correlations in every degree of freedom of the photon pairs , i.e. to make the two - photon state factorable@xcite .
if this is not done , the heralded photons will be in mixed states , and therefore unsuitable for use in quantum logic gates , which rely on hong - ou - mandel ( hom ) interference between independent photons . typically , photon pairs generated in spontaneous processes exhibit significant spectral and spatial correlations due to the energy and momentum conservation constraints that are typical in parametric nonlinear optics .
spatial correlations may be minimized by use of guided - wave configurations , such as those exploiting nonlinear waveguides and optical fibers@xcite .
spectral correlations are , however , more difficult to eliminate .
it is possible to eliminate all correlations in this degree of freedom for photon pairs generated by means of pdc , as first shown in ref .
@xcite , and later extended in @xcite .
this is achieved by the method of group - velocity matching using a broadband pump pulse . in this paper
we generalize this method to the case of sfwm , by engineering the group velocities of the sideband photons using photonic crystal fibers ( pcfs ) .
pcfs are comprised of a solid silica core surrounded by a silica cladding containing a regular array of air holes .
this leads to an exceptionally high core - cladding index contrast , creating strong waveguide dispersion that can be tailored for a wide - range of applications @xcite .
spontaneous four wave mixing occurs in single - mode fibers with a third - order optical nonlinearity . in this process
, two pump photons are scattered from one or two distinct pump fields into a pair of fields , labeled signal and idler , which are spectrally and/or polarization distinct .
the single - mode waveguide geometry leads to the suppression of correlations in the transverse momentum degree of freedom , as well as suppression of mixed spectral - transverse momentum correlations .
thus , by imposing appropriate group velocity matching constraints on the sfwm process in a single - mode fiber , it becomes possible to eliminate correlations in _ all _ degrees of freedom , resulting in factorable two - photon states .
previous experimental work has explored this type of source in a number of geometries including standard single - mode fiber and pcf .
photon number correlations between signal and idler fields were first observed by pumping in the anomalous dispersion region @xcite , although there was a significant background of noise photons generated by raman scattering of the pump light @xcite .
spontaneous raman scattering generates photons shifted to the red of the pump wavelength by up to 50 thz , corresponding to the frequency of an optical phonon in glass .
thermal population of the phonons gives rise to scattering over a broad frequency range , and the process does not require phase matching , so it occurs at all pump wavelengths .
photonic crystal fibers allow the signal and idler fields to be widely separated in frequency from the pump , so that raman noise at the lower - frequency photon wavelength is greatly reduced @xcite .
phase - matching of sfwm with widely separated sidebands is enabled by pumping in the normal dispersion region , which requires the favorable dispersion properties of strongly guided waves @xcite .
these experiments did not , however , address the requirement of purity for generating heralded single - photon states .
in particular , the photons generated in those experiments would require tight spectral filtering to eliminate correlations between signal an idler frequencies prior to heralding .
using such tight filtering ensures high - visibility hom interference between photons from separate sources , but greatly reduces the count rates . to avoid the need for filtering , the sfwm process must produce a state whose two - photon component can be written as the product of the signal state and the idler state , i.e.,@xmath2 , in the photon number basis with @xmath3 indicating signal and idler modes respectively .
we call this property `` factorability , '' which corresponds to the absence of correlations between the frequencies ( and momenta ) of the idler and signal photons . in physical terms
, factorability implies that no information about , say , the idler photon ( apart from its existence ) can be extracted from the detection of the signal photon , or vice - versa . to date , such factorable states have been produced only in pdc at a single wavelength using a particular crystal having special dispersion properties @xcite , and other techniques could in principle be used to extend possible operation wavelengths @xcite .
we show here that factorability can be achieved at a much wider range of wavelengths by using pcf .
in addition to factorability , pcf enables the production of a wide range of spectral correlations in the two - photon component of the state .
two extremes are possible , a spectrally correlated state and a spectrally anti - correlated state .
the former is a resource for quantum - enhanced quantum positioning @xcite , whereas the latter is of importance for applications relying on time - of - arrival differences between two optical modes , such as optical coherence tomography @xcite .
we will show that the anti - correlated case can be made ultra - broadband by tailoring the higher - order dispersion .
thus we see that the ability to engineer the dispersion in pcf leads to a flexible system that can generate states in two widely disparate regimes ( factorable and highly correlated ) , and indeed in a very general class of intermediate regimes . in all these cases ,
the light is produced in well - defined fiber modes , which is convenient for integration with waveguide devices .
this holds the promise of combining many quantum logic gates using integrated optics , leading to the possibility of scalable quantum optical devices .
in this paper we study sfwm in a single - mode optical fiber with a third - order nonlinear susceptibility @xmath0 . in this process , two - photons from pump fields @xmath4 and @xmath5
are jointly annihilated to create a photon pair comprised of one photon in the signal mode , @xmath6 , and one photon in the idler mode @xmath7 .
we assume that all fields propagate in the fundamental spatial mode of the fiber .
this assumption is justified if the fiber core radius is small enough that it only supports the fundamental mode , or alternatively if only this fundamental mode is excited . following a standard perturbative approach @xcite , the two - photon state produced by spontaneous four - wave mixing in an optical fiber of length @xmath8
can be shown to be given by @xcite @xmath9 here , @xmath10 is a constant which represents the generation efficiency ( linearly proportional to the fiber length , electric field amplitude for each of the pump fields and dependent on the relative polarizations of the pump and created pair fields ) and @xmath11 is the joint spectral amplitude function ( jsa ) , which describes the spectral entanglement properties of the generated photon pair @xmath12 \mbox{exp } \left [ i\frac{l}{2}\delta k\left ( \omega^{\prime},\omega_{s},\omega_{i}\right ) \right ] , \end{aligned}\ ] ] which is given in terms of the pump spectral amplitudes @xmath13 , and the phase mismatch function @xmath14 , that in the case where the two pumps , signal and idler are co - polarized , is given by @xmath15 which includes a self / cross - phase modulation contribution for the two pumps with peak powers @xmath16 and @xmath17 , characterized by the nonlinear parameters @xmath18 and @xmath19 , which depend on specific fiber used and pump wavelength@xcite .
the energy conservation constraint is apparent in the argument of the second term of the phase mismatch ( see eq.([eq : delk ] ) ) .
a `` factorable '' state is defined to be a state for which @xmath11 is equal to a product of two functions , @xmath20 , where the functions @xmath21 and @xmath22 depend only on the signal and idler frequencies respectively . by making use of a linear approximation for the phase mismatch ,
in addition to modeling @xmath23 as gaussian functions with bandwidth @xmath24 respectively , it is possible to obtain an expression for the joint spectral amplitude in closed analytical form .
expanding @xmath25 in a first - order taylor series about frequencies @xmath26 for which perfect phase - matching is attained ( where @xmath27 ) , and defining the detunings @xmath28 and @xmath29 , the approximate phase mismatch @xmath30 is defined by @xmath31 where @xmath32 , given by eq.([eq : delk ] ) evaluated at the frequencies @xmath26 , must vanish to guarantee phase - matching at these center frequencies .
the coefficients @xmath33 are given by @xmath34 , where @xmath35 represent group - velocity mismatch terms between the pump centered at frequency @xmath36 and the generated photon centered at the frequency @xmath26 , and @xmath37 is the group velocity mismatch between the two pumps @xmath38 , \nonumber\\ \tau_{p } = l\left [ k_{1}^{(1)}\left ( \omega_{1}^{0}\right ) - k_{2}^{(1)}\left ( \omega_{2}^{0}\right ) \right ] , \end{aligned}\ ] ] written in terms of @xmath39 . note that this approach requires _ a priori _ knowledge , for given pump fields , of the signal and idler frequencies ( @xmath40 and @xmath41 ) at which perfect phase - matching is achieved .
these frequencies can be determined by solving ( for example numerically ) the condition @xmath42 ( see eq.([eq : delk ] ) ) .
it can be shown that within the linear approximation , the integral in eq .
( [ eq : jsa ] ) can be carried out analytically , yielding @xmath43 where the pump envelope function @xmath44 is derived from the pump spectral amplitudes for the two individual pump fields through the integral in eq .
( [ eq : jsa ] ) , and is given by @xmath45 , \ ] ] and where @xmath46 describes the phase - matching properties in the fiber .
for degenerate pumps ( where @xmath47 ) , it may be shown that the phase - matching function is given by @xmath48 \exp\left [ i\frac{l\delta k_{lin}}{2 } \right ] , \ ] ] where @xmath49 is given in eq .
( [ eq : ldelk ] ) with @xmath50 . for non - degenerate pumps , @xmath51 and @xmath52 , with @xmath53 , \end{aligned}\ ] ] where erf(@xmath54 ) is the error function , the parameter @xmath55 is defined as @xmath56 , and @xmath57 is a normalization coefficient .
in the present paper , we concentrate on the important class of factorable states for which @xmath58 . for a fiber characterized by @xmath59 m , @xmath60 and @xmath61 cm .
( b ) phasematching function @xmath46 ; for a relatively small region of \{@xmath62 } space the phase - matching function contours are essentially straight lines , with slope @xmath63 ( with @xmath33 given by eq .
( [ eq : taus ] ) ) . ( c )
resulting joint spectral intensity .
, width=13 ]
our analysis focuses on pcfs consisting of a fused silica core surrounded by silica cladding with a pattern of air holes which remains constant along the fiber length .
this mixture of air and and glass in the cladding results in an average refractive index that is considerably lower than that of the core , providing a high dielectric contrast , resulting in strong optical confinement .
this leads to high peak irradiances even for modest input powers , which enhances nonlinear optical effects such as sfwm .
in addition , the dispersion characteristics of the pcfs can be engineered by variations of the distribution , size and shape of the air holes surrounding the core . in particular
, it becomes possible to choose the zero dispersion wavelength(s ) ( zdw ) , to tailor the fwm phase - matching properties @xcite and to design fibers approaching endlessly single - mode behavior @xcite . )
contour for sfwm in the degenerate pump case .
colored background : phase - matching orientation angle .
black , dashed line : symmetric group velocity matching ( gvm ) contour . along the phase - matching contour
we have indicated particular orientation angles of interest .
( b ) close up , near the lower zero group velocity dispersion frequency.,width=453 ] the fwm phase - matching properties are determined by the fundamental mode propagation constant , given in terms of the effective refractive index @xmath64 by @xmath65 .
we adopt a step - index model , where the core has radius @xmath66 , its index is that of fused silica @xmath67 , and the cladding index is calculated as @xmath68 , where @xmath69 is the air - filling fraction .
this fiber dispersion model has been shown to be accurate for @xmath69 from @xmath70 to @xmath71 according to ref .
@xcite . in the context of our work
, this model permits a straightforward exploration of the spectral entanglement properties in @xmath72 parameter space .
the structure of the sfwm two - photon state , in the degenerate - pump , co - polarized case ( for which @xmath73 ) is illustrated in fig .
[ fig : pmillus ] , for a specific fiber with @xmath59 m , @xmath60 and length @xmath61 cm . fig .
[ fig : pmillus](a ) shows the pump envelope function plotted as a function of @xmath74 and @xmath75 ( see eq.([eq : pe ] ) ) where we have assumed that the pump is centered at @xmath76 nm , has a bandwidth of @xmath77 nm , the incident power is @xmath78w and @xmath79 .
[ fig : pmillus](b ) shows the phasematching function ( see eq.([eq : phinsnii ] ) ) for this choice of parameters . note that the phasematched region forms a strip on @xmath80 space , oriented at an angle @xmath81 with respect to the @xmath74 axis .
[ fig : pmillus](c ) shows the resulting joint spectral intensity @xmath82 ( see eq.([eq : jsanus ] ) ) .
it is apparent from fig .
[ fig : pmillus ] that the properties of the two photon state are determined by the i ) relative orientations and ii ) widths of the strips representing the phasematching and pump envelope functions . in this paper
we explore how the interplay of the various design parameters can lead to two - photon states with engineered spectral entanglement properties . a plot of the perfect phase - matching contour ( @xmath83 ) for degenerate pumps versus @xmath74 and @xmath84 gives for each pump frequency the expected signal and idler central frequencies @xmath40 and @xmath41 .
such a phase - matching contour is illustrated in fig .
[ fig : contordk ] for @xmath85 m and @xmath86 , with pump power @xmath87w and @xmath88 , where the generated frequencies are expressed as detunings from the pump frequency @xmath89 .
note that energy conservation implies that @xmath90 .
pcfs often have two zdws , where one can be as low as 500 nm ( in comparison with 1270 nm for bulk silica ) .
the phase - matching contours take the form of closed loops , with inner branches near the pump frequency and outer branches that can be hundreds of nanometers from the pump wavelength .
note that four wave mixing relying on outer - branch phasematching has been observed in previous work , in the context of classical non - linear optics @xcite . for this specific fiber geometry ,
a continuous pump wavelength range of approximately @xmath91 nm exists , in which parametric generation can be observed , almost completely contained between the two zdws ( @xmath92 m and @xmath93 m ) .
the power - induced phase modulation terms in @xmath94 split the trivial @xmath95 branch , leading to frequency - distinct inner - branch signal / idler frequencies close to the pump ( used for sfwm in @xcite ) , whereas the outer branch exhibits comparatively little dependence on pump power @xcite . in what follows
we will concentrate on phase - matching configurations where the signal and idler frequencies are sufficiently removed from the pump frequency , or where the pump is orthogonally polarized to the signal and idler photons , so as to avoid the generation of background photons by spontaneous raman scattering at the lower - frequency photon of the pair @xcite . from eq .
( [ eq : pe ] ) it is clear that the pump envelope function in @xmath80 space has contours of equal amplitude which have negative unit slope in all of @xmath80 space .
in contrast , the contours of the phase - matching function @xmath96 are characterized by a slope in @xmath80 space given by @xmath63 ( with @xmath33 given according to eq .
( [ eq : taus ] ) ) .
the relationship between the slope @xmath97 of the curve at a point in \{@xmath98 } space to the resulting phasematching function in \{@xmath62 } space is @xmath99 .
for example , @xmath100 corresponds to zero slope on the contour in fig . [
fig : contordk](a ) or @xmath101 . the colored background in fig .
[ fig : contordk ] indicates the slope of the phase - matching contour , or orientation angle , ranging from @xmath102 to @xmath103 , indicated in blue and red respectively .
the type of spectral correlations observed in a sfwm two - photon state is determined in part by the slope @xmath104 of the phase - matching contour . if the phase - matching contour is given by a closed loop ( which is true for most fibers , i.e. range of values of @xmath72 , of interest ) , all phase - matching orientation angles @xmath104 are possible , controlled by the pump frequency .
thus , for certain relative orientations and widths of these two functions , it becomes possible to generate factorable two - photon states .
it can be shown that a factorable state is possible if @xmath105 among those states which fulfil eq.([eq : codp_i ] ) those exhibiting a phasematching angle of @xmath106 , or @xmath107 , are of particular interest .
for these states , in the degenerate pumps case , a factorable , symmetric state is guaranteed if @xmath108 where @xmath109 .
the condition in eq .
( [ eq : codp_i ] ) constrains the group velocities : either @xmath110 , or @xmath111 must be satisfied . the condition in eq .
( [ eq : codp_ii ] ) constrains the bandwidths .
thus , the region in @xmath80 space in which factorability is possible is bounded by the conditions @xmath112 and @xmath113 . note that in the case of pdc ( in second - order nonlinear crystals ) , it is in general terms challenging to obey eq .
( [ eq : codp_i ] ) , for it implies that the pump , say with frequency @xmath84 , must propagate at a higher group velocity than one of the generated photons at @xmath114 , which can be interpreted as anomalous group - velocity dispersion . in practice , this can be achieved for certain materials , within a restricted spectral region ( usually with pdc in the infrared ) in a type - ii process , where the polarization of the pump is orthogonal to that of one of the generated photons @xcite . in the case of sfwm , because the frequencies ( in the degenerate pump regime ) will in general obey @xmath115 or @xmath116 , no such anomalous group - velocity dispersion is needed , making it much more straightforward to fulfill the group - velocity matching conditions required for factorable photon pair generation .
once the dispersion relation for a given fiber geometry has been determined , the next step in designing a factorable photon pair source consists of identifying the pump wavelengths that satisfy the required group - velocity matching , amongst those which also satisfy phase - matching . for this purpose
, we refer to the colored background in fig .
[ fig : contordk ] .
each color corresponds to a different phase - matching orientation angle @xmath104 . in particular ,
symmetric factorable two - photon states , for which the signal and idler photons have identical spectral widths , are possible if @xmath117 . in this case symmetric group - velocity matching @xmath118 ( or equivalently @xmath119 ) is attained , and the phase - matching contours are oriented so that its contours have unit slope .
the frequency values that fulfill this condition are represented by the dashed line .
the pump frequency that permits symmetric factorable states can be determined from the intersection of the phase - matching contour with the group - velocity matching contour . similarly , asymmetric factorable two - photon states , for which the signal and idler photons have greatly different spectral widths , are possible if @xmath120 or @xmath121 .
this is the case of asymmetric group - velocity matching , for which @xmath122 or @xmath123 ( or equivalently @xmath112 or @xmath124 ) is required . in this case
the phase - matching contours are oriented parallel to the @xmath74 or @xmath75 axes .
in addition , eq .
( [ eq : codp_ii ] ) leads to the condition that @xmath125 ( for @xmath112 ) or @xmath126 ( for @xmath124 ) .
both symmetric and asymmetric gvm are possible using a number of different configurations .
we provide examples of each , exploiting co - polarized as well as cross - polarized sfwm , in the following sections .
note however , that the examples to be shown do not include all possible source designs .
the synthesis of a symmetric , factorable state is illustrated in fig .
[ fig : estsime ] for the same fiber geometry as described above ( @xmath127 m and @xmath86 ) .
figure [ fig : estsime](a ) shows the pump envelope function , for a single pump centered at @xmath128 nm with a relatively narrow bandwidth of @xmath70 nm . figure [ fig : estsime](b ) shows the phase - matching function assuming a fiber length of @xmath129 cm , while fig .
[ fig : estsime](c ) shows the joint spectral intensity @xmath82 , exhibiting an essentially factorable character . for comparison ,
[ fig : estsime](d ) shows the joint spectral intensity obtained by numerical integration of eq .
( [ eq : jsa ] ) using the full dispersion ( rather than relying on a linear approximation ) , revealing that this case the linear approximation of the phase - matching condition is in fact an excellent approximation .
this source leads to a numerically - obtained state purity defined as tr[@xmath130 , where @xmath131 is the reduced density operator for the signal state , of @xmath132 .
in fact , the departure from ideal purity is mainly due to sidelobes ( related to the sinc function ) displaced from the central portion of joint spectral intensity shown in figs . [
fig : estsime](c ) and ( d ) .
the purity can be increased by filtering out these sidelobes , which in general contain a small fraction of the total flux .
specifically , for the example shown in fig .
[ fig : estsime ] , two separate narrowband rectangular - profile spectral filters for the signal and idler modes with equal frequency bandwidth of @xmath133 ( corresponding to wavelength widths @xmath134 nm and @xmath135 nm ) increases the purity to @xmath136 , while reducing the flux by @xmath137 .
( @xmath138 m and @xmath86 ) where the pump central wavelength ( @xmath139 m ) is obtained by imposing simultaneous phase - matching and group - velocity matching .
we consider a pump bandwidth of @xmath70 nm and a fiber length of @xmath140 m .
the function values are normalized such that white @xmath141 and black @xmath142 .
( a ) pump envelope function @xmath143 .
( b ) phase - matching function @xmath144 .
( c ) analytic jsi , obtained with approximation from eq .
( [ eq : ldelk ] ) .
( d ) jsi obtained by numerical integration of eq .
( [ eq : jsa ] ) .,width=434 ] as pointed out already , one of the advantages of generating factorable , symmetric photon pairs through sfwm rather than pdc in a second - order nonlinear crystal , is that it presents greater ( but still not arbitrary ) choice of the pump , signal and idler frequencies .
thus , for the example shown above , we have chosen the pump so that it is compatible with , say , a ti : sapphire laser .
one of the generated photons is centered at @xmath145 nm , which corresponds to the standard telecommunications band and could be detected with an ingaas - based avalanche photo - detector .
the conjugate generated photon is centered at @xmath146 nm , which can be detected with a silicon - based avalanche photodiode ( though with non - optimal efficiency , since this wavelength is relatively far from the sensitivity peak of such detectors ) .
a natural application for such a source would be the heralding of pure , single photons for transmission over standard optical telecommunication fibers .
the use of non - degenerate pumps represents an additional degree of freedom , not available for pdc in second - order crystals , which can be exploited for state engineering purposes .
the two pumps may differ both in terms of their central frequencies ( @xmath147 and @xmath36 ) and their bandwidths ( @xmath148 and @xmath149 ) ; in what follows we exploit both of these aspects of non - degeneracy . the orientation of the phase - matching function ( see eqs .
( [ eq : ldelk ] ) and ( [ pmf1 ] ) ) in \{@xmath62 } space is determined by the angle @xmath150 .
note that the orientation of the phase - matching function depends on @xmath148 and @xmath149 ( through @xmath151 and @xmath152 ) , while the shape of its profile and its width also depend on @xmath148 and @xmath149 ( through parameter @xmath55 ) . a considerable simplification results by imposing the condition @xmath153 .
in this case , @xmath33 reduces to the corresponding degenerate - pump values @xmath35 ( with @xmath154 ) .
furthermore , parameter @xmath55 reduces to @xmath155 , so that the phase - matching function width exhibits no dependence on the broader pump bandwidth ( @xmath149 ) , while the effective pump envelope function depends only on @xmath149 .
this de - coupling of @xmath148 and @xmath149 translates into a more straightforward exploration of possible fiber geometries for the generation of states with specific spectral entanglement properties . a phase - matching diagram similar to that presented for degenerate pumps ( see fig .
[ fig : contordk ] ) , may be prepared by maintaining one of the two pump frequencies fixed ( in this case pump @xmath156 ) , where it is convenient to express the generated frequencies as detunings from the _ mean _ pump frequency i.e. @xmath157 ; note that energy conservation implies that @xmath90 .
figure [ cont_ndp ] illustrates the phase - matching properties for a specific non - degenerate geometry , with @xmath158 m , @xmath159 and @xmath160rad / sec ( that is , @xmath161 nm ) .
the black solid curve represents the resulting phase - matching contour .
the two straight lines represent trivial phase - matching branches ( in which the created photons are degenerate with the two pumps ) , while the loop represents the non - trivial phase - matching branch . depending on the fiber geometry , the non - trivial branch may become large enough to overlap the trivial branches , or may shrink down to a single point . for the former case ,
if self - phase modulation becomes appreciable , this contour splits into three distinct loops ( rather than two as for degenerate pumps ) .
the colored background represents the orientation angle @xmath104 ranging from @xmath162 in blue to @xmath163 in red . ) contour for sfwm in the non - degenerate pump regime .
colored background : phase - matching orientation angle .
black , dotted line : frequencies that satisfy the group - velocity matching condition ( see eq .
( [ eq : gvmcond ] ) ) . along the phase - matching contour
we have indicated particular angles of orientation of interest .
, width=434 ] a factorable , symmetric two - photon state is produced if i ) @xmath100 which implies @xmath107 and ii ) the spectral widths of the phase - matching function @xmath164 and the pump envelope function @xmath143 are equal to each other . from eqs .
( [ eq : taus ] ) it is straightforward to show that the first condition is satisfied if the four fields satisfy @xmath165 this represents a generalization of the equivalent group - velocity matching condition for degenerate pumps ( see eq .
[ eq : codp_i ] , and the text which follows ) . if @xmath166 , the right - hand side of eq .
( [ eq : gvmcond ] ) vanishes , leading to a condition that is identical in form to that obtained for degenerate pumps ( where the broader pump now plays the role of the degenerate pump ) . in fig .
[ cont_ndp ] , pairs @xmath167 that satisfy this generalized group - velocity matching condition are represented by a dashed line .
note that the position and shape of this contour in general depend on @xmath148 and @xmath149 ; however , in the case @xmath166 , the contour becomes decoupled from these bandwidths .
intersection points between the two contours determine the center frequency for pump field @xmath77 that satisfies the requisite group - velocity matching , in addition to phase - matching .
the values of @xmath66 and @xmath69 are selected so that @xmath106 is satisfied _ and _ so that the broadband pump frequency @xmath147 takes a certain desired value ( in this case , @xmath168 m ) .
this leads to the generated signal and idler wavelengths @xmath169 nm and @xmath170 nm . the fiber length , along with the two bandwidths ( subject to @xmath171 ) are selected so as to match the widths of the pump envelope and phasematching functions .
for the specific geometry shown , these values are @xmath172 cm , while @xmath148 and @xmath149 are given in terms of the corresponding fwhm bandwidths : @xmath173 nm and @xmath174 nm . an important motivation for exploiting pump non - degeneracy is that it permits sfwm geometries with pumps which are sufficiently distinct from the signal and idler photons , and yet where the latter are in relative proximity .
this is important on the one hand for the suppression of contamination from spontaneous raman scattering and on the other hand to ensure that both generated photons can be detected with available efficient single - photon detectors .
the pump wavelengths were selected so that they can be obtained from a modelocked cr : forsterite laser and its second harmonic , while the generated light is within the detection bandwidth of silicon avalanche photo - diodes .
figure [ ndpsim](a ) shows the pump envelope function @xmath143 .
figure [ ndpsim](b ) shows the phase - matching function @xmath175 ( see eq.([pmf1 ] ) ) for which @xmath176 .
figure [ ndpsim](c ) shows the resulting joint spectral intensity which exhibits a factorable character , obtained using the linear - dispersion approximation ( see eq .
( [ eq : ldelk ] ) ) .
for comparison , fig .
[ ndpsim](d ) shows the joint spectral intensity obtained by numerical integration of eq.([eq : jsa ] ) , exhibiting excellent agreement with fig .
[ ndpsim](c ) .
this source leads to a numerically - obtained state purity of @xmath177 . as in the case of a factorable symmetric state obtained with degenerate pumps ( see fig . [
fig : estsime ] ) , the purity can be increased by filtering out the sidelobes in the joint spectral intensity at small cost in terms of collected flux .
two separate narrowband rectangular - profile spectral filters for the signal and idler modes with equal frequency bandwidth of @xmath178 ( corresponding to wavelength widths @xmath179 nm and @xmath180 nm ) increases the purity to @xmath181 , while reducing the flux by @xmath182 .
obtained for the fiber geometry assumed for fig .
[ cont_ndp ] .
( a ) pump envelope function @xmath143 .
( b ) phase - matching function @xmath164 .
( c ) analytic jsi .
( d ) jsi obtained by numerical integration of eq .
( [ eq : jsa ] ) . ,
birefringence is an additional degree of freedom that we can use to satisfy the conditions necessary for generating factorable two - photon states .
typical hexagonal pattern pcf can exhibit significant birefringence due to slight asymmetries in the hole pattern or internal stresses in the glass .
birefringence can also be purposefully added to the structure through defects such as two larger holes or elliptical cores . in pcfs ,
the birefringence typically ranges from @xmath183 @xmath184 , for normal pcf , to @xmath183 @xmath185 @xcite for polarization maintaining pcfs .
conservation of angular momentum restricts the types of fwm processes that can occur in a fiber , with the result that the polarizations of the four fields must appear in pairs in the process @xcite .
it follows that there are three distinct pair generation processes that can occur in fibers : @xmath186 , @xmath187 , @xmath188 , where @xmath189 and @xmath190 can be any two linear orthogonal polarizations .
photon pair generation with frequency separated pumps in the @xmath187 process was analyzed in 2004 @xcite and demonstrated @xcite in 2005 .
in contrast , we limit our discussion to a single pump pulse polarized along one fiber axis ( e.g. @xmath188 , @xmath191 ) . the first realization of birefringent fwm was a demonstration of this process in regular fibers @xcite .
it was demonstrated in pcfs in 2006 @xcite .
single - photon detectors operating in the visible region of the spectrum are generally less costly and have higher quantum efficiencies and lower noise than single - photon detectors operating in the near - infrared region .
therefore for some applications it is desirable for both photons from a sfwm pair to be produced in the visible region .
this limits the usefulness for these applications of pcfs which have outer - branch solutions far separated from the pump frequency , as shown in fig .
[ fig : contordk](a ) , as one of the photons is generated in the infrared .
on the other hand , the photons should be at least 50 thz from the pump to reduce background from spontaneous raman scattering .
this limits the usefulness of the inner branch solutions , for which the generated photons are typically less then 5 thz from the pump @xcite .
the benefit of birefringence is that it creates new factorable solutions inside the visible region yet still separated from the pump by more than the raman bandwidth . unlike the nondegenerate case , it accomplishes this without the experimental complexity of two pumps .
this can be seen by considering the relevant phase - matching equation , @xmath192 where the power - induced term is 1/3 of that in the co - polarized case @xcite .
some insight can be gained by separating the birefringent contribution to @xmath94 from the individual wavevectors as in eq .
( [ approxbipm ] ) .
it was experimentally observed in @xcite that @xmath193 is approximately independent of frequency and we note that @xmath84 is relatively constant in comparison to @xmath194 . consequently , the birefringent contribution to the wavevector mismatch functions much like the power phase - modulation term in eq .
( [ eq : delk ] ) , introducing in the @xmath195 curve a splitting of the @xmath196 solution ( creating the inner branch in fig .
[ fig : contordk ] ) . unlike the power term
, the birefringence can create a large splitting and can be either positive or negative .
( thick ) and group - velocity matching @xmath197 ( thin ) curves ( @xmath198 , @xmath199 zdws @xmath200 nm , @xmath201 nm similar to nl-1.7 - 790 from crystal fiber ) . with power @xmath202 ,
three pairs of curves are plotted with @xmath203 @xmath204 respectively resulting in a birefringence of @xmath205 ( red , green , blue ) .
points a through f bound regions in which factorability is possible .
b ) full dispersion numerical jsis for the asymmetric states corresponding to points d ( @xmath206 @xmath207 nm and @xmath208 @xmath209 nm , @xmath210 @xmath211 nm , and @xmath61 m ) and f ( @xmath121 , @xmath212 @xmath213 nm , @xmath208 @xmath214 @xmath215 nm , @xmath210 @xmath211 nm , and @xmath61 m).,width=453 ] we begin by extending the step index model reviewed in section 3 to birefringent fiber . in this extension , each transverse axis of the fiber is modeled as a separate fiber .
the axes have the same air - filling fraction parameter @xmath69 but different core diameters @xmath216 and @xmath217 simulating the birefringence @xcite .
we use this model for all the figures in this section .
we also neglect the power term in eq .
( [ bipm ] ) since it is typically much smaller than the birefringent contribution .
although this model is accurate only for small birefringence where the fiber asymmetry is small , it allows us to predict universal properties of the sfwm across a broad range of configurations . in fig .
[ birefring ] , the phase - matching ( @xmath218 ) and group - velocity matching ( @xmath197 ) curves are plotted for a pcf fiber with zdws at 790 nm and 1404 nm , where @xmath219 m and @xmath198 .
three pairs of curves are plotted for @xmath220 @xmath1 , resulting in a birefringence of @xmath221 ( where @xmath222 ) .
the corresponding shift in the zdw is -0.07 , 0 , 0.07 nm and @xmath223 ps / km , respectively .
the segments of the @xmath224 curve bounded by intersections with the @xmath197 curve are regions where factorability is possible . for both positive and negative @xmath193
the outer branch intersections remain approximately the same as in the copolarized degenerate case . in addition ,
negative @xmath193 creates two intersections ( points c and d ) in between the two zdws . in between these points
factorability is possible . likewise
, positive @xmath193 also creates two intersections ( points e and f ) . however , now it is outside the region bounded by the intersections that factorability is possible .
the splitting of the phase - matching ( @xmath225 ) solution from the @xmath226 line in the @xmath227 case is similar to that from the power - induced phase modulation term @xmath228 in the phase matching equation eq .
( [ eq : delk ] ) , but is an order of magnitude larger . unlike power - induced inner branch fwm , birefringence allows for phase - matching beyond the raman peak and is relatively insensitive to pump laser power fluctuations .
in contrast , the @xmath229 case is qualitatively different from either of these since @xmath228 is always positive in silica .
moreover , it allows for factorable state generation over an unprecedented pump wavelength range , essentially anywhere outside the zdws .
are plotted as a function of the birefringence @xmath230 thus , the solid dark blue curves give the pump wavelength @xmath231 of asymmetric factorable solutions . the corresponding idler and signal wavelengths @xmath208 and @xmath212
are given by the magenta dashed curves ( for points d and f ) and light - blue dash - dotted curves ( for points c and e ) .
the shaded regions indicate the range in which factorable states can be generated provide the pump bandwidth is matched to the fiber length .
the fiber parameters are the same as in fig .
[ birefring].,width=453 ] figure [ birefring](b ) shows the joint spectral intensity in two cases that lead to high degrees of factorability .
factorability corresponds in both cases to the absence of correlation between signal and idler frequencies , as can be seen in the shape of the jsi . in either case , if the idler photon is detected ( without spectral filtering ) , then a nearly pure - state signal photon is heralded .
numerically calculated heralded photon purity for the cases at point f and point d are 0.988 and 0.993 , respectively . in fig .
[ gvmsol ] , the intersection of the @xmath195 and @xmath197 curves is plotted as a function of the @xmath193 at 800 nm .
thus for each @xmath193 the figure indicates the pump wavelength @xmath231 that creates an asymmetric factorable state ( @xmath120 or @xmath232 ) , as well as the center wavelengths @xmath212 and @xmath208 of the generated photons . at these points
, the pump bandwidth must satisfy condition @xmath233 but is otherwise arbitrary .
the shaded regions indicate the @xmath231 range in which factorability is possible ( including the symmetric state ) if the pump bandwidth is set according to eq .
( [ eq : codp_ii ] ) .
it has been pointed out in the fwm literature that birefringent fwm has the added benefit of reducing raman contamination in the signal and idler @xcite .
recently this has also been forwarded as a strategy to reduce background for pair - generation @xcite .
raman gain @xmath234 is reduced by as much as an order of magnitude in the axis orthogonal to the strong pump pulse @xcite .
however , the relevant @xmath0 for birefringent pair production is also reduced by a factor of three compared to co - polarized fwm @xcite . tripling the power @xmath235 to compensate , ref .
@xcite studied the pair - correlation produced by the fiber , @xmath236 where @xmath237 is the intensity of signal or idler mode .
they found that at the raman peak , @xmath238 was 7 for co - polarized sfwm as compared to 60 for cross - polarized sfwm , indicating that the cross - polarized case can be a higher quality photon - pair source .
the flexibility afforded by pcfs in engineering the spectral correlations of photon pairs permits the generation of a wide class of states .
for instance , this may be illustrated by considering the generation of quantum states in which the photon pair component has an extraordinarily high degree of frequency anti - correlation@xcite .
the photons individually have an ultrabroad bandwidth , yet are very tightly anticorrelated in frequency , and highly correlated in time ( to the level of a single optical cycle ) .
we have identified conditions for which ultra - broadband two - photon states can be generated in the degenerate pump regime by sfwm .
such ultrabroad states have recently been generated by pdc @xcite . as pointed out above , for fiber geometries where the phase - matching diagram occurs in the form of loops ( which is true for most cases of interest ) , it is possible to obtain an arbitrary phase - matching orientation on @xmath80 space . by imposing constraints on higher dispersive orders ,
it becomes possible to generate states with yet different character .
in particular , in analogy with the generation of ultra - broadband two - photon states via pdc @xcite , it can be shown that in the case of sfwm , the bandwidth of the generated light can be made particularly large if the fourth - order @xmath239 and second order @xmath240 coefficients vanish simultaneously . in this case
, a degenerate pump at the zero - group - velocity - dispersion frequency can lead to a state with a remarkably broad bandwidth ( hundreds of nanometers ) , even for a nearly monochromatic pump .
for example , for the pcf described previously with @xmath85 m and @xmath86 , whose phase - matching curve is shown in fig .
[ fig : contordk ] , pumping at a frequency around @xmath241 rad / sec ( wavelength @xmath242 nm ) , where the curve is nearly vertical , would create photons with a bandwidth of around @xmath243 rad / sec ( @xmath244 nm ) centered at the pump frequency , depending on the bandwidth of the pump .
previously , light produced by co - polarized pumps was shown to have a bandwidth of 50 nm for cw pumps @xcite and 100 nm for pulsed pumps @xcite . for cross - polarized cw pumps broadband light of 34 nm
was produced @xcite .
we have studied theoretically the spontaneous four wave mixing process in single mode photonic crystal fiber . we have shown that it is possible to design a source so as to yield photon pairs with a broad class of spectrally engineered properties , including factorable and ultrabroadband states .
these results are achieved by adjusting the source configuration , including the dispersion ( determined by the core radius and air filling fraction ) and birefringence of the fiber and the pump frequency or frequencies , bandwidth(s ) and polarization(s ) .
the flexibility of the two - photon state engineering arises from the interplay of this large number of control parameters .
factorable signal - idler pair generation allows the production of heralded , high purity single - photon wavepackets , with minimal or no need for spectral filtering .
factorable state generation is possible if certain group - velocity matching conditions are satisfied , in addition to standard phase - matching .
most fibers ( all those for which the phase - matching curve occurs in loops ) can achieve all appropriate group - velocity matching conditions by suitable choice of the pump wavelength .
this allows one to realize all possible orientations of the joint spectral amplitude of the generated photon pairs .
the size of the phase - matching loop is largely determined by the dispersion of the fiber , hence the utility of pcfs .
therefore we find that for specific choices of pump frequency most fiber designs can yield factorable states .
this is in contrast with parametric downconversion in second - order nonlinear crystals where such conditions are met at best at isolated frequencies for certain materials , or using more complicated microstructures @xcite .
an important design consideration is the possible contamination of the frequency sideband modes by photons from spontaneous raman scattering .
this problem can be circumvented in part by cooling the fiber to cryogenic temperatures , thereby suppressing the raman scattering @xcite .
we have shown how to circumvent the problem by exploiting a phase - matching configuration where the signal and the idler frequencies are widely separated from the pump frequency ( see fig .
[ fig : contordk ] ) .
if desired , the phase matching in non - birefringent fiber can be engineered to create photon pairs having one photon in the infrared and one photon in the visible .
alternative approaches also discussed in this paper are cross - polarized fwm in birefringent fiber , where the pump and generated photons have orthogonal polarizations , and the use of non - degenerate pumps in a co - polarized regime .
we have shown that these approaches lead to sources capable of generating factorable photon pairs where the signal and idler photons are both in the visible , and at the same time are both sufficiently separated from the pump frequency to avoid raman contamination .
we expect these results to be useful in the design of sources for practical implementations of quantum information processing technologies .
kgp , and rrr acknowledge conacyt - mxico for partial funding through project no . 46492 , and abu acknowledges support from conacyt - mexico through grant 46370 .
iaw , jl and oc were supported by the epsrc ( uk ) through the qip irc ( gr / s82716/01 ) and through project ep
/ c013840/1 and by the european commission under the integrated project qubit applications ( qap ) funded by the ist directorate as contract number 015848 .
hjm and mgr were supported by the national science foundation grants numbers ecs-0621723 and phy-0554842 . | we study theoretically the generation of photon pairs by spontaneous four - wave mixing ( sfwm ) in photonic crystal optical fiber .
we show that it is possible to engineer two - photon states with specific spectral correlation ( `` entanglement '' ) properties suitable for quantum information processing applications .
we focus on the case exhibiting no spectral correlations in the two - photon component of the state , which we call factorability , and which allows heralding of single - photon pure - state wave packets without the need for spectral post filtering .
we show that spontaneous four wave mixing exhibits a remarkable flexibility , permitting a wider class of two - photon states , including ultra - broadband , highly - anticorrelated states .
99 see , for example , the review by p. kok , w. j. munro , k. nemoto , t. c. ralph , j. p. dowling and g. j. milburn , `` linear optical quantum computing with photonic qubits , '' rev .
mod .
phys . * 79 * , 135174 ( 2007 ) .
s. e. harris , m. k. oshman , and r. l. byer , `` observation of tunable optical parametric fluorescence , '' phys .
rev .
lett .
* 18 * , 732734 ( 1967 ) .
a. b. uren , c. silberhorn , k. banaszek , i. a. walmsley , r. erdmann , w. p. grice and m. g. raymer , `` generation of pure - state single - photon wavepackets by conditional preparation based on spontaneous parametric downconversion , '' laser phys . * 15 * , 146161 ( 2005 ) .
m. g. raymer , j. noh , k. banaszek and i. a. walmsley , `` pure - state single - photon wave - packet generation by parmametric down - conversion in a distributed microcavity , '' phys . rev . a * 72 * , 023825 ( 2005 ) .
a. b. uren , c. silberhorn , k. banaszek and i.a .
walmsley , `` efficient conditional preparation of high - fidelity single photon states for fiber - optic quantum networks , '' phys .
rev .
lett . * 93 * , 093601 ( 2004 ) .
k. banaszek , a. b. uren and i. a. walmsley , `` generation of correlated photons in controlled spatial modes by downconversion in nonlinear waveguides , '' opt . lett .
* 26 * , 13671369 ( 2001 ) .
j. fan and a. migdall , `` a broadband high spectral brightness fiber - based two - photon source , '' opt . express * 15 * , 29152920 ( 2007 ) .
j. rarity , j. fulconis , j. duligall , w. wadsworth , and p. st .
j. russell , `` photonic crystal fiber source of correlated photon pairs , '' opt .
express * 13 * , 534544 ( 2005 ) .
j. fan and a. migdall , `` generation of cross - polarized photon pairs in a microstructure fiber with frequency - conjugate laser pump pulses , '' opt .
express * 13 * , 57775782 ( 2005 ) .
x. li , j. chen , p. voss , j. sharping , and p. kumar , `` all - fiber photon - pair source for quantum communications : improved generation of correlated photons , '' opt .
express * 12 * , 37373744 ( 2004 ) .
w. p. grice , a. b. uren and i. a. walmsley , `` eliminating frequency and space - time correlations in multiphoton states , '' phys . rev . a * 64 * , 063815 ( 2001 ) . p. russell , `` photonic crystal fiber , '' science * 299 * , 358362 ( 2003 ) . m. fiorentino , p. l. voss , j. e. sharping and p. kumar , `` all - fiber photon - pair source for quantum communications , '' ieee photon . technol
. lett .
* 14 * , 983985 ( 2002 ) .
r. h. stolen , `` fundamentals of raman amplification in fibers , '' in _ raman amplifiers for telecommunications 1 _ , edited by m. n. islam ( springer , 2003 ) , pp . 3559 .
r. jiang , r. saperstein , n. alic , m. nezhad , c. j. mckinstrie , j. ford , s. fainman and s. radic , `` parametric wavelength conversion from conventional near - infrared to visible band , '' ieee photon . technol . lett . * 18 * , 24452447 ( 2006 ) .
j. d. harvey , r. leonhardt , s. coen , g. k. l. wong , j. c. knight , w. j. wadsworth and p. st .
j. russell , `` scalar modulation instability in the normal dispersion regime by use of a photonic crystal fiber , '' opt .
lett .
* 28 * , 22252227 ( 2003 ) .
m. yu , c. j. mckinstrie and g. p. agrawal , `` modulational instabilities in dispersion - flattened fibers , '' phys . rev .
e * 52 * , 10721080 ( 1995 ) .
peter j. mosley , jeff s. lundeen , brian j. smith , ian a. walmsley , piotr wasylczyk , alfred b. uren , christine silberhorn , in coherence and quantum optics ix , ( kluwer academic / plenum , new york ) ( accepted ) . z.d .
walton , a. v. sergienko , b. e. a. saleh , and m. c. teich , `` polarization - entangled photon pairs with arbitrary joint spectrum '' phys .
rev .
a * 70 * , 052317 ( 2004 ) j. p. torres , f. macia , s. carrasco , and l. torner , `` engineering the frequency correlations of entangled two - photon states by achromatic phase matching '' opt
. lett . *
30 * , 314 ( 2005 ) o. kuzucu , m. fiorentino , m. a. albota , f. n. c. wong , and f. x. krtner , phys .
rev .
lett . * 94 * , 083601 ( 2005 ) a.b .
uren , k. banaszek and i. a. walmsley , `` photon engineering for quantum information processing '' quantum information and computation * 3 * , 480 ( 2003 ) a. b. uren , r. erdmann , m. de la cruz and i. a. walmsley , generation of two - photon states with an arbitrary degree of entanglement via nonlinear crystal superlattices , phys .
rev .
lett . * 97 * , 223602 ( 2006 ) .
v. giovanetti , s. lloyd and l. maccone , `` quantum - enhanced positioning and clock synchronization , '' nature * 412 * , 417419 ( 2001 ) .
m. b. nasr , b. e. a. saleh , a. v. sergienko and m. c. teich , `` demonstration of dispersion - canceled quantum - optical coherence tomography , '' phys .
rev .
lett .
* 91 * , 083601 ( 2003 ) .
l. mandel and e. wolf , _ optical coherence and quantum optics _ ( cambridge university press , 1995 ) .
j. chen , x. li and p. kumar , `` two - photon - state generation via four - wave mixing in optical fibers , '' phys .
rev .
a * 72 * , 033801 ( 2005 ) .
j. chen , k. f. lee and p. kumar r , `` quantum theory of degenerate @xmath0 two - photon state , '' e - print arxiv : quant - ph/0702176v1 .
g. p. agrawal , _ nonlinear fiber optics , 4th ed . _
( elsevier , 2007 ) . c. j. mckinstrie
, h. kogelnik and l. schenato , `` four - wave mixing in a rapidly - spun fiber , '' opt . express * 15 * , 85168534 ( 2006 ) .
this paper also reviews scalar and vector fwm in strongly - birefringent and randomly - birefringent fibers .
k. p. hansen , `` dispersion flattened hybrid - core nonlinear photonic crystal fiber , '' opt .
express * 11 * , 15031509 ( 2003 ) .
t. a. birks , j. c. knight and p. st .
j. russell .
`` endlessly single - mode photonic crystal fiber , '' opt .
lett .
* 22 * , 961963 ( 1997 ) .
g. k. l. wong , a. y. h. chen , s. w. ha , r. j. kruhlak , s. g. murdoch , r. leonhardt , j. d. harvey and n. y. joly , `` characterization of chromatic dispersion in photonic crystal fibers using scalar modulation instability , '' opt .
express * 13 * , 86628670 ( 2005 ) .
a. ortigosa - blanch , a. diez , m. delgado - pinar , j. l. cruz and miguel v. andres , `` ultrahigh birefringent nonlinear microstructured fiber , '' ieee photon .
. lett .
* 16 * , 16671669 ( 2004 ) .
a. l. berkhoer and v. e. zakharov , `` self - excitation of waves with different polarizations in nonlinear media , '' sov .
phys .
jetp * 31 * , 486493 ( 1970 ) . c. j. mckinstrie and s. radic , `` phase - sensitive amplification in a fiber , ''
opt . express * 12 * , 49734979 ( 2004 ) .
r. h. stolen , m. a. bosch and c. lin , `` phase matching in birefringent fibers , '' opt . lett . *
6 * , 213215 ( 1981 ) .
r. j. kruhlak , g. k. l. wong , j. s. y. chen , s. g. murdoch , r. leonhardt , j. d. harvey , n. y. joly and j. c. knight , `` polarization modulation instability in photonic crystal fibers , '' opt
. lett . * 31 * , 13791381 ( 2006 ) .
s. g. murdoch , r. leonhardt and j. d. harvey , `` polarization modulation instability in weakly birefringent fibers , '' opt
. lett . * 20 * , 866868 ( 1995 ) .
q. lin , f. yaman and g. p. agrawal , `` photon - pair generation by four - wave mixing in optical fibers , '' opt .
lett .
* 31 * , 12861288 ( 2006 ) .
q. lin , f. yaman and g. p. agrawal , `` photon - pair generation in optical fibers through four - wave mixing : role of raman scattering and pump polarization , '' phys .
rev .
a * 75 * , 023803 ( 2007 ) .
such a state is typically referred to as highly entangled , but one should keep in mind that the large vacuum component of the state renders this `` entanglement '' useful only in a post - selection experiment .
k. a. odonnell and a. b. uren , `` observation of ultrabroadband , beamlike parametric downconversion , '' opt .
lett .
* 32 * , 817819 ( 2007 ) .
l. zhang , a. b. uren , r. erdmann , k. a. odonnell , c. silberhorn , k. banaszek and i. a. walmsley , `` generation of highly entangled photon pairs for continuous variable bell inequality violation , '' j. mod . opt . * 54 * , 707719 ( 2007 ) .
r. jiang , n. alic , c. j. mckinstrie and s. radic,``two - pump parametric amplifier with 40 db of equalized gain over a bandwidth of 50 nm , '' proc .
ofc 2007 , paper owb2 .
j. m. chavez boggio , j. d. marconi , s. r. bickham and h. l. fragnito , `` spectrally flat and broadband double - pumped fiber optical parametric amplifiers , '' opt .
express * 15 * , 52885309 ( 2007 ) .
s. radic , c. j. mckinstrie , r. m. jopson , j. c. centanni , q. lin and g. p. agrawal , `` record performance of parametric amplifier constructed with highly nonlinear fibre , '' electron
. lett . * 39 * , 838839 ( 2003 ) .
h. takesue and k. inoue , `` 1.5-@xmath1 m band quantum - correlated photon pair generation in dispersion - shifted fiber : suppression of noise photons by cooling fiber , '' opt .
express * 13 * , 78327839 ( 2005 ) . |
sharpe ratio @xcite has become a `` gold standard '' for measuring performance of hedge funds and other institutional investors ( this note uses the generic term `` portfolio '' ) .
it is sometimes argued that it is applicable only to i.i.d .
gaussian returns , but we will follow a common practice of ignoring such assumptions . for simplicity
we assume that the benchmark return ( such as the risk - free rate ) is zero .
the ( ex post ) _ sharpe ratio _ of a sequence of returns @xmath0 is defined as @xmath1 , where @xmath2 ( none of our results will be affected if we replace , assuming @xmath3 , @xmath4 by @xmath5 , as in @xcite , ( 6 ) . )
intuitively , the sharpe ratio is the return per unit of risk .
another way of measuring the performance of a portfolio whose sequence of returns is @xmath6 is to see how this sequence of returns would have affected an initial investment of 1 assuming no capital inflows and outflows after the initial investment .
the final capital resulting from this sequence of returns is @xmath7 .
we are interested in conditions under which the following anomaly is possible : the sharpe ratio @xmath8 is large while @xmath9 .
( more generally , if we did not assume zero benchmark returns , we would replace @xmath9 by the condition that in the absence of capital inflows and outflows the returns @xmath6 underperform the benchmark portfolio . )
suppose the return is @xmath10 over @xmath11 periods , and then it is @xmath12 in the @xmath13th period . as @xmath14 , @xmath15 and @xmath16 .
therefore , making @xmath13 large enough , we can make the sharpe ratio @xmath17 as large as we want , despite losing all the money over the @xmath13 periods .
if we want the sequence of returns to be i.i.d .
, let the return in each period @xmath18 be @xmath10 with probability @xmath19 and @xmath12 with probability @xmath20 , for a large enough @xmath13 . with probability one
the sharpe ratio @xmath21 will tend to a large number as @xmath22 , despite all money being regularly lost .
of course , in this example the returns are far from being gaussian ( strictly speaking , returns can not be gaussian unless they are constant , since they are bounded from below by @xmath23 ) .
it is easy to see that our examples lead to the same conclusions when the sharpe ration is replaced by the _ sortino ratio _
@xcite @xmath24 , where @xmath25
the examples of the previous section are somewhat unrealistic in that there is a period in which the portfolio loses almost all its money . in this section
we show that only in this way a high sharpe ratio can become compatible with losing money .
for each @xmath26 $ ] , define @xmath27 where @xmath28 ranges over the positive integers and @xmath29 over @xmath30 . in other words
, @xmath31 is the best achievable sharpe ratio for sequences of returns that lose money , assuming that none of the returns falls below @xmath32 .
it is not difficult to show that @xmath33 , and in the previous section we saw that @xmath34 . in this section
we are interested in the behaviour of @xmath31 for the intermediate values of @xmath35 , @xmath36 .
the function @xmath31 in the ranges @xmath37 $ ] ( left ) and @xmath38 $ ] ( right).,title="fig:",scaledwidth=48.0% ] the function @xmath31 in the ranges @xmath37 $ ] ( left ) and @xmath38 $ ] ( right).,title="fig:",scaledwidth=48.0% ] figure [ fig : f1 ] shows the graph of @xmath39 over @xmath37 $ ] and over @xmath38 $ ] . over the interval
@xmath37 $ ] the slope of @xmath39 is roughly 1 .
we can see that even for a relatively large value of @xmath40 , the sharpe ratio of a losing portfolio never exceeds 0.5 ; according to table [ tab : f1 ] , @xmath41 ( much less than the conventional threshold of 1 for a good sharpe ratio @xcite ) .
.[tab : f1]the approximate values of @xmath31 , @xmath42 , and @xmath43 for selected @xmath35 . [ cols="^,^,^,^,^,^,^,^",options="header " , ] the values of @xmath44 and @xmath45 for selected @xmath35 are shown in table [ tab : g ] , @xmath46 on the left and @xmath47 on the right .
the meaning of @xmath43 is the same as in tables [ tab : f1 ] and [ tab : f2 ]
. we do not give the values of @xmath42 ; they are huge on the left - hand side of the table and equal to @xmath35 on the right - hand side .
the left - hand side suggests that @xmath48 , and this can be verified analytically .
figures [ fig : f1][fig : g2 ] can be regarded as a sanity check for the sharpe and sortino ratio .
not surprisingly , they survive it , despite the theoretical possibility of having a high sharpe and , _ a fortiori _ , sortino ratio while losing money . in the case of the sharpe ratio , such an abnormal behaviour can happen only when some one - period returns are very close to @xmath23 . in the case of the sortino ratio
, such an abnormal behaviour can happen only when some one - period returns are very close to @xmath23 or when some one - period returns are huge . | a simple example shows that losing all money is compatible with a very high sharpe ratio ( as computed after losing all money ) .
however , the only way that the sharpe ratio can be high while losing money is that there is a period in which all or almost all money is lost .
this note explores the best achievable sharpe and sortino ratios for investors who lose money but whose one - period returns are bounded below ( or both below and above ) by a known constant . |
hypercat maintains catalogues of data collected in the literature or at the telescope , concerning the photometry , kinematics and spectrophotometry of galaxies .
some catalogues contain `` global '' properties as total magnitude and other spatially resolved data .
they give basic data to study the scaling relations of galaxies , as for instance the fundamental plane , and contain all the information needed to make the necessary corrections and normalizations in order to compare measurements of galaxies at different redshifts .
the catalogues of global properties are : * _ the catalogue of central velocity dispersions _
( for galaxies and globular clusters ) has been presented in a preliminary form in prugniel & simien 1996 .
the present version gives 5470 measurements published in 352 references for 2335 objects .
hypercat allows one to retrieve the published measurements as well as homogenized ( ie .
corrected for systematics effects between datasets ) and aperture corrected data .
* _ the catalogue of magnitudes and colours _
( published in prugniel & heraudeau , 1998 ) presents the photometry of 7463 galaxies in the u to i bands .
the global parameters , asymptotic magnitude , surface brightness , photometric type ( ie .
shape of the growth curve ) , colour and colour gradients were computed from circular aperture photometry . * _ the catalogue of mg2 index _
( published in golev & prugniel , 1998 ) have 3712 measurements for 1416 galaxies .
aperture corrections and homogenization are available . * _ the maximum velocity of rotation _ is available for the stellar rotation of 720 galaxies ( mostly early - type ) .
they represents 1491 measurements taken in 224 dataset . a bibliographical catalogue of spatially resolved kinematics ( prugniel et al . 1998 ) indexes 6214 measurements for 2677 galaxies .
in addition , other parameters , like the recession velocity , galactic absorption or environment parameters , are automatically extracted from other databases and hypercat provides procedures to compute derived parameters .
however , the present understanding of the scaling relations becomes limited by the quality of the parameterization restricted to these `` global '' values .
for instance , in prugniel et al .
( 1996 ) we have shown that a more detailed description , including rotation and non - homology of the structure , must be taken into account when studying the fundamental plane of early - type galaxies .
for this reason , hypercat has also embarked in the gathering and distribution of spatially resolved data such as _ multi - aperture photometry _ for 20537 galaxies ( 222045 measurements ) , _ kinematic profiles _
`` rotation curve '' , velocity dispersion profiles ... ) for 1761 galaxies ( 73520 measurements ) and _ catalogue of line strength profiles _ ( currently under development ) .
an original aspect in the development of hypercat is that the different catalogues are _ separately maintained in different sites_. the database is automatically updated by procedures running over the network at time of low - traffic . at present , observatories participating to the project are : capodimonte ( napoli ) , sternberg ( moscow ) , brera ( milano ) , university of sofia and lyon .
the distribution , over several astronomers , of the work to maintain this database makes the individual charge affordable and we can foresee that we will be able to continue this part of the project .
in addition , as hypercat becomes known in the community , people begin to send us their data in a form making them easy to implement .
the usual approach when new measurements are needed is to make new observations .
this is justified when the past observations do not have the required quality , but archived observations offer in many cases a serious alternative _ if _ the data can be accessed easily and have a good enough description .
we started in 1998 the construction of a fits archive in hypercat ( hfa ) coupled to data - mining procedures aimed at distributing data at any desired stage of processing or even measurements . at present hfa
contains 29366 fits files ( for 14631 galaxies ) mainly from the our medium resolution spectra of galaxies ( golev et al 1998 for details ) and eso - lv survey ( lauberts et al .
1989 ) . in the near future
, we will archive other datasets , and in particular we call for contributions from astronomers outside our group which may be interested to distribute their data through this channel .
golev , v. , prugniel , ph . , 1998 132 , 255 lauberts , a. , valentijn , e. a. , 1989 , eso ( 1989 ) , 0 maubon , g. , prugniel , ph . , golev , v. , simien , f. , 1998 _ la lettre de lohp _ 18 prugniel , ph . , simien , f. , 1996 309 , 749 prugniel , ph .
heraudeau , ph . , 1998 128 , 299 | the hypercat database is developed at _ observatoire de lyon _ and is distributed on the web(www - obs.univ - lyon1.fr / hypercat ) through different mirrors in europe .
the goal of hypercat is to gather data necessary for studying the evolution of galaxies ( dynamics and stellar contains ) and particularly for providing a @xmath0 reference for these studies . |
the balance between excitatory and inhibitory elements is of great relevance for processes of self - organization in various neurological structures @xcite .
synchronization is one prominent phenomenon that is critically influenced by the balance of excitation and inhibition @xcite and is itself an important mechanism involved in processes such diverse as learning and visual perception on the one hand @xcite and the occurrence of parkinson s disease and epilepsy on the other hand @xcite .
exploring the preconditions for synchronization on a theoretical level can be effectively performed within the framework of complex dynamical networks a field that has gained much attention during the last decades @xcite .
the influence of a particular network architecture can be rich and is widely studied .
however , the interaction between the elements of the network can be modeled not only by the strength of the connecting links , but in addition , time - delayed coupling can be incorporated to take into account finite signal transmission times .
a single constant delay time , referred to as discrete time - delay , already influences the system s dynamics @xcite and its synchronization properties @xcite dramatically . yet
, more complex concepts of time - delay have been considered recently to achieve a better correspondence between the model and natural systems @xcite .
a distribution of delay - times arises naturally , if a model is constructed from experimental data .
for instance , the delay distribution in a specific avian neural feedback loop can be well approximated by a gamma distribution @xcite .
additionally , time - dependent delay - times can as well be represented by distributed delay , if the delay - time varies rapidly compared to intrinsic time - scales of the system @xcite .
recent theoretical studies of systems with distributed delay have addressed amplitude death @xcite and non - invasive stabilization of periodic orbits @xcite . in this work ,
we study the effects of distributed time - delay on synchronization in a network model with inhibitory nodes .
the stability of synchronization is investigated using the master stability function ( msf ) approach @xcite . with this technique
the effect of the network topology can be separated from the dynamics of the network s constituents , which is described by the master stability function of a complex parameter @xmath0 .
the eigenvalues of the coupling matrix representing the network structure then yield the stability for a specific network when they are inserted for the parameter @xmath0 .
thus , distributed delay , which affects the dynamics , and inhibition , which alters the eigenvalue spectrum of the coupling matrix , can both be investigated independently .
this work extends previous works @xcite on inhibition - induced desynchronization in time - delayed networks in two ways .
first , distributed delay is considered instead of discrete delay .
second , a more realistic network model based on inhibitory nodes instead of inhibitory links is proposed and studied
. such a model might be more appropriate especially in the context of neuroscience , where for the majority of neurons it is established that the same set of neurotransmitters is released at each synapse , a rule which is often referred to as dale s law .
thus , it is reasonable to assume that in most cases a neuron either excites all its neighbors or inhibits all its neighbors , and not a combination of both .
in contrast to previous work @xcite where the inhibitory links have been added randomly , we develop a highly symmetric model , which has two main advantages . first , the eigenvalue spectrum of the coupling matrix is real and can be calculated analytically .
second , in networks with randomly added inhibition , the increase of inhibition is accompanied by the increase of asymmetry .
thus , it is not possible to study the pure effect of inhibition , but only a combined effect of asymmetry and inhibition .
this difficulty does not arise in the network model proposed here . as the network
s constituents we choose stuart landau oscillators .
in contrast to simple phase - oscillators like the kuramoto oscillator , this paradigmatic non - linear oscillator has an additional radial degree of freedom , which leads to more complex dynamics , i.e. , coupled amplitude and phase dynamics .
further , it can be used to describe generally any system close to a hopf bifurcation , which occurs in a variety of prominent systems used to model e.g. , semiconductor lasers @xcite , neural systems like the fitzhugh - nagumo model @xcite or the morris - lecar model @xcite .
this paper is structured as follows . in section [ sec_balance ]
, we introduce the network model .
we derive and characterize the eigenvalue spectrum of its coupling matrix .
in section [ sec_sync ] , we investigate the existence of synchronous oscillations with respect to the distributed delay parameters .
subsequently , we analyze their stability in section [ sec_stable ] . in section [ sec_trans ] , we discuss how the interplay of distributed delay and inhibitory nodes influences the stability for the specific network model proposed here .
the robustness of our results against asymmetric perturbations of the network topology is investigated numerically in section [ sec_asym ] .
we conclude our work in section [ sec_conc ] .
the topology of a network is encoded in its coupling matrix @xmath1 . here , we consider networks without self - coupling @xmath2 and normalized row sum @xmath3 which allows for the existence of synchronous solutions . as
the stability of the synchronous state depends on the eigenvalue spectrum of @xmath1 , we are interested in the question how the balance between excitatory and inhibitory nodes influences the latter . it is known that inhibition corresponding to negative entries in the coupling matrix increases the spreading of the eigenvalue distribution in the complex plane @xcite . for a symmetric network with excitatory nodes only ,
all eigenvalues are real and located within the interval @xmath4 $ ] due to gershgorin s circle theorem @xcite . adding inhibition can cause the existence of eigenvalues with absolute values larger than one .
furthermore , introducing inhibition randomly causes asymmetric coupling matrices that have complex eigenvalues .
these two effects are usually intertwined .
however , it is desirable to investigate the effects of inhibition independently from the effect of asymmetry . to this end
, we introduce a simple network model with excitatory and inhibitory nodes , whose coupling matrix has a real eigenvalue spectrum that can be calculated analytically .
thus , it is capable of providing clear statements about the influence of inhibitory nodes that underline similar numerical results found previously @xcite .
the network model we propose is composed of two regular rings .
one consists of @xmath5 excitatory nodes coupled to their @xmath6 nearest neighbors with equal weight @xmath7 of the links .
the other consists of @xmath8 inhibitory nodes coupled to their @xmath9 nearest neighbors with links of weight @xmath10 .
all excitatory nodes are coupled to all inhibitory nodes with equal link strength @xmath11 .
likewise , all inhibitory nodes are coupled to all excitatory ones with link strength @xmath12 .
this yields a coupling matrix of the form @xmath13 where @xmath14 and @xmath15 represent regular bidirectional ring matrices of link strength @xmath7 with @xmath6 and @xmath9 neighbors , respectively , and @xmath16 with @xmath17 is an @xmath18 matrix .
we impose an equal coupling radius @xmath19 of the two regular rings , i.e. , @xmath20 .
thus , together with the unity row sum condition from eq .
( [ rowsum ] ) , the weights @xmath7 and @xmath11 are fully determined @xmath21 introducing the inhibition ratio , i.e. , the ratio of inhibitory to excitatory nodes @xmath22 , eqs .
( [ aa ] ) , ( [ bb ] ) can be solved as @xmath23 and @xmath24 . exploiting the block structure of @xmath1
, the eigenvalue spectrum can be constructed explicitly . for a detailed derivation
we refer to the appendix [ appendixa ] .
the @xmath25 eigenvalues are given by @xmath26\ ] ] for @xmath27 and @xmath28 \;,\ ] ] for @xmath29 .
the remaining two eigenvalues are @xmath30 and @xmath31 .
the latter is present for any matrix with unity row sum .
it is called longitudinal eigenvalue and is associated with the dynamics within the synchronization manifold .
hence it can be neglected for the consideration of the stability of a synchronous state @xcite .
thus , in the following description it will be omitted .
the spectrum is depicted in fig .
[ fig_evs ] as a function of the inhibition ratio @xmath32 for two different coupling radii @xmath19 .
in the limit of large networks @xmath33 with fixed coupling radius @xmath19 and inhibition ratio @xmath32 , the eigenvalues @xmath34 are well defined for @xmath35 and yield @xmath36 as they correspond to the extremal eigenvalues depicted by red ( black ) lines in fig .
[ fig_evs ] , they will turn out to be of particular importance regarding the stability of synchronization . due to this fact , it is possible to make statements about the stability for arbitrarily large network sizes within this model .
focusing on the large network limit , we note the following observations .
the absolute values of the eigenvalues increase with the inhibition ratio @xmath32 and decrease with the coupling radius @xmath19 . for all @xmath19
, there exists a critical inhibition ratio @xmath37 above which at least one eigenvalue larger than unity exists .
thus , we can confirm and quantify the observation that inhibition leads to a spreading of the eigenvalue spectrum .
further , we note that the distance between the largest and the second largest eigenvalue increases with @xmath19 . for large @xmath19
, this causes the appearance of a gap in the eigenvalue spectrum , which will be relevant for resynchronization at high inhibition ratios . in fig .
[ fig_evs ] the features mentioned are clearly displayed . for two different coupling radii ( a ) @xmath38 , ( b ) @xmath39 .
network size @xmath40 , with @xmath41 excitatory nodes and @xmath42 inhibitory nodes .
the red ( grey ) lines show the extremal eigenvalues in the large network limit given by eqs .
( [ ex1 ] ) and ( [ ex2 ] ) .
, height=151 ]
apart from the network topology , the stability of a synchronous state depends on the dynamics of the network s constituents and the delay . here ,
we consider coupled stuart
landau oscillators with distributed time delay @xmath43 \ ; , \label{z}\end{aligned}\ ] ] where @xmath44 is the normal form of a supercritical hopf bifurcation @xmath45 @xmath46 is the coupling strength and @xmath1 denotes the coupling matrix .
the delay distribution @xmath47 is normalized and positive - semidefinite .
we consider uniformly distributed delay @xmath48 \\ 0 & \text{otherwise } \end{cases } \;,\ ] ] where @xmath49 defines the width of the distribution and @xmath50 is the mean delay time , and gamma distributions defined by @xmath51 where @xmath52 is the gamma function . here
, @xmath53 describes a fixed time offset , which shifts the distribution allowing for coupling without instantaneous parts and the integer @xmath54 as well as the real parameter @xmath55 characterize the shape of the distribution .
the mean delay time @xmath50 is given by @xmath56 . in the following only the strong kernel , @xmath57 ,
is considered .
generalizations for the weak kernel , i.e. , exponential decay ( @xmath58 ) and higher values of @xmath54 , however , can be obtained in a straightforward fashion and exhibit the same qualitative features , since for all gamma distributions the exponentially decaying tail dominates the behavior .
it is convenient to rephrase eq .
( [ z ] ) in polar coordinates using the amplitude @xmath59 and the phase @xmath60 . with
the relation @xmath61 one obtains @xmath62 r_j(t - t ' ) -r_i(t ) \right\ } { \mathrm{d}}t ' \ ; , \nonumber \\ \dot \varphi_i(t ) & = \omega + \sigma \sum_{j=1}^n g_{ij } \int_0^\infty g(t ' ) \times \label{phidot } \\ & \sin [ \varphi_j(t - t')-\varphi_i(t ) ] \frac{r_j(t - t')}{r_i(t ) } { \mathrm{d}}t ' \nonumber \;.\end{aligned}\ ] ] a synchronous zero - lag oscillation with constant common amplitude @xmath59 and constant common frequency @xmath63 implies @xmath64 and @xmath65 for all @xmath66 .
this yields two coupled equations , which define the synchronous state . for uniformly distributed delay , they are transcendental and read @xmath67 in the eigenfrequency dominated regime , i.e. , @xmath68
, the existence and the magnitude of the amplitude @xmath59 follows a periodic structure in both the mean delay @xmath50 and the distribution width @xmath49 ( cf .
[ fig_sol ] ) .
even ( odd ) multiples of the eigenperiod @xmath69 yield maximum ( minimum ) amplitudes , which can be understood intuitively in terms of resonance ( antiresonance ) phenomena .
in contrast , for the @xmath70-distribution ( @xmath57 ) the solution determined by @xmath71 \ ; , \\
\omega & = \omega - \sigma \left [ \frac { ( l^2-\omega^2)l^2 \cos ( \omega \tau_0 - \nu ) } { l^2+\omega^2 } -1 \right ] \;,\end{aligned}\ ] ] with @xmath72 is periodic in the time - offset @xmath53 , but not periodic in the distribution width , which can be measured by the standard deviation @xmath73 and is inversely proportional to the shape parameter @xmath9 . both distributions reproduce known results for discrete time delay in the limit of small distribution widths , e.g. , the existence of multiple solutions at large delay times @xcite , and yield a universal weak coupling solution in the limit of large widths , where @xmath74 vanishes as a consequence of the riemann - lebesgue lemma @xcite .
the amplitude and frequency in this limit correspond to the solution of uncoupled stuart landau oscillators with eigenfrequency @xmath75 and a shifted bifurcation parameter @xmath76 . in fig .
[ fig_sol ] , the amplitude of the synchronous oscillation for ( a ) a uniform delay kernel and ( b ) a gamma distributed kernel is displayed as a function of ( a ) the mean delay time @xmath50 and the distribution width @xmath49 and ( b ) the time offset @xmath53 and the standard deviation @xmath73 , respectively .
of the synchronous oscillation for ( a ) uniformly distributed delay , ( b ) delay distributed by a @xmath70-distribution ( @xmath57 ) .
parameters : @xmath77 .
bottom right white region in ( a ) : forbidden because of causality .
other white regions : no solution to eqs .
( [ r ] ) , ( [ o ] ) .
, height=136 ]
the stability of the synchronous solution is determined by the master stability function method @xcite , which has recently been extended to systems with distributed time - delay @xcite . in this approach
, a linear stability analysis of the synchronous solution is performed while the coupling matrix is diagonalized .
a variational equation defines the floquet exponents @xmath78 , which determine the stability .
-plane for uniformly distributed delay of width @xmath49 with different mean delay times @xmath50 ; @xmath0 corresponds to the eigenvalues of the coupling matrix .
parameters : @xmath79 .
white region in ( a ) : no solution ( @xmath80 ) to eqs .
( [ r ] ) , ( [ o]).,scaledwidth=63.0% ] as we consider a periodic orbit , a description in radial and phase coordinates reflects the symmetry of the problem and yields time - independent jacobi matrices , which simplifies the stability analysis drastically .
however , in this coordinate system , the coupling term is non - linear , which requires a technical modification of the approach given in ref .
the case considered here in eqs .
( [ rdot ] ) , ( [ phidot ] ) is of the general form @xmath81 where the coupling scheme @xmath82 reduces to a matrix vector multiplication for linear coupling as in ref .
the linear stability of the synchronous solution is determined by the following variational equation @xmath83 where the matrices @xmath84 and @xmath85 are calculated as @xmath86 and @xmath0 corresponds to the eigenvalues of the coupling matrix that emerge from the diagonalization of @xmath1 , and @xmath87 represents the synchronous solution . in the case
studied here , the variational state vector is defined by @xmath88 and due to the choice of the coordinate system , the matrices @xmath84 and @xmath85 do not explicitly depend upon time @xmath89 , i.e. , @xmath90 and @xmath91 .
they read @xmath92 a(t ' ) & = \sigma \begin{pmatrix } \cos \omega t ' & \sin \omega t ' \\ - \sin \omega t ' & \cos \omega t ' \end{pmatrix } \;,\end{aligned}\ ] ] with the abbreviations @xmath93 applying the ansatz @xmath94 yields a generalized characteristic equation for the floquet exponents @xmath78 @xmath95 = 0 \ ; , \label{det}\ ] ] where @xmath96 is the identity matrix .
the maximum real part of the floquet exponents determines the stability .
evaluated as a function of the eigenvalue @xmath0 , it is referred to as master stability function ( msf ) @xmath97 when all eigenvalues @xmath0 of a given coupling matrix @xmath1 yield a negative value of the master stability function , the synchronous state is stable for this network . since for the network model considered here the eigenvalues
are real , the msf is evaluated only for real @xmath0 .
the generalized characteristic equation ( [ det ] ) is equivalent to determining the roots of an exponential polynomial . exploiting the analyticity of these functions yields an elegant procedure to determine the number of roots with positive real parts , which answers the question of stability without calculating the floquet exponents explicitly . for uniformly distributed delay this procedure
is explained in appendix [ appendixb ] .
the msf for uniformly distributed delay inherits some of the characteristic structures from the solution itself ( cf .
[ fig_msf_rect ] ) .
its qualitative appearance is periodic in the mean delay and the distribution width .
evaluated in the @xmath98-plane , for mean delay times equal to multiples of half the eigenperiod @xmath99 , i.e. , @xmath100 ( fig .
[ fig_msf_rect]a , d ) the periodicity in @xmath49 resembles the checkerboard structure of resonances and antiresonances , which determines the amplitude of the synchronous solution ( cf . fig .
[ fig_sol ] ) .
in particular , antiresonant regions are characterized by the existence of a stability island only for @xmath101 , a condition that can not be matched by all eigenvalues of a coupling matrix of the type considered here and thus corresponds to unstable synchronization . and
the delay width @xmath49 for uniformly distributed delay in a network with coupling radius ( a ) @xmath38 , ( b ) @xmath39 .
eigenvalue spectra ( dots ) and master stability function ( color code ( grey scale ) ) with ( c ) coupling radius ratio @xmath38 and delay distribution width @xmath102 and ( d ) @xmath103 .
network size @xmath104 with @xmath105 excitatory and @xmath42 inhibitory nodes .
other parameters as in fig .
[ fig_msf_rect ] ( b ) . vertical lines in ( c ) ,
( d ) denote synchronization - desynchronization transitions . ] in the resonant regions the stability island ( light color ) contains the interval @xmath4 $ ] .
hence , networks below the critical inhibition ratio given in eq .
( [ inhi ] ) exhibit stable synchronous oscillations .
the same applies to mean delays exactly between multiples of half the eigenperiod , i.e. , multiples of a quarter eigenperiod ( cf .
[ fig_msf_rect]c ) . here ,
the unit interval is contained within the stability regions for any @xmath49 .
additionally , a second stability island appears above a certain delay distribution width @xmath49 .
the second stability island can induce stable synchronization for networks with an inhibition ratio above the critical one .
this is the case if the eigenvalues outside the unit interval happen to lie inside the additional stability island .
for any mean delay different from the special cases mentioned above , the msf shows a very rich structure ( fig . [ fig_msf_rect]b ) .
depending on @xmath49 a single connected or two disconnected stability regions of various size can occur .
further , the gap between the two islands can lie left or right to the line @xmath106 .
note that similar behavior has been found in case of a discrete delay and different local dynamics @xcite .
the stability gap in the master stability function immediately implies several interesting consequences for the stability of synchronous solutions in inhibitory networks , which are discussed in detail in the following .
as the interval @xmath107 $ ] in fig . [ fig_msf_rect]b covers all qualitatively different regimes , further analysis is restricted to this interval . in fig .
[ fig_desync ] the stability of synchronous solutions is evaluated for the concrete network type proposed here in the parameter setting that yields the msf of fig .
[ fig_msf_rect]b .
the stability is evaluated as a function of the inhibition ratio @xmath32 and the distribution width @xmath49 of the delay for two different coupling radii @xmath19 .
the result elucidates in detail the complex interplay between distributed delay and the excitatory - inhibitory node balance . for the higher coupling radius @xmath38 ( cf .
[ fig_desync]a ) , the stability of synchronization has a strong structural agreement with the msf itself ( cf .
[ fig_msf_rect]b ) . both can be categorized into five different @xmath49-regimes to get a better understanding of how their features are related .
\(i ) for small distribution widths , i.e. , @xmath108 only one stability island of the msf exists .
it contains the interval @xmath4 $ ] and is bounded by @xmath106 .
thus , all networks up to the critical inhibition ratio exhibit stable synchronous solutions and above the critical inhibition ratio , the stability collapses and the network desynchronizes .
\(ii ) for larger @xmath49 , i.e. , @xmath109 a second stability island appears in the msf for @xmath101 and its extension grows with @xmath49 .
this causes a resynchronization at relatively high inhibition ratios , when the largest eigenvalue increasing with @xmath32 reenters the stability region . for @xmath102
this is depicted in fig .
[ fig_desync]c . since for stability
all eigenvalues of the coupling matrix are required to lie inside the stability regions , the gap of the eigenvalue spectrum of @xmath1 is crucial . it must contain the stability gap of the msf in order to induce resynchronization above the critical inhibition ratio .
\(iii ) in a narrow @xmath49-range around @xmath110 , the two stability islands of the msf almost touch each other . here , synchronization is stable up to extremely high inhibitory ratios of about @xmath111 .
\(iv ) for even larger @xmath49 , i.e. , @xmath112 , the stability gap of the msf lies within the interval @xmath4 $ ] .
hence , synchronization is suppressed for comparably small inhibition ratios when the maximum eigenvalue is no longer contained inside the stability region . on the other hand , resynchronization occurs at lower inhibition ratios . in an extreme case
, this leads to the situation that synchronization is unstable up to a certain inhibition ratio , which is just the inverse behavior of what occurs for small @xmath49 in case ( i ) .
\(v ) finally , synchronization is unstable for any network , if the stability island on the left is too small , which is the case for @xmath113 . in particular if the eigenvalue @xmath114 , which is independent of @xmath19 and @xmath32 , is no longer contained inside the stability region . for networks with lower coupling radii ,
e.g. @xmath39 , the eigenvalue spectrum has no pronounced gap below the largest eigenvalue .
thus , several of the largest and smallest eigenvalues leave and reenter the gapped stability regions of the msf when the inhibition ratio is increased .
this leads to qualitative deviations from the case @xmath38 as displayed in fig .
[ fig_desync]b .
apart from the decreased critical inhibition ratio defined in eq .
( [ inhi ] ) , the most striking difference appears for intermediate values of @xmath49 , where instead of a single resynchronization , multiple synchronization - desynchronization transitions appear for increasing @xmath32 .
each of them is caused by a different eigenvalue @xmath115 .
this mechanism is visualized in fig .
[ fig_desync]d .
the last two examples illustrate how crucial the delay distribution width is for the stability of synchronization in a network with inhibitory nodes . depending on the distribution width
, the inhibition ratio can induce a single synchronization transition , a single desynchronization transition or even multiple synchronization - desynchronization transitions . for small coupling radii the effect of inhibition is enhanced , and multiple transitions are more likely to occur .
delay distribution with standard deviation @xmath116 in the @xmath117-plane for two different time - offsets ( a ) @xmath118 , ( b ) @xmath119 .
other parameters as in fig .
[ fig_msf_rect].,scaledwidth=55.0% ] we emphasize that for the network type chosen here , every transition can be predicted analytically from the eigenvalue spectrum in the continuum limit once the msf is known .
thus , we can conclude universal statements from the msfs by categorizing them into the five distinct regimes listed above .
this procedure is independent of the specific form of the distribution .
for instance , the msf for the @xmath70-distribution can be categorized by the same means . in the @xmath120-plane of the topological eigenvalue @xmath0 and the standard deviation @xmath116 of the delay distribution , evaluated at different time - offsets @xmath53
it exhibits similar structures and regularities like e.g. , the stability gap and the particular importance of the line @xmath106 ( cf .
[ fig_msf_gam ] ) . for suitable values of @xmath53 ,
e.g. , @xmath119 , the msf shows qualitatively the same features as the msf for uniformly distributed delay from fig .
[ fig_msf_rect]b .
thus , it will generate a similar stability diagram as depicted in fig .
[ fig_desync ] .
this implies two useful consequences .
first , the rather artificial uniform distribution can be used to approximate the more realistic gamma distribution for suitable choices of the parameters .
second , the stability diagram in fig .
[ fig_desync ] possesses a certain universality in the sense that for this type of model qualitatively similar regimes are expected .
the network model considered here has the advantage that it can be treated analytically due to its high degree of symmetry .
however , real - world networks will typically possess a certain degree of asymmetry .
thus , it is interesting to investigate whether the qualitative nature of the synchronization desynchronization transitions observed for this specific network model is robust against asymmetric perturbations of the network topology . in order to address this question , we propose a rewiring algorithm which breaks the network symmetry and investigate the synchronization desynchronization transitions for those perturbed networks . for an arbitrary asymmetric network the stability of synchronization can no longer be calculated analytically since the topological eigenvalues of the coupling matrix are not known in closed form and have to be computed numerically .
they are complex and thus have to be compared to the master stability function evaluated in the @xmath121-plane .
we choose a rewiring algorithm that leaves the number of inhibitory and excitatory nodes constant such that the inhibition ratio @xmath32 remains a characteristic network parameter . to this end
we pick a node randomly and rewire one of its outgoing links to another randomly chosen node , with duplicated links forbidden .
the weight of the new link is chosen according to the four connection classes , i.e. , excitatory - inhibitory ( exc .- inh . ) links of weight @xmath11 , exc .- exc .
links of weight @xmath7 , inh .- inh .
links of weight @xmath10 , and inh .- exc .
links of weight @xmath12 . to ensure synchronizability the row sums are normalized after the rewiring process . note that in networks with inhibition the normalization procedure is ambiguous and a normalized row sum in principle still allows for inputs with arbitrarily large absolute values as long as excitation and inhibition compensate each other . to avoid these unphysical situations we impose the condition that in addition the row sum of absolute values is kept constant .
this ensures that the eigenvalue spectrum of the coupling matrix is bounded by the same gershgorin discs before as well as after the rewiring procedure .
cases where the row sum is no longer positive are excluded by the algorithm since the row sum can not be normalized in a straightforward fashion . by rewiring repeatedly ,
the total number of inhibitory and excitatory nodes as well as the total number of excitatory and inhibitory links is kept constant , but the topology is steadily randomized under these constraints .
the global coupling between nodes of different categories acquires defects , while the regular ring networks defining the topology of nodes of the same category transform into small - world networks @xcite . in the limit of infinitely many rewirings ,
the signatures of the regular rings and of the global coupling are lost and the network evolves into a random network with a uniform wiring probability per connection category . in fig .
[ fig_asym ] we show the fraction of synchronization @xmath122 as a function of the inhibition rate @xmath32 for the unperturbed symmetric network ( @xmath123 ) as well as for perturbed networks generated by more and more rewiring iterations @xmath124 .
we consider two different coupling radii @xmath19 and delay distribution widths @xmath49 that show non trivial synchronization
desynchronization transitions for the unperturbed network in fig .
[ fig_desync ] , i.e. , @xmath38 , @xmath125 in fig .
[ fig_asym]a and @xmath39 , @xmath126 in fig
. [ fig_asym]b .
the qualitative nature of the transitions is preserved for rewiring iterations of up to @xmath127 in ( a ) , and , up to @xmath128 in ( b ) , respectively . at @xmath128 and for coupling radius @xmath39 considered in ( b ) , the parts of the network connecting nodes of the same category are no longer well described by regular rings , but exhibit the characteristics of a small - world network namely small shortest pathlengths and high clustering . it is interesting that for these networks the synchronization - desynchronization transition is qualitatively the same as compared to the fully symmetric network
. however , the transitions occur at higher inhibition ratios for rewired networks compared to unperturbed networks . for a very large number of rewirings ( @xmath129 ) the synchronization
desynchronization transition changes qualitatively for @xmath39 ( fig .
[ fig_asym]b ) and synchronization is stable up to much larger inhibition ratios for both @xmath38 and @xmath39 .
this behavior can be understood by considering how the eigenvalue spectrum transforms under the rewiring process . in fig .
[ fig_msf_ev ] , we show the master stability function compared to the eigenvalue spectra of networks generated by an increasing number of rewirings .
the following trends are visible : for an increased number of rewirings , the imaginary parts of the eigenvalues increase due to the emergence of asymmetry as expected .
in addition , the extremal eigenvalues are dragged towards the origin and the spectrum is contracted to a spherical distribution of comparably small radius , which is characteristic for random networks @xcite .
the contraction explains why the synchronization - desynchronization transitions occur for higher inhibition ratios compared to the unperturbed networks , since the shrinking of the maximum real part counteracts the spreading of the eigenvalues along the real axis caused by increasing inhibition ratios ( see fig .
[ fig_desync]c ) .
since the multiple synchronization desynchronization transitions occur due to the gap structure of the eigenvalue spectrum for the symmetric model , we infer that they do not appear for strongly randomized networks whose eigenvalue spectra show less pronounced gaps .
this explains why we observe a qualitative change of the synchronization
desynchronization transition in fig .
[ fig_asym]b from @xmath128 to @xmath130 .
we conclude that the analytic results obtained for the symmetric network model are transferable to networks with asymmetric perturbations . however , qualitative deviations from the latter are possible if the information about the original network structure is destroyed by sufficiently large perturbations . as a function of the inhibition ratio @xmath32 for increasing numbers of repeated rewirings @xmath124 from top to bottom .
the blue ( black ) dashed line marks parameters which are considered in fig .
[ fig_msf_ev ] .
parameters : @xmath131 realizations , network size @xmath104 with @xmath41 excitatory and @xmath42 inhibitory nodes , ( a ) @xmath132 , @xmath38 , ( b ) @xmath125 , @xmath39 , other parameters as in fig .
[ fig_msf_rect]b .
, title="fig : " ] as a function of the inhibition ratio @xmath32 for increasing numbers of repeated rewirings @xmath124 from top to bottom .
the blue ( black ) dashed line marks parameters which are considered in fig . [ fig_msf_ev ] .
parameters : @xmath131 realizations , network size @xmath104 with @xmath41 excitatory and @xmath42 inhibitory nodes , ( a ) @xmath132 , @xmath38 , ( b ) @xmath125 , @xmath39 , other parameters as in fig .
[ fig_msf_rect]b .
, title="fig : " ] , ( b ) @xmath130 , and ( c ) @xmath133 rewirings .
parameters : @xmath134 , @xmath38 ( see fig .
[ fig_asym]a blue ( black ) dashed line ) , @xmath41 , @xmath132 , others as in fig .
[ fig_msf_rect]b .
, title="fig : " ] , ( b ) @xmath130 , and ( c ) @xmath133 rewirings .
parameters : @xmath134 , @xmath38 ( see fig .
[ fig_asym]a blue ( black ) dashed line ) , @xmath41 , @xmath132 , others as in fig .
[ fig_msf_rect]b .
, title="fig : " ] , ( b ) @xmath130 , and ( c ) @xmath133 rewirings .
parameters : @xmath134 , @xmath38 ( see fig .
[ fig_asym]a blue ( black ) dashed line ) , @xmath41 , @xmath132 , others as in fig .
[ fig_msf_rect]b .
, title="fig : " ]
in this paper we have introduced a dynamic network model , where the influence of the excitatory - inhibitory node balance could be directly tracked in the eigenvalue spectrum of the coupling matrix . for this model
we could thus confirm and extend previous results that increasing inhibition ratios lead to a spreading of the eigenvalue spectrum which likely causes desynchronization .
the existence of synchronous oscillations has been investigated in a network of coupled stuart - landau oscillators with distributed delay . for uniformly distributed delay
, the amplitude of oscillations depends in a periodically modulated way on the distribution width and the mean delay , which we can understand as a consequence of resonance and antiresonance phenomena .
this characterization into resonant and antiresonant regimes could also be applied to classify the structure of the master stability function ( msf ) , which determines the stability of synchronous oscillations . here
, we have identified regions of resonance and antiresonance determined by the position of the stability island of the msf . however , for mean delays incommensurable with half the eigenperiod stability gaps are observed .
the resulting consequences for the stability of synchronous oscillations in networks with inhibitory nodes were analyzed in detail and the combined effects of distributed delay and the inhibition ratio was investigated .
we have found that both the delay distribution width as well as the inhibition ratio can induce synchronization - desynchronization transitions . for small delay distribution widths ,
strong inhibition causes desynchronization .
this result is in agreement with other works , e.g. , by battaglia et al.@xcite , where sufficiently strong long - range excitations cause full synchronization . as a tendency
, we found that distributed delay enhances the possibility of stable synchronization for increasing inhibition ratios .
the robustness of our results against asymmetric perturbations of the network topology is confirmed numerically .
one main result is that the synchronization - desynchronization transitions do not change qualitatively when random rewiring of the links introduces a small - world structure for the connections between nodes of the same category .
this regime is of particular interest since it is known that neural networks exhibit small - world character @xcite .
this work indicates the relevance of distributed delay beyond the stabilization of fixed points and sheds some light on its interesting effects on synchronization in non - linear systems .
this work was supported by dfg in the framework of sfb 910 .
to calculate the eigenvalues of @xmath13 where @xmath14 and @xmath15 are regular bidirectional ring @xmath135 ) and @xmath136 matrices , respectively , and @xmath16 is an @xmath137 matrix with entry @xmath11 at every position , we make the following ansatz for the @xmath138-dimensional eigenvector @xmath139 . here , @xmath140 is an @xmath5-dimensional vector and @xmath141 is an @xmath8-dimensional vector .
the eigenvalue problem then reads @xmath142 setting @xmath143 , we obtain @xmath144 as @xmath14 is a circulant matrix , its eigenvectors are known and the @xmath9-th entry of the @xmath145-th eigenvector is given by @xmath146 since @xmath16 has the same entry at every position , eq .
( [ b ] ) yields @xmath147 , which is fulfilled for every @xmath148 defined in eq .
( [ a6 ] ) with @xmath27 .
hence , the first @xmath149 eigenvectors are found and the corresponding eigenvalues can be calculated from the general formulas known for circulant matrices . with the ansatz @xmath150 , further @xmath151 eigenvectors can be constructed analogously .
the eigenvector @xmath152 yields the eigenvalue @xmath153 due to the unit row sum condition and the last eigenvector can be calculated explicitly with the ansatz @xmath154 , @xmath155 . a straightforward calculation yields @xmath156 and the corresponding eigenvalue is again calculated explicitly .
for uniformly distributed delay the problem of solving eq .
( [ det ] ) is essentially a problem of determining the roots of an exponential polynomial with frequencies @xmath157 and @xmath158 . to see this
, one has to multiply eq .
( [ det ] ) with @xmath159
. then , problem eq .
( [ det ] ) can be written in the following form @xmath160 where @xmath161 with real coefficients @xmath162 , given by @xmath163 \ ; , \\
a_{001 } & = \frac{2 \gamma \sigma \omega}{\rho } \left [ \sin ( \omega \rho - \omega \tau ) ( 2 r^2 + 2 g_1 \sigma ) \right . \\ & \left .
\quad-2 g_2 \sigma \cos ( \omega \rho - \omega \tau ) \right ] \ ; , \\ a_{011 } & = -\frac { \gamma^2 \sigma^2 } { 2 \rho^2 } \cos ( 2 \rho \omega ) \;,\\ a_{020 } & = a_{002 } = \frac{\gamma^2 \sigma^2}{4 \rho^2 } \;,\\ a_{100 } & = 2 \omega^2 ( g_1 \sigma + r^2)\ ; , \\
a_{110 } & = \frac { \gamma \sigma } { 2 \rho } \left [ \cos ( \omega \rho + \omega \tau ) ( -2 r^2 - 2 g_1 \sigma ) \right .
\\ & \quad \left . -2 g_2 \sigma \sin ( \omega \rho + \omega \tau ) - \omega \sin ( \omega \rho + \omega \tau ) \right]\ ; , \\ a_{101 } & = -\frac { \gamma \sigma } { 4 \rho } \left [ \cos ( \omega \rho - \omega \tau ) ( -2 r^2 - 2 g_1 \sigma ) \right . \\ & \left .
\quad -2 g_2 \sigma \sin ( \omega \rho - \omega \tau ) + \omega \sin(\omega \rho -\omega \tau ) \right ] \;,\\ a_{200 } & = c_0 + \omega^2 \ ; , \\ a_{210 } & = - \frac { \sigma \gamma}{4 \rho } \cos ( \omega\rho+\omega \tau)\ ; , \\
a_{201 } & = \frac { \sigma \gamma}{4 \rho } \cos ( \omega\rho -\omega \tau)\ ; , \\ a_{300 } & = 2 \sigma g_1 + 2 r^2 \;,\end{aligned}\ ] ] where @xmath164 the synchronous state is stable if @xmath165 has no roots with positive real part .
this question can be answered using fundamental results of complex analysis since @xmath165 is an analytic function , which decays exponentially with @xmath166 . following the work by habets @xcite with a slight modification regarding the treatment of the roots on the imaginary axis @xmath167 ,
we find that the number of roots @xmath168 with positive real parts can be calculated as @xmath169 where @xmath170 with @xmath171 and @xmath172 and @xmath173 .
this reduces the multidimensional root search to the calculation of a single integral of a real - valued function on a finite interval , which is an enormous simplification of the problem . additionally , this method is very robust against numerical errors and produces a reliable boundary of the stability islands of the msf .
51ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo
secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ] + 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty @noop * * , ( ) link:\doibase 10.1038/nature01616 [ * * , ( ) ] link:\doibase 10.1523/jneurosci.5297 - 05.2006 [ * * , ( ) ] link:\doibase 10.1038/nn.2276 [ * * , ( ) ] \doibase http://dx.doi.org/10.1016/0959-4388(95)80012-3 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physrevlett.101.078102 [ * * , ( ) ] link:\doibase 10.1126/science.1216483 [ * * , ( ) ] , link:\doibase 10.1103/physrevlett.106.194101 [ * * , ( ) ] link:\doibase 10.1103/physreve.86.061903 [ * * , ( ) ] \doibase doi : 10.1016/s0896 - 6273(00)80821 - 1 [ * * , ( ) ] @noop * * , ( ) @noop _ _ , ed .
( , , ) link:\doibase 10.1103/revmodphys.74.47 [ * * , ( ) ] \doibase doi : 10.1016/j.physrep.2005.10.009 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physreve.79.056207 [ * * , ( ) ] link:\doibase 10.1103/physreve.81.025205 [ * * , ( ) ] link:\doibase 10.1209/0295 - 5075/96/60013 [ * * , ( ) ] link:\doibase 10.1140/epjb / e2012 - 30810-x [ * * , ( ) ] in @noop _ _ , ( , , ) chap . , pp .
@noop * * , ( ) link:\doibase 10.1137/060673813 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.94.158104 [ * * , ( ) ] link:\doibase 10.1140/epjb / e2014 - 40985 - 7 [ * * , ( ) ] link:\doibase 10.1007/s00422 - 008 - 0239 - 8
[ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physreve.88.032912 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1140/epjb / e2011 - 20677 - 8 [ * * , ( ) ] link:\doibase 10.1098/rsta.2012.0466 [ * * , ( ) ] link:\doibase 10.1007/s40435 - 013 - 0049 - 2 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.80.2109 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) link:\doibase 10.1103/physrevlett.105.254101 [ * * , ( ) ] @noop _ _ , springer theses ( , , ) @noop _ _ , annals of mathematics studies ( , ) link:\doibase 10.1103/physreve.89.032915 [ * * , ( ) ] @noop * * , ( ) link:\doibase 10.1103/physreve.84.046107 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.99.238106 [ * * , ( ) ] @noop * * , ( ) @noop * * , ( ) http://opac.inria.fr/record=b1049107[__ ] , ( , ) | we investigate the combined effects of distributed delay and the balance between excitatory and inhibitory nodes on the stability of synchronous oscillations in a network of coupled stuart
landau oscillators . to this end
a symmetric network model is proposed for which the stability can be investigated analytically .
it is found that beyond a critical inhibition ratio synchronization tends to be unstable .
however , increasing distributional widths can counteract this trend leading to multiple resynchronization transitions at relatively high inhibition ratios .
the extended applicability of the results is confirmed by numerical studies on asymmetrically perturbed network topologies .
all investigations are performed on two distribution types , a uniform distribution and a gamma distribution . |
the inclination of a spiral galaxy ( i.e. , the angle between the galactic plane and the tangent plane of the celestial sphere ) is not only an important parameter , but also difficult to determine .
a spiral galaxy consists of a thin disk , a bulge and spiral arms that are thought to be situated in the disk .
if we assume that the thickness of the spiral plane is rather negligible in comparision to its extension , and that when a spiral galaxy is inclined moderately to the plane of sky , the thickness of the nucleus can be omitted , the inclination @xmath1 can be obtained by : @xmath2 where @xmath3 and @xmath4 are the apparent major and minor isophotal diameters respectively . when a spiral galaxy is seen edge - on , it is not possible to consider the thickness of the nuclear part as negligible
thus , eq . ( 1 ) can not be used to calculate the inclination .
the reason for this is that the apparent minor isophotal - diameter consists of two parts .
one is attributed by the disk and , another by the bulge , the latter of which will decrease the real value of the inclination . considering that the disk is not infinitely thin , aaronson et al .
( 1980 ) corrected eq .
( 1 ) by @xmath5 the constant of @xmath6 is added in accordance with an empirical recipe . a more elaborate specification of the axial ratio for an edge - on system that depends on the hubble type could be justified .
the thinnest galaxies are sc spirals , earlier types have larger bulges .
giovanelli et al .
( 1997 ) provided an example to justify why they assumed the axial ratio of sc galaxies to be 0.13 .
a smaller value of the axial ratio for an edge - on system results in smaller derived inclinations , where the spirals are more face - on . besides , if the values @xmath3 and @xmath4 are approximations due to errors , the inclination obtained by eq .
( 2 ) is not an exact value .
ma et al . ( 1997 , 1998 ) proposed a method to determine the inclination of a spiral galaxy by fitting a spiral arm with a logarithmic spiral form with constant pitch angle .
they obtained the inclinations of 72 northern spiral galaxies .
the question of the mathematical form of spiral arms was recognized at the beginning of this century ( von der pahlen , 1911 ; groot , 1925 ) .
then , danver ( 1942 ) , kennicutt ( 1981 ) and kennicutt & hodge ( 1982 ) systematically studied the shapes of spiral arms . using the method of the least squares and as many points as possible situated on the spiral arm in question , danver ( 1942 ) studied a sample of 98 nearby spirals by drawing the projected images on white paper and then , by copying it on the paper to be used for the measurement thanks to transparent illumination .
kennicutt ( 1981 ) measured the shapes of spiral arms in 113 nearby sa - sc galaxies by disposing directly of photographic enlargement and using an iterative procedure to correct for inclination effects .
he gave an initial estimate of the inclination and pitch angle to orient the spiral to a face - on geometry , and then used any residual sinusoidal deviations in the arm shapes to make small corrections to the derived orientation . using the iraf software , ma et al .
( 1997 , 1998 ) fitted the shapes of spiral arms on the images , so that they could show clearly whether the fitting was good or not .
the display program of iraf software can enlarge the image and change its grey scale to minimize any personal prejudice about the regularity and prominence of arms .
but we must emphasize that the display program of iraf has many variables , so the results are not always objective . in our program , we modify z1 ( minimum greylevel to be displayed ) and z2 ( maximum greylevel to be displayed ) in the display program in order to display the images clearly . in the procedure of fitting , we emphasize the global spiral structure , where , except for the small - scale distortions , the arms can be represented by the logarithmic spiral forms . there has been much interest concerning the separation of disk and bulge components in the observed surface brightness distribution of spiral galaxies .
de vaucouleurs ( 1958 ) , for instance , established an isophotic map of m 31 in blue light by means of direct photoelectric scans , spaced at 10@xmath7 intervals in declination from + 39@xmath831@xmath7 to 42@xmath830@xmath7 . from photoelectric photometry , he determined that the thickness of the flat component is about 0.8 kpc . by assuming that a galaxy has an infinitesimally thin disk , freeman ( 1970 ) and sandage et al .
( 1970 ) collected and studied the radial distribution of the surface brightness @xmath9 for thirty - six s0 and spiral galaxies , and showed that @xmath9 distribution for these galaxies can be presented by two main components : an inner spheroidal component which follows the law of @xmath10 and an outer exponential component ( disk ) , with @xmath11 where @xmath12 is defined as a radial scale length .
van der kruit and searle ( 1981a ) proposed a model for the light distribution in the disks of edge - on spiral galaxies , assuming that a galaxy has a locally isothermal , self - gravitating and truncated exponential disk .
this model has the feature of being isothermal in @xmath13 at all radii with a scale parameter @xmath14 and has an exponential dependence of surface brightness upon @xmath15 with a scale length @xmath12 .
the space - luminosity of this model can be described by @xmath16 with this model , van der kruit & searle ( 1981a , 1981b , 1982a , 1982b ) determined @xmath12 and @xmath14 for seven disk - dominated and one spheroid - dominated spiral galaxies by using three - color surface photometry .
peng et al ( 1979 ) investigated three - dimensional disk galaxies , based on the fundamental assumption by parenago that the density distribution along @xmath13-direction for a finite thickness disk is @xmath17 where @xmath18 is defined as an exponential scale height , @xmath19 is defined as a thickness of disk and equals @xmath20 , and @xmath21 is the surface density . by solving poission s equation for a logarithmic density perturbation , peng et al .
( 1979 ) obtained a criterion for density waves to appear , which is @xmath22 where @xmath23 is the polar coordinate of the starting point from which arms of a galaxy stretch outward on the galactic plane , and @xmath24 is the number of the arms in a spiral galaxy .
based on this criterion , peng ( 1988 ) proposed a method for estimating the thickness of a non - edge - on spiral , and derived the thicknesses of four galaxies .
guthrie ( 1992 ) derived the axial ratios @xmath25 of disc components for 262 edge - on spiral galaxies on print copies of the blue palomar sky survey plates by using a microscope fitted with a micrometer eyepiece .
he then analyzed the distribution of isophotal axial ratios for 888 diameter - limited normal sa - sc galaxies to give information on the true axial ratios @xmath26 , and at last presented the mean value of @xmath27 is @xmath28 .
( 1997 , 1998 ) derived the thicknesses of 72 spirals by using peng s proposal ( peng , 1988 ) and presented some statistical correlations between thickness or flatness and other parameters .
the value of @xmath26 derived by guthrie ( 1992 ) is based on observational work and , should be much more reliable .
so , it is important to compare ma et al.s results ( 1997 , 1998 ) to guthrie s ( 1992 ) . in ma
et al . ( 1997 , 1998 ) , the flatnesses for 72 galactic disks ( @xmath29 ) , which is measured at or reduced down to the surface brightness level @xmath30 magnitudes per square arcsecond , and corrected to the `` face - on '' ( @xmath31 ) . for the galactic extinction , but not for redshift , @xmath32 is from the third catalogue of bright galaxies by de vaucouleurs et al .
( hereafter rc3 ) . ] were given , and the mean value is @xmath33 .
suppose that the value of ratio of radial scale length ( @xmath12 ) over exponential scale height ( @xmath18 ) for an average exponential galactic disk is equal to @xmath26 from guthrie ( 1992 ) , which is 9 .
from freeman ( 1970 ) , we can derive @xmath34 .
thus , we obtain @xmath35 , which is in relative agreement with the mean value of ma et al .
( 1997 , 1998 ) . at the same time
, we calculate the values of @xmath36 for spirals of various types t and list them in table 1 .
t , from the rc3 , is morphological types , and @xmath37 is the dispersion . from this table and table 3 of guthrie ( 1992 ) , we can see that our results are in agreement with guthrie s ( 1992 ) .
except for the data concerning scd galaxies , our results also agree with de grijs ( 1998 ) .
c|ccc t & @xmath36 & @xmath37 + 2 & @xmath38 & 0.09 + 3 & @xmath39 & 0.21 + 4 & @xmath40 & 0.21 + 5 & @xmath41 & 0.18 + 6 & @xmath42 & 0.33 + the structure of this paper is as following : in sect .
2 , we outline the principles of obtaining an inclination and a pitch angle ; sect .
3 presents some statistical properties ; and conclusions will be shown in sect .
as we know , when the line of intersection ( namely , the major axis of the image ) between the galactic plane and tangent plane is taken as the polar axis , it is easily proved that @xmath43 and @xmath44 where @xmath15 and @xmath45 are the polar co - ordinates in the spiral plane and @xmath46 , @xmath47 are the corresponding co - ordinates in the tangent - plane , and @xmath48 is the inclination . if it is possible to represent arms by equiangular spirals as @xmath49 and @xmath50 where @xmath15 and @xmath45 are the polar co - ordinates on the spiral arm in the spiral plane , and @xmath51 , defined as a pitch angle , is the angle between the tangent to the arm and the circle with the constant @xmath15 .
the mathematical form of eq .
( 10 ) in the tangent plane of the celestial sphere is @xmath52 here @xmath53 and @xmath54 where @xmath55 @xmath56 where @xmath57 is an integer .
supposing that ( @xmath58 , @xmath59 ) are the measured co - ordinates of points of the arms , and making use of the least squares method , we set @xmath60=minimum.\ ] ] by direct differentiation with respect to @xmath61 , we obtain the equation for determining the parameter @xmath61 of @xmath62=0.\ ] ] this is a transcendental equation . using the least squares method ,
we can obtain @xmath48 and @xmath51 .
details have been shown in ma et al .
( 1997 , 1998 ) .
our statistical sample contains 72 northern spiral galaxies for which the values of disk thickness and inclination and pitch angle of individual spiral arms are from ma et al .
( 1997 , 1998 ) . except for m 31 , the images ,
which have distinguishable spiral arms , are from the digitized palomar sky survey .
the image of m 31 is digitized from possii ( field no .
295 ) via pds measurement at purple mountain observatory of china .
most galaxies in this sample have the grand design structure .
all the galaxies have the total color indexes ( @xmath63 ) , which are corrected for the differential galactic and internal extinction to `` face - on '' , and for redshift between @xmath64 and @xmath65 bands .
the mean numerical hubble stage indexes ( t ) of the galaxies , which are quoted in the rc3 , are from 2 to 6 , and 2 , 3 , 4 , 5 and 6 correspond to sab , sb , sbc , sc and scd respectively .
this is an inclination - limited sample , i.e. , the values of @xmath66 ) for these galaxies are smaller than the limited - value ( 0.76 ) , where @xmath67 and @xmath68 , taken from the rc3 , are the apparent major and minor isophotal diameters measured at or reduced down to the surface brightness level @xmath30 magnitudes per square arcsecond . clearly , this is not a diameter - limited sample .
( 1998 ) have presented some statistical correlation between thickness or flatness and other parameters . in this section
, we will investigate some other statistical correlations .
1 plots the correlation between the thickness and the total ( `` face - on '' ) absolute magnitude in the b band .
the latter is derived by using the formula @xmath69 here @xmath70 is the total `` face - on '' magnitude from the rc3 , corrected for the galactic and internal extinction , as well as for redshift , d the distance , in units of pc , from ma et al .
( 1997 , 1998 ) .
one can find that there is a significant correlation between thickness and b - band absolute magnitude , i.e. , thicker galaxies are brighter .
based on the careful studies of binggeli et al .
( 1985 ) about the virgo cluster , binggeli et al .
( 1988 ) have thoroughly reviewed what is currently known about the luminosity function @xmath71 .
they have derived the luminosity function @xmath72 for each morphological type separately in the virgo cluster .
furthermore , according to the detailed analysis of the member galaxies for the virgo cluster ( zhao & shao , 1994 ; shao & zhao , 1996 ) , shu et al .
( 1995 ) also investigated its lf for individual morphological types . combining with the luminosity function @xmath72 of the virgo cluster ( binggeli et al .
, 1988 ; shu , et al . , 1995 ) , fig .
1 automatically implies that early - type spirals , which are on average brighter than late - type ones , are also thicker . in this section
, we will present some statistical correlations between @xmath0 and morphological type , luminosity and color for spiral galaxies .
( 1997 , 1998 ) derived the flatnesses ( @xmath29 ) for 72 spiral galaxies .
when we take @xmath73 ( @xmath74 is defined as @xmath75 ) , we can derive the values of @xmath0 for these spirals .
2 shows @xmath0 as a function of different hubble types .
it can be seen that @xmath0 of spiral galaxies decreases smoothly an average along the hubble type , but the dispersion in @xmath0 among galaxies of the same hubble sequence is very large .
this , perhaps , reflects that the intrinsic flattening of spiral disks is smaller for later type galaxies . de grijs ( 1998 ) also shows that galaxies become systematically thinner when going from type s0 to sc .
fig . 3 plots @xmath0 as a function of total ( `` face - on '' ) absolute magnitude in the b band . in fig .
3 , we can see a clear trend reflecting an increase in the values of @xmath0 as the galaxies become brighter for all but a few galaxies . combining this data with the luminosity function @xmath72 of the virgo cluster ( binggeli et al .
, 1988 ; shu et al . , 1995 ) , we find that early - type spirals , which are brighter on average than late - type ones , have larger values of @xmath0 .
4 plots @xmath0 as a function of total galactic color index ( @xmath63 ) in the rc3 and shows a strong trend which suggests that a bluer galaxy has a smaller value of @xmath0 .
roberts & haynes ( 1994 ) studied the physical parameters along the hubble sequence systematically by making use of two primary catalogues , the rc3 and a private catalogue maintained by r. giovanclli & m. haynes .
he presented the well - established trend between morphology and mean color which reveals that the e and s0 galaxies are clearly redder than spirals and that late - type spirals are bluer on average than early - type ones .
so , fig .
4 also implies that later - type galaxies have smaller values of @xmath0 .
tully & fisher ( 1977 ) showed that if a spiral structure is well defined , the opening of the spiral structure could be used to define the inclination of the disk .
it has been shown by danver ( 1942 ) , kennicutt ( 1981 ) and kennicutt & hodge ( 1982 ) that the spiral arm can be represented by a logarithmic spiral form with a constant pitch angle .
grosb@xmath45l ( 1985 ) studied 605 galaxies with inclination angles inferior to @xmath76 , and estimated the position and inclination angles of these galaxies based on the bisymmetric intensity distribution in their outer parts applying a one dimensional fourier transform method .
ma et al . ( 1997 , 1998 ) obtained the inclinations of 72 northern spiral galaxies by fitting the arms on the images .
the rc3 lists the values of @xmath77 and their errors , @xmath67 and @xmath68 are the apparent major and minor isophotal diameters measured at or reduced down to the surface brightness level @xmath78 b magnitudes per square arcsecond .
we calculate the inclinations by use of eq .
5 presents the correlation between them and those obtained by ma et al .
( 1997 , 1998 ) .
this figure shows that , when we take into account the errors in the inclination due to errors in @xmath77 , and although there is some scatter , ma et al.s result ( 1997 , 1998 ) is consistent with what is derived from eq . ( 2 ) .
hubble ( 1926 , 1936 ) introduced an early scheme to classify galaxies .
its concepts are still in use , a sequence starting from elliptical to spiral galaxies , and including lenticular .
this scheme has been extended by some astronomers ( holmberg , 1958 ; de vaucouleurs , 1956 , 1959 ; morgan , 1958 , 1959 ; van den bergh , 1960a , b , 1976 ; sandage , 1961 ; sandage & tammann , 1981 , 1987 ; sandage & bedke , 1993 ) over the years , who tried to employ multiple classification criteria .
the two main systems commonly used are derived from hubble s original classification criteria .
one is the hubble system as explained in detail by sandage ( 1961 ) , sandage & tammann ( 1981 , 1987 ) and sandage & bedke ( 1993 ) , and another , developed by de vaucouleurs ( 1956 , 1959 ) , adds more detailed descriptions to the notation and makes a division and extension of the sc and sbc families by introducing the scd , sd , sdm , sm and i m subdivisions .
galaxy morphological classification is still mainly done visually by dedicated astronomers , based on the hubble s original scheme and its modification .
it is possible for each observer to give slightly different weights to various criteria , although the criteria for classification are accepted generally by them .
lahav et al . (
1995 ) and naim et al .
( 1995 ) investigated the consistency of visual morphological classifications of galaxies from six independent observers .
they found that individual observers agree with one another with combined rms dispersions of between 1.3 and 2.3 type units , typically about 1.8 units of the revised hubble numerical type index t , although there are some cases where the range of types given to it was superior to 4 units in width .
the tightness of the spiral arm , in addition to the degree of resolution in the arm and the relative size of the unresolved nuclear region , are the fundamental criteria in hubble s ( 1926 , 1936 ) classification of spiral galaxies . in order to better understand the nature and origin of the hubble sequence and evaluate the difference of classification , kennicutt ( 1981 ) compared the arm pitch angles as measured by him with the hubble type as determined by sandage & tammann ( 1981 , hereafter st ) and yerkes class by morgan ( 1958 , 1959 ) . from figs . 7 and 8 in kennicutt ( 1981 )
, we can see that , although the st s classification is based almost solely upon disk resolution and the morgan s is based entirely on the central concentration of the galaxies , the trends between the arm pitch angles and the two types are almost the same . in the classifications by de vaucouleurs et al .
( 1976 , rc2 ; 1991 , rc3 ) , all these three criteria are considered .
a most important feature of spiral galaxies is that the stellar content between the spiral arms and the spheroidal component is very different , and the central part is redder than the arm region .
the yerkes system by morgan ( 1958 , 1959 ) contains the information on the state of stellar evolution in the central region of galaxies and it complements the information in the hubble sequence by de vaucouleurs ( 1956 , 1959 ) , which , in cases of conflicting criteria , emphasizes more strongly the strength of population in the arm .
there are other sets of classifications in which disk - to - bulge is the primary discriminant ( van den bergh , 1976 ) or the ratio of disk - to - bulge to spiral arm morphology is primarily considered ( dressler , 1980 ; wilkerson , 1980 ) . among these three criteria , perhaps , spiral arm morphology may be distance independent . until now , there are only a few papers dealing with pitch angles .
so , the mean value of pitch angle for the different hubble types , which is an important parameter in the decision of whether the wkb approximation can be satisfied ( binney & tremaine , 1987 ) , has not been presented .
as we know , the wkb approximation is an indispensable tool for understanding the origin and evolution of a density wave in spiral galaxies .
c|cccc hubble type & @xmath79 & @xmath37 & n + s0/a & @xmath80 & 2.86 & 8 + sa & @xmath81 & 2.80 & 5 + sab & @xmath82 & 3.71 & 14 + sb & @xmath83 & 4.35 & 30 + sbc & @xmath84 & 5.36 & 51 + sc & @xmath85 & 4.25 & 48 + scd & @xmath86 & 5.34 & 21 + sd & @xmath87 & 0.00 & 2 + before deriving the mean value of pitch angle for the different types in the rc3 , it is useful to compare the pitch angles of the common galaxies derived by kennicutt ( 1981 ) and ma et al .
22 of the spirals measured by ma et al .
( 1998 ) were previously done by kennicutt ( 1981 ) , and the average pitch angles are compared in fig .
6 . except for 5 galaxies ,
the results are consistent between kennicutt ( 1981 ) and ma et al .
( 1998 ) . some of the discrepancies can be attributed to our measured error , and the others may be expected from the galaxies themselves ; for example , when a spiral galaxy has two asymmetric arms : different authors ( kennicutt , 1981 ; ma et al .
1998 ) did not measure the same arm in the same galaxy .
table 2 presents the mean values of pitch angle ( @xmath51 ) from the ones measured by kennicutt ( 1981 ) and ma et al .
( 1997 , 1998 ) for the different hubble sequences in the rc3 , where @xmath37 is the dispersion and n the number of galaxies .
the important result from table 2 is that the mean pitch angles of sa - sc are not larger than @xmath88 , so that cotan@xmath89 .
thus the wkb approximation is satisfied at the mean pitch angles from sa - sc , but not by very much .
7 presents the mean values versus the hubble types in the rc3 , and it shows that , from the early to late hubble types , the mean values of the pitch angle increase , although the value for s0/a s is larger than that for sa s . in fact , it is difficult to distinguish between s0/a and sa . besides hubble classification systems , elmegreen & elmegreen ( 1982a , b ; 1987 ) introduced another classification system which is designed to emphasize arm continuity , length and symmetry , and it is related to the presence or strength of density waves .
this system is not dependent on the galaxy hubble sequence and contains 12 distinct arm classes , corresponding to a systematic change from the ragged and patchy arms in ` flocculent ' galaxies to the two symmetric and continuous arms in ` grand design ' ones .
intermediate arm classes show characteristics of both the ` flocculent ' and ` grand design ' types . based on their spiral arm classification system , elmegreen & elmegreen ( 1982a , b ; 1987 ) classified the spiral galaxies in the field , in binary systems , in groups and in clusters , and suggested that bars tend to correlate with spiral density waves , companions may influence ( or generate ) symmetric density waves , grand design galaxies are physically larger than flocculent ones by a factor of @xmath90 1.5 , and grand design galaxies are also preferentially in dense groups .
in this paper , we investigate some statistical correlations about the thickness of galactic disks and pitch angles of spiral arms .
the main conclusions are : ( 1 ) .
early - type spirals , that are brighter on the average , are thicker ; ( 2 ) . the axis ratio ( @xmath0 , here @xmath91 and @xmath92 are defined as two times the radial scale length ( @xmath12 ) and exponential scale height ( @xmath18 ) ) of galactic disk tends to be smaller along the hubble sequence ; ( 3 ) . except for a few galaxies , early - type spiral galaxies have larger values of @xmath0 ; ( 4 ) . @xmath0
correlates strongly with galaxy color ; ( 5 ) .
the inclinations obtained by fitting the pattern of spiral structure with a logarithmic spiral form are nearly the same as those obtained by using the formulas of aaronson et al .
( 1980 ) ; ( 6 ) . the mean measured pitch angles for different hubble sequences in the rc3 are derived , and the results show that the mean pitch angles of sa - sc s are not larger than @xmath88 , so that cotan@xmath89
. thus the wkb approximation can be satisfied at the mean pitch angles from sa - sc s , but not by very much ; ( 7 ) . from early
to late hubble types , the mean value of pitch angle increases , despite some scattering .
although the method , proposed by peng ( 1988 ) for deriving the thickness of a face - on disc , is effective and simple , it relies on a spiral structure theory that predicts that a spiral arm has to end somewhere in the disk .
however , this theory might not be completely right .
for example , zaritsky et al . (
1993 ) has presented k - band ( 2.2 @xmath93 ) images of m 51 , which reveal remarkable dynamical structure not visible in the conventional optical observations , and show that the spiral arms extend significantly further towards the galaxy s center than previously observed . in the optical images ,
the spiral arms begins at a radius of about @xmath94 from the center .
but , the k - band residual image showed the spiral arms wind through an additional @xmath95 beyond that seen in the b - band images of the entire galaxy , and end at about @xmath96 from the center of the galaxy .
if spiral galaxies whose spiral arms do not stop anywhere in the disk exist , the parameters @xmath97 and @xmath98 ( peng , 1988 ; ma et al .
, 1997 , 1998 ) can not be found . in the k - band images , which minimize the effect of dust and maximize sensitivity to the dominant stellar population
, we can derive our reliable values of @xmath97 and @xmath98 .
as an example , we apply our method to the image of m 51 in the k - band and derive the flatness of disk ( @xmath99 ) , which is @xmath100 . comparing this value with peng s ( 1988 ) ( @xmath101 )
, we can find that the value of the flatness based on the k - band image is smaller .
the reason is that , in the k - band image , the effect of dust can be minimized and the values of @xmath97 and @xmath98 may be reliably derived .
we also emphasize that the images , which ma et al .
( 1998 ) used , are from the digitized palomar sky survey , in which many images have burnt - out centers .
there are some pictures that have burnt - out centers in ma et al .
( 1998 ) , but we can change the parameters of display program to minimize the effect .
we are indebted to the anonymous referee for many critical comments and helpful suggestions that have greatly improved our paper , and for correcting our english .
we are grateful to prof .
zaritsky for providing us the image of m 51 in the k - band .
j. ma gratefully acknowledges the support of k. c. wong education foundation , hong kong . this work is supported partly by the chinese national science foundation , no .
19603003 .
aaronson m. , mould j. , huchra j. , 1980 , apj 237 , 655 binggeli b. , sandage a. , tammann g. a. , 1985 , aj 90 , 1681 binggeli b. , sandage a. , tammann g. a. , 1988 , ara&a 26 , 509 binney j. , tremaine s. , 1987 , galactic dynamics .
princeton univ .
press , princeton danver c.c .
, 1942 , ann .
10 de vaucouleurs g. , 1956 , mem .
commonwealth obs .
( mount stromolo ) , vol .
13 de vaucouleurs g. , 1958 , apj 128 , 465 de vaucouleurs g. , 1959 , handbuch der physik 53 , 275 de vaucouleurs g. , de vaucouleurs a. , corwin , h.g.jr , 1976 , the second reference catalogue of bright galaxies .
university of texas , austin de vaucouleurs g. , de vaucouleurs a. , corwin , h.g.jr , et al .
, 1991 , the third reference catalogue of bright galaxies . springer
verlag , new york ( rc3 ) dressler a. , 1980 , apjs 42 , 565 de grijs r. , 1998 , mnras 299 , 595 elmegreen d. m. , elmegreen b. g. , 1982a , mnras 201 , 1021 elmegreen d. m. , elmegreen b. g. , 1982b , mnras 201 , 1035 elmegreen d. m. , elmegreen , b. g. , 1987 , apj 314 , 3 freeman k. c. , 1970 , apj 160 , 811 giovanelli r. , haynes m.p .
, herter t. , et al , 1997 , aj 113 , 22 groot h. , 1925 , mnras lxxxv . 6 , 525 grosb@xmath45l p.j . , 1985 ,
a&ass 60 , 261 guthrie b.n.g . , 1992 ,
a&ass 93 , 255 holmberg e. , 1958 , medd . lund .
ser.2 , no .
136 hubble e.p . , 1926 ,
apj 64 , 321 hubble e.p . , 1936 , the realm of the nebuloe .
yale univ .
press , new haven ma j. , peng qiu - he , gu qiu - sheng , 1997 , apj 490 , l51 ma j. , peng qiu - he , gu qiu - sheng , 1998 , a&ass 130 , 449 kennicutt r.c . , 1981 ,
aj 86 , 1847 kennicutt r.c . , hodge p. , 1982
, apj 253 , 101 morgan w.w . , pasp 1958 , 70 , 364 morgan w.w . ,
pasp 1959 , 71 , 394 lahav o. , naim a. , buta r.j . , et al .
, 1995 , sci 267 , 859 naim a. , lahav o. , buta r.j . , et al .
, 1995 , mnras 274 , 1107 ( sci 267 , 859 ) roberts m.s . ,
haynes m.p . , 1994 , ara&a 32 , 115 peng qiu - he , li xiao - qing , huang ke - liang , et al . , 1979 , sci . in china xxii , 925 peng qiu - he , 1988 , a&a 206 , 18 sandage a.r . , 1961 , the hubble atlas of galaxies .
carnegie institute of washington , washington sandage a.r .
, freeman k.c .
, stokes n.r . , 1970 ,
apj 160 , 831 sandage a.r . , tammann g.a . , 1981 , a revised shapley - ames catalog of galaxies .
carnegie institute of washington , washington sandage a.r .
, tammann g.a .
, 1987 , a revised shapley - ames catalog of bright galaxies .
carnegie institute of washington , washington sandage a.r . ,
bedke j. , 1993 , carnegie atlas of galaxies . carnegie institute of washington , washington shao z.y . , zhao.j.l . , 1996 ,
caa 20 , 273 shu c.g . , zhao j.l .
, tian k.p . ,
1995 , acta astrophy . sin .
15 , 110 tully r.b .
, fisher j.r . , 1977 , a&a 54 , 661 van der kruit p.c . , searle l. , 1981a , a&a 95 , 105 van der kruit p.c . , searle l. , 1981b , a&a 95 , 116 van der kruit p.c . , searle l. , 1982a , a&a 110 , 61 van der kruit p.c . ,
searle , l. , 1982b , a&a 110 , 79 ven den bergh s. , 1960a , apj 131 , 215 ven den bergh s. , 1960b , apj 131 , 558 ven den bergh s. , 1976 , apj 206 , 883 von der pahlen e. , 1911 , astron .
188 , 249 wilkerson m.s . , 1980 ,
apj 241 , l115 zaritsky d. , rix h.w . , rieke m. , 1993 , nat 364 , 313 zhao j.l . , shao z.y . , 1994 , a&a 289 , 89 | this paper presents some statistical correlations of 72 northern spiral galaxies .
the results show that early - type spirals that are brighter , and thicker , and the axis ratios ( @xmath0 ) of the disk tend to be smaller along the hubble sequence .
we also find that @xmath0 correlates strongly with the galaxy s color , and early - type spirals have larger values of @xmath0 .
the inclinations obtained by fitting the pattern of a spiral structure with a logarithmic spiral form are nearly the same as those obtained by using the formulas of aaronson et al .
( 1980 ) . finally , the mean measured pitch angles for the different hubble sequences in the third catalogue of bright galaxies by de vaucouleurs et al .
( 1991 ) are derived . |
pushing the resolution limits of light microscopy , and understanding optical phenomena on scales below the diffraction limit , has been the driving force of what is known today as nano - optics @xcite . to overcome this limit ,
most of the early work was focused on near - field optical microscopy and related techniques @xcite . however , in recent years , new concepts in fluorescence microscopy have pushed the resolution of far - field imaging down to the nanometer range @xcite .
most of these methods @xcite rely on the accurate localization of individual fluorescent markers , that are isolated from one another on the basis of one or more distinguishing optical characteristics , or by selective or random activation of a bright and a dark state @xcite .
determining the location of an isolated fluorescent marker is only limited by photon noise , and not by the diffraction barrier .
a key issue affecting these subwavelength imaging methods is the optical transparency of the media surrounding the light emitters .
taking advantage of the transparency of cells , fluorescence microscopy uniquely provides noninvasive imaging of the interior of cells and allows the detection of specific cellular constituents through fluorescence tagging
. however , certain biological tissues or soft - matter systems ( such as foams or colloidal suspensions ) look turbid due to intense scattering of photons traveling through them @xcite .
the image formed at a given point in the observation plane consists in a superposition of multiple fields , each arising from a different scattering sequence in the medium .
this gives rise to a chaotic intensity distribution with numerous bright and dark spots known as a speckle pattern , producing a blurred image carrying no apparent information about the source position @xcite .
techniques to measure the distance between individual nano - objects without actually imaging their position exist @xcite , fluorescence resonance energy transfer ( fret ) being the most widespread example @xcite .
it relies on the near - field energy transfer between two fluorophores ( donor and acceptor ) emitting at different wavelengths .
the fret signal ( _ e.g. _ the ratio between the intensities emitted by the donor and the acceptor at different wavelengths ) depends on the donor - acceptor distance in the range @xmath0 nm . as such , it is not very sensitive to scattering problems .
however , determining distances between two emitters in the range of 10 to 500 nm in a scattering medium still remains a challenging problem , not accessible either by fluorescence microscopy or fret techniques .
our main goal here is to introduce a new approach to obtain information about the relative distance between two identical incoherent point sources in a disordered environment , based on the analysis of the fluctuations of the emitted light .
this is an issue of much interest , for example , in the study of conformational changes in biomolecules in living tissues . sensing the distance between two incoherent sources in a complex medium
could also provide an alternative to green s function retrieval techniques based on the correlations of the isotropic ambient noise measured at two receivers @xcite . in this paper
, we propose a method to capture the interaction between two identical sources in a scattering environment , based only on the measurement of intensity fluctuations .
the principle of the method is schematically illustrated in fig . 1 , and is based on the analysis of the intensity - intensity correlation function and the intensity fluctuations in the speckle pattern formed by two identical and mutually incoherent point sources .
this approach permits , in principle , to monitor the relative distance between the sources in the range 10 - 500 nm , with a precision that is not limited by diffraction , but by the microstructure of the scattering medium . in application to green s function retrieval in complex media , the approach replaces the two - point field - field correlation of the background noise by a measurement at a single point of the intensity noise due to the two fluctuating sources .
this might simplify the technique , in particular at visible or near - ir frequencies where time - domain field - field correlations are not easy to measure .
the result in this paper also illustrate the fact that multiple scattering , that had long been considered as an unavoidable nuisance , can actually enhance the performance of sensing , imaging and communication techniques @xcite , as already demonstrated in the context of spatio - temporal focusing by time reversal @xcite , wavefront shaping of multiply scattered waves @xcite , or improvement of information capacity of telecommunication channels @xcite .
we consider two point sources of light ( electric dipoles ) located at @xmath1 and @xmath2 in a disordered medium .
the sources are characterized by their electric dipole moments @xmath3 and @xmath4 , that are fluctuating quantities of the form @xmath5 \exp(-i\omega t ) { { \bf u}}_k$ ] with @xmath6 a slowly varying random phase , @xmath7 a complex amplitude and @xmath8 a unit vector defining the orientation of the dipole moment .
this corresponds to a classical model for a quasi - monochromatic temporally incoherent source , such as a fluorescent source emitting at frequency @xmath9 .
we assume that the two sources are uncorrelated ( or mutually incoherent ) , so that @xmath10\exp[-i\phi_2(t)]}=0 $ ] , where the bar denotes averaging over the fluctuations of the sources . using the ( dyadic ) green function @xmath11 of the disordered medium ,
the electric fields at any point @xmath12 can be written : @xmath13 the intensity associated to this field is a time fluctuating and spatially varying quantity that forms a time - dependent speckle pattern .
let us first consider the total power @xmath14 emitted by the two sources .
it reads @xmath15 where @xmath16 is a sphere with radius @xmath17 that encloses the disordered medium and @xmath18 is the speed of light in vacuum . for a _ non
absorbing _ medium , the following relation can be derived from the vector form of green s second identity @xcite ( the frequency dependence in the green function is dropped for simplicity ) : @xmath19 { { \bf p}}_2^ * \ .
\label{eq : green_identity}\end{aligned}\ ] ] from eqs .
( [ eq : field_general]-[eq : green_identity ] ) , we obtain @xmath20 where the notation @xmath21 { { \bf u}}_{j^\prime}$ ] has been introduced for the sake of simplicity .
we first assume that a temporal averaging over the fluctuations of the sources can be performed , in one configuration of the disordered medium ( frozen disorder ) .
the fluctuation time scale of the emitted power can be associated to the coherence time , as usually defined for partially coherent sources @xcite . for fluorescent sources ,
this time is on the order of the lifetime @xmath22 of the excited state . for emission in the visible range , expected orders of magnitude are @xmath23 ns for dye molecules or quantum dots , and @xmath24s for rare - earth ions .
the averaged power @xmath25 is simply the sum of the averaged power emitted by each source independently , since the terms with @xmath26 in eq .
( [ eq : p2 ] ) vanish upon time averaging .
it reads as = ( |p_1|^2 _ 11 + |p_2|^2 _ 22 ) [ eq : pmoy ] where @xmath27 \mathrm{im } g_{jj}$ ] is the electric part of the local density of states ( ldos ) at point @xmath28 @xcite .
however , a cross - term survives in the fluctuations of the total emitted power . indeed , calculating the variance of @xmath14 from eq .
( [ eq : p2 ] ) , one obtains @xcite @xmath29 \ .
\label{eq : varp_time}\ ] ] the imaginary part of the two - point green function @xmath30 in eq .
is known to enter the expression of field - field spatial correlations in random fields , such as blackbody radiation or volume speckle patterns @xcite , the description of time - reversed fields @xcite , and is at the core of green s function retrieval techniques based on ambient noise correlations @xcite .
it is proportional to the cross density of states ( cdos ) , that was introduced in a different context for the description of spatial coherence in complex systems @xcite .
physically , the cdos counts the number of photonic eigenmodes connecting two points ( in our case the source points ) at a given frequency @xcite .
more precisely , the cdos connecting @xmath31 to @xmath28 is given by @xmath32 \mathrm{im } g_{jk}$ ] @xcite . using the cdos , and assuming that the two sources have the same amplitude ( @xmath33 ) , the variance of the total emitted power can be rewritten as @xmath34 this equation is the first result in this paper .
it provides a direct relationship between the temporal fluctuations of the total power emitted by two incoherent sources and the cdos connecting the source points in an arbitrary environment .
this suggests that a retrieval of the amplitude of the cdos ( or equivalently of the imaginary of the green function at two different points ) is possible in a structured medium from a measurement of temporal fluctuations of the emitted power emerging from two incoherent sources .
such a measurement would resemble the green function retrieval from ambient noise correlations , based on the relationship between field - field correlations and the imaginary part of the green function given by the fluctuation - dissipation theorem , initially introduced in the context of electromagnetic thermal fluctuations @xcite .
the generality of this relationship , also valid for field fluctuations in speckle patterns @xcite , has stimulated the development of green s function retrieval techniques in acoustics , seismology or with low - frequency electromagnetic waves @xcite . in this approach ,
the statistical isotropic ambient noise is detected by two receivers , while in the method suggested here the green function is encoded in the noise due to the fluctuations of the two sources .
the possibility to measure power fluctuations instead of field - field correlations might be an advantage for electromanetic green s function retrieval in the visible or near - ir frequency range .
equation ( [ eq : varp_time_2 ] ) also shows that the power fluctuations encode the interdistance between the sources . at this stage , since the cdos @xmath35 is specific to the sample under study and unknown , changes in power fluctuations could reflect changes in the interdistance , but the interdistance could not be determined without solving a difficult inverse problem .
we will see how the problem can be simplified in the presence of multiple scattering in a disordered medium by peforming an ensemble averaging over the configurations of disorder .
we now assume that in addition to a temporal averaging over the fluctuations of the sources , an average over the configurations of the disordered medium can be performed .
a specific situation would be that of sources embedded in a dynamic medium , with configurational changes occurring on a time scale much larger than the characteristic time of the fluctuations of the sources .
an equivalent situation is that of sources moving inside a frozen disordered medium , also on a sufficiently large time scale , as schematically shown in fig .
if both the sources and the disordered medium are fixed ( i.e. the medium itself does not fluctuate ) , an artificial configurational averaging process could be induced by an external moving diffuser surrounding the medium , as shown in fig .
2(b ) . in all these situations both averaging processes can be performed independently and subsequently .
the ( temporal ) variance of the total emitted power averaged over the configurations of the disordered medium is readily obtained from eq .
: @xmath36 where we use the notation @xmath37 for configurational averaging . in practice
, it is often convenient to work with normalized statistical quantities . for later convenience
, we can introduce the contrast of the power fluctuations , that we define as _
[ eq : def_sigmap ] let us point out again the specific use of two non - commuting averaging processes in this definition , one over the temporal fluctuations of the sources and subsequently one over the configurations of the disordered medium . from eqs . and
, the power contrast can be written in terms of the ldos and cdos , leading to _
[ eq : expr_sigmap ] in the next section we will derive a simple relationship between the power contrast @xmath38 and the speckle contrast @xmath39 deduced from the radiated intensity measured in a single direction , or equivalently at a single point in the far field speckle pattern .
a measurement of the total emitted power @xmath14 requires a detection integrated over @xmath40 steradians .
measuring fluctuations of the intensity radiated in a given direction , or intensity correlations between two different directions , could be a more convenient approach in practice .
the time - averaged far - field intensity emerging in a given direction @xmath41 , per unit solid angle , and for a polarization state @xmath42 , is given by & = & + [ inocros ] where @xmath43 is the intensity radiated by the point source @xmath44 .
the latter reads as i_j(*u * ) = _ r r^2 |*e*_*g*(*r*,*r*_j ) p_j * u*_j|^2 [ ijsigma ] with @xmath45 .
the configurational average of the ( time averaged ) total emitted power and fluctuations can be rewritten as & = & _ 4 * u*[exact1 ] + & = & _ ^ _ 4 * u * * u*^ [ exact2 ] where @xmath46 means integration over the solid angle , @xmath47 runs over the two orthogonal polarization states in the far field , and @xmath48 .
we now assume that the radiated field is a random statistically isotropic field , under the only constraints given by eqs . and .
in this case one has @xmath49 and = ( 1 + _ ^ _ * u**u*^ ) [ avip1ip1 ] where @xmath50 is the usual kronecker delta and @xmath51 is a kronecker delta with respect to detection angles .
the derivation of eq . is given in appendix a. at this stage , it is worth noticing that relation between the speckle intensity - intensity correlation function and the fluctuations of the total emitted power is the origin of the so - called @xmath52 correlation known for a speckle produced by a single source @xcite .
if we consider the particular case of a single source at @xmath1 ( @xmath53 ) that does not fluctuate in time , the normalized intensity - intensity correlation function @xmath54 follows directly from eq . and
is given by c(*u*,;*u*^,^ ) & = & -1 + & = & c_0 + ( c_0 + 1)_^ _ * u**u*^ [ c1]where c_0 & = & is the normalized variance of the ldos at the position of the source , that is at the origin of the infinite - range @xmath52 contribution to the speckle correlation function @xcite .
therefore , for a single source , eq . provides another way of deriving the well - known results related to the @xmath52 correlation .
moreover , from eq . , the speckle contrast ( defined as the normalized variance of the intensity in a specific direction and a given polarization channel @xmath47 ) is given by = 1 + 2 c_0 a result already obtained by shapiro @xcite , based on a microscopic diagrammatic approach for scalar waves .
this result leads to non - universal corrections to the rayleigh statistical distribution of the intensity in speckle patterns produced by multiple scattering @xcite . for two fluctuating sources , assuming again that configurational changes in the disordered medium occur on time scales larger than the characteristic time of the source fluctuations
, we also have [ the derivation is similar to that leading to eq .
( [ avip1ip1 ] ) , see appendix a ] : @xmath55 making use of eqs . , and for two sources with the same amplitude @xmath56
, we obtain @xmath57 \nonumber \\ & = \frac{\omega^4}{8 ^ 3 \ , \epsilon_0 ^ 2}|p|^4 \left\langle \rho_{12}^2\right\rangle \ .
\label{eq : fluct_iu}\end{aligned}\ ] ] this equation shows that the intensities corresponding to two different speckle spots ( @xmath58 ) in the angular speckle pattern formed by two incoherent point sources are strongly correlated , with a correlation given by the fluctuations @xmath59 of the cdos .
we end up with the surprising result that for two incoherent sources a cross - term survives averaging , and induces infinite range correlations in the speckle pattern .
this correlation is formally very similar to that given by the @xmath52 contribution in the case of a single source , the fluctuations of the cdos replacing the fluctuations of the ldos .
since the cdos is a two - point quantity , connecting in this specific situation the two source points , the speckle correlations encode the distance between the two sources .
another implication of eq .
is that in the presence of multiple scattering , cdos fluctuations can be accessed from measurements of the directional intensity , and not only from measurements of the total power @xmath14 as suggested initially by eq . .
more precisely , we will now show that the speckle contrast measured in a single speckle spot ( hereafter denoted by @xmath60 ) contains the same information as the contrast @xmath61 in eq .
, that assumed a measurement of the total emitted power . to proceed , we define the speckle contrast for a detection in a given polarization channel @xmath47 as @xmath62 in a similar way as the contrast of the total emitted power in eq . . from eq . , and the relation = one immediatly obtains _
s= = _ p [ sc2 ] or equivalently _
[ scfinal ] this equation is the second important result in this paper .
it shows that for two incoherent sources in a disorder medium , the speckle contrast measured in a single speckle spot , and computed from both temporal averaging over the fluctuations of the sources _ and _ configurational averaging over disorder , is proportional to @xmath63 that characterizes the fluctuations of the cdos connecting the two sources.this means that the information on the interdistance between the sources is encoded in the speckle contrast @xmath60 measured at a single point .
changes in the speckle contrast could therefore be used to detect changes in the interdistance between the two sources .
compared to a measurement of the total emitted power @xmath14 , a measurement of the speckle contrast @xmath39 would require a simpler instrumentation , but at the cost of a substantial reduction of the signal level .
note that a parallel detection of the fluctuations in several speckle spots could be performed using a ccd camera . in practice , a compromise between simplicity in instrumentation and signal - to - noise ratio should be found .
the contrast in eq .
( [ scfinal ] ) would decrease to zero when increasing the interdistance .
this change occurs on a range given by the width of the cdos ( green function ) considered as a function of the interdistance @xmath64 .
this width depends on the microscopic structure of the disordered medium , providing in principle resolution capabilities beyond the free - space diffraction limit .
the determination of the absolute value of the interdistance from the speckle contrast would require an expression of @xmath63 .
the crucial issue in this case is to find the expression of @xmath65 in a disordered medium .
this can be done at least in the weak scattering limit , as discussed in the next section .
another consequence of eq .
( [ scfinal ] ) , or eq . , is that it provides a way to measure @xmath66 in a complex medium from the speckle noise recorded at a single point , which in the context of green s function retrieval might provide an alternative to the two - point field - field correlation technique , as discussed previously in section ii .
when the sources are embedded in a weakly disordered , homogeneous and isotropic medium , it is possible to give an explicit approximate expression of the fluctuations of the cdos @xmath67 \langle ( \mathrm{im } g_{12})^2 \rangle$ ] . in the limit @xmath68 , where @xmath69 is the effective refractive index , @xmath70 ( with @xmath71 the wavelength in vacuum ) and @xmath72 the boltzmann scattering mean free path , the averaged green function is given by the well known result for homogeneous and isotropic media and we can write @xmath73 as @xmath74 where @xmath75 is the effective wavenumber .
the description of the scattering medium by a complex effective wavenumber breaks down when the distance @xmath76 between the emitters approaches the size of the homogeneities ( more precisely the correlation length of disorder , see the discussion in ref . @xcite and references therein ) .
although this might look like a severe limitation , this approximation is in practice very robust and has been shown to model accurately light diffusion through biological tissues .
to lowest order , we can further approximate @xmath77 which , from eqs . , or , and provides an explicit expression of the intensity correlation function , or the speckle contrast , in terms of the relative distance between the sources .
such an explicit expression should be of practical interest , e.g. , for the sensing of the interdistance between emitters embedded in biological tissues .
let us point out that in terms of the detection of _ changes _ in the interdistance ( without measuring its absolute value ) , the method suggested in this paper does not requite any explicit expression of @xmath63 and relies only on eqs .
( [ eq : varp_time_2 ] ) , ( [ eq : expr_sigmap ] ) or ( [ scfinal ] ) .
the results derived in this paper suggest the intriguing possibility that intensity measurements at only one point in a speckle pattern produced by two incoherent sources can provide information about the relative distance between the sources .
moreover , this information is in principle not limited by diffraction .
it can be extracted from the speckle contrast @xmath60 obtained after a proper time and configurational averaging process .
the results also suggests an alternative approach to the green function retrieval technique . in the later the statistical isotropic ambient noise
is detected by two receivers , while in the method suggested here the green function is encoded in the noise due to two fluctuating incoherent sources measured at a single point .
finally , let us note that for fluorescent emitters , the quantum optical equivalent of eqs . , or might be established in terms of decay rate or photocount statistics of a coupled system , and fluctuations could be associated with super - radiant and sub - radiant states .
this quantum treatment is left for future work .
we acknowledge helpful discussions with m. fink and f. scheffold .
this work was supported by the spanish mec through grant no .
fis2012 - 36113 , by the comunidad de madrid through grant no .
s2009/tic-1476 ( microseres project ) , by labex wifi ( laboratory of excellence within the french program `` investments for the future '' ) under references anr-10-labx-24 and anr-10-idex-0001 - 02 psl*. jjs acknowledges an ikerbasque visiting fellowship and rc and gc the hospitality of the dipc at donostia - san sebastian ( spain ) where part of this work was done .
the far - field intensity radiated by two dipoles emerging in a given direction @xmath41 , per unit solid angle , and for a polarization state @xmath42 , can be written as i_(*u * ) = _ r r^2 | _
j=1 ^ 2 * e*_*g*(*r*,*r*_j ) * p*_j |^2 with @xmath45 . assuming that the @xmath40 solid angle is divided in a finite number @xmath78 , of ( solid angle ) pixels , the total power emitted by the two sources is given by p & = & _ 4 i(*u * ) * u*= _ _
u^n i_(*u * ) which can be written in compact matrix form , p & = & ^^ where @xmath79 and @xmath80 is a @xmath81 matrix ( @xmath82 comes from the two orthogonal polarizations of the field for each observation angle in the far - field ) . by using the singular value decomposition ( svd ) , the @xmath80 matrix can be factorized as & = & ^ + ^&= & ^2 ^ where @xmath83 is an unitary @xmath84 matrix , @xmath85 is a @xmath81 rectangular diagonal matrix with nonnegative real numbers on the diagonal and @xmath86 is a @xmath87 unitary matrix .
equation in the main text can now be written as @xmath88= \frac{4\pi}{n } [ \hat{{{\bf p}}}^\dagger\mathbf{v } \mathbf{\sigma}^2 \mathbf{v}^\dagger \hat{{{\bf p } } } ] \nonumber \\ & = \frac{\mu_0 \ , \omega^3}{2 } \sum_{j , j ' = 1}^2 p_{j'}^*p_j \mathrm{im } g_{jj ' } \nonumber \\ & = \frac{\mu_0 \ , \omega^3}{2 } \begin{pmatrix } { { \bf p}}_1^\dagger & { { \bf p}}_2^\dagger \end{pmatrix } \begin{bmatrix } \mathrm{im}{{\bf g}}(11 ) & \mathrm{im}{{\bf g}}(12 ) \\ \mathrm{im}{{\bf g}}(21 ) & \mathrm{im}{{\bf g}}(22 ) \end{bmatrix } \begin{pmatrix } { { \bf p}}_1 \\ { { \bf p}}_2 \end{pmatrix } \label{eq : mat1}\end{aligned}\ ] ] where @xmath89 . in terms of these new matrices ,
the intensity at a given angle @xmath41 and polarization @xmath47 can be computed as @xmath90 where @xmath91 corresponds to the @xmath92 component of the dipole @xmath93 ( with @xmath94 $ ] and @xmath95 ) .
if we assume that the radiated field is statistically isotropic , we can consider @xmath96 as a random unitary matrix statistically independent of the @xmath85 and @xmath97 matrices ( isotropy hypothesis ) .
this is one of the key assumptions in the macroscopic scaling approach of transport theory @xcite . by using the averages over the unitary group ( evaluated by mello in ref .
@xcite ) , u_b u_^ u_u_ b = _ b b _ * u**u* _ and the ensemble average of the intensity is given by @xmath98 which , from eq .
, gives @xmath99 with @xmath100 and @xmath101 i.e. , as expected , under the isotropy hypothesis , the intensity at a given angle is simply proportional to the average of the total radiated power , @xmath102 . after averaging eq .
( [ eq : isum ] ) over the temporal fluctuations of the sources we have @xmath103 and @xmath104 is simply the sum of the individual intensities [ eq . ] i.e. the non - diagonal boxes of @xmath105 do not contribute ( no crosstalk term ) .
the same applies for @xmath106 let us now consider eq .
, & = & _ _*u*,*u* , [ exact3 ] with @xmath107 given by eq . .
computation of eq . involves again @xmath103 together with averages of 4 elements of an unitary random matrix @xcite @xmath108 it is easy now to find : @xmath109 which , in the large @xmath110 limit , gives eq . in the main text .
following a similar procedure , one also obtains @xmath111 which , in the large @xmath110 limit , gives eq . .
e. betzig _
et al . _ ,
science , * 313 * 1642 ( 2006 ) ; s.t .
hess , t.p.k .
girirajan and m.d .
mason , biophys .
j. , * 91 * 4258 ( 2006 ) ; m.j .
rust , m. bates and x. zhuang , nat .
methods * 3 * , 793 ( 2006 ) ; e. toprak and p.r .
selvin , annu .
* 36 * , 349 ( 2007 ) ; j. antelman , c. wilking - chang , s. weiss and x. michalet , nano letters * 9 * 2199 ( 2009 ) ; a. pertsinidis , y. zhang and s. chu , nature * 466 * 647 ( 2010 ) .
r. l. weaver and o. i. lobkis , phys .
lett . * 87 * , 134301 ( 2001 ) ; b.a .
van tiggelen , phys .
* 91 * , 243904 ( 2003 ) ; a. derode , e. larose , m. tanter , j. de rosny , a. tourin , m. campillo , and m. fink , j. acoust .
113 * , 2973 ( 2003 ) ; m. campillo and a. paul , science * 299 * , 547 ( 2003 ) ; m. davy , m. fink and j. de rosny , phys . rev . lett . *
110 * , 203901 ( 2013 ) .
vellekoop and a.p .
mosk , opt .
lett . * 32 * , 2309 ( 2007 ) ; z. yaqoob , d. psaltis , m.s .
feld and c. yang , nature photon . * 2 * , 110 ( 2008 ) ; s. m. popoff , g. lerosey , r. carminati , m. fink , a.c .
boccara and s. gigan phys . rev .
lett . * 104 * , 100601 ( 2010 ) ; i.m .
vellekoop , a. lagendijk and a.p .
mosk , nature photon . * 4 * , 320 ( 2010 ) .
g. j. foschini and m. gans , wireless personal commun .
* 6 * , 311 ( 1998 ) ; i.e. telatar , eur .
telecommun . ,
* 10 * , 585 ( 1999 ) ; s. h. simon , a. l. moustakas , m. stoytchev , and h. safar , phys . today *
54*(9 ) , 38 ( 2001 ) . the cdos defined here is a `` partial '' cdos defined from a particular matrix element @xmath114 of the tensor green function @xmath115 . in ref .
@xcite , a `` global '' cdos was introduced from the trace of @xmath115 in order to wash out the polarization degrees of freedom .
j. j. senz , l. s. froufe - prez , and a. garca - martn , in _ wave scattering in complex media : from theory to applications _ , ed . by s. skipetrov and b. van tiggelen , nato science series ii ,
vol . 107 ( kluwer , dordretch , 2003 ) , p. 175 | we study the fluctuations of the light emitted by two identical incoherent point sources in a disordered environment .
the intensity - intensity correlation function and the speckle contrast , obtained after proper temporal and configurational averaging , encode the relative distance between the two sources .
this suggests the intriguing possibility that intensity measurements at only one point in a speckle pattern produced by two incoherent sources can provide information about the relative distance between the sources , with a precision that is not limited by diffraction .
the theory also suggests an alternative approach to green s function retrieval technique , where the correlations of the isotropic ambient noise detected by two receivers are replaced by a measurement at a single point of the noise due to two fluctuating incoherent sources . |
thermoelectricity has recently received enormous attention due to the constant demand for new and powerful ways of energy conversion .
increasing the efficiency of thermoelectric materials , in the whole range spanning from macro- to nano - scales , is one of the main challenges , of great importance for several different technological applications @xcite .
progress in understanding thermoelectricity at the nanoscale will have important applications for ultra - sensitive all - electric heat and energy transport detectors , energy transduction , heat rectifiers and refrigerators , just to mention a few examples .
the search for optimisation of nano - scale heat engines and refrigerators has hence stimulated a large body of activity , recently reviewed by benenti _
et al . _
@xcite . while most of the investigations have been carried out in two - terminal setups , thermoelectric transport in multi - terminal devices
just begun to be investigated @xcite since these more complex designs may offer additional advantages .
an interesting perspective , for instance , is the possibility to exploit a third terminal to `` decouple '' the energy and charge flows and improve thermoelectric efficiency @xcite .
furthermore , fundamental questions concerning thermodynamic bounds on the efficiency of these setups has been investigated@xcite , also accounting for the effects of a magnetic field breaking the time - reversal symmetry@xcite . in most of the cases
studied so far , however , all but two - terminal were considered as mere probes ; i.e. no net flow of energy and charge through them was allowed .
in other works a purely bosonic reservoir has been used , only exchanging energy ( and not charge ) current with the system @xcite .
is taken as the reference for measuring temperature and energy : @xmath0 ; @xmath1 .
the reservoirs @xmath2 and @xmath3 have small variations in temperature and chemical potential : @xmath4 , @xmath5 .
with @xmath6 we denote a generic coherent scattering region.,width=264 ] a genuine multi - terminal device will however offer enhanced flexibility and therefore it might be useful to improve thermoelectric efficiency .
a full characterization of these systems is still lacking and motivates us to tackle this problem . here
we focus on the simplest instance of three reservoirs , which can exchange both charge and energy current with the system .
a sketch of the thermal machine is shown in fig.[fig : system ] , where three - terminal are kept at different temperatures and chemical potentials connected through a scattering region .
our aim is to provide a general treatment of the linear response thermoelectric transport for this case , and for this purpose we will discuss local and non - local transport coefficients .
note that non - local transport coefficients are naturally requested in a multi - terminal setup , since they connect temperature or voltage biases introduced between two - terminal to heat and charge transport among the remaining terminals .
we will then show that the third terminal could be exploited to improve thermoelectric performance with respect to the two - terminal case .
we will focus our investigations on the efficiency at maximum power @xcite , i.e. of a heat engine operating under conditions where the output power is maximized . such quantity , central in the field of finite - time thermodynamics @xcite , is of great fundamental and practical relevance to understand which systems offer the best trade - off between thermoelectric power and efficiency .
the paper is organized as follows . in section [ sec:3terminals ]
we briefly review the linear response , onsager formalism for a generic three - terminal setup . we will discuss the maximum output power and trace a derivation of all the local and non - local transport coefficients . in section [ sec : efficiency ]
we extend the concept of carnot bound at the maximum efficiency to the three - terminal setup and we derive analytical formulas of the efficiency at maximum power in various cases , depending on the flow of the heat currents .
these expressions are written in terms of generalized dimensionless figures of merit .
note that the expressions derived in section [ sec:3terminals ] and [ sec : efficiency ] are based on the properties of the onsager matrix and on the positivity of the entropy production .
therefore they hold for non - interacting as well as interacting systems .
this framework will then be applied in section [ sec : models ] to specific examples of non - interacting systems in order to illustrate the salient physical picture .
namely , we will consider a single quantum dot and two dots in series coupled to the three - terminal . finally section [ sec : conclusions ] is devoted to the conclusions .
the system depicted in fig . [ fig : system ] is characterized by three energy and three particle currents ( @xmath7 and @xmath8 , respectively ) flowing from the corresponding reservoirs , which have to fulfill the constraints : @xmath9 ( positive values being associated with flows _ from _ the reservoir _ to _ the system ) . in what follows
we will assume the reservoir 3 as a reference and the system to be operating in the linear response regime , i.e. set @xmath10 and write @xmath11 with @xmath12 and @xmath13 for @xmath14 , and @xmath15 is the boltzmann constant .
under these assumptions the relation between currents and biases can then be expressed through the onsager matrix @xmath16 of elements @xmath17 via the identity : @xmath18 where @xmath19 and @xmath20 are the generalized forces , and where @xmath21 are the heat currents of the system , the corresponding currents to reservoir 3 being determined from @xmath22 and @xmath23 via the conservation laws of eq .
( [ particle ] ) . in our analysis
we take @xmath16 to be symmetric ( i.e. @xmath24 ) by enforcing time reversal symmetry in the problem .
we also remind that , due to the positivity of the entropy production rate , such matrix has to be semi - positive definite ( i.e. @xmath25 ) and that it can be used to describe a two - terminal model connecting ( say ) reservoir 1 to reservoir 3 by setting @xmath26 for all @xmath27 . for a two - terminal model
the elements of the onsager matrix @xmath16 can be related to four quantities which gauge the transport properties of the system under certain constraints .
specifically these are the electrical conductance @xmath28 and the peltier coefficient @xmath29 ( evaluated under the assumption that both reservoirs have the same temperature ) , and the thermal conductance @xmath30 and the thermopower ( or seebeck coefficient ) @xmath31 ( evaluated when no net charge current is flowing through the leads ) .
when generalized to the multi - terminal model these quantities yield to the introduction of non - local coefficients , which describe how transport in a reservoir is influenced by a bias set between two other reservoirs . for a two - terminal configuration
the thermopower relates the voltage @xmath32 that develops between the reservoirs to their temperature difference @xmath33 under the assumption that no net charge current is flowing in the system , i.e. @xmath34 . a generalization of this quantity to the multi - terminal scenario can be obtained by introducing the matrix of elements @xmath35 with local ( @xmath36 ) and non - local ( @xmath37 ) coefficients , @xmath38 being the electron charge . in this definition , which does not require the control of the heat currents , we have imposed that the particle currents in all the leads are zero ( the voltages are measured at open circuits ) _ and _ that all but one temperature differences are zero ( of course this last condition is not required in a two - terminal model ) .
it is worth observing that eq .
( [ eq : thermopower ] ) differs from other definitions proposed in the literature .
for example in ref .
@xcite a generalization of the two - terminal thermopower to a three - terminal system , was proposed by setting to zero one voltage instead of the corresponding particle current . while operationally well defined
, this choice does not allow one to easily recover the thermopower of the two - terminal case ( in our approach instead this is rather natural , see below ) .
finally in the probe approach presented in refs . @xcite it was possible to study a multi - terminal device by using an effective two - terminal system only , because the heat and particle currents of the probe terminals are set to vanish by definition . therefore , within this approach , there are no chances of having non - local transport coefficients . in the three - terminal scenario
we can use eq .
( [ omatrix ] ) to rewrite the elements of the matrix ( [ eq : thermopower ] ) . in particular introducing the quantities @xmath39 we get ( see [ app : transport ] for details )
@xmath40 which yields , correctly , @xmath41 as the only non - zero element , by taking the two - terminal limits detailed at the end of the previous section . in a two - terminal configuration the electric conductance describes how the electric current depends upon the bias voltage between the two - terminal under isothermal conditions , i.e. @xmath42 . the generalization to many - terminal systems is provided by the following matrix : @xmath43 using the three - terminal onsager matrix ( [ omatrix ] ) we find @xmath44 which , in the two - terminal limit where reservoir 2 is disconnected from the rest , gives @xmath45 as the only non - zero element . the thermal conductance for a two - terminal is the coefficient which describes how the heat current depends upon the temperature imbalance @xmath33 under the assumption that no net charge current is flying through the system , i.e. @xmath46 . in the multi - terminal scenario
this generalizes to @xmath47 where one imposes the same constraints as those used for the thermopower matrix ( [ eq : thermopower ] ) , i.e. no currents and @xmath48 for all terminals but the @xmath27-th . for a three - terminal case , using eq .
( [ defl2 ] ) this gives @xmath49 @xmath50 and @xmath51 once more , in the two - terminal limit where the reservoir 2 is disconnected from the rest , the only non - zero element is @xmath52 . in a two - terminal configuration
the peltier coefficient relates the heat current to the charge current under isothermal condition , i.e. @xmath53 . for multi - terminal systems this generalizes to the matrix @xmath54 which can be shown to be related to the thermopower matrix ( [ eq : thermopower ] ) , through the onsager reciprocity equations , i.e. @xmath55 ( @xmath56 being the magnetic field on the system),@xcite from which , using eqs .
( [ sloc ] ) and ( [ sloc1 ] ) , one can easily derive for the three - terminal case the dependence upon the onsager matrix @xmath16 .
in order to characterize the properties of a multi - terminal system as a heat engine we shall now analyze its efficiency . generalizing the definition for the efficiency of a two - terminal machine @xcite
, we define the steady state heat to work conversion efficiency @xmath57 , for a three - terminal machine , as the power @xmath58 generated by the machine ( which equals to the sum of all the heat currents exchanged between the system and the reservoirs ) , divided by the sum of the heat currents _ absorbed _ by the system , i.e. @xmath59 where the symbol @xmath60 in the denominator indicates that the sum is restricted to positive heat currents only , and where in the last expression we used eq .
( [ particle ] ) to express @xmath61 in terms of the other two independent currents@xcite .
the definition ( [ eff ] ) applies only to the case in which @xmath62 is positive . since the signs of the heat currents
@xmath63 are not known _ a priori _ ( they actually depend on the details of the system ) , the expression of the efficiency depends on which heat currents are positive .
for the three - terminal system depicted in fig .
[ fig : system ] we set for simplicity @xmath64 and focus on those situations where @xmath65 is negative ( positive values of @xmath65 being associated with regimes where the machine effectively works as a refrigerator which extract heat from the coldest reservoir of the system ) . under these conditions
the efficiency is equal to @xmath66 when both @xmath67 and @xmath68 are positive , or @xmath69 when for @xmath70 or @xmath3 only @xmath71 is positive .
the carnot efficiency represents an upper bound for the efficiency and is obtained for an infinite - time ( carnot ) cycle . for a two - terminal thermal machine the carnot efficiency
is obtained by simply imposing the condition of zero entropy production , namely @xmath72 . if the two reservoirs are kept at temperatures @xmath73 and @xmath74 ( with @xmath75 ) , from the definition of the efficiency , eq .
, one gets the two - terminal carnot efficiency @xmath76 .
the carnot efficiency for a three - terminal thermal machine is obtained analogously by imposing the condition of zero entropy production , when a reservoir at an intermediate temperature @xmath77 is added . if @xmath67 only is positive as in eq .
( [ case1 ] ) , one obtains @xmath78 where @xmath79 .
note that eq .
( [ eq : carnot3t_a ] ) is the sum of the two - terminal carnot efficiency @xmath80 and a term whose sign is determined by @xmath81 . since @xmath82 , @xmath83 and @xmath84 , it follows that @xmath85 is always _ reduced _ with respect to its two - terminal counterpart @xmath80 .
analogously if only @xmath68 is positive , one obtains @xmath86 , \ ] ] which again can be shown to be reduced with respect to @xmath80 , since @xmath87 , @xmath88 , @xmath89 , and @xmath90 .
we notice that this is a hybrid configuration ( not a heat engine , neither a refrigerator ) : the hottest reservoir absorbs heat , while the intermediate - temperature reservoir releases heat .
however , the heat to work conversion efficiency is legitimately defined since generation of power ( @xmath91 ) can occur in this situation . finally , if both @xmath67 and @xmath68 are positive as in eq .
( [ case2 ] ) one obtains @xmath92 since @xmath64 , the term that multiplies @xmath93 is positive so that @xmath94 is reduced with respect to the two - terminal case .
it is worth noticing that , in contrast to the two - terminal case , the carnot efficiency can not be written in terms of the temperatures only , but it depends on the details of the system .
moreover , note that the carnot efficiency is unchanged with respect of the two - terminal case if @xmath95 in ( [ eq : carnot3t_a ] ) or if @xmath96 in ( [ eq : carnot3t_b ] ) . indeed ,
in this situation the quantities @xmath97 are equal to one , making the extra terms in eqs . or to vanish .
the above results for the carnot efficiency could be generalized to many - terminal systems . in particular
, we conjecture that , given a system that works between @xmath73 and @xmath74 ( with @xmath75 ) and adding an arbitrary number of terminals at intermediate temperatures will in general lead to carnot bounds smaller than @xmath80 .
on the other hand , adding terminals at higher ( or colder ) temperatures than @xmath73 and @xmath74 will make @xmath98 increase .
notice that within linear response and via eq .
( [ omatrix ] ) we can express the carnot efficiencies ( [ eq : carnot3t_a])-([eq : carnot3t_b ] ) in terms of the generalized forces @xmath99 .
the efficiency at maximum power is the value of the efficiency evaluated at the values of chemical potentials that maximize the output power @xmath58 of the engine . in the two - terminal case the efficiency at maximum power
can be expressed as @xcite @xmath100 where @xmath101 is a dimensionless figure of merit which depends upon the transfer coefficient of the system .
the positivity of the entropy production imposes that such quantity should be non - negative ( i.e. @xmath102 ) , therefore @xmath103 is bounded to reach its maximum value @xmath104 only in the asymptotic limit of @xmath105 ( curzon - ahlborn limit @xcite within linear response @xcite ) .
for the three - terminal configuration the output power is a function of the four generalized forces @xmath106 introduced in eq .
( [ omatrix ] ) , i.e. @xmath107 in the linear regime this is a quadratic function which can be maximized with respect to @xmath108 and @xmath109 while keeping @xmath110 and @xmath111 constant ( the existence of a maximum being guaranteed by the the positivity of the entropy production ) .
the resulting expression is @xmath112 where @xmath113 $ ] is a positive semi - definite matrix , see [ app : cholesky ] , whose elements depends on the onsager coefficients via the identities @xmath114 indicating with @xmath115 the eigenvalues of @xmath116 we can then further simplify eq .
( [ wmax1 ] ) by writing it as @xmath117 where @xmath118 is the geometric average of system temperatures , while the angle @xmath119 identify the rotation in the @xmath110 , @xmath111 plane which defines the eigenvectors of @xmath116 .
we call the parameter @xmath120 _ three - terminal power factor_. it relates the maximum power to the temperature difference : by construction it fulfills the inequality @xmath121 , the maximum being achieved for @xmath122 ( i.e. by ensuring that @xmath123 coincides with the eigenvector of @xmath116 associated with its largest eigenvalue @xmath124 ) . note that in the two - terminal limit we have @xmath125 , @xmath126 , @xmath127 , so that the usual two - terminal power factor @xmath128 is recovered . exploiting eq .
( [ neweq1 ] ) we can now write the efficiency at maximum power for the three cases detailed in eqs .
( [ case2 ] ) and ( [ case1 ] ) .
specifically we have @xmath129 @xmath130 and @xmath131 where we have defined the parameters @xmath132 , @xmath133 and @xmath134 and introduced the following generalized @xmath135 coefficients : @xmath136 the efficiencies ( [ eta1 ] ) , ( [ eta2 ] ) and ( [ eta12 ] ) can also be expressed in terms of the corresponding carnot efficiencies given in eqs . , and , obtaining the following equations which mimic eq .
( [ etamax ] ) of the two - terminal case : @xmath137 @xmath138 @xmath139 where we have introduced the constants @xmath140 and the combinations of figures of merit @xmath141 notice also that in writing eq .
( [ eq : eta / etac12 ] ) we retained only the leading order neglecting contributions of order @xmath142 or higher .
the above expressions can be used to provide a generalization of the curzon - ahlborn limit efficiency for a multi - terminal quantum thermal device . indeed using the cholesky decompositions on the onsager matrix
, we can prove that the constants @xmath143 , @xmath144 defined in eqs .
( [ eq : c1 ] ) , ( [ eq : c2 ] ) are positive , see [ chsec ] for details .
this fact together with the positivity of the quantities @xmath145 , that we have checked numerically , implies that the efficiencies @xmath146 are always upper bounded by half of the associated carnot efficiencies , i.e. @xmath147 the inequality being saturated when the generalized @xmath135 coefficients ( [ ztgen ] ) diverge .
an analogous conclusion can be reached also for ( [ eq : eta / etac12 ] ) , yielding @xmath148 in this case @xmath149 is no longer guaranteed to be positive due to the presence of @xmath150 .
still the inequality ( [ ca12 ] ) can be derived by observing that the quantities @xmath149 and @xmath151 entering in the rhs of eq .
( [ eq : eta / etac12 ] ) have always the same sign .
in this section we shall apply the theoretical framework developed so far to two specific non - interacting systems attached to three terminals .
namely , we will discuss the case of a single dot and the case of two coupled dots , in the absence of electron - electron interaction ( which can not be dealt within the landauer - bttiker formalism ) .
our aim is to show that one can easily find situations where the efficiency and output power are enhanced with respect to the two - terminal case .
furthermore , through the example of the single dot , we find the conditions that guarantee the non - local thermopowers to vanish . the coherent flow of particles and heat through a non - interacting conductor can be described by means of the landauer - bttiker formalism . under the assumption that all dissipative and phase - breaking processes take place in the reservoirs ,
the electric and thermal currents are expressed in terms of the scattering properties of the system @xcite .
for instance , in a generic multi - terminal configuration the currents flowing into the system from the @xmath152-th reservoir are : @xmath153 , \label{curr1}\\ & j^q_i=\frac{1}{h}\sum_{j \neq i } \int_{-\infty}^{\infty } de \ , ( e-\mu_i ) \
, \mathcal{t}_{ij}(e)\ , [ f_i(e)-f_j(e ) ] , \label{curr2 } \end{aligned}\ ] ] where the sum over @xmath27 is intended over all but the @xmath152-th reservoir , _ h _ is the planck s constant , @xmath154 is the transmission probability for a particle with energy @xmath155 to transit from the reservoir @xmath27 to reservoir @xmath152 , and where finally @xmath156 + 1\}^{-1}$ ] is the fermi distribution of the particles injected from reservoir @xmath152 ( notice also that we are considering currents of spinless particles ) . in what follows
we will use the above expressions in the linear response regime where @xmath157 and @xmath158 , and compute the associated onsager coefficients ( [ omatrix ] ) , see [ app : scattering ] .
in this section we study numerically a simple model consisting of a quantum dot with a single energy level @xmath159 , coupled to three fermionic reservoirs , labeled @xmath2 , @xmath3 , and @xmath160 , see fig . [
fig : dot ] . for simplicity ,
the coupling strength to electrodes 1 and 3 are taken equal to @xmath161 , while the coupling strength to electrode 2 is denoted by @xmath162 . in particular
we want to investigate how the efficiencies , output powers and transport coefficients evolve when the system is driven from a two - terminal to a three - terminal configuration , that is by varying the ratio @xmath163 .
the two - terminal configuration corresponds to @xmath164 and the third terminal is gradually switched on by increasing @xmath163 . as detailed in [ app : scattering ] , the transmission amplitudes between each pair of terminals can be used to evaluate the onsager coefficients @xmath17 the resulting expression being provided in eqs .
( [ eq : l_ij_2b ] ) .
once the matrix @xmath17 is known , all the currents flowing through the system , efficiencies , output powers and transport coefficients can be calculated within the framework developed in the previous sections .
is connected to three fermionic reservoirs @xmath2 , @xmath3 , and @xmath160 . the chemical potential and temperature of the reservoir 3 are assumed as the reference values @xmath165 and @xmath166 .
the constants @xmath161 and @xmath162 represent the coupling between the system and the various reservoirs ( see [ appesin ] for details ) .
a zero value of @xmath162 corresponds to disconnecting the reservoir 2 from the system : in this regime the model describes a two - terminal device where reservoirs 1 and 3 are connected through the single dot.,width=264 ] in fig .
[ fig : carnot2d ] we show how the carnot efficiency @xmath98 depends on the temperature differences @xmath167 and @xmath168 , when the chemical potentials are chosen to guarantee maximum output power , i.e. , fixing the generalized forces @xmath169 in order to maximize @xmath58 .
as we can see , @xmath98 increases linearly along any `` radial '' direction defined by a relation @xmath170 , where @xmath171 is a constant .
in particular , the dashed lines corresponding to @xmath172 , @xmath173 , and @xmath174 separate the different regimes discussed in sec . [ carnot ] : for @xmath175 the system absorbs heat only from reservoir @xmath2 ( if @xmath176 ) or from @xmath3 and @xmath160 ( if @xmath177 ) ; for @xmath178 the system absorbs heat from reservoirs from @xmath2 and @xmath3 ( if @xmath176 ) or from @xmath160 only ( if @xmath177 ) ; finally , for @xmath179 and @xmath180 the system absorbs heat only from reservoir @xmath3 ( if @xmath181 ) or from @xmath2 and @xmath160 ( if @xmath182 ) . in the case when only one heat flux is absorbed the carnot efficiency is given by eq .
( [ eq : carnot3t_a ] ) or eq .
( [ eq : carnot3t_c ] ) , while it is given by eq .
( [ eq : carnot3t_b ] ) if two heat fluxes are absorbed .
( density plot ) of the three - terminal system depicted in fig .
[ fig : dot ] , as a function of the gradients of temperature in reservoirs @xmath2 and @xmath3 ( the chemical potentials @xmath183 and @xmath184 being chosen to guarantee maximum output power @xmath62 ) . the coupling with the reservoirs have been set to have a symmetric configuration with respect to @xmath2 and @xmath3 ( i.e. @xmath185 ) .
note that @xmath98 increases linearly along any radial direction defined by a relation @xmath186 , where @xmath171 is a constant .
in particular , the dashed lines corresponding to @xmath172 , @xmath173 and @xmath174 separate different regimes discussed in sec . [ carnot ] . the numbers in each region identify the reservoirs from which the heat is absorbed .
parameter values : @xmath187 , @xmath188.,width=377 ] in figs .
[ fig : eta_w_vs_gammab ] and [ fig : etawmax_wmax_vs_gammab ] , we show how the efficiency eq .
( [ eff ] ) , the output power eq .
( [ w ] ) , the efficiency at maximum output power eqs .
( [ eta1])-([eta12 ] ) and the maximum output power eq .
( [ wmax1 ] ) , vary when the system is driven from a two - terminal to a three - terminal configuration , _
i. e. _ by varying the ratio @xmath163 .
we set opposite signs for @xmath189 and @xmath190 , so that the system absorbs heat only from the hottest reservoir @xmath2 , and @xmath191 , in such a way that the carnot efficiency @xmath98 coincides with that of a two - terminal configuration , namely @xmath192 .
interestingly , we proved that increasing the coupling @xmath162 to the reservoir @xmath3 may lead to an improvement of the performance of the system .
in particular , as shown in fig . [
fig : eta_w_vs_gamma ] , the efficiency and the output power can be enhanced _ at the same time _ at small couplings @xmath162 , exhibiting a maximum around @xmath193 and @xmath194 , respectively . in fig .
[ fig : etawmax_wmax_vs_gammab ] we show results for the same quantities but at the maximum output power [ @xmath195 and @xmath196 . in this case , while @xmath197 still increases with @xmath162 , the corresponding efficiency decreases approximately linearly . ,
normalized over the associated carnot limit computed as in sec .
[ carnot ] , as a function of the coupling to the reservoir @xmath3 .
note that as @xmath163 is switched on , the efficiency increases until it reaches a maximum around @xmath198 , and then it decreases .
right panel : output power @xmath62 extracted from the system , as a function of the coupling to the reservoir @xmath3 .
parameters : @xmath199 , @xmath188 , @xmath200 , @xmath201 , and @xmath191.,width=453 ] , normalized over the carnot limit , as a function of the coupling to the reservoir @xmath3 .
right panel : maximum output power @xmath197 extracted from the system , as a function of the coupling to the reservoir @xmath3 .
parameters : @xmath199 , @xmath202 , @xmath203 , and @xmath191.,width=453 ] in figs .
[ fig : eta_w_vs_gamma ] and [ fig : etawmax_wmax_vs_gamma ] we show the same quantities as in figs .
[ fig : eta_w_vs_gammab ] and [ fig : etawmax_wmax_vs_gammab ] , but as a function of the coupling @xmath161 for two values of @xmath162 ( @xmath164 and @xmath204 ) . from fig .
[ fig : eta_w_vs_gamma ] we can see that at small @xmath161 the coupling to a third terminal can enhance both the efficiency ( for @xmath205 ) and the power ( for @xmath206 ) . in fig .
[ fig : etawmax_wmax_vs_gamma ] we note that , both for the two- and the three - terminal system , the efficiency at maximum power tends to @xmath207 in the limit @xmath208 , while the output power vanishes . for two - terminal the result
is well known , since a delta - shaped transmission function leads to the divergence of the figure of merit @xmath135 @xcite . correspondingly , the efficiency at maximum power saturates the curzon - ahlborn bound @xmath207 .
the same two - terminal energy - filtering argument explains the three - terminal result .
indeed , we found numerically that for @xmath208 the chemical potentials optimizing the output power are such that @xmath209 . since
also the temperatures are chosen so that @xmath95 , we can conclude that terminals @xmath3 and @xmath160 can be seen as a single terminal . , normalized over the carnot limit , as a function of the coupling energy @xmath161 .
right panel : output power @xmath62 extracted from the system , as a function of the coupling energy @xmath161 . in both cases ,
the full red curves correspond to a three - terminal configuration with @xmath204 , while the dashed blue curve refer to the two - terminal case ( @xmath164 ) .
parameters : @xmath188 , @xmath210 , @xmath211 , and @xmath191.,width=453 ] , normalized over the carnot limit , as a function of the coupling energy @xmath161 .
right panel : maximum output power @xmath197 extracted from the system , as a function of the coupling energy @xmath161 . in both cases ,
the full red curves correspond to a three - terminal configuration with @xmath204 , while the dashed blue curves refer to the two - terminal case ( @xmath164 ) .
parameters : @xmath188 , @xmath212 , and @xmath191.,width=453 ] in this section we show analytically that the non - local thermopowers are always zero in this model , while the local ones are equal .
we consider a general situation , with three different coupling parameters : @xmath213 , @xmath214 and @xmath215 , with @xmath216 . under these assumptions ,
the transmissions are given at the end of [ app : scattering ] .
substituting these expressions in eqs . and , we find : @xmath217 this result is a direct consequence of the factorization of the energy dependence of the transmission probabilities , which are all proportional to the same function @xmath218 , as shown in eq .
( [ eq : t_cd ] ) .
such factorization allows us to rewrite the onsager s coefficients as in eq .
( [ eq : l_ij_2b ] ) and derive eq .
( [ eq:1dotthermopower ] ) .
the fact that the non - local thermopowers , for example @xmath219 , are zero can be understood as follows . consider first the case in which @xmath220 and terminal @xmath3 behaves as a voltage probe .
if so , from the condition @xmath221 we derive @xmath222 . due to the factorization of the energy dependence in the transmissions we obtain @xmath223 .
hence , @xmath224 does not depend on the coupling @xmath162 .
if in particular we consider @xmath225 , because of the symmetry of the system under exchange of the terminal @xmath2 and @xmath160 we have @xmath226 .
we can therefore conclude that , independently of the coupling @xmath162 , the probe voltage condition for terminal @xmath3 implies @xmath227 .
it can be shown that such result remains valid even when @xmath228 but @xmath229 , as requested in the calculation of the thermopower @xmath219 . as a result , @xmath230 .
the same argument can be repeated for the current @xmath231 with the terminal @xmath2 acting as a voltage probe , leading to @xmath232 .
let us now consider a system made of two quantum dots in series , each with a single energy level , coupled to three fermionic reservoirs .
this system is described by the hamiltonian : @xmath233.\ ] ] we call @xmath234 the hopping energy between the dots , and we assume that dot @xmath16 is coupled to the left lead ( @xmath2 ) , dot @xmath235 is coupled to the right lead ( @xmath160 ) and that both are coupled to a third lead ( @xmath3 ) ( see fig.[fig : bidot ] ) .
the self energies describing these couplings are : @xmath236 , \quad \sigma_2 = \left [ \begin{array}{cc } \sigma_{2 } & 0 \\ 0 & \sigma_2 \end{array } \right],\quad \sigma_3 = \left [ \begin{array}{cc } 0 & 0 \\ 0 & \sigma_3 \end{array } \right].\ ] ] in the wide - band approximation , we assume that these quantities are energy - independent and they can be written as purely imaginary numbers @xmath237 .
the self energies thus become : @xmath238 , \quad \sigma_2 = \left [ \begin{array}{cc } -i\frac{\gamma_2}{2 } & 0 \\ 0 & -i\frac{\gamma_2}{2 } \end{array } \right ] , \quad \sigma_3 = \left [ \begin{array}{cc } 0 & 0 \\ 0 & -i\frac{\gamma_3}{2 } \end{array } \right].\ ] ] the retarded green s function of the system is then : @xmath239^{-1 } & = \left [ \begin{array}{cc } e - e_{l}-\sigma_1-\sigma_2 & t \\ t & e - e_r-\sigma_3-\sigma_2 \end{array } \right]^{-1}=\cr & = \frac{1}{\det [ \mathcal{g}]}\left [ \begin{array}{cc } e - e_r + i\frac{\gamma_3+\gamma_2}{2 } & -t \\ -t & e - e_l + i\frac{\gamma_1+\gamma_2}{2 } \end{array } \right],\end{aligned}\ ] ] with , @xmath3 and @xmath160 . the chemical potential and temperature of reservoir @xmath160
are assumed as the reference values @xmath165 and @xmath166 .
a two - terminal configuration is obtained in the case in which the coupling to reservoir @xmath3 ( equal for both the dots ) vanishes ( @xmath164).,width=264 ] @xmath240=\big(e - e_l + i\frac{\gamma_1+\gamma_2}{2}\big)\big(e - e_r + i\frac{\gamma_3+\gamma_2}{2}\big ) - t^2.\ ] ] let us now define the broadening matrices as @xmath241 : @xmath242 , \quad \quad
\gamma_2 = \left [ \begin{array}{cc } \gamma_2 & 0 \\ 0 & \gamma_2 \end{array } \right ] , \quad \gamma_3 = \left [ \begin{array}{cc } 0 & 0 \\ 0 & \gamma_3 \end{array } \right].\ ] ] the matrix of transmission probability @xmath243 between each pair of reservoirs is then given by the fisher - lee formula @xcite @xmath244 .\ ] ] for the system under consideration we obtain @xmath245|^2}\,t^2,\ ] ] @xmath246|^2}\,\left[(e - e_r)^2 + \left(\frac{\gamma_3 + \gamma_2}{2}\right)^2 + t^2\right],\ ] ] @xmath247|^2}\,\left[(e - e_l)^2 + \left(\frac{\gamma_1 + \gamma_2}{2}\right)^2 + t^2\right].\ ] ] at this point , it is clear that the energy dependence of the transmission matrix can not be factorized as for the single dot case .
this model is hence the simplest in which we can observe finite non - local thermopowers and an increase of both the power and the efficiency of the corresponding thermal machine .
we find that the behavior of such quantities as functions of the various parameters is qualitatively very similar to the case of the single dot , thus confirming that a third terminal could improve the performance of a quantum machine .
since in this system all the transport coefficients are different from zero , it is instructive to study the behavior of the generalized figures of merit defined in eq .
( [ ztgen ] ) . in fig .
[ fig : zt_bidot ] we show , in the configuration with only one positive heat flux ( @xmath248 ) , @xmath249 ( dotted line ) , @xmath250 ( dashed line ) and @xmath251 ( full line ) .
we investigate their behavior as a function of the coupling @xmath162 and of the total coupling @xmath161 .
note that in the two - terminal limit ( @xmath252 ) @xmath251 reduces to the standard thermoelectric figure of merit @xmath135 , while @xmath249 and @xmath250 tend to zero .
when we turn on the interaction with the reservoir @xmath3 ( left panel ) , we notice that the figure of merit @xmath251 decreases , while the figures of merit @xmath250 and @xmath249 increase their absolute values . from the behavior as a function of the total coupling @xmath161 we can see that in the limit of @xmath253-shaped transmission function ( @xmath254 )
, the figures of merit diverge , leading to the carnot efficiency , while in the limit of broad transmission window ( @xmath255 ) , all the figures of merit go to zero and we recover the case of zero efficiency .
( dotted line ) , @xmath250 ( dashed line ) and @xmath251 ( full line ) as a function of the coupling to the bottom reservoir @xmath162 ( left panel ) , and as a function of the total coupling @xmath161 ( right panel ) .
parameter values : @xmath256 @xmath257 , @xmath258 @xmath257 , @xmath259 @xmath257 ( left panel ) and @xmath260 @xmath257 ( right panel).,width=453 ] to lead 2 .
the full red line corresponds to @xmath261 , while the dashed ( blue ) line corresponds to @xmath262 .
( right panel ) local thermopowers as a function of the coupling @xmath162 to lead 2 .
the full ( red ) line corresponds to @xmath263 , while the dashed ( blue ) line corresponds to @xmath264 .
parameter values as in fig .
[ fig : zt_bidot ] .
, width=453 ] as mentioned before , the fact that the energy - dependence of the transmission matrix for the double dot can not be factorized is sufficient to guarantee finite non - local thermopowers , as shown in the left panel of fig .
[ fig : thermopowers_bidot ] . as a function of @xmath162
, @xmath219 starts from zero , while @xmath265 starts from a finite value .
this different behavior for the two non - local thermopowers is due to the different role played by @xmath162 in the two cases . as far as @xmath219 is concerned , when we set a temperature difference @xmath168 in lead 2 , a chemical potential difference @xmath224 develops in lead 1 to annihilate the current that flows out of the lead 2 .
when the coupling @xmath162 goes to zero , that current goes to zero and so does the chemical potential difference @xmath224 .
this argument does not hold for @xmath265 , because the temperature difference @xmath167 is set in lead 1 , and @xmath162 does not control the current anymore .
therefore when the coupling @xmath162 approaches zero the current still have a finite value , and so the chemical potential difference @xmath266 needed to annihilate it .
furthermore , the local thermopowers are no more equal , as shown in the right panel of fig .
[ fig : thermopowers_bidot ] .
in this paper we have developed a general formalism for linear - response multi - terminal thermoelectric transport .
in particular , we have worked out analytical expressions for the efficiency at maximum power in the three - terminal case . by means of two simple non - interacting models ( single- and double - dot )
, we have shown that a third terminal can be useful to improve the thermoelectric performance of a system with respect to the two - terminal case .
moreover , we have discussed conditions under which non - local thermopowers could be observed .
our analysis could be extended also to cases in which time - reversal symmetry is broken by a magnetic field or including bosonic or superconducting terminals .
it is an interesting open problem to understand in such instances both thermoelectric performance in realistic systems and fundamental bounds on efficiency for power generation and cooling .
the authors would like to acknowledge h. linke , m. molteni and s. valentini for the useful discussions .
this work has been supported by eu project thermiq and by miur - prin : collective quantum phenomena : from strongly correlated systems to quantum simulators .
to compute the multi - terminal thermopowers defined in eqs .
( [ sloc ] ) and ( [ sloc1 ] ) we have to express one of the temperatures as a function of a thermal current . for example
let us start from the inversion between @xmath110 and @xmath67 . in the onsager
s formalism this can be expressed as : @xmath267 where @xmath268 is a permutation matrix that switches @xmath110 and @xmath67 , @xmath269 and @xmath270 are column vectors with components @xmath106 and @xmath271 , respectively , and @xmath272 is the @xmath273 identity matrix .
then we obtain : @xmath274 where @xmath275 and @xmath276 are the vectors @xmath269 and @xmath270 after the action of @xmath277 , that is , with @xmath278 ; @xmath279 and @xmath280 are the matrices determined by the product @xmath281 .
we can now define the thermopower from the following equations : @xmath282 for this choice of the parameters we have inverted , two different thermopowers can be defined , the non local @xmath219 : @xmath283 and the local @xmath284 : @xmath285 the two - terminal limit in which reservoirs 2 and 3 only are connected is obtained after setting in the onsager matrix @xmath286 if @xmath287 or @xmath14 . in this limit , the previous expressions reduce to : @xmath288 the non - local term goes to zero , while the local one goes to the correct value of the 2-terminal system .
the two other terms of these generalized thermopowers are obtained with the inversion of @xmath111 and @xmath68 .
then we can define @xmath265 as the non local quantity , and @xmath289 as the local one : @xmath290 in a similar way all the other transport coefficients can be defined , by inverting a generalized force with a current .
[ app : cholesky ] in linear algebra , the cholesky decomposition @xcite is a tool which allows to write a hermitian , positive - definite ( or semipositive - definite ) matrix @xmath16 as a product of a lower triangular matrix @xmath291 and its conjugate transpose @xmath292 : @xmath293 ( in particular , if @xmath16 is _ real _ , @xmath292 is simply the transpose of @xmath291 ) .
it turns out that the sign of some quantities defined throughout this work as combinations of products of onsager coefficients @xmath17 can be easily studied by using the cholesky decomposition on the onsager matrix @xmath16 . as an example , by writing @xmath294 it can be shown that the coefficient @xmath295 and @xmath296 defined in eq .
are equal to @xmath297 and therefore are non - negative .
the coefficient @xmath298 instead has undefined sign .
still one can prove that it is such that the determinant of the matrix @xmath116 which appears in eq .
is non - negative .
indeed we have @xmath299 which , together with the positivity of @xmath295 and @xmath296 entails that @xmath116 is semi - positive definite .
the same procedure can be used to study the sign of the constants @xmath143 , @xmath144 and @xmath149 defined in eqs . , , and , respectively . as it is shown here below ,
@xmath143 and @xmath144 are always non - negative , while @xmath149 has undefined sign : @xmath300}{\rho_{22}^2\rho_{33}^2}\ ; , \nonumber \\ c_2 & = & \frac{2[(\delta \rho_{22 } \rho_{33}+\rho_{24}\rho_{33}-\rho_{23}\rho_{34})^2+(\rho_{23}^2+\rho_{33}^2)\rho_{44}^2]}{(\rho_{24}\rho_{33}-\rho_{23}\rho_{34})^2+(\rho_{23}^2+\rho_{33}^2)\rho_{44}^2}\ ; , \nonumber \\ c_{12 } & = & \frac{(\rho_{22}\rho_{33}+\delta \rho_{24}\rho_{33}-\delta \rho_{23}\rho_{34})^2+\delta^2(\rho_{23}^2+\rho_{33}^2)\rho_{44}^2}{\rho_{22}\rho_{33}(\rho_{24}\rho_{33}-\rho_{23}\rho_{34})}. \nonumber \\\end{aligned}\ ] ]
for a three - terminal configuration , as in the previous sections , we choose the right reservoir @xmath160 as the reference ( @xmath301 , @xmath302 ) , and characterize the problem in terms of the particle / heat currents flowing in linear response between the system and leads @xmath2 ( held at @xmath303 and @xmath304 ) and @xmath3 ( held at @xmath305 and @xmath306 ) .
the onsager s coefficients are obtained from the linear expansion of the currents @xmath307 and @xmath71 ( @xmath287 ) given by eqs .
( [ curr1 ] ) and ( [ curr2 ] ) .
they can be written in terms of the transmission probabilities @xmath243 , @xmath308 as : @xmath309 where @xmath166 is the temperature , @xmath310 denotes the fermi - dirac distribution at @xmath165 , and @xmath311 is the partial derivative with respect to the energy .
+ for a scattering region consisting of a quantum dot with a single energy level , connected to three - terminal , we can express the transmission function as @xcite @xmath312 where @xmath313 is the contribution to the broadening due to the coupling to lead @xmath152 , defined by @xmath314 , @xmath315 being the self - energy of lead @xmath152 . in the wide - band limit approximation
, we set @xmath316 , where @xmath317 does not depend on the energy . note
that this choice leads to the identification @xmath318 .
furthermore , at the denominator , @xmath319 is the total broadening due to the coupling to _ all _ leads .
if we denote @xmath320 , @xmath321 , and @xmath322 the couplings to the three leads , we obtain for the transmissions the values @xmath323 the onsager coefficients then read as follows : @xmath324 with @xmath325 . the numerical data shown in sec . [ sec : singledot ] are obtained for @xmath326 , i.e. for @xmath327 .
the two - terminal configuration corresponds to @xmath328 ( @xmath164 ) , while the coupling to terminal 2 is switched on progressively by increasing @xmath296 .
88 note that eq .
( [ eff ] ) can be easily generalized to @xmath116 terminals , after appropriate change of the numerator . by setting for instance @xmath329 , @xmath330 , and @xmath331 ( @xmath332 ) ,
the output power reads as follows : @xmath333 | the efficiency of a thermal engine working in linear response regime in a multi - terminal configuration is discussed . for the generic three - terminal case
, we provide a general definition of local and non - local transport coefficients : electrical and thermal conductances , and thermoelectric powers . within the onsager formalism
, we derive analytical expressions for the efficiency at maximum power , which can be written in terms of generalized figures of merit .
furthermore , using two examples , we investigate numerically how a third terminal could improve the performance of a quantum system , and under which conditions non - local thermoelectric effects can be observed . |
recent surveys of the high - redshift universe have identified a substantial population of massive quiescent galaxies , already in place at @xmath10 ( e.g. , labb et al . 2005 ; daddi et al .
2005 ; van dokkum et al .
2006 ; kriek et al .
studies of their structural parameters have convincingly ruled out simple monolithic collapse models , in which little to no structural evolution is expected . instead
, observational studies find galaxies , and especially quiescent galaxies , to grow significantly in size as time progresses ( e.g. , trujillo et al .
2006 ; toft et al . 2007 ;
zirm et al .
2007 ; van dokkum et al . 2008 ; cimatti et al . 2008
; van der wel et al .
2008 ; franx et al . 2008
; buitrago et al .
2008 ) . at @xmath11 ,
massive quiescent galaxies are typically 5 times more compact , and two orders of magnitude more dense than local ellipticals of the same mass ( e.g. , van dokkum et al .
2008 , hereafter vd08 ) .
bezanson et al . (
2009 ) note that , even though their effective densities ( measured within one effective radius @xmath12 ) are strikingly high compared to local ellipticals , the central densities measured within a fixed aperture of 1 kpc exceed those of local ellipticals by no more than a factor 2 - 3 ( see also hopkins et al .
this observation suggests an inside - out growth , in agreement with stacking results by van dokkum et al .
( 2010 ) . motivated by these recent observational developments ,
several mechanisms have been proposed to incorporate the constraints on structural evolution into galaxy formation theories . in most cases ,
minor and/or major mergers are invoked to explain the observed size evolution ( khochfar & silk 2006a ; naab et al .
2007 , 2009 ; hopkins et al .
briefly , mergers were more gas - rich at high redshifts , and hence formed a larger fraction of their stars in a nuclear dissipational component , explaining their compact nature .
subsequent ( dry ) merging activity puffs up the system without adding too much mass or new stars , which would violate observational constraints .
however , alternative scenarios involving an expansion of the stellar distribution as response to significant mass losses have been suggested as well ( fan et al .
2008 ) . in either case
, an accurate observational characterization of the size - mass relation provides a crucial test for galaxy formation models . given the paucity of kinematic mass measurements based on absorption - line spectra of @xmath13 galaxies ( although see cenarro & trujillo 2009 ; cappellari et al .
2009 ; van dokkum , kriek & franx 2009 ) , studies of the high - redshift size - mass scaling relation to date have focussed on stellar mass estimates from spectral energy distribution ( sed ) modeling ( e.g. , williams et al .
significant systematic uncertainties related to the assumption of an imf ( e.g.,van dokkum 2008 ; dav 2008 ; wilkins et al .
2008 ) and the choice of a stellar population synthesis code ( e.g. , maraston et al . 2006 ;
wuyts et al .
2007 ; muzzin et al .
2009a ) remain .
uncertainties related to the star formation history , metallicity , and dust attenuation are relatively modest for the quiescent population , they only contribute significantly to the overall uncertainty during earlier , actively star - forming phases ( wuyts et al . 2009a ) .
however important an accurate characterization of mass , we focus in this paper on the measurement of the second parameter of the scaling relation : galaxy size .
observations probe the projected distribution of light , sampling it by a discrete number of pixels after it was smeared by a point spread function ( psf ) .
in addition , the signal is superposed by noise .
the translation to a physically more meaningful mass profile involves the assumption of a mass - to - light ratio @xmath5 .
although often for simplicity assumed to be a constant , spatial variations in @xmath5 may occur due to age , metallicity and/or dust gradients .
furthermore , since the total size of a galaxy is ill - defined , one refers to ( circularized ) size as the radius @xmath12 containing half the mass .
given the finite image resolution , this quantity is generally obtained by fitting a template profile , taking pixelization and psf smearing into account . in most of the literature ,
a one - component sersic ( 1968 ) profile has been adopted , providing satisfyingly flat residual images given the noise level of the observations .
numerical simulations provide an excellent tool for the interpretation of galaxy structure .
the simulated data offers a three - dimensional view of the mass , age , and metallicity profile at high resolution , free of sky noise . by feeding the output to a radiative transfer code and producing mock observations ,
each of the above aspects related to the nature of observational data can be isolated , and its effect analyzed .
for example , contrasting the light profiles of local gas - rich merger remnants and ellipticals with those of simulated merger remnants , hopkins et al .
( 2008b , 2009b ) demonstrated that a two - component profile ( consisting of an inner dissipational , and outer violently relaxed component ) provides both a better fit and a physically more meaningful interpretation of their structure than a single sersic profile . in this paper , we compare the structure of simulated merger remnants to the best observations of compact quiescent galaxies at @xmath0 to date .
in addition , we discuss the presence of m / l ratio gradients that may bias measurements of the half - mass radius , and can be revealed by multi - wavelength structural studies with the high - resolution cameras onboard _
hubble space telescope_. this study complements the comparison between observations and merger simulations by wuyts et al .
( 2009b ) that focussed on the integrated colors , number and mass densities of high - redshift quiescent galaxies .
we describe the simulations , and mock observations based thereupon in [ simulations.sec ] .
there , we also caution for artificial heating in simulating regimes of extreme density .
we discuss the relation between size and mass for quiescent galaxies with a range of formation histories in
[ masssize.sec ] , and address their kinematics in
[ sigma.sec ] .
next , we analyze radial variations in the @xmath5 ratio , their origin in terms of stellar populations , and color gradients through which they manifest themselves ( [ mlgradient.sec ] ) . finally , we exploit realistic mock observations of the simulations to fit their surface brightness profiles alongside real high - redshift compact galaxies . [ profiles.sec ] highlights the cuspiness of the simulated profiles , and explores possible origins of an apparent profile mismatch with respect to the observations .
we summarize the results in [ summary.sec ] . throughout this paper , we adopt the following cosmological parameters : @xmath14 .
this work is based on a suite of smoothed particle hydrodynamic ( sph ) simulations representing isolated and merging galaxies at redshifts @xmath15 to 3 that were performed with the gadget-2 code ( springel 2005 ) .
the code uses an entropy - conserving formalism of sph ( springel & hernquist 2002 ) , and includes gas cooling , a multi - phase model for the interstellar medium ( ism ) to describe star formation and supernova feedback ( springel & hernquist 2003 ) , and a prescription for supermassive black hole growth and feedback ( springel , di matteo & hernquist 2005 ; di matteo , springel & hernquist 2005 ) .
the progenitor disks in our simulations are embedded in an extended dark matter halo with a hernquist ( 1990 ) profile .
they span a range of baryonic masses from @xmath16 to @xmath17 , and initial gas fractions of 20 - 80% .
we mostly focus on the subset of massive gas - rich merger simulations in which the progenitors were scaled to approximate the structure of disk galaxies at redshift @xmath18 , following robertson et al .
briefly , this means that the mass- and redshift - dependent halo concentration measured by bullock et al .
( 2001 ) was adopted : @xmath19 where @xmath20 is the linear collapse mass at z=0 .
the virial mass and virial radius of the progenitors were scaled as follows : @xmath21 where @xmath22 is the virial velocity and @xmath23 is the hubble parameter .
disk sizes were initialized according to the mo et al .
( 1998 ) formalism for dissipational disk galaxy formation assuming the fraction of the total angular momentum contained in the disk equals the fraction of the total mass contained in the disk .
the disk scale length is then derived from the halo concentration @xmath24 ( eq . [ cvir.eq ] ) and the galaxy spin @xmath25 , where we adopt a default @xmath26 , as motivated by cosmological n - body simulations ( vitvitska et al .
2002 ) . in practice
, this means that the @xmath18-scaled progenitor disks have effective radii that are a factor 1.7 smaller than similar mass galaxies today , in agreement with the observed size evolution of star - forming galaxies ( franx et al .
2008 ) .
equal - mass mergers were simulated for a range of orbital configurations , from coplanar to polar to various tilted disk orientations , following cox et al .
( 2006 ) . specifically , we analyzed simulations with the range of spin axis orientations of the progenitor disks listed , in standard spherical coordinates , in table[spinorientations.tab ] .
in typical runs , each of the two progenitor galaxies initially consists of 60000 to 120000 dark matter particles , 40000 gas and 40000 stellar disk particles , and one black hole sink particle . for the mass range probed by our simulations
, this corresponds to typical mass resolutions of the baryonic and dark matter particles of @xmath27 and @xmath28 respectively.in addition to equal - mass mergers , we also ran simulations in which the disk galaxy was left to evolve in isolation , simulations of unequal - mass mergers , and scenarios where a merger remnant undergoes subsequent merging .
lccccl[b ] h & 0 & 0 & 0 & 0 & both prograde + b & 180 & 0 & 0 & 0 & prograde - retrograde + d & 90 & 0 & 0 & 0 & polar + e & 30 & 60 & -30 & 45 & tilted 1 + g & 150 & 0 & -30 & 45 & tilted 2 + i & 0 & 0 & 71 & 30 & barnes orientations + j & -109 & 90 & 71 & 90 & @xmath29 + k & -109 & -30 & 71 & -30 & + l & -109 & 30 & 180 & 0 & + m & 0 & 0 & 71 & 90 & + n & -109 & -30 & 71 & 30 & + o & -109 & 30 & 71 & -30 & + p & -109 & 90 & 180 & 0 & + we adopted a gravitational softening length of @xmath30 pc . a resolution study for one of our fiducial merger simulations , where we increased the mass resolution by a factor 4 and repeated the simulations with two and three times the spatial resolution ( i.e. , adopting softening lengths that are half or a third of our default ) and otherwise identical conditions confirms that the adopted resolution is sufficient for the purposes of this paper ( figure[resol.fig ] ) .
the total stellar mass of the merger remnant varies by a few percent only when increasing the spatial resolution . at radii larger than the softening length
, the surface mass density profiles are converged at the 10% level . within a radius of 140 pc
, the amount of stellar mass assembled in the nucleus tends to be somewhat larger in the higher resolution runs .
we furthermore refined the gadget-2 timestep resolution by a factor @xmath31 with respect to the gadget-2 default , resulting in runtimes that are longer by the same factor .
figure [ timestep.fig ] illustrates the importance of the latter choice when studying ( sub)structure at high spatial densities .
the panels illustrate four realizations of a gas - rich merger simulation with a final stellar mass of @xmath32 and half - light radius of @xmath2 kpc , identical except for the timestep resolution with which particle orbits are integrated
. the timestep scales with @xmath33 , where the respective value of the gadget-2 parameter @xmath34 = errtolintaccuracy is shown in the panel . for each realization
, we show the 3d ( top line ) and 2d ( bottom polygon ) half - mass radius evolution as function of time since the merger , which we define as the radius of a sphere ( respectively circle ) encompassing half of the ( projected ) stellar mass . throughout this paper , we define time of merging as the moment when the peak in star formation activity is reached .
different colors represent the size evolution of the galaxy as a whole ( black ) , the young component of stars that formed within a 250 myr interval around final coalescence ( blue ) , and the old component of stars that were already formed prior to the nuclear starburst ( red ) .
several conclusions can be drawn from figure [ timestep.fig ] .
first , numerical heating can lead to artificial growth of the simulated merger remnant . since this effect is not uniform over the galaxy , but manifests itself particularly in the dense and young central component , it can also artificially alter the profile shape . furthermore , the stellar mass formed during the nuclear starburst steadily decreases when a finer timestep resolution is adopted , by 26% when comparing simulation ( a ) to ( d ) from figure[timestep.fig ] , and by 10% when comparing simulation ( b ) to ( d ) .
this trend suggests that , in addition to artificial growth during the post - merger phase , the implementation of star formation and/or feedback processes may also depend on the length of the integration time step .
this result has important implications for studies of structural evolution of massive galaxies based on cosmological simulations , where finite computational power imposes a delicate tradeoff between the box size ( essential for reliable number statistics at the massive end ) , the spatial resolution , and the timestep resolution .
clearly , by compromising the timestep resolution to a level where no numerical convergence is reached , spurious growth in size and smoothing of cusps will be superposed on galaxy growth by real physical processes .
a second conclusion to draw from figure [ timestep.fig ] is that , once a converging timestep resolution is adopted , we find no evidence for size evolution after final coalescence in our binary merger simulations .
third , the young dissipational component in this simulation that started with a gas fraction of 80% and was still very gas - rich ( @xmath35 ) at the time of final coalescence is more than an order of magnitude smaller than the old component of previously formed stars .
finally , by definition the 2d half - mass radius ( which depends on viewing angle ) is smaller than the 3d half - mass radius .
we find a typical @xmath36 ratio of 1.4 .
furthermore , we find that the ratio @xmath36 depends more strongly on viewing angle for the young component than for the old component , implying that the recent , dissipational star formation event took place in a disk - like structure , whereas the violently relaxed component has a more spherical shape .
we find similar properties for the merger remnants produced by simulations that started with different orbital configurations .
the fluxes , colors and light profiles of the merger remnants are computed from the simulation output in two steps .
first , the intrinsic stellar emission is derived from a stellar population synthesis code , where we treat each particle as a simple stellar population ( ssp ) with its stellar mass , age and metallicity computed by the gadget-2 code .
we adopt a kroupa ( 2001 ) imf and compute the photometry using bruzual & charlot ( 2003 ) or maraston ( 2005 ) models .
we find that our results are independent of the choice of stellar population synthesis code .
initial conditions ( stellar age and metallicity ) of the stellar and gas particles present at the start of the simulation were set by a simple closed box formalism detailed by wuyts et al .
( 2009a ) .
the precise choice of these initial stellar population properties has a negligible impact on the nature of the merger remnants that are the focus of this paper . in step two
, we use the information on the gas distribution and enrichment to compute the attenuating effect of dust on the emerging galaxy light .
we perform our analysis using two independent codes that are frequently used in the literature .
the first is a line - of - sight attenatuation code ( los ) that simply computes the wavelength - dependent absorption to each stellar particle from the metallicity - weighted diffuse gas column density between that stellar particle and the observer ( see hopkins et al .
2005 ; robertson et al . 2007 ;
wuyts et al .
2009a , b ) .
the second code , sunrise ( jonsson 2006 ; jonsson , groves & cox 2010 ) , uses a monte carlo methodology to track photon packets on their way through the dusty ism , and models the effects of both absorption and scattering .
in addition , sunrise uses a sub - grid model to account for the attenuation by birthclouds ( hii and photodissociation regions ) that surround young star clusters , computed using the photoionization code mappings ( groves et al .
2008 ) . furthermore , sunrise tracks dust temperatures and re - emission of absorbed light at longer wavelengths ( see , e.g. , younger et al .
2009 ; narayanan et al .
however , since no high - resolution structural information is available in the far - infrared , and dust re - emission is negligible during the quiescent merger remnant phase , this aspect of the radiative transfer is not discussed in this paper .
likewise , the gadget-2 simulation keeps track of gas accretion onto a central supermassive black hole , but its emission is negligible during the quiescent merger remnant phase of interest .
we compute realizations of the radiative transfer with attenuation laws representative of the milky way and the small magellanic cloud .
unless specifically mentioned , our conclusions are qualitatively the same independent of the radiative transfer code or attenuation law used .
we use the vd08 sample of compact quiescent galaxies at @xmath11 as reference .
all 9 galaxies have a spectroscopically confirmed balmer/4000 break ( kriek et al .
seven of them were observed for 3 orbits in the f160w filter using the nic2 camera on _ hst _ , probing their rest - frame optical emission .
the two brighest galaxies were exposed for 2 orbits . at a drizzled pixel scale of 0.0378 , the point - spread function ( psf ) is well sampled , and the observations yielded the highest - resolution surface brightness maps of such galaxies to date . in order to establish a fair comparison , we produce mock observations placing the simulated merger remnants at the same redshift , using the same observed filter and an identical pixel scale .
we convolved the resulting postage stamps with a psf extracted from the vd08 nic2 images , applied poisson noise , and added them into empty regions of the nic2 images in order to guarantee similar noise properties .
we randomized the sub - pixel position of the simulated galaxies , but find that the recovered structural properties are independent of this treatment . in the following ,
we first analyze the _ true _ half - mass ( [ masssize.sec ] ) and half - light ( [ mlgradient.sec ] ) radii of the simulated galaxies as measured from the full ( noise - free and high - resolution ) information available from simulation and radiative transfer .
next , we include the effects of pixelization , finite resolution , and limited signal - to - noise by running the two - dimensional fitting code galfit ( peng et al .
2002 ) on the real and mock nic2 images ( [ profiles.sec ] ) .
wuyts et al . ( 2009b ) demonstrated that the integrated stellar population properties ( rest - frame optical and optical - to - nir colors , as well as specific star formation rates ) of simulated merger remnants are consistent with those of observed quiescent galaxies at @xmath37 . here
, we investigate whether a major merger formation scenario can also account for the remarkable compactness of the observed quiescent systems . a useful diagnostic to address
this question is the size - mass relation . in figure
[ sizemass.fig ] , we mark the local spheroid size - mass relation ( shen et al .
2003 ) with a solid line . as reliable size measurements of complete samples of high - redshift quiescent galaxies
are only available over a narrow mass range , we indicate a fiducial size - mass relation at redshift @xmath38 ( dashed line ) by assuming the @xmath39 slope and adopting the median size and mass of the vd08 sample as zero - point anchor .
we note that both observational ( trujillo et al .
2006 ; ryan et al .
2010 ) and theoretical ( khochfar & silk 2006a ) studies have suggested that the size - mass relation becomes shallower with redshift .
ryan et al .
( 2010 ) report a mass dependence of the power - law index @xmath40 in the size evolution @xmath41 of @xmath42 .
we caution that this fit is driven by measurements of galaxies at @xmath43 , and , as stated by ryan et al .
( 2010 ) , should be considered preliminary at best .
nevertheless , it is illustrative to plot the corresponding size - mass relation at @xmath11 ( dotted line ) .
we conclude that the ultra - compactness of massive ( @xmath44 ) quiescent galaxies at @xmath0 is a robust result .
tighter constraints on the sizes of high - redshift spheroids of lower mass will be essential to understand whether they formed through similar mechanisms .
the main focus of this paper is on compact galaxies at the high - mass end .
overplotted in figure [ sizemass.fig ] , we show the 2d ( i.e. , projected ) half - mass radii of simulated merger remnants as a function of their stellar mass , as seen from 100 viewing angles uniformly spread over a sphere .
we consider merger simulations with a range of stellar masses , and in each case only plot the snapshots more than 100 myr after the final starburst .
encircled are the remnants produced by merging progenitors that were scaled to represent @xmath45 disks ( i.e. , reaching the remnant phase around @xmath11 ) .
the other simulations started out with progenitors scaled to represent local disk galaxies .
for every simulation , we define a dissipational fraction @xmath46 as the gas fraction of the system 125 myr before the peak in star formation rate ( sfr ) is reached .
even though some of our simulations start with initial gas fractions as high as 80% , @xmath46 rarely reaches values above 50% , because rapid star formation in the progenitor disks , especially during first passage , consumes significant amounts of gas before final coalescence .
figure [ sizemass.fig ] clearly illustrates that , at a given mass , the size of a merger remnant is smaller when the dissipational fraction is higher , and even more so when the progenitor disks had a more compact nature to start with .
tidal torques are responsible for channeling large amounts of gas to the central region ( barnes & hernquist 1991 , 1996 ) where it is consumed in a starburst ( mihos & hernquist 1994 , 1996 ) . we conclude that the location of the observed @xmath0 quiescent galaxies on the size - mass diagram ( the vd08 sample has median properties @xmath47 kpc , @xmath48 ) can straightforwardly be explained by major merger activity provided the progenitors at high redshift were more gas - rich and had scalelengths smaller than today s disk galaxies .
this idea was first formulated by khochfar & silk ( 2006a ) , and is further detailed by hopkins et al .
( 2009a ; 2010a ) . significant obervational support for increased gas fractions in star - forming galaxies towards higher redshift was inferred from h@xmath40 spectroscopy by erb et al .
( 2006 ) , and more recently confirmed on the basis of molecular line measurements ( baker et al . 2004 ; coppin et al . 2007 ;
daddi et al .
2008 ; tacconi et al .
likewise , a decrease in the size of star - forming galaxies at a given mass is observationally well established ( trujillo et al .
2006 ; franx et al .
2008 ; williams et al .
a proper characterization of galaxy structure requires knowledge of the mass - to - light ratio to determine the stellar mass , and of its gradient to translate the measured half - light radius to a radius containing half the mass ( see [ mlgradient.sec ] ) .
one way to circumvent @xmath5 ratio effects , is probing the central potential directly through kinematics .
however , measuring the velocity dispersions of quiescent @xmath0 galaxies from their absorption line spectra is extremely expensive in terms of telescope time . to date , one such measurement ( van dokkum , kriek & franx 2009 ) at @xmath49 has been carried out , totaling 29 hours on a 8 m class telescope . at somewhat lower redshift and mass ( @xmath50 ) , a stacked measurement of similar red nuggets at @xmath51 , as well as two individual measurements at @xmath52
were presented by cappellari et al .
( 2009 ) . in figure [ sigma.fig ]
, we compare the measured velocity dispersion to the virial estimator @xmath53 , where @xmath54 is the stellar mass derived from sed modeling .
the dashed line indicates the proportionality followed by present - day early - type galaxies ( van dokkum & stanford 2003 ; cappellari et al .
color - coded by their gas fraction shortly before the final starburst , we overplot the same simulated merger remnants whose progenitors were scaled to represent high - redshift star - forming galaxies as displayed in figure [ sizemass.fig ] . both the observed and simulated quiescent galaxies show a clear correlation between the measured velocity dispersion and what would be estimated based on virial arguments . within the error bars , the observational results are all consistent with having the same scaling between @xmath53 and @xmath55 as early - type galaxies in the nearby universe .
simulations of gas - rich mergers are able to produce remnants with similar velocity dispersions . at a given velocity dispersion ,
the virial estimator @xmath53 is larger for runs with a higher dissipational fraction , implying that mergers of varying dissipational fraction are non - homologous ( see also robertson et al .
2006 ; hopkins , cox & hernquist 2008 ) .
briefly , the scaling factor @xmath56 in the relation @xmath57 where @xmath54 is the stellar mass and @xmath12 the stellar half - mass radius , depends on the profile shape of the stellar mass distribution ( lower @xmath56 for cuspier systems ) , and on the baryon - to - dark matter ratio within the stellar effective radius ( lower @xmath56 for more baryon - dominated centers ) .
our simulations show that both factors are a strong function of the dissipational fraction @xmath46 of the merger .
systems with larger @xmath46 are increasingly baryon - dominated in their centers .
their cuspy profile shapes are discussed at length in section [ cuspiness.sec ] .
although such measurements are not yet available observationally at @xmath58 , it is interesting to consider the degree of rotation expected in high - redshift compact quiescent galaxies . to this end , we study the ratio of rotational over random motion ( @xmath59 , where the rotation velocity is measured along the major axis ) as a function of ellipticity in figure[vsig_eps.fig ] .
the gray - coded distribution indicates the locus occupied by quiescent merger remnants that match the vd08 galaxies in size and mass .
these remnants are the product of gas - rich merger simulations with a range of orbital configurations , and we observe them from 100 sightlines uniformly distributed over a sphere .
the green line marks the relation expected for an oblate isotropic rotator ( binney 1978 ) : @xmath60 vertical dashed lines indicate the ellipticities of the @xmath11 compact galaxies of vd08 , as measured on the high - resolution nic2 images . for reference
, we also plot the location of massive early - type galaxies ( of larger size than those at @xmath0 ) at @xmath61 ( van der wel & van der marel 2008 ) and at @xmath39 ( a complete sample extracted from the hyperleda database by paturel et al .
2003 , as detailed by van der wel & van der marel 2008 ) .
the closed and open circles represent galaxies with an e and s0 morphology respectively .
the rotation parameter of the simulated merger remnants increases with ellipticity . at a given ellipticity ,
the compact remnants show a larger degree of rotational support than the bulk of early - type galaxies today .
their kinematics resemble those of the fastest rotators of the @xmath39 sample at each ellipticity .
cox et al . ( 2006 ) demonstrated that the rotation parameters and flattening of local spheroids with less rotation can be reproduced by merger simulations tuned to lower redshift , with lower gas fractions ( see also naab , jesseit & burkert 2006 ) .
a hint of increasing rotational support towards higher redshifts , where mergers are expected to be increasingly dissipational , may already be observed when comparing the measurements at @xmath62 to those at @xmath39 .
it remains to be seen whether this trend extends to @xmath0 , and whether similarly high values of @xmath63 , of up to unity , are present as anticipated by our simulations .
the large range of ellipticities in the vd08 sample , reaching values of @xmath64 which are larger than found in our simulations , may point to rotation being indeed important in these systems .
size measurements have traditionally been carried out using one waveband only , typically the longest wavelength available at high resolution . by definition , this technique probes a monochromatic surface brightness profile . evidently , the half - light radius derived from it equals the physically more meaningful half - mass radius only under the assumption of a spatially constant @xmath5 ratio .
having established that gas - rich mergers are a viable mechanism to collect @xmath65 of baryonic material in a @xmath2 kpc radius , we now use our simulations to test this assumption for the quiescent remnants . using the radiative transfer methods described in [ sim_phot.sec ] , we compute the half - light radius in three rest - frame wavebands @xmath66 , @xmath6 , and @xmath67 , and contrast it with the half - mass radius of the system for 100 lines of sight .
figures[rlight_rmass_los.fig ] and [ rlight_rmass_sunrise.fig ] show the resulting light - to - mass size ratio @xmath68 as a function of time since the merger for the same simulation shown in figure [ timestep.fig](c ) .
the central line indicates the median evolution over all viewing angles , whereas the light and dark shaded regions mark the central 50% and 100% percentiles respectively .
panels on the left illustrate the time evolution of @xmath69 ignoring attenuation by dust . in this case
, all scatter at a given time is due to projection effects .
the scatter is limited to a few percent in rest - frame @xmath6 , which is the band most closely corresponding to f160w at @xmath11 .
initially , as many new stars are formed in the nuclear starburst , the intrinsic light profile is more centrally concentrated than the stellar mass profile ( i.e. , @xmath70 ) . on timescales of a few hundred myr ,
as the bright and short - lived o and b stars die out , the nuclear emission dims with respect to the outer parts of the galaxy , and the @xmath5 gradient flattens .
in fact , around 400 myr after the peak in star formation rate , the light profile in the rest - frame optical becomes more extended than the mass profile .
this trend is more pronounced as we consider shorter wavelengths , reaching @xmath71 in rest - frame @xmath66 .
the reason for this reversed @xmath5 gradient is that the young stars in the center formed out of more metal - enriched gas than the stars that constitute most of the galaxy outskirts , making them fainter and cooler ( binney & merrifield 1998 , and references therein ) .
we illustrate the physical conditions underlying the @xmath5 ratio gradients in figure [ physgradient.fig ] for the snapshot 480 myr after the merger . at this time
, the remnant has a specific star formation rate ( @xmath72 ) and broad - band colors similar to those of observed quiescent galaxies . plotting stellar age as function of distance from the stellar particles to the center ( figure [ physgradient.fig]a ) , three loci in age can be distinguished , corresponding to stars formed in the progenitor disks , during first passage , and during final coalescence . the radial position in the merger remnant
is clearly correlated with the epoch at which the stars were formed .
figure [ physgradient.fig]a also illustrates that the low level trickle of star formation present after the nuclear starburst happens mostly outside the stellar half - mass radius ( @xmath73 kpc ) .
this is a secondary effect contributing to the larger extent of the light profile compared to the mass profile .
figure [ physgradient.fig]b demonstrates the presence of a negative stellar metallicity gradient , of slope @xmath74 .
this slope is a factor 1.5 to 2.5 steeper than that of typical massive early - type galaxies today ( rawle , smith & lucey 2010 ; kuntschner et al .
2010 ) , although those exhibit a large scatter in gradients . together with their compact nature ( [ masssize.sec ] ) and fast rotation ( [ sigma.sec ] ) , this implies the gas - rich merger remnants at @xmath58 can not evolve passively into present - day massive ellipticals .
subsequent ( dry ) merging has been proposed as a mechanism to slow the rotation , grow the size , and dilute the metallicity gradient , with the extent of the dilution dependent on the properties of the merger ( di matteo et al .
2009 ) .
in addition to stellar age and metallicity , the gas and dust distribution and extinction column towards stellar particles also vary radially .
the remaining gas in the @xmath75 kpc range has a negative metallicity gradient ( figure [ physgradient.fig]c ) , while at even larger radii the sph gas particles show a wide spread in metallicities .
the normalized cumulative distribution of dust ( taken to be proportional to the metallicity - weighted gas distribution ) is contrasted with that of stars and gas in figure[physgradient.fig]d .
the corresponding spherically averaged density profiles , normalized to the density at 1 kpc , are presented in figure [ physgradient.fig]e .
the gas is distributed over scales that are an order of magnitude larger than that of the stellar distribution . both gas and dust ( @xmath76 metallicity - weighted gas )
reach a peak density near the stellar half - mass radius ( @xmath77 kpc ) .
the tail of young stars at @xmath2 kpc visible in figure[physgradient.fig]a shows that it is also here that the small amount of star formation that is still present after the merger ( @xmath78yr@xmath79 ) takes place .
gas that during earlier phases resided within this radius has largely been consumed by star formation , or has been blown out by feedback processes from supernovae and agn .
although this is not visible in the spherically averaged density profile of figure[physgradient.fig]d , the central cavity devoid of gas and dust is not spherically symmetric , but rather axisymmetric .
the resulting radial distribution of optical depths towards the stellar particles therefore depends on the line of sight .
we illustrate this distribution for two characteristic viewing angles in panel ( f ) and ( g ) . for the viewing angle presented in panel ( f ) , the effective attenuation of the region within the half - mass radius is larger than that outside .
consequently , the attenuated @xmath6-band half - light radius viewed from this angle is 4.8 times larger than the half - mass radius . for the viewing angle of panel ( g ) ,
the opposite is the case and the resulting attenuated @xmath6-band half - light radius is smaller than the half - mass radius by a factor 1.4 .
the right - hand panels of figure [ rlight_rmass_los.fig ] illustrate how the superposition of a dust , age , and metallicity gradient translates into the light - to - mass size ratio @xmath80 , and how this quantity depends on time relative to the merger , line of sight , and wavelength . during the nuclear starburst , the @xmath5 ratio gradient is completely reversed with respect to the intrinsic ( unattenuated ) @xmath5 ratio gradient , because of central dust obscuration .
the median light - to - mass size ratio then drops , reaching @xmath81 values below 1 around 300 myr after the peak in star formation rate , as the central dust content decreases while the young , massive stars in the center are still alive .
finally , as the system reaches a quiescent remnant phase ( @xmath82 myr ) , the age gradient has faded and the combination of a negative extinction and metallicity gradient results in typical half - light radii of @xmath83 , @xmath84 , and @xmath85 times the half - mass radius at @xmath86 myr in rest - frame @xmath66 , @xmath6 , and @xmath67 respectively . at @xmath87 gyr , the half - light radii in the rest - frame @xmath66- , @xmath6- , and @xmath67-band are @xmath88 , @xmath89 , and @xmath90 times the half - mass radius . here ,
the error bar indicate the central 50% percentile of the scatter due to line - of - sight variations .
typically , the distribution shows an extended tail towards large @xmath80 ratios . both the line - of - sight scatter and the median value of @xmath81 increase towards shorter wavelengths .
computing the synthetic photometry with the independent radiative transfer code sunrise ( figure [ rlight_rmass_sunrise.fig ] ) gives results that are qualitatively consistent with those obtained from the los code .
sunrise predicts typical light - to - mass size ratios of 3.6 , 2.4 , and 1.3 in rest - frame @xmath66 , @xmath6 , and @xmath67 respectively during the quiescent phase .
the sightline dependence shows a similar behavior as seen in the results from the los code , i.e. , with a tail towards large @xmath81 ratios ( ratios of 5 and above ) .
we tested the dependence of our results on the adopted attenuation law , on the assumed age and metallicity of stars present at the start of the simulation , and whether an age and/or metallicity gradient was already in place in the initial progenitor disks , and find consistent results with variations much smaller than those from sightline to sightline . likewise
, similar results were obtained from simulations where the progenitor disks were merged with a different orbital configuration .
peirani et al .
( 2010 ) simulate minor mergers between an elliptical galaxy and a satellite ( spiral galaxy ) , and find qualitatively similar stellar age gradients : the young stars that formed by dissipational processes reside in the center of the remnant . the evolution of half - light radii presented by peirani et al .
( 2010 ) resembles that seen in intrinsic light in our major merger simulations .
however , the dust columns , and therefore the impact of attenuation on the observerd @xmath5 gradients , are larger for the major than for the minor merger simulations . the assumption of a fixed dust - to - gas rather than dust - to - metal ratio , and the absence of agn feedback in the peirani et al .
( 2010 ) simulations may also contribute to differences in the predicted @xmath5 gradients . in summary , we find that the interplay between an age , metallicity , and dust gradient manifests itself as a @xmath5 ratio gradient that depends on time , line of sight and wavelength .
since an observer can only probe the latter parameter directly , the detection of color gradients showing preferentially red cores in compact quiescent galaxies would support the presented scenario .
moreover , from the simulations we expect the internal color gradients to be correlated with the integrated rest - frame optical color ( see figure[grad_integrat.fig ] ) .
we stress that the presence of a red core does not necessarily imply inside - out growth .
in fact , in this scenario , and in the absence of subsequent merging ( see [ subsequent.sec ] ) , the bulk of the youngest stars resides in the center , but the effect of the age gradient on the internal color profile is compensated by the presence of metallicity and extinction gradients . finally ,
if negative @xmath5 ratio gradients are indeed present , our results may imply that the observed quiescent galaxies may be an order of magnitude ( @xmath91 ) more dense than previously inferred from hst / nicmos observations ( vd08 ) .
we established that gas - rich mergers are a viable mechanism to assemble dense stellar systems of @xmath65 and @xmath92 kpc size ( [ masssize.sec ] ) , and that such a scenario would leave a color and @xmath5 ratio gradient as imprint ( [ mlgradient.sec ] ) .
now , we go beyond the zeroth - order structural measurement of size and compare how the surface brightness profiles of simulated merger remnants compare to those of observed quiescent galaxies at @xmath0 . for this purpose
, we use the mock f160w observations described in
[ sim_mock.sec ] that are matched in terms of psf , pixel size , wavelength , redshift , and noise properties to the nic2 observations of high - redshift spheroids by vd08 .
we analyze the real and mock observations in concert with the two - dimensional fitting code galfit ( peng et al .
2002 ) using identical settings . in figure[vd08_stamps.fig ]
, we show postage stamps of the real nic2 images by vd08 .
next to each image , we plot the residual images after subtracting the best - fit sersic profile when leaving the sersic index @xmath93 free , or fixing it to @xmath94 , @xmath95 , @xmath96 , and @xmath97 respectively .
we rank the objects by best - fit sersic index and find they span a large range from @xmath98 to @xmath99 , reproducing the vd08 results . in most cases , fixing the sersic index to @xmath97 leads to significantly larger residuals than leaving @xmath93 free . repeating the analysis for random viewing angles of a simulated merger remnant ( figure [ sim_stamps.fig ] ) ,
we obtain strikingly different results . for the majority of sightlines
, we find that exponential ( @xmath100 ) and even de vaucouleurs ( @xmath96 ) profiles provide a poor fit to the mock data .
their @xmath101 residual images ( illustrated in the top 6 rows of figure [ sim_stamps.fig ] ) are characterized by a central positive peak , surrounded by a negative ring , outside of which a positive wing is barely visible above the noise .
our results remain unchanged when running galfit with the sky level left as a free parameter .
the origin of this characteristic pattern becomes clear when considering a one - dimensional representation of the surface brightness , free of the psf , noise , and pixelization effects of the mock observations ( figure [ n_re.fig]a ) . here
, the black solid line indicates the median surface brightness profile of 100 sightlines uniformly distributed over a sphere .
the dashed and dotted gray lines decompose the light into that of stars formed prior and during final coalescence respectively . as discussed extensively by hopkins et al .
( 2008b ; 2009b ) , the young component produces a central cusp , and the combination of the young and old component yields a profile that is poorly characterized by a simple sersic index .
the colored curves in figure [ n_re.fig](a ) illustrate sersic profiles with a total luminosity and half - light radius identical to that of the simulated merger remnant . the radially alternating positive - negative - positive residual pattern is immediately apparent , as is the tendency to fit high values of @xmath93 .
it is important to note that this tendency is driven both by the presence of a central cusp ( the amplitude of which depends on the amount of gas present at final coalescence ) and by the presence of the strong wings to the profile ( the amplitude of which is determined by the amount of gas consumed during the progenitor phase ) . for a smaller number of sightlines ( illustrated by the bottom 3 panels of figure [ sim_stamps.fig ] ) a flat residual image
is obtained when fitting a sersic index in the range @xmath102 .
obscuration of the central young component by dust is responsible for this sightline - dependent effect .
we demonstrate our findings with more statistical robustness ( more lines of sight and timesteps , and simulations with different orbital configurations ) in figure [ n_re.fig](b ) . here , we plot the profile shape ( characterized by the best - fit sersic index ) as function of the half - light radius recovered by galfit .
large black circles are the vd08 compact galaxies .
they have a median @xmath103 , well below the @xmath96 de vaucouleurs profile that is characteristic for nearby spheroids .
van dokkum et al . (
2010 ) attribute this evolution to minor merging building up the wings of the high - redshift compact nuggets over cosmic time .
the sersic profiles fitted to the simulated merger remnants ( red dots ) are significantly cuspier ( higher @xmath93 , adopting an upper bound of @xmath104 ) . for the subset ( about one fifth ) of sightlines where the central cusp is sufficiently obscured by dust , the inferred half - light radius is several kpc , inconsistent with the observed compact systems .
we draw a similar conclusion from a direct comparison of the surface brightness profiles of our simulated merger remnants to that of a massive compact quiescent galaxy at @xmath105 in the hubble ultra deep field ( szomoru et al .
2010 , see figure [ szomoru.fig ] ) .
szomoru et al .
( 2010 ) derive the deconvolved profile of this galaxy from very deep wide field camera 3 ( wfc3 ) imaging using a novel technique that corrects the best - fit sersic profile with the residual of the fit to the observed image .
they find that the surface brightness profile is well approximated by an @xmath106 sersic profile .
typically , our simulated merger remnants have a higher surface brightness in the @xmath107 kpc wings relative to the surface brightness in the @xmath108 kpc range . the ability to correctly recover the true half - light radius known from the simulation depends on the upper bound set on the sersic index . for @xmath109 ,
we find a systematic overestimate of the size by @xmath3% , whereas setting @xmath110 improves the size measurement to about @xmath111% in the median , with a scatter in @xmath112 of @xmath113 .
so , whereas the mock images for the majority of sightlines are best fit by high values of @xmath93 , these best fits do not properly recover the true half - light radius of the system .
this emphasizes the fact that an individual sersic profile poorly describes its structure .
we assessed the impact of observational depth by artificially decreasing the signal - to - noise ratio of our mock observations ( in practice , we reduce the source brightness , and add them to empty regions of the same nic2 images ) . running galfit with @xmath114 ,
this yields an increase in both the systematic size overestimate and the scatter in @xmath112 over a range of viewing angles . adopting @xmath115 , we find that , in the median , the recovered size is consistent with the true size over the full range of input source magnitudes @xmath116
. however , the scatter in @xmath112 does increase by a factor 2 - 3 as we compare sources with @xmath117 to sources of identical profile shape with @xmath118 .
we caution that different biases may be at play in the size measurement of observed compact quiescent galaxies such as the vd08 sample .
their intrinsic profile shape ( i.e. , noise - free and at infinite resolution ) is unknown , but we did demonstrate it differs from that of our sample of simulated merger remnants as they translate into different ( mock ) observed properties . as an independent test
, we also inserted pure sersic profiles of @xmath2 kpc size , with @xmath119 and a range of luminosities into empty regions of the nicmos images . running galfit on them
, we find qualitatively similar results regarding potential biases in size measurements .
the scatter in the recovered size increases with decreasing signal - to - noise ratio . whereas profiles of similar magnitude as the galaxies from vd08 ( @xmath120 ) were recovered correctly , systematic overestimates of the size by a factor 1.06 ( 1.2 ) in the median and 1.1 ( 2.2 ) in the mean are found when decreasing the brightness of non - exponential ( @xmath121 ) sources by 1 ( 2 ) magnitudes .
in other words , perhaps counter - intuitively , size measurements from 2d surface brightness fitting on low signal - to - noise data may be biased towards size estimates that are too large , rather than too small .
computing the mock observations of our simulated merger remnants using the los and sunrise radiative transfer codes yields qualitatively similar results .
namely , that the simulated surface brightness profiles tend to be cuspier , and in general occupy a different locus in ( @xmath93 , @xmath12)-space than the observed compact galaxies at @xmath37 .
moreover , we confirm that this difference is notable even given the limitations in depth and resolution of present - day observations . in detail
, we note that , for a given simulation , the los code produces cuspier profiles than sunrise , yielding significant residuals after subtracting a @xmath97 fit ( e.g. , rows 1 - 3 of figure[sim_stamps.fig ] ) whereas these are mostly flat when running galfit on sunrise images ( e.g. , rows 4 - 6 of figure[sim_stamps.fig ] ) . finally , we carried out a similar procedure on mock images of simulations where we let disk galaxies evolve in isolation .
the initial disks are designed to have an exponential mass profile .
we checked that galfit correctly recovers their @xmath100 profile shape from mock images of the first simulation snapshot ( values of @xmath122 are recovered for some sightlines due to dust obscuration in the central region ) .
eventually , since gas is consumed by star formation and we did not allow for gas replenishment , the simulated disk galaxies also reach low specific star formation rates . by the time this quiescent phase is reached ,
the disks have evolved in such a way that we find a range of sersic indices similar to that of the vd08 sample ( see figure [ n_re.fig]b ) . however , in all cases , their recovered size is a factor of several too large to reproduce the observed quiescent galaxies .
we now consider a few variations on the formation scenario for compact quiescent galaxies , with an emphasis on how they alter the consistency with observational constraints on size and profile shape .
first , the apparent discrepancy in profile shape of the merger remnants may stem from inadequate assumptions on the structure of the progenitors from which they formed .
figure [ progenitor.fig ] presents a qualititative , but by no means complete , exploration of the vast parameter space in which progenitor properties can vary .
first , we ran a simulation where the spin parameter of the dark matter halos was increased from @xmath123 to @xmath124 ( _ green _ ) , with otherwise default settings .
the higher spin parameter has a stabilizing effect on the gas disk , lowering the star formation rate during the progenitor phase , and increasing the gas fraction at the time of coalescence ( @xmath125 compared to @xmath126 for our fiducial model ) .
the result is an even more concentrated remnant .
less stars formed prior to the merger , but they were redistributed in such a manner that it is mainly at intermediate radii ( around 1 kpc ) that the surface brightness profile is devoid of light with respect to our default simulation .
this is where the residual to a sersic fit with @xmath127 was already negative ( see figure [ n_re.fig]a ) , causing the best fit to run into the upper bound set on @xmath93 again .
second , we investigate a simulation with the same total baryonic mass , but with progenitor galaxies that are built up of a stellar bulge in addition to a gas - rich disk , each contributing half of the baryonic mass ( _ red _ ) .
secular formation processes to grow such a bulge by coalescence of clumps in gas - rich disks ( i.e. , without invoking additional merging ) have recently been proposed by genzel et al .
( 2008 ) , elmegreen , bournaud & elmegreen ( 2008 , and dekel et al .
( 2009b ) .
we find that merging the bulge@xmath128disk systems leads to a compact ( @xmath129 kpc ) remnant with significantly smoother central profile , but the strong wings remain in place .
analyzing its mock observations with galfit yields a locus in ( @xmath93 , @xmath12)-space that does not overlap with the vd08 sample ( see figure[progenitor.fig]b ) .
the same conclusion can be drawn from a third and fourth variation where we increase the disk dispersion twofold ( _ orange _ ) and adjust the merger mass ratio from equal mass to 1:3 ( _ blue _ ) respectively .
the latter variations are inspired by the small @xmath130 ratios of star - forming disk galaxies at @xmath58 observed by the sins survey ( frster schreiber et al .
2009 ) , and by the paucity of equal - mass mergers compared to mergers of mass ratio 1:3 ( naab et al .
2007 ; guo & white 2008 ; hopkins et al .
we do not consider smaller mass ratios as they would not quench the star formation and produce a quiescent spheroid . finally , we considered the effect of adopting a lower pressurization of the ism ( @xmath131 instead of @xmath132 in the nomenclature of springel , di matteo & hernquist 2005 ) .
this leads to enhanced star formation in the outer regions during the early stages of the merging process , resulting in more pronounced wings in the remnant .
this is in qualitative agreement with bournaud et al .
( 2010 ) who find that the remnants of mergers of clumpy disks are cuspier ( i.e. , have higher sersic indices and more material at large radii ) than those produced by merging disks in which the effective ism is not allowed to fragment so completely .
overall , the considered variations in progenitor properties have only a modest impact on the structure of the merger remnants : they occupy a similar region in the size - mass diagram , and have @xmath5 ratio gradients that translate into similar color gradients ( i.e. , red cores ; the dashed and solid lines in figure [ progenitor.fig]a indicate the rest - frame @xmath66 and @xmath6 half - light radii respectively ) .
moreover , whereas the amplitude of the central cusp shows a dependence on the specific variations in progenitor structure explored here , the extended outskirts of the light distribution are present in each of the above model variations . as discussed in section[summary.sec ] , a wider exploration of parameter space , involving physics leading to relative changes in the star formation efficiency between low- and high - density environments , may be appropriate to structurally alter the outer regions of the merger remnants .
in this section , we address whether subsequent merging can alter the structure of the simulated merger remnants in such a way that their profile shape shows a closer resemblance to that of observed quiescent galaxies . to this extent ,
we run a set of re - merger simulations in which the remnant of the initial gas - rich merger collides with a star - forming galaxy of lower mass .
we consider mass ratios 1:6 to 1:2 , and initial gas fractions of 80% , 40% and 5% for the star - forming galaxies with which the merger remnant interacts .
the resulting surface mass densities are presented in figure[subsequent.fig ] . a more extended analysis , involving merger rates as function of mass , mass ratio , redshift and ( redshift - dependent ) gas fraction is required to properly model spheroid evolution ( khochfar & silk 2006b ; naab et al . 2009 ; hopkins et al .
2010a ) , but for our present purpose it is sufficient to highlight two characteristic features notable in our re - merger simulations .
first , the re - merger is consistently larger than the original merger remnant . the relation between the half - mass radius before and after depends on the nature ( mass ratio , gas fraction , ... ) of the re - merger .
as expected , the size growth is larger for the 1:2 than for the 1:6 mass ratios . at a given mass ratio ,
the growth is more pronounced when the amount of dissipation is minimized ( i.e. , dry merging is more efficient at expanding galaxies ) .
second , the re - merger remnant is not homologous to the original gas - rich merger remnant ( see also hopkins et al . 2009c ; hopkins & hernquist 2010 ) .
typically , we see the build - up of an outer wing while the dense central cusp is only little affected .
this effect is more pronounced as we consider re - mergers with a lower dissipational fraction .
the net effect is that , if anything , higher values of @xmath93 are preferred in fitting the products of subsequent merging .
in addition , their size at a given mass grows too big to be consistent with the observed spheroids at @xmath0 . while subsequent ( minor ) mergers might well drive the structural evolution of spheroids from @xmath58 to the present day ( naab et al .
2009 ; van dokkum et al . 2010 ; hopkins et al . 2010a ; carrasco et al . 2010 ) , they are not a viable mechanism to produce the @xmath0 compact galaxies in the first place . even if such an external process would yield sizes and profile shapes that matched the observations , it would not have offered a satisfactory solution , since it is unlikely that every quiescent galaxy observed at @xmath0 experienced a second interaction after the initial quenching event . in light of the last comment ,
it is interesting to consider internal , rather than external , processes that may alter the light distribution .
one such process is stellar mass loss .
the simulations presented in this paper were run with gadget-2 s default settings that account for a small amount of mass loss .
ten percent of the gas mass converted into stars is instantaneously returned to the interstellar medium ( ism , springel & hernquist 2003 ) . with it , energy is injected in the surrounding ism to model the effect of short - lived stars that explode as supernovae . for currently favored stellar initial mass functions ( imfs ) such as a kroupa ( 2001 ) or chabrier ( 2003 ) imf , however ,
eventually as much as half of the gas mass initially converted into stars is returned to the ism .
initially , the mass loss proceeds fast and in an explosive fashion , reaching 10% in 7 myr and 25% in 60 myr . at later times ,
agb stars drive more gradual winds , increasing the released fraction to @xmath3% after a gigayear ( bruzual & charlot 2003 ) . in principle
, three things can happen to the mass lost due to winds : either it escapes the galaxy s potential entirely , or it settles within the galaxy and stays there as gas , or it is consumed by star formation again .
in addition , the redistribution of baryonic mass due to mass loss can potentially change ( lower ) the central potential , causing the distribution of stars to adjust ( expand ) accordingly .
fan et al .
( 2008 ) propose such a scenario of adiabatic expansion due to mass loss driven by agn and/or stellar feedback as a mechanism for spheroid growth between @xmath0 and @xmath39 .
others ( bezanson et al . 2009 ; van dokkum et al .
2010 ; hopkins et al .
2010a ) have objected to this process as a driver of structural evolution since @xmath0 , amongst other reasons because most of the mass loss is expected to take place within 500 myr after the stars are formed , i.e. , prior to the phase when the galaxies are observed as being quiescent .
the same timescale argument works in favor of our present discussion .
if mass loss modifies the surface brightness profile substantially , it will have happened by the time we observe the galaxies as being quiescent .
the precise effect of such extensive mass loss is however not trivial to predict , as the energetics , the amount of mass loss , and the location of the stars when they lose mass all depend on time .
moreover , the timescales of stellar evolution ( i.e. , mass loss ) , star formation history of the system , and merger dynamics all overlap . it will be interesting to investigate in future numerical work how a time - dependent treatment of extensive stellar mass loss and its feedback affects the outcome of simulations as the ones studied here .
since this is beyond the scope of this paper , we now limit ourselves to simple toy models addressing the impact of , e.g. , feedback on how the simulated quiescent galaxies would be observed . in these toy models ,
we apply a simple transformation to a simulation snapshot of a merger remnant : an expansion by a fixed factor of the central component where the young stars dominate ( figure[feedback_rt.fig]a ) , and additional extinction that decreases ( figure [ feedback_rt.fig]c ) or increases ( figure[feedback_rt.fig]e ) with radius .
the first toy model may be thought of as an ad hoc implementation of feedback , whereas the latter two explore dramatic differences in the effect of radiative transfer .
the corresponding diagrams of recovered sersic index versus recovered size when feeding mock nic2 observations of the toy models to galfit are presented in the right - hand panels .
the color - coding traces the expansion or extinction factor , where the legend quotes the extinction at @xmath133 kpc@xmath134 and @xmath135 kpc@xmath134 for the radially decreasing and increasing extinction respectively .
figure [ feedback_rt.fig](b ) shows that , if a physical mechanism that is currently not or improperly modeled is capable of puffing up the central stellar cusp by over an order of magnitude , the best - fit sersic indices enter the regime of those of the observed compact galaxies .
however , the same transformation would lead to a violation of the observational constraints on size by a factor of several .
likewise , larger sizes are inferred from toy model images where the central region is heavily obscured ( figure [ feedback_rt.fig]d ) . moreover , as we already noted in the case of merging bulge@xmath128disk progenitors ( [ progenitor.sec ] ) , a reduction of the central cusp without any change to the outer wings does not necessarily lower the sersic index inferred from the mock observations of limited depth and resolution . a better correspondence with the locus of the vd08 sample in ( @xmath93
, @xmath12)-space is obtained when we lower the amplitude of the profile wings ( figure [ feedback_rt.fig]e ) .
the plotted transformations reduce the true half - light radius ( vertical lines in figure[feedback_rt.fig]e ) by a factor 2 to 7 , but the sizes inferred by galfit are of order 1 kpc . although these toy models have observed properties similar to the vd08 galaxies , the required transformation factors are large ( an order of magnitude or more ) and it is hard to think of a radiative transfer effect related to the dust distribution or intrinsic stellar population properties in the wings that can account for such a deviation from our fiducial model .
there is another way of interpreting the last toy model .
the outskirts are built up of old stars that formed in the progenitor disks and were redistributed by violent relaxation during the merger .
consequently , limiting the amount of star formation at early times may lower the amplitude of the wings . in the binary merger simulations considered here ,
this is hard to accomplish since all gas is already present at the start of the simulation and the compact gas - rich disks are sensitive to gravitational instabilities giving rise to star formation .
this is the case even for the fully - pressurized multiphase ism model adopted throughout this paper ( @xmath136 ) .
a way to circumvent this problem would be to delay the supply of gas or spread it more continuously over time .
indeed , recent cosmological simulations suggest that galaxies during the first 3 billion years after the big bang were not closed systems , and supply of gas through cold filaments and streams was a generic feature of their evolution ( keres et al . 2005 ; dekel & birnboim 2006 ; ocvirk et al . 2008 ; dekel et al .
as such , reproducing not only the stellar population properties and size - mass relation , but also the profile shape of high - redshift quiescent galaxies may require simulations in a cosmological setting .
this is numerically very challenging as a high spatial and timestep resolution ( see [ sim_main.sec ] ) remain essential to model the detailed structure of the merger remnants , as well as the accretion processes onto the supermassive black hole(s ) that contribute to the quenching of their star formation .
consequently , a resimulation technique ( see , e.g. , tormen , bouchet & white 1996 ) may be appropriate , as applied for example by naab et al .
( 2007 , 2009 ) in the context of spheroid formation .
the existence of ultra - compact , massive ( @xmath65 ) quiescent galaxies already in place merely 3 billion years after the big bang has been one of the great surprises of galaxy evolution studies in the last few years .
their discovery has triggered questions regarding their origin and fate , as well as the reliability with which their properties such as mass and size are characterized . in this paper
, we address the formation and nature of such systems in the context of ( gas - rich ) merger scenarios . as such
, this study is part of a larger effort that couples hydrodynamic simulations with radiative transfer in order to investigate how different types of observed high - redshift galaxies may tie together as stages of one evolutionary sequence .
figure [ summary.fig ] illustrates a typical star formation history of a gas - rich merger simulation .
gas - rich star - forming galaxies ( _ blue _ ) are abundantly present at high redshift ( erb et al . 2006 ; tacconi et al .
when in the process of merging ( _ yellow _ ) , the morphology of such systems may ( but will not always ) appear disturbed .
the timescales of a morphological merger signature depend on mass ratio ( lotz et al .
2010a ) and gas fraction ( lotz et al .
2010b ) of the merger .
narayanan et al . (
2010 ) describe how during a relatively short ( @xmath137 myr ) period at final coalescence ( _ orange _ ) a peak in emission at sub - millimeter wavelengths is reached .
continuing over a somewhat longer ( @xmath138 myr ) timespan , the system may be identified as a so - called dust - obscured galaxy ( dog ) , characterized by its large 24 @xmath139 m to optical flux ratio , although not every dog , and particularly those with low 24 @xmath139 m luminosity , is a gas - rich merger ( narayanan et al .
2009 ) . shortly after the peak in star formation rate
( at @xmath140 myr for the particular simulation shown ) , the accretion rate onto the central supermassive black hole peaks .
the role of merging in galaxy - black hole coevolution has been discussed extensively by hopkins et al .
( 2006 , 2008a ) .
finally , wuyts et al . ( 2009b ) investigated the role major mergers may have played in producing the above mentioned compact systems , in terms of their number and mass densities as well as their integrated stellar population properties .
iterating on the latter aspect , the inset panel in figure[summary.fig ] contrasts the distribution of rest - frame uv to near - infrared seds of merger remnants computed from our simulations ( _ median , 50th and 100th percentiles are displayed in red _ ) to the mean rest - frame sed of quiescent galaxies from the vd08 sample ( muzzin et al .
2009b , _ black data points , with error bars representing the error on the mean _ ) .
the model and observations show an excellent agreement over the full wavelength range probed , boosting confidence that the modeled stellar populations reflect reality at least in an integrated sense . in this paper , we focussed on the structure and resolved stellar populations of merger remnants .
we used the sample of massive quiescent galaxies from vd08 as reference sample .
their near - infrared spectra are characterized by a balmer/4000 break ( kriek et al .
deep nic2 observations ( vd08 ) have revealed their compact nature at the highest resolution currently available . analyzing a suite of binary merger simulations of varying mass , gas fraction , progenitor scaling , and orbital configuration ,
we confirm the idea originally proposed by khochfar & silk ( 2006a ) that major mergers can explain their location in the size - mass diagram provided they are gas - rich .
merging galaxies scaled to represent high - redshift star - forming disks , we find that systems of @xmath65 with half - mass radius @xmath2 kpc can be formed when the gas fraction by the time of final coalescence is about @xmath3% .
observational evidence for gas fractions of this magnitude has been accumulating in recent years ( e.g. , tacconi et al .
the corresponding velocity dispersions of these simulated massive compact galaxies are of order 300 - 400 km s@xmath79 .
they show considerable rotation ( @xmath141 of up to unity ) compared to the majority of lower redshift early - type galaxies , a result that has yet to be confirmed observationally . running radiative transfer on the output of our sph simulations , we find that the merger remnants have a radially dependent mass - to - light ratio
. typically , when observing the remnant 500 myr to 1 gyr after the peak in star formation rate , the half - mass radius is a factor @xmath7 smaller than the rest - frame @xmath6-band half - light radius .
this implies that the high effective densities inferred from nicmos and wfc3 observations of high - redshift quiescent galaxies may in fact only be lower limits . in the rest - frame @xmath66-band ,
the typical light - to - mass size ratio increases to a factor 3 - 4 .
the ratio @xmath142 shows a significant sightline - dependence , with a tail to high values ( @xmath143 ) .
our conclusions are robust to the choice of radiative transfer code , input stellar population synthesis models , dust attenuation law , and whether or not the progenitor disks had stellar population gradients .
the @xmath5 ratio gradient is expressed as a color gradient that correlates with the integrated color of the system ( redder galaxies hosting a more pronounced red core ) .
we caution that the presence of a red core can not unambiguously be interpreted as evidence for inside - out growth .
in fact , in the simulations studied in this paper , the central stellar population formed last , rather than first .
the effect of the negative age gradient is compensated by the fact that the stars inside the half - mass radius are more metal - rich , and suffer more extinction . finally , we perform two - dimensional parametric fitting with the galfit code ( peng et al . 2002 ) on real and mock nic2 observations with the same limitations of resolution and depth .
we find that our simulated merger remnants occupy a different region in ( @xmath93 , @xmath12)-space than the vd08 quiescent galaxies .
so do simulations of disk galaxies that are left to evolve in isolation .
the high sersic indices ( @xmath8 ) inferred from the mock observations are driven by both the central cusp ( consisting of young stars ) and the extended wings ( consisting of old stars ) .
we explored a number of variations in progenitor properties that influence the amplitude of the central cusp , but have little effect on the outskirts .
subsequent ( minor ) merging tends to build up the profile wings even further while leaving the central cusp in the mass distribution relatively unaffected . by lack of cosmological context ,
the binary merger simulations start with large amounts of gas in the progenitor disks ( up to 80% of the baryonic mass content ) in order to have a sufficiently large gas fraction by final coalescence to model the dissipational merger event . given this large gas reservoir at the start of the simulation , it is hard to prevent it from forming stars and ending up forming the extended wings of the remnant profiles ( wings that are not seen to that degree in the observations ) . as a consequence , while gas - rich mergers as simulated in this paper reproduce the basic structural diagnostic ( the size - mass diagram ) , second order structural properties such as profile shape may differ because too many stars are formed at early times .
this could be prevented if the gas was not all present initially but instead accreted more gradually over time , or if star formation and/or stellar mass loss and its feedback operated in such a way that the efficiency of converting gas into stars was lower at low densities than assumed here ( i.e. , in the progenitor disks and particularly their outer parts ) .
one could speculate that a low gas - phase metallicity , turbulence from accretion ( bournaud et al . 2010 ) , or the elevated intensity of the ionizing background radiation at @xmath0 inhibit the formation of molecular gas and thus stars during this early phase , especially at low densities where the gas may still be pristine .
we demonstrate that , in addition to spatial resolution , timestep resolution is important in modeling the detailed structure of such extreme systems properly . therefore , running spheroid evolution simulations in a full cosmological context is computationally very expensive .
pioneering efforts to simulate spheroid formation in a cosmological context ( without agn feedback ) have recently been carried out by naab et al .
( 2007 ; 2009 ) .
these authors apply a resimulation technique on three target halos extracted from a low - resolution dark matter simulation to follow the formation and evolution of spheroids within those halos from high redshift to the present day . in agreement with our findings ,
naab et al . ( 2007 ; 2009 ) argue that at high redshift dissipational processes on short timescales are a plausible mechanism to produce the compact galaxies observed at @xmath58 . at later times
, they find that minor and gas - poor merging forms an efficient means to build up an envelope around this core ( see also section [ subsequent.sec ] of this paper , and hopkins et al .
2009a ) . in the scenario investigated in this paper
, the initial dissipational event occurs in the form of a major merger .
such a scenario connects three types of galaxies observed at high redshift in one evolutionary sequence : gas - rich disks , dusty starbursts with star formation rates peaking to @xmath144 , and quiescent remnants .
the dissipational event modeled by naab et al .
( 2007 ; 2009 ) is driven by efficient radiative cooling and collapsing of gas , not necessarily involving major merging .
it remains to be seen whether such a model can account for the peak star formation rates observed in , e.g. , sub - millimeter galaxies , and the post - starburst nature of the stellar populations of compact galaxies .
observationally , measurements of clustering , from large scales to pair counts , based on wide and deep cosmological surveys could potentially provide support for a merger - driven evolutionary scenario .
baker , a. j. , tacconi , l. j. , genzel , r. , lehnert , m. d. , & lutz , d. 2004 , apj , 604 , 125 barnes , j. e. , hernquist , l. 1991 , apj , 370 , l65 barnes , j. e. 1992 , apj , 393 , 484 barnes , j. e. , hernquist , l. 1996 , apj , 471 , 115 bezanson , r. , van dokkum , p. g. , tal , t. , marchesini , d. , kriek , m. , franx , m. , & coppi , p. 2009
, apj , 697 , 1290 binney , j. 1978 , mnras , 183 , 501 binney , j. , & merrifield , m. 1998 , galactic astronomy , princeton university press , princeton bournaud , f. , et al .
2010 , submitted to apjl ( astro - ph 1006.4782 ) bruzual , g , & charlot , s. 2003 , mnras , 344 , 1000 buitrago , f. , trujillo , i. , conselice , c. j. , bouwens , r. j. , dickinson , m. , & yan , h. 2008 , apj , 687 , l61 bullock , j. s. , kolatt , t. s. , sigad , y. , somerville , r. s. , kravtsov , a. v. , klypin , a. a. , primack , j. r. , & dekel , a. 2001 , mnras , 321 , 559 cappellari , m. , et al .
2006 , mnras , 366 , 1126 cappellari , m. , et al .
2009 , apj , 704 , l34 carrasco , e. r. , conselice , c. j. , & trujillo , i. 2010 , mnras , in press ( astro - ph 1003.1956 ) cenarro , a. j. , & trujillo , i. 2009 , apj , 696 , l43 chabrier , g. 2003 , pasp , 115 , 763 cimatti , a. , et al .
2008 , a&a , 482 , 21 coppin , k. e. k. , et al . 2007 , apj , 665 , 936 cox , t. j. , dutta , s. n. , di matteo , t. , hernquist , l. , hopkins , p. f. , robertson , b. , & springel , v. 2006 , apj , 650 , 791 daddi , e. , et al .
2005 , apj , 626 , 680 daddi , e. , dannerbauer , h. , elbaz , d. , dickinson , m. , morrison , g. , stern , d. , & ravindranath , s. 2008 , apj , 673 , l21 dav , r. 2008 , mnras , 385 , 147 dekel , a. , & birnboim , y. 2006 , mnras , 368 , 2 dekel , a. , et al .
2009a , nature , 457 , 451 dekel , a. , sari , reem , & ceverino , d. 2009b , apj , 703 , 785 di matteo , t. , springel , v. , & hernquist , l. 2005 , nature , 433 , 604 di matteo , p. , pipino , a. , lehnert , m. d. , combes , f. , & semelin , b. 2009 , a&a , 499 , 427 elmegreen , b. g. , bournaud , f. , elmegreen , d. m. 2008 , apj , 688 , 67 erb , d. k. , steidel , c. c. , shapley , a. e. , pettini , m. , reddy , n. a. , & adelberger , k. l. 2006 , apj , 646 , 107 fan , l. , lapi , a. , de zotti , g. , & danese , l. 2008 , apj , 689 , l101 frster schreiber , n. m. , et al .
2009 , apj , 706 , 1364 franx , m. , van dokkum , p. g. , frster schreiber , n. m. , wuyts , s. , labb , i. , & toft , s. 2008 , apj , 688 , 770 genzel , r. , et al .
2008 , apj , 687 , 59 groves , b. , dopita , m. a. , sutherland , r. s. , kewley , l. j. , fischera , j. , leitherer , c. , brandl , b. , & van breugel , w. 2008 , apjs , 176 , 438 guo , q. & white , s. d. m. 2008 , mnras , 384 , 2 jonsson , p. 2006 , mnras , 372 , 2 jonsson , p. , groves , b. , & cox , t. j. 2010 , mnras , 403 , 17 hernquist , l. 1990 , apj , 356 , 359 hopkins , p. f. , hernquist , l. , martini , p. , cox , t. j. , robertson , b. , di matteo , t. , & springel , v. 2005 , apj , 625 , l71 hopkins , p. f. , hernquist , l. , cox , t. j. , di matteo , t. , robertson , b. , & springel , v. 2006 , apjs , 163 , 1 hopkins , p. f. , hernquist , l. , cox , t. j. , keres , d. 2008a , apjs , 175 , 356 hopkins , p. f. , hernquist , l. , cox , t. j. , dutta , s. n. , & rothberg , b. 2008b , apj , 679 , 156 hopkins , p. f. , cox , t. j. , & hernquist , l. 2008 , apj , 689 , 17 hopkins , p. f. , hernquist , l. , cox , t. j. , keres , d. , & wuyts , s. 2009a , apj , 691 , 1424 hopkins , p. f. , cox , t. j. , dutta , s. n. , hernquist , l. , kormendy , j. , & lauer , t. r. 2009b , apjs , 181 , 135 hopkins , p. f. , lauer , t. r. , cox , t. j. , hernquist , l. , & kormendy , j. 2009c , apjs , 181 , 486 hopkins , p. f. , bundy , k. , murray , n. , quataert , e. , lauer , t. r. , & ma , c .- p .
2009d , mnras , 398 , 898 hopkins , p. f. , bundy , k. , hernquist , l. , wuyts , s. , & cox , t. j. 2010a , mnras , 401 , 1099 hopkins , p. f. , et al . 2010b , apj , 715 , 202 hopkins , p. f. , & hernquist , l. 2010 , mnras , in press ( astro - ph 1006.0488 ) keres , d. , katz , n. , weinberg , d. h. , & dav , r. 2005 , 363 , 2 khochfar , s. , & silk , j. 2006a , apj , 648 , l21 khochfar , s. , & silk , j. 2006b , mnras , 370 , 902 kriek , m. , et al .
2006 , apj , 649 , 71 kroupa , p. 2001 , mnras , 322 , 231 kuntschner , h. , et al .
2010 , mnras , in press ( astro - ph 1006.1574 ) labb , i. , et al .
2005 , apj , 624 , l81 lotz , j. m. , jonsson , p. , cox , t. j. , & primack , j. r. 2010a , mnras , 404 , 575 lotz , j. m. , jonsson , p. , cox , t. j. , & primack , j. r. 2010b , mnras , 404 , 590 maraston , c. 2005 , mnras , 362 , 799 maraston , c. , daddi , e. , renzini , a. , cimatti , a. , dickinson , m. , papovich , c. , pasquali , a. , & pirzkal , n. 2006 , apj , 652 , 85 mihos , j. c. , hernquist , l. 1994 , apj , 431 , l9 mihos , j. c. , hernquist , l. 1996 , apj , 464 , 641 mo , h. j. , mao , s. , & white , s. d. m. 1998 , mnras , 295 , 319 muzzin , a. , van dokkum , p. g. , franx , m. , marchesini , d. , kriek , m. , & labb , i. 2009a , apj , 706 , 188 muzzin , a. , marchesini , d. , van dokkum , p. g. , labb , i. , kriek , m. , & franx , m. 2009b , apj , 701 , 1839 naab , t. , jesseit , r. , & burkert , a. 2006 , mnras , 372 , 839 naab , t. , johansson , p. h. , ostriker , j. p. , & efstathiou , g. 2007 , apj , 658 , 710 naab , t. , johansson , p. h. , & ostriker , j. p. 2009 , apj , 699 , 178 narayanan , d. , et al .
2009 , mnras , in press ( astro - ph 0910.2234 ) narayanan , d. , hayward , c. c. , cox , t. j. , hernquist , l. , jonsson , p. , younger , j. d. , & groves , b. 2010 , mnras , 401 , 1613 ocvirk , p. , pichon , c. , & teyssier , r. 2008 , mnras , 390 , 13260 paturel , g. , et al .
2003 , a&a , 412 , 45 peng , c. y. , ho , l. c. , impey , c. d. , & rix , h .- w .
2002 , aj , 124 , 266 peirani , s. , crockett , r. m. , geen , s. , khochfar , s. , kaviraj , s. , & silk , j. 2010 , mnras , 405 , 2327 rawle , t. d. , smith , r. j. , & lucey , j. r. 2010 , mnras , 401 , 852 robertson , b. , cox , t. j. , hernquist , l. , franx , m. , hopkins , p. f. , martini , p. , & springel , v. 2006 , apj , 641 , 21 robertson , b. , li , y. , cox , t. j. , hernquist , l. , & hopkins , p. f. 2007 , apj , 667 , 60 ryan , r. e. , et al .
2010 , submitted to apj ( astro - ph 1007.1460 ) sersic , j. l. 1968 , atlas de galaxias australes ( cordoba , argentina : observatorio astronomico , 1968 ) shen , s. , mo , h. j. , white , s. d. m. , blanton , m. r. , kauffmann , g. , voges , w. , brinkmann , j. , & csabai , i. 2003 , mnras , 343 , 978 springel , v. , & hernquist , l. 2002 , mnras , 333 , 649 springel , v. , & hernquist , l. 2003 , mnras , 339 , 289 springel , v. 2005 , mnras , 364 , 1105 springel , v. , di matteo , t. , & hernquist , l. 2005 , mnras , 361 , 776 szomoru , d. , et al .
2010 , apj , 714 , l244 tacconi , l. j. , et al . 2010 , nature , 463 , 781 toft , s. , et al .
2007 , apj , 671 , 285 tormen , g. , bouchet , f. r. , & white , s. d. m. , 1997 , mnras , 286 , 865 trujillo , i. , et al .
2006 , apj , 650 , 18 van der wel , a. , holden , b. p. , zirm , a. w. , franx , m. , rettura , a. , illingworth , g. d. , & ford , h. c. 2008 , apj , 688 , 48 van der wel , a. , & van der marel , r. p. 2008 , apj , 684 , 260 van dokkum , p. g. , & stanford , s. a. 2003 , apj , 585 , 78 van dokkum , p. g. , et al .
2006 , apj , 638 , 59 van dokkum , p. g. 2008 , apj , 674 , 29 van dokkum , p. g. , et al .
2008 , apj , 677 , 5 van dokkum , p. g. , kriek , m. , & franx , m. 2009 , nature , 460 , 717 van dokkum , p. g. , et al .
2010 , apj , 709 , 1018 vitvitska , m. , klypin , a. a. , kravtsov , a. v. , wechsler , r. h. , primack , j. r. , & bullock , j. s. 2002 , apj , 581 , 799 wilkins , s. m. , hopkins , a. m. , trentham , n. , & tojeiro , r. 2008 , mnras , 391 , 363 williams , r. j. , quadri , r. f. , franx , m. , van dokkum , p. , toft , s. , kriek , m. , & labb , i. 2010 , apj , 713 , 738 wuyts , s. , et al .
2007 , apj , 655 , 51 wuyts , s. , franx , m. , cox , t. j. , hernquist , l. , hopkins , p. f. , robertson , b. e. , & van dokkum , p. g. 2009a , apj , 696 , 348 wuyts , s. , et al . 2009b , apj , 700 , 799 younger , j. d. , hayward , c. c. , narayanan , d. , cox , t. j. , hernquist , l. , & jonsson , p. 2009 , mnras , 396 , 66 zirm , a. w. , et al .
2007 , apj , 656 , 66 | we present a detailed analysis of the structure and resolved stellar populations of simulated merger remnants , and compare them to observations of compact quiescent galaxies at @xmath0 .
we find that major merging is a viable mechanism to produce systems of @xmath1 and @xmath2 kpc size , provided the gas fraction at the time of final coalescence is high ( @xmath3% ) , and provided that the progenitors are compact star - forming galaxies , as expected at high redshift .
their integrated spectral energy distributions and velocity dispersions are in good agreement with the observations , and their position in the @xmath4 diagram traces the upper envelope of the distribution of lower redshift early - type galaxies .
the simulated merger remnants show time- and sightline - dependent @xmath5 ratio gradients that result from a superposition of radially dependent stellar age , stellar metallicity , and extinction .
the median ratio of effective radius in rest - frame @xmath6-band light to that in mass surface density is @xmath7 during the quiescent remnant phase .
this is typically expressed by a negative color gradient ( i.e. , red core ) , which we expect to correlate with the integrated color of the system . finally , the simulations differ from the observations in their surface brightness profile shape .
the simulated remnants are typically best fit by high ( @xmath8 ) sersic indices , whereas observed quiescent galaxies at @xmath0 tend to be less cuspy ( @xmath9 ) . limiting early star formation in the progenitors
may be required to prevent the simulated merger remnants from having extended wings . |
the aim of this letter is to investigate the thermodynamic stability problem of a schwarzschild black hole based on a recent approach @xcite , where the equilibrium compatible entropy function of the black hole is considered to be the rnyi one @xcite
. the nonextensive nature of the bekenstein - hawking entropy of black hole event horizons has been noticed @xcite very early on after the thermodynamic theory of black holes had been formulated @xcite , and the corresponding thermodynamic stability problem has also been investigated many times with various approaches .
the standard stability analysis of extensive systems however ( with the criteria that the _ hessian _ of the entropy function has no positive eigenvalues ) , is not applicable for black holes , as it strongly depends on the _ additive _ property of the entropy function , which condition clearly fails to hold in this case .
the standard thermodynamic functions of a schwarzschild black hole are given by @xmath1 and @xmath2 where @xmath3 is the bekenstein - hawking entropy , @xmath4 is the hawking temperature and @xmath5 is the corresponding heat capacity of the black hole . in the classical approach ( concluding from a _ hessian _
analysis ) , schwarzschild black holes appear to be thermodynamically unstable in the canonical treatment , since the heat capacity of the hole is always negative . on the other hand , this approach is clearly not reliable , as the bekenstein - hawking entropy is not additive , and the corresponding hawking temperature is also not compatible with thermal equilibrium requirements @xcite . for a better understanding on the problem , one needs to consider the consequences of nonadditive thermodynamic effects as well . to circumvent this issue , kaburaki _
et al_. @xcite have used an alternative approach , and investigated the thermodynamic stability of black holes by the poincar turning point method @xcite , which is a topological approach , and does not depend on the additivity of the entropy function .
later on , this method has been used to study critical phenomena of higher dimensional black holes and black rings as well @xcite .
in @xcite , we investigated the bekenstein - hawking entropy problem of a schwarzschild black hole by considering the so - called formal logarithm approach @xcite ( discussed below ) , and found that ( if the classical picture can be taken seriously without any quantum corrections in the small energy limit ) , the equilibrium compatible entropy function of the black hole is linear in the hole s mass , and the corresponding zeroth law compatible temperature is constant , i.e. it is independent of the hole s energy .
we also analyzed the thermodynamic stability of the problem , and showed that isolated schwarzschild black holes are stable against spherically symmetric perturbations within this approach . in the present letter however , we are focusing on the direction that we proposed in @xcite , where we regarded the bekenstein - hawking formula as a nonextensive tsallis entropy @xcite .
this model was motivated by the requirement of the existence of an empirical temperature in thermal equilibrium , or in other words , by the satisfaction of the zeroth law of thermodynamics . by applying the formal logarithm method @xcite
, we showed that the zeroth law compatible entropy function of black holes in this model is the rnyi one @xcite , and the corresponding temperature function has an interesting similarity to the one of an ads black hole in standard thermodynamics @xcite . in the general case
, both the tsallis- and the rnyi entropies contain a constant free parameter , whose physical meaning may depend on the concrete physical situation .
in particular , for the problem of black hole thermodynamics , it may arise e.g. from quantum corrections to micro black holes ( a semi - classical approach has been obtained from the bekenstein bound @xcite in @xcite ) , or from finite size reservoir corrections in the canonical ensemble @xcite .
many other parametric situations are also possible .
the purpose of this letter is to extend our study on the tsallis - rnyi problem by investigating the corresponding thermodynamic stability of black holes . in the stability analysis
we consider both the poincar turning point- and the _ hessian _ methods because the rnyi entropy is additive for factorizing probabilities , and hence the standard approach is also applicable . in the obtained results we find perfect agreement from both directions . throughout the paper we use units such as @xmath6 .
nonextensive approaches to black hole thermodynamics have been investigated several times with various methods ( see eg .
@xcite and references therein ) , on the other hand , a zeroth law compatible formulation of nonextensive thermodynamics is a long standing problem , and a possible solution has been proposed only very recently . based only on the concept of composability , abe showed @xcite that the most general nonadditive entropy composition rule which is compatible with homogeneous equilibrium has the form @xmath7 where @xmath8 is a differentiable function of @xmath9 , @xmath10 is a constant parameter , and @xmath11 , @xmath12 and @xmath13 are the entropies of the subsystems and the total system , respectively . by extending this result , bir and
vn investigated non - homogeneous systems as well @xcite , and developed a formulation to determine the most general functional form of those nonadditive entropy composition rules that are compatible with the zeroth law of thermodynamics .
they found that the general form is additive for the formal logarithms of the original quantities , which in turn , also satisfy the familiar relations of standard thermodynamics .
they also showed , that for homogeneous systems the most general , zeroth law compatible entropy function has the form @xmath14,\ ] ] which is additive for composition , i.e. @xmath15 and the corresponding zeroth law compatible temperature function can be obtained as @xmath16 if one assumes additivity in the energy composition . for the classical black hole case
, it is easy to show from the area law of the entropy function , that the bekenstein - hawking formula satisfies the equilibrium compatibility condition of ( [ abe ] ) , as it follows the nonadditive composition rule @xmath17 which is equivalent with the choices of @xmath18 and the @xmath19 limit in the abe formula @xcite .
the corresponding thermodynamic and stability problem for the case of a schwarzschild black hole ( applying also the formal logarithm method ) has been studied in @xcite . in the more general case however , when the parameter @xmath20 , ( originating e.g. from finite size reservoir corrections in the canonical approach @xcite , or from quantum corrections to micro black holes ( see e.g. @xcite and references therein ) ) , the rnyi entropy formula may arise quite generally , when the conditions @xmath21 and @xmath22 are also imposed , due to the consequence of some natural physical requirements ( e.g. triviality , and leading order additivity for small energies ) @xcite .
the rnyi entropy @xcite , defined as @xmath23 , is equivalent with the choices of @xmath24 and @xmath25 in ( [ formlog ] ) , if the original entropy functions follow the nonadditive composition rule @xmath26 which is known as the tsallis composition rule , and @xmath27 is the so - called nonextensivity parameter .
the tsallis entropy is defined as @xmath28 @xcite , and it is easy to show that the formal logarithm of the tsallis formula provides the rnyi entropy @xmath29.\ ] ] in the limit of @xmath30 ( @xmath31 ) , both the tsallis- and the rnyi formulas reproduce the standard boltzmann - gibbs entropy , @xmath32 .
based on the parametric tsallis - rnyi model , we investigated the thermodynamic properties of a schwarzschild black hole in @xcite .
we found that the temperature horizon radius relation is identical to the one obtained from a black hole in ads space by using the original entropy formula , in both cases the temperature has a minimum . according to the rnyi entropy function of black holes
can be obtained by taking the formal logarithm of the bekenstein - hawking entropy , which in the leading order of the @xmath33 parameter follows the nonadditive tsallis composition rule .
therefore , the rnyi entropy of a black hole can be computed as @xmath34,\ ] ] and for the schwarzschild solution it results @xmath35 for comparison ( not presented in @xcite ) , on fig .
[ fig : t ] we plot the temperature functions versus the black hole mass for the schwarzschild
rnyi , the ads boltzmann and the standard schwarzschild
boltzmann cases . on fig .
[ fig : s ] the corresponding entropy functions are also plotted . on all plots in this paper
we use the parameter value @xmath36 , and the corresponding curvature parameter of the ads space is chosen such as to obtain the same @xmath37 . in the following section we discuss the thermodynamic stability of this problem first by considering pure , isolated black holes in the microcanonical treatment , and then by focusing on the canonical ensemble , where the black holes are surrounded by a bath of thermal radiation . for the purpose of generality
, our analysis will be based on the poincar turning point method @xcite following the works of kaburaki et al .
@xcite , however the canonical approach is also considered from the _ hessian _ point of view .
to separate stable and unstable configurations for cases of a one - parameter series of equilibria , poincar developed a powerful analytic approach @xcite .
it has been applied several times to problems in astrophysical and gravitating systems , in particular for the study of the thermodynamic stability of black holes in standard four- @xcite and also in higher dimensions @xcite .
let us only quote here the main results of this method and omit all the details and proofs which can be found in the original references .
suppose @xmath38 is a distribution function whose extrema @xmath39 define stable equilibrium configurations if the extremal value of @xmath40 is a maximum .
consider now the equilibrium value @xmath41 $ ] , where @xmath42 is a solution of @xmath39 .
if the derivative function @xmath43 , plotted versus @xmath44 has the topology of a continuous and differentiable curve , it can be shown that changes of stability will occur only at points where the tangents are vertical . the @xmath40 distribution function
is called _ massieu _ function , the points with vertical tangents are called _ turning points _ , and the points with negative tangents near the turning points are a branch of _ unstable _ configurations , while the points with positive slopes near the turning points are a branch of _ more stable _ configurations . pure , isolated black holes are described by kaburaki et al .
@xcite when a perfectly reflecting , spherical mirror covers the hole just above its event horizon . in this idealistic case
, the black hole can be described without radiation in the microcanonical treatment with @xmath45 being the bekenstein - hawking entropy .
a thermodynamic variable , @xmath44 , is the total mass - energy , @xmath46 , and the conjugate variable of @xmath46 with respect to the entropy is the derivative @xmath47 , which is the inverse temperature . in the standard picture of black hole thermodynamics ,
@xmath48 is the inverse hawking temperature , and the stability curve of a schwarzschild black hole is the linear function @xmath49 .
this straight line represents all the equilibrium configurations without any turning point , and hence the isolated schwarzschild solution in vacuum is stable against spherically symmetric perturbations in the classical ( microcanonical ) treatment @xcite . for the parametric rnyi case ,
the equilibrium compatible entropy function is given in ( [ renyi_sbhst ] ) , and @xmath48 is the inverse rnyi temperature @xmath50 the corresponding stability curve @xmath51 is plotted on fig .
[ fig : pmc ] together with the classical schwarzschild curve for comparison .
as it can be seen , similarly to the standard result , the rnyi curve has no vertical tangent , and since this curve represents all the equilibrium configurations , we can conclude that isolated schwarzschild black holes are thermodynamically stable against spherically symmetric perturbations ( in the microcanonical treatment ) in the rnyi approach as well .
when the black hole is surrounded by an infinite bath of thermal radiation , the appropriate thermodynamic approach to consider is the canonical one . in the traditional picture , the radiation is treated as an ideal reservoir with infinite size and an infinitely large heat capacity , so it can emit or absorb all the heat that is needed by any change of the black hole without modifying its temperature .
recently , more realistic approaches have also been developed by considering large but finite size reservoirs instead of the infinite approximation . in particular , bir showed that finite size reservoir corrections can provide modifications to the standard canonical theory in the form of nonadditive thermodynamics @xcite .
these effects are usually neglected in the classical thermodynamic limit ( when an infinite number of degrees of freedom is present ) , however they can provide relevant modifications to the behavior of finite size systems . in @xcite ,
it has been shown , that in the case of a finite heat bath with a large but finite and constant heat capacity , the inverse heat capacity of the bath may serve as the @xmath33 parameter in the equilibrium compatible entropy composition rule ( [ abe ] ) , and provide the corresponding tsallis- and rnyi entropy formulas in the leading order .
these results provide additional motivation to investigate the rnyi approach for black hole thermodynamics in the canonical treatment .
let us now consider the black hole in the canonical approach .
the black hole entropy is no longer the appropriate massieu function , @xmath40 , which takes its maximum at a stable equilibrium , rather it is @xmath52 where @xmath53 is the helmholtz free energy @xcite . the parameter @xmath44 now is @xmath48 , and the conjugate variable , @xmath43 , is @xmath54 the corresponding stability curve is then @xmath55 , which we plotted on fig .
[ fig : pc ] . , which is the sign of a stability change .
] the canonical stability curves are simply the @xmath56 clockwise rotated versions of the microcanonical ones , and it is immediate to see that a vertical tangent appears at @xmath57 , which belongs to the minimum temperature @xmath58 .
black holes with mass parameter smaller than @xmath37 are _ unstable _ against spherically symmetric perturbations within this approach , however larger black holes with @xmath59 are _ stable _ , as opposed to the standard result where all solutions appear to be unstable
. the stability change at @xmath37 can also be confirmed by the _ hessian _
analysis , i.e. in this case simply by checking the signature of the heat capacity of the schwarzschild black hole in the bath .
the corresponding curves are plotted on fig .
[ fig : c ] .
it can be seen ( as shown in @xcite ) , that the heat capacity has a pole at @xmath37 , analogous to the ads - boltzmann case , and black holes with larger mass parameter have positive heat capacities ( i.e. negative _ hessian _ ) and hence these solutions are thermodynamically stable .
black holes with smaller masses however are unstable solutions .
in their classic paper @xcite , hawking and page investigated the thermodynamic properties of black holes in asymptotically ads spacetimes .
they showed that due to the presence of the gravitational potential of the ads space , the stability properties of black holes are different from their corresponding ones in asymptotically flat spacetimes .
in particular , the canonical ensemble exists for schwarzschild black holes , and they can be in stable equilibrium with thermal radiation at a fixed temperature .
in addition , it has been found that a thermodynamic phase transition occurs between the radiation and the black hole phases at a critical temperature which depends on the ads curvature parameter only .
based on the similarity of our findings to the ads problem , it is motivated to discuss the question of a possible phase transition in the tsallis - rnyi approach as well . by plotting the free energy of the black hole versus the temperature ( for asymptotically flat black holes , i.e. at zero pressure , the gibbs free energy is equivalent with the helmholtz one ) , a hawking - page - like transition can be revealed . on fig .
[ fig : hp ] , the blue , dotted curve represents the branch of black holes with @xmath60 , which are unstable at any temperature .
black holes with @xmath61 ( red , continuous curve ) are stable above the minimum temperature , but they posses positive free energy in the @xmath62 region , and therefore the pure radiation phase ( purple , dashed line ) is the thermodynamically preferred state with its approximately zero free energy . on the other hand , for @xmath63 , the large ( and stable ) black hole configuration becomes the thermodynamically preferred state because its free energy is smaller than zero , and hence , a phase transition occurs at @xmath64 from the radiation- to the black hole phase . from the @xmath65 condition we get @xmath66 , which depends on the @xmath33 parameter only .
this picture is completely analogous to the hawking - page transition of schwarzschild black holes in ads space , where the curvature parameter plays a similar role that is played by the @xmath33 ( @xmath67 ) parameter in the rnyi model .
a relevant difference however , compared to the ads case , is that in our model the rnyi entropy is additive , and hence the corresponding stability results are reliable .
in this letter we studied the thermodynamic stability problem of schwarzschild black holes described by the rnyi formula as equilibrium- and zeroth law compatible entropy function .
first we considered the question of a pure , isolated black hole in the microcanonical approach , and showed that these configurations are stable against spherically symmetric perturbations , just like in the classical picture .
we also considered the problem when the black hole is surrounded by a bath of thermal radiation in the canonical treatment , and found that , as opposed to the standard picture , asymptotically flat , schwarzschild black holes can be in stable equilibrium with thermal radiation at a fixed temperature , and a stability change occurs at a certain value of the mass - energy parameter which belongs to the minimum temperature solution .
black holes with smaller masses are unstable in this model , however larger black holes become stable . these results
are very similar to the ones obtained by hawking and page in ads space within the standard boltzmann entropy approach @xcite . based on this similarity
, we also analysed the question of a possible phase transition in the canonical picture , and showed that a hawking - page - like black hole phase transition occurs in a very similar fashion as in ads space , and the corresponding critical temperature depends only on the @xmath0-parameter of the rnyi formula .
our findings are relevant in many aspects .
parametric corrections to the bekenstein - hawking entropy formula may arise in various kind of physical situations , most importantly from quantum considerations ( stemming either from string theory , loop quantum gravity , or other semi - classical theories ) . in these corrections ,
the perturbation parameters are small , and by connecting them to the @xmath33 parameter in abe s formula ( [ abe ] ) , it can be expected that the tsallis composition rule ( [ tc ] ) is obtained quite generally in the leading order .
in addition , from the requirement of the existence of an empirical temperature in thermal equilibrium , the rnyi entropy arises very naturally via the formal logarithmic method @xcite . based on these lines ,
our obtained results seem to be quite generic for parametric corrections to the bekenstein - hawking model in the small parameter and small energy limit . as a different direction
, we also mentioned that finite size reservoir corrections can result the same tsallis - rnyi model in the canonical picture @xcite , and we expect that many other parametric situations are also possible .
one of the motivations of this approach has been to satisfy the zeroth law of thermodynamics , and it is an important question whether the model is also in accordance with the remaining laws . in particular , bekenstein s generalized second law @xcite states that the sum of the black hole entropy and the common ( ordinary ) entropy in the black hole exterior never decreases on a statistical average . in obtaining this result , bekenstein considered the boltzmann - gibbs formula for estimating the entropy of the system . within our approach
, bekenstein s computations may be repeated by using the rnyi formula instead ( see sec .
[ nte ] for the definition ) , and due to the properties of the rnyi entropy measure @xcite , we expect that the generalized second law remains valid in this approach as well .
we postpone this study for a future work . as for the third law
, the question is even more open .
recent results suggest ( see e.g. @xcite and references therein ) , that the generalized entropy formulas , in particular the rnyi entropy , violate the third law of thermodynamics even when the @xmath0-parameter is close to 1 .
the validity and applicability of the generalized entropies is therefore clearly an unsettled problem , and it is in the focus of active investigations today .
there are many open questions in this field and also many research directions to consider . in this letter
, we studied the simplest problem of a standard 3 + 1 dimensional , schwarzschild black hole , and our plan is to extend our work to more general settings , e.g. rotating , kerr black holes , or dynamic black holes formed by collapsing shells . due to its similarity to the ads problem , our present results might have some relevance from the ads / cft correspondence @xcite point of view as well . further studies to address these questions are in progress and we hope to report on them in a forthcoming publication .
v.g.cz is grateful for discussions with prof .
t.s . bir .
the research leading to this result has received funding from : the european union seventh framework programme ( fp7/2007 - 2013 ) under the grant agreement no .
pcofund - ga-2009 - 246542 ; the fct project sfrh / bcc/ 105835/2014 ; the japanese ministry of education , science , sports , and culture grant - in - aid for scientific research ( c ) ( no .
23540319 ) , and also from the japan society for the promotion of science l14710 grant . t.s .
bir and v.g .
czinner , _ phys .
_ b 726 ( 2013 ) 861 .
a. rnyi , _ acta .
sci . hung . _ 10 ( 1959 ) 193 ; _ probability theory _ , north holland , amsterdam , 1970 .
davies , proc .
r. soc .
lond . a 353 ( 1977 ) 499 - 521 . j.d .
bekenstein , phys .
d 7 ( 1973 ) 2333 ; j.m .
bardeen et al . , commun .
phys . 31 ( 1973 ) 161 ; s.w . hawking , commun . math .
phys . 43 ( 1975 ) 199 ; phys .
d 13 ( 1976 ) 191 .
bir and p. vn , phys . rev .
e 83 ( 2011 ) 061147
. o. kaburaki , i. okamoto and j. katz , phys .
d 47 ( 1993 ) 2234 ; _ class .
_ 10 ( 1993 ) 1323 ; o. kaburaki , gen .
( 1996 ) 843 .
h. poincar , acta .
( 1885 ) 259 .
g. arcioni and e. lozano - tellechea , phys .
d 72 ( 2005 ) 104021 .
czinner , int .
j. mod .
d 24 ( 2015 ) 1542015 .
lett . a 78 ( 1980 ) 219 ; p.t .
landsberg and n.t .
bishop , _ gen .
_ 19 ( 1987 ) 1083 ; p.t .
landsberg , j. stat .
35 ( 1984 ) 159 ; d. pavn and j.m .
rub , gen . rel .
18 ( 1986 ) 1245 ; j. maddox , nature , 365 ( 1993 ) 103 ; g. gour , class .
( 2003 ) 3403 ; j. oppenheim , phys .
e 68 ( 2003 ) 016108 ; a. pesci , _ class .
_ 24 ( 2007 ) 2283 ; r.f .
aranha et al . , int .
j. mod .
d 17 ( 2008 ) 2049 ; c. tsallis and l.j.l .
cirto , eur .
phys . j. c 73 ( 2013 ) 2487 . | thermodynamic stability of black holes , described by the rnyi formula as equilibrium compatible entropy function , is investigated .
it is shown that within this approach , asymptotically flat , schwarzschild black holes can be in stable equilibrium with thermal radiation at a fixed temperature .
this implies that the canonical ensemble exists just like in anti - de sitter space , and nonextensive effects can stabilize the black holes in a very similar way as it is done by the gravitational potential of an anti - de sitter space .
furthermore , it is also shown that a hawking page - like black hole phase transition occurs at a critical temperature which depends on the @xmath0-parameter of the rnyi formula .
non - extensive entropy , black hole thermodynamics , stability |
imaging detector arrays based on transition edge sensors ( tes ) are a maturing technology for a range of both earth and space - base applications , such as infrared astronomy , x - ray astronomy , and material analysis .
squid - based multiplexed readout is needed to obtain a detector noise limited readout chain .
multiplexing in this context implies that multiple tes signals which overlap in time and phase space are made mathematically independent so that they can be transported through a shared readout chain without losing information , by multiplying each tes signal with an independent carrier and subsequently adding these modulated tes signals .
the various multiplexing variants which are being developed can be ordered based on their modulating element .
we distinguish the variants in which the squid is the multiplying element which mounts the signal on a carrier , and the variant in which the tes mounts the signal on the carrier ( fdm ) .
for the first category the tes is biased with a direct voltage ( dc ) , and the modulated signals are separated in the time domain ( tdm)@xcite , or frequency domain ( microwave squid multiplexing @xcite or code domain multiplexing @xcite ) . for the second category the tes is biased with an alternating voltage ( ac ) , and the modulated signals are separated in the frequency domain ( fdm)@xcite . for both variants of multiplexing the signals
have to be bandwidth limited to prevent addition of wide - band noise by using an inductor as low - pass filter , or an inductor and capacitor in series as band - pass filter , respectively .
an important difference between the two categories of multiplexing within the context of this paper , is the property that for case of squid - based modulation the bias sources of the teses are commonly shared between pixels to lower the total wire count , with the consequence that the set points of the pixels can not be chosen individually , whereas for the case of tes - based modulation the tes bias sources are independent and therefore multiplexed .
this reduces the wire count and makes the working points of the pixels individually adjustable .
the individual bias setting per pixel can also be used to change the impedance of the tes bias source for each pixel by applying feedback to the applied bias voltage based on the measured tes current .
this type of feedback has been proposed in different incarnations and for different purposes .
it has been shown @xcite that with this type of feedback the effective thermal decay time of x - ray detectors can be made shorter when the effective internal resistance of bias source is made negative , i.e. @xmath2 , where @xmath3 is the operating resistance of the tes , and @xmath4 the total series resistance in the tes bias circuit .
the same feedback topology has also been applied to compensate parasitic resistances in the tes bias circuit @xcite , to increase the stiffness of the voltage bias and thereby the electro thermal feedback ( etf ) loop gain , leaving a ( slightly ) net positive resistance @xmath5 in the voltage source .
recently , the authors of this paper proposed @xcite to use the regime of @xmath6 for bolometer operation , so that the tes resistance is kept constant under optical loading , and that the small signal parameters of the tes change minimally under optical loading .
as a result , the linearity of the detector is optimised , and the dependence of the cross talk on the signal level is minimised . in this paper
, we show the first experimental results which demonstrate that operation in the regime where @xmath6 is stable both for single pixel and for multiplexed operation , and that the signal - to - noise ratio of the tes - based detector is conserved under the rll with respect to voltage bias . for practical reasons
the experiments have been divided over two setups .
the comparison of the signal - to - noise ratio was performed in a bolometer pixel with a low nep of @xmath7w/@xmath1 .
the multiplexing experiment was performed with x - ray micro calorimeters , because pulse - tube induced micro - vibrations allowed for stable operation of only one pixel in the bolometer multiplexer during this particular experiment .
-curve of a bolometer pixel as taken under voltage bias .
the pixel consists of a tes and absorber on a sin island , which is suspended by four sin legs of @xmath8 .
the size of the tiau tes equals @xmath9 m , with @xmath10 mk , a bias power level of @xmath11 fw , and a measured dark nep of @xmath7w/@xmath1 @xcite .
the bias points under which the pixel has been characterised using the rll are indicated with the crosses . ]
the experiments were performed with a 3-pixel fdm setup with integrated lithographic 8-pixel @xmath12 bandpass filters @xcite , and a two - pixel x - ray fdm setup , both with baseband feedback ( bbfb ) electronics which provides squid linearisation and signal demodulation .
the rll was integrated in the firmware of the bbfb electronics , following the block diagram as sketched in fig .
[ fig : setup ] .
the system is adjusted such that the dissipative tes current @xmath13 is aligned fully in the @xmath14-channel of the readout .
the @xmath15-channel feedback is required to ensure stability of the bbfb system .
the output of the @xmath14-channel , i.e. the measured tes current , is multiplied with a constant ( @xmath16 ) , which is proportional to the resulting set point resistance of the tes , as the applied bias voltage @xmath17 .
the switch toggles between the ( standard ) voltage bias condition and the rll mode .
to demonstrate the stability of the rll , and to compare the signal - to - noise ratio of a bolometer pixel under voltage bias and the rll , a typical bolometer pixel under development for the sw band of safari @xcite was used , at a bias frequency of 1.46 mhz .
the pixel consists of a tes and absorber on a sin island , which is suspended by four sin legs of @xmath8 .
the size of the tiau tes equals @xmath9 m , with @xmath10 mk , a bias power level of @xmath11 fw , and a measured dark nep of @xmath7w/@xmath1 @xcite .
the pixel has been characterised for a range of bias points in the transition under both voltage bias and the rll .
the current - voltage characteristic of the pixel as obtained under voltage bias is shown in fig . [
fig : ivcurve ] .
the bias points which have been characterised using the rll are marked with crosses .
the latter range is limited because of electrothermal stability constraints , which originate from the much higher electrothermal loop gain under the rll@xcite , in combination with the fact that the circuit was originally designed for operation under voltage bias only .
before discussing the results , we need to consider that under voltage bias the observed current noise spectra density and the noise equivalent power ( nep ) scale linearly , as the bias voltage is kept constant . under the rll
the situation is different , as not the bias voltage is kept constant , but the tes resistance instead . as a result ,
the detector nep is proportional to the tes bias current @xmath18 squared , as the dissipated power equals @xmath19 . in the small signal limit
this implies that to calculate the nep the observed current noise spectral density must be multiplied by 2 , as the square of the observed current @xmath18 equals @xmath20 in the small signal limit . from the rll small signal model it follows that the detector nep is equal under voltage bias and the rll
this implies that also the information bandwidth of the detector , i.e. the frequency at which the phonon noise and johnson noise of the detector are equal , is the same under rll and voltage bias .
it also follows from the small signal model that the measured current noise spectral densities under voltage bias and the rll cross sect at the information bandwidth . for frequencies above the information bandwidth of the detector , but below the electrical bandwidth , the behaviours diverge . for that regime under voltage bias
we expect to observe the current noise spectral density corresponding to a resistor of @xmath21 , whereas under the rll this resistor value equals @xmath22 .
the measured current noise power spectral densities for four different bias points ( @xmath23 0.55 , 0.58 , 0.62 , and 0.74 ) in the transition are shown in fig .
[ fig : rll - vb - noise ] . to make a direct comparison with the noise power spectral density under voltage bias possible ,
the spectra as measured under the rll have been multiplied by a factor of two , as discussed above . as a result of this multiplication ,
the information band is not found anymore at the cross section between the spectra , but at the frequency where the difference equals a factor of 2 .
note that as we intend to do a relative comparison , and because we do nt have an optical signal source available to measure the responsivity of the detector , the observed current noise spectral density has not been converted into calibrated nep values .
instead , we present the current noise spectral densities .
it is clear that in the low frequency region , i.e. @xmath24 hz which is within the information bandwidth of the detector , the noise spectra overlap , as was to be expected as discussed above .
we also observe that for higher frequencies the observed noise power is much higher for the rll than under voltage bias .
this qualitatively scales with the expected factor @xmath25 , assuming that @xmath26 .
unfortunately impedance data is lacking for this detector so that a quantitative comparison is not possible .
however , the onset of electrothermal oscillations between 1 and 2 khz shows that the thermal bandwidth of the detector , which is approximately equal to the information bandwidth ( @xmath27 hz ) , has indeed shifted with a significant factor of @xmath28 with respect to the situation under voltage bias . to demonstrate that the rll can also be applied for multiple pixels simultaneously , we put two x - ray microcalorimeter pixels simultaneously under resistance locking .
it was found that the pixels operated stable , and that the observed dynamic behaviour and nep did not change when the second pixel was added ( not shown ) .
note that the observed currents are used for biasing the tess , to close the rll ( see fig .
[ fig : setup ] ) . as an illustration of the resulting beating in the current
is shown in the time domain plot in fig .
[ fig:2pix ] .
frequency domain multiplexing makes it possible to create a resistance locked loop per pixel , as the pixel bias voltages are multiplexed and therefore individually tuneable .
the rll improves detector speed , linearity and dynamic range for bolometer operation .
it was shown that for fdm resistance locking can easily be integrated in the digital baseband feedback electronics which is used for demodulation and squid linearisation .
it was found that within the information bandwidth of the detector the observed neps are equal , and that closing multiple loops simultaneously does not impair the stability of the system . when combined with the earlier observation @xcite that for tes - based bolometers the performance under ac bias is equal to under dc bias , we believe that this mode of operation is suitable for larger tes - based bolometer applications .
o. noroozian , j. a. b. mates , d. a. bennett , j. a. brevik , j. w. fowler , j. gao , g. c. hilton , r. d. horansky , k. d. irwin , z. kang , d. r. schmidt , l. r. vale , and j. n. ullom , `` high - resolution gamma - ray spectroscopy with a microwave - multiplexed transition - edge sensor array , '' _ applied physics letters _ , vol .
103 , no . 20 , pp . ,
[ online ] .
available : http://scitation.aip.org/content/aip/journal/apl/103/20/10.1063/1.4829156 m. a. dobbs , m. lueker , k. a. aird , a. n. bender , b. a. benson , l. e. bleem , j. e. carlstrom , c. l. chang , h .-
cho , j. clarke , t. m. crawford , a. t. crites , d. i. flanigan , t. de haan , e. m. george , n. w. halverson , w. l. holzapfel , j. d. hrubes , b. r. johnson , j. joseph , r. keisler , j. kennedy , z. kermish , t. m. lanting , a. t. lee , e. m. leitch , d. luong - van , j. j. mcmahon , j. mehl , s. s. meyer , t. e. montroy , s. padin , t. plagge , c. pryke , p. l. richards , j. e. ruhl , k. k. schaffer , d. schwan , e. shirokoff , h. g. spieler , z. staniszewski , a. a. stark , k. vanderlinde , j. d. vieira , c. vu , b. westbrook , and r. williamson , `` frequency multiplexed superconducting quantum interference device readout of large bolometer arrays for cosmic microwave background measurements , '' _ review of scientific instruments _ , vol .
83 , no . 7 , pp . ,
[ online ] .
available : http://scitation.aip.org/content/aip/journal/rsi/83/7/10.1063/1.4737629 j. van der kuur , j. beyer , m. bruijn , j. gao , r. den hartog , r. heijmering , h. hoevers , b. jackson , b. van leeuwen , m. lindeman , m. kiviranta , p. de korte , p. mauskopf , p. de korte , h. van weers , and s. withington , `` the spica - safari tes bolometer readout : developments towards a flight system , ''
_ j. low temp . phys . _ , vol .
167 , pp . 561567 , 2012 .
s. w. nam , b. cabrera , p. colling , r. clarke , e. figueroa - feliciano , a. miller , and r. romani , `` a new biasing technique for transition edge sensor with electrothermal feedback , '' _ ieee trans .
_ , vol . 9 , no . 2 ,
pp . 42094212 , 1999 .
j. van der kuur and m. kiviranta , `` operation of transition edge sensors in a resistance locked loop , '' _ applied physics letters _
102 , no . 2 , p. 023505 , 2013 .
[ online ] .
available : http://link.aip.org/link/?apl/102/023505/1 m. bruijn , l. gottardi , r. den hartog , j. van der kuur , a. van der linden , and b. jackson , `` , '' _ _ , vol .
176 , no . 3 - 4 ,
pp . 421425 , 2014 .
[ online ] .
available : http://dx.doi.org/10.1007/s10909-013-1003-6 b. d. jackson , p. a. j. de korte , j. van der kuur , p. d. mauskopf , j. beyer , m. p. bruijn , a. cros , j .-
gao , d. griffin , r. den hartog , m. kiviranta , g. de lange , b .-
van leeuwen , c. macculi , l. ravera , n. trappe , h. van weers , and s. withington , `` the spica - safari detector system : tes detector arrays with frequency division multiplexed squid readout , '' _ ieee trans .
_ , vol . 2 , no . 1 , pp . 1221 , 2012 .
l. gottardi , h. akamatsu , m. bruijn , j .-
gao , r. den hartog , r. hijmering , h. hoevers , p. khosropanah , a. kozorezov , j. van der kuur , a. van der linden , and m. ridder , `` weak - link phenomena in ac - biased transition edge sensors , '' _ journal of low temperature physics _ , vol .
176 , pp . 279284 , | tes - based bolometer and microcalorimeter arrays with thousands of pixels are under development for several space - based and ground - based applications .
a linear detector response and low levels of cross talk facilitate the calibration of the instruments . in an effort to improve the properties of tes - based detectors , fixing the tes resistance in a resistance - locked loop ( rll ) under optical loading
has recently been proposed .
earlier theoretical work on this mode of operation has shown that the detector speed , linearity and dynamic range should improve with respect to voltage biased operation .
this paper presents an experimental demonstration of multiplexed readout in this mode of operation in a tes - based detector array with noise equivalent power values ( nep ) of @xmath0w/@xmath1 .
the measured noise and dynamic properties of the detector in the rll will be compared with the earlier modelling work .
furthermore , the practical implementation routes for future fdm systems for the readout of bolometer and microcalorimeter arrays will be discussed . |
stochastic processes with an absorbing state arise frequently in statistical physics @xcite . in systems with spatial structure , phase transitions to an absorbing state , as exemplified by the contact process @xcite , are widely studied in connection with self - organized criticality @xcite , the transition to turbulence @xcite , and issues of universality in nonequilibrium critical phenomena @xcite .
interest in such transitions should continue to grow in the wake of experimental confirmation in a liquid crystal system @xcite .
this letter presents a new theoretical approach to absorbing - state phase transitions via analysis of exact ( numerical ) quasistationary ( qs ) probability distributions .
the quasistationary probability distribution ( qsd ) provides a wealth of information about systems exhibiting an absorbing - state phase transition @xcite .
( since the only true stationary state for a finite system is the absorbing one , stationary - state " simulations in fact probe qs properties , that is , conditioned on survival . ) in particular , the order parameter and its moments , static correlation functions , and the qs lifetime are all accessible from the qsd . until now , qs properties of systems with spatial structure have been determined only via simulation @xcite ; here i develop an effective scheme for determining the qsd on rings of @xmath3 sites .
the qsd is defined as follows .
consider a continuous - time markov process @xmath4 with state @xmath5 absorbing : if @xmath6 , then @xmath7 at all subsequent times .
the transition rates @xmath8 ( from state @xmath9 to state @xmath10 ) are such that @xmath11 , @xmath12 .
( some processes have several absorbing states , @xmath13 . )
let @xmath14 denote the probability of state @xmath9 at time @xmath15 , given some initial state @xmath16 .
the _ survival probability _ @xmath17 is the probability that the process has not visited the absorbing state up to time @xmath15 . we suppose that as @xmath18 the @xmath19 , normalized by the survival probability , attain a time - independent form , allowing us to define the qsd : in principle , one could integrate the master equation numerically and extract the qsd in the long - time limit .
such an approach is very time - consuming , and essentially useless for processes with a large state space .
i use instead the iterative scheme demonstrated in @xcite .
given some initial guess for the distribution @xmath23 , the following relation is iterated until convergence is achieved : here @xmath25 is the probability flux ( in the master equation ) into state @xmath9 ( @xmath26 is the flux to the absorbing state , so that @xmath27 gives the lifetime of the qs state ) , and @xmath28 is the total rate of transitions out of state @xmath9 .
the parameter @xmath29 can take any value between 0 and 1 ; in practice rapid convergence is obtained with @xmath30 .
the iterative scheme is used to determine the qsd of the _ contact process _ ( cp ) on rings of @xmath3 sites . in the cp @xcite , each site @xmath31 of a lattice
is either occupied ( @xmath32 ) , or vacant ( @xmath33 ) .
transitions from @xmath34 to @xmath35 occur at a rate of unity , independent of the neighboring sites .
the reverse transition is only possible if at least one neighbor is occupied : the transition from @xmath35 to @xmath34 occurs at rate @xmath36 , where @xmath37 is the fraction of nearest neighbors of site @xmath31 that are occupied ; thus the state @xmath35 for all @xmath31 is absorbing .
( @xmath38 is a control parameter governing the rate of spread of activity . )
although no exact results are available , the cp has been studied intensively via series expansion and monte carlo simulation . since
its scaling properties have been discussed extensively @xcite we review them only briefly here .
the best estimate for the critical point in one dimension is @xmath39 , as determined via series analysis @xcite . approaching the critical point ,
the correlation length @xmath40 and lifetime @xmath41 diverge , following @xmath42 , where @xmath43 is the relative distance from the critical point .
the order parameter ( the fraction of active sites ) , scales as @xmath44 for @xmath45 . near the critical point , finite - size scaling ( fss ) @xcite , implies that average properties such as @xmath46 depend on @xmath3 through the scaling variable @xmath47 , leading , at the critical point , to @xmath48 , with dynamic exponent @xmath49 , and @xmath50 .
the computational algorithm for determining the qsd consists of three components .
the first ( applicable to any model having two states per site ) enumerates all configurations on a ring of @xmath3 sites .
configurations differing only by a lattice translation are treated as equivalent . in subsequent stages
only one representative of each equivalence class is used , yielding a considerable speedup and reduction in memory requirements .
( the multiplicity , or number of configurations associated with each equivalence class , is needed for calculating observables . )
the second component runs through the list of configurations , enumerating all possible transitions .
here proper care must be taken to determine the weight of each transition , due to the varying multiplicity of initial and final configurations .
the exit rate for each configuration @xmath9 is simply : @xmath51 , where @xmath52 is the number of occupied sites and @xmath53 the number of occupied - vacant nearest - neighbor pairs . to determine @xmath54 one enumerates , for each configuration , all transitions from some other state @xmath10 to @xmath9 .
( each vacant site @xmath31 in @xmath9 implies a transition from a configuration @xmath10 , differing from @xmath9 only in that site @xmath31 is occupied ; each nearest - neighbor pair of occupied sites @xmath55 in @xmath9 implies transitions from a @xmath10 in which either @xmath31 or @xmath56 is vacant .
transitions between the same pair of configurations @xmath10 and @xmath9 are grouped together , with the proper multiplicity stored in the associated weight . )
the final part of the algorithm determines the qsd via the iterative procedure described above .
the specific rules of the model enter only in the second stage ; extension to other models is straightforward .
i determined the qsd for the contact process on rings of up to 23 sites .
the number of configurations scales as @xmath57 ( for @xmath3 prime this formula is exact ) .
the number of annihilation transitions is @xmath58 ( on average half the sites are occupied ) and that of creation transitions is @xmath59 .
( for @xmath60 , there are 364@xmath61723 configurations , and @xmath62 transitions ; the calculation takes about 16 hours on a 3 ghz processor . )
the qs order parameter @xmath46 ( fig .
[ qsex1 ] ) , follows the anticipated trend ( i.e. , @xmath63 is a sigmoidal function ) , but does not show any clear sign of the critical point ; indeed , no such sign is expected for the small systems considered here . a precise estimate of the critical value @xmath64 can nevertheless be obtained through analysis of the moment ratios .
let @xmath65 denote the @xmath66-th moment of the occupied site density , and @xmath67 .
the values @xmath68 , marking the crossing of @xmath69 and @xmath70 , approach the critical value systematically , as shown in fig .
[ cplms ] .
( @xmath68 is plotted , for convenience , versus @xmath71 as this leads to an approximately linear plot . )
once preliminary estimates of @xmath68 ( with uncertainty @xmath72 ) have been obtained , i perform high - resolution studies , with @xmath73 , in the vicinity of each crossing ; precise estimates of the crossing values ( uncertainty @xmath74 ) are then obtained applying neville s algorithm @xcite to the data for @xmath75 . using the bulirsch - stoer ( bst ) extrapolation technique @xcite , the data for sizes 8 to 23 furnish @xmath76 and the critical moment ratio @xmath77 .
these values compare well with the best available estimates of @xmath78 and @xmath79 @xcite ; the associated errors are @xmath80 0.05% , remarkably small , in light of the system sizes used .
the moment ratios @xmath81 also exhibit crossings that converge to @xmath64 .
more surprisingly , the product @xmath82 exhibits crossings and appears to approach a well defined limit , 1.366(1 ) , as @xmath83 at the critical point ; simulations ( @xmath84 and 2000 ) , yield 1.374(2 ) for this quantity .
the reduced fourth cumulant , or kurtosis , given by @xmath85 , where @xmath86 ( the variance of the order parameter ) and @xmath87 , does not exhibit crossings but instead takes a pronounced minimum at a value @xmath88 that converges to the critical value as @xmath89 .
( this property has been verified in simulations using @xmath84 , and 2000 ; departures from the minimum value are evident for @xmath90 @xcite . ) the sharpness of the minimum , as gauged by @xmath91 , appears to increase rapidly with size : @xmath92 .
( this is consistent with @xmath93 , as expected from fss . ) since a negative kurtosis reflects a probability distribution that is broader at the maximum , and with shorter tails ( compared to a gaussian distribution with the same mean and variance ) , it is natural that @xmath94 should be minimum at the critical point , where fluctuations are dominant .
the statistical entropy per site , @xmath95 , is plotted in fig .
1 . in the large-@xmath3 limit
, @xmath96 should be zero for @xmath97 , since the qsd is concentrated on a set of configurations with vanishing density .
as @xmath83 , one expects @xmath98 to diverge ( as is the case for @xmath99 ) , and to attain a maximum at some @xmath100 , approaching zero as @xmath101
. the numerical data are consistent with these trends . encouraged by the good results of the moment - ratio analysis , i examine three quantities expected , on the basis of fss , to exhibit scaling at the critical point : the qs lifetime @xmath41 , the order parameter , and the derivative @xmath102 .
( the latter should diverge @xmath103 . ) despite the small system sizes , these quantities indeed appear to follow power laws ( see fig . [ rhl ] ) . to obtain precise estimates of the associated exponents ,
i calculate the finite difference ratios @xmath104/[\ln l - \ln ( l-1)]$ ] , ( and similarly for @xmath41 and @xmath105 ) .
linear regression of these ratios versus @xmath106 , using the seven largest sizes , yields @xmath107 , @xmath108 , and @xmath109 .
these estimates differ by 0.06% , 0.01% , and 0.5% , respectively , from the literature values of 0.25208 , 1.5807 , and 1.0968 @xcite .
thus data on qs properties of systems with 23 sites or fewer yield estimates of critical exponents to within half a percent or better !
precise estimates are also found for @xmath110 and @xmath94 at the critical point : using bst extrapolation i find values of 1.5306(5 ) and -0.5015(5 ) , compared with the simulation values of 1.526(3 ) and -0.505(3 ) , respectively @xcite . to test the robustness of this approach ,
i apply it to the _ pair contact process _ ( pcp ) . in the pcp @xcite
, each site is again either occupied or vacant , but all transitions involve a pair of particles occupying nearest - neighbor sites , called a _
pair _ in what follows .
a pair annihilates itself at rate @xmath111 , and with rate @xmath112 creates a new particle at a randomly chosen site neighboring the pair , if this site is vacant .
any configuration lacking a pair of nearest - neighbor occupied sites is absorbing .
simulation results @xcite place the pcp in the same universality class as the cp ( namely , that of directed percolation ) .
unlike the cp , for which quite precise results have been derived via series expansions , there are no reliable predictions from series or other analytic methods . using , as before , the parameter values associated with crossings of the moment ratio @xmath113 , i obtain ( for system sizes @xmath114 to 23 ) , the estimate @xmath115 , about 0.3% above the best available estimate of 0.077092(1 ) @xcite .
the estimates @xmath116 and @xmath117 , obtained via the same procedure as used for the cp , are also in good agreement with the accepted values .
analysis of the qs lifetime however , yields the unexpectedly large value @xmath118 . in fact , the finite - difference ratios @xmath119 vary erratically with @xmath3 , indicating that the result for @xmath0 is unreliable .
this may be associated with the large number of absorbing configurations in the pcp ( growing exponentially with @xmath3 ) , so that the extinction rate has not reached its asymptotic limiting behavior at the system sizes considered here .
extrapolation of the moment ratio at @xmath120 yields @xmath121 , and reduced fourth cumulant @xmath122 , again in good accord with the expected values .
( as in the case of the cp , the value of @xmath123 at which @xmath94 takes its minimum approaches @xmath64 with increasing system size . )
thus the qs properties of the pcp ( for @xmath124 ) permit one to assign the model to the directed percolation class , despite the lack of a clear result for the dynamic exponent @xmath0 .
it is natural to inquire whether the qs probability distribution exhibits any simplifying features . in an equilibrium lattice gas with interactions that do not extend beyond nearest neighbors , for example
, the probability of a configuration depends only on the number of particles @xmath125 and nearest - neighbor pairs @xmath126 .
the the cp , by contrast , i find that the qs probability of each configuration in a given @xmath127 class is distinct ( the probabilities typically vary over an order of magnitude or more , even far from the critical point ) . in a broad sense , this is because , unlike in equilibrium , not all annihilation events possess a complementary creation event . for similar reasons , it does not appear likely that the qsd could be obtained via the maximization of the statistical entropy , subject to some simple set of constraints . in summary ,
i show that analysis of exact ( numerical ) quasistationary properties on relatively small rings yields remarkably precise results for critical properties at an absorbing - state phase transition . deriving
the qs distribution involves rather modest programming and computational effort : the results reported here can be obtained in a few days on a fast microcomputer .
applied to the contact process , the analysis yields most critical properties with an error well below 0.1% . for the more complicated pcp ,
errors are generally @xmath128 1% .
application to other absorbing - state phase transitions , including some belonging to other universality classes , is in progress
. the method may also be useful in the study of metastable states , provides a valuable check on simulations , and may serve as the basis for phenomenological renormalization group approaches . | i derive precise results for absorbing - state phase transitions using exact ( numerically determined ) quasistationary probability distributions for small systems .
analysis of the contact process on rings of 23 or fewer sites yields critical properties ( control parameter , order - parameter ratios , and critical exponents @xmath0 and @xmath1 ) with an accuracy of better than 0.1% ; for the exponent @xmath2 the accuracy is about 0.5% .
good results are also obtained for the pair contact process . |
critical phenomena @xcite in anisotropic systems without equivalent nearest neighbors constitute an interesting research topic @xcite . a universal formula for percolation thresholds , that involves the dimension of the anisotropic lattice and an arithmetic average of the coordination number for different anisotropic lattices , has been recently postulated in ref.@xcite .
the extension of these studies to more complex problems , such as directed percolation ( dp ) , and more complex systems , such as anisotropic random systems , is yet to be addressed . in this context ,
random systems are good candidates to model anisotropy since they do not have equivalent nearest neighbors nor equivalent sites at all lengths . in this work
we propose a simple simulation model to study the properties of dp in two - dimensional ( 2d ) anisotropic random media .
the degree of anisotropy is computed by means of the ratio @xmath3 between the axes of a semi - ellipse enclosing the bonds that promote percolation in one direction , such that @xmath4 ( see fig.1 ) . as a function of the order parameter @xmath0 and at the percolation threshold , we measure the correlation length exponent @xmath5 and the fractal dimension @xmath6 of the largest percolating clusters ( in systems of up to 51200 random sites ) . in the present model ,
the well - known scaling exponents of isotropic dp follow by simply setting @xmath2 . at percolation threshold ,
our model shows that the average number of bonds per site for dp in anisotropic 2d random systems is an invariant ( @xmath7 ) independently of @xmath0 .
this result suggests that the sinai theorem , proposed originally for isotropic percolation ( ip ) , is also valid for anisotropic dp problems .
the new invariant also yields a constant @xmath1 for all @xmath0 , which corresponds to the value of isotropic dp .
the paper is organized as follows . in the next section
we outline our model . in sec.iii , we present the results of our simulations and discuss the effects of @xmath0 on the scaling exponents .
in order to simulate dp in 2d anisotropic random media we develop a simulation algorithm similar to the one used in ref.@xcite .
the coordinates of @xmath8 sites are generated at random in a square box of size @xmath9 .
the simulation length unit is chosen such that the density of sites , namely @xmath10 , in the box is always unity regardless of the total number of sites @xmath8 .
the percolation is then checked over sites from the left edge towards the right edge of the simulation box ( _ i.e. _ , along the x - axis in fig.1 ) . a periodical boundary condition is applied in the vertical @xmath11-direction . in fig.1 we show a ` particle ' that moves from @xmath12 to @xmath13 .
the moving is allowed whenever the site @xmath13 is contained within the shaded elliptical area . in our simulations , the degree of anisotropy is given by the parameter @xmath14 , where @xmath15 is the longer and @xmath16 is the shorter axis of a semi - ellipse , _
i.e. _ , @xmath0 is the ratio of the maximum ` hopping distances ' along the @xmath17- and @xmath11-axes . in the standard 2d isotropic dp
there are three possible equivalent directions to move : up , down and forward .
this situation in our model is attained by setting @xmath0 = 1 . in the limit @xmath18
, the model tends to the one - dimensional ( 1d ) percolation problem .
thus , simulation results using the present 2d percolation model will reveal features of the crossover from the standard ( say , isotropic ) dp to the 1d percolation problem . for intermediate values of @xmath19
our model features anisotropic dp . for a given value of the anisotropy parameter @xmath0 and for a given realization of random site coordinates @xmath20 , in a sample of size @xmath21 , we study percolation from the left- to the right- simulation box edge . at the percolation threshold , we obtain the critical value of the semi - ellipse - axis @xmath16 : @xmath22 and the mass @xmath23 of the critical cluster : @xmath24 = `` _ total number of sites belonging to the largest cluster at percolation _ '' .
these quantities , are then averaged over a great number @xmath25 of random realizations of site coordinates for the same sample size @xmath8 which result on the average quantities @xmath26 and @xmath27 , respectively . in general , the dependence of the averages @xmath28 and @xmath29 on the samples size @xmath8 is a consequence of the finite size effects of the percolation problem . in order to quantify these effects ,
the present simulations were performed at different @xmath8 = 400 , 800 , 1600 , 3200 , 6400 , 12800 , 25600 and 51200 .
accordingly , the number @xmath25 decreases from @xmath30 to @xmath31 such that the product of the numbers @xmath32 is approximately the same for all sample sizes @xmath8 in our study . along with these average quantities ,
we also calculate the moments @xmath33^{2 } > ^{1/2 } \;\;\ ; , \;\;\ ; \label{eq : delta } \\
\delta m(n ) & = & < [ m(n ) - { \cal m}(n)]^{2 } > ^{1/2 } \;\;\ ; , \end{aligned}\ ] ] and also the next - order moments , which are used to estimate the statistical errors of our simulation results . the present measurements are performed for various values of @xmath34 and @xmath35 . as can be seen from the results discussed in the next section ,
the greater the value of @xmath0 , the stronger the finite size effects are .
we verify that for @xmath36 simulations can only been carried out in samples of size @xmath37 . following the well - known finite - size scaling procedure suggested in ref.@xcite
, the critical exponent @xmath5 of the percolation problem is defined from the scaling expression @xmath38 where @xmath39 is given in eq.([eq : delta ] ) .
note that in the present study percolation is checked by the longitudinal direction only ( the @xmath17-axes in fig.1 ) , then the exponent @xmath5 in eq.([eq : mu ] ) should be identified with the parallel @xmath40 ( see @xcite ) .
in fig.2(a ) the quantities @xmath41 are plotted versus @xmath42 for different values of the order parameter @xmath0 .
the slopes of the fitting lines give the corresponding values for the exponent .
thus we found that for the largest @xmath43 , @xmath44 and for @xmath45 we measured @xmath46 .
other values are given in the figure . from these calculated moments and the linear fitting procedure
, we estimate the statistical error to be less than @xmath47 for all values of @xmath5 shown in this figure .
results for the dp limiting case @xmath2 has been previously reported by one of us @xcite . in this case , the value @xmath48 is known as the universal value of @xmath40 for a whole class of istropic dp models in 2d . as the amount of anisotropy increases , _
i.e. _ @xmath49 , the correlation length exponent @xmath5 decreases .
since this decrease is initially very fast to then become smoothly , it is not possible to obtain the whole crossover from 2d to 1d directed percolation for the behaviour of @xmath5 .
that is , the decrease from @xmath50 in 2d - dp to the limit @xmath51 in 1d is limited by the size of our simulation box .
the finite - size effects in correspondence to the different values of @xmath0 ( and , therefore , to the different degrees of anisotropy in the 2d random systems ) are in fact equivalent to those discussed in great detail in refs.@xcite for anisotropic percolation and isotropic dp . by using the values of @xmath5 in fig.2(a ) ,
the critical ` _ radius _ ' @xmath52 is determined from the scaling expression @xmath53 are plotted versus @xmath54 . from these plots
we obtain @xmath55 by taking the asymptotic values @xmath56 for all @xmath0 studied .
the estimated values of @xmath55 are also shown in this figure .
very remarkably , our simulations show that for all @xmath0 considered the quantity @xmath57 is in fact a constant .
since @xmath58 is the area of the critical semi - ellipse at percolation , then our results suggests that sinai s theorem @xcite , proposed originally for ip , is also valid for 2d anisotropic dp problems . in this respect
, we emphasize again that our length unit should be taken as @xmath59 for a system with site concentration @xmath10 .
thus , our simulations leads to the new invariance @xmath60 where @xmath10 is the site concentration ( _ e.g. _ , the donor concentration in doped semiconductors ) , @xmath61 is the area of the critical semi - ellipse and @xmath62 is the mean number of connected bonds per site at percolation .
the invariance of eq.([eq : inv ] ) may be somehow related to the fractal behavior of the critical clusters as we shall discuss below .
let us determine first the fractal dimension @xmath6 of the critical percolation cluster using a standard procedure based on the scaling expression @xcite @xmath63 in fig.2(b ) the quantities @xmath64 are plotted against @xmath65 for different values of the anisotropy parameter @xmath0 .
very surprisingly we found that the fractal dimensions @xmath6 , as determined from the slopes of the fitting lines for various values of @xmath0 in fig.2(b ) , seem indeed to be constant and independent of @xmath0 within our simulation errors .
we estimate @xmath66 for all @xmath0 , which corresponds to about the same value of the isotropic dp model with @xmath2 . at a first glance
this result might rise some doubts , but we believe it can be understood in connection with the invariant given in eq.([eq : inv ] ) .
the invariance of @xmath62 , with respect to changes in the anisotropy parameter @xmath0 , implies that the average number of connected bonds at percolation is independent of @xmath0 .
if we assume the percolation process within an elementary semi - ellipse ( as in fig.1 ) to be the ` originating percolation rule ' , then the invariance of eq.([eq : inv ] ) could mean that the law to generate percolation clusters remains unchanged as @xmath0 varies .
if this conjecture is right , we could suggest here a more general statement for all types of percolation models which are related to each other by the sinai theorem ; in these cases , the fractal dimensions of the percolation clusters could all be the same .
it should be noted that our simulations are limited to @xmath67 .
it is in this range that we observed the invariance of @xmath68 and the constant value for @xmath6 .
we believe that these features are maintained for a larger range of @xmath0 values .
however , it is not feasible to increase @xmath0 and the sample size @xmath8 simultaneously and get to the point where the present 2d simulation model crosses to the 1d case ( _ i.e. _ , @xmath69 ) . to conclude ,
we have suggested a model for anisotropic directed percolation ( adp ) and have presented the first simulation results for the main critical exponents of the model in 2d random systems .
quite surprisingly , we have found a new invariance for the average number of connected bonds at percolation due to presence of a suitable external force ( _ e.g. _ , shear stress , magnetic field , _ etc _ ) .
our simulations show that the product @xmath70 is a constant for all @xmath0 s considered .
this invariance should be in close relation to the value of @xmath6 .
we strongly believe the present model of adp could be important to describe some physical phenomena such as hopping conduction in anisotropic @xmath10-ge and @xmath10-si under strong electrical fields , where the impurity wave functions are anisotropic and the conduction band splits into one ellipsoid @xcite .
our measurements could be useful , for instance , in the expressions for the hopping resistivity in 2d anisotropic random media .
the new invariance @xmath71 could be used in these systems similarly to the invariance @xmath72 for ip in a circle problem @xcite .
we hope the present model will stimulate further investigations on this direction .
one of the authors ( n.v.l . ) would like to thank the condensed matter group at ictp , trieste , for financial support .
99 for a review , see d. stauffer , phys .
rep . * 54 * 1 ( 1979 ) ; physica a * 242 * 1 ( 1997 ) ; also j.w .
essam , rep .
prog . phys . * 43 * 833 ( 1980 ) .
i.g . enting and j. oitmaa , j. phys . a * 10 * , 1151 ( 1977 ) .
s. galam and a. mauger , phys .
e * 56 * , 322 ( 1997 ) . v. lien nguyen and a. rubio , solid state commun . * 95 * 833 ( 1995 ) .
levinstein , b.i .
shklovskii , m.s .
shur and a.l .
efros , sov .
jetp * 42 * 197 ( 1975 ) v. lien nguyen , b.i .
shklovskii and a.l .
efros , sov .
* 13 * 1281 ( 1979 ) .
ya.g . sinai ( _ theorem _ ) : `` if a surface @xmath73 can be obtained from the surface @xmath74 via a linear transformation of coordinates that involves both rotation and dilatation , then the values @xmath68 ( _ i.e. _ , the mean number of bonds per site ) for the two surfaces are identical '' , see @xcite page 116 .
b. mandelbrot in _ fractal geometry of nature _ , freeman ( san francisco 1982 ) b.i .
shklovskii and a.l .
efros in _ electronic properties of doped semiconductors _ , springer - verlag ( berlin , 1984 ) chapters 5 and 6 . | we propose a simulation model to study the properties of directed percolation in two - dimensional ( 2d ) anisotropic random media .
the degree of anisotropy in the model is given by the ratio @xmath0 between the axes of a semi - ellipse enclosing the bonds that promote percolation in one direction .
at percolation , this simple model shows that the average number of bonds per site in 2d is an invariant equal to 2.8 independently of @xmath0 .
this result suggests that sinai s theorem proposed originally for isotropic percolation is also valid for anisotropic directed percolation problems .
the new invariant also yields a constant fractal dimension @xmath1 for all @xmath0 , which is the same value found in isotropic directed percolation ( _ i.e. _ , @xmath2 ) . |
since defects exist in any real system , the understanding of their influence on ordering and structural phase transitions is important .
a signature of the presence of defects in a crystal near a phase transition is the so - called `` two length - scale '' behavior , in which , in the critical diffuse scattering ( cds ) of x - rays or neutrons , a narrow `` central peak '' is found on top of a broad peak @xcite .
previous theoretical studies of this behavior have established that one cause of this is the presence of dislocation lines @xcite .
these theories argue that the strain field associated with a dislocation line results in the growth of a roughly cylindrical ordered region near the dislocation line itself .
such regions order at a temperature higher than the defect - free crystal . accordingly , while the order occurs in the cylindrical regions , the broad peak in the cds is due to thermal fluctuations in regions of the material which are relatively unaffected by the strain field , while the narrow central peak is due to the fluctuations in regions where the enhanced ordering occurs .
unaccounted for in these theories , however , is the fact that in many real systems defects do not exist uniformly throughout the crystal
. often defects are caused by surface treatments or surface reconstructions and in this case they accumulate near the surface and their density decays with depth . when this happens , the ordering properties and two length - scale behavior depend on depth as well .
indeed , with high resolution x - ray diffraction measurements , we have previously found that v@xmath0h has two length - scale and associated behavior that is depth dependent @xcite .
these measurements were performed in both reflection and transmission geometries , allowing us to compare the behavior of the crystal at different depths . in this paper , we propose a theoretical explanation of these experimental results that accounts for the depth dependence of the observed behavior .
systematic studies of many materials in which two length - scale behavior has been found @xcite , including previous studies of v@xmath0h @xcite , have concluded that the narrow central peak of the cds only occurs in the scattering from a defective `` skin layer '' , that is a region of the material that starts a few hundred below the surface and extends several tens of @xmath1 m below the surface .
however , to the best of our knowledge , the two - length - scale behavior in v@xmath0h is different from that which has been observed in any other material , because in v@xmath0h the phase transition in the bulk is a first - order transition . in the skin layer , instead , the ordering is more complicated as found experimentally by a number of unusual phenomena including : ( 1 ) diffuse scattering which , as the temperature is lowered toward a critical value , consists of a broad peak that changes only slightly with temperature and a narrow central peak with an amplitude that diverges @xcite ; ( 2 ) an effective critical temperature @xmath2 for the behavior of the central peak that changes with the depth @xmath3 below the surface and extrapolates to a temperature @xmath4 that always exceeds the bulk transition temperature @xmath5 @xcite ; ( 3 ) a crossover in the universal critical behavior shown by the central peak from three - dimensional mean field critical behavior to a different universality class as the temperature increases from @xmath2 @xcite ; ( 4 ) a narrow two - phase region and a weak first - order transition observed at temperatures @xmath6 slightly below the critical value @xcite .
( color online ) schematic illustration of the arrangement of dislocation lines in v@xmath0h .
edge dislocations are arranged in walls of parallel lines that extend in the directions normal to the surface , whose density decreases with depth .
two colors are used for clarity of presentation to distinguish the lines extending in the two directions , but they do nt correspond to any physical difference.,scaledwidth=35.0% ] in order to explain these experimental findings we present a theory which accounts for the distribution of defects experimentally detected @xcite : edge dislocations occur mostly in the skin layer , accumulating near the surface ; they are arranged in arrays of parallel lines which we refer to as `` walls '' ; each wall consists of lines that are oriented in either of the two directions _ parallel _ to the surface ; the walls extend into the crystal and are thus oriented in either of the two directions _ perpendicular _ to the surface . in fig .
[ wallsketch ] we show a schematic of this arrangement of defects .
bulk measurements of the peak height @xmath7 of the @xmath8 superstructure reflection vs. @xmath9 , where @xmath5 is the transition temperature for the bulk
. the sudden jump of a few orders of magnitude in @xmath7 is a clear indication that the transition is first - order.,scaledwidth=45.0% ] a ) peak height @xmath7 of the @xmath8 superstructure reflection vs. reduced temperature @xmath10 .
the value of @xmath7 is proportional to @xmath11 , and thus shows the crossover of the critical exponent @xmath12 from a mean - field - compatible value of @xmath13 for small @xmath14 to @xmath15 for higher @xmath14 .
the measurements were carried out at a depth @xmath3 of @xmath16 m .
+ b ) inverse correlation length @xmath17 vs. reduced temperature @xmath10 , showing the crossover of the critical exponent @xmath18 from a mean - field - compatible value of @xmath19 for small @xmath14 to @xmath20 for higher @xmath14 .
the measurements were carried out at a depth @xmath3 of @xmath21m.,title="fig:",scaledwidth=45.0% ] a ) peak height @xmath7 of the @xmath8 superstructure reflection vs. reduced temperature @xmath10 .
the value of @xmath7 is proportional to @xmath11 , and thus shows the crossover of the critical exponent @xmath12 from a mean - field - compatible value of @xmath13 for small @xmath14 to @xmath15 for higher @xmath14 .
the measurements were carried out at a depth @xmath3 of @xmath16 m .
+ b ) inverse correlation length @xmath17 vs. reduced temperature @xmath10 , showing the crossover of the critical exponent @xmath18 from a mean - field - compatible value of @xmath19 for small @xmath14 to @xmath20 for higher @xmath14 .
the measurements were carried out at a depth @xmath3 of @xmath21m.,title="fig:",scaledwidth=45.0% ] ( color online ) intensities @xmath22 of the @xmath23 fundamental reflection in a range of temperatures of @xmath24 k around the weak first - order transition temperature @xmath25 in the skin - layer . from top to bottom , the measurements were taken at @xmath26 k , @xmath27 k , @xmath28 k and @xmath29 k. the dashed blue and dashed - dotted green lines show individual gaussian peak fits , while the red lines are the convolutions of the single peaks . at all the temperatures we see peaks at @xmath30 @xmath31 and @xmath32 @xmath31 ( indicated by arrows ) , corresponding to the @xmath33 and the @xmath34 phases , respectively .
the measurements were carried out at a depth of @xmath35m.,title="fig:",scaledwidth=36.0% ] ( color online ) intensities @xmath22 of the @xmath23 fundamental reflection in a range of temperatures of @xmath24 k around the weak first - order transition temperature @xmath25 in the skin - layer . from top to bottom , the measurements were taken at @xmath26 k ,
@xmath27 k , @xmath28 k and @xmath29 k. the dashed blue and dashed - dotted green lines show individual gaussian peak fits , while the red lines are the convolutions of the single peaks . at all the temperatures we see peaks at @xmath30 @xmath31 and @xmath32 @xmath31 ( indicated by arrows ) , corresponding to the @xmath33 and the @xmath34 phases , respectively .
the measurements were carried out at a depth of @xmath35m.,title="fig:",scaledwidth=36.0% ] ( color online ) intensities @xmath22 of the @xmath23 fundamental reflection in a range of temperatures of @xmath24 k around the weak first - order transition temperature @xmath25 in the skin - layer . from top to bottom , the measurements were taken at @xmath26 k ,
@xmath27 k , @xmath28 k and @xmath29 k. the dashed blue and dashed - dotted green lines show individual gaussian peak fits , while the red lines are the convolutions of the single peaks . at all the temperatures we see peaks at @xmath30 @xmath31 and @xmath32 @xmath31 ( indicated by arrows ) , corresponding to the @xmath33 and the @xmath34 phases , respectively .
the measurements were carried out at a depth of @xmath35m.,title="fig:",scaledwidth=36.0% ] ( color online ) intensities @xmath22 of the @xmath23 fundamental reflection in a range of temperatures of @xmath24 k around the weak first - order transition temperature @xmath25 in the skin - layer . from top to bottom , the measurements were taken at @xmath26 k , @xmath27 k , @xmath28 k and @xmath29 k. the dashed blue and dashed - dotted green lines show individual gaussian peak fits , while the red lines are the convolutions of the single peaks .
at all the temperatures we see peaks at @xmath30 @xmath31 and @xmath32 @xmath31 ( indicated by arrows ) , corresponding to the @xmath33 and the @xmath34 phases , respectively .
the measurements were carried out at a depth of @xmath35m.,title="fig:",scaledwidth=36.0% ] ( color online ) integrated intensities @xmath22 vs. absolute value of reduced temperature @xmath10 .
the integrated intensities , proportional to @xmath36 , allow us to estimate the long range critical exponent @xmath37 .
the black circles correspond to the @xmath8 superstructure reflection , depth @xmath3 of @xmath38 m ; the red squares to the @xmath39 superstructure reflection , depth @xmath3 of @xmath40 m .
the like - colored lines show the results of fits which gave values for the critical exponent @xmath37 of @xmath41 and @xmath42 , respectively , obtained neglecting a narrow ( @xmath24 k ) region of two - phase coexistence.,scaledwidth=45.0% ] as with previous theories , the strain due to dislocation lines enhances the ordering in their vicinity . as the temperature is lowered toward the critical temperature , ordering in the skin layer
first occurs in cylindrically shaped ordered regions ( ors ) near the individual dislocation lines .
however , in contrast to previous theories , which assumed the ors grow freely , in our theory the interaction between lines constrains their collective growth .
the ordering process responsible for the formation of the ors is continuous and this explains the change in the order of the phase transition from first - order in the bulk to continuous in the skin layer .
then , as the ors grow and occupy more than a certain fraction of the material , the crossover in the universal critical behavior of the central peak occurs .
since the density of dislocation lines decreases with depth , the effective critical temperature also varies with depth .
finally , as the temperature is reduced further , below the effective critical temperature at any particular depth , the parts of the material in between the network of ors undergo a weak first - order transition . before presenting our theory , in the next section we recount in some detail the unusual experimental facts of the two length - scale phenomena and associated behavior found in v@xmath0h .
as in prior works @xcite , we focus on the transition from the ordered monoclinic @xmath33 phase to the disordered body centered tetragonal @xmath34 phase in which the c - axis is along z ( for the phase diagram see ref . ) . in the @xmath33 phase ,
one half of the z - axis octahedral sites halfway between two v atoms , namely the @xmath43 sites , are mostly occupied by hydrogen atoms , while in the @xmath34 phase both @xmath43 and @xmath44 sites are , on average , equally occupied @xcite . as first reported in ref .
, in transmission geometry there is clear evidence of a first - order transition in the bulk material ( see fig .
[ firstorder ] ) . the diffuse scattering from the skin layer can be measured at various depths in reflection geometry by varying beam energy and reflection order . at temperatures above the bulk transition temperature ,
what we find is a broad peak corresponding to the bulk diffuse scattering , which increases slowly with decreasing temperature , coexisting with a central peak whose amplitude diverges at a temperature still higher than the bulk transition one @xcite .
the cds of the central peak indicates that the transition is continuous in a skin layer that has a depth of several tens of @xmath1 m .
remarkably , the temperature at which the cds diverges depends on the depth probed .
thus , we find a depth dependent critical temperature @xmath2 , where @xmath3 is the effective scattering depth @xcite .
furthermore , as shown in ref . , there is a two - length scale effect in the cds from the skin layer .
as the temperature is lowered toward @xmath2 , the height of the central peak , which is proportional to the susceptibility , shows a crossover in critical scaling behavior .
in fact , for temperatures close to the critical temperature , the value of the critical exponent @xmath12 in the law @xmath45 , describing the divergence of the susceptibility with the reduced temperature @xmath46 , is always consistent with 1 ; similarly , the full width at half maximum of the central peak , which is proportional to the inverse correlation length @xmath17 , and which scales as @xmath47 , has its exponent always compatible with 0.5 ( see two examples in fig . [ crosso ] ) .
also , we notice that , regardless of the depth , the value of @xmath12 is always compatible with @xmath48 .
note that these values of @xmath12 and @xmath18 , measured for small @xmath14 , correspond to three - dimensional mean field values .
we have considered the integrated intensities @xmath22 of the superstructure reflections , which are proportional to @xmath36 .
we find that the intensities exhibit power - law variations for temperatures below the critical value , but only if one neglects a narrow temperature range of @xmath24 k close to the critical point , which corresponds to a two - phase coexistence region ( see figs . [ 2phase ] and [ beta ] ) .
the existence of this region indicates that , even in the skin - layer , there is an actual first - order transition , albeit with a weak character , as indicated by the power law behavior outside the aforementioned small range of temperatures .
the measured value of @xmath49 , however , is not mean field but it is in agreement with an earlier measurement on a different crystal @xcite . above the crossover , at larger @xmath14 ,
different universal critical behavior is observed .
measurements indicating values of @xmath50 and @xmath51 have previously been reported @xcite .
however , other data fits at these larger values of @xmath14 give a range of values for @xmath12 and @xmath18 .
for example , in fig .
[ crosso ] values of @xmath52 and @xmath53 are found .
thus , unfortunately , the existing experimental data sets do nt allow us to determine accurately the critical exponents for temperatures higher than the crossover .
the available data show values for @xmath12 as low as about 1.8 and as high as 3 and values of @xmath18 from about @xmath54 to 2 . regarding the presence of defects in the material , dislocation walls are the only kind of defect that extends into the crystal for several @xmath1 m .
other kinds of defects are not found after the first 150 , and the influence of this upper region on the scattering is negligible @xcite .
the arrangement of dislocation lines into walls , in the skin layer , is indicated by the mosaic spread .
walls occur inhomogeneously across the surface , so that in planes parallel to the surface there is an inhomogeneous distribution of dislocation lines @xcite .
in order to give a picture that describes the unusual experimental behavior of v@xmath0h detailed above , let us consider a ginzburg - landau free energy density expansion for a crystal with an anisotropic distribution of dislocation lines .
this free energy density is of the form @xmath55 where @xmath56 is the order parameter field , and @xmath57 , @xmath58 and @xmath59 are coefficients that depend on position @xmath60 , thermodynamic parameters , and , in some cases , the strain fields created by dislocations and structural ordering . in particular , the order parameter field is defined to be 0 for regions of the material which are in the disordered phase , 1 for points in the ordered phase with the hydrogen atoms in the @xmath43 sites , and @xmath61 for the ordered phase with the hydrogen atoms in the @xmath44 sites .
also note that in the above equation the odd powers are missing as their presence is disallowed by the symmetry of the crystal structures .
much of the unusual phenomenological behavior observed in v@xmath0h can be explained through the behavior of the coefficients @xmath57 and @xmath58 , which depends on the strain fields caused by the edge dislocations and on the structural ordering that occurs preferentially near them . in order to provide such explanation
, we must not only understand the mechanism by which strain fields modify @xmath57 and @xmath58 in general , but also what specific modifications result from the particular morphological arrangement of dislocation lines that occurs in v@xmath0h .
we will see that while the behavior of @xmath57 can explain the spatial variation of the critical temperature , the corrections to the fourth - order term justify instead the change in order of the transition .
in particular , while a ginzburg - landau expansion yields critical behavior for vanishing odd - order terms , we can still describe a first - order transition by letting @xmath58 go negative . in the remainder of this section
, we will first discuss the effects of the defects on @xmath57 , explaining the behavior of the critical temperature .
then , we will describe the shape of the ordered regions near the dislocation lines .
next , we will show how the ordering strains affect @xmath58 , how this makes the transition in the skin - layer continuous , and how the crossover occurs .
finally , we will briefly discuss the effects of the depth - dependent crossover temperature on the size of the critical region and give an argument for the presence of a weak first - order transition in the skin - layer .
the compression - dilatation effect of an edge dislocation on a crystal is responsible for a change in critical temperature @xcite .
we can assume that the atoms of the crystal interact more strongly where they are pushed closer to each other , and vice versa that they interact more weakly in the opposite direction .
as a first approximation we can assume that the correction to the critical temperature is proportional to the elastic strain in the crystal .
the elastic strain can be found following the general procedure described in ref . .
in the case of a single dislocation line this yields a trace of the stress tensor @xmath62 that in polar coordinates @xmath60 and @xmath63 centered on the dislocation line is @xmath64 where @xmath1 is the shear modulus , @xmath18 is poisson s ratio and @xmath65 is the magnitude of the burgers vector . making use of the relations between elastic constants @xcite , we can rewrite eq .
[ polartrace ] as @xmath66 where @xmath67 is the bulk modulus . the above equation , divided by @xmath68 , yields the strain , which , in our treatment , is proportional to the local relative critical temperature change @xmath69 @xmath70 where @xmath71 is the new critical temperature , and which is defined as @xmath72 with @xmath73 the transition temperature for an undifected crystal . notice that this result is in agreement with ref . .
in the case of multiple dislocations , the effects of the single dislocation lines get superimposed . for walls of dislocations ,
this allows us to estimate the above quantities quite easily . without loss of generality
, we will consider a wall that extends in the @xmath3 direction while the lines are parallel to @xmath74 ; we will indicate by @xmath75 the local inverse linear density of defects , i.e. , the local average distance between two lines .
we rewrite then eq .
[ newredloctc ] in cartesian coordinates as @xmath76 where we used @xmath77 and @xmath78 .
if @xmath75 changes smoothly along the wall , we can then write @xmath79 where @xmath60 is the radial distance outwards from the closest dislocation line and the dependence of @xmath7 on @xmath3 has been omitted for simplicity of writing .
the sum of the series above is @xmath80 where @xmath81 is the unit of length used .
it should be noticed that the contribution decays exponentially with the normal distance @xmath82 from the wall , in agreement with the experimental results in ref . .
also , the rapid convergence of eq . [ series ] suggests that @xmath7 needs to vary slowly only over a short distance , in order for the error made by considering it constant to be small . substituting eq .
[ series ] in eq .
[ walls ] , we can then conclude that , in the case of dislocation walls , the local relative critical temperature change @xmath83 now , an expression can be found for the behavior of the critical temperature with depth . in first approximation ,
@xmath84 which can be rewritten as @xmath85 where @xmath86 note that @xmath87 can be computed from experimental data , such as the ones reported in fig . 5 of ref . , while the other derivative can be computed from eq
. [ critempwall ] , yielding the following expression for @xmath88 : @xmath89}{{\left\lbrace x\left[\cosh\left(\frac{2\pi x}{h}\right)-\cos\left(\frac{2\pi y}{h}\right)\right]\right\rbrace}^2}-\frac{y\cos\left(\frac{2\pi y}{h}\right)}{h\left[\cosh\left(\frac{2\pi x}{h}\right)-\cosh\left(\frac{2\pi y}{h}\right)\right]}\right)\:.\ ] ] notice that using typical values for the elastic constants , the proportionality factors in front of the functions in the above equations are of the order of unity . as for @xmath57 , the second order coefficient in the free energy density expansion ,
we recall that without defects it is @xmath90 , with @xmath91 generic proportionality constant . from this expression and the definition of @xmath69
, we can write the new second order coefficient of the expansion as @xmath92 the spatial dependence of the parameter is contained in the local relative critical temperature change , that we wrote here as @xmath69 for sake of simplicity , but which is actually @xmath93 . thus , in contrast to @xmath57 , @xmath94 varies in space .
furthermore , summing the contributions coming from single dislocations or dislocation walls , one gets a field of relative critical temperature changes , which features in the free energy density expansion as follows : @xmath95 \rbrace \eta^2+a_4\eta^4+a_6\eta^6\:.\ ] ] in this treatment the strain field surrounding a dislocation line is spatially inhomogeneous , and this leads to a spatial inhomogeneity of the ordering near a dislocation line .
in fact , the strain field caused by a line is dipole - like . because of this , at temperatures above the transition temperature of an undefected crystal , ordered regions exist below the dislocation lines , where the transition temperature is increased .
on the other hand , the transition temperature on the other side of the line is symmetrically decreased .
thus , disordered regions exist within a dislocation wall even at a temperature @xmath96 below the transition temperature of an ideal crystal . then
, at any depth , the correlation length along @xmath3 can not exceed a value of the order of @xmath97 ; conversely , the local ordering temperature can only change when the density of defects changes , i.e. , over a distance of at least a few @xmath7 . to find the shape of the ordered regions , we first fix a value for the relative critical temperature change @xmath69 .
then , we solve eq .
[ critempwall ] in order to find expressions relating @xmath82 and @xmath3 as functions of each other .
these solutions correspond to the border of the ordered region for the chosen value of @xmath69 , and , up to a multiplicative factor that is of order of unity for typical values of the elastic constants , they are @xmath98}}{\tau_c\left[\cosh\left(\frac{2\pi x}{h}\right)+1\right]}\right\rbrace\:;\\ x\left(y\right)=\pm\frac{h}{2\pi}\operatorname{arccosh}\left\lbrace\frac{2\pi\tan\left(\frac{\pi y}{h}\right)+\tau_c\left[1-\tan^2\left(\frac{\pi y}{h}\right)\right]}{\tau_c\left[1+\tan^2\left(\frac{\pi y}{h}\right)\right]}\right\rbrace\:.\label{xofy}\end{gathered}\ ] ] note that the regions are symmetric with respect to the @xmath82 and @xmath3 directions . from eq .
[ yofx ] it follows that the maximum extension of the regions along the @xmath3 direction , @xmath99 , happens at @xmath100 , and its value is @xmath101 notice that @xmath102 when @xmath103 , that is , the regions can never extend in @xmath3 for a distance greater than @xmath104 , as already stated in the previous section . from eq .
[ xofy ] , it follows that the maximum estension along the @xmath82 direction , @xmath105 , happens at @xmath106 , and its value is @xmath107+\tau_c\left\lbrace1-\tan^2\left[\frac{1}{2}\arctan\left(\frac{\pi}{\tau_c}\right)\right]\right\rbrace}{\tau_c\left\lbrace1+\tan^2\left[\frac{1}{2}\arctan\left(\frac{\pi}{\tau_c}\right)\right]\right\rbrace}\right)\:,\ ] ] that can be rewritten as @xmath108 note that the center of the region is at @xmath109 .
now it is possible to show that the ordered regions grow from a rod - shaped nucleus with a circular cross - section .
with growth , their aspect ratio changes , making the cross - section effectively elliptical . finally , the constraint constituted by the order - inhibited zones above the dislocation lines introduces distortions in the shape . to show this
, we prove that , for large values of @xmath69 , corresponding to the beginning of the growth of the regions , @xmath110 , where @xmath111 , @xmath112 and @xmath113 is the @xmath3 coordinate measured from the center of the region . in this regime
, we can expand @xmath65 as follows : @xmath114 then , we use a similar treatment for @xmath91 .
first we expand the argument of the @xmath115 for large values of @xmath69 : @xmath116\:.\ ] ] then , we expand the @xmath115 for values of the argument close to 1 , getting @xmath117 expanding the argument of the @xmath118 in eq .
[ yofx ] for large @xmath69 we get @xmath119\:,\ ] ] where we have imposed the condition that the argument be real .
note that a big @xmath69 implies a small @xmath82 .
then , it is @xmath120 shape of the cross section of the ordered regions for different values of @xmath69 . from the outermost line inwards , the lines correspond to values of @xmath69 that range from @xmath121 to 1 , in steps of @xmath121 , except for the innermost line that corresponds to @xmath122.,scaledwidth=45.0% ] also note that in the beginning of the growth @xmath123 .
this means that the regions nucleate in the shape of narrow cylindrical rods along the dislocation lines .
figure [ cylreg ] shows the shape of the cross section of the ordered regions for different values of @xmath69 . substituting eq .
[ nucy ] into eq .
[ xofy ] , and expanding for high values of @xmath69 we find : @xmath124 finally , we have @xmath125 that is , for small sizes , the regions are effectively elliptical , and @xmath91 and @xmath65 can be identified with the major and minor semiaxes , respectively .
( color online ) schematic illustration of the shape and size of the ordered regions in a wall at a fixed temperature .
the surface of the crystal is at the top of the image .
the crystal extends beyond what pictured , that is just a region in the vicinity of a single wall .
while in the skin - layer the regions have an elliptical cross - section whose aspect ratio is constant , at greater depths their shape becomes increasingly circular . also , while the size of the regions increases with depth , their number density decreases.,scaledwidth=30.0% ] to understand the relative shape and size of the ordered regions in a wall at a particular fixed temperature , first notice that from the definition of @xmath69 ( eq . [ tauc ] ) it follows that any given temperature @xmath96 is critical for points in the crystal that satisfy the equation @xmath126 that is to say , for points in which @xmath127 .
then , from eqs . [ minor2 ] and [ major2 ] , for a fixed value of @xmath69 the major and minor semiaxes of the ordered regions grow linearly with @xmath7
consequently , for a fixed temperature , the density of defects decreasing with depth makes the size of the ordered regions grow with depth .
their aspect ratio , in contrast , remains the same .
of course , this argument is valid as long as the approximation employed to obtain eq .
[ critempwall ] using eq .
[ series ] holds .
deep in the skin - layer , where the density of defects almost vanishes , the strain field associated with any ordered region is restricted to that if a single dislocation line .
therefore , at great depths , the expression for @xmath69 tends to the form given in eq .
[ taucart ] . solving eq .
[ taucart ] yields @xmath128 we can then calculate @xmath99 and @xmath105 as above : @xmath129 thus , as the depth increases , the shape of the cross - section of the regions becomes increasingly circular ( see fig .
[ dislwall ] ) . to determine how the defects affect the fourth - order term in the free energy density expansion , we start with finding the displacement field due to a point source of expansion by means of the tensor green s function for the equilibrium equation . the displacement , which is purely radial , is @xmath130 where @xmath131 is the work associated with this expansion , @xmath132 is lam s constant , and @xmath1 is the shear modulus @xcite .
then , consider the particular case of a local expansion due to the difference of structure between 2 points .
given our order parameter field @xmath133 , this expansion , purely hydrostatical since it s applied to an infinitesimal volume , can be thought of as due to an equivalent pressure @xmath134 = k\frac{\delta v}{v}\eta\left ( \vec{r}\right)\:,\ ] ] where @xmath67 is the bulk modulus . since we are associating a change of volume with a point , what we want to consider is the smallest volume for which it makes sense to think about a change of structure , that is , the volume of a unit cell .
so , calling @xmath135 the volume of a unit cell in the ordered phase and @xmath136 the one in the disordered phase , we have @xmath137 , hence @xmath138 the work done to achieve this deformation is @xmath139 and we can use this expression in eq .
[ raddispl ] to find the contribution to the displacement at the point @xmath140 due to a point at @xmath141 , which is @xmath142 ^2}{v_d\left ( \vec{r'}\right ) } \frac{\eta\left ( \vec{r}\right ) } { r^2}\:.\ ] ] notice that we have made the dependence of @xmath136 on the point explicit .
for any particular @xmath140 , @xmath143 vanishes in the limit of @xmath144 . then , knowing the field in every point , the total displacement at @xmath140 is @xmath145 ^2}{v_d\left ( \vec{r'}\right ) } \frac{\vec{r'}}{{r'}^3}\ : , \end{split}\ ] ] where the integral is extended over all the points in the disordered phase . strictly speaking , eq .
[ pressure ] is valid when the transition is from the disordered to the ordered phase . for the inverse transition
, @xmath146 has the opposite sign and the term in parentheses in eq .
[ pressure ] is @xmath147 . yet , since @xmath148 , the error committed in using the same formula is very small ; in fact , in the subsequent equations the volume difference ends up squared , so that the error is actually of second order and is neglected in this linear treatment .
notice also that , after integration , the displacement is not necessarily purely hydrostatic . from the displacement vector we can find the components of the strain tensor in cartesian coordinates as @xmath149 then , from the generalized hooke s law , we can find the correction to the free energy density due to the deformation induced by the phase transition , @xmath150 , as @xmath151 it s to be stressed that this quadratic form is always positive , since @xmath67 and @xmath1 are always positive .
in fact , since the body is in equilibrium , in the absence of forces the energy must have a minimum at @xmath152 , that is , the quadratic form must be positive .
since this has to happen in any case , it also has to happen when the stress is a pure shear or a pure hydrostatic compression , that is when only one of the two addenda is non - zero .
this is the reason why both @xmath67 and @xmath1 must always be positive @xcite .
moreover the squares make this correction actually proportional to @xmath153 .
then , when it is applied to the @xmath154 order coefficient in the free energy density expansion , where it belongs since its functional form is that of a strain , it yields a positive effective correction to the @xmath155 order term .
this correction , which corresponds exactly to the elastic free energy , does not depend on the actual structure of defects that causes it , but merely on the order parameter field and , of course , on the elastic constants of the material .
this means that , regardless if it s coming from a single dislocation , a wall of dislocations , or a more complicated framework of defects , once the elements of the strain tensor are known , they can be used to correct directly the free energy density expansion as discussed and as follows : @xmath156 \rbrace \eta^2\\ & \quad+\left [ a_4+\frac{b}{\eta^2}\right ] \eta^4+a_6\eta^6\ : , \end{split}\ ] ] where @xmath69 is the local relative critical temperature change resulting from the superposition of the contributions of single dislocation lines ( eq . [ newredloctc ] ) and dislocation walls ( eq . [ critempwall ] ) .
while no quantitative solution of the above equations is offered here , notice that an explicit calculation would be very difficult .
in fact , even in simpler cases of lattices in only 2 dimensions featuring regularly arranged point defects with short range interactions , the results can be highly non - trivial @xcite . we know , however , that , from a purely phenomenological point of view , the change of order in a phase transition can be associated with the change of the sign of @xmath58 , which determines whether the transition is continuous or discontinuous .
the fact that the correction to this coefficient is always positive , shows how an originally negative @xmath58 can turn positive , thus resulting in a first - order transition becoming continuous in the presence of defects , as observed in v@xmath0h .
note that a correction to @xmath58 due to the ordering process exists also in an undefected material .
however , at least in the beginning of the nucleation of the new phase in the disordered material , this correction is not greater than the one in the skin - layer discussed above@xcite .
in fact , when the ordered phase starts to nucleate in a perfect crystal , the shape of the nucleus is that of an infinitely thin platelet @xcite .
the strain energy density contribution of the infinitesimal platelet is @xmath157 where @xmath158 is a linear function of the strains and @xmath159 is the unitary vector minimizing @xmath158 @xcite .
as before , the strains being proportional to @xmath153 , the equation above yields an effective correction on the fourth - order term . on the other hand , we have seen that the shape of the nucleus in the skin - layer is that of an infinitesimal cylindrical rod . then , eq . [ a4b ] becomes @xmath160 where the brackets indicate the angular average of @xmath158 .
since the average of a quantity is never smaller than its minimum , the correction in the skin layer due to an infinitesimal nucleus can not be smaller than the one in the bulk .
( color online ) schematic illustration of the ordering process near dislocation lines at an arbitrary depth .
red and light blue regions of the ordered regions distinguish different ordering domains , corresponding to the two different signs of the order parameter . at higher temperatures , shown in a )
, the regions do nt touch each other , the correlation length ( average length of an ordered domain ) is smaller than the distance between dislocation walls , and there is little or no correlation in the ordering in different regions . at lower temperatures , shown in b ) ,
the ordered regions are thicker and now touch , the correlation length is longer than the average distance between dislocation walls , and the order is spreading throughout the network of regions.,title="fig : " ] ( color online ) schematic illustration of the ordering process near dislocation lines at an arbitrary depth .
red and light blue regions of the ordered regions distinguish different ordering domains , corresponding to the two different signs of the order parameter . at higher temperatures , shown in a )
, the regions do nt touch each other , the correlation length ( average length of an ordered domain ) is smaller than the distance between dislocation walls , and there is little or no correlation in the ordering in different regions . at lower temperatures , shown in b ) ,
the ordered regions are thicker and now touch , the correlation length is longer than the average distance between dislocation walls , and the order is spreading throughout the network of regions.,title="fig : " ] recalling eq . [ totaldispl ] , we notice that the more the order spreads , the more the domain over which the intergral is carried out is reduced .
moreover , the contributions in the integral are weighted with the square of the distance to the point to which they apply .
thus , with growing regions , the magnitude of the displacement vector will steadily decrease .
also , the displacement vector field will become spatially increasingly homogeneous , especially in points not too close to the borders .
therefore , the components of the strain tensor will decrease as well , since we know from eq .
[ displtens ] that they depend on the derivatives of the total displacement vector components , which are changing smoothly .
the fact that the correction to @xmath58 is proportional to the square of a linear combination of strain tensor components ( eq . [ a4corr ] ) guarantees that its value will be small and positive throughout the new transition .
if this is added to a small negative `` original '' fourth order coefficient , as it is the case in v@xmath0h , whose bulk transition is first order , it will make @xmath58 effectively vanish , thereby making the transition behavior tricritical , and , because the upper critical dimension of this tricritical theory is 3 , causing the mean - field values of the exponents that are observed experimentally .
note that below a certain temperature where the wall lines have ordered independently , the walls consist of domains with order parameter of opposite signs . in the case of a complex framework of dislocation walls which intersect and interact with each other , on further lowering the temperature ,
one of the two signs will eventually spread throughout the network until the order parameter has the same sign everywhere ( see also fig .
[ cylreg ] ) .
the mechanism for tricriticality also provides an explanation for the depth - dependence of the crossover temperature .
in fact , the density of defects decreases with depth .
thus , in order for the domain of the integral in eq . [ totaldispl ] to be sufficiently reduced , the temperature needs to be decreased more at greater depths .
consequently , the correlation length @xmath161 has to increase more at greater depths than it has to in layers closer to the surface in order to reach the critical point .
in other words , for the same change in reduced temperature , the change in correlation length is higher for shallower depths .
the corresponding scaling law should then be modified so that the coefficient depend on depth : @xmath162 in particular , it is reasonable to expect @xmath163 to be proportional to the density of defects , as it is in fact observed experimentally @xcite .
this mechanism provides an explanation as to why experimental measurements of @xmath161 at different depths yield different values even if taken at corresponding reduced temperatures , once the depth - dependence of the critical temperature has been accounted for , and even though the critical exponent remains the same .
a treatment to collapse the data onto a single curve , including corrections for the experimental method used , has been proposed and verified in ref . .
as the temperature is lowered below the critical point , the remaining material between the walls orders and a second , distinct , phase transition occurs .
however , considering eq .
[ totaldispl ] again , we notice that the correction only applies to the material inside the `` skeleton '' of ors .
therefore , for the material that is still between the ors , the fourth - order coefficient is still negative . in such regions
the transition is still first - order , but it is weaker than it is in an undefected crystal . because this transition is only weakly first - order , a critical exponent @xmath37 associated with it can still be measured as shown in fig . [ beta ] .
the measured values are @xmath41 and @xmath42 .
such values confirm that the transition regime is not tricritical , since in that case one would expect a mean field exponent @xmath164 @xcite .
in conclusion , we have proposed a theoretical model which explains the two - length - scale phenomena and related behavior observed experimentally in many materials , including the unusual ordering behavior observed in v@xmath0h . in particular , the depth - dependence of the critical temperature in the skin layer , reported recently @xcite , is shown to be caused by the strain field induced by the presence of walls of dislocations .
the ordering process itself crosses over between two regimes , causing the experimentally observed crossover in the values of the critical exponents .
the additional strains induced by the formation of the ordered regions are responsible for increasing the effective value of the fourth - order coefficient in a ginzburg - landau free energy density expansion , thereby allowing the transition to become continuous in the skin layer even if it is first - order in the bulk as we find in v@xmath0h .
furthermore , the strength of this correction weakens during the spreading of the order through the network , driving the value of the fourth - order coefficient to zero , thus producing tricritical behavior and mean - field values of the critical exponents .
thus , when the temperature is lowered , first a continuous transition happens along the dislocations as described .
then , at a still lower temperature , the material between the cylindrical regions orders . however , it undergoes a transition that is first order , similar to the one that takes place in the bulk but weaker .
the model also explains why , in the mean - field regime , the critical region has a depth - dependent size .
although in this paper we have specifically considered the case of v@xmath0h , the theoretical framework that we have developed should be broadly applicable for understanding ordering behavior in defective materials , particularly those that have an anisotropic distribution of defects .
we considered a situation in which the bulk , and the material in between the interconnected network of cylindrical regions in the skin layer , orders through first - order transitions .
however , the model we have developed is easily adaptable to a situation where the transition in between the cylindrical regions is continuous , and to a situation where both transitions are continuous . in the latter case , though , we would not expect to see a crossover in the scaling behavior of the central peak of the skin layer cds .
this work is supported by the nsf through grants # dmr-0406323 and # dmr-0908286 ( keb and cidg ) .
keb , cidg and scm also acknowledge support by the texas center for superconductivity at the university of houston ( t@xmath165suh ) .
rib is supported by the division of materials science and engineering , office of basic energy science , u.s .
department of energy .
we gratefully acknowledge m. e. fisher for his valuable comments and discussions .
10 i. m. dubrovski and m. a. krivoglaz , zh .
jetp * 77 * , 1017 ( 1979 ) .
a. l. korzhenevskii , k. herrmanns and h .- o .
heuer , europhys .
* 45 * , 195 ( 1999 ) .
j. trenkler , r. barabash , h. dosch and s. c. moss , phys .
b * 64 * ( 2001 ) 214101 .
j. trenkler , p. c. chow , p. wochner , h. abe , k. e. bassler , r. paniago , h. reichert , d. scarfe , t. h. metzger , j. peisl , j. bai , s. c. moss , phys .
lett . * 81 * , 2276 ( 1998 ) . c. i. del genio , j. trenkler , k. e. bassler , p. wochner , d. r. haeffner , g. f. reiter , j. bai and s. c. moss , phys .
b * 79 * ( 2009 ) 184113 .
s. r. andrews , j. phys .
c * 19 * , 3721 ( 1986 ) .
r. a. cowley , physica scripta * t66 * , 24 ( 1996 ) .
m. gehring , k. hirota , c. f. majkrzak , g. shirane , phys .
71 * , 1087 ( 1993 ) .
m. gehring , a. vigilante , d. f. mcmorrow , d. gibbs , c. f. majkrzak , g. helgesen , r. a. cowley , r. c. c. ward , m. r. wells , physica b * 221 * , 398 ( 1996 ) . k. hirota , g. shirane , p. m. gehring , c. f. majkrzak , phys .
b * 49 * , 11967 ( 1994 ) .
k. hirota , j. p. hill , s. m. shapiro , g. shirane , y. fujii , phys .
b * 52 * , 13195 ( 1995 ) .
d. f. mcmorrow , n. hamaya , s. shimomura , y. fujii , s. kishimoto , h. iwasaki , solid state commun .
* 76 * , 443 ( 1990 ) .
neumann , u. rtt , j. r. schneider , g. shirane , phys . rev .
b * 52 * , 3981 ( 1995 ) .
u. rtt , a. diederichs , j. r. schneider , g. shirane , europhysics letters * 39 * , 395 ( 1997 ) .
t. r. thurston , g. helgesen , d. gibbs , j. p. hill , b. d. gaulin , g. shirane , phys .
* 70 * , 3151 ( 1993 ) .
r. h. wang , y. m. zhu , s. m. shapiro , phys .
lett . * 80 * , 2370 ( 1998 ) .
j. trenkler , s. c. moss , h. reichert , r. paniago , u. gebhardt , h. d. carstanjen , t. h. metzger , j. peisl , in _ proceedings of the international advanced studies institute conference on exploration of subsurface phenomena by particle scattering ( asi-002 ) , monterey , 1998 _ , edited by n. q. lam , c. a. melendres and s. k. sinha ( iasi press , north east , md , 2000 ) , p. 155 .
j. trenkler , h. abe , p. wochner , d. haeffner , j. bai , h. d. carstanjen , s. c. moss , modelling simul .
* 8 * , 269 ( 2000 ) .
t. schober and w. pesch , zeit .
- wiesbaden * 114 * , 21 ( 1979 ) .
s. c. moss , in _ electronic structure and properties of hydrogen in metals _ , edited by p. jena and c. b. sattertwaite ( plenum , new york , 1983 ) .
y. fukai , _ the metal - hydrogen system _ ( springer - verlag , berlin , 1993 ) vol 21
. b. schnfeld , s. c. moss and k. kjaer , phys .
b * 36 * , 5466 ( 1987 ) .
j. p. hirth and j. lothe , _ theory of dislocations _
( mcgraw - hill , new york , 1978 ) .
l. d. landau and e. m. lifshitz , _ theory of elasticity _
( pergamon press , london , 1959 ) .
m. e. fisher and h. au - yang , j. phys .
c : solid state phys .
* 8 * , l418 ( 1975 ) .
a. d. bruce and r. a. cowley , _ structural phase transitions _ ( taylor & francis ltd . , london , 1981 ) .
a. g. khachaturyan , _ theory of structural transformations in solids _
( dover publications , inc . ,
mineola , 1983 ) .
d. r. nelson and m. e. fisher , phys .
b * 11 * , 1030 ( 1975 ) . | structural defects in a crystal are responsible for the `` two length - scale '' behavior , in which a sharp central peak is superimposed over a broad peak in critical diffuse x - ray scattering . we have previously measured the scaling behavior of the central peak by scattering from a near - surface region of a v@xmath0h crystal , which has a first - order transition in the bulk . as the temperature is lowered toward the critical temperature , a crossover in critical behavior is seen , with the temperature range nearest to the critical point being characterized by mean field exponents . near the transition , a small two - phase coexistence region
is observed .
the values of transition and crossover temperatures decay with depth .
an explanation of these experimental results is here proposed by means of a theory in which edge dislocations in the near - surface region occur in walls oriented in the two directions normal to the surface .
the strain caused by the dislocation lines causes the ordering in the crystal to occur as growth of roughly cylindrically shaped regions .
after the regions have reached a certain size , the crossover in the critical behavior occurs , and mean field behavior prevails . at a still lower temperature , the rest of the material between the cylindrical regions orders via a weak first - order transition . |
it is well known that under some circumstances the phenomenology of atomic liquids finds an almost perfect correspondence in the phenomenology of colloidal fluids @xcite .
this seems to be particularly true regarding the rather complex dynamic behavior of these systems as they approach the glass transition @xcite . although it is clear that this analogy has some fundamental limitations ( such as the presence of many - body hydrodynamic interactions in colloidal systems ) , one can be confident , for example , that the phase behavior of colloidal and atomic systems with identical interaction potentials will , of course , be the same . thus
, if one approaches this problem with a dynamic simulation technique , one is confident that the equilibrium phase diagram of a specific system ( say a lennard - jones liquid ) will be independent of the simulation technique employed in its determination ( either molecular or brownian dynamics ) @xcite .
time - dependent and dynamic properties , on the other hand , are expected in general to depend on the specific microscopic transport mechanisms .
nevertheless , some features associated with the collective , long - time behavior of the system also seem to be rather insensitive to the microscopic short - time dynamics .
for example , it has been suspected , and partially corroborated , that for a given model system ( i.e. , same pair potential ) standard molecular dynamics will lead to essentially the same dynamic arrest scenario as brownian dynamics @xcite .
determining the range of validity of this dynamic analogy , however , continues to be a relevant topic in the study of the dynamics of liquids . from the theoretical side , for example , one would like to have a unified description of the macroscopic dynamics of both , colloidal and atomic liquids , which explicitly exhibits the origin of the similarities and differences in their macroscopic dynamics .
this topic has been addressed in the framework of the mode coupling theory of the ideal glass transition @xcite , originally developed for newtonian liquids , but also adapted to brownian systems .
such attention was focussed on the similarity of the long - time dynamics of newtonian and brownian systems in the neighborhood of the glass transition @xcite .
a number of issues , however , remain open @xcite , one important question referring to the validity and limitations of this long - time similarity under general conditions , such as those involving ordinary thermodynamically stable fluids , and not necessarily associated with the glass transition . in this sense ,
one possible general framework for such theoretical analysis is the concept of the generalized langevin equation ( gle ) @xcite .
the gle formalism describes the dynamics of the thermal fluctuations @xmath1 of the instantaneous value of the macroscopic variables @xmath2 ( @xmath3 ) , around its equilibrium value @xmath4 , and has the structure of the most general linear stochastic equation with additive noise for the vector @xmath5^{\dagger } $ ] ( with the dagger indicating transpose ) .
the gle equation has been widely used in the description of thermal fluctuation phenomena in simple liquid systems , and boon and yip s textbook @xcite contains a detailed account of its early use to describe the dynamics of simple liquids .
although this stochastic equation is conventionally associated with the mori - zwanzig projection operator formalism @xcite , in reality its structure is not a consequence of the hamiltonian basis of mori - zwanzig s derivation ; instead , it is essentially equivalent to the mathematical condition of stationarity @xcite
. understood in the latter manner , the gle formalism was first employed in ref .
@xcite to derive the equation of motion of an individual tracer particle in a colloidal suspension without hydrodynamic interactions .
such an equation reads @xmath6 where @xmath7 is the mass and @xmath8 the velocity of the tracer particle , while @xmath9 is the friction coefficient caused by the frictional resistance of the supporting solvent and @xmath10 the associated random force .
the memory term involving the time - dependent friction function @xmath11 , and its associated random force @xmath12 , are the friction and fluctuating forces that originate in the time - evolution of the cage of surrounding colloidal particles . under well defined approximations , the exact result for the time - dependent friction function @xmath13 derived in ref .
@xcite was shown there to reduce to the following approximate expression in terms of the collective and self intermediate scattering functions ( isfs ) @xmath14 and @xmath15 , @xmath16}{s(k)}\right ] ^{2}f(k , t)f_s(k , t ) .
\label{dzdt0}\ ] ] in this equation @xmath17 is the temperature , @xmath18 the number concentration , and @xmath19 the static structure factor of the bulk suspension .
this result , together with similarly general expressions for @xmath14 and @xmath15 also derived within the gle formalism @xcite , was later employed in the construction of the self - consistent generalized langevin equation ( scgle ) theory of colloid dynamics @xcite , eventually applied to the description of dynamic arrest phenomena @xcite , and more recently , to the construction of a first - principles theory of equilibration and aging of colloidal glass - forming liquids @xcite . with the aim of investigating the relationship between the dynamics of atomic and brownian liquids , in this work
we apply the gle formalism to derive the generalized langevin equation that describes the motion of individual tracer particles in simple _ atomic _ liquids , thus extending to these systems the results of ref .
@xcite reviewed above .
the most remarkable prediction of the derivation presented here is that the resulting stochastic equation for the velocity @xmath8 of the atomic tracer turns out to be formally identical to the colloidal case described by the two equations above , with the solvent friction coefficient @xmath20 replaced by a kinetic ( or ` doppler ' ) friction coefficient @xmath21 determined by kinetic - theoretical arguments . since the concept of kinetic friction may involve a rather subtle use of otherwise simple and well established concepts , we start this paper in section ii by providing a simple and intuitive description of the short - time random motion performed by an individual tracer particle in an _ atomic _ liquid , as a consequence of molecular collisions .
such description exhibits the fact that the resulting random motion must be described by the same stochastic mathematical model that describes the brownian motion of a tracer particle in a colloidal fluid .
this is just the mathematical model underlying the ordinary langevin equation ( eq .
( [ gletracerdif ] ) above without the time - dependent friction term and its associated random force ) .
thus , also in the atomic case , a relaxation time @xmath22 of the velocity , due to the friction force @xmath23 , defines the crossover from ballistic to diffusive motion .
the fundamental difference lies in the physical origin of the friction force @xmath24 and in the determination of the friction coefficient @xmath25 . in a brownian liquid
the friction force @xmath26 is caused by the supporting solvent ; hence , @xmath9 assumes its stokes value .
in contrast , as discussed in section ii , in a newtonian liquid the friction force @xmath24 is not caused by any external material agent but by the unimpeded tendency to establish or restore , through molecular collisions , the equipartition of the energy available for distribution among the kinetic energy degrees of freedom of the system .
thus , the corresponding value of @xmath25 that emerges from these considerations is provided by einstein s relation , with a kinetic - theoretically determined diffusion coefficient .
the discussion of the langevin equation for atomic liquids is continued in sections [ sectioniii ] and [ sectioniv ] , where a more formal and complete derivation is provided .
thus , in section [ sectioniii ] we show that the force on the tracer particle can be written as an integral of the divergence of the stress tensor @xmath27 , whose kinetic component @xmath28 is the origin of the doppler friction force @xmath23 and its associated random force , whereas the term involving the configurational component @xmath29 describes the effects of the ordinary conservative direct forces ( electrostatic , van der waals , etc , ) exerted by the surrounding particles .
it is also seen that the latter effects enter additively in the langevin equation of the atomic liquid , as an integral term that is linear in the instantaneous local density of the surrounding particles . in section [ sectioniv ]
we demonstrate that this linear coupling of the motion of the tracer with the local density of the surrounding particles leads to a time - dependent configurational friction term and to its corresponding random force , thus resulting in a generalized langevin equation for the velocity of a tracer particle , which turns out to be formally identical to its colloidal counterpart in eq .
( [ gletracerdif ] ) .
the main predictions of the resulting generalized langevin equation for atomic liquids are then discussed in section [ sectionv ] .
these include well defined scaling rules that exhibit the identity between the long - time dynamics of atomic and colloidal liquids .
there we test these scalings by comparing the simulation results for a given model system ( the hard sphere fluid ) using both , molecular dynamics and brownian dynamics simulations .
the last section summarizes the most relevant conclusions , and discusses some limitations and potential applications of the results of the paper .
let us consider a simple atomic fluid , formed by @xmath0 spherical particles in a volume @xmath30 whose microscopic dynamics is described by newton s equations , @xmath31 where @xmath7 is the mass and @xmath32 the velocity of the @xmath33th particle at position @xmath34 , and in which the interactions between the particles are represented by a sum of pairwise forces , with @xmath35 being the force exerted on particle @xmath33 by particle @xmath36 .
our general aim is to establish a connection between the microscopic dynamics described by these ( newton s ) equations , and the equation that describes the random motion of any representative individual tracer particle of the liquid . in this section
we discuss a simple and intuitive ( albeit possibly subtle ) physical picture , which plays a central role in our theoretical effort to establish such connection .
we start by recalling the physical meaning of two fundamental concepts in the dynamics of an atomic fluid , namely , the mean free time , @xmath22 , and the mean free path , @xmath37 , which represent the characteristic time- and length - scales at which the crossover from the short - time ballistic motion of the atoms to their long - time diffusive transport occur .
it is well known @xcite that for correlation times @xmath38 much shorter than @xmath22 , and for distances much shorter than @xmath37 , all the particles move ballistically , so that for @xmath39 , the mean squared displacement ( msd ) is given by @xmath40 , with @xmath41 being the thermal velocity . for times
@xmath38 much longer than @xmath22 , each particle has undergone many collisions , and its motion can be represented as a sequence of ( ballistic ) random flights of mean length @xmath37 and mean flight - time @xmath22 , traveled at a random velocity that has zero mean and covariance @xmath42 . from the theory of random flights , however , it is well - known @xcite that the motion represented by such a sequence of random displacements , will become diffusive in the long - time limit .
this means that it will be characterized by a mean squared displacement that , for @xmath43 , will increase linearly with time , @xmath44 .
furthermore , the corresponding diffusion coefficient @xmath45 will be given by @xmath46 .
thus , we conclude that the msd of a representative tracer particle will exhibit two well - defined limiting behaviors in two opposite time regimes , namely , it will be ballistic at short times , @xmath40 for @xmath47 , and diffusive at long times , @xmath44 for @xmath48 . the simplest mathematical model that provides a full description of the crossover of @xmath49 from the first to the second of these two exact limits ,
is provided by a gaussian stationary stochastic process , described by a linear stochastic equation with additive noise for the instantaneous velocity @xmath8 of the tracer particle , i.e. , by @xcite @xmath50 with @xmath51 being a purely random " ( or white " ) noise , i.e. , a stationary and gaussian stochastic process with zero mean ( @xmath52 ) , uncorrelated with the initial value @xmath53 of the velocity fluctuations , and delta - correlated with itself .
in fact , the stationarity condition is in reality equivalent to the fluctuation - dissipation relation between the random and the dissipative terms in eq .
( [ langevin0 ] ) , @xmath54 , where @xmath55 is the stationary covariance of the initial velocities . within the additional physical assumption that identifies the stationary state described by eq .
( [ langevin0 ] ) with the thermodynamic equilibrium state , we have that this covariance is determined by the equipartition theorem , @xmath56 ( with @xmath57 being the 3@xmath583 cartesian unit tensor ) . clearly , this equation is formally identical to the ordinary langevin equation @xcite for the instantaneous velocity @xmath8 of a colloidal particle in a solvent , in which case , stoke s result @xmath59 ( with @xmath60 being the viscosity of the solvent and @xmath61 the diameter of the colloidal particle ) provides an independent determination of the friction coefficient @xmath20 @xcite . in the present case ,
however , there is no supporting solvent to produce friction , and hence , identifying the origin and determining the value of @xmath25 requires slightly more subtle arguments . the simplest manner to describe its physical origin
may be found in uhlenbeck and ornstein s brief reference to the so - called _ doppler _ friction @xcite .
these authors point out that any tracer particle colliding with the particles of a gas , whose size @xmath61 is smaller than the mean free path @xmath62 , will be subjected to doppler friction , caused by the fact that when the tracer particle is moving , say to the right , will be hit by more molecules from the right than from the left " . in the following section we shall provide a more formal derivation of this kinetic friction effect . at this point ,
however , we provide simple arguments for its quantitative determination . to determine @xmath25 in the present case we first notice that the mathematical solution of eq .
( [ langevin0 ] ) for the msd is such that at long times , @xmath63 .
this determines a self - diffusion coefficient @xmath45 in terms of @xmath25 through einstein s relation , @xmath64 .
thus , we must only determine either @xmath25 or @xmath45 .
in our case , we write einstein s relation as @xmath65 and determine @xmath45 independently , borrowing the arguments developed in the elementary kinetic theory of gases @xcite . for this , we recall that @xmath46 , which , since @xmath66 , can be written as @xmath67 .
we then estimate the mean free path @xmath37 to be given by @xmath68 , with @xmath69 and with @xmath61 being the collision diameter of the particles .
thus , we must have that @xmath70 .
in fact , the rigorous value of @xmath45 is @xcite @xmath71 this expression , together with einstein s relation above , determines the value of the kinetic friction coefficient @xmath25 of an atomic fluid .
let us now discuss some of the implications of the _ atomic _ langevin equation defined by eqs .
( [ langevin0 ] ) , ( [ einstein ] ) , and ( [ dkinetictheory ] ) .
since the langevin equation itself is mathematically identical to the ordinary ( i.e. , colloidal ) langevin equation , its solution is also formally the same .
for example , from eq .
( [ langevin0 ] ) , and the assumed properties of @xmath72 one can evaluate the velocity auto - correlation function ( vaf ) @xmath73 with the result @xmath74 with @xmath75 being the velocity relaxation time . the msd , normalized as @xmath76 is related with the vaf by means of the exact relationship @xmath77 or , in terms of the laplace transforms ( lt ) @xmath78 and @xmath79 , as @xmath80 this exact relationship can be written , for the particular form of the vaf in eq .
( [ vdt0 ] ) , as the following differential equation , @xmath81 whose solution reads @xmath82 .
\label{wdt0}\ ] ] this expression interpolates @xmath83 between its corresponding short- and long - time asymptotic limits , @xmath84 and @xmath85 in addition , it also exhibits the fact that the crossover from ballistic to diffusive motion is most naturally described using @xmath86 as the unit of time .
let us notice , however , that for atomic liquids the relaxation time @xmath86 is identical to the mean free time @xmath22 , since @xmath87 .
thus , for an atomic liquid the mean free time is the most natural time unit , and the mean free path @xmath88 the most natural length unit since , for an atomic liquid , eq .
( [ wdt0 ] ) can be rewritten in terms of the scaled time @xmath89 and the scaled msd @xmath90 as @xmath91 .
\label{wdtstar}\ ] ] to illustrate the validity of this result , in fig .
1 we compare it with the molecular dynamics simulation data ( solid circles ) for @xmath92 in a fluid of hard spheres of diameter @xmath61 at a small but finite volume fraction @xmath93 , namely , at @xmath94 . clearly , the scaled solution ( [ wdtstar ] ) of the atomic langevin equation ( solid curve ) lies very close to the simulation data of the msd .
$ ] as a function of the scaled time @xmath95 $ ] , with @xmath62 and @xmath96 being the mean free path and mean free time , of a hard - sphere fluid at volume fractions @xmath97 ( solid circles ) and @xmath98 ( empty circles ) generated by molecular dynamics simulations of soft sphere systems using the velocity verlet s algorithm @xcite and the soft - to - hard - sphere dynamic correspondence of ref . @xcite .
the solid line corresponds to the low density limit , eq . ( [ wdtstar ] ) , and the dashed line corresponds to the expression in eq .
( [ wdtstar1 ] ) with @xmath99 . ]
the dimensionless quantities @xmath92 and @xmath100 serve to highlight an important scaling that derives directly from the atomic langevin equation in eq .
( [ langevin0 ] ) . at the same time , however , these scaled variables hide the specificity of the actual magnitudes involved in concrete real physical situations . for example , let us think of a typical atomic liquid , such as argon ( @xmath101 amu , @xmath102 ) , and rewrite the expression for @xmath45 in eq .
( [ dkinetictheory ] ) as @xmath103 . \label{dkinetictheory2}\ ] ] then , for @xmath104 and @xmath105 , representative of a moderately dilute gas at room temperature , we find that @xmath106 , @xmath107 , and @xmath108ps .
let us now compare this crossover timescale @xmath86 with the structural relaxation timescale @xmath109 , the time it would take any particle to _ diffuse _ one interparticle mean distance @xmath110 with the diffusion coefficient @xmath45 . for the conditions under consideration we find that @xmath111ps , so that @xmath112 is only about twice longer than @xmath86 .
this means that there is nothing like a time - scale separation between the crossover from ballistic to diffusive motion and the crossover from free to collective relaxation ( i.e. , from the so - called @xmath113-processes to the slower @xmath114-processes ) .
although this timescale degeneration disappears gradually as the density increases , giving rise to the full separation of time scales characteristic of the approach to the glass transition , this degeneration of time scales is a characteristic feature of atomic dynamics throughout the whole stable liquid phase of these simple systems . to illustrate this , notice that taking @xmath115 and @xmath116 , representative of the freezing conditions of argon , leads to @xmath117 , @xmath118 , @xmath119ps , and @xmath120ps .
, so that now @xmath121 .
this situation must now be compared with its corresponding colloidal analog . in this case , the velocity relaxation time @xmath122 is no longer identical to the mean free time .
instead , it is determined by the mass m of the particle and by the solvent friction coefficient @xmath20 , now given by its stokes value @xmath123 , with @xmath60 being the viscosity of the solvent .
thus , consider a micron - sized colloidal particle in water at room temperature and with the same mass density as water itself , so that @xmath124 , @xmath125 kg , and @xmath126 kg / s ( taking @xmath127 kg / m@xmath128s for the viscosity of water ) .
this leads to the estimate @xmath129 for the ( @xmath130-independent ) einstein diffusion coefficient @xmath131 .
the corresponding estimate of @xmath132 is , then , @xmath133s , whereas the value of @xmath134 ( for @xmath97 ) is @xmath135s .
we thus immediately notice a really dramatic separation of timescales , expressed by the fact that now @xmath136 . as a consequence , in a colloidal suspension ,
the crossover from ballistic to diffusive motion ( occurring at very early times @xmath137 ) and the crossover from free to collective difusion ( occurring at times @xmath138 ) , are separated by nearly 8 decades .
thus , if one were interested in observing the ballistic - to - diffusive crossover in our illustrative colloidal suspension , one would have to measure @xmath83 in the time window corresponding to fig . 1 , i.e , for times centered in the regime @xmath139s . this crossover , however , will be completely shifted to unobservable , extremely short times , when probing only the structural relaxation of the colloidal liquid for times @xmath38 in the milisecond range and above ( i.e. , @xmath138 ) .
for this reason in the description of the dynamics of suspensions the inertial term of the langevin equation is normally neglected , @xmath140 , leading to the _ overdamped _ langevin equation , @xmath141 , whose solution for the msd is @xmath142 .
this , however , is just the diffusive limit in eq .
( [ wdt0long ] ) .
the main conclusion of these illustrative estimates is , thus , that in colloidal liquids this wide separation of timescales is always present , whereas it is virtually nonexistent in an atomic liquid .
as mentioned above , however , this disparity between atomic and colloidal liquids will disappear in the metastable regime due to the dominant effect of interparticle interactions in the determination of the structural relaxation processes . to introduce the discussion of these effects , in fig .
1 we have also included the molecular dynamics simulation results for the msd of the hard - sphere liquid at freezing conditions , @xmath98 ( empty circles ) . these results illustrate the deviations from the ideal behavior described by eq .
( [ wdtstar ] ) , which originate from the structural effects of the interparticle interactions ( very moderate , and almost imperceptible in the results for @xmath97 ) .
the comparison between the results for the msd at these two volume fractions clearly show that in both cases the ballistic short - time behavior ( i.e. , for times @xmath143 ) is accurately described by eq .
( [ wdtstar ] ) . beyond the crossover time @xmath22 , however , the results for @xmath97 exhibit only the kinetic effects of the interparticle interactions , whereas the results for @xmath98 exhibit the combination of both , the kinetic and the structural effects of the interparticle forces .
one of the results of the following sections will be a simple expression for @xmath83 that preserves the short - time limit @xmath144 , but replaces the ideal long - time limit @xmath142 by the correct one , @xmath145 where @xmath146 is the long - time self - diffusion coefficient ( see end of sec .
[ sectioniv ] ) .
this expression reads @xmath147 , \label{wdtstar1}\ ] ] with the parameter @xmath148 being the ratio of the _ long - time _ to the short - time self - diffusion coefficients .
in fact , it is the solution of @xmath149 which extends eq .
( [ wdtintdifec0 ] ) to finite densities .
the dashed line in fig .
1 corresponds to this expression with the value @xmath150=0.099 .
the comparison with the simulation data indicates that eq .
( [ wdtstar1 ] ) provides an accurate representation of @xmath83 at short and at long times compared with @xmath22 . in this comparison ,
the parameter @xmath150 was employed as a mere fitting parameter .
the idea , however , is to build a first - principles theory that _ predicts _ the value of @xmath150 and the behavior of @xmath83 in the intermediate - time regime .
the results of the present paper will eventually constitute an essential ingredient in the construction of the _ atomic _ version of the self - consistent gle theory .
the most immediate task , however , is to provide a more formal understanding of the physical meaning and the fundamental nature of the kinetic friction represented by @xmath25 , and in general , of the langevin equation for atomic liquids in eq .
( [ langevin0 ] ) .
this task is addressed in the following two sections .
the fact that the langevin equation in eq .
( [ langevin0 ] ) adequately describes the results for the msd of newtonian systems strongly suggests a more profound equivalence between the dynamics of newtonian and brownian liquids , at least regarding tracer diffusion phenomena . in both cases
the relaxation time @xmath151 $ ] of the velocity , due to the friction force @xmath23 , defines the crossover from ballistic ( @xmath152 ) to diffusive ( @xmath153 ) motion .
the fundamental difference lies in the physical origin of the friction force @xmath23 : in a brownian liquid , this friction is assumed to be caused by an external material agent , namely , the supporting solvent , which also acts as a heat reservoir . in a newtonian liquid , in contrast
, the ( ` doppler ' ) friction force @xmath23 is not caused by any external material agent but by the molecular collisions responsible to impose the equipartition of energy among the kinetic energy degrees of freedom of the system .
in other words , the underlying physical origin of the kinetic friction effects is that for times @xmath38 much longer than @xmath22 , in which each particle has undergone many molecular collisions , the kinetic energy has indeed been distributed according to the most probable distribution .
this then means that partial thermal equilibrium , consisting of this equipartition of the kinetic energy , is achieved within the time scale represented by @xmath22 .
such partial thermalization must involve the transport of heat through molecular collisions .
the instantaneous fluctuations of this local temperature , however , lead to the random force @xmath154 that cause the instantaneous fluctuations in the velocity of the tracer particle .
the same molecular mechanisms are also responsible for the emergence of the systematic friction force @xmath155 on the tracer particle whenever this particle has a non - zero mean velocity @xmath156 .
let us now develop a more microscopic and precise discussion of these physical effects .
for this , let us now consider @xmath157 particles in a volume @xmath30 , the first of them ( the tracer particle ) having momentum @xmath158 and the rest @xmath0 particles having momentum @xmath159 , with @xmath160 . in the absence of external forces ,
the total momentum @xmath161 is conserved , @xmath162 = 0,\ ] ] or @xmath163 = 0,\ ] ] with @xmath164 and with @xmath165 being the total confining volume @xmath30 minus the excluded volume @xmath166 around the center of the tracer particle .
thus , @xmath167 is bounded by an outer surface @xmath168 of the confining walls , which remain fixed in space , and by the surface @xmath61 around @xmath167 , which follows the motion of this particle , and hence , is not fixed in space . clearly , then
, the momentum conservation equation can also be written as @xmath169 on the other hand , taking the time derivative of eq .
( [ current ] ) , one can write the continuity equation for the momentum density @xmath170 as @xmath171 where @xmath172 and @xmath173 are the kinetic and configurational components of the stress tensor @xmath27 , i.e. , @xmath174 the kinetic part @xmath172 , which describes the change of momentum due to particles crossing the boundaries of @xmath167 , is a second rank tensor with components @xcite @xmath175^{\alpha \beta } \equiv -\sum _ { i=1}^n \frac{p_i^\alpha(t ) p_i^\beta(t)}{m}\delta(\mathbf{r}-\mathbf{r}_i(t ) ) . \label{pikin}\ ] ] the configurational part of @xmath27 , on the other hand , is defined by the condition @xmath176 so that the continuity equation for @xmath170 , eq .
( [ cont1 ] ) , can actually be written most simply as @xmath177 thus , substituting this equation in eq .
( [ cont0 ] ) , we have @xmath178 the second term on the right side of eq .
( [ cont2 ] ) is just @xmath179=-\sum_{i=1}^n { \mathbf{f}}_{it } ( t)=\sum_{i=1}^n { \mathbf{f}}_{ti } ( t)$ ] , where @xmath180 is the force exerted on particle @xmath33 by particle @xmath36 and @xmath181 is the force exerted _ by _ particle @xmath33 on the tracer particle , and where we have used the fact that @xmath182 .
we may then write this term as @xmath183\sum_{i=1}^n \delta ( \mathbf{r}-\mathbf{r}_i(t ) ) \
d^3r$ ] . by shifting the origin of the coordinate system to the center of the tracer particle ( including the variable of integration @xmath184 ) , it is not difficult to see that eq .
( [ cont2 ] ) can be rewritten as @xmath185 n^*(\textbf{r},t)\ d^3r , \label{cont3}\ ] ] with @xmath186 being the local particle density around the tracer particle described , however , from a reference frame _ whose origin moves _ together with the center of this particle .
( [ cont3 ] ) shows that there is a very simple and exact coupling between the force on the tracer particle and two collective variables of the surrounding fluid , namely , the kinetic component @xmath28 of its stress tensor ( whose trace is a measure of the instantaneous local kinetic energy per particle , i.e. , of the local instantaneous temperature ) and the local number density @xmath187 .
thus , the interatomic forces affect the motion of a tracer particle through two physically distinct channels .
the first of them is a kinetic constraint imposed by momentum conservation , and the second is the ordinary configurational effect of interatomic forces .
each of these two variables contribute additively and linearly to the total force on the tracer particle . taking the equilibrium average ( indicated by an overbar ) of eq .
( [ cont3 ] ) , we have that @xmath188\overline { n^*(\textbf{r})}\ d^3r , \label{meancont3}\ ] ] where the mean value of @xmath28 can be obtained averaging eq .
( [ pikin ] ) , with the result @xmath189 with @xmath57 being the 3@xmath583 cartesian unit tensor and with @xmath190 given by @xmath191 where @xmath192 is the bulk radial distribution function of the system . from these results
it is not difficult to realize that for symmetry reasons each of the two mean forces on the right side of eq .
( [ meancont3 ] ) vanish independently , so that the tracer particle experiences a vanishing total mean force , and @xmath193 . we may now write the state variables @xmath194 and @xmath195 as the sum of their equilibrium mean value plus the corresponding fluctuations , namely , as @xmath196 and @xmath197 this allows us to rewrite eq .
( [ cont3 ] ) as an exact relationship between these thermal fluctuations and the instantaneous momentum of the particle , namely @xmath198\delta n^*(\textbf{r},t)\ d^3r .
\label{contfluct}\ ] ] this exact equation will now be taken as the starting point for a formal statistical mechanical derivation of the ordinary langevin equation of a the tracer particle in our atomic liquid .
for this , let us first recall that the basis of the gle formalism are the general mathematical conditions stated by the theorem of stationarity @xcite .
this theorem states that the equation describing the dynamics of the thermal fluctuations @xmath1 of the instantaneous value of the macroscopic variables @xmath2 ( @xmath3 ) around its equilibrium value @xmath4 must have the structure of the most general linear stochastic equation with additive noise for the vector @xmath5^\dagger $ ] , namely , @xmath199 in this equation @xmath200 is the matrix of static correlations , @xmath201 , @xmath202 is an anti - hermitian matrix ( @xmath203 ) , and the matrix @xmath204 is determined by the fluctuation - dissipation relation @xmath205 , where @xmath206 is the @xmath33th component of the vector of random forces @xmath207 . besides the selection rules imposed by these symmetry properties of the matrices @xmath200 , @xmath202 , and @xmath204 , other selection rules are imposed by other symmetry conditions . for example @xcite , if the variables @xmath2 have a definite parity upon time reversal , @xmath208 with @xmath209 1 or -1 , then @xmath210 and @xmath211 .
let us now apply this mathematical infrastructure to the physical context involving the exact momentum conservation equation , eq .
( [ cont3 ] ) , and let us define the vector @xmath212 , partitioned as @xmath213^\dagger , \label{vectora}\ ] ] in terms of the sub - vectors @xmath214 , and @xmath215 , defined by their components @xmath216 @xmath217^{\alpha \beta}(\textbf{r } ) = \delta \pi_k ^{\alpha \beta}(\textbf{r},t ) \ \ \ \ ( \textrm{with}\ \ \ \alpha , \beta = x , y , z,\ \ \ \ \textrm{and } \ \ \ \textbf{r}\in v ) \label{vectorpikin}\ ] ] and @xmath218(\textbf{r } ) = \delta n^*(\textbf{r},t ) \ \ \ \ ( \textrm{with } \ \ \ \textbf{r}\in v ) .
\label{vectorpikin}\ ] ] with this definition of the vector @xmath212 one can calculate the static correlation matrix @xmath219 using the microscopic definitions of @xmath215 in eqs .
( [ nstar ] ) , ( [ meanlocaln ] ) , and ( [ dmeanlocal ] ) and of @xmath220 in eqs .
( [ pikin ] ) , ( [ meanpikin ] ) , and ( [ dpikin ] ) .
the result for @xmath221 can be written as the following partitioned matrix @xmath222,\ ] ] whose sub - matrices have elements defined as @xmath223^{\alpha \beta}\equiv \langle p^\alpha p^\beta \rangle $ ] , @xmath224^{\alpha \beta,\mu \nu}(\textbf{r},\textbf{r}')\equiv \langle \delta \pi_k ^{\alpha \beta}(\textbf{r})\delta \pi_k ^{\mu \nu}(\textbf{r}')\rangle $ ] , @xmath225^{\alpha \beta}(\textbf{r},\textbf{r}')\equiv \langle \delta \pi_k ^{\alpha \beta}(\textbf{r})\delta n^*(\textbf{r}')\rangle $ ] , @xmath226^{\mu \nu}(\textbf{r},\textbf{r}')\equiv \langle \delta n^*(\textbf{r})\delta \pi_k ^{\mu \nu}(\textbf{r}')\rangle $ ] , and @xmath227(\textbf{r},\textbf{r}')\equiv \langle \delta n^*(\textbf{r})\delta n^*(\textbf{r}')\rangle $ ] , given , respectively , by @xmath228^{\alpha \beta}=(mk_{b}t ) \delta _ { \alpha \beta } , \label{chipp}\ ] ] @xmath229^{\alpha \beta,\mu \nu}(\textbf{r},\textbf{r}')=(k_{b}t)^2\ { [ \delta _ { \alpha \beta}\delta _ { \mu \nu } + \delta _ { \alpha \mu}\delta _ { \beta \nu } + \delta _ { \alpha \nu}\delta _ { \mu \beta}]\chi_s(\textbf{r},\textbf{r } ' ) + \delta _ { \alpha \beta}\delta _ { \mu \nu}\chi_d(\textbf{r},\textbf{r}')\ } , \label{chipipi}\ ] ] @xmath230^{\alpha \beta}(\textbf{r},\textbf{r}')=[\chi _ { n \pi } ] ^{\alpha \beta}(\textbf{r},\textbf{r}')=-(k_{b}t ) \delta _ { \alpha \beta}\chi_d(\textbf{r},\textbf{r } ' ) , \label{chipin}\ ] ] and @xmath231(\textbf{r},\textbf{r}')=\chi_s(\textbf{r},\textbf{r } ' ) + \chi_d(\textbf{r},\textbf{r } ' ) , \label{chinn}\ ] ] where the self and the distinct parts of @xmath232 are defined as @xmath233 and @xmath234 we then write up the generalized langevin equation for our vector @xmath235 in the format of eq .
( [ gle0 ] ) . for this
, we first notice that all the variables , except @xmath236 , are even functions under time - reversal .
according to onsager s reciprocity relations , and the general anti - hermiticity of @xmath202 and hermiticity of @xmath237 @xcite , we have that the only possibly non - zero submatrices of @xmath202 and @xmath204 are @xmath238\ ] ] and @xmath239.\ ] ] the determination of some of the non - zero elements of @xmath202 and @xmath204 is rather straightforward .
thus the previous selection rules , along with the general format imposed by the gle equation ( [ gle0 ] ) , allows us to write the time - evolution equation for the sub - vector @xmath236 as @xmath240\delta\pi ( t)-\left[\omega_{p\pi}(\chi^{-1})_{\pi n } + \omega_{p n}(\chi^{-1})_{n
n } \right]\delta n^ * ( t ) \nonumber \\ & & -\int_0^t dt ' l_{pp}(t - t')(\chi^{-1})_{pp}\mathbf{p}(t ' ) + f_p(t ) \label{dpdtgral}\end{aligned}\ ] ] by comparing with the exact momentum conservation equation in eq .
( [ dpdtgral ] ) , we immediately conclude that @xmath241 , and that the remaining terms correspond , respectively , to the kinetic and configurational forces on the right side of this equation .
in addition , for simplicity we approximate @xmath242 in eqs .
( [ chipin ] ) and ( [ chid ] ) , so as to neglect at this point the static cross - correlation @xmath243 and @xmath244 , @xmath230^{\alpha \beta}(\textbf{r},\textbf{r}')=[\chi _ { n \pi } ] ^{\alpha \beta}(\textbf{r},\textbf{r}')\approx 0,\ ] ] so that the previous equation is rewritten as @xmath245 comparing this equation with eq .
( [ dpdtgral ] ) one can determine the sub - matrices @xmath246 and @xmath247 and hence , also the sub - matrices @xmath248 and @xmath249 . in a similar manner , from the exact format imposed by the gle , and using the previous selection rules ( as well as the approximation @xmath250 ) , one can also write the time - evolution equations for the other two variables , @xmath251 and @xmath215 , as @xmath252 and @xmath253 eqs .
( [ dpdtgralp])-([dndtgralp ] ) provide a non - contracted description of the thermal fluctuations in an atomic liquid , which involves the tracer particle s momentum @xmath254 as one of the variables . contracting this description to the state subspace spanned by @xmath254 itself
will finally lead to the complete generalized langevin equation for a tracer particle in such atomic liquid .
the result , however , is rather involved , but the essence may be best appreciated if we introduce an additional simplification , which consists of neglecting the dissipative coupling between the variables @xmath251 and @xmath215 , i.e. , by setting @xmath255 under these circumstances , the solution of eq .
( [ dpidtgralp ] ) can be written as @xmath256 where the green s function @xmath257 is the solution of @xmath258 with initial condition @xmath259 , i.e. , its laplace transform ( lt ) @xmath260 will be given , in terms of the lt of @xmath261 , by @xmath262^{-1}. \label{ltgreen}\ ] ] substituting the expression above for @xmath263 in eq .
( [ dpdtgralp ] ) we arrive at the following langevin equation @xmath264 where we have defined the time - dependent kinetic friction coefficient @xmath265 as @xmath266 and the kinetic random force @xmath267 as @xmath268 .
\label{funcgreenf}\ ] ] according to the theorem of contractions @xcite , @xmath267 and @xmath269 must satisfy the fluctuation - dissipation relationship @xmath270 . for future reference , let us notice that eq .
( [ ltgreen ] ) allows us to write the ft of the kinetic friction function @xmath271 in eq .
( [ deltazetak ] ) directly in terms of the ft of the memory function @xmath261 as @xmath272^{-1 } \omega_{\pi p}\chi^{-1}_{pp}. \label{deltazetakp}\ ] ] clearly , the exact determination of the memory function @xmath261 , and hence , of the kinetic friction coefficient @xmath273 , is perhaps impossible , but some properties can be drawn from the expressions just derived , at least in certain limits and within well - defined approximations .
for example , if @xmath273 relaxes to zero within a finite relaxation time , then for times @xmath38 much longer than such relaxation time we can approximate @xmath273 by its markov limit , @xmath274 where @xmath275 in addition , due to the radial symmetry of the interparticle interactions , @xmath276 must be isotropic ( i.e. , diagonal ) , so that @xmath277 with @xmath278 given by @xmath279^{xx}. \label{zetamarkov}\ ] ] thus , we conclude that in the markov limit eq .
( [ dpdtgralpp ] ) can be written as @xmath280 in the following section we discuss the additional contraction process that leads to the elimination of the variable @xmath281 from the description consisting of this equation and of eq .
( [ dndtgralp ] ) above . before that , however , let us mention that eq .
( [ deltazetakp ] ) , and in particular its markov limit in eq .
( [ zetamarkov ] ) , involves the lt of the memory function @xmath282 as the only unknown quantity , which the gle formalism is unable to determine .
although one could introduce additional approximations to determine this memory function , this is not the main objective of the present paper ; instead , the derivation above was only meant to provide a more formal explanation of the origin of the kinetic friction and random forces , introduced and discussed in more efficient and intuitive terms in the previous section .
after all , such arguments did provide a simple and accurate zeroth - order approximate determination of the kinetic friction coefficient @xmath278 , namely , the use of the kinetic - theory value of the self - diffusion coefficient , eq .
( [ dkinetictheory ] ) , in einstein s relation , eq .
( [ einstein ] ) .
one important contribution of the previous section was to make a point that in an atomic liquid the force on a tracer particle couples linearly with the kinetic component @xmath28 of the stress tensor and with the local number density @xmath187 of the surrounding fluid , as indicated by eq .
( [ cont3 ] ) . as a consequence ,
the interatomic forces affect the motion of a tracer particle through two physically distinct channels , namely , the kinetic constraint imposed by momentum conservation and the ordinary configurational effect of interatomic forces .
another important conclusion was to notice that the former is the origin of the kinetic friction , finally formatted in eq .
( [ dpdtgralppp ] ) as a dissipative friction term @xmath283 plus the corresponding random force @xmath51 .
let us now discuss the effects of the coupling with @xmath284 . for this ,
let us resume the formal process of contraction of the description initiated in the previous section .
we thus recall that after projecting out the variable @xmath285 , the time - evolution equations for the remaining fluctuating variables are eqs .
( [ dpdtgralpp ] ) and ( [ dndtgralp ] ) . for clarity , we rewrite here these equations as @xmath286 and @xmath287 where the kinetic friction term of eq .
( [ dpdtgralpp ] ) has been written in its markov limit ( as in eq .
( [ dpdtgralppp ] ) ) , and where the term of eq .
( [ dndtgralp ] ) involving @xmath288 has been neglected , as discussed immediately above eq .
( [ funcgreenpi ] ) .
we now formally project out the variable @xmath215 by solving eq .
( [ dndtgralpp ] ) for this variable , and substituting the resulting solution in the third term of the right side of eq .
( [ dpdtgralpppp ] ) .
this contraction process results in the following generalized langevin equation for the velocity @xmath289 of a tracer particle in the atomic liquid , @xmath290 where the _ configurational _ time - dependent friction function @xmath291 is given by @xmath292 and the new ( configurational ) random force @xmath293 as @xmath294 .
\label{funcgreenconf}\ ] ] in these equations , the green s function @xmath295 is the solution of @xmath296 with initial condition @xmath297 .
according to the contraction theorem @xcite , @xmath293 and @xmath298 must satisfy the fluctuation - dissipation relationship @xmath299 . although the previous statements are physically accurate and well - defined , it is also useful to rephrase this abstract derivation in more concrete and intuitive terms . for this
we rewrite eq .
( [ dpdtgralpppp ] ) as @xmath300\delta n^*(\textbf{r},t ) , \label{eq3}\ ] ] to recover the original notation in eq .
( [ contfluct ] ) for the configurational force term .
the comparison with eq .
( [ dpdtgralpppp ] ) then implies that @xmath301^{\alpha}(\textbf{r } ) = [ \nabla^\alpha u(r)]$ ] and , convoluting this equation with @xmath227(\textbf{r},\textbf{r}')\equiv \langle \delta n^*(\textbf{r})\delta n^*(\textbf{r}')\rangle $ ] determines that @xmath302 is given by @xmath303^{\alpha}(\textbf{r } ) & & = - \int d^3r ' [ \nabla^{'\alpha } u(r')]\chi _ { n n}(\textbf{r}',\textbf{r } ) \nonumber \\ & & = k_bt [ \nabla^{\alpha}n^{eq}(\textbf{r})],\end{aligned}\ ] ] where the second equality is a direct consequence of the exact equilibrium condition referred to as the wertheim - lovett s relation , namely @xcite , @xmath304= - \beta \int d^3r ' \chi _ { n n}(\textbf{r}',\textbf{r})[\nabla^{'\alpha } u(r')].\end{aligned}\ ] ] in the previous equations , the equilibrium mean value @xmath305 has been denoted by @xmath306 .
now , since @xmath307^\dagger$ ] , we find that @xmath308^{\alpha}(\textbf{r } ) = - k_bt [ \nabla^{\alpha}n^{eq}(\textbf{r})]$ ] .
using this result , along with the value @xmath223^{\alpha \beta}=(mk_{b}t ) \delta _ { \alpha \beta}$ ] ( eq .
( [ chipp ] ) ) , we can write eq .
( [ dndtgralpp ] ) more concretely as @xmath309 \cdot { \bf v}(t ) -\int_0^t dt ' \int d^3 r ' d^*(\textbf{r},\textbf{r}';t - t')\delta n^*(\textbf{r}',t ' ) + f(\textbf{r},t ) , \label{eq4}\ ] ] with @xmath310 being the elements of the matrix " @xmath311 $ ] .
the first term on the right side of this equation is a linearized streaming term and @xmath312 is a fluctuating term , related to @xmath310 by @xmath313 . formally solving eq .
( [ eq4 ] ) and substituting the solution for @xmath284 in eq .
( [ eq3 ] ) , leads again to the generalized langevin equation in eq .
( [ eq6 ] ) with the time - dependent friction tensor @xmath314 of eq .
( [ deltazetaconf ] ) given more concretely by @xmath315 g^*(\textbf{r},\textbf{r}';t)[\nabla ' n^{eq}(r ' ) ] , \label{eq7}\ ] ] where @xmath316 is the propagator , or green s function , of eq .
( [ eq4 ] ) , i.e. , it solves the equation @xmath317 with initial value @xmath318 .
notice that , since the initial value @xmath319 is statistically independent of @xmath320 and @xmath312 , the density - density time - correlation function @xmath321 , which is the van hove function of the particles surrounding the tracer particle , and observed from the tracer particle s reference frame , is also a solution of the same equation with initial value @xmath322 . to simplify the notation ,
let us re - write eq .
( [ eq7 ] ) as @xmath323 \cdot g ^*(t)\cdot[\nabla n^{eq } ] $ ] , where the convolution @xmath324 between two arbitrary functions @xmath325 and @xmath326 is written as the inner product @xmath327 , and similarly with ( column ) vectors " such as @xmath328 and @xmath329 . in this notation , the dagger means transpose . with this notation ,
wertheim - lovett s relation reads @xmath330
= -\beta \chi_{nn } \cdot [ \nabla u]$ ] . with this relation , and the definition of the the inverse matrix @xmath331 by the equation @xmath332 , with @xmath333 being the unit matrix ( @xmath334 = dirac s delta function )
, one can write eq .
( [ eq7 ] ) in a variety of different but equivalent manners .
in particular , we will employ the following : @xmath335\cdot \chi_{nn}^{-1 } \cdot \chi^*(t ) \cdot \chi_{nn}^{-1 } \cdot [ \nabla n^{eq } ] , \label{eq9}\ ] ] where we have used the fact that the van hove function @xmath336 can be written as @xmath337 .
let us now notice that for spherical particles @xmath298 is isotropic and diagonal , @xmath338 so that we only have to calculate the scalar time - dependent friction function @xmath339 .
the exact expressions for @xmath13 can then be given a more concrete and tractable appearance if some approximations are introduced , related to the general properties of the functions @xmath340 and @xmath341 .
the latter is just the two - particle distribution function of the colloidal particles surrounding the tracer particle , but subjected to the external " field @xmath342 exerted by this tracer particle .
thus , it is effectively a three - particle correlation function . only if one ignores the effects of such external " field , one can write @xmath343 $ ] . similarly , we may also approximate @xmath340 by @xmath344 .
this is referred to as the _ homogeneous fluid approximation " _
@xcite , which then allows us to write @xmath345 ns(k)\ ] ] and @xmath346 nf^*(k , t),\ ] ] with @xmath347 \rangle .
\label{fstardkt}\ ] ] and @xmath348 using these expressions in eq .
( [ eq9 ] ) , along with the fact that @xmath349 ( so that its ft is @xmath350 $ ] ) , we have that eq .
( [ eq9 ] ) becomes @xmath16}{s(k)}\right ] ^{2}f^*(k , t ) .
\label{dzdt00}\ ] ] the function @xmath351 in this equation is just the intermediate scattering function , but the asterisk indicates that the position vectors @xmath352 and @xmath353 have the origin in the center of the tracer particle . denoting by @xmath354 the position of the tracer particle referred to a laboratory - fixed reference frame , we may re - write @xmath355)\right]\cdot\left[\exp ( i\textbf{k}\cdot[\textbf{x}_t(t)-\textbf{x}_t(0)])\right]\ \rangle , \label{decouplingapprox}\ ] ] where @xmath352 is the position of the @xmath33th particle in the fixed reference frame .
approximating the average of the product in this expression by the product of the averages , leads to @xmath356 where @xmath357\ } \rangle$ ] is the _ self _ isf .
this is referred to as the _ decoupling approximation _
thus , from the exact result in eq .
( [ eq9 ] ) above , plus the introduction of the two approximations just described , we finally arrive at the following general but approximate expression for the time - dependent friction function @xmath13 , @xmath16}{s(k)}\right ] ^{2}f(k , t)f_s(k , t ) .
\label{dzdt0p}\ ] ] this expression is reminiscent of the corresponding mode coupling theory ( mct ) result @xcite . its derivation above , however , follows a completely different conceptual route . in the following section we discuss important implications of our results above
the main result of the previous sections is , of course , the generalized langevin equation ( eq . ( [ eq6 ] ) )
describing the ballistic to diffusive crossover of the brownian motion of individual tracer particles in an atomic liquid .
taking into account the isotropy of the configurational time - dependent friction function ( eq .
( [ isotropydzeta ] ) ) , this stochastic equation reads @xmath358 the configurational effects of the interparticle interactions is embodied in the time - dependent friction function @xmath359 , which eq .
( [ dzdt0p ] ) writes in terms of the isfs @xmath14 and @xmath15 .
thus , the full analysis of this stochastic equation requires in principle the previous determination of these more complex dynamic properties .
some important implications , however , can be drawn without a detailed knowledge of @xmath359 .
the most remarkable conclusion is that the brownian motion of individual tracer particles in atomic and colloidal liquids is described by the same mathematical model , namely , the generalized langevin equation derived here for atomic systems ( eq .
( [ eq6p ] ) with eq .
( [ dzdt0p ] ) ) , and the gle derived in ref .
@xcite for colloidal fluids ( eq .
( [ gletracerdif ] ) with eq .
( [ dzdt0 ] ) ) . according to this formal mathematical similarity , the properties that describe the tracer s random motion in atomic and in colloidal liquids , such as the mean squared displacement @xmath83 , should collapse onto each other when expressed in dimensionless units that absorb the mass @xmath7 and the short - time friction coefficient ( @xmath20 or @xmath25 ) . to see this ,
let us first notice that from eq .
( [ eq6p ] ) one can write the velocity autocorrelation function @xmath360 in terms of @xmath359 , in laplace space , as @xmath361 with the friction coefficient @xmath362 representing either the kinetic friction coefficient @xmath25 in atomic fluids or the solvent friction coefficient @xmath20 in colloidal liquids , @xmath363 using this result in the exact relationship in eq .
( [ wdzvdz ] ) one can derive the following integro - differential equation for @xmath83 , @xmath364w(t')dt ' , \label{wdtintdifec}\ ] ] where @xmath365 is the crossover timescale from ballistic to diffusive motion ( and which equals the mean free time @xmath22 only in atomic liquids ) , and where the short - time self - diffusion coefficient @xmath366 is defined by einstein s relation , @xmath367 thus , @xmath368 is given by the kinetic theoretical result in eq .
( [ dkinetictheory ] ) only for atomic liquids . for both , atomic and brownian tracers , in the absence of interactions @xmath369 vanishes and eq .
( [ wdtintdifec ] ) becomes eq .
( [ wdtintdifec0 ] ) , discussed in sec . [ sectionii ] for atomic liquids , and whose solution is given by eq .
( [ wdt0 ] ) . in the presence of interactions , however , @xmath370 , but at very short times ( @xmath371 ) the solution of eq .
( [ wdtintdifec ] ) is still identical to that of a freely - flying particle , i.e. , the short - time asymptotic expression for the msd is also given by @xmath372 , as illustrated by the molecular dynamics simulations in fig .
[ fig1 ] . in the opposite regime , @xmath373 ,
the interparticle interactions change the long - time asymptotic limit of @xmath83 from its free - diffusion value @xmath374 to the new value @xmath375 , which defines the long - time self - diffusion coefficient @xmath146 . in this regime ,
the convolution @xmath376 on the right side eq .
( [ wdtintdifec ] ) can be approximated by its markov limit @xmath377w(t ) = \delta \zeta w(t)$ ] , so that eq .
( [ wdtintdifec ] ) reads @xmath378 where @xmath379 with the constant @xmath380 defined as @xmath381 this equation is precisely eq .
( [ wdtintdifec1 ] ) of sec .
[ sectionii ] , whose analytic solution in eq .
( [ wdtstar1 ] ) was shown in fig . 1 to provide a simple interpolation between the short- and long - time limits of the molecular dynamics simulation data for the msd of the hs liquid throughout its thermodynamically stable liquid regime , @xmath382 .
to continue the discussion of the dynamic equivalence between atomic and brownian liquids , let us define in both cases a length @xmath383 in terms of @xmath86 and @xmath366 as @xmath384 for atomic liquids @xmath383 is identical to the mean free path @xmath62 , but not for colloidal systems , for which @xmath383 is only given by this equation ( together with eqs .
( [ taub ] ) and ( [ einsteinshort ] ) ) .
we may now use @xmath383 and @xmath86 as the units of length and time , respectively , and rewrite eq .
( [ wdtintdifec ] ) in terms of the scaled time @xmath385 and the scaled msd @xmath386 as @xmath387 where the configurational timescale @xmath112 , the time it takes a particle to diffuse a mean distance @xmath110 with a diffusion coefficient @xmath366 , is given by @xmath388 and in which we have defined the dimensionless function @xmath389 as @xmath390 = \frac{1}{3\left ( 2\pi \right ) ^{3}n^{5/3}}\int d { \bf k}\left[\frac { k[s(k)-1]}{s(k)}\right ] ^{2}f(k , t)f_s(k , t).\ ] ]
let us notice that the purpose of using @xmath383 and @xmath86 as the units of length and time is to focuss on the crossover time - regime from ballistic to diffusive motion .
if , instead , we were interested in focussing on the crossover from free - diffusion to structural relaxation , the best would be to use the mean inter - particle distance @xmath110 as the unit of length and @xmath112 as the time unit , and to rewrite eq .
( [ wdtintdifec ] ) in terms of the scaled time @xmath391 and the scaled msd @xmath392 , to read @xmath393 for dense atomic liquids ( e.g. , hard spheres at @xmath394 ) either choice is perfectly adequate to observe within the same time window both , the crossover from ballistic to diffusive motion and the crossover from free to correlated motion , as illustrated by the simulation results for @xmath83 in fig .
the reason is , of course , that in this case @xmath86 and @xmath112 do not greatly differ from each other , and hence , there is no important time - scale separation .
the situation is , however , dramatically different in the corresponding colloidal liquid since , as discussed in sect .
[ sectionii ] , the ratio @xmath395 may be as small as @xmath396 .
thus , these two crossover timescales are separated by about 8 decades , and can not be analyzed in the same time window .
in fact , this implies that if we focus on the crossover from ballistic to diffusive motion , as in eq .
( [ wdtintdifecstar ] ) , then the term involving the configurational friction will be completely negligible .
in contrast , if we focus on the crossover from free to correlated motion , as in eq .
( [ wdtintdifecstardiff ] ) , then it is the inertial term involving the time derivative of the msd what can be neglected .
this is referred to as the overdamped limit , which also amounts to ignoring the inertial term @xmath397 $ ] on the left side of the gle in eq .
( [ eq6p ] ) .
thus , in this limit eq .
( [ wdtintdifecstardiff ] ) reads @xmath398 this only changes the true short - time limit @xmath399 of the solution of eq .
( [ wdtintdifecstardiff ] ) to @xmath400 , but leaves unaltered the long - time limit @xmath401 .
this equation thus describes the diffusive motion of colloidal tracer particles .
it should also be clear , however , that even though in atomic liquids there is not an appreciable timescale separation , eq .
( [ wdtintdifecstaroverdamped ] ) also describes the long - time motion of atomic tracer particles .
hence , except for the referred short - time differences , the msd of an atomic and a colloidal liquid with the same interactions and the same @xmath19 should be indistinguishable when plotted in terms of these dimensionless units , provided that the intermediate scattering functions @xmath14 and @xmath15 , which enter in the previous expression for @xmath389 above , also share a similar long - time scaling property .
thus , we may embark on a study of these dynamic properties to see if they indeed exhibit the desired scalings , or else , we can check directly if the msd itself exhibits the expected universality of atomic and colloidal liquids .
here we have adopted the second approach , and have compared the molecular and the brownian dynamics simulation data of @xmath83 for the hs system at the volume fractions @xmath97 and 0.5 . for this comparison
we use the same molecular dynamics data as in fig .
[ fig1 ] , now plotted in terms of the dimensionless msd @xmath402 $ ] as a function of the dimensionless time @xmath403 $ ] , and the brownian dynamics data generated for this comparison using the methodology explained in ref .
@xcite . as we can see from this comparison , for each volume fraction the molecular dynamics and
the brownian dynamics data agree at long times , within a high degree of numerical precision .
the short - time difference between the molecular and brownian dynamics simulation data originates , of course , from the fact that the latter are based on the conventional ermak and mccammon s brownian dynamics algorithm @xcite , in which the overdamped " limit is previously taken in the microscopic equations of motion .
thus , this agreement is also an indirect indication that the intermediate scattering functions @xmath14 and @xmath15 must also share similar scaling properties . the analysis of this issue , however , is addressed separately @xcite .
$ ] as a function of the dimensionless time @xmath404 $ ] for the volume fractions @xmath97 and 0.5 .
the squares correspond to brownian dynamics simulations @xcite generated as described in ref .
@xcite , and the circles to the molecular dynamics simulations of fig . 1 . ]
in summary , in this paper we have explained a simple argument that exposes a dynamic equivalence between the long - time dynamic properties of atomic and colloidal liquids .
such simple arguments were complemented by a more formal fundamental derivation of the generalized langevin equation for a tracer particle in an atomic liquid , which is the atomic counterpart of the gle for tracer diffusion derived in ref .
@xcite for colloidal liquids in the absence of hydrodynamic interactions . the dynamic equivalence suggested by the fact that the gle for tracer diffusion in both cases has the same mathematical structure , need in reality that other dynamic properties , such as the intermediate scattering functions @xmath14 and @xmath15 , also share a similar long - time scaling property . for the time being , here we have tested the predicted dynamic equivalence at the level of the mean squared displacement @xmath83 in the context of a specific model system , namely , the hard sphere liquid , in its dynamic version corresponding to molecular and brownian dynamics .
let us state , however , that the present work does not settle the question of the generality of this dynamic equivalence .
instead , it only contributes to stimulate the corresponding discussion .
for example , it is important to discuss the manifestation of this dynamic equivalence on properties other than the msd . as indicated above , verifying that similar scalings are exhibited by the intermediate scattering functions @xmath14 and @xmath15 is an issue that must still be addressed in detail .
in fact , our group has already approached this issue within the gle formalism @xcite , and the results turn out to be completely consistent with those of the present paper .
the atomic - to - brownian long - time dynamic equivalence thus seems to be a very robust prediction .
the most relevant implications of this dynamic equivalence have been corroborated by the systematic comparisons between molecular and brownian dynamics simulations of the sort illustrated in this paper .
a summary of this analysis has been advanced in a recent brief communication @xcite .
another important issue refers to the actual universality of the atomic - colloidal dynamic equivalence discussed in this paper , since the only validation of these predictions involved a specific model system , namely , the hard sphere fluid . in this direction , let us mention that the present colloidal atomic dynamic correspondence is not restricted to the hard - sphere fluid , but it actually extends over to systems with soft repulsive interactions .
this is a direct result of combining the present colloidal atomic correspondence for the hard sphere system , with another important scaling rule , which derives from the principle of dynamic equivalence between soft - sphere and hard - sphere liquids @xcite .
the extension of this scaling to atomic systems is immediate once the collision diameter @xmath61 entering in the expression for @xmath405 in eq .
( [ dkinetictheory ] ) is given a proper definition @xcite for the soft - sphere potential @xmath342 considered .
still another issue refers to the possible limitations of this long - time dynamic equivalence , imposed by the fact that the present derivation apparently assumed colloidal systems in the absence of hydrodynamic interactions . in reality , however , the validity of this dynamic equivalence should extend over to systems with hydrodynamic interactions , provided that the corresponding effects enter only through the value of the short - time self - diffusion coefficient @xmath366 , as suggested in @xcite .
other interesting directions along which to question the applicability and universality of this dynamic equivalence refers to the realm of liquid mixtures and to the effects of attractive interactions .
the answer to these questions , however , will only come from the comparison between the dynamic properties of atomic and colloidal liquids , similar to that presented here in fig .
2 , or by expanding the theoretical analysis that led us to the present proposal .
acknowledgments : the authors are grateful to g. prez - ngel and m. chvez - pez for their advise and assistance with the molecular and brownian dynamics simulations , and to l. yeomans - reyna , for valuable discussions .
we also acknowledge the kind hospitality of the joint institute for neutron sciences ( oak ridge , tn ) , where part of this manuscript was written .
we are grateful to w .- r .
chen and t. egami for stimulating discussions .
this work was supported by the consejo nacional de ciencia y tecnologa ( conacyt , mxico ) through grants 84076 and 132540 and through the red temtica de la materia condensada blanda .
l. lpez - flores , p. mendoza - mndez , l. e. snchez - daz , l. l. yeomans - reyna , a. vizcarra - rendn , gabriel prez - ngel , m. chvez - pez , and m. medina - noyola , _ dynamic equivalence between atomic and colloidal liquids _
, arxiv:1106.2475v1 [ cond-mat.soft ] . | we derive the time - evolution equation that describes the brownian motion of labeled individual tracer particles in a simple model atomic liquid ( i.e. , a system of @xmath0 particles whose motion is governed by newton s second law , and interacting through spherically symmetric pairwise potentials ) .
we base our derivation on the generalized langevin equation formalism , and find that the resulting time evolution equation is formally identical to the generalized langevin equation that describes the brownian motion of individual tracer particles in a colloidal suspension in the absence of hydrodynamic interactions .
this formal dynamic equivalence implies the long - time indistinguishability of some dynamic properties of both systems , such as their mean squared displacement , upon a well - defined time scaling .
this prediction is tested here by comparing the results of molecular and brownian dynamics simulations performed on the hard sphere system . |
the lattice formulation of quantum chromodynamics ( qcd ) is presently considered as the only first principle tool to investigate a transition between the confined and deconfined states of strongly interacting matter .
such a phase transition ( pt ) is also expected in a pure non abelian su(n ) gauge theory which is known as gluodynamics .
the svetitsky - jaffe hypothesis [ ] relates the deconfinement pt in ( d+1)-dimensional su(n ) gluodynamics to the magnetic pt in the d - dimensional z(n ) symmetric spin model .
the key element of this correspondence is that the role of spin in the original su(n ) gluodynamics is played by the so called polyakov loop .
the latter is interpreted as the time propagator of an infinitely heavy static quark . a high level of understanding of the spin systems along with the svetitsky - jaffe hypothesis led to a significant progress in studies of the su(n ) gluodynamics properties in the pt vicinity .
formation of geometrical clusters composed of the polyakov loops is an important feature of the pure gauge theories [ , ] .
a similar phenomenon is well known in spin systems and it is responsible for percolation of clusters . moreover , the deconfinement pt in gluodynamics already was studied within the percolation framework in refs . [ , ] , where the main attention was paid to the largest and the next to the largest clusters whereas the smaller clusters were ignored . however , in many respects the pt details are encoded in the properties of smaller clusters which is well - known after formulating the fisher droplet model ( fdm ) [ ] .
an important finding of the fdm is that at the critical point the size distribution of physical clusters obeys a power law which is controlled by the fisher topological exponent @xmath0 .
hence the value of this exponent is rather important in order to develop a consistent theory of pt in strongly interacting matter and to localize the critical point of qcd phase diagram which is a hot topic of the physics of heavy ion collisions .
therefore , in this work we study the geometrical clusterization in su(2 ) gluodynamics and analyze the properties of clusters of all possible sizes .
such an approach allows us to explain the deconfinement of color charges as a specific kind of the liquid - gas pt [ ] . [ cols="^ " , ]
in this contribution we present a novel approach to study the deconfinement pt in the su(2 ) pure gauge theory in terms of the geometrical clusters composed of the polyakov loops of the same sign .
we demonstrate that the separation of ( anti)clusters into liquid " droplet and gas " of smaller fragments is well justified and reflects the physical properties of the lattice system .
this concept allows us to explain the deconfinement pt as a special kind of the liquid - gas transition .
however , in contrast to the ordinary liquids , the su(2 ) gluodynamics contains two types of liquid whose behavior is drastically different in the region of broken global z(2 ) symmetry .
the cluster liquid droplet evaporates above pt whereas the anticluster liquid droplet experiences the condensation of the accompanying gas of anticlusters .
a successful application of the ldf to the description of the size distributions of gaseous ( anti)clusters is the main result of this study .
surprisingly , even the monomers are qualitatively described by eq .
( [ eqii ] ) .
the fit of the ( anti)cluster size distributions by the ldf formula allows us to determine the @xmath2-dependences of the reduced chemical potential and the reduced surface tension coefficient .
while in the symmetric phase this quantities are identical for fragments of both kinds , their behavior is drastically different in the deconfined phase .
another important finding of this study is a high precision determination of the fisher topological constant @xmath3 which is the same both for clusters and for anticlusters .
this result is in line with the exactly solvable model of the nuclear liquid - gas pt [ ] . at the same time it disproves the fdm prediction that @xmath4 [ ] .
we showed that the reduced surface tension coefficient and the mean size of the largest ( anti)cluster can be used as the new order parameters of deconfinement pt in su(2 ) gluodynamics .
in contrast to the fdm the power law for the size distribution is found only for the gas of clusters and not at the pt point , but at @xmath5 .
the authors thank d. b. blaschke , o. a. borisenko , v. chelnokov , ch .
gattringer , d. h. rischke , l. m. satarov , h. satz and e. shuryak for the fruitful discussions and valuable comments .
the present work was supported in part by the national academy of sciences of ukraine and by the nas of ukraine grant of grid simulations for high energy physics . | the liquid droplet formula is applied to an analysis of the properties of geometrical ( anti)clusters formed in su(2 ) gluodynamics by the polyakov loops of the same sign . using this approach
, we explain the phase transition in su(2 ) gluodynamics as a transition between two liquids during which one of the liquid droplets ( the largest cluster of a certain polyakov loop sign ) experiences a condensation , while another droplet ( the next to the largest cluster of the opposite sign of polyakov loop ) evaporates .
the clusters of smaller sizes form two accompanying gases , which behave oppositely to their liquids .
the liquid droplet formula is used to analyze the size distributions of the gas ( anti)clusters . the fit of these distributions allows us to extract the temperature dependence of surface tension and the value of fisher topological exponent @xmath0 for both kinds of gaseous clusters .
it is shown that the surface tension coefficient of gaseous ( anti)clusters can serve as an order parameter of the deconfinement phase transition in su(2 ) gluodynamics .
the fisher topological exponent @xmath0 of ( anti)clusters is found to have the same value @xmath1 .
this value disagrees with the famous fisher droplet model , but it agrees well with an exactly solvable model of nuclear liquid - gas phase transition .
this finding may evidence for the fact that the su(2 ) gluodynamics and this exactly solvable model of nuclear liquid - gas phase transition are in the same universality class .
+ * kewords : * geometrical clusters , size distributions , liquid droplet model formula , surface tension |
the centaurus cluster ( abell 3526 ) is x - ray bright , being the nearest cluster ( redshift @xmath1 ) with a 210 kev luminosity exceeding @xmath2 .
our earlier 31.7 ks chandra image of the centaurus cluster revealed a complex structure in the innermost few arcmin of the core , centred on the brightest cluster galaxy ngc4696 ( sanders & fabian 2002 ) .
the iron abundance of the gas was found to peak at a radius of about 1 arcmin from the centre .
the temperature drops from 3.5 to about 1 kev over this whole region . a plume - like structure swirls clockwise to the ne beyond which there is an abrupt temperature increase ( i.e. a cold front ) .
the central x - ray emission is surrounded by marked dips in emission , or bubbles , which coincide with the complex radio source ( taylor , fabian & allen 2002 ) .
arcsec . using an image smoothed with a gaussian of @xmath3 arcsec , we subtracted some of the larger scale structure to improve contrast .
various features are marked . ]
previous x - ray observations ( e.g. allen & fabian 1994 ) show a system with smooth , elliptical , x - ray isophotes , indicating the system is relatively relaxed .
however , there is evidence for a current or past merger event ( allen & fabian 1994 ; churazov et al . 1999 ; furusho et al . 2001 ; dupke et al 2001 ) in the form of shifts in x - ray isophote centroids with radius and bulk motions in the x - ray gas .
a neighbouring subcluster , cen 45 centred on ngc4709 which is about 15 arcmin e of ngc4696 , has a velocity which is 1500 higher than the main centaurus cluster , cen 30 ( lucey , currie & dickens 1986 ) .
observations of the centaurus cluster using _ rosat _ and _ asca _ show that the central region of the cluster is particularly rich in metals , with a large abundance gradient ( fukazawa et al .
1994 ; ikebe et al .
1998 ; allen et al . 2001 ) .
cluster cores are in detail complex but provide us with an observable analogue of the cooling and heating processes implicit in the formation of massive galaxies .
the nearness , intermediate temperature , short radiative cooling time and high metallicity make the centaurus cluster an excellent candidate for studying these processes and also the enrichment of the intracluster gas . here
we present images of the centaurus cluster from a recent 200 ks chandra observation .
we adopt @xmath4 which means that one arcsec corresponds to 210 pc at the redshift of the centaurus cluster .
the data presented here are based on _ chandra _ obsids 504 , 5310 , 4954 and 4955 .
obsid 504 was first presented in sanders & fabian ( 2002 ) .
the standard lc_clean tool was used to remove periods in the observations with possible flares yielding a total good time of 199.3 ks .
each of the datasets were reprocessed to have the latest gain file , and time dependent gain correction was applied .
we used standard blank sky observations to create background spectra for use in spectral fitting . a 3-band x - ray image of the central core is shown in fig . 1 ,
with the separate band images in fig . 2 .
the images here have been accumulatively smoothed ( sanders et al in preparation ; smoothing with a circular top hat kernel with radius determined so that the minimum signal to noise is constant ) .
a whole - band image showing the outer parts is in fig . 3 and temperature and abundance maps in fig . 4 .
the distribution of abundances ( scaled to those of anders & grevesse 1989 ) as a function of temperature is shown in fig . 5 .
x - ray and optical images of a disc galaxy seen in x - ray absorption are in fig .
an existing vla 1.4 ghz image has been overlaid on a whole band x - ray image in fig 7 .
the x - ray image has been adaptively - smoothed using an algorithm due to h. ebeling ; features should be significant above the 3-sigma level .
new vla observations were obtained on 2004 october 24 with the vla in its a configuration . at the observed frequency of 326 mhz
this yielded an angular resolution of 17.7 @xmath5 5.6 arcsecond in postion angle @xmath64.3 deg .
the total time on source was 178 min .
the bandwidth used was 12.5 mhz in a 4 if spectral line mode so as to allow for interference excision using the aips task flgit .
calibration and imaging of the data were performed in the standard way within aips .
the final image has an rms noise of 4.8 mjy / beam . this image ( fig
. 8 , top ) compares well with a 1.4 ghz vla image previously published ( fig . 7 ,
see also taylor et al .
2002 ) and shows that the radio emission extends nearly 2 arcmin ( 25 kpc ) to the south of the nucleus . at the extremities the spectrum is quite steep with a power law index of @xmath61.5 ( fig . 8 , lower ) .
strong faraday rotation is observed in the central radio source indicating a mean magnetic field of about @xmath7 at 10 kpc radius ( taylor et al 2002 ) .
this corresponds to a magnetic pressure there which is about 2 per cent of the thermal pressure .
the longer x - ray exposure has enabled us to see more deeply into the cluster core and resolve much more detail .
new features include : @xmath8 filaments extending to the e and ne from the centre are found in the soft x - ray image below 1 kev ( figs .
1 and 2 left ) .
the inner parts of the filaments correspond to the optical filaments and dust lane seen in ngc4696 ( fabian et al 1982 ; sparks , macchetto & golombek 1989 ) .
comparison with new h@xmath9 images of this region will be presented elsewhere ( crawford et al . , in preparation ) .
@xmath8 in the 12 kev band the holes corresponding to the radio lobes are very clear and above 2 kev the rims of these ` holes ' or ` bubbles ' appear bright .
the rims do not appear in projection to be hotter ( fig . 4 ; confirmed by a higher resolution temperature map ) and are therefore not shocked .
this is similar to results on bubbles found in the perseus cluster ( fabian et al 2002 , 2003 ) .
possible weak ripples in the x - ray surface brightness centred on the nucleus can be seen in fig . 3 , particularly on the ne of the e edge ( cf .
fabian et al 2003 ; forman et al 2003 ) .
5.6 arcseconds in position angle @xmath64.3 degrees . the radio spectral index ( defined @xmath10 ) has been computed between 330 and 1565 mhz after convolving the 1565 mhz image to the resolution of the 330 mhz image .
, title="fig : " ] 5.6 arcseconds in position angle @xmath64.3 degrees .
the radio spectral index ( defined @xmath10 ) has been computed between 330 and 1565 mhz after convolving the 1565 mhz image to the resolution of the 330 mhz image .
, title="fig : " ] @xmath8 a loop of gas to the se ( fig . 2 ,
left ) does not coincide with any obvious radio structure ( figs .
7 and 8) and could either surround relativistic plasma where the electrons have aged to the extent that they no longer radiate in the observed bands ( a ghost bubble ) , or could just be a loop of cooler x - ray gas . a slight deficit in this region seen in the harder x - ray bands ( figs
. 1 and 2 ) suggests that it is most likely a ghost bubble
. a larger loop to the ne appears in the adaptively - smoothed image ( fig .
radio emission is leaking into the inner end ( fig .
8) , suggesting that it too is a ghost bubble .
the lop - sided larger scale diffuse emission in the total 0.47 kev band was apparent in the earlier image but is now seen much more clearly .
it shows a semicircular edge to the e at 83 arcsec ( 17.5 kpc ) from the nucleus ( fig .
3 ) and a larger semicircular edge to the w at 153 arcsec ( 32.1 kpc ) . while the edge to the e is centred on the nucleus of ngc4696 , that to the w is centred 52 arcsec ( 10.9 kpc ) to the w of the nucleus , suggesting that the gas there is in motion with respect to the nucleus .
the edges are clearly evident in the projected temperature map ( fig .
4 ) and emphasise that the temperature abruptly jumps there by almost 1 kev .
they appeared in the radial temperature profiles made from the earlier data ( sanders & fabian 2002 ) to both e and w separately .
the simplest explanation for them is as cold fronts , such as seen in other clusters ( e.g. markevitch 2000 ; vikhlinin et al 2002 ) .
separate deprojection of spectra across the edges to the e and w shows that the pressure across the edges is consistent with being continuous .
the abundance drops across both edges ; steeply in the case of the w edge ( fig . 4 )
, unlike the cold front in a496 ( dupke & bregman 2003 ) .
the sharp abundance change is similar to behaviour seen in ngc507 by kraft et al ( 2004 ) , although the edge there coincides with a radio lobe , which is not the case here .
the large extent of the edges argues against them being simple cold fronts . sloshing motions of the gas within the central potential well ( e.g.
markevitch et al 2001 ) could account for them and for the e - w asymmetry of the appearance .
however , the semicircular shape of the eastern one ( see particularly fig .
3 ) suggests that this , at least , may have been shaped by disturbances from the nucleus .
the radio source on the other hand appears to be moving to the n , although that may just be an impression from its outer structure ( fig .
8) which may be following the path of least resistance ( the optical image of ngc4696 shows it significantly flattened in the n - s direction ) .
the temperature and abundance maps ( fig .
4 ) reveal that the core of the cluster consists of at least 4 distinct parts .
they are a ) the region immediately around the nucleus , b ) a high abundance region extending 18 kpc to the end of the eastern plume , c ) a quasi - spherical region of radius 32 kpc with its centre displaced 11 kpc to the w of the nucleus , and d ) the outer region .
each part occupies a different part of the abundance temperature plane ( fig .
5 ) with a ) @xmath11 kev , b ) @xmath12 kev , c ) @xmath13 kev and finally d ) @xmath14 kev . in the present multi - temperature analysis ,
the abundance peaks at @xmath15 in b ) and drops by about @xmath16 between b ) and c ) and c ) and d ) .
there can not be any large turbulent or stochastic motions in the core or these high abundances would have been smoothed out .
following the work of rebusco et al ( 2005 ) , who have estimated the diffusion coefficient @xmath17 in the core of the perseus cluster .
we can obtain a limit from the much higher gradient seen in centaurus .
assuming stochastic diffusion only , then steep gradients over a distance @xmath18 are smoothed out on a timescale of @xmath19 .
if we first look at the outer , western edge and rely only on diffusion to smooth it and adopt a minimum time equal to the local crossing time of the structure of @xmath20 then @xmath21 this is smaller than the result , @xmath22 of rebusco et al ( 2005 ) .
since the region is displaced from the central galaxy it is unlikely that any current enrichment has a noticeable effect .
detailed modelling including metal enrichment should yield a tight constraint from the inner , high abundance , region .
the above constraint from the outer region means that the level of turbulence and stochastic motions in the centaurus cluster is low ( the product of velocity and lengthscale of the motions being @xmath23 ) .
smallscale motions need continual pumping or they would rapidly die out , temporarily heating the gas , as noted by rebusco et al ( 2005 ) . the lowest energy image reveals delicate filaments of x - ray emitting gas which are only 13 arcsec ( 200600 pc ) wide .
the temperature of these structures is about 0.7 kev while the gas they are embedded in has a temperature of about 1.5 kev .
conduction is presumably much reduced and magnetic fields help maintain the integrity of the structures . in the very core around the nucleus , we can now see that the temperature drops down to less than 0.7 kev , with the inner region being multiphase .
the whole temperature range detectable in the centaurus cluster therefore exceeds a factor of 5 , which is significantly larger that the factor of three seen in many other clusters ( peterson et al 2003 ; but see morris & fabian 2004 ) . a disc galaxy to the se of ngc4696 ( at ra 12 49 03.8 , dec -41 20 27 , j2000.0 )
can just be seen in silhouette in fig.3 and , more clearly , in fig .
it has a radial heliocentric velocity of @xmath24 ( dickens , currie & lucey 1986 ) , so lies in the cen30 cluster , and is 322-g93 in the eso / uppsala catalogue .
photoelectric absorption by galaxies projected onto intracluster gas is also seen in the perseus ( gillmon et al 2003 ) and a2029 ( clarke et al 2004 ) clusters .
the inner region is complex with radio bubbles , ghost bubbles and cool filaments .
the frothy x - ray appearance of the centre resembles the centre of the virgo cluster ( young et al 2002 ; forman et al 2003 ) which also has several inner bubbles .
the region beyond is smoother but appears different to the east and west . in those directions
clear semicircular edges are seen .
the one to the e is concentric with the nucleus indicating that it may have been triggered by a disturbance from there .
the larger one to the w is centred west of the nucleus .
its shape suggests that some disturbance from the nucleus may produce it as well .
the western edge appears as a marked abundance drop , which is not easily accounted for , although large buoyant bubbles dragging iron - rich mattter outward , as seen in the perseus cluster ( sanders , fabian & dunn 2004 ) , or fast - moving , outer gas sweeping past as a result of the ongoing merger with cen45 ( which is not otherwise seen ) , could contribute .
the dense , metal - rich centre of the centaurus cluster is revealed as more complex than in previous observations .
the radio source clearly has an impact immediately around the centre in the frothy bubble structure and the eastern edge , although the long - term nature of the interaction and its contribution to the energy balance of the central intracluster medium is elusive .
given that in equilibrium temperature contours follow equipotentials , the asymmetric temperature structure around the central galaxy ngc4696 means that some major part of the mass or gas distribution there is dynamically changing .
the national radio astronomy observatory is operated by associated universities , inc .
, under cooperative agreement with the national science foundation .
gbt acknowledges support for this work from the national aeronautics and space administration through chandra award number go4 - 5135x issued by the chandra x - ray observatory center , which is operated by the smithsonian astrophysical observatory for and on behalf of the national aeronautics space administration under contract nas8 - 03060 .
gbt also thanks the institute of astronomy for hospitality while working on this project .
acf and swa thank the royal society for support . | x - ray images and gas temperatures taken from a deep @xmath0200 ks chandra observation of the centaurus cluster are presented .
multiple inner bubbles and outer semicircular edges are revealed , together with wispy filaments of soft x - ray emitting gas .
the frothy central structure and eastern edge are likely due to the central radio source blowing bubbles in the intracluster gas .
the semicircular edges to the surface brightness maps 32 kpc to the east and 17.5 kpc to the west are marked by sharp temperature increases and abundance drops .
the edges could be due to sloshing motions of the central potential , or are possibly enhanced by earlier radio activity .
the high abundance of the innermost gas ( about 2.5 times solar ) limits the amount of diffusion and mixing taking place .
ps .
= -0.4 in x - rays : galaxies galaxies : clusters : individual : centaurus intergalactic medium |
the class of b[e ] supergiants consists of about 20 luminous evolved b stars with a rich emission line spectrum and a strong infrared excess ( zickgraf et al .
1986 , lamers et al .
most of the confirmed members of this class are located in the magellanic clouds , mainly for two reasons : the luminosities of the galactic objects can not be precisely determined due to the uncertain distances , and the difficulty to resolve the objects of this class from other b - type emission line stars ( be stars , herbig be stars , and other types of b[e ] stars ) .
gummersbach et al .
( 1995 ) were able to place 14 magellanic cloud b[e ] supergiants in the hr diagram .
there , they appear to define two distinct groups , one at relatively low luminosity ( @xmath4 ) and low effective temperature ( @xmath5k ) , and the other at higher luminosities ( @xmath6 ) and temperatures ( @xmath7k @xmath8k ) . the spectral properties of the b[e ] supergiants are best explained within the model by zickgraf et al .
( 1985 ) , who propose that these stars have a two component wind : a fast polar wind responsible for the high excitation uv resonance lines , and an equatorial slow and cool wind producing the narrow permitted and forbidden lines .
the equatorial wind is associated with the highest mass - loss rate and usually identified with an outflowing disk where dust can condense and emit at infrared wavelengths .
such disk might be produced by rotational wind compression ( bjorkman & cassinelli 1993 , bjorkman 1999 ) .
nevertheless , it remains to be shown that disk inhibition due to non - radial components of the wind driving line force and gravity darkening ( owocki & gayley 1998 ) can be overcome , perhaps by a combination of rotational compression and wind bi - stability proposed by lamers & pauldrach ( 1991 ) which predicts a sudden increase in the wind mass flux and decrease in the wind velocity at a critical temperature ( @xmath9k ) when the stellar surface temperature decreases gradually from the pole towards the equator .
langer & heger ( 1998 ) have connected the b[e ] supergiant stage with phases in the evolution of rotating massive stars during which the star can possibly reach the @xmath10-limit , i.e. its surface rotation rate ( which also takes into account the radiation force of the star ) is able to destabilize the stellar surface at the equator ( langer 1997 ) .
they found that the most luminous and hot b[e ] stars might be related to core hydrogen burning models which arrive at the @xmath10-limit due to increasing surface opacities during their main sequence evolution , which is possible for stars relatively close to the eddington - limit even if they are slow rotators ( langer 1998 ) .
they proposed further that stars below @xmath11 could reach the @xmath10-limit during core helium burning ( on the so called blue loops ) due to efficient angular momentum transport from the stellar interior to the stellar surface during this phase ( heger & langer 1998 ) .
finally , the outbursts of luminous blue variables have been associated with these stars hitting the @xmath10-limit ( langer 1997 , langer et al .
1999 ) , a conjecture which is strongly supported by the bi - polarity of virtually all circumstellar nebulae of lbvs ( nota et al .
1995 ) . whether all massive stars go through a b[e ] supergiant stage , and whether they are connected to luminous blue variables is unclear
empirically , the distribution of the group of luminous b[e ] supergiants in the hr diagram overlaps with that of the lbvs ( bohannan 1997 ) .
a connection between b[e ] supergiants and lbv stars has been early suggested by shore ( 1990 ) and schulte - ladbeck & clayton ( 1993 ) from their analysis of s22 , in the large magellanic cloud .
classified as a b[e ] supergiant by zickgraf et al .
( 1986 ) , s22 shows an intrinsic polarization of 0.52 @xmath12 due to electron scattering in an aspherical wind . the polarization degree is variable and this is probably linked to variations in the mass - loss rate of the star ( schulte - ladbeck & clayton 1993 ) .
a similar result has been found for the galactic lbv hr carinae , which is characterized by an intrinsic continuum polarization of about 0.4@xmath12 , possibly variable ( clampin et al . 1995 ) .
this can again be explained as due to a non - spherical wind geometry ( the presence of a circumstellar disk has been also discussed by nota et al .
1997 ) and a time dependent mass loss rate .
in addition , shore ( 1990 ) has detected almost a factor of two variation in the uv flux of s22 longward of 1600 and a factor between 2 and 3 variation shortward of 1600 .
the amplitude of the uv variability is quite similar to that observed in lbvs during their shell ejection phase ( pasquali & nota 1999 ) . as an alternative approach , to study the occurrence of the lbv phase in the evolution of massive stars , we have undertaken a longslit spectroscopy campaign of galactic and mc evolved supergiants whose stellar properties ( m@xmath13 and log t@xmath14 ) are in the range set by confirmed lbvs .
the aim of the observations is to detect the presence of circumstellar nebulae and to determine whether these are ejected by the star and possibly establish an evolutionary connection with lbvs . here , we present the first results obtained for the r4 , in the small magellanic cloud . with @xmath15 and t@xmath16k ( zickgraf et al .
1996 ) , r4 is the hottest and least luminous star within the high luminosity group of b[e ] supergiants .
zickgraf et al .
showed that r4 is a spectroscopic binary ( a = 23 a.u . )
comprising a b[e ] supergiant with spectrophotometric variability characteristic of a lbv , and an evolved a type companion star which is about 10 times less luminous ( @xmath17 ) . in sect . 2 we present the spectroscopic data taken for r4 , while in sect .
3 we describe the results obtained from our observations .
a discussion of our results , of the implications for the evolutionary history of r4 , and of the connection of b[e ] supergiants with lbvs follows in sect . 4 .
we observed r4 at the 3.5 m eso / ntt telescope , on the nights of july 27 - 30 , 1998 .
we used the emmi spectrograph in the remd ( red medium dispersion ) configuration and acquired longslit spectra through gratings 6 and 7 in the following wavelength ranges : 4470 - 5820 , 5080 - 6420 and 6240 - 6870 .
we observed in three different positions : on the star , at 3@xmath0 north and 3@xmath0 south , respectively .
these two offsets were computed by translating the typical linear sizes of known lbv nebulae into angular sizes at the smc distance ( 57.5 kpc , van den bergh 1989 ) , in order to be able to detect a circumstellar nebula , if present , when no high - resolution imaging could be performed .
we employed a 1@xmath0 x 180@xmath0 longslit oriented east - west in all the cases but in the 4470 - 5820 spectrum , for which the slit had been oriented north - south .
the complete journal of the observations is reported in table 1 , where the wavelength range , the dispersion and effective resolution , the exposure time and the offset are summarized for each spectrum .
the red arm of emmi was equipped with a tk2048eb4 ccd , whose spatial scale is 0.27@xmath0 per pixel .
we obtained the usual set of bias and flat - field images , together with comparison spectra of hear for the wavelength calibration .
the spectra were first cleaned to remove cosmic rays and bad columns , and then corrected for bias and flat - fielded .
their subsequent reduction was performed following the procedure outlined in the iraf longslit package .
we modeled the sky background in each frame by fitting a surface described by chebyshev polinomials of low order in x and y. this surface was then subtracted from the frame itself .
the wavelength calibration was achieved in three steps : first , we identified the emission lines at the central row of the comparison spectrum and derived the dispersion correction by fitting a chebyshev function of order 3 .
the same lines were then reidentified across the entire frame adopting a step of 2 rows and readjusting the dispersion correction if necessary .
finally , the dispersion corrections obtained through the frame were fitted into a surface using chebyshev polynomials of order 5 .
this surface was then applied to each spectrum for the final wavelength calibration .
the same procedure was of course repeated for each grating .
the effective resolution is reported in column 5 of table 1 for each grating : a fwhm = 2.72 and 2.36 corresponds to a velocity resolution of 84 and 61 km s@xmath1 at h@xmath18 and [ nii]@xmath195755 respectively , while a fwhm of 1.12 defines a velocity resolution of 25 km s@xmath1 at h@xmath20 .
measurements of the sky lines indicate that the wavelength calibration is certain at @xmath21 10 km s@xmath1 at any position within the longslit spectra .
the long - slit spectra of the r4 region show the presence of extended nebular lines , such as [ nii ] @xmath226548 , 6584 and [ sii ] @xmath226717 , 6731 .
these lines show a well defined spatial extension and velocity structure , clearly distinguishable from the underlying emission due to the local interstellar medium .
we show in figure 1 the [ nii ] @xmath196584 line as detected at the position 3@xmath0 north with respect to the star ( top panel ) , and 3@xmath0 south ( bottom panel ) . the spatial extent of the nebular emission is approximately 32 pixels , which corresponds to an angular size of 8.6@xmath0 ( given the emmi / ccd spatial scale of 1 pixel = 0.27@xmath0 ) and to a linear size of 2.4 pc , in the assumption of a distance for the smc of 57.5 kpc ( van den bergh 1989 ) .
the peculiar velocity structure of the nebular lines can be immediately noted in a quick look inspection of the two - dimensional spectra : while at 3@xmath0 north the bulk of the emission is redshifted , at 3@xmath0 south the line is blueshifted .
a remarkable symmetry is present , both along the spatial direction ( north compared with south ) , and in the velocity structure ( redshifted versus blueshifted ) .
we used the [ nii]@xmath196584 line to derive the nebular radial velocity map at the two offset pointings .
we binned the spectra by a factor of 2 along the spatial axis ( 2 pixels corresponding to 0.54@xmath0 ) and extracted an individual spectrum from each bin .
we measured the peak wavelengths in the [ nii ] profile by multi - gaussian fitting and computed the corresponding radial expansion velocities .
our fits are characterized by a typical error of @xmath21 4 km s@xmath1 . in this wavelength region ,
the spectra have a velocity resolution of @xmath23 25 km s@xmath1 and the absolute wavelength calibration is within @xmath21 10 km s@xmath1 across the entire slit .
the derived radial velocities , corrected for the heliocentric motion , have been plotted in figure 2 as a function of distance from the star , in arcseconds , for both slit positions ( 3@xmath0 north - top panel ; 3@xmath0 south - bottom panel ) . on the abscissae , east is to the left and west is to the right .
the star is at position 0 . from figure 2
it is clear that the local interstellar medium dominates the velocity distribution at distances larger than @xmath245@xmath25 from the star , since we measure the mean radial velocity of this overall motion to be 110 km s@xmath1 , in agreement with the fabry - perot h@xmath20 observations of le coarer et al .
however , in the distance range -5@xmath26 d @xmath27 from the star , two additional components are clearly resolved : at the position 3@xmath0 north we find a component which is redshifted , and spans between 200 and 280 km s@xmath1 , covering the entire spatial range . at the position 3@xmath0 south , a second component is also present , which varies between 20 and 90 km s@xmath1 .
this component extends between 4@xmath0 e and 4@xmath0 w from the star . the remarkable symmetry of these radial velocity structures can be better appreciated in figure 3 ( top panel ) where we have plotted on the same spatial scale the two velocity profiles obtained at the two positions .
first , we notice that the northern velocity distribution mirrors the southern .
the n component displays two radial velocity maxima at @xmath23 2.5@xmath0 e and @xmath23 1.5@xmath0 w. in correspondence to the same two positions , the s component shows two radial velocity minima .
the n component reaches radial velocity minima in correspondence to the star and at the outer boundaries of its spatial extension .
the s component follows a symmetrical trend .
the peak - to - peak radial velocity amplitude of both components is very similar ( @xmath23 80 km s@xmath1 ) .
in addition , there is symmetry between the e and w regions of the nebula with respect to the central star .
indeed , the eastern portion of both velocity curves , when folded , significantly overlaps the western . the observed velocity structure is likely to be indicative of a nebula surrounding r4 .
the nebula is dynamically associated with the central star , for which zickgraf et al .
( 1996 ) determined a radial velocity of 147 km s@xmath1 , and most likely has been ejected by r4 in a previous phase .
therefore , with respect to the central star , the northern component of the nebula turns out to be red - shifted by 84 km s@xmath1 while the southern is blue - shifted by 118 km s@xmath1 on average . assuming a mean expansion velocity of 100 km s@xmath1 and a full linear size of 2.4 pc , we derive a dynamical age of the nebula of @xmath28 1.2 @xmath2 10@xmath3 yr . without a direct image , and on the basis of kinematics considerations alone
, it is difficult to make definite conclusions on its structure .
compared with the information available on nebulae around lbvs , the nebula around r4 appears more complicated in nature . from the kinematics
, it is fair to conclude that : * the nebula is _ not _ a simple expanding shell .
an expanding shell would result in a radial velocity map which has maximum dispersion in correspondence to the position of the star , and minimum velocity at the shell boundaries ( eg .
ag car : smith 1991 , nota et al .
* the nebula is not strictly bipolar .
compared with the radial velocity maps derived for hr car , a prototypical bipolar outflow ( nota et al .
1997 ) , the situation is very different : in the case of hr car , there is a clear demarcation between the two sides of the bipolar outflow , with the redshifted region limited to the nw quadrant , and the blueshifted to the se quadrant ( see figure 9 in nota et al .
1997 ) . in order to provide a consistent explanation for the radial velocity maxima and minima observed in the r4 , a more complicated structure needs to be invoked , which is symmetrical around two axes . in the bottom panel of figure 3 ,
we provide a cartoon of one possible structure , in which a cloverleaf morphology is aligned with the peculiar radial velocity features . in this proposed structure ,
the outflow occurs in four directions along two axes , perpendicular to each other , and oriented at a pa @xmath23 45@xmath29 .
this complicated structure , although speculative , has been observed in planetary nebulae . only a direct image will confirm whether such speculation is correct .
in addition to the kinematical properties , the nebular lines provide also some information on the chemical composition of the nebula . we have computed the line ratio [ nii]6584/[sii]6717 for the r4 nebula and the local interstellar medium from the long - exposure spectra which we corrected for atmospheric extinction .
we derived a ratio [ nii]6584/[sii]6717 of about 3 ( in agreement also with the data of zickgraf et al .
1996 ) and 0.3 for the r4 nebula and the local interstellar medium , respectively .
such result is inconsistent with the same line ratio measured in hii regions and sn remnants in the smc by russell & dopita ( 1990 ) .
typically , the [ nii]6584/[sii]6717 ratio is 0.6 for the hii regions ( with one exception : n84c is characterized by a value of 2.4 ) and varies between 0.2 and 0.4 in the case of snr , independently of the position in the galaxy .
a [ nii]6584/[sii]6717 value of 0.3 is considered to reflect the intrinsic lower n content of the smc with respect to the galaxy and the lmc .
we may then conclude that the r4 nebula is n - enriched by a factor 10 with respect to both the local interstellar medium and hii regions / sn remnants .
> from section 3 , we conclude that r4 is surrounded by a bipolar circumstellar nebula , nitrogen enriched , with a dynamical age of @xmath28 1.2 @xmath2 10@xmath3yr .
this nebula appears to have been ejected from the central star , and its morphological , kinematic and chemical properties are comparable to the average properties of lbv nebulae ( cf . , nota et al . 1995 ) .
although esteban & fernandez ( 1998 ) have detected a circumstellar nebula around the galactic b[e ] star mcw@xmath30137 which appears not to be chemically enriched , our findings for r4 provide the first evidence for an ejected nebula around a b[e ] supergiant . in order to investigate the implications of our finding for
the connection of b[e ] supergiants and lbvs , let us recall the proposed evolutionary scenarios for the b[e ] supergiants .
table 3 summarises the expected stellar and circumstellar nebula properties of b[e ] supergiants according to the very massive main sequence star scenario , the blue loop scenario and the binary merger scenario ( langer & heger 1998 ) . in the first scenario ,
the star reaches the @xmath10-limit ( cf .
section 1 ) during its main sequence evolution due to its high luminosity , i.e. its proximity to the eddington limit ( langer 1998 ) .
this appears possible only for the most massive stars . since at low metallicities
the eddington - limit is even higher at higher luminosity than for young stars in the solar neighborhood ( ulmer & fitzpatrik 1998 ) , this scenario would require an extraordinarily fast rotation to apply to the case of r4 . in the second scenario ,
the star reaches the @xmath10-limit on a blue loop evolving off the hayashi line during core helium burning ( heger & langer 1998 ) .
this scenario can not apply to r4 since stars on blue loops do not exceed effective temperatures of @xmath9k ( cf .
langer & maeder 1995 ) , i.e. , the blue loops never extend to the main sequence band . in the third scenario ,
an equatorial disk or ring is created by mass overflow through the second lagrangian point in a close binary system in the course of a close binary merger .
this scenario was supposed to be most appropriate for the case of r4 by langer & heger ( 1998 ) for the following reasons .
the b[e ] supergiant r4 has an evolved a type companion star with a mass of about 12.9@xmath31 ( zickgraf et al .
1996 ) . while the b[e ] star mass has been determined to a similar value ( @xmath32 ) its bolometric luminosity outshines that of the a star by a factor of @xmath33 .
zickgraf et al .
conclude from the high luminosity of r4 ( @xmath34 ) , and from its strong surface enrichment in cno processed material , that it must have lost large amounts of mass in a previous red supergiant phase .
however , as noted by langer & heger ( 1998 ) , even if a @xmath35 red supergiant at the very low metallicity of the smc had lost about @xmath36 via its stellar wind ( which appears unlikely for several reasons ; e.g. , the large envelope mass of the progenitor of supernova 1987a ) , there would remain a basic puzzle in the r4 binary system .
the a star is clearly beyond core hydrogen exhaustion : how then can the b star have an _ evolved _ companion which is ten times less luminous ?
the b star should have long exhausted its fuel and exploded as a supernova . clearly , binary mass transfer must have occured in this system .
now , the orbital separation in the system is presently about 23 a.u . or 5000@xmath37 ( zickgraf et al .
1996 ) , which may be too large to allow any mass transfer from the a star to the b star .
also , mass transfer in very wide systems is supposed to be unstable and would not leave the two stars at a large separation ( e.g. , podsiadlowski et al .
1992 ) , and can therefore be excluded . a viable binary scenario for r4
may be that the b[e ] star has been formed by a recent binary merger , and the a star is not involved in any mass transfer but only serves as a suitable clock ( langer & heger 1998 ) .
wellstein et al .
( 2000 ) find in a parameter study of binary evolution models that a system which starts out with a 12@xmath31 and a 11@xmath31 star on a 40 day orbit would evolve into contact after core hydrogen exhaustion in the 12@xmath31 star , which leads to mass overflow through the outer lagrangian point @xmath38 and then most likely to a merger .
the @xmath38 overflow , which is likely to comprise several solar masses of material ( wellstein et al .
2000 ) , gives rise to a circumstellar nebula , which is then shaped by the ensueing b star wind , like the homunculus nebula around @xmath39 carinae in the wind interaction scenario of langer et al .
the large amount of angular momentum in the stellar merger remnant may lead to a disk wind according to the bjorkman - cassinelli mechanism , which then is responsible for the b[e ] morphology of the stellar spectrum .
furthermore , the merger star would be expected to contain an overly large helium fraction in its interior which would give it an unusually large l / m - ratio and a surface strongly enriched in cno products . that all these details are observational facts makes r4 the strongest massive star candidate for a binary merger .
within this picture , the system started out as a triple system with three very similar stars , a close pair of , say , 12@xmath31 and 11@xmath31 , in a wide orbit with a @xmath40 star . soon after the latter has evolved off the main sequence into an a type supergiant
, the close pair would merge to form the b[e ] supergiant , about 1.2 @xmath2 10@xmath3 yr ago . in order to understand the relevance of r4 for massive stars in general
, we should establish whether r4 is a peculiar or a typical b[e ] supergiant .
the latter case would have the dramatic implication that most b[e ] supergiants might be the result of a binary merger . however , there are at least two arguments against this proposition .
first , r4 has an extreme location in the hr diagram compared to all other b[e ] supergiants in that its location is by far the closest to the main sequence band .
zickgraf et al .
( 1996 ) compared it with evolutionary tracks of charbonnel et al .
( 1993 ) , where r4 falls exactly on the terminal age main sequence of a 20@xmath31 track .
second , r4 is today the only b[e ] supergiant with an ejected circumstellar nebula . before generalizing the evolutionary scenario of its progenitor
, one would certainly want to detect more examples with ejected nebulae . on the other hand , the similarity of the properties of the r4 nebula and that of lbv nebulae in general which provides a remarkably homogeneous class is striking .
these nebulae are all very similar in terms of morphological and physical properties .
they are all typically 1 parsec in size , with morphologies which are mildly to extremely bipolar .
they expand in the surrounding medium with velocities of the order of 50 100 km s@xmath1 .
their size and expansion velocities identify dynamical timescales which are of the order of several thousand of years .
densities are generally found to be low ( 500 - 1000 @xmath41 ) and so are temperatures , found in the range 5000 - 10000 k. in terms of overall physical and chemical properties , the nebula surrounding r4 would fit this category well . however , two facts preclude the possibility that the r4 nebula and all lbv nebulae have been formed the same way .
the first is r4 s location in the hr diagram , which of all the stars in the group of luminous b[e ] supergiants is farthest away from the lbv regime .
the second , even stronger argument is the fact that some lbv nebulae appear to have multiple shells ( nota et al .
2000 ) and thus the central stars most likely experienced multiple outbursts , which appears not possible within the binary merger scenario . although we argue that not all b[e ] supergiants are formed by stellar mergers , we conclude that at least two intrinsically very different nebula formation mechanisms , the binary merger and the single star lbv outburst mechanism , can produce nebulae with very similar properties . seemingly , the prerequisites for the nebular structure are the same in both cases , i.e. a massive disk ejected by the central star which is then shaped by its strong wind . while the disk forms through the @xmath38-roche lobe overflow in the case of the merger ( hardcore hydrodynamic calculations for such events exist so far only for low mass stars , which , however , support the general idea
yorke et al .
1995 ) , it may be formed through rotational wind compression ( bjorkman & cassinelli 1993 ) in the single star case ( cf .
langer et al . 1999 ) .
our longslit spectroscopic observations reveal that r4 is embedded in a circumstellar nebula whose full spatial extension is 8.6@xmath0 , i.e. 2.4 pc at the smc distance of 57.5 kpc ( van den bergh 1989 ) .
the emission lines show a significant structure in velocity indicating that the nebula is expanding at 100 km s@xmath1 on average .
this implies a dynamical age of 1.2 @xmath2 10@xmath3 yr .
the radial velocity maps obtained for the nebula at 3@xmath0 north and 3@xmath0 south from the central star are characterized by two velocity maxima and two velocity minima , respectively , located at the same positions with respect to the star .
hence , the two velocity distribution appear to be symmetric not only in the velocity field but also along the e - w direction ( cf . figure 3 ) .
such a symmetry excludes that the r4 nebula is either a simple expanding shell ( i.e. ag car ) or a simple bipolar outflow ( i.e. hr car ) .
it rather suggests a cloverleaf morphology , where the outflow occurs in four directions along two axes , perpendicular to each other and oriented at pa @xmath23 45@xmath42 .
since the line flux ratio [ nii]6584/[sii]6717 is almost independent of electron temperature and density , it can be used to estimate any overabundance of nitrogen with respect to the `` unprocessed '' sulphur .
the nebular [ nii]6584/[sii]6717 ratio turns out to be 3 against a value of 0.3 as measured for the local interstellar medium in the same observed spectra , and a value between 0.6 and 0.2 as derived for hii regions and sn remnants in the smc by russell & dopita ( 1990 ) .
this factor of 10 discrepancy clearly indicates that the r4 nebula is nitrogen enriched and therefore , since it is also kinematically associated with the central b[e ] supergiant , it is an ejected nebula .
_ r4 is surrounded by a bipolar and n - enriched nebula , ejected from the central star , whose morphological , kinematical and chemical properties well compare with the mean properties of lbv nebulae . _ we have shown that the central star most likely formed through a binary merger , as proposed by langer & heger ( 1998 ) .
this makes r4 the strongest observational counterpart of such event among massive stars , which has long been sought for .
e.g. , in a comprehensive binary and single star evolution and population sysnthesis study , podsiadlowski et al .
( 1992 ) conclude that @xmath43% of all massive binaries undergo a merging process , most likely so just after the initially more massive star has terminated core hydrogen burning . in section 4.2 , we concluded from the properties of r4 that two distinct mechanisms can form circumstellar nebulae of exactly that type found around lbvs .
this has two consequences .
first , it implies that some lbv outbursts and nebulae may in fact be due to the merging process of two massive stars .
the predicted amount of expected binary mergers ( see above ) implies that this is in fact likely .
second , it may deepen the understanding of why bipolar circumstellar nebulae are such a frequent phenomenon .
e.g. , what holds for massive stars may be true for the progenitors of planetary nebulae , and in the end the dispute of whether bipolar planetaries are formed through binaries ( e.g. , soker 1998 ) or single stars ( e.g. , garca - segura et al .
1999 ) may end in a draw .
we conclude by urging for imaging observations of the r4 nebula
. they will not only provide for the first time the morphological details of a nebula ejected by a massive stellar merger and will thereby allow to constrain the hydrodynamical processes at work in such phenomenon , but it will perhaps reveal clues of how to empirically discriminate binary and single star ejection mechanisms and thus allow for a better understanding of bipolar circumstellar nebulae in general .
the authors wish to thank guillermo garca - segura and the staff of the tonantzintla institute for their hospitality during their visit where this project was perceived .
ap , an and mc would like to thank stsci and st - ecf for their hospitality during their visits where the paper was finalized . nl acknowledges supporting from the deutsche forschungsgemeinschaft through grants la 587/15 - 1 and 16 - 1 .
rsl acknowledges funding from the hst grant to go proposal 6540 .
bjorkman , j.e .
1999 , in variable and non - spherical stellar winds in luminous hot stars , eds .
b. wolf , o. stahl & a.w .
fullerton , lecture notes in physics ( springer : heidelberg ) , p. 121 bjorkman , j.e . ,
cassinelli , j.p .
1993 , , 409 , 429 bohannan , b. 1997 , in luminous blue variables : massive stars in transition , eds .
a. nota & h.j.g.l.m .
lamers , asp conf .
ser . , 120 , p. 3
charbonnel , c. , meynet , g. , maeder , a. , schaller , g. , schaerer , d. 1993 , a&as , 101 , 415 clampin , m. , schulte - ladbeck , r.e . , nota , a. , robberto , m. , paresce , f. , clayton , g.c .
1995 , apj , 448 , 788 esteban , c. , fernandez , m. 1998 , mnras , 298 , 185 garca - segura , g. , langer , n. , ryczka , n. , franco j. 1999 , apj 517 , 767 gummersbach , c.a .
, zickgraf , f .- j . ,
wolf , b. 1995 , a&a , 302 , 409 heger , a. , langer , n. 1998 , a&a 334 , 210 lamers , h.j.g.l.m .
, pauldrach , a.w .
1991 , a&a , 244 , l5 lamers , h.j.g.l.m . ,
zickgraf , f .- j .
, de winter , d. , houziaux , l. , zorec , j. 1998 , a&a , 340 , 117 langer , n. 1997 , in luminous blue variables : massive stars in transition , eds .
a. nota & h.j.g.l.m .
lamers , asp conf .
ser . , 120 , p. 83
langer , n. 1998 , a&a 329 , 551 langer , n. , maeder a. 1995 , a&a 295 , 685 langer , n. , heger , a. 1998 , in b[e ] stars , eds .
a.m. hubert & c. jaschek , assl , 233 , p. 235
langer , n. , garca - segura , g. , mac low , m .-
1999 , apj 520 ,
l49 le coarer , e. , rosado , m. , georgelin , y. , viale , a. , goldes , g. 1993 , a&a , 280 , 365 nota , a. et al .
2000 , in prep .
nota , a. , leitherer , c. , clampin , m. , greenfield , p. , golimowski , d.a .
1992 , , 398 , 621 nota , a. , livio , m. , clampin , m. , schulte - ladbeck , r.e .
1995 , , 448 , 788 nota , a. , smith , l.j . ,
pasquali , a. , clampin , m. , stroud , m. 1997 , , 486 , 338 owocki , s.p . ,
gayley , k.g .
1998 , in boulder - munich ii : properties of hot , luminous stars , ed .
howarth , asp conf .
, 131 , p. 237 pasquali , a. , nota , a. 1999 , mem sait , in press podsiadlowski ph . ,
joss p.c . ,
hsu j.j.l .
1992 , apj 391 , 246 russell , s.c . , dopita , m.a .
1990 , , 74 93 schulte - ladbeck , r.e . , clayton , g.c .
1993 , aj , 106 , 790 shore , s.n .
1990 , iau circ .
, 5005 , 1 smith , l.j .
1991 , in iau symp .
143 , wolf - rayet stars and interrelations with other massive stars in galaxies , eds .
k.a . van der hucht & b. hidayat ( dordrecht : kluwer ) , p. 385 soker , n. 1998 , apj 496 , 833 ulmer , a. , fitzpatrick , e.l .
1998 , apj 520 , 200 van den bergh , s. 1989 , a&ar , 1 , 111 wellstein s. , braun h. , langer n. 2000 , in prep . for a&a yorke h.w .
, bodenheimer p. , taam r.e .
1995 , apj 451 , 308 zickgraf , f .- j . ,
wolf , b. , stahl , o. , leitherer , c. , klare , g. 1985 , a&a , 143 , 421 zickgraf , f .- j . ,
wolf , b. , stahl , o. , leitherer , c. , appenzeller , i. 1986 , a&a , 163 , 119 zickgraf , f .- j .
, kovacs , j. , wolf , b. , stahl , o. , kaufer , a. , appenzeller , i. 1996 , a&a , 309 , 505 c c c c c c c c july 30 & 7 & 4470 - 5820 & 0.66 & 2.72 & 1800 & on the star july 31 & 7 & 5080 - 6420 & 0.66 & 2.36 & 1500 & 3@xmath0 north `` & 6 & 6240 - 6870 & 0.31 & 1.12 & 60 & 3@xmath0 north '' & `` & '' & `` & '' & 1800 & `` '' & `` & '' & `` & '' & 80 & on the star `` & '' & `` & '' & `` & 60 & 3@xmath0 south '' & `` & '' & `` & '' & 1800 & " c c c c & & & 10@xmath30000@xmath30k ... @xmath44 & @xmath45k & @xmath46k & 30@xmath30000@xmath30k & & & luminosity & @xmath47 & @xmath48&@xmath49 & & & time scale & some 10@xmath50yr & some 10@xmath3yr & ( ? ) & & & time integr . & @xmath51 & @xmath52 & @xmath51 ( ? ) disk mass & & & & & & nebular chem . &
non - enriched to & moderately & strongly composition & moderately enriched & enriched & enriched & & & | we present new , ntt longslit spectroscopy of the b[e ] supergiant in the binary system r4 in the small magellanic cloud .
the data show extended , forbidden n and s emissions which are typical signatures of circumstellar matter .
their extension along the space axis of the slit defines an angular size of 8.6@xmath0 which translates into a linear size of 2.4 pc .
the n emission lines also show the velocity structure of a bipolar outflow expanding at 100 km s@xmath1 on average .
this implies that , for a measured radius of 1.2 pc , the outflow originated about 1.2 @xmath2 10@xmath3 yr ago . the line flux ratio [ nii]6584/[sii]6717 indicates that the nebula is nitrogen enriched and therefore it has been ejected from the central star .
this is the first bipolar , ejection nebula detected around a well - established b[e ] supergiant .
the bipolar morphology and the chemical enrichment shown by the nebula associated with r4 are consistent with the picture of a binary merger ( langer & heger 1998 ) , in which r4 was originally a system composed by a close pair and a third star ( the observed a companion ) . the close pair merged into a single star and the merging process produced a circumstellar nebula that was later shaped by the ensueing b star wind . since the bipolar morphology
, the kinematics and the enriched chemical composition make the nebula surrounding r4 very similar to the observed lbv nebulae , our findings imply that at least a few lbv outbursts and nebulae might well be the result of the merging process of two massive stars . |
the non - baryonic dark matter in the form of weakly interacting massive particles ( wimps ) still eludes detection despite recent achievements in the detection technology @xcite . aside from scaling up the size of existing detectors ,
the improvement in the detection sensitivity is possible by detecting the direction of the incoming dark matter particles . as the earth moves in the galactic halo
, the dark matter particles appear to come from cygnus constellation .
the direction tag of the of the incoming particle , often referred to as the effect , increases the sensitivity of a directional detector by one order of magnitude @xcite . in this paper we present
improved results for tagging the direction of low - energy nuclear recoils created by neutrons from a @xmath0cf source by using a time - projection chamber with optical readout .
the neutrons are used in lieu of the dark matter particles because they create similar distributions of recoil energies and angles .
the measurement of directionality tag relies on the fact that the ionization rate of recoiling nuclei depends on their residual energy , and therefore the direction of the recoil can be tagged from the light distribution along the track .
the detector is in more details described in @xcite .
the chamber utilizes @xmath1 wire frames .
the drift region between the cathode mesh and the ground wire plane is 2.6 cm with average electric field of 580 v / cm , while the amplification region between the ground and the anode wire plane ( + 2.2 kv ) is about 3 mm .
the pitch of the wires for the ground ( anode ) plane is 2 mm ( 5 mm ) and the wire diameter is 15 @xmath2 m ( 50 @xmath2 m ) .
the chamber is filled with 4 at 200 torr .
the scintillation light is recorded with a cooled ccd camera equipped with a photographic lens that images approximately 2 @xmath3 of the anode plane .
the spread of pixel yields due to adc noise and dark current is 25 counts .
images are corrected for adc bias and hot channels are identified and excluded from analysis .
neutrons are created in the fission of the nucleus , which occurs in approximately 3% of all decays and produces 3.8 neutrons per fission @xcite .
the radioactivity of our source is 3.4 mci and we estimate the total flux of @xmath4 neutrons per second into the solid angle ( @xmath5 sr ) of the detector .
the wires of the tracking chamber are aligned with the direction of the neutron beam .
the recoil length projected to the wire axis is longer in case of wimp scattering , therefore , of effect in neutron scattering is expected to be harder .
we take sequential 1-second exposures with the ccd camera .
we reject images that have segments shorter than 0.7 mm , and recoil tracks that fall close to the boundary of the ccd field of view .
the energy of the recoil segment is determined from the projection of the light intensity to the axis perpendicular to the wire .
the relation between the light intensity and the energy is determined using alpha particles that travel perpendicular to the wire and deposit a known amount of energy .
the range of the recoil segment is calibrated using the known pitch of anode wires and the observed distance between wires in the ccd images .
, width=283 ] an image of a nuclear recoil in figure [ fg::recoil_images ] shows noticeable asymmetry of the light yield along the wire . in order to quantify this effect
, we define the skewness @xmath6 as the dimensionless ratio between the third and second moments of the light yield along the wire coordinate ( @xmath7 ) : @xmath8 the sign indicates the slope of the light intensity along the track : recoils that travel in the direction of the incoming neutrons have a negative skewness . [ cols= " < , < " , ] a plot of the measured skewness as a function of the segment length is shown in the top plot of figure [ fg::recoil_energy_vs_skewness ] .
the data in this plot corresponds to 3.6 h of live time using 5 mg of 4 gas .
the head - tail asymmetry is easier to observe for longer tracks that are better aligned with the anode wires and create more scintillation light .
the bottom plot in figure [ fg::recoil_energy_vs_skewness ] shows the fraction of events with negative skewness as a function of the track length .
since the measured light yield is proportional to the energy of the recoil segment and the length is proportional to the track range projected to the wire , these two quantities should be correlated .
figure [ fg::recoil_energy_vs_length ] shows clear correlation between the light yield versus length of the recoil segments . ,
width=283 ] we collect 1 day of live - time of data without sources and find two events that pass our standard selection cuts .
we verify good rejection of gammas by collecting 1/3 day of live - time of data with @xmath9cs source ( 8 @xmath2ci ) placed near the sensitive area of our detector and find zero events passing the cuts .
we assign a conservative error of 10% to the density of the 4 gas .
the statistical uncertainty on the energy measurements is about 10% . the systematic error on the energy comes from non - uniformity in wire gain , stability of the gain over time , the pressure measurement and the calibration method that assumes the energy - independent proportionality of the stopping power with the ionization rate .
the error on the recoil range comes from the analysis technique that overestimates the range for low - energy recoils with the range close to the diffusion width .
we have presented improved results for tagging the direction of low - momentum nuclear recoils generated in the elastic scattering of low - energy neutrons with 4 gas .
we have shown that in our current experimental setup the tag of incoming particle can be determined for recoil energies above 200 kev .
this threshold can be further reduced with expected improvements in the detector preformance .
this study has profound implications for the development of dark matter detectors , as the directionalty will be essential to produce convincing evidence for dark matter particles in the presence of backgrounds . | we present new results with a prototype detector that is being developed by the dmtpc collaboration for the measurement of the direction tag ( ) of dark matter wind .
we use neutrons from a source to create low - momentum nuclear recoils in elastic scattering with the residual gas nuclei . the recoil track
is imaged in low - pressure time - projection chamber with optical readout .
we measure the ionization rate along the recoil trajectory , which allows us to determine the direction tag of the incoming neutrons . |
in many applications , in particular in the pricing of financial securities , we are interested in the effective computation by monte carlo methods of the quantity @xmath0 , where @xmath1 is a diffusion process and @xmath2 a given function .
the monte carlo euler method consists of two steps .
first , approximate the diffusion process @xmath3 by the euler scheme @xmath4 with time step @xmath5 . then approximate @xmath6 by @xmath7 , where @xmath8 is a sample of @xmath9 independent copies of @xmath10 .
this approximation is affected , respectively , by a discretization error and a statistical error @xmath11 on one hand , talay and tubaro @xcite prove that if @xmath2 is sufficiently smooth , then @xmath12 with @xmath13 a given constant and in a more general context , kebaier @xcite proves that the rate of convergence of the discretization error @xmath14 can be @xmath15 for all values of @xmath16 $ ] ( see , e.g. , kloeden and platen @xcite for more details on discretization schemes ) . on the other hand ,
the statistical error is controlled by the central limit theorem with order @xmath17 .
further , the optimal choice of the sample size @xmath9 in the classical monte carlo method mainly depends on the order of the discretization error .
more precisely , it turns out that for @xmath18 the optimal choice of @xmath9 is @xmath19 .
this leads to a total complexity in the monte carlo method of order @xmath20 ( see duffie and glynn @xcite for related results ) .
let us recall that the complexity of an algorithm is proportional to the maximum number of basic computations performed by this one .
hence , expressing this complexity in terms of the discretization error @xmath14 , we get @xmath21 . in order to improve the performance of this method ,
kebaier introduced a two - level monte carlo method @xcite ( called the statistical romberg method ) reducing the complexity @xmath22 while maintaining the convergence of the algorithm .
this method uses two euler schemes with time steps @xmath5 and @xmath23 , @xmath24 and approximates @xmath25 by @xmath26 where @xmath27 is a second euler scheme with time step @xmath23 and such that the brownian paths used for @xmath28 and @xmath29 has to be independent of the brownian paths used to simulate @xmath30 .
it turns out that for a given discretization error @xmath31 ( @xmath16 $ ] ) , the optimal choice is obtained for @xmath32 , @xmath33 and @xmath34 . with this choice ,
the complexity of the statistical romberg method is of order @xmath35 , which is lower than the classical complexity in the monte carlo method .
more recently , giles @xcite generalized the statistical romberg method of kebaier @xcite and proposed the multilevel monte carlo algorithm , in a similar approach to heinrich s multilevel method for parametric integration @xcite ( see also creutzig et al .
@xcite , dereich @xcite , giles @xcite , giles , higham and mao @xcite , giles and szpruch @xcite , heinrich @xcite , heinrich and sindambiwe @xcite and hutzenthaler , jentzen and kloeden @xcite for related results ) .
the multilevel monte carlo method uses information from a sequence of computations with decreasing step sizes and approximates the quantity @xmath36 by @xmath37 where the fine discretization step is equal to @xmath5 thereby @xmath38 . for @xmath39 , processes @xmath40 , @xmath41 , are independent copies of @xmath42 whose components denote the euler schemes with time steps @xmath43 and @xmath44 .
however , for fixed @xmath45 , the simulation of @xmath46 and @xmath47 has to be based on the same brownian path .
concerning the first empirical mean , processes @xmath48 , @xmath49 , are independent copies of @xmath50 which denotes the euler scheme with time step @xmath51 . here , it is important to point out that all these @xmath52 monte carlo estimators have to be based on different independent samples . due to the above independence assumption on the paths , the variance of the multilevel estimator is given by @xmath53 where @xmath54 .
assuming that the diffusion coefficients of @xmath55 and the function @xmath56 are lipschitz continuous , then it is easy to check , using properties of the euler scheme that @xmath57 for some positive constant @xmath58 ( see proposition [ p1 ] for more details ) .
giles @xcite uses this computation in order to find the optimal choice of the multilevel monte carlo parameters .
more precisely , to obtain a desired root mean squared error ( rmse ) , say of order @xmath15 , for the multilevel estimator , giles @xcite uses the above computation on @xmath59 to minimize the total complexity of the algorithm .
it turns out that the optimal choice is obtained for ( see theorem 3.1 of @xcite ) @xmath60 hence , for an error @xmath18 , this optimal choice leads to a complexity for the multilevel monte carlo euler method proportional to @xmath61 .
interesting numerical tests , comparing three methods ( crude monte carlo , statistical romberg and the multilevel monte carlo ) , were processed in korn , korn and kroisandt @xcite . in the present paper ,
we focus on central limit theorems for the inferred error ; a question which has not been addressed in previous research .
to do so , we use techniques adapted to this setting , based on a central limit theorem for triangular array ( see theorem [ lindeberg ] ) together with toeplitz lemma .
it is worth to note that our approach improves techniques developed by kebaier @xcite in his study of the statistical romberg method ( see remark [ rem - keb ] for more details ) .
hence , our main result is a lindeberg
feller central limit theorem for the multilevel monte carlo euler algorithm ( see theorem [ cltsr ] ) .
further , this allows us to prove a berry esseen - type bound on our central limit theorem . in order to show this central limit theorem
, we first prove a stable law convergence theorem , for the euler scheme error on two consecutive levels @xmath62 and @xmath63 , of the type obtained in jacod and protter @xcite .
indeed , we prove the following functional result ( see theorem [ th - acc ] ) : @xmath64 where @xmath65 is the same limit process given in theorem 3.2 of jacod and protter @xcite .
our result uses standard tools developed in their paper but it can not be deduced without a specific and laborious study .
further , their result , namely @xmath66 is neither sufficient nor appropriate to prove our theorem [ cltsr ] , since the multilevel monte carlo euler method involves the error process @xmath67 rather than @xmath68 .
thanks to theorem [ cltsr ] , we obtain a precise description for the choice of the parameters to run the multilevel monte carlo euler method . afterward , by a complexity analysis we obtain the optimal choice for the multilevel monte carlo euler method .
it turns out that for a total error of order @xmath18 the optimal parameters are given by @xmath69 this leads us to a complexity proportional to @xmath70 which is the same order obtained by giles @xcite . by comparing relations ( [ gilesparam ] ) and ( [ bakparam ] )
, we note that our optimal sequence of sample sizes @xmath71 does not depend on any given constant , since our approach is based on proving a central limit theorem and not on obtaining an upper bound for the variance of the algorithm .
however , some numerical tests comparing the runtime with respect to the root mean square error , show that we are in line with the original work of giles @xcite .
nevertheless , the major advantage of our central limit theorem is that it fills the gap in the literature for the multilevel monte carlo euler method and allows to construct a more accurate confidence interval compared to the one obtained using chebyshev s inequality .
all these results are stated and proved in section [ main ] .
the next section is devoted to recall some useful stochastic limit theorems and to introduce our notation .
let @xmath72 be a sequence of random variables with values in a polish space @xmath73 defined on a probability space @xmath74 .
let @xmath75 be an extension of @xmath74 , and let @xmath76 be an @xmath73-valued random variable on the extension .
we say that @xmath72 converges in law to @xmath76 stably and write @xmath77 , if @xmath78 for all @xmath79 bounded continuous and all bounded random variable @xmath65 on @xmath80 .
this convergence is obviously stronger than convergence in law that we will denote here by `` @xmath81 . ''
according to section 2 of jacod @xcite and lemma 2.1 of jacod and protter @xcite , we have the following result .
[ lemma ] let @xmath82 and @xmath83 be defined on @xmath80 with values in another metric space @xmath84 .
@xmath85 conversely , if @xmath86 and @xmath83 generates the @xmath87-field @xmath88 , we can realize this limit as @xmath89 with @xmath76 defined on an extension of @xmath90 and @xmath91 .
now , we recall a result on the convergence of stochastic integrals formulated from theorem 2.3 in jacod and protter @xcite .
this is a simplified version but it is sufficient for our study . let @xmath92 be a sequence of @xmath93-valued continuous semimartingales with the decomposition @xmath94 where , for each @xmath95 and @xmath96 , @xmath97 is a predictable process with finite variation , null at @xmath98 and @xmath99 is a martingale null at @xmath98 .
[ th - conv - int ] assume that the sequence @xmath100 is such that @xmath101 is tight .
let @xmath102 and @xmath103 be a sequence of adapted , right - continuous and left - hand side limited processes all defined on the same filtered probability space . if @xmath104 then @xmath76 is a semimartingale with respect to the filtration generated by the limit process @xmath105 , and we have @xmath106 .
we recall also the following lindeberg feller central limit theorem that will be used in the sequel ( see , e.g. , theorems 7.2 and 7.3 in @xcite ) .
[ lindeberg ] let @xmath107 be a sequence such that @xmath108 as @xmath109
. for each @xmath110 , let @xmath111 be @xmath112 independent random variables with finite variance such that @xmath113 for all @xmath114 .
suppose that the following conditions hold : @xmath115 .
lindeberg s condition : for all @xmath116 , @xmath117 moreover , if the @xmath118 have moments of order @xmath119 , then the lindeberg s condition can be obtained by the following one : lyapunov s condition : @xmath120 .
let @xmath121 be the process with values in @xmath93 , solution to @xmath122 where @xmath123 is a @xmath124-dimensional brownian motion on some given filtered probability space @xmath125 with @xmath126 is the standard filtration , @xmath127 and @xmath128 are , respectively , @xmath93 and @xmath129 valued functions .
we consider the continuous euler approximation @xmath130 with step @xmath131 given by @xmath132\delta.\ ] ] it is well known that under the global lipschitz condition @xmath133\\[-10pt ] \eqntext{x , y\in\mathbb r^d,}\end{aligned}\ ] ] the euler scheme satisfies the following property ( see , e.g. , bouleau and lpingle @xcite ) : @xmath134\\[-8pt ] & & \mathbb { e } \bigl[{\sup_{0
\leq t \leq t}}\bigl| x_t - { x}^n_t\bigr|^{p } \bigr ] \leq\frac{k_p(t)}{n^{p/2}},\qquad k_p(t)>0.\nonumber\end{aligned}\ ] ] note that according to theorem 3.1 of jacod and protter @xcite , under the weaker condition @xmath135 we have only the uniform convergence in probability , namely the property @xmath136 following the notation of jacod and protter @xcite , we rewrite diffusion ( [ 1 ] ) as follows : @xmath137 where @xmath138 is the @xmath139th column of the matrix @xmath128 , for @xmath140 , @xmath141 and @xmath142
. then the continuous euler approximation @xmath130 with time step @xmath131 becomes @xmath143\delta.\ ] ]
let @xmath144 denotes the euler scheme with time step @xmath43 for @xmath145 , where @xmath146 .
noting that @xmath147 the multilevel method is to estimate independently by the monte carlo method each of the expectations on the right - hand side of the above relation .
hence , we approximate @xmath148 by @xmath149 here , it is important to point out that all these @xmath52 monte carlo estimators have to be based on different , independent samples . for each @xmath39
the samples @xmath150 are independent copies of @xmath151 whose components denote the euler schemes with time steps @xmath43 and @xmath44 and simulated with the same brownian path . concerning the first empirical mean , the samples @xmath152 are independent copies of @xmath153 .
the following result gives us a first description of the asymptotic behavior of the variance in the multilevel monte carlo euler method .
[ p1 ] assume that @xmath127 and @xmath128 satisfy condition ( [ hbsigma ] ) .
for a lipschitz continuous function @xmath154 , we have @xmath155 we have @xmath156_{\mathrm{lip } } \sum_{\ell=1}^{l } n_{\ell}^{-1}\mathbb
e \bigl [ \sup_{0 \leq t \leq t } \bigl{\vert}x^{m^{\ell } } _ t - x_t\bigr{\vert}^2 + \sup_{0\leq t \leq t}\bigl{\vert}x^{m^{\ell-1}}_t - x_t \bigr{\vert}^2 \bigr],\end{aligned}\ ] ] where @xmath157_{\mathrm{lip}}:=\sup_{u\neq v}\frac{|f(u)-f(v)|}{|u - v|}$ ] .
we complete the proof by using property ( [ pbold ] ) on the strong convergence of the euler scheme .
inequality ( [ ordre ] ) indicates the dependence of the variance of @xmath158 on the choice of the parameters @xmath159
. this variance can be smaller than the variance of @xmath160 , so that @xmath158 appears as a good candidate for the variance reduction .
the main result of this section is a lindeberg
feller central limit theorem ( see theorem [ cltsr ] below ) .
in order to prove this result , we need to prove first a new stable law convergence theorem for the euler scheme error adapted to the setting of multilevel monte carlo algorithm .
this is crucial and is the aim of the following subsection . in what follows
, we prove a stable law convergence theorem , for the euler scheme error on two consecutive levels @xmath62 and @xmath63 , of the type obtained in jacod and protter @xcite .
our result in theorem [ th - acc ] below is an innovative contribution on the euler scheme error that is different and more tricky than the original work by jacod and protter @xcite since it involves the error process @xmath67 rather than @xmath68 .
note that the study of the error @xmath161 as @xmath162 can be reduced to the study of the error @xmath163 as @xmath164 where @xmath165 and @xmath166 stand for the euler schemes with time steps @xmath167 and @xmath5 constructed on the same brownian path . [ th - acc ] assume that @xmath127 and @xmath128 are @xmath168 with linear growth then the following result holds : @xmath169 with @xmath170 the @xmath171-dimensional process satisfying @xmath172,\ ] ] where @xmath173 and @xmath174 is the @xmath175 valued process solution of the linear equation @xmath176.\ ] ] here , @xmath177 is a @xmath178 matrix with @xmath179 is the partial derivative of @xmath180 with respect to the @xmath181th coordinate , and @xmath182 is a standard @xmath183-dimensional brownian motion independent of @xmath184 .
this process is defined on an extension @xmath185 of the space @xmath186 .
note that by letting formally @xmath187 tend to infinity , we recover the jacod and protter s result @xcite .
proof of theorem [ th - acc ] consider the error process @xmath188 , defined by @xmath189.\ ] ] combining relation ( [ forme - euler ] ) , for both processes @xmath165 and @xmath166 , together with a taylor expansion @xmath190 where @xmath191 is the @xmath178 matrix whose @xmath192th row is the gradient of the real - valued function @xmath180 at a point between @xmath193 and @xmath194 .
therefore , the equation satisfied by @xmath195 can be written as @xmath196 with @xmath197 in the following , let @xmath198 be the @xmath199 valued solution of @xmath200 theorem 48 , page 326 in @xcite , ensures existence of the process @xmath201 defined as the solution of @xmath202 thanks to theorem 56 , page 333 in the same reference @xcite , we get @xmath203 since the increments of the euler scheme satisfy @xmath204 and @xmath205 with @xmath206 and @xmath207 , it is easy to check that @xmath208\\[-8pt ] & & { } -\sum_{i , j=1}^{q } z^{m n , n}_t \int_0^t \widetilde h^{i , j , m n , n}_s \bigl(y^i_s - y^i_{\eta_{m n}(s)}\bigr ) \,dy^j_s
-\widetilde r_{t,1}^{m n , n } -\widetilde r_{t,2}^{m n , n}\nonumber\end{aligned}\ ] ] with @xmath209 and @xmath210 where , for @xmath211 , @xmath212 and @xmath213 now , let us introduce @xmath214 moreover , @xmath215 exists and satisfies the following explicit linear stochastic differential equation : @xmath216 thanks to the uniform convergence in probability of the euler scheme and according to theorem 2.5 in jacod and protter @xcite , we have @xmath217 furthermore , in relation ( [ eq1erreur ] ) , one can replace , respectively , @xmath218 and @xmath219 by their common limit @xmath220 given by relation ( [ eq - defh ] ) .
so that relation ( [ eq1erreur ] ) becomes @xmath221 with @xmath222 where @xmath223 and @xmath224 , @xmath225 , are introduced by relation ( [ eq1erreur ] ) and @xmath226 the remainder term process @xmath227 vanishes with rate @xmath228 in probability .
more precisely , we have the following convergence result .
[ lem ] the rest term introduced in relation ( [ eq1biserreur ] ) is such that @xmath229 converges to zero in probability as @xmath110 tends to infinity . for the reader s convenience , the proof of this lemma
is postponed to the end of the current subsection .
the task is now to study the asymptotic behavior of the process given by relation ( [ eq1biserreur ] ) @xmath230 in order to study this process , we introduce the martingale process , @xmath231 and we proceed to a preliminary calculus of the expectation of its bracket .
let @xmath232 and @xmath233 , we have : * for @xmath234 , the bracket @xmath235 , * for @xmath236 and @xmath237 , @xmath238 , * for @xmath236 and @xmath239 , @xmath240 , @xmath241 $ ] and we have @xmath242 - 1}\int_{(mk+\ell ) \delta /m}^{(mk+\ell+1)\delta / m } \bigl(\eta_{m n}(s)-\eta_n(s)\bigr ) \,ds+o\biggl ( \frac { 1}{n^2}\biggr ) \nonumber \\ & = & \sum_{\ell=0}^{m-1}\sum _
{ k=0}^{[t/\delta]-1}\frac{\delta ^2}{m } \biggl ( \frac{mk+\ell}{m}-k \biggr)+o\biggl(\frac{1}{n^2}\biggr ) \\
& = & \frac{(m-1)\delta ^2}{2m}[t/\delta ] + o\biggl(\frac{1}{n^2}\biggr ) \nonumber \\ & = & \frac{(m -1)t}{2 m n } t + o\biggl(\frac{1}{n^2}\biggr).\nonumber\end{aligned}\ ] ] having disposed of this preliminary evaluations , we can now study the stable convergence of @xmath243 . by virtue of theorem 2.1 in @xcite
, we need to study the asymptotic behavior of both brackets @xmath244 and @xmath245 , for all @xmath241 $ ] and all @xmath246 .
the case is obvious and we only proceed to prove that : * for @xmath236 , @xmath247 , for all @xmath241 $ ] , * for @xmath236 and @xmath237 , @xmath248 , for all @xmath241 $ ] , * for @xmath236 and @xmath249 , @xmath250 , for all @xmath241 $ ] . for the first point , we consider the @xmath251 convergence @xmath252 with @xmath253\\[-8pt ] & & { } -\eta_n(s)\wedge \eta_{m n}(u)+ \eta_n(s)\wedge\eta_n(u).\nonumber\end{aligned}\ ] ] it is worthy to note that @xmath254 hence , @xmath255 , for @xmath256 , @xmath257 , for @xmath258 , and @xmath259 this yields the desired result . concerning the second point ,
the @xmath251 norm is given by @xmath260 with the same function @xmath261 given in relation ( [ eq - defg ] ) .
using the properties of function @xmath261 developed above , we have in the same manner @xmath262 which proves our claim . for the last point , that is the essential one , we use the development of @xmath263 given by relation ( [ eqn - expectation ] ) to get @xmath264\\[-8pt]\nonumber & & \qquad = n^2 \mathbb e\bigl\langle m^{n , i , j } \bigr\rangle_t^2- \frac{(m -1)^2t^2}{4m^2 } t^2+o\biggl(\frac{1}{n}\biggr).\end{aligned}\ ] ] otherwise
, we have @xmath265 with @xmath266 on one hand , for @xmath256 , using property ( [ eq - prop ] ) together with the independence of the increments @xmath267 and @xmath268 , yields @xmath269 on the other hand , in relation ( [ eq - defh ] ) we use the cauchy
schwarz inequality to get @xmath270 and this yields @xmath271 now , noting that @xmath272 , relation ( [ eqn2-expectation ] ) becomes @xmath273 once again thanks to the development of @xmath274 given by relation ( [ eqn - expectation ] ) , we deduce that @xmath275 by ( [ eq - dev1 ] ) and ( [ eq - dev2 ] ) , we deduce the convergence in @xmath251 of @xmath276 toward @xmath277 . by theorem 2.1 in jacod @xcite , @xmath278 converges in law stably to a standard @xmath183-dimensional brownian motion @xmath279 independent of @xmath184 .
consequently , by lemma [ lemma ] and theorem [ th - conv - int ] , we obtain @xmath280 finally , we complete the proof using relations ( [ eq - doleansdade ] ) , ( [ eq1biserreur ] ) , lemma [ lem ] and once again lemma [ lemma ] to obtain @xmath281 proof of lemma [ lem ] at first , we prove the uniform convergence in probability toward zero of the normalized rest terms @xmath282 for @xmath225 .
the convergence of @xmath283 @xmath225 is a straightforward consequence of the previous one .
the main part of these rest terms can be represented as integrals with respect to three types of supermartingales that can be classified through the following three cases : @xmath284 where @xmath285 and @xmath241 $ ] . in the first case , the supermartingale is deterministic of finite variation and its total variation on the interval @xmath286 $ ] has the following expression : @xmath287 so , the process @xmath288 converges to @xmath98 and is tight . in the second case , for @xmath289 ,
the supermartingale is also of finite variation and its total variation on the interval @xmath286 $ ] has the following expression : @xmath290 it is clear that @xmath291 , which ensures the tightness of the process @xmath292 .
therefore , we only need to establish the convergence of @xmath293 toward @xmath98 in @xmath294 , for @xmath241 $ ] .
in fact , we have @xmath295 when @xmath256 , we have @xmath296 and by independence of the brownian motion increments , we deduce that the integrand term is equal to @xmath98 . otherwise , when @xmath297 , we apply the cauchy
schwarz inequality to get @xmath298 it follows from all these that @xmath299 . in the last case , for @xmath300 , the process
@xmath301 is a square integrable martingale and its bracket has the following expression : @xmath302 it is clear that @xmath303 , so we deduce the tightness of the process @xmath304 and the convergence @xmath305 .
now thanks to property ( ) and relation ( [ eq - doleansdade ] ) , it is easy to check that the integrand processes @xmath306 and @xmath307 , introduced in relation ( [ eq1erreur ] ) , converge uniformly in probability to their respective limits @xmath308 and @xmath309 , where @xmath310 and @xmath311 . therefore , by theorem [ th - conv - int ] we deduce that the integral processes given by @xmath312 vanish .
consequently , we conclude using relation ( [ eq - doleansdade ] ) that @xmath313 for @xmath225 .
we now proceed to prove that @xmath314 .
the convergence of the process @xmath315 toward @xmath98 is obviously obtained from the previous one .
the main part of this rest term can be represented as a stochastic integral with respect to the martingale process given by @xmath316 with @xmath317 .
it was proven in jacod and protter @xcite that @xmath318 where @xmath182 is a standard @xmath183-dimensional brownian motion defined on an extension probability space @xmath185 of @xmath319 , which is independent of @xmath184 .
thanks to property ( ) and relation ( [ eq - doleansdade ] ) , the integrand process @xmath320 and once again by theorem [ th - conv - int ] we deduce that the integral processes given by @xmath321 vanish
. all this allows us to conclude using relation ( [ eq - doleansdade ] ) .
let us recall that the multilevel monte carlo method uses information from a sequence of computations with decreasing step sizes and approximates the quantity @xmath36 by @xmath322 in the same way as in the case of a crude monte carlo estimation , let us assume that the discretization error @xmath323 is of order @xmath15 for any @xmath16 $ ] .
taking advantage from the limit theorem proven in the above section , we are now able to establish a central limit theorem of lindeberg feller type on the multilevel monte carlo euler method .
to do so , we introduce a real sequence @xmath324 of positive terms such that @xmath325\\[-10pt ] \eqntext{\mbox{for } p>2}\end{aligned}\ ] ] and we assume that the sample size @xmath326 depends on the rest of parameters by the relation @xmath327 we choose this form for @xmath326 because it is a generic form allowing us a straightforward use of toeplitz lemma that is a crucial tool used in the proof of our central limit theorem . indeed , property ( [ con - weights - e ] ) implies that if @xmath328 is a sequence converging to @xmath329 as @xmath45 tends to infinity then @xmath330 in the sequel , we will denote by @xmath331 , respectively , @xmath332 the expectation , respectively , the variance defined on the probability space @xmath75 introduced in theorem [ th - acc ] .
we can now state the central limit theorem under strengthened conditions on the diffusion coefficients .
[ cltsr]assume that @xmath127 and @xmath128 are @xmath333 functions satisfying the global lipschitz condition ( [ hbsigma ] ) .
let f be a real - valued function satisfying @xmath334 assume @xmath335 , where @xmath336 is differentiable at @xmath337 , and that for some @xmath16 $ ] we have @xmath338 then , for the choice of @xmath326 , @xmath339 given by equation ( [ eq - size ] ) , we have @xmath340 with @xmath341 and @xmath342 denotes a normal distribution .
the global lipschitz condition ( [ hbsigma ] ) seems to be essential to establish our result , since it ensures property ( [ pbold ] ) .
otherwise , hutzenthaler , jentzen and kloeden @xcite prove that under weaker conditions on @xmath127 and @xmath128 the multilevel monte carlo euler method may diverges whereas the crude monte carlo method converges .
proof of theorem [ cltsr ] to simplify our notation , we give the proof for , the case @xmath343 is a straightforward deduction . combining relations ( [ eq - biais ] ) and ( [ eq - estimateur ] ) together
, we get @xmath344 where @xmath345 using assumption ( [ hen ] ) , we obviously obtain the term @xmath346 in the limit .
taking @xmath347 , we can apply the classical central limit theorem to @xmath348 .
then we have @xmath349 . finally , we have only to study the convergence of @xmath350 and we will conclude by establishing @xmath351 to do so
, we plan to use theorem [ lindeberg ] with the lyapunov condition and we set @xmath352\\[-8pt ] z^{m^{\ell},m^{\ell -1}}_{t , k}&:= & f\bigl ( x^{\ell , m^{\ell}}_{t , k } \bigr)-f\bigl ( x^{\ell , m^{\ell-1}}_{t , k}\bigr)-\mathbb e \bigl(f\bigl ( x^{\ell , m^{\ell}}_{t , k}\bigr)-f\bigl ( x^{\ell , m^{\ell-1}}_{t , k } \bigr ) \bigr).\nonumber\end{aligned}\ ] ] in other words , we will check the following conditions : * @xmath353 . *
( lyapunov condition ) there exists @xmath119 such that @xmath354 . for the first one , we have @xmath355 otherwise , since @xmath356 ,
applying the taylor expansion theorem twice we get @xmath357 the function @xmath358 is given by the taylor
young expansion , so it satisfies@xmath359 and @xmath360 . by property ( [ pbold ] ) , we get the tightness of @xmath361 and @xmath362 and then we deduce @xmath363 so , according to lemma [ lemma ] and theorem [ th - acc ] we conclude that @xmath364\\[-10pt ] \eqntext{\mbox{as } \ell\rightarrow\infty.}\end{aligned}\ ] ] using ( [ hf ] ) it follows from property ( [ pbold ] ) that @xmath365 we deduce using relation ( [ cle - euler ] ) that @xmath366 consequently , @xmath367 hence , combining this result together with relation ( [ tcl - lindebergfeller ] ) , we obtain the first condition using toeplitz lemma . concerning the second one , by burkhlder s inequality and elementary computations , we get for @xmath119 @xmath368 where @xmath369 is a numerical constant depending only on @xmath370 . otherwise , property ( [ pbold ] )
ensures the existence of a constant @xmath371 such that @xmath372 therefore , @xmath373\\[-8pt ] & \leq & \frac{\widetilde c_p } { ( \sum_{\ell=1}^{l}a_\ell ) ^{p/2}}\sum_{\ell = 1}^{l}a_\ell^{p/2 } \mathop{\longrightarrow}_{n\rightarrow\infty } 0.\nonumber\end{aligned}\ ] ] this completes the proof .
[ rem ] from theorem 2 , page 544 in @xcite , we prove a berry esseen - type bound on our central limit theorem .
this improves the relevance of the above result . indeed , take @xmath374 as in the proof , for @xmath375 and @xmath376 given by relation ( [ xnl - berryessen ] ) , with @xmath377 , put @xmath378 and denote by @xmath379 the distribution function of @xmath380 .
then for all @xmath329 and @xmath381 @xmath382 where @xmath383 is the distribution function of a standard gaussian random variable . if we interpret the output of the above inequality as sum of independent individual path simulation , we get @xmath384 according to the above proof , it is clear that @xmath385 behaves like a constant but getting lower bounds for @xmath385 seems not to be a common result to our knowledge .
concerning @xmath386 , taking @xmath387 in both inequalities ( [ eq - momentinequality ] ) and ( [ eq - momentinequalitybis ] ) gives us an upper bound .
in fact , when @xmath2 is lipschitz , there exists a positive constant @xmath388 depending on @xmath127 , @xmath128 , @xmath51 and @xmath2 such that @xmath389 for the optimal choice @xmath390 , given in the below subsection , the obtained berry esseen - type bound is of order @xmath391 .
[ rem - keb ] note that the above proof differs from the ones in kebaier @xcite .
in fact , here our proof is based on the central limit theorem for triangular array which is adapted to the form of the multilevel estimator , whereas kebaier used another approach based on studying the associated characteristic function .
further , this latter approach needs a control on the third moment , whereas we only need to control a moment strictly greater than two .
also , it is worth to note that the limit variance in theorem [ cltsr ] is smaller than the limit variance in theorem 3.2 obtained by kebaier in @xcite . from a complexity analysis point of view , we can interpret theorem [ cltsr ] as follows .
for a total error of order @xmath15 , the computational effort necessary to run the multilevel monte carlo euler method is given by the sequence of sample sizes specified by relation ( [ eq - size ] ) . the associated time complexity
is given by @xmath392 the minimum of the second term of this complexity is reached for the choice of weights @xmath393 , @xmath39 , since the cauchy
schwarz inequality ensures that @xmath394 , and the optimal complexity for the multilevel monte carlo euler method is given by @xmath395 it turns out that for a given discretization error @xmath31 to be achieved the complexity is given by @xmath396 . note that this optimal choice @xmath393 , @xmath39 , with taking @xmath397 corresponds to the sample sizes given by @xmath398 hence , our optimal choice is consistent with that proposed by giles @xcite . nevertheless , unlike the parameters obtained by giles @xcite for the same setting
[ see relation ( [ gilesparam ] ) ] , our optimal choice of the sample sizes @xmath399 does not depend on any given constant , since our approach is based on proving a central limit theorem and not on getting upper bounds for the variance .
otherwise , for the same error of order @xmath18 the optimal complexity of a monte carlo method is given by @xmath400 which is clearly larger than @xmath401 .
so , we deduce that the multilevel method is more efficient .
also , note that the optimal choice of the parameter @xmath187 is obtained for @xmath402 .
otherwise , any choice @xmath403 , @xmath404 , leads to the same result .
some numerical tests comparing original giles work @xcite with the one of us show that both error rates are in line . here in figure
[ fig1 ] , we make a simple log log scale plot of cpu time with respect to the root mean square error , for european call and with @xmath405 .
it is worth to note that the advantage of the central limit theorem is to construct a more accurate confidence interval .
in fact , for a given root mean square error rmse , the radius of the @xmath406-confidence interval constructed by the central limit theorem is @xmath407rmse .
however , without this latter result , one can only use chebyshev s inequality which yields a radius equal to @xmath408rmse .
finally , note that , taking @xmath409 still gives the optimal rate and allows us to cancel the bias in the central limit theorem due to the euler discretization .
the multilevel monte carlo algorithm is a method that can be used in a general framework : as soon as we use a discretization scheme in order to compute quantities such as @xmath410 , we can implement the statistical multilevel algorithm . and
this is worth because it is an efficient method according to the original work by giles @xcite .
the central limit theorems derived in this paper fill the gap in literature and confirm superiority of the multilevel method over the classical monte carlo approach .
the authors are greatly indebted to the associate editor and the referees for their many suggestions and interesting comments which improve the paper . | this paper focuses on studying the multilevel monte carlo method recently introduced by giles [ _ oper .
res . _
* 56 * ( 2008 ) 607617 ] which is significantly more efficient than the classical monte carlo one .
our aim is to prove a central limit theorem of lindeberg feller type for the multilevel monte carlo method associated with the euler discretization scheme .
to do so , we prove first a stable law convergence theorem , in the spirit of jacod and protter [ _ ann .
probab . _
* 26 * ( 1998 ) 267307 ] , for the euler scheme error on two consecutive levels of the algorithm .
this leads to an accurate description of the optimal choice of parameters and to an explicit characterization of the limiting variance in the central limit theorem of the algorithm .
a complexity of the multilevel monte carlo algorithm is carried out . . |
from the very first days of astrophysics , astronomers have tried to increase their understanding of the universe by building more and more powerful telescopes .
however , there is another way to increase the performances which consists in keeping more or less the same type of instruments but by to install them on an observational site presenting better characteristics in terms of background , transparency , image quality , etc .
this quest led astronomers to build telescopes on secluded , remote sites on mountain tops or even to launch them in space .
however , if we try to summarize these characteristics , keywords are probably cold , dry , stable , high , dark ( wavelength dependent ) and it seems that there is one site on earth that qualifies for all of them : antarctica . to make the best use of the above exceptional characteristics , we propose to build a wide - field ( 0.5-degree in diameter ) high - resolution ( @xmath6 arcsec using ground layer adaptive optics ) infrared ( from 0.8 to 5 @xmath7 ) 2.4-m telescope that we call white .
quite a number of science topics can take advantage of those figures .
burton et al .
( 2005 ) reviewed possible science programmes for a 2-m class telescope at dome c. however , we will try to show that three well focused key surveys would provide us with the very best and unique science that be carried out within a few years .
of course , beyond the pure science , such surveys would be very helpful to pre - select interesting targets for jwst and alma .
defining the science objectives , estimating the limits in terms of performance , of time of completion and , in terms of cost is crucial to get the necessary support from the related science community .
we will define three key projects that would wonderfully benefit from the observational characteristics available at dome c. the two magellanic clouds are small satellite galaxies of the milky way .
as so , they provide us with the closest extragalactic objects that we can observe in detail from the south hemisphere .
the total lmc has been surveyed with spitzer ( sage project : http://sage.stsci.edu/ ) in the infrared but the angular resolution is only spitzer s , i.e. 1 - 2 arcsec . we need to go beyond sage in the very dense stellar environments that we observe .
the two magellanic clouds are ideally suited to study star formation in a global scale and a survey of the clouds in infrared at a 0.3-arcsec angular resolution will provide a unique database to statistically study star formation outside the milky way and resolve stars .
the interaction between the magellanic clouds and the milky way is an instance of a minor merger that might have been a common phenomenon in the evolution of large galaxies such as the milky way .
so , understanding this galaxy - galaxy interaction on a large scale in our neighbourhood is crucial in a cosmological sense .
the infrared wavelength range is particularly interesting to study young stellar formation regions which are embedded into dust clouds . indeed , since the optical depth of dust decreases with increasing wavelengths ( figure 1 ) , it is possible to detect and study objects that would be otherwise deeply buried into dust and , therefore , undetected at optical wavelengths . .
the white logo indicates the wavelength range that we would like to use.,width=226,height=151 ] the high sensitivity in the nir / mir opens up the possibility to study very large populations of high redshift galaxies ( half a million galaxies per sq .
deg . for the shallow night time survey ) that will be observed in their optical frame .
the observations also offers a very interesting opportunity to detect a new population of dusty supernovae that can hardly be observed in the visible range ( e.g. maiolino et al . 2002 ) . we know that most of the star formation in the universe at @xmath9 is hidden in dust ( figure 2 ) .
the infrared is therefore a very promising way to explore the history of the star formation in galaxy and the young phases of the formation of galaxies .
this topics undergoes a large interest because of the availability of infrared facilities in the recent years ( e.g. chary et al .
2005 , pozzo et al .
2006 , elias - rosa et al .
2006 ) . as compared to the local universe.,width=226,height=151 ] the non detection of dusty supernovae is very likely due to the fact that the dust attenuation is much higher for some supernovae .
maiolino et al .
( 2002 ) estimated that the dust attenuation in the v - band ( @xmath10 ) of sn2001db is about 5 magnitudes higher than the average dust attenuation for a more usual sample of supernovae .
astier et al .
( 2006 ) stated that `` there is no consensus on how to correct for host galaxy extinction affecting high redshift sneia '' .
so , we propose with this survey to decrease its influence by moving to the infrared where the dust attenuation is much lower ( by @xmath11 mag . ,
figure 1 ) .
going in the infrared will help in the interpretation of snae light curves by neglecting dust effects .
an additional positive effect would be that light curves in the infrared are flatter for a longer time than in the visible ( krisciunas et al .
2003 , di carlo 2004 ) .
that means that we have a better opportunity to study supernovae at a higher signal - to - noise ratio in the near infrared than in the visible because the light curve are much more favourable to detect these objets .
chary et al .
( 2005 ) studied supernovae observed with the hst , spitzer / irac and spitzer / mips .
they show that an efficient survey of dusty supernovae should be carried out in the nir / mir to detect embedded objects but they also stress that high angular resolution is needed to identify supernovae close to the core of galaxies in case of a nuclear starburst .
figure 3 presents the maximum redshift to detect dusty supernovae estimated in the j , k and l bands .
dashed - dotted horizontal lines show the limiting magnitude for the shallow survey ( i.e. using the temporal information ) in the three above bands . j and k bands provide about the same maximum redshift of @xmath12 in 1-hour exposures within 10 days from the light maximum while the detection of supernovae at maximum would be possible up to z = 0.7 - 0.8 , with a slight advantage for the j - band that might be compensented by the fact that the dust attenuation is lower in k than in j. however , the l - band maximum redshift is @xmath13 but with the advantage of observing dustier supernovae .
figure 4 shows that at @xmath14 , a few dusty snia / sq .
deg would be observed .
however , if we do not restrict ourselves to snia but open up to core - collapsed supernovae that are more relevant to study the formation of dust and the star formation of galaxies , a few tens to a few hundreds core collapsed dusty supernovae ( within 10 days from the light maximum i.e. 0.2 mag . ) can be detected at z = 0.7 - 0.8 .
this unique sample can only be observed if we can take simultaneously advantage of infrared observations ( to look into the dust ) , of time - dependent observations ( supernovae detections ) , of wide - field observations ( for statistics ) and for high - resolution ( to resolve supernovae from their parent galaxies ) .
.,width=226,height=151 ] although only a few snia supernovae per square degree could be detected at z = 0.4 , the number of core collapsed supernovae could reach more respectable numbers up to a few tens .
the situation is much more favourable at z = 0.7 - 0.8 where several tens snia per square degree and several hundreds core collapsed supernovae per square degree to build the first large sample of dusty supernovae . an additional survey would be to use some of the time that might be available if we relax the constraint on the astronomical twilight
. a very wide survey over 1000 square degrees in the @xmath15 band would be possible down to @xmath16 abmag .
this type of survey would bring a very large sample of galaxies useful , for instance , to constrain the cosmology through the weak lensing .
this key project is detailed by saunders ( this volume ) and we refer to this paper for a presentation of this very promising topics .
classic adaptive optics enables large telescopes to provide diffraction limited images , but their corrected field is restrained by the angular decorrelation of the turbulent wave - fronts .
however many scientific goals would benefit from a wide and uniformly corrected field , even with a partial correction of about 0.3 .
ground layer adaptive optics ( glao ) systems are supposed to provide such a correction by compensating the lower part of the atmosphere only and a glao correction appears as a good and adapted system in the specific case of dome c. a more detailed description of glao can be found in andersen et al .
we performed two independant simulations based on two sets of assumptions .
the first set assumed a 4 m telescope with a 8x8-actuators deformable mirror from the v - band to the k - band and a turbulence seeing on the ice of 1.1 arcsec .
the second set assumed a 2 m telescope with a 10x10-actuators deformable mirror from the i - band to the k - band and a turbulence seeing of 1.9 arcsec that are carried out using paola ( jolissaint , veran , conan 2006 ) .
the first simulations are relatively optimistic while the second one are more pessimistic .
the number of actuators does not pose any strong technological constraints .
these simulations are encouraging : the gain in encircled energy is still larger than 6 at a radius of 30 arcmin in the red and infrared wavelength ranges ( burgarella et al .
2007 ) .
the study and development plan of the ground layer adaptive optics is therefore a priority .
after the initial simulations carried out in 2007 , we need to test the concept on an antarctized optical bench during 2008 and part of 2009 .
the phase is necessary to check that the glao concept allows to reach the requirements in the lab .
then , it will be necessary to develop a prototype that will be tested on the sky before shipping the prototype to dome c where it could be tested on a small telescope ( e.g. irait ) .
finally , if the tests are positive the telescope + glao + instrumentation would be on the sky in the window 2012 - 2014 to start building the key - project surveys .
the completion of this phase should occur in the 2015 timeframe .
two types of telescopes are under study at the observatoire astronomique marseille provence that would allow observing in the nir+mir over a wide field of view while not degrading the image quality above 0.3 arcsec . *
a first one is the three - reflection telescope ( trt ) designed by lemaitre et al .
the size of the spot is still 0.25 arcsec rms at a radius of 1 degree , which is well within our specifications .
two down - scaled models of this telescope already exist at ias , frascati and at oamp / lam , marseille .
* another option would be an off - axis design adapted from space telescope designs ( e.g. moretto et al .
2004 ) which bears several advantages : no obstruction of the beam by the secondary mirror , no diffraction / emission from the secondary support structure to degrade images / sensitivities .
this type of design is studied by g. moretto .
the feasibility of manufacturing such mirrors for both designs are well understood by several optical manufacturers such as reosc ( paris - france ) and mirror lab ( tucson , arizona , usa ) in particular for the 2 m m1 . in the case of the smaller aspherical mirrors - from 0.43 to 1.1 m
- we are confident that such mirrors can be made with active vase mirror technology pioneered by lemaitre . a concept for an instrument is presented by mora et al .
in this volume . estimating the performances of white is crucial to evaluate the scientific interest of this project . to compute the performances presented in figure 4 , we assume a 2.4 m primary mirror , a night time ir background from walden et al .
( 2005 ) , a day time ir background ( beyond the k - band ) has been estimated by interpolating between the k - band background quoted by gillett ( august 1995 ) : 9.9 mag / arcsec2 and 35 jy / arcsec2 at 11.5 m ( storey et al .
the temperature of the telescope at 210k ( emissivity set to 3% ) , a pixel scale of 0.15 , jwst - type detectors ( 1 - 5 @xmath7 ) , a ground layer ao , integration over 2 * fwhm ( ee = 80% ) to optimize the snr . the angular resolution is set by the glao device up to the k - band and by the limit of diffraction at longer wavelength .
each individual image of the shallow survey will permit to have a very important temporal information that will be used to track time - dependent phenomenons such as variable stars or supernovae .
on the other hand , by summing up all the individual exposures , we will be able to reach much deeper sensitivies . here
, we assume a total exposure time of 42h in each band but the total exposure time varies from 42 hours at the shorter wavelength to more than 160 hours in j.
we present in this paper a concept and the science objectives for the project white that would observe in the approximate range @xmath17 using a glao device to reach an angular resolution of @xmath6 arcsec over a field of fiew of @xmath18 sq .
it is important to notice that white will efficiently use all the observing time by carrying out observations beyond @xmath19 even during daytime . to optimize white ,
we define a few scientific key projects that will be carried out in a limited ( 2 - 3 years ) time .
we must note , however , that other similar projects have been are are proposed like amidst ( epchtein 2004 ; le bertre , vauglin & epchtein 2006 ) , pilot ( storey et al .
2007 ) and a smaller telescope ( irait , tosti 2007 ) is now at dome c that will permit to characterize the site from astronomical observations in the wavelength range ( @xmath20 ) .
white moves along the same way as the previous ones . because of environmental and funding issues , a cooperation between several institutes around the world is mandatory .
99 andersen d.r . , stoesz j. , morris s. , et al .
2006 , astro - ph/061009 astier p. , guy j. , regnault n. et al . 2006 , a & a 447 , 31 burgarella d. , ferrari m. , fusco t. , et al .
2007 , eas publications series , vol .
25 burton m.g . , lawrence j.s . ,
ashley m.c.b . ,
2005 , pasa 22 , 199 chary r. , dickinson m.e . , teplitz h.i . ,
et al . 2005 , apjs 635 , 1022 di carlo e. , 2004 , mmsai 75 , 150 elias - rosa n. , benetti s. , cappelaro e. , et al .
2006 , mnras 369 , 1880 epchtein n. 2004 , eas publicaions series , vol .
14 , 193 jolissaint l. , vran j .-
, conan r. 2006 , josa a 23 , 382 krisciunas k. , suntzeff n.b . ,
candia , p. et al .
2003 , aj 125 , 166 le bertre t. , vauglin i .
, epchtein n. 2006 , eas .
publications series , vol .
25 , 179 lemaitre g.r . ,
montiel , p. , joulie , p. , et al .
2004 , spie 5494 , 426 maiolino r. , vanzi l. , mannucci f. , et al .
2002 , a & a 389 , 84 moretto g. , langlois , m. , ferrari , m. 2004 , spie 5487 , 1111 storey j.v.w . , ashley m.c.b . , burton m.g . ,
1998 , spie 3354 , 1158 storey j.w.v . ,
ashley m.c.b . ,
burton m.g . ,
2007 , eas publications series , vol .
25 , 255 pozzo m. , meikle w.p.s .
, rayner j.t . , et al .
2006 , mnras 368 , 1169 takeuchi t.,t .
, buat v. , burgarella d. 2005 , a & a 440 , l17 , 147 tosti g. 2007 , eas publications series , vol . 25 , 209 walden v.p . , town m.s . , halter b. , et al .
2005 , pasp 117 , 300 | the wide - field high - resolution infrared telescope ( white ) will be dedicated in the first years of its life to carrying out a few ( well focused in terms of science objectives and time ) legacy surveys .
white would have an angular resolution of @xmath0 homogeneous over @xmath1 sq . deg . in the wavelength range @xmath2 to @xmath3 , which means that we will very efficiently use all the available observational time during night time and day time .
moreover , the deepest observations will be performed by summing up shorter individual frames .
we will have a temporal information that can be used to study variable objects .
the three key science objectives of white are : 1 ) a complete survey of the magellanic clouds to make a complete census of young stellar objects in the clouds and in the bridge and to study their star formation history and the link with the milky way .
the interaction of the two clouds with our galaxy might the closest example of a minor merging event that could be the main driver of galaxy evolution in the last 5 gyrs .
2 ) the building of the first sample of dusty supernovae at @xmath4 in the near infrared range ( @xmath5 ) to constrain the equation of state from these obscured objects , study the formation of dust in galaxies and build the first high resolution sample of high redshift galaxies observed in their optical frame 3 ) a very wide weak lensing survey over that would allow to estimate the equation of state in a way that would favourably compete with space projects . |
neutrino oscillations have been experimentally observed in a variety of situations , and are among the first evidence of physics beyond the standard model . typically , these oscillations are explained by attributing mass to neutrinos ; however , not all experiments can be explained using the same masses - notably , lsnd@xcite and miniboone@xcite require a larger mass - squared difference than the other experiments , and can not be explained using a three - flavor theory of mass .
furthermore , recent results at minos and miniboone have hinted at an asymmetry between neutrinos and antineutrinos @xcite , which would be evidence for lorentz violation .
it has already been shown that models incorporating lorentz violations can reproduce many of the results of the mass model@xcite .
examples include the bicycle model@xcite and the tandem model@xcite . here
, a new model is introduced to attempt to explain these experiments .
we consider three generations of light , left - handed neutrinos , and three generations of light , sterile , right - handed neutrinos , and their antiparticles .
we allow for small mass and small general lorentz violations . to first order , the general hamiltonian is a 12 @xmath2 12 matrix , given in block form by @xmath3\ ] ] where @xmath4 \\ h_{12 } & = & \left [ \begin{array}{cc } -ig^{\lambda \mu \nu}_d p_\lambda q_\mu p_\nu + ih_d^{\lambda \mu}p_\lambda q_\mu & -ig_m^{\lambda \mu \nu}p_\lambda q_\mu
p_\nu + ih_m^{\lambda \mu}p_\lambda q_\mu\\ -ig_m^{\dagger \lambda \mu \nu}p_\lambda q_\mu p_\nu + ih_m^{\dagger \lambda \mu}p_\lambda q_\mu & -ig_d^{t \lambda \mu \nu}p_\lambda q_\mu p_\nu - ih_d^{t \lambda \mu}p_\lambda q_\mu\\ \end{array } \right ] \\ h_{21 } & = & \left [ \begin{array}{cc } ig_d^{\lambda \mu \nu}p_\lambda q^*_\mu p_\nu - ih_d^{\lambda \mu}p_\lambda q^*_\mu & ig_m^{\lambda \mu \nu}p_\lambda q^*_\mu p_\nu - ih_m^{\lambda \mu}p_\lambda q^*_\mu \\
ig_m^{\dagger \lambda \mu \nu}p_\lambda q^*_\mu p_\nu - ih_m^{\dagger \lambda \mu}p_\lambda q^*_\mu & ig_d^{t\lambda \mu \nu}p_\lambda q^*_\mu p_\nu + ih_d^{t \lambda \mu}p_\lambda q^*_\mu \\ \end{array } \right ] \\ h_{22 } & = & \left [ \begin{array}{cc } -c_r^{\mu \nu}p_\mu p_\nu + a_r^\mu p_\mu & -c_m^{t\mu\nu}p_\mu p_\nu - a_m^{t\mu}p_\mu \\ -c_m^{*\mu \nu}p_\mu p_\nu - a^{*\mu}_m p_\mu & -c_l^{t \mu \nu}p_\mu p_\nu - a_l^{t\mu}p_\mu \\ \end{array } \right]\end{aligned}\ ] ] in the basis @xmath5 .
note that mass does not appear because it enters only at second order .
such a hamiltonian allows many unusual features , including neutrino - antineutrino mixing , neutrino - sterile neutrino mixing , strange energy dependence and direction dependence .
we propose a tricycle model , in which we assume that @xmath6 so that the off - diagonal terms can be ignored .
this allows us to restrict our attention to one quadrant of the matrix , the @xmath7 sector ( @xmath8 ) .
note that we have only considered the isotropic part of each of the lorentz - violating coefficients .
in particular , the model to be investigated has the form @xmath9\ ] ] where @xmath10 is taken to be hermitian and @xmath10 and @xmath11 commute , so that they can be simultaneously diagonalized .
we assume that the diagonalizing matrix has the conventional form@xcite @xmath12\ ] ] ( we are assuming that @xmath13 ) . the model is then fixed by 8 parameters : the two mixing angles and the three eigenvalues of each block , which we call @xmath14 and @xmath15 respectively .
the eigenvalues of the hamiltonian are @xmath16 the model employs a seesaw mechanism to produce very different behavior at high and low energies . at high energy the @xmath17 matrix dominates , which cuts off left - right oscillations and allows the left - handed neutrinos to oscillate among themselves as normal . at low energies ,
however , the @xmath10 terms dominate , and oscillations into sterile neutrinos are predicted . observe that @xmath18 so that three of the eigenvalues have the expected @xmath19 energy dependence at high energies .
transition probabilities can be calculated exactly .
for example , @xmath20\end{aligned}\ ] ] where @xmath21\ ] ] and @xmath22\ ] ] the mixing angles and two of the eigenvalues are determined by the high - energy , mass - like behavior widely detected .
there remain four independent parameters in the model , which can be adjusted to control the low energy behavior without disrupting the high - energy limit . as energy decreases ,
the probabilities diverge smoothly from the standard mass predictions .
the model introduced above ( 6 ) will never produce observable cpt violations .
this is because @xmath10 is a cpt - odd variable , but @xmath11 is cpt - even , so that under cpt transformations , @xmath23 goes to @xmath24\ ] ] however , the eigenvalues and mixing angles on which the probability depends do not observe the sign of @xmath10 , as can be seen from their definitions ( 8 , 11 , 12 ) .
this causes the probabilities to be the same whether @xmath10 or @xmath25 is used .
in fact , even if @xmath10 does not commute with @xmath11 , cpt symmetry will still be preserved ; to introduce cpt violation , @xmath10 and @xmath11 terms must be mixed ( for example , an ordinary rotation of ( 6 ) will introduce cpt violations ) .
this model is intended to show that behavior typical of mass models can be reproduced by a lorentz - violating model without mass .
a variety of low - energy behavior is consistent with the same behavior at high energy .
however , it remains difficult to explain all experiments , even with four free parameters .
we would like to thank swarthmore college for funding the research reported here .
thanks also to matt mewes for his advice and assistance in this project .
xx lsnd collaboration , a. aguilar , phys .
d * 64 * , 112007 ( 2001 ) .
miniboone collaboration , a. a. aguilar - arevalo , phys .
lett . * 103 * , 111801 ( 2009 ) .
p. vahle ( minos collaboration ) , neutrino 2010 plenary talk , june 14 , 2010 ; to be published in the proceedings .
v. alan kosteleck and matthew mewes , phys . rev .
d * 69 * , 016005 ( 2004 ) . v. alan kosteleck and matthew mewes , phys .
d * 70 * , 031902 ( 2004 ) .
teppei katori , v. alan kosteleck and rex tayloe , phys .
d * 74 * , 105009 ( 2006 ) . c. amsler , phys .
. * b667 * , 1 ( 2008 ) . | a new model for neutrino oscillations is introduced , in which mass - like behavior is seen at high energies , but various behavior can be predicted at low energies . the model employs no neutrino masses , but instead relies on the lorentz - violating parameters @xmath0 and @xmath1 .
oscillations into sterile neutrinos and into antineutrinos are also considered . |
the minimal supersymmetric standard model ( mssm ) may be described by a lagrangian containing interactions consistent with invariance under the gauge group @xmath3 and global supersymmetry plus a lagrangian containing a restricted set of soft supersymmetry breaking terms @xcite .
these terms break supersymmetry while maintaining a useful property of a supersymmetric theory , namely the cancellation of quadratic divergences @xcite .
the absence of these divergences is necessary in order to define the renormalized mass of a fundamental scalar , such as the higgs boson , without a fine - tuning of the cancellation between the bare mass and the scalar self - energy @xcite .
the presence of fundamental scalar fields in the mssm , besides the higgs bosons , leads to the possibility that these fields may acquire non - zero vacuum expectation values ( vevs ) .
since this would violate the conservation of color and/or electric charge symmetry , this leads to forbidden regions of the parameter space of the theory .
we will calculate numerical estimates of the boundary of the allowed region of soft - breaking parameters using both the tree - level potential and the one - loop effective potential .
many studies of the mssm mass spectrum neglect these charge / color breaking , or ccb , bounds in their analyses .
previously , ccb bounds were obtained for various supersymmetric models , however no systematic numerical study of ccb constraints for a realistic approximation to the mssm using the one - loop effective potential has been done @xcite .
one may assume that there are relations among the soft breaking terms , such as in the minimal supergravity model in which all scalar masses and scalar trilinear couplings are the same at the unification scale , of order @xmath4 @xcite .
however we will find constraints on the soft - breaking parameters at a low - energy scale , @xmath5 , with @xmath6 .
this is an indeterminate upper limit on particle masses if the mssm is to explain the gauge hierarchy problem .
we will not make any assumptions about the theory near the gut scale nor the particle spectrum above @xmath5 .
we will use an approximation to the mssm that includes only the top flavor supermultiplets .
this follows from evidence that the top quark mass @xmath7 @xcite .
we use the conventional definition @xmath8 , with @xmath9 , @xmath10 the vevs for the higgs scalar fields , @xmath11 and @xmath12 , respectively . assuming a small value for @xmath1 , near @xmath13 , gives the top quark yukawa coupling , @xmath14 . the contributions from the bottom supermultiplets
may then be ignored .
there are various reasons to choose these particular values of @xmath1 and to consider only the top squarks as acquiring a non - zero vev .
first of all , there is an infrared quasi - fixed point in the renormalization group equation for @xmath15 which corresponds to a value @xmath16 @xcite .
the mass relation m_t(m_t ) = h_t^fp(m_t)(2)vgives @xmath17 if one uses the relation between the top quark mass defined by the pole in its propagator and its running mass to first - order in the qcd gauge coupling @xcite .
m_pole = m_t(m_t ) .
therefore a value of @xmath1 at @xmath5 in the range @xmath18 results from a large range of @xmath19 values at the gut scale .
although @xmath1 is not required to be in this range , it indicates that this is a natural choice .
one motivation for considering only the top sector comes from assuming common soft - breaking parameters at the gut scale .
a large value of @xmath19 causes the third generation parameters to undergo the largest change as they are evolved from @xmath20 down to @xmath5 . for this same reason
, @xmath19 also gives the largest contribution to the radiative gauge symmetry breaking of @xmath21 @xcite .
therefore , if one assumes that the minimum of the effective potential at energy scales @xmath22 gives zero vevs for the scalar fields , such as in the case of universality at the gut scale , as one evolves to @xmath5 the third - generation parameters undergo the largest change and the ccb constraints from third generation scalar fields will be the most restrictive .
finally , as discussed in ref .
@xcite , the potential barrier height for tunneling from the the symmetric vacuum at a high temperature ( @xmath23 ) , early in the expansion of the universe , to a lower ccb minimum is proportional to @xmath24 where @xmath25 is the smallest of the yukawa couplings for the slepton and squark fields that have non - zero vevs at the ccb minimum .
this implies that one should consider ccb vacua in which only the higgs fields and the top squarks have non - zero vevs in order for the tunneling from the symmetric to the ccb vacuum to have occurred in a time less than the present age of the universe ( @xmath26 years ) .
we use a consistent approximation to the mssm with @xmath27 as a small value near the fixed point value and interactions with the bottom quark superfields are ignored .
we use all mssm interactions between the following fields @xmath11 , @xmath12 , @xmath28 , @xmath29 , @xmath30 , @xmath31 , @xmath32 , @xmath33 , @xmath34 , and @xmath35 .
@xmath11 and @xmath12 are respectively the hypercharge @xmath36 and @xmath37 higgs boson doublets .
the corresponding field variables with a tilde are the higgsino doublets . @xmath30 and @xmath31 are the left - handed component of the top quark and the right - handed component of the charge conjugate top quark field respectively . again , the corresponding field variables with tildes are the top squarks .
@xmath34 is the gluon field and @xmath35 is the gluino field .
notice that the field content in this approximation is supersymmetric .
this arises from including all interactions with the top quark supermultiplet involving the parameters @xmath19 and @xmath38 .
the potential in this approximation as well as the definitions of the parameters appearing in it are shown in the appendix .
we use the values of the gauge couplings at the weak scale @xmath39 : @xmath40 , @xmath41 , and @xmath42 for all calculations @xcite .
since we will take the renormalization scale to be @xmath43 , the running of the gauge couplings from @xmath39 to this scale is negligible .
we also omit the couplings @xmath44 and @xmath45 in the calculation of the one - loop effective potential .
however , we retain the terms with these couplings at tree - level since they include the quartic higgs scalar interactions responsible for the non - zero higgs vev of the standard model ( sm ) .
quantum corrections may affect whether spontaneous symmetry breaking occurs in a field theory .
the vevs of the scalar fields are the values of the classical fields at the minimum of the effective potential .
we use the one - loop correction to the effective potential in the dimensional reduction , @xmath46 , renormalization scheme @xcite [ one - loop veff ] v_eff^1-loop = 1str , where the supertrace , @xmath47 , is over all color and spin degrees of freedom with a minus sign for fermions .
@xmath48 is the square of the renormalization scale , which we take to be equal to @xmath49 .
@xmath50 is the field dependent mass matrix ; for instance , for a theory with scalar fields @xmath51 _ ij = .^2v |_=_c . @xmath52 as well as @xmath53
are function of the classical fields @xmath54 .
since the top yukawa coupling , @xmath19 , and the strong gauge coupling , @xmath55 , are large one may expect significant contributions to the effective potential from one - loop corrections .
the one - loop effective potential is also more stable under a change of renormalization scale @xcite . since the parameters appearing in the equation for the one - loop correction to @xmath53 are the renormalized parameters we consider the resulting ccb constraints as being limits on these renormalized parameters in the @xmath56 scheme at momentum scale @xmath5 .
if large logs of ratios of massive parameters appear in the effective potential , such as @xmath57 or @xmath58 , then one may , in principle , sum these using the renormalization group equation @xcite . however , this is difficult when there are multiple scalar fields and masses .
we only consider masses , @xmath59 , and the renormalization scale , @xmath5 , which differ by less than two orders of magnitude .
furthermore , since we are only interested in the effective potential near its minimum where @xmath60 , i.e. , the classical field near the minimum of @xmath53 is of the same order of magnitude as the masses , the logarithms appearing in @xmath53 are small .
renormalization group improvement of the effective potential in this case is unnecessary .
we also assume that the lagrangian parameters are real . in a field basis where the higgs fields are real and positive and @xmath19 and the scalar masses , @xmath61 , real , only one of the parameters @xmath0 , @xmath38 , or @xmath62 may be made real by redefining the fields . however ,
it was shown in ref .
@xcite that the presence of complex phases in these parameters , greater than about @xmath63 , gives too large of a contribution to the neutron s electric dipole moment due to gluino loops .
we evaluate the one - loop effective potential using eq .
( [ one - loop veff ] ) by diagonalizing the mass matrix @xmath52 .
if one chooses the landau gauge for the gluons , in which the propagator is - , the generalized mass matrix @xmath52 is block diagonal in the gluon fields , bosonic fields , and fermionic fields .
the top squark fields may also be rotated by a global @xmath64 transformation so that they are of the form _
i = ( c q_1r + 0 + 0 ) , _ i = ( c t_1r + t_1i + t_2r + 0 ) , with @xmath65 , @xmath66 , @xmath67 , and @xmath68 are real .
we then use only the real part of the one - loop effective potential ; the imaginary part is related to the decay rate of the vacuum as shown in ref .
@xcite .
the global minimum of the effective potential is found by calculating the local minimum using a standard algorithm from ref .
@xcite starting from a set of field values on a rectangular grid .
if this global minimum occurs for non - zero values of the squark fields then there is a ccb vacuum .
this process is repeated for a set of soft - breaking parameter values on a rectangular grid in parameter space .
a quadratic surface is then fit to the parameter points on the boundary between those that give a symmetric vacuum and those that lead to a ccb vacuum .
the surface is fit to the boundary points by varying the coefficients in the equation of the surface , @xmath69 , until the average of the distance squared from the boundary points to the nearest point on the surface , @xmath70 , is minimized .
the equation of the surface is [ quadratic_surface ] a_t^4+p_1a_t^2+p_2m_2 ^ 4+p_3m_2 ^ 2 + p_4m_^4+p_5m_^2+p_6m_^4+p_7m_^2 -p_8=0 with @xmath71 a point in parameter space .
these soft - breaking parameters are defined in the appendix .
this particular parameterization of the surface is chosen solely because of its simplicity , but as will be seen later it gives an accurate characterization of the ccb boundary .
because of the large dimension of the parameter space in which we want to find ccb bounds , it is necessary to initially constrain some of the parameters values in order to reduce the calculation time .
two of the soft - breaking parameters , @xmath72 and @xmath73 , are constrained using two different methods . in method
s the values of these parameters are chosen such that there is a standard model minimum in the effective potential with @xmath74 and @xmath75 , @xmath76 with @xmath77 .
this is done by solving the analytical expression for the one - loop effective potential with zero vevs for the top squarks in terms of these parameters .
these relations are shown in eq .
( [ one_loop_m1m3 ] ) . in the other procedure ,
method f , we simply fix @xmath72 and @xmath73 at constant values , i.e. , independent of the other soft - breaking parameters .
we begin with a numerical analysis of the model with three real scalar fields @xmath78 , @xmath79 , and @xmath80 of ref .
the potential in this model is @xmath81 \nonumber\end{aligned}\ ] ] this is a simplified version of the potential described in the appendix with @xmath82 , @xmath83 , and @xmath84 .
the ccb bound [ analytical_bound ] a^2 < 3(m_2 ^ 2 + m_q^2 + m_t^2 ) , derived in ref .
@xcite follows from minimizing the potential only in the equal - field direction , _
i.e. _ , @xmath85 .
this is valid , in general , only if the d - term dominates , or if @xmath86 with @xmath87 the smallest gauge coupling .
we perform a numerical analysis of the potential in eq .
( [ ghs_pot ] ) using the parameter value ranges shown in table [ ghs_range ] and with @xmath88 .
the gauge couplings are fixed at their running values at the weak scale @xmath39 as stated earlier .
the resulting best fit to the ccb boundary is shown in table [ ghs_fits ] . clearly , this is close to the analytical result of eq .
( [ analytical_bound ] ) , as it should be with a small value for @xmath89 .
next we repeat the analysis for the potential of eq .
( [ ghs_pot ] ) with the same soft - breaking parameter ranges of table [ ghs_range ] , but we set @xmath90 .
this is the proper value for the top quark yukawa coupling for small @xmath1 .
the corresponding ccb bound is given in table [ ghs_fits ] .
the ccb bound for @xmath90 is more stringent than the bound for @xmath88 , i.e. , for a given set of parameter values within the range given in table [ ghs_range ] , @xmath91 , the value of the @xmath92 parameter at which the vacuum becomes ccb is lower .
the fractional difference in the @xmath92 parameter bound between these cases is largest when the other parameters , @xmath93 , are near the lower end of their range in table [ ghs_range ] , where @xmath94 .
also the value of @xmath92 at the ccb bound in this case is small .
the average fractional difference over the entire parameter ranges shown in table [ ghs_range ] for the two bounds is @xmath95 next we examine the ccb bounds obtained from of our approximation to the mssm with the lagrangian given in the appendix . in all of the following analyses the mass units are @xmath96 .
we also fix @xmath97 and @xmath98 .
the renormalization scale for the one - loop calculation of the effective potential is set at @xmath99 .
we first examine the bound for method f with @xmath100 and @xmath101 .
these masses result in a value of @xmath102 if a s.m .
minimum were required , i.e. , zero squark vevs .
these values for @xmath72 and @xmath73 were chosen since they are consistent with a s.m .
minimum for @xmath97 and they are close to the renormalization scale , @xmath103 .
the ranges of parameter values for which the effective potential was calculated are shown in table [ mssm_range ] .
the ccb bounds for the analyses of both the tree - level and the one - loop effective potential are shown in table [ method_f_fits ] .
parameter values which give a potential that is unbounded below are not used in determining the best fit surface for the ccb bound .
this includes values with @xmath104 for the tree - level potential since in this case the potential is unbounded for @xmath105 and zero squark vevs .
we do not examine negative @xmath106 values since in this case , for the small range of @xmath107 and @xmath108 values for which the potential is bounded below , the vevs for the c.c.b .
vacuum are too small for the numerical methods to distinguish this minimum from the minimum with all vevs vanishing . however , according to eq .
( [ higgs_mass_def ] ) , the higgs soft - breaking mass , @xmath109 , is negative over part of the parameter range examined .
next we calculate the ccb bounds using method s , in which @xmath72 and @xmath73 are fixed by requiring a s.m .
the parameter values listed in table [ mssm_range ] were used in determining the bounds .
since there is no standard model minimum for @xmath110 , we do not examine @xmath106 values in this range .
the ccb bounds calculated by finding the global minimum of the both the tree - level potential and the one - loop effective potential are given in table [ method_s_fits ] .
the quadratic surface of eq .
( [ quadratic_surface ] ) is sufficient to provide an accurate characterization of the numerical ccb bound .
all of the least - square fits give the average distance squared , @xmath111 .
the one - loop ccb bound calculations yield a larger value for @xmath112 because the longer calculation time requires using less parameter grid points than for the tree - level calculation .
typically it takes around @xmath113 days of cpu time on a digital alpha workstation to calculate the one - loop ccb boundary points and perform a least - squares fit so we are limited by the computation time . as stated in section [ simplified model ] , the numerical ccb bound for the simplified lagrangian of ref .
@xcite with yukawa coupling @xmath114 is significantly different from the numerical bound with @xmath115 , and hence also different from the analytical bound of eq .
( [ analytical_bound ] ) , when the other soft - breaking parameters are small . however , for the remainder of the parameter values tested , the numerical bound with @xmath114 is quite close to the analytical bound .
one would not expect the ccb bound of eq .
( [ analytical_bound ] ) to be correct for large @xmath89 .
one possible explanation is that the @xmath116 d - term is large enough at least to insure that the minimum is in the direction @xmath117 .
also for @xmath114 the numerical bound does give a more stringent ccb bound than that of eq .
( [ analytical_bound ] ) over the entire range of parameters tested .
we present some contour plots of the ccb bounds with @xmath118 for comparison .
the contours show the value of @xmath0 on the ccb boundary , i.e. , lower values of @xmath0 result in a symmetric vacuum whereas higher values result in a ccb vacuum with nonzero squark vevs . figs .
[ tree_f_fig ] and [ loop_f_fig ] show the bounds for method f using the tree - level and one - loop effective potential respectively
[ tree_s_fig ] and [ loop_s_fig ] show the tree - level and one - loop method s ccb bounds . fig .
[ analytical_fig ] shows the analytical bound of eq .
( [ analytical_bound ] ) with @xmath118 .
since there is not the additional constraint of requiring a standard model minimum for the higgs field for both the method f ccb bounds and for the analytical bound , one may compare these ccb bounds .
the tree - level ccb bound on the @xmath0 parameter for method f is lower than the analytical bound of eq .
( [ analytical_bound ] ) for the entire range of parameter values , @xmath91 , shown in table [ mssm_range ] .
the one - loop correction to the effective potential raises the value of @xmath0 for the ccb bound over about @xmath119 of the parameter range considered .
however , even with the one - loop corrections , the ccb bound for the mssm potential is more stringent , i.e. , gives a lower value for @xmath0 , than the bound of eq .
( [ analytical_bound ] ) for @xmath120 of the parameter range .
the one - loop corrections for the ccb bound calculated using method s give a lower @xmath0 value than the tree - level bound over the entire range of parameters examined .
for values of the parameters , @xmath121 , that give a small value for the @xmath0 parameter ccb bound , the one - loop corrections to the effective potential makes the ccb bound significantly stricter . in conclusion ,
ccb bounds on the soft - breaking parameters of the higgs and top quark / squark sectors of the mssm provide important constraints for these parameters .
these constraints may be expressed as a maximum value of the @xmath0 parameter for given values of the remaining soft - breaking parameters .
the numerical ccb constraints that we calculated give more stringent ccb bounds than the analytical constraint of eq .
( [ analytical_bound ] ) for most of the ranges of parameter values considered . because of the large top yukawa coupling , one - loop corrections to the effective potential may result in significantly different ccb bounds than those for the tree - level potential
the author wishes to thank y. okada for suggesting this topic and b. wright for many useful discussions .
this work was partially supported by the japan society for the promotion of science .
the potential may be divided into several parts . a sum over group and spinor indices , where applicable , is implied . the supersymmetric d - terms are @xmath122 where @xmath44 , @xmath45 , and @xmath55 are respectively the @xmath123 , @xmath124 , and @xmath64 couplings , @xmath125 are the pauli matrices and @xmath126 are the antihermitian generators of @xmath64 . using the relations _ ij^a_kl^a = 2_il_jk - _ ij_kl and t_ij^at_kl^a = ( _ il_jk -_ij_kl ) the @xmath124 contribution becomes @xmath127 \nonumber\end{aligned}\ ] ] and the @xmath64 one is v_su(3 ) = g_3 ^ 2 .
the superpotential or f term is v_f = h_t^2(||^2|h_2 ^ 0|^2 + ||^2|h_2 ^
0|^2 + ||^2 ) + h_t(h_1 ^ 0 * ) + h.c . , with @xmath128 denoting the hermitian conjugate and @xmath129 and @xmath130 are the neutral components of the higgs scalar doublets .
the higgs scalar and fermion doublet components are h_1 = ( ) .
the quark - squark - gluino interaction terms are v_q = g_3(p_l^(a)t^a^ * -t^a^(a ) p_rt + ^*t^ap_lt - t^ap_r^(a ) ) .
@xmath131 are the projection operators for left- and right - handed chiral spinors , @xmath132 , @xmath133 is the four component spinor field for the top quark , and @xmath134 are the majorana gluino fields .
the quark - squark - higgsino terms are v_q = h_t(|_2 ^ 0p_lt + ^*p_r_2 ^ 0 - p_l_2 ^ 0 - ^*|_2 ^ 0p_rt ) .
the higgsino interaction terms are v _
= ( _ 1 ^ 0_2 ^ 0 - _ 1 ^ -_2^+ ) + h.c
. finally the susy soft - breaking terms are @xmath135 with the addition of the supersymmetric higgs interactions , the masses for @xmath11 and @xmath12 become @xmath136 respectively .
if the squark vevs are zero the one - loop contributions to the higgs effective potential from top squark and quark loops may be written in an analytical form @xcite .
after requiring that the minimum of the effective potential be at @xmath137 and @xmath138 and solving for @xmath72 and @xmath73 we obtain @xmath139 , \nonumber \\
m_{1}^{2 } & = & m_{3}^{2}\tan\beta - m_{z}^{2}\cos 2\beta \nonumber \\ & + & { 3\over{16\pi^{2}}}h_{t}^{2}\mu(a_{t}\tan\beta + \mu ) { { f(m_{1}^{'2})-f(m_{2}^{'2})}\over{m_{2}^{'2}-m_{1}^{'2 } } } , \nonumber\end{aligned}\ ] ] with @xmath140 .
the definition @xmath141 and the tree - level z boson mass , @xmath142 , were used .
the tree - level relation follows by including only the first term in the above equations for @xmath72 and @xmath73 .
p. fayet , phys .
b * 69 * ( 1977 ) 489 ; j. rosiek , phys . rev .
d * 41 * ( 1990 ) 3464 .
l. girardello , and m.t .
grisaru , nucl .
* b194 * ( 1982 ) 65 .
nilles , phys . rep . * 110 * ( 1984 ) 1 .
frere , d.r.t .
jones , s. raby , nucl .
* 222 * ( 1983 ) 11 ; l. alvarez - gaum , j. polchinski , m.b .
wise , _ ibid .
_ * 221 * ( 1983 ) 495 ; c. kounnas , a.b .
lahanas , d.v .
nanopoulos , m. quiros , _ ibid . _
* 236 * ( 1984 ) 438 ; j.p .
derendinger and c.a .
savoy , _ ibid . _
* 237 * ( 1984 ) 307 ; h. komatsu , phys .
b * 215 * ( 1988 ) 323 ; p. langacker and n. polonsky , phys .
d * 50 * ( 1994 ) 2199 .
gunion , h.e .
haber , and m. sher , nucl .
b306 * ( 1988 ) 1 .
l. alvarez - gaum and j. polchinski , nucl . phys .
* b221 * ( 1983 ) 495 .
cdf collaboration , phys .
* 73 * ( 1994 ) 225 ; _ ibid .
_ , `` observation of top quark production in pp collisions with the cdf detector at fermilab '' , preprint fermilab - pub-95/022-e , hep - ex/9503002 . b. pendleton and g.g .
ross , phys .
b * 98 * ( 1981 ) 291 ; c.t .
hill , phys .
b * 98 * ( 1981 ) 291 ; v. barger , m. berger , p. ohmann , and r.j.n .
phillips , phys . lett .
b * 314 * ( 1993 ) 351 .
r. tarrach , nucl . phys .
* b183 * ( 1981 ) 384 .
ibez and g.g .
ross , phys .
b * 110 * ( 1982 ) 215 ; k. inoue , a. kakuto , h. komatsu , and s. takeshita , prog .
* 68 * ( 1982 ) 96 ; l. alvarez - gaum , m. claudson , and m.b .
wise , nucl .
b207 * ( 1982 ) 96 .
g. gamberini , g. ridolfi , and f. zwirner , nucl .
b331 * ( 1990 ) 331 .
m. claudson , j. hall , and i. hinchliffe , nucl .
b228 * ( 1983 ) 501 .
particle data group , phys .
d * 50 * ( 1994 ) 1173 .
w. siegel , phys .
b * 84 * ( 1979 ) 193 ; d.m .
capper , d.r.t .
jones , and p. nieuwenhuizen , nucl
b167 * ( 1980 ) 479 .
m.b . einhorn and d.r.t .
jones , nucl .
b230 * ( 1984 ) 261 ; h. nakano and y. yoshida , phys .
d * 49 * ( 1994 ) 5393 ; c. ford , _ ibid .
_ 50 ( 1994 ) 7531
. j. polchinski and m. wise , phys .
b * 125 * ( 1983 ) 393 .
e. weinberg and a. wu , phys .
d * 36 * ( 1987 ) 2474 . w. press , b. flannery , s. teukolsky , and w. vetterling , _ numerical recipes - the art of scientific computing _ , ( cambridge university press , cambridge , 1986 ) .
j. ellis , g. ridolfi , and f. zwirner , phys .
b * 257 * ( 1991 ) 83 ; j. ellis _
_ , _ ibid .
_ * 262 * ( 1991 ) 477 ; a. brignole , j. ellis , g. ridolfi , f. zwirner , _ ibid . _
* 271 * ( 1991 ) 123 .
c|dd parameter & & + @xmath145 & 0.7246 & 2.723 + @xmath146 & 0.1608 & 3.527 + @xmath147 & -3.808 & -8.310 + @xmath148 & 0.1991 & 3.370 + @xmath149 & -3.830 & -8.228 + @xmath150 & 0.1985 & 3.452 + @xmath151 & -3.836 & -8.310 + @xmath152 & 0.4911 & 0.5877 + @xmath153 & 2.@xmath154 & 4.@xmath155 cdddd parameter & min .
value & max .
value & tree - level grid spacing & one - loop grid spacing + @xmath0 & 0.0 & 3.0 & 0.1 & 0.4 + @xmath106 & 2.@xmath156 & 0.7525 & 0.125 & 0.25 + @xmath107 & 2.@xmath156 & 0.7525 & 0.125 & 0.25 + @xmath108 & 2.@xmath156 & 0.7525 & 0.125 & 0.25 c|dd parameter & & + @xmath145 & 6.@xmath157 & 5.@xmath158 + @xmath146 & 3.@xmath159 & 1.@xmath160 + @xmath147 & -3.@xmath161 & -4.@xmath162 + @xmath148 & 6.@xmath163 & 6.@xmath164 + @xmath149 & -1.@xmath165 & -1.@xmath166 + @xmath150 & 6.@xmath167 & 7.@xmath168 + @xmath151 & -1.@xmath169 & -1.@xmath170 + @xmath152 & -1.@xmath171 & -1.@xmath172 + @xmath153 & 1.@xmath173 & 4.@xmath174 c|dd parameter & & + @xmath145 & 4.@xmath175 & 3.@xmath176 + @xmath146 & 9.@xmath177 & -1.@xmath178 + @xmath147 & -1.@xmath179 & 5.@xmath180 + @xmath148 & 3.@xmath181 & 7.@xmath182 + @xmath149 & -6.@xmath183 & -9.@xmath184 + @xmath150 & 3.@xmath185 & -7.@xmath186 + @xmath151 & -6.@xmath187 & 3.@xmath188 + @xmath152 & -2.@xmath189 & -2.@xmath190 + @xmath153 & 1.@xmath191 & 9.@xmath192 | we calculate limits on the trilinear soft - breaking parameter , @xmath0 , in the minimal supersymmetric standard model by requiring the absence of nonzero top squark vacuum expectation values . assuming a low @xmath1 , which implies a large top yukawa coupling
, we also calculate one - loop corrections to the effective potential .
the resulting numerical calculations of the charge / color breaking limits are presented as best - fit surfaces .
we compare these results with the analytical limit , @xmath2 , and find that although this is a good estimate of the charge / color breaking bounds for a simplified model of the top sector , stricter bounds are found by a numerical minimization of the minimal supersymmetric standard model potential . |
the infrared astronomical satellite ( iras , * ? ? ? * ; * ? ? ? * ) was extremely successful in characterizing for the first time the global properties of the mid- and far - infrared sky , carrying out an all - sky survey at wavelengths of 12 , 25 , 60 and 100 @xmath0 m and leading to discoveries such as those of luminous , ultraluminous and hyperluminous infrared galaxies ( lirgs , ulirg and hlirgs , respectively ) , a substantial population of evolving starbursts and the detection of large - scale structure in the galaxy distribution @xcite .
unfortunately , the iras view was typically limited to the very local universe ( @xmath2 ) , thus hampering statistical studies of infrared - luminous galaxies at cosmological redshifts .
only few sources were detected by iras at higher redshifts , typically ulirgs magnified by gravitational lenses , like f10214 + 4724 ( @xmath3 , * ? ? ? * ) .
in particular only about 1000 galaxies were detected all over the sky in iras 12 band .
infrared source counts based on iras data @xcite showed some marginally significant excess of faint sources with respect to no evolution models @xcite , but not enough statistics and dynamic range in flux to discriminate between evolutionary scenarios were available . although conceived as an observatory - type mission , the infrared space observatory ( iso , * ? ? ?
* ) was in many ways the natural successor to iras , bringing a gain of a factor @xmath4 in sensitivity and @xmath5 in angular resolution in the mid - infrared . a substantial amount of iso observing time was therefore devoted to field surveys aimed at detecting faint infrared galaxies down to cosmological distances .
such surveys were conceived as complementary in flux depth and areal coverage , allowing a systematic investigation of the extragalactic sky down to so far unattainable flux densities at both mid and far infrared wavelengths , whose results are summarized by @xcite .
in particular , extragalactic 15 source counts determined with isocam @xcite have revealed a significant departure from euclidean slope within the 1 - 5 mjy flux range , which has been interpreted as evidence for a strongly evolving population of starburst galaxies .
the european large area iso survey ( elais , * ? ? ?
* ; * ? ? ?
* ) was the most ambitious non - serendipitous survey and the largest open time project carried out with iso , aimed at bridging the flux gap between iras all - sky survey and iso deeper surveys .
elais observations mapped areas of about 12 at 15 and 90 @xmath0 m and smaller areas at 7 and 175 @xmath0 m with the isocam ( * ? ? ?
* and 15 @xmath0 m ) and isophot ( * ? ? ?
* and 175 @xmath0 m ) cameras .
most importantly , elais 15 @xmath0 m observations are the only ones allowing to sample the 1 - 5 mjy flux range , where most of the source evolution appears to take place . since the project approval ,
the elais consortium , grown in time to a total of 76 collaborators from 30 european institutes , has undertaken an extensive program of ground - based optical and near - infrared imaging and spectroscopy .
thanks to such an extensive multi - wavelength coverage , the elais fields have now become among the best studied sky areas of their size , and natural targets of on - going or planned large - area surveys with the most powerful ground- and space - based facilities .
further details on elais multi - wavelength observations and catalogues are presented in @xcite . after the loss of the wire satellite , notwithstanding the observations at several infrared wavelengths soon to come from spitzer and later from sofia and herschel
, iso observations will remain a valuable database for many years to come .
in particular , until the advent of jwst , elais 15 observations will provide a complementary view on three areas ( s1 , n1 and n2 ) which will be covered at different wavelengths as part of the spitzer wide - area extragalactic survey ( swire , * ? ? ? * ) .
thus the need of reducing such data with the uttermost care and provide the community with an agreed - upon legacy from the elais project .
this paper presents the final analysis of elais 15 observations , and is structured as follows . in section [ elais15.sec ] a brief description of the most relevant aspects of elais 15 @xmath0 m dataset is given .
section [ datared.sec ] describes the data reduction method and its improvements . in section [ autosim.sec ]
the technique employed for flux determination and its results are presented .
section [ simulations.sec ] details the results of the simulations that were carried out in order to assess the performance of the data reduction method and thus the quality of the resulting catalogue . in sections
[ astroacc.sec ] and [ photoacc.sec ] , respectively , estimates of the achieved astrometric and photometric accuracy are given .
section [ optids.sec ] summarizes the identification of 15 @xmath0 m sources in optical and near - infrared images , while section [ photocal.sec ] describes the procedure adopted to establish the catalogue photometric calibration .
finally , section [ catalogue.sec ] describes gives a basic description of the catalogue contents .
the elais 15 @xmath0 m main dataset is made up of 28 rasters ( iso basic imaging observations ) , each covering an area of about @xmath6 , divided into 4 fields , one ( s1 ) in the southern hemisphere and three ( n1 , n2 and n3 ) in the northern one .
small superpositions at the boundaries and a limited degree of redundancy on portions of the fields give a total covered area of 10.85 .
the fields were selected on the basis of their high ecliptic latitude ( @xmath7 , to reduce the impact of zodiacal dust emission ) , low cirrus emission ( @xmath8 ) and absence of any bright ( @xmath9 jy ) iras 12 @xmath0 m source . in figure [ fields.fig ] the location on the sky of the survey fields is shown , overlaid on cirrus maps ( cobe normalized iras maps of @xcite ) .
nearby iras sources with 12 @xmath0 m fluxes brighter than are also plotted . the overall sky coverage , highlighting the position and redundancy of single rasters , is illustrated in figure [ coverage.fig ] . + + elais 15 @xmath0 m observations were carried out operating the isocam instrument in raster mode using the lw3 filter . in this observing mode ,
the isocam @xmath10 pixel lw detector was stepped across the sky in a grid pattern , with about half detector width steps in one direction and the whole detector width steps in the other .
thus , reliability was improved as each sky position ( apart from those at the boundaries of the raster region ) was observed twice in successive pointings and overheads were reduced because each raster covered a large area ( @xmath11 ) . at each raster pointing ( i.e. grid position ) the detector was read out several ( typically 10 ) times , to increase the redundancy in order to be able to identify cosmic rays impacts and distinguish their severe effects on the electronics from real sources .
furthermore , on the raster first pointing , 80 readouts were carried out to allow the detector to approach stabilization .
table [ lw3pars.tab ] describes the observation parameters for the lw3 observations , while table [ 15mumfields.tab ] lists the fields and related iso data archive filenames making up the dataset .
lc parameter & value + band @xmath12 & 14.3 @xmath0 m + band fwhm range & 12.018.0 @xmath0 m + detector gain & 2 @xmath13 + integration time & 2 s + number of exposures per pointing & 10 + number of stabilization exposures & 80 + pixel field of view & 6@xmath14 + number of pixels & 32 @xmath15 32 + number of horizontal and vertical steps & 28 , 14 + number of rasters ( including repetitions ) & 28 + horizontal and vertical step sizes & 90@xmath16 , 180@xmath14 + total area & 10.85 + ccccc field & raster & ida & ra ( j2000 ) & dec ( j2000 ) + s1 & s1_1 & 23200251 & 00 30 25.40 & -42 57 00.3 + & s1_2 & 23200353 & 00 31 08.20 & -43 36 14.1 + & s1_3 & 41300955 & 00 31 51.90 & -44 15 27.0 + & s1_4 @xmath13 & 23300257 @xmath13 & 00 33 59.40 & -42 49 03.1 + & s1_5_a & 23300459 & 00 34 44.40 & -43 28 12.0 + & s1_5_b & 77500207 & 00 34 44.40 & -43 28 12.0 + & s1_5_c & 78502406 & 00 34 44.40 & -43 28 12.0 + & s1_6 @xmath13 & 41001161 @xmath13 & 00 35 30.40 & -44 07 19.8 + & s1_7 & 40800663 & 00 37 32.50 & -42 40 41.2 + & s1_8 & 40800765 & 00 38 19.60 & -43 19 44.5 + & s1_9 & 41001867 & 00 39 07.80 & -43 58 46.6 + n1 & n1_1 & 30200101 & 16 15 01.00 & + 54 20 41.0 + &
n1_2_a @xmath13 & 30400103 @xmath13 & 16 13 57.10 & + 54 59 35.9 + & n1_2_b & 67200103 & 16 13 57.10 & + 54 59 35.9 + & n1_3 & 30500105 & 16 10 34.90 & + 54 11 12.7 + & n1_4 & 30600107 & 16 09 27.00 & + 54 49 58.7 + & n1_5 & 31000109 & 16 06 10.80 & + 54 01 08.0 + & n1_6 & 30900111 & 16 04 59.00 & + 54 39 44.3 + n2 & n2_1 & 50200119 & 16 32 59.80 & + 41 13 33.2 + & n2_2 & 51100131 & 16 34 44.50 & + 40 38 45.0 + & n2_3 & 50000723 & 16 36 05.50 & + 41 33 11.8 + & n2_4 & 50200225 & 16 37 48.90 & + 40 58 13.1 + & n2_5 & 50100727 & 16 39 13.80 & + 41 52 31.6 + & n2_6 & 50200429 & 16 40 55.50 & + 41 17 22.7 + & n2_r_a & 11600721 & 16 35 45.00 & + 41 06 00.0 + & n2_r_b & 77900101 & 16 35 45.00 & + 41 06 00.0 + n3 & n3_3 & 42500237 & 14 29 38.30 & + 33 24 49.6 + & n3_5 & 43800341 & 14 32 38.20 & + 33 11 10.3 +
reduction of data obtained with iso instrumentation has always proved very difficult for a number of reasons . as far as isocam observations carried out using its long wavelength ( lw ) detector are concerned , the two most important instrumental phenomena one has to deal with are the qualitatively very different effects produced on the detector s electronics by the frequent and severe cosmic ray impacts , which have long been known and referred to as * glitches * , and its sizable transient behaviour after changes in the incident photon flux , which we will hereafter simply refer to as * transients*. in both cases , the cryogenic operational temperatures of the detector caused it to very slowly respond after these events .
lack of an accurate modeling of these effects can thus lead to spurious detections or errors in flux determination . in elais 15 @xmath0 m data reduction
, the impact of these effects is increased by the instrumental parameters which were chosen in order to maximize the survey area . the short integration time ( @xmath17 ) , the large raster step ( half the detector size along one axis and the whole detector size along the other one ) and the large pixel size ( 6 arcsec ) all contribute to reduce the redundancy and to increase the undersampling
. low redundancy and high undersampling , in turn , increase difficulties in distinguishing sources from strong glitches ( low reliability ) and in correctly determining source fluxes ( low photometric accuracy ) .
roughly speaking , glitches can be divided into three categories according to the way they shape the detector s output signal , their decay time and influence on the pixel responsivity : glitches belonging to these different classes are respectively dubbed * common glitches * , * faders * and * dippers*. slow decreases of the signal following cosmic ray impacts are called faders , while prominent reductions of the pixel responsivity very slowly recovering afterwards are called dippers .
these two effects are believed to be associated with proton or @xmath18 particle impacts on the detector , and have a fairly long lasting impact on pixel responsivities .
conversely , the much more frequent impacts of cosmic ray electrons produce common glitches characterized by a relatively fast decay time , lasting only a few readouts .
therefore , the number of frames affected by a single fader or dipper is much higher than in the case of a common glitch , the pixel responsivity taking from tens to hundreds of seconds to recover completely .
however , common glitches are much more frequent than faders and dippers and , may all the same hamper the quality of data reduction .
thus , all kinds of glitches , if not correctly removed ( or , more properly , corrected for ) , can lead to spurious detections , or * unreliability * , and to the loss of genuine sources , or *
incompleteness*. on the other hand , transients all follow the same pattern , due to the fact that they arise from the non negligible time it takes for the output signal to reach the stabilization value after a change in the incident photon flux has taken place . the measured signal is thus always lower than the true one .
failing to model this time effect in data reduction can lead to a systematic underestimation of source fluxes .
for these reasons , the data cleaning and modeling is an extremely delicate process requiring great care in order to produce highly reliable sky maps and source lists . while it was variously demonstrated that it is possible , at least to a certain extent , to describe the detector s behaviour adopting some physical model , the large number of readouts involved in raster observations and the peculiar nature and strength of noise patterns also require efficient and robust algorithms to be developed so as to make the actual data reduction undertaking feasible in a nearly - automatic way .
a number of data reduction methods has thus been developed and tested , mostly on deep fields ( e.g. the preti method by @xcite and the triple beam switch method by @xcite ) .
unfortunately , such methods did not prove completely reliable on shallower fields , leading to a high number of false detections .
besides , these methods suffered from the lack of an efficient way to interactively check the quality of the data reduction when needed .
the preliminary analysis of elais 15 @xmath0 m data @xcite was thus carried out adopting a more traditional approach involving the corroboration of automated detections through visual inspection by different observers .
the data reduction described in this paper was carried out using the lari method @xcite , a new technique developed to overcome these difficulties and provide a robust interactive technique for the reduction and analysis of isocam and isophot data , particularly suited for the detection of faint sources and thus for the full exploitation of their scientific potential .
the method was variously refined , and significantly better results are now obtained , with respect to the technique used in @xcite for the reduction of the s1 field , so that a thorough re - reduction of all elais fields seemed appropriate and was thus carried out . as before , data reduction is carried out within an idl environment using mostly purpose - built routines , exploiting cam interactive analysis ( * ? ? ?
* cia , ) software for basic operations only .
the lari method describes the sequence of readouts , or time history , of each pixel of isocam lw3 detector in terms of a mathematical model for the charge release towards the contacts .
such a model is based on the assumption of the existence , in each pixel , of two charge reservoirs , a short - lived one @xmath19 , also known as * breve * , and a long - lived one @xmath20 , also known as * lunga * , evolving independently with a different time constant and fed by both the photon flux and the cosmic rays .
the model is fully charge - conservative , i.e. no decay of accumulated charges is considered , except towards the contacts , and thus the observed signal @xmath21 is related to the incident photon flux @xmath22 and to the accumulated charges @xmath19 and @xmath20 by the @xmath23 where the evolution of these two quantities is governed by the same differential equation , albeit with a different efficiency @xmath24 and time constant @xmath25 @xmath26 so that @xmath27 the values of the parameters @xmath24 and @xmath25 depend on the detector s physics , and are thus assumed to depend on the exposure time of a given observation and on the stabilization background level of a given pixel , according to the relation @xmath28 where @xmath29 is the value of @xmath25 relative to a reference exposure time @xmath30 and average signal level @xmath31 . in practice , an additive offset signal attributable to thermal dark current ( a component which is otherwise removed in standard dark current subtraction ) , is actually added to both @xmath21 and @xmath22 in the equation above when it is estimated to be important , i.e. in the rare cases when the deepest dippers depth otherwise exceeds 10% of the stabilization background level .
data reduction begins with a preliminary * pipeline * incorporating all necessary steps in order to prepare the data for the temporal fitting procedure which is the critical step of the lari method .
raw data downloaded from the iso data archive are first imported into the idl ` raster ` structure containing all observational information using cia routines .
likewise , dark current subtraction and conversion from adu to adu / gain / s are carried out using cia .
then a dedicated idl structure called ` liscio ` is built in order to contain not only all ` raster ` information but also all ancillary arrays needed in order to carry out the following reduction . at this stage
potential glitches are identified through a two - step median filtering process of each pixel time history , or * deglitching*. then a separate routine determined the stabilization background ( or global background , as opposed to the local background described below ) and the aforementioned offset signal .
such routine was carefully devised to filter out in so far as possible long - term effects such as stabilization , faders and dippers and thus provide a reliable estimate of these two parameters the fitting procedure is particularly sensitive to .
the same routine also identifies the few potential bright sources which at times the fit failed to recognize by itself , which are then interactively checked by eye to assess their reliability .
the signal as a function of time is finally processed independently for each pixel .
the fitting procedure models the transients attributable to changes in incident flux and the features on both short and long time scales produced by cosmic ray impacts on the time history , modeling glitches as discontinuities in the charge release . as seen in section [ model.sec ] , the same values for the @xmath24 and @xmath25 parameters are used for all pixels , apart from the scaling of the @xmath25 according to the exposure time and the stabilization background level , leaving as free parameters only the charges at the beginning of the observation and at the `` peaks '' of glitches .
glitches from nearby pixels are also considered when their height is substantially ( i.e. 20 times ) higher than the chosen threshold .
the fitting algorithm starts with the strongest potential glitches identified in pipeline deglitching , assumes discontinuities at these positions and tries to find a fit to the time history that satisfies the model assumed to describe the solid - state physics of the detector .
if no acceptable fit is found , the next fainter glitch is considered as a potential discontinuity , and so on .
iteration of the fitting procedure is interrupted when either a satisfactory ( typically 0.2 adu / gain / s ) data - model rms deviation is achieved or the maximum number of allowed iterations is reached . at this stage
the code estimates several quantities needed to build the sky maps on which source extraction will then be performed .
all of these quantities are `` recovered '' from glitches , i.e. already take into account discontinuities in charge release assumed to describe glitches during fit , and their list includes : * the charges stored into the * breve * and * lunga * reservoirs at each readout . * the local background , i.e. the signal to be expected on the basis of the previously accumulated charges if only the stabilization background were hitting the detector . * the model signal , produced by the incident flux coming from both the stabilization background and detected sources .
* the * `` unreconstructed '' * signal , i.e. the detected flux recovered from glitches but not from transients .
this is computed as the difference between the measured signal and the local background . * the * `` reconstructed '' * signal , i.e. the detected flux recovered not only from glitches but also from transients associated with changes in incident flux .
for the sake of clarity , one must emphasize here the differences between the two kinds of signals ( and the corresponding fluxes and sky maps they will finally turn into ) defined above , namely * unreconstructed * and * reconstructed * signals .
both quantities take into account the effects of the stabilization background and glitches on the detector , and their values are thus expected to be negligible if a source is not illuminating the pixel at the given pointing . the difference between the two quantities only appears when an additional signal the code is not able to model otherwise ( i.e. as the effect of a glitch ) is detected and attributed to a source of a given flux .
the code then models the transients expected from such an additional flux and `` reconstructs '' the signal one would detect if they did not affect the detector , i.e. if its response were instantaneous . in other words ,
unreconstructed signals do not take into account the effects of this modeling of changes in incident flux , thus representing the effective charge collected during the exposure , whereas reconstructed signals recover the charge `` loss '' due to the slow detector response .
therefore , the former are systematically lower than the latter . actually , simulations show that the code is not actually able to properly model transients below a certain intensity threshold , thus suggesting the use of unreconstructed signals only to carry out further processing .
figure [ fits.fig ] shows how a successful fit is able to describe cosmic - ray - induced violent changes in the signal level and thus recover useful information ( specifically , source fluxes ) from the pixel time history .
panels a ) and b ) show two examples of how glitches ( a fader and a dipper , respectively ) are described as discontinuities in the signal level slowly recovering towards the stabilization background , while panels c ) and d ) show how sources are detected even on the top of strong glitches .
the solid line represents the observed data , the short - dashed line the best - fit model and the long - dashed line the detected ( unreconstructed ) flux .
the dot - dashed line finally represents the stabilization background + * a ) fader * + + * b ) dipper * + + * c ) bright source * + + * d ) faint source * + after running the fitting procedure on the time history of all pixels making up a raster , the interactive analysis of fitting results is carried out , looking in detail at portions of the time history that were not well - fit by the automated analysis .
the details of the interactive analysis process need to be tuned to the quality of the specific raster under consideration .
in particular , the choice of thresholds in interactive checks is closely related with observing parameters such as the exposure time but also with the varying frequency and severeness of cosmic ray impacts . generally speaking ,
more careful checks could be profitably carried out on intrinsically higher - quality data . on the other hand , it is desirable to ensure the most uniform data reduction as practicable .
it was thus decided to adopt a common trade - off between the data quality of all rasters making up our dataset and to apply it to all rasters . as a first step , fitting failures flagged by substantial data - model rms deviations ( higher that 0.23 adu / gain / s ) or negative signals ( lower than -0.6 adu / gain / s ) are checked .
then all sizeable signal excesses ( unreconstructed signals higher than 0.5 adu / gain / s ) are individually inspected .
whenever the need arises , a further fit extending to the whole pixel time history or to a smaller portion of it is carried out . particularly noisy regions or very strong individual features completely preventing data reduction
are masked .
the massive work of interactive analysis is carried out with an easy - to - use idl widget - based graphical user interface , a screenshot of which is shown in figure [ gui.fig ] , which allows any kind of operation that could be necessary : data visualization and browsing , glitch addition and correction , time history masking and re - fitting .
such software is the main factor allowing the increase in the volume and quality of the obtained catalogues with respect to @xcite . once a satisfactory fit is obtained over the whole time history for all pixels
, one can proceed to the generation of sky maps and to source extraction . after masking glitches and other noisy parts of the time history identified during interactive analysis ,
flat - fielding is carried out and a signal estimate for each raster pointing is computed by averaging all readouts relative to that pointing .
the result is converted from adu / gain / s to mjy using isocam lw3 standard sensitivity 1 mjy = 1.96 adu / gain / s .
the `` images '' thus computed relative to all raster pointings are then projected onto a @xmath32 pixel sky map adopting nominal astrometric information and using a new mapping technique optimized to partly overcome the severe psf undersampling due to the large pixel size adopted in observations . here and in the following , all necessary projections are carried out using the ` projection ` c++ code included in cia . in so doing ,
isocam severe field distortion @xcite was taken into account as determined by @xcite , whereas psf modeling followed the prescriptions given by @xcite for stellar psfs but adopting a spectrum of the form @xmath33 , i.e. a closer match to the expected galaxy spectrum than the rayleigh - jeans form used for stellar spectra .
the resulting psf is larger than the one computed for stars . following this procedure ,
both unreconstructed and reconstructed sky maps are produced . in both cases ,
two ancillary maps are also constructed . the sky coverage ( or `` npix '' ) map contains the number of independent `` images '' added together to obtain a pixel value in the final map . on the other hand ,
the noise ( or `` rms '' ) map contains the estimate of the root mean square noise of each single pixel in the final map , computed by scaling the overall root mean square noise of the map as measured at its centre according to the inverse square root of the noise map .
source detection is then performed on unreconstructed sky maps .
first , all pixels above a conservatively low flux threshold ( 60 @xmath0jy ) are selected , then the idl astronomy user s library ` find ` routine ( based on daophot s homonymous routine , particularly suited for the detection of point - like sources in crowded fields ) is used to identify positive brightness perturbations around these pixels and return their peak flux , @xmath34 , centroid and shape parameters such as roundness and sharpness . at this stage ,
further interactive checks are performed on all sources detected with a @xmath34 greater than 5 to assess their reliability .
more specifically , sources detected on the sky map are projected back onto the time history to identify all raster pointings where the source is supposed to contribute a significant signal . then all these pointings are individually checked and , if necessary , refitted , to improve the fit and thus recover lost signals or remove spurious ones .
the overall results of these further checks are an increased reliability and an improved astrometric and photometric accuracy . up to this stage ,
all rasters are reduced and processed individually . however , in order to fully exploit the limited redundancy of the observations , a technique to build a mosaic out of rasters covering the same field was devised .
this is carried out as follows .
once the reduction of all rasters of interest is completed according to the above procedure , the necessary corrections to nominal astrometry are determined as the median offset between elais source positions expressed with respect to nominal astrometry and the positions of usno a2.0 sources found in the field .
this is done through a two - step process .
first , the two catalogues are cross - correlated , assigning to each elais sources its closest usno association .
the median of the positional differences thus determined , excluding elais sources with no usno association within 12 arcsec , is computed and assumed as a first - order correction to elais source positions .
the cross - correlation procedure is then repeated to calculate a second - order astrometric correction in exactly the same way , the only difference being that elais sources with no usno identification within 3 times the root mean square deviation of elais - usno association distance are also excluded during this second step .
the sum of the two corrections is assumed as the raster offset with respect to nominal astrometry , and the root mean square deviation of elais - usno distances as the error in the offset determination .
deviations from nominal astrometry have long been known to be significant in iso raster observations , and this is confirmed by our results , which are summarized in table [ offsets.tab ] .
most total offsets are greater than or of the order of the pixel field of view size ( 6 arcsec ) , whereas associated errors have a mean value of 0.39 arcsec .
the small errors are not only due to the careful data reduction , but also to the large number of elais sources , and largely contribute to the very good overall astrometric accuracy quantified in section [ astroacc.sec ]
. a common mosaic grid is then built on which different rasters belonging to the same field are projected taking into account astrometric offsets .
mapping , source extraction , projection of sources detected with a @xmath34 greater than 5 back on time history and interactive checks are furtherly performed on this mosaic sky map exactly as on single raster sky maps .
the quality of data reduction is thus improved through cross - checks of sources detected on different rasters , increasing reliability and completeness in repeatedly observed regions and partly overcoming the otherwise severe problems at raster boundaries .
the final @xmath34 sky maps of the four fields are shown in figures [ s1_map.fig ] , [ n3_map.fig ] , [ n1_map.fig ] and [ n2_map.fig ] .
typical noise levels are between 20 and 30 .
such maps , together with noise and sky coverage maps , will soon be available in fits format at ` http://astro.imperial.ac.uk/\simvaccari/elais ` .
ccccccc raster & & ra ( @xmath14 ) & dec ( @xmath14 ) & total ( @xmath14 ) & + + & ra ( j2000 ) & dec ( j2000 ) & offset @xmath35 error&offset @xmath35 error&offset @xmath35 error + s1_1 & 00 30 25.40 & -42 57 00.3 & -2.06 @xmath35 0.40 & -4.46 @xmath35 0.38 & 4.91 @xmath35 0.55 + s1_2 & 00 31 08.20 & -43 36 14.1 & -3.24 @xmath35 0.22 & + 6.86 @xmath35 0.29 & 7.59 @xmath35 0.36 + s1_3 & 00 31 51.90 & -44 15 27.0 & + 1.57 @xmath35 0.29 & -7.75 @xmath35 0.33 & 7.91 @xmath35 0.44 + s1_4 & 00 33 59.40 &
-42 49 03.1 & + 0.23 @xmath35 0.22 & -4.01 @xmath35 0.27 & 4.02 @xmath35 0.35 + s1_5_a & 00 34 44.40 & -43 28 12.0 & -3.50 @xmath35 0.23 & + 9.63 @xmath35 0.27 & 10.24 @xmath35 0.35 + s1_5_b & 00 34 44.40 & -43 28 12.0 & -0.52 @xmath35 0.21 & -8.10 @xmath35 0.26 & 8.12 @xmath35 0.33 + s1_5_c & 00 34 44.40 & -43 28 12.0 & -3.04 @xmath35 0.24 & + 5.34 @xmath35 0.29 & 6.14 @xmath35 0.38 + s1_6 & 00 35 30.40 & -44 07 19.8 & + 0.60 @xmath35 0.43 & -7.14 @xmath35 0.24 & 7.17 @xmath35 0.49 + s1_7 & 00 37 32.50 & -42 40 41.2 & + 1.26 @xmath35 0.22 & -5.62 @xmath35 0.38 & 5.76 @xmath35 0.44 + s1_8 & 00 38 19.60 & -43 19 44.5 & + 0.72 @xmath35 0.22 & -5.31 @xmath35 0.24 & 5.36 @xmath35 0.36 + s1_9 & 00 39 07.80 & -43 58 46.6 & -2.34 @xmath35 0.19 & + 4.61 @xmath35 0.23 & 5.17 @xmath35 0.30 + n1_1 & 16 15 01.00 & + 54 20 41.0 & + 1.96 @xmath35 0.34 & -9.34 @xmath35 0.21 & 9.54 @xmath35 0.40 + n1_2_a & 16 13 57.10 & + 54 59 35.9 & + 3.27 @xmath35 0.31 & -9.48 @xmath35 0.22 & 10.02 @xmath35 0.38 + n1_2_b & 16 13 57.10 & + 54 59 35.9 & + 4.17 @xmath35 0.23 & -7.89 @xmath35 0.15 & 8.92 @xmath35 0.27 + n1_3 & 16 10 34.90 & + 54 11 12.7 & -1.04 @xmath35 0.42 & -6.18 @xmath35 0.25 & 6.27 @xmath35 0.49 + n1_4 & 16 09 27.00 & + 54 49 58.7 & + 4.60 @xmath35 0.44 & -6.71 @xmath35 0.23 & 8.14 @xmath35 0.50 + n1_5 & 16 06 10.80 & + 54 01 08.0 & -8.90 @xmath35 0.27 & + 3.77 @xmath35 0.18 & 9.67 @xmath35 0.32 + n1_6 & 16 04 59.00 & + 54 39 44.3 & -8.71 @xmath35 0.32 & + 3.88 @xmath35 0.22 & 9.54 @xmath35 0.39 + n2_1 & 16 32 59.80 & + 41 13 33.2 & -4.68 @xmath35 0.23 & + 7.18 @xmath35 0.21 & 8.57 @xmath35 0.31 + n2_2 & 16 34 44.50 & + 40 38 45.0 & + 2.93 @xmath35 0.27 & -2.87 @xmath35 0.27 & 4.10 @xmath35 0.38 + n2_3 & 16 36 05.50 & + 41 33 11.8 & -5.50 @xmath35 0.46 & + 7.30 @xmath35 0.37 & 9.14 @xmath35 0.59 + n2_4 & 16 37 48.90 & + 40 58 13.1 & -5.36 @xmath35 0.24 & + 7.46 @xmath35 0.21 & 9.19 @xmath35 0.32 + n2_5 & 16 39 13.80 & + 41 52 31.6 & -5.43 @xmath35 0.29 & + 7.32 @xmath35 0.24 & 9.11 @xmath35 0.38 + n2_6 & 16 40 55.50 & + 41 17 22.7 & -2.45 @xmath35 0.28 & + 2.32 @xmath35 0.24 & 3.38 @xmath35 0.37 + n2_r_a & 16 35 45.00 & + 41 06 00.0 & + 1.29 @xmath35 0.33 & + 6.01 @xmath35 0.21 & 6.15 @xmath35 0.39 + n2_r_b & 16 35 45.00 & + 41 06 00.0 & + 5.91 @xmath35 0.34 & + 4.24 @xmath35 0.22 & 7.27 @xmath35 0.40 + n3_3 & 14 29 38.30 & + 33 24 49.6 & -1.39 @xmath35 0.30 & -3.44 @xmath35 0.16 & 3.71 @xmath35 0.34 + n3_5 & 14 32 38.20 & + 33 11 10.3 & -0.17 @xmath35 0.31 & -3.52 @xmath35 0.24 & 3.52 @xmath35 0.39 +
even after a very careful reduction , flux determination is a very delicate process owing to the interplay of different factors , the most important being mapping effects related to psf undersampling , detector s transients and other trends due to the limits of the adopted reduction method .
mapping effects are particularly tricky , since they determine the way one obtain source total fluxes from corresponding peak fluxes measured on sky maps by source detection software .
clearly , the conversion factor between these two quantities is very much dependent on the actual source position with respect to the centre of the pixel , a source total flux being nearer to its peak flux the nearer the source is to the center of the sky map pixel .
the technique we developed , which we called * autosimulation * , was conceived to model and insofar as possible take into account mapping and data reduction effects on flux determination by first constructing both real and theoretical ( i.e. noise - free ) sky maps , and then use the latter ones to correct fluxes obtained from the former ones .
autosimulation consists in simulating sources , on top of noise - free maps , at each position where a source was actually detected ( or , for simulations described in section [ simulations.sec ] , at their randomly generated positions ) , calculating the ratio between peak and total flux for the simulated source on such theoretical sky maps , and converting the measured peak flux of the real source to its total flux using such ratio .
source simulation is carried out in a straightforward way using the same mathematical model the lari method is based on , thus also allowing to verify its predictions . in order to describe how autosimulation works in some greater detail ,
a few definitions and relations which will also be useful to discuss the results of simulations described in section [ simulations.sec ] must be summarized : * @xmath36 is the measured peak flux obtained from real ( i.e. containing glitches , noise and transients ) sky maps .
its value therefore depends on mapping effects , transients and the adopted data reduction method ; * @xmath37 is the `` theoretical '' peak flux obtained from simulated ( i.e. containing neither glitches nor noise but taking into account source transients ) maps .
its value depends on mapping effects and transients only . *
@xmath38 and @xmath39 are analogous to @xmath36 and @xmath37 but are built from reconstructed maps , thus recovering the effects of transients . as already mentioned , this strictly holds above a certain flux threshold only , where the values of these quantities actually cease to depend on transients .
the correction factor for mapping effects is computed as follows .
a simulated source is generated on each position where a real source was actually detected , assuming a total flux @xmath40 based on the measured peak flux @xmath36 and on the average peak flux / total flux ratio @xmath41 obtained from simulations ( see section [ simulations.sec ] and figure [ flussivsfs.fig ] ) .
then the flux estimate corrected for mapping effects is computed as @xmath42 .
source simulation is carried out using the same model for charge release adopted in fitting the data . in so doing , however
, autosimulation does not correct for more subtle and elusive effects arising from data reduction , such as flux loss due to bad fits , which can only be assessed through simulations .
this is described in section [ fluxdet.sec ] .
such a procedure provides reliable flux estimates which are well compatible with estimates obtained through conventional aperture photometry for most sources , but is clearly non - optimal when dealing with extended sources . for these , as well as for a few very close or blended sources ,
aperture photometry usually provide a better estimate of the source flux . after correcting fluxes for mapping effects ,
potentially extended sources are therefore identified through both visual inspection of sky maps and the calculation of parameters connected to source extension .
sources flagged by these criteria ( totalling 67 out of 1923 making up the catalogue , or 3.5% ) were then individually treated , aperture photometry with a suitable aperture radius was carried out and the result was adopted as their most reliable flux estimate . as to the overall performance of the autosimulation process when compared with aperture photometry , autosimulated fluxes and their counterparts calculated through aperture photometry are compared for all sources in section [ autoaper.sec ] .
due to the peculiar nature of iso data and of the reduction method employed , it is important to carefully test the performance of the latter on `` ideal '' data and sources . in particular ,
systematic effects on flux estimates related to the data reduction method can only be probed by these means . due to the strong peculiarities of our dataset , which is characterized by several noise features on different time scales ,
only real data can effectively be taken as representative of instrumental behaviour .
therefore , the effects of artificial sources must be somehow simulated on the top of real pixel time histories , and data reduction must then be carried out exactly as done for real sources .
source confusion in the field is thus slightly increased , but this effect is not critical for elais data .
the lari method is based on a physical model of the detecto s behaviour , so can make straightforward predictions of the effect on the detector of the additional photons from simulated sources . on this basis ,
an extensive set of simulations was carried out to assess the effects on flux estimates and the overall performance of data reduction in a statistically meaningful way .
the s1 field was chosen for this purpose as the most representative , being the largest and including both regions observed only once and repeatedly observed regions ( hereafter , * non - repeated * and * repeated * regions , respectively ) , therefore allowing to assess the differences in performance warranted by higher redundancy .
simulations were carried out as follows .
first a set of @xmath43 random sky positions ( excluding regions near to real sources , mosaic sky map boundaries and masked regions of mosaic sky map , i.e. preventing simulated sources to appear nearer than 15 arcsec to any of these regions ) was generated .
then a logarithmically uniform flux distribution covering the range between a lower limit @xmath44 and an upper limit @xmath45 @xmath46 was coupled with the random positions above to characterize the set of simulated sources .
following the same procedure , 100 more sources were simulated ( adopting a flux distribution following the same analytical formula but covering the 0.353.5 mjy flux range ) in the repeated regions of the field , thus increasing the otherwise low number of simulated sources in these regions .
all 700 sources were then projected from their sky position back onto the pixel time history , simulating their effects as superposed on glitches and noise characterizing the real data . in so doing ,
all portions of time history where simulated sources were predicted to give a sizeable additional signal were identified .
the fitting procedure was thus re - run on these regions only , and all steps of data reduction and flux determination were then carried out on simulated sources exactly as described in section [ datared.sec ] for real sources . at this stage the positions and fluxes of simulated sources that had been detected with a @xmath34 greater than 5 were compared with the input values to calibrate and evaluate the performance provided by our data reduction method in different respects .
the overall number of simulated and detected sources is given in table [ simbasic.tab ] , whereas the following sections detail all relevant aspects of simulations , apart from the completeness estimates which will be described by @xcite in order to obtain extragalactic source counts .
.number of simulated and detected sources .
total figures are given together with those for non - repeated and repeated regions . [ cols="<,^,^,^",options="header " , ]
a technique for iso - cam / phot data reduction , the lari method , was variously refined and applied to elais 15 @xmath0 m observations . the mathematical model for the detector s behaviour
is the same as originally presented in @xcite , but thanks to various improvements , and particularly to a new graphical user interface , the method is now more robust and , most importantly , quicker and easier to apply to large datasets .
its application , in the new form , to the four fields composing the dataset ( including a re - reduction of s1 observations already presented in @xcite ) has produced a catalogue of 1923 sources spanning the 0.5 100 mjy range , detected with a @xmath34 greater than 5 over a total area of @xmath47 .
optical identification of 15 @xmath0 m sources has been carried out on heterogeneous optical and near - infrared imaging material , allowing to determine a robust association for about 85% of the sources and identify 22% of them as bona fide stars , furtherly demonstrating the reliability of our data reduction process .
the evaluation of the catalogue s quality has been carried out through both accurate simulations and multi - wavelength identification .
the astrometric accuracy is of order 1 arcsec in both ra and dec for @xmath48 , while it increases up to about 2 arcsec in both ra and dec for @xmath49 , and somewhat better for sources detected in higher - redundancy sky regions .
the photometric accuracy is estimated to be below 25% over the whole range of fluxes and redundancy levels probed by our catalogue , and better than 15% for @xmath48 sources . the comparison of measured stellar fluxes with fluxes estimated on the basis of stellar atmospehere models calibrated on iras data and on near - infrared photometry allowed to achieve an iras / iso relative photometric calibration .
an iras / iso relative calibration factor of @xmath50 was determined , shedding doubts on the goodness of the two independently determined calibrations at the 10% level . for lack of a simple way to identify error sources in iras and/or iso calibratioon process ,
it was decided to put our catalogue on the more commonly used iras flux scale . in a forthcoming paper @xcite
completeness estimates and extragalactic source counts from this catalogue will be presented , covering the crucial flux range 0.5 100 mjy between isocam 15 @xmath0 m deep surveys and iras all sky survey .
this paper is based on observations with iso , an esa project with instruments funded by esa member states ( especially the pi countries : france , germany , the netherlands and the united kingdom ) and with the participation of isas and nasa .
the isocam data presented in this paper were analysed using cia , a joint development by the esa astrophysics division and the isocam consortium .
the isocam consortium is led by the isocam pi , c. cesarsky .
99 aussel h. , cesarsky c.j .
, elbaz d. , starck j.l . , 1999 , a&a , 342 , 313 aussel h. , 2004 , in preparation bertin e. , dennenfeld m. , moshir m. , 1997 , a&a , 323 , 685 cesarsky c.j .
et al . , 1996 ,
a&a , 315 , l32 dsert f.x .
et al . , 1999 ,
a&a , 342 , 363 elbaz d. et al . , 1999 , a&a , 351 , 37 fadda d. et al . , 2004 , in preparation franceschini a. et al . , 1988 , mnras , 233 , 175 genzel r. and cesarsky c.j . , 2000 ,
ara&a , 38 , 761 gonzlez - solares e. et al . , 2004 , mnras , submitted , astro - ph/0402406 gregorich d.t .
et al . , 1995 ,
aj , 110 , 259 gruppioni c. , lari c. , pozzi f. , zamorani g. , franceschini a. , oliver s. , rowan - robinson m. , serjeant s. 2002 , mnras , 335 , 831 hacking p. , houck j.r . , condon j.j . , 1987 , apj , 316 , 15 hg e. et al . , 2000 ,
a&a , 355 , 27 kessler m.f .
et al . , 1996 ,
a&a , 315 , l27 lari c. et al
. , 2001 , mnras , 325 , 1173 la franca f. et al . , 2004 ,
aj , accepted , astro - ph/0403211 lari c. et al .
, 2003 , esa sp-511 , 349 lari c. et al . , 2004 , in preparation lejeune t. , cuisinier f. , buser r. , 1998 , a&ass , 130
, 65 lemke d. et al . , 1996 , a&a , 315 , l64 lonsdale c.j .
et al . , 1990 ,
apj , 358 , 60 lonsdale c.j .
et al . , 2003 , pasp , 115 , 897 manners j.c .
et al . , 2003 ,
mnras , 343 , 293 mcmahon r.g .
. , 2001 , new astronomy review , 45 , 97 neugebauer g. , habing h. j. , van duinen r. et al . , 1984 , apj , 278 , l1 okumura k. , 1998 , isocam psf report , available at ` http://www.iso.vilspa.esa.es/users/expl\_lib/cam\_list.html ` okumura k. 2000 , isocam field distortion report , available at ` http://www.iso.vilspa.esa.es/users/expl\_lib/cam\_list.html ` omont a. , ganesh s. , alard c. , blommaert j.a.d .
, caillaud b. et al . , 1999 ,
a&a , 348 , 755 oliver s. , rowan - robinson m. et al . , 2000 , mnras , 316 , 749 ott s. , gastaud r. , ali b. , delaney m. , miville - desch^ enes m .- a . , okumura k. , sauvage m. , guest s. , 2001 , asp conf .
238 , 170 pozzi f. et al . , 2003 , mnras , 343 , 1348 rowan - robinson m. et al . , 1984 ,
apj , 278 , 7 rowan - robinson m. et al . , 1991 , nature , 351 , 719 rowan - robinson m. et al . , 2003 , mnras , accepted , astro - ph/0308283 saunders w. et al . , 1991 , nature , 349 , 32 schlegel d.j . , finkbeiner d.p .
, davis m. , 1998 , apj , 500 , 525 serjeant s. , oliver s. , rowan - robinson m. et al .
, 2000 , mnras , 316 , 768 soifer b.t .
et al . , 1984 , apj , 278 , 71 soifer b.t . , houck j.r . , neugebauer g. , 1987 , ara&a , 25 , 187 starck j.l . , aussel h. , elbaz d. , fadda d. , cesarsky c. , 1999 , a&as , 138 , 365 visnen p. et al . , 2002 ,
mnras , 337 , 1043 | we present the final analysis of the european large area iso survey ( elais ) 15 @xmath0 m observations , carried out with the isocam instrument on board the infrared space observatory ( iso ) .
the data reduction method , known as lari method , is based on a mathematical model of the detector s behaviour and was specifically designed for the detection of faint sources in iso - cam / phot data .
the method is fully interactive and leads to very reliable and complete source lists .
the resulting catalogue includes 1923 sources detected with @xmath1 in the flux range and over an area of 10.85 split into four fields , making it the largest non - serendipitous extragalactic source catalogue obtained to date from iso data .
this paper presents the concepts underlying the data reduction method together with its latest enhancements .
the data reduction process , the production and basic properties of the resulting catalogue are then discussed .
[ firstpage ] infrared : galaxies galaxies : formation , evolution , active , starburst cosmology : observations methods : data analysis catalogues . |
in many studies of cell mechanics and dynamics , the cell is characterized as a viscoelastic body @xcite .
it is an intriguing question to what extent such mechanical behaviour can be rationalized in terms of comparatively simple polymer physics models . in this respect ,
the comparison of cell rheological data and minimalistic _ in vitro _ reconstituted constructs of the cytoskeleton , such as pure actin solutions @xcite or crosslinked actin networks @xcite , has recently provided many new insights .
important progress has also been achieved in the development of phenomenological mathematical descriptions .
this includes approaches related to the tube model @xcite , tensegrity - based approaches @xcite , effective - medium models @xcite , and some others @xcite . in particular
, the glassy wormlike chain ( ) model @xcite , a phenomenological extension of the standard model of semiflexible polymers @xcite has been successful in describing a plethora of rheological data for polymer solutions @xcite and living cells @xcite over many decades in time with a minimum of parameters .
however , all these studies were primarily concerned with _ viscoelastic _
behaviour , while the latest investigations have underscored the glassy @xcite fragile @xcite , and inelastic @xcite character of the mechanical response of living cells .
even for biopolymer networks _ in vitro _
, experiments operating in the nonlinear regime had so far to resort to special protocols that minimize plastic flow @xcite in order to make contact with dedicated theoretical models . the aim of the present contribution is to overcome this restriction by extending the to situations involving inelastic deformations . as a first step , we concentrate onto _ reversible inelastic _
behaviour , where the deformation does not alter the microscopic ground state .
the protocol applied by trepat _
et al . _
@xcite provides a paradigmatic example .
cells are subjected to a transient stretch such that , after some additional waiting time in the unstretched state , the ( linear ) material properties of the initial state are recovered . the simplification for the theoretical modelling results from the assumption that not only the macro - state but also the micro - state of the system may to a good approximation be treated as reversible under such conditions ; i.e.
, we assume that the complete conformation of the polymer network , including the transiently broken bonds between adjacent polymers , is constrained to eventually return to its original equilibrium state . for the time - delayed hysteretic response of the network to such protocols one could thus still speak of a viscoelastic ( `` anelastic '' ) response in an operational sense , but we refrain from doing so in view of the fundamentally inelastic nature of the underlying stochastic process in contrast to the reversible softening effects observed in @xcite , for example .
indeed , by simply allowing bonds to reform in new conformational states , the model developed below can readily be extended to arbitrary irreversible plastic deformations , as will be demonstrated elsewhere @xcite . before entering the discussion of our model
, we would also like to point out that the proposed ( inelastic ) extension of the is strongly constrained by identifying the newly introduced parameters with those of the original ( viscoelastic ) model , where possible . despite its increased complexity , the extended model will therefore enable us to subject the underlying physical picture to a more stringent test than hitherto possible by comparing its predictions to dedicated experiments . moreover , unlike current state - of - the - art simulation studies @xcite it is not limited to rod networks but is firmly routed in a faithful mathematical description of the underlying brownian polymer dynamics . this paper is organized as follows .
first , we review some basic facts about the in section [ sec : gwlc ] . next , in section [ sec : interaction ]
, we introduce our extended reversible inelastic version , which we formulate using the notion of an effective interaction potential as in the original construction of the in @xcite .
( a preliminary account of the basic procedure and some of its cell - biological motivation including reversible bond - breaking kinetics has recently been given in a conference proceedings @xcite . ) sections [ sec : viscoelastic ] and [ sec : fluidization ] explain the physical mechanism underlying the mechanical response under pulsed and periodically pulsed loading , while section [ sec : remodelling ] illustrates its phenomenology . we demonstrate that the model exhibits the hallmarks of nonlinear cell mechanics : strain / stress stiffening , fluidization , and cyclic softening @xcite .
section [ sec : intr_lengths ] investigates the relevance of the lately quantified structural heterogeneities in networks of semiflexible polymers @xcite for the mechanical properties , before we conclude and close with a brief outlook .
the glassy wormlike chain ( ) is a phenomenological extension of the wormlike chain ( ) model , the well - established standard model of semiflexible polymers .
a broad overview over and dynamics can be found elsewhere @xcite .
the describes the mechanics of an isolated semiflexible polymer in an isothermal viscous solvent . in the weakly bending rod approximation ,
a solution of the stochastic differential equations of motion for the is possible _ via _ a mode decomposition ansatz for the transverse displacement of the polymer contour from the straight ground state .
the individual modes labelled by an index @xmath0 are independent of each other and decay exponentially with rates @xmath1 . for convenience ,
we set the thermal energy @xmath2 , so that the bending rigidity can be identified with the persistence length @xmath3 , in the following . using this convention ,
the expression for the transverse susceptibility of a polymer of contour length @xmath4 ( with respect to a point force ) reads @xcite @xmath5 here , @xmath6 is an optional backbone tension , and @xmath7 the euler buckling force for a hinged rod of length @xmath4 .
the different powers of @xmath0 in the denominator give notice of the competition of bending and stretching forces .
in the , interactions of the polymer with the surrounding network ( e.g. excluded volume interactions or sticky contacts ) are reflected in an altered mode relaxation spectrum @xcite . the intuitive albeit not fully microscopic picture underlying the formulation of the is illustrated in figure [ fig : cartoon ] , which depicts a test polymer reversibly bound to the background network _ via _ potential wells at all topological contacts , the so - called entanglement points .
consider now a generic point somewhere along the polymer backbone ( not an entanglement point ) .
it can relax freely until the constrictions are being felt , which slow down the contributions from long wavelength bending modes .
the translates this intuition into a simple prescription for the mode spectrum : the short - wavelength modes are directly taken over from the wormlike chain model , while modes of wavelength @xmath8 longer than the typical contour length @xmath9 between adjacent bonds are modified to account for the slowdown . motivated by the physical picture illustrated in figure [ fig : cartoon ] , the slowdown of the relaxation of a wavelength @xmath8 is expressed by an arrhenius factor @xmath10 $ ] for breaking @xmath11 potential energy barriers of height @xmath12 simultaneously .
accordingly , the phenomenological recipe to turn a into a reads : @xmath13 & \lambda_n \geq \lambda \end{array}\right . . \ ] ] upon inserting this into ( [ eq : sus_wlc ] ) one obtains the susceptibility @xmath14 .
the microscopic `` modulus '' for transverse point excitations of a generic backbone element on a test polymer is then defined as the inverse susceptibility , @xmath15 .
an approximate expression for the macroscopic shear modulus is obtained along similar lines @xcite . in the original equilibrium theory , @xmath9 was assumed to be a constant on the order of the entanglement length of the network , @xmath16 .
note , however , that @xmath17 is a geometric quantity ( which is determined by the polymer concentration and the persistence length ) while the contour length @xmath9 between closed bonds clearly depends on the state of the bond network .
one therefore has to allow for an increase of @xmath9 with the number of broken bonds in non - equilibrium applications .
this issue is explored in the following .
all mechanical quantities calculated within the model crucially depend on the interaction length @xmath9 . in previous applications of the model @xcite it was assumed that @xmath9 remains constant equal to its equilibrium value and unaffected by the deformation of the sample . in other words ,
the equilibrium theory allowed for statistical bond fluctuations but not for a dynamical evolution of the parameters characterizing the thermodynamic state of the bond network .
an obvious starting point for generalizations of the model to non - equilibrium situations is therefore to consider the number of closed bonds , and therefore also @xmath9 , as _ dynamic variables _ , dependent on the strain- and stress - history of the network .
we now provide a mean - field description to account for such a dynamically evolving bond network . for clarity , we return to the intuitive picture underlying the where the ( possibly crosslinker- or molecular - motor - mediated ) complex interactions between the polymers are summarized into an effective interaction potential for a test segment against the background , as sketched in figure [ fig : potential ] .
the same idea has also been used before in many related situations ( e.g.@xcite ) . between both states .
the transition rates @xmath18 and @xmath19 between the bound and unbound state depend on the barrier heights @xmath12 and @xmath20 , respectively .
an externally applied force can tilt the potential and favour binding or unbinding events.,width=9 ] the generic potential exhibits three essential features : a well - defined bound state , a well - defined unbound state , and a barrier in between ( figure [ fig : potential ] ) .
the confining well corresponding to the unbound state represents the tube - like caging of the test polymer within the surrounding network @xcite .
for ease of notation , we further introduce the dimensionless fraction of closed bonds or , _ bond fraction _
@xmath21 which is simply the ensemble - averaged fraction of sticky contacts that are actually in the bound state .
the minimum bond distance @xmath22 is typically on the order of @xmath17 , but may be somewhat larger in situations where the bonds are mediated by crosslinker molecules or by partially sterically inaccessible sticky patches ( as e.g. for helical molecules @xcite ) . a quantitatively fully consistent way of calculating the dynamics of @xmath23 would involve solving the full fokker - planck equation for a -contact a formidable program to be pursued elsewhere @xcite . here ,
for the sake of our qualitative purposes , we chose to concentrate onto the physical essence of the discussion , and keep the mathematical structure as simple and transparent as possible .
we therefore approximate the dynamics by a simple exponential relaxation as familiar from the standard example of reacting brownian particles , conventionally described by kramers theory with a bell - like force dependence @xcite . using this simplification and assuming a schematic interaction potential as depicted in figure [ fig : potential ] , the value of @xmath23 is determined by a competition of bond breaking and bond formation with reaction rates @xmath18 and @xmath19 , respectively .
both rates are represented in the usual adiabatic approximation ( meaning that the equilibration _ inside _ the wells is much faster than the barrier crossing and external perturbations ) by @xcite @xmath24 where @xmath25 is an intrinsic characteristic brownian time scale for bond breaking and formation and @xmath26 are characterized by different brownian time scales depending on the width of the potential wells . at the present stage , we do not bother to distinguish these time scales nor to fine - tune their numerical values , and identify them , for the sake of simplicity , with the entanglement time scale @xmath27 in our numerical calculations .
] , and @xmath6 is the force pulling on the bond . noting that the fraction of open bonds is @xmath28 , we can then write down the following rate equation for the fraction of closed bonds @xmath29 the time dependence of @xmath30 leads via ( [ eq : nu_lambda ] ) to an implicit time dependence of the -parameter @xmath31 and thereby of all observables derived from the .
the time - dependent force @xmath32 in ( [ eq : rate ] ) may be thought to result from an externally imposed stress protocol or from internal dynamical elements such as molecular motors setting the network under dynamic stress .
hence , via an appropriately chosen set of slowly changing state parameters @xmath32 , @xmath33 , @xmath34 , the model can in principle accommodate for the active biological remodelling in living cells and tissues @xcite .
( for a discussion of the relation between the microscopic @xmath6 and the macroscopic stress , see @xcite . )
note that for constant force , the stationary value of @xmath23 , @xmath35\right)^{-1 } , \ ] ] obtained by setting @xmath36 , does not depend on @xmath12 , as it should be ( the steady state is independent of the transition state ) .
on the bond fraction @xmath23 for various values of @xmath12 ( brown : @xmath37 , red : @xmath38 , blue : @xmath39 , green : @xmath40 , black : @xmath41 ) at a frequency @xmath42 and a force @xmath43 ( @xmath44 is the relaxation time of a mode of length @xmath22 and @xmath45 is the euler buckling force for a rod of length @xmath22 ) ; _ left : _
normalized real part @xmath46 ; _ right : _ loss angle.,title="fig:",width=7 ] on the bond fraction @xmath23 for various values of @xmath12 ( brown : @xmath37 , red : @xmath38 , blue : @xmath39 , green : @xmath40 , black : @xmath41 ) at a frequency @xmath42 and a force @xmath43 ( @xmath44 is the relaxation time of a mode of length @xmath22 and @xmath45 is the euler buckling force for a rod of length @xmath22 ) ; _ left : _ normalized real part @xmath46 ; _ right : _ loss angle.,title="fig:",width=7 ] whenever the bond kinetics can be disregarded ( @xmath47 ) , the viscoelastic properties are simply those of the bare @xcite , which can basically be characterized as short time dynamics followed by a slow , highly stretched logarithmic relaxation resembling power - law rheology with a non - universal exponent within the typical experimental time windows .
strong perturbations ( stress or strain ) then give rise to a pronounced stiffening due to the steeply nonlinear response of semiflexible polymers under elongation @xcite .
this well established behaviour can be understood as the canvas against which we aim to discern the characteristic mechanical signatures of the bond kinetics .
an observable that will be of particular interest in the following is the complex microscopic `` modulus '' @xmath48 , introduced in the previous section .
it is used to determine the linear as well as ( via a nonlinear update scheme , see [ sec : technical ] ) the nonlinear force response of the system . to understand how the time - dependence ( [ eq : rate ] ) of @xmath23 affects this important quantity ,
it is instructive to first examine the dependence of @xmath49 on @xmath23 . to this end , we formally consider @xmath23 temporarily as an independent variable instead of determining it from ( [ eq : rate ] ) .
note that this approach nevertheless makes sense as for a fixed set of parameters , @xmath23 can still take any value between zero and one .
this is due to the freedom of choosing an initial state , which can be imagined to have evolved from the prior deformation history . for fixed other parameters , both the real part @xmath46 ( figure [ fig : mod_lam]a ) and the imaginary part @xmath50 of @xmath49 increase monotonically with @xmath23 .
we emphasize that @xmath50 is not simply proportional to @xmath46 and therefore the loss angle @xmath51 also depends on @xmath23 ( figure [ fig : mod_lam]b ) . for small @xmath23 , the loss angle is large , corresponding to fluid - like behaviour . with increasing @xmath23 ,
the system becomes more solid - like .
increasing the barrier height @xmath12 makes the dependence of @xmath49 on @xmath23 more pronounced ( figure [ fig : mod_lam ] ) .
conversely , as can be expected , the dependence on the barrier height @xmath12 vanishes with decreasing bond fraction @xmath23 .
note that due to the boundary conditions the limit of a newtonian fluid ( @xmath52 ) is not recovered when formally taking the limit of vanishing bond fraction ( @xmath53 , @xmath54 ) .
finally , the potential difference @xmath55 solely determines the dynamics of @xmath23 via ( [ eq : rate ] ) , and therefore influences the modulus @xmath49 solely through the equilibrium value for @xmath23 .
after these general considerations , we now concentrate on the nonlinear and non - equilibrium dynamic response of the extended inelastic model , which results from the coupled relaxation of the viscoelastic polymer network _ and _ the transient bond network . a convenient way to characterize the mechanical properties of an inelastic material is a force - displacement curve . for a perfectly linear elastic ( hookean ) body , it would simply consist of a straight line , whereas a perfectly plastic body would feature as a rectangle delineated by a yield force @xmath56 and an arbitrary plastic strain value . for our qualitative purpose
, we identify the average transverse displacement of the test polymer segment ( which is used to determine the force response , see [ sec : technical ] ) with the reaction coordinate @xmath57 of the schematic potential sketched in figure [ fig : potential ] ; and we identify the force @xmath6 entering the reaction rates with the backbone tension of the test polymer . as a characteristic length scale for the transverse displacements we use the width @xmath58 of the effective confinement potential , which is a measure of a typical mean - square displacement of the polymer in the unbound state . using this convention
, we now consider the effect of a time - symmetric displacement pulse on the evolution of the force @xmath32 and bond fraction @xmath30 . for technical convenience
, we use a gaussian shape for the displacement pulses , but the qualitative conclusions to be drawn are largely independent of the precise protocol .
the duration of the displacement pulse , which sets the time scale for the dynamic response , is used as the unit of time in the following . here , we are not interested in short - time tension - propagation and single - polymer dynamics @xcite , hence a lower bound for pertinent pulse durations is provided by the interaction time scale @xmath59 . on the other hand , for pulse durations longer than @xmath60
the system deformation would be quasistatic so that no genuinely dynamic effects of the bond kinetics could be observed . for a small gaussian displacement pulse of relative amplitude @xmath61 ( c.f .
figure [ fig : force_displacement]a and [ sec : technical ] ) , the shape of the curve predicted by our inelastic model shows all features of a _ viscoelastic _ medium ( figure [ fig : force_displacement]b , blue dashed curve ) .
it starts with a nearly linear regime for relative displacements @xmath62 , where a weak stiffening sets in .
due to dissipation in the medium , the path back to the initial state takes its course at lower force , which causes a weak hysteresis .
this is essentially the viscoelastic response that already the bare model would have predicted . :
gaussian deformation pulses ( @xmath63 ) with relative displacement amplitudes of @xmath61 ( dashed ) and @xmath64 ( solid ) ; corresponding force - displacement curves and bond fraction evolutions ( @xmath65 , @xmath66 ; dotted curve in @xmath65 : response of a without bond breaking ) ; order - of - magnitude estimate for the rate - dependence of the yield force , ( [ eq : rate_fc ] ) with const.@xmath67 , versus the numerically obtained maximum forces of the force - displacement curves for various mean force rates ( @xmath68 ) ; the minimum bond fraction ( @xmath68 , inset ) depends non - monotonically on the force rate ; @xmath69 , @xmath70 , @xmath71 , @xmath72 , @xmath73 . ,
title="fig:",width=7 ] : gaussian deformation pulses ( @xmath63 ) with relative displacement amplitudes of @xmath61 ( dashed ) and @xmath64 ( solid ) ; corresponding force - displacement curves and bond fraction evolutions ( @xmath65 , @xmath66 ; dotted curve in @xmath65 : response of a without bond breaking ) ; order - of - magnitude estimate for the rate - dependence of the yield force , ( [ eq : rate_fc ] ) with const.@xmath67 , versus the numerically obtained maximum forces of the force - displacement curves for various mean force rates ( @xmath68 ) ; the minimum bond fraction ( @xmath68 , inset ) depends non - monotonically on the force rate ; @xmath69 , @xmath70 , @xmath71 , @xmath72 , @xmath73 . ,
title="fig:",width=7 ] : gaussian deformation pulses ( @xmath63 ) with relative displacement amplitudes of @xmath61 ( dashed ) and @xmath64 ( solid ) ; corresponding force - displacement curves and bond fraction evolutions ( @xmath65 , @xmath66 ; dotted curve in @xmath65 : response of a without bond breaking ) ; order - of - magnitude estimate for the rate - dependence of the yield force , ( [ eq : rate_fc ] ) with const.@xmath67 , versus the numerically obtained maximum forces of the force - displacement curves for various mean force rates ( @xmath68 ) ; the minimum bond fraction ( @xmath68 , inset ) depends non - monotonically on the force rate ; @xmath69 , @xmath70 , @xmath71 , @xmath72 , @xmath73 .
, title="fig:",width=7 ] : gaussian deformation pulses ( @xmath63 ) with relative displacement amplitudes of @xmath61 ( dashed ) and @xmath64 ( solid ) ; corresponding force - displacement curves and bond fraction evolutions ( @xmath65 , @xmath66 ; dotted curve in @xmath65 : response of a without bond breaking ) ; order - of - magnitude estimate for the rate - dependence of the yield force , ( [ eq : rate_fc ] ) with const.@xmath67 , versus the numerically obtained maximum forces of the force - displacement curves for various mean force rates ( @xmath68 ) ; the minimum bond fraction ( @xmath68 , inset ) depends non - monotonically on the force rate ; @xmath69 , @xmath70 , @xmath71 , @xmath72 , @xmath73 . , title="fig:",width=7 ] for a larger relative deformation amplitude of @xmath64 , however , the predictions of the bare and the inelastic diverge .
the bare predicts a strong stiffening ( figure [ fig : force_displacement]b , red dotted curve ) , whereas the full model exhibits an initial stiffening regime followed by a pronounced softening ( figure [ fig : force_displacement]b , solid red curve ) .
a very interesting observation is that `` softening '' in this case does not only mean a decrease of the slope of the force - extension curve , but that the slope actually becomes _ negative _ over an extended parameter region , once an operationally defined threshold force @xmath56 is exceeded .
this `` flow state '' bears more resemblance to plastic flow than to viscoelastic relaxation .
the reason for this qualitatively new phenomenon is the force - induced bond breaking .
theoretically this is best exemplified by the time - dependence of the ( experimentally not easily accessible ) fraction @xmath30 of closed bonds ( figure [ fig : force_displacement]c ) .
a glance at the bond fraction during the weak deformation scenario ( figure [ fig : force_displacement]c , blue dashed curve ) reveals how the limit of a bare is recovered , namely whenever the deformation is not sufficiently violent and persistent to significantly decrease the fraction of closed bonds . for the large deformation scenario , in contrast ,
the bond fraction stays only initially constant ( figure [ fig : force_displacement]c , solid red curve ) . as a consequence of the large strain
the force rises steeply , as can be seen from the strong stiffening in figure [ fig : force_displacement]b ( solid red curve ) . at the yield force
, the bond fraction suddenly decreases to nearly half of its initial value during a very short time .
the decrease in the bond fraction is accompanied by a somewhat slower drop in the force , which is reflected in the sudden softening of the force - displacement curve .
the bond fraction eventually recovers on a much slower time scale , which is roughly given by @xmath74 .
note that even though the effects of the inelastic response dominate the stress - strain curve , the viscoelastic relaxations from the underlying model are still present .
they could hardly be disentangled from the inelastic contributions , though , without an underlying faithful model of the viscoelastic response at hand . in summary , we observe a competition between force - amplitude dependent stress - stiffening and force - rate dependent yielding events .
if the backbone force stays much smaller than the yield force , the adaptation of the bond network requires an adiabatically long time , on the order of the bond lifetime @xmath75 , so that plain viscoelastic behaviour is observed on the pulse time scale . in contrast , if the backbone tension @xmath6 reaches the yield force @xmath56 , the bond fraction declines sharply , thereby switching the response from the /-typical stiffening to a pronounced softening .
the rate dependence of the yield force @xmath56 can be estimated by setting the time it takes to reach the yield force , @xmath76 , equal to the force - dependent time scale of bond opening , @xmath77 , from ( [ eq : kpm ] ) , @xmath78 here , lw@xmath79 is the positive real branch of the lambert w function . in figure
[ fig : force_displacement]d , this estimate is compared with results from numerical evaluations for gaussian displacement pulses at different average rates .
apart from the numerical errors , the slight deviations from the estimate may be attributed to the fact that the force rate is not constant for the gaussian protocol .
they can be mostly eliminated by using force ramp protocols of various slopes instead of the gaussian displacement pulses . for not too low rates ,
the rate dependence can be approximated by the even simpler relation @xmath80 where the force rate has to be normalized by a suitable force scale .
the minimum bond fraction reached during the application of the pulse , which we interpret as a measure of the degree of fluidization , depends _ non - monotonically _ on the rate ( figure [ fig : force_displacement]d , inset ) .
qualitatively , this can be understood by noticing that two different factors influence the fluidization , namely the maximum force attained during deformation and the time over which the force is applied .
the maximum force is simply the yield force @xmath56 , with the rate dependence in ( [ eq : rate_fc ] ) and figure [ fig : force_displacement]d , while the reciprocal rate itself sets the time scale . for high rates @xmath81 ( in our example )
the rate - dependence of the maximum force wins and the minimum @xmath23 reached decreases with increasing rate . for low rates , where the rate - dependence of the maximum force is much weaker ( figure [ fig : force_displacement]d ) the minimum @xmath23 decreases with increasing pulse duration , _ viz .
_ decreasing rate . note that for slow pulses ( low rates ) , the bond fraction after the pulse may be quite different from the minimum @xmath23 , as significant recovery may already take place during the pulse .
the substantial changes in the material properties accompanying bond breaking can be exemplified by monitoring the linear elastic modulus @xmath82 ( measured at a fixed frequency @xmath83 ) in response to a strain pulse ( figure [ fig : modulus]a ) .
apart from the usual /-typical stress - stiffening below the threshold force @xmath84 , the modulus is seen to drop _ below _ the value it had before the deformation pulse , where it apparently saturates ( remember that the deformation vanishes for @xmath85 and that the recovery takes roughly a time @xmath86 ) .
we thus observe what we call a `` passive '' , `` physical '' remodelling of the bond network as opposed to the `` active '' , `` biological '' remodelling of the cytoskeleton of living cells in response to external stimuli .
these passive remodelling effects have recently been observed for human airway smooth muscle cells @xcite .
( a more quantitative discussion of the relation to the experimental observations will be given elsewhere . )
the fact that the deformation pulse leads to a decrease in @xmath46 and an increase in the loss angle @xmath87 ( figure [ fig : modulus]c ) suggests the notion of _ fluidization _ @xcite . and
the bond fraction @xmath23 largely determine the storage modulus @xmath46 and the loss angle @xmath87 : normalized storage modulus @xmath88 ( @xmath63 ) and loss angle @xmath87 ( @xmath66 ) during displacement pulses of large and small amplitudes , 2.2 ( solid ) , 0.73 ( dashed ) ; the stronger deformation pulse causes a strong and persistent overshoot of the loss angle , indicating _ fluidization _ ; slow recovery after the pulse , ( @xmath65 , @xmath68 ) ; parameters as in figure [ fig : force_displacement ] , @xmath89.,title="fig:",width=7 ] and the bond fraction @xmath23 largely determine the storage modulus @xmath46 and the loss angle @xmath87 : normalized storage modulus @xmath88 ( @xmath63 ) and loss angle @xmath87 ( @xmath66 ) during displacement pulses of large and small amplitudes , 2.2 ( solid ) , 0.73 ( dashed ) ; the stronger deformation pulse causes a strong and persistent overshoot of the loss angle , indicating _ fluidization _ ; slow recovery after the pulse , ( @xmath65 , @xmath68 ) ; parameters as in figure [ fig : force_displacement ] , @xmath89.,title="fig:",width=7 ] and the bond fraction @xmath23 largely determine the storage modulus @xmath46 and the loss angle @xmath87 : normalized storage modulus @xmath88 ( @xmath63 ) and loss angle @xmath87 ( @xmath66 ) during displacement pulses of large and small amplitudes , 2.2 ( solid ) , 0.73 ( dashed ) ; the stronger deformation pulse causes a strong and persistent overshoot of the loss angle , indicating _ fluidization _ ; slow recovery after the pulse , ( @xmath65 , @xmath68 ) ; parameters as in figure [ fig : force_displacement ] , @xmath89.,title="fig:",width=7 ] and the bond fraction @xmath23 largely determine the storage modulus @xmath46 and the loss angle @xmath87 : normalized storage modulus @xmath88 ( @xmath63 ) and loss angle @xmath87 ( @xmath66 ) during displacement pulses of large and small amplitudes , 2.2 ( solid ) , 0.73 ( dashed ) ; the stronger deformation pulse causes a strong and persistent overshoot of the loss angle , indicating _ fluidization _ ; slow recovery after the pulse , ( @xmath65 , @xmath68 ) ; parameters as in figure [ fig : force_displacement ] , @xmath89.,title="fig:",width=7 ] the passive remodelling described here is a _ transient _ phenomenon , because , after cessation of the external perturbations , the bond fraction will eventually recover its equilibrium value .
this indicates that also the change in the system properties is a transient effect , which is demonstrated by the recovery from fluidization in figure [ fig : modulus ] b & d. thus , while the fluidization resembles a plastic process on short time scales , on long times scales , the phenomenology is more similar to pseudo- and superelasticity as observed in shape - memory alloys @xcite . to demonstrate that the transient passive remodelling also affects the nonlinear material properties
, we consider a series of three pulses , next ( figure [ fig : stress_strain ] ) .
the force response to such a protocol is depicted in figure [ fig : stress_strain]b .
the force - displacement curve for the second stretching ( second left branch of the solid curve ) is less steep than for the first stretching ( very left branch of the solid curve ) and the yield force is lower .
this indicates a cyclic softening , or viscoelastic shake - down effect , closely related to the fluidization of the network by strain .
the strength of the shake - down depends on the fraction of transiently broken bonds , and hence on the rate as well as on the amplitude of the applied deformation ( c.f .
figure [ fig : stress_strain]c , solid red curve ) . upon repeated application of deformation pulses
the force - displacement curves settle on a limit - curve that is essentially preconditioned by the initial deformation pulse and the inelastic work it performed on the sample . for more gentle protocols that only break a smaller fraction of bonds ,
the initial fluidization would be not as pronounced and one would obtain a gradual shake - down , which converges to a limit curve only after many cycles . ) and response ( @xmath65 , @xmath66 ) for three subsequent gaussian strain pulses ; force - displacement curves ( @xmath65 ) for a single minimum interaction distance ( @xmath90 , solid ) and for a distribution of @xmath22 according to @xcite with mean @xmath91 ( dashed ) ; the bond fraction @xmath23 ( @xmath66 ) for @xmath92 ( dash - dotted ) , @xmath93 ( solid ) , and @xmath94 ( dashed ) ; other parameters as in figure [ fig : force_displacement ] .
, title="fig:",width=14 ] ) and response ( @xmath65 , @xmath66 ) for three subsequent gaussian strain pulses ; force - displacement curves ( @xmath65 ) for a single minimum interaction distance ( @xmath90 , solid ) and for a distribution of @xmath22 according to @xcite with mean @xmath91 ( dashed ) ; the bond fraction @xmath23 ( @xmath66 ) for @xmath92 ( dash - dotted ) , @xmath93 ( solid ) , and @xmath94 ( dashed ) ; other parameters as in figure [ fig : force_displacement ] . , title="fig:",width=7 ] ) and response ( @xmath65 , @xmath66 ) for three subsequent gaussian strain pulses ; force - displacement curves ( @xmath65 ) for a single minimum interaction distance ( @xmath90 , solid ) and for a distribution of @xmath22 according to @xcite with mean @xmath91 ( dashed ) ; the bond fraction @xmath23 ( @xmath66 ) for @xmath92 ( dash - dotted ) , @xmath93 ( solid ) , and @xmath94 ( dashed ) ; other parameters as in figure [ fig : force_displacement ] . ,
title="fig:",width=7 ] so far , we have assumed that there is one well - defined characteristic length scale @xmath22 for the polymer interactions , which is on the order of the entanglement length @xmath17 and interpreted as the minimum average contour distance between adjacent bonds of the test polymer with the background network .
this is clearly a mean - field assumption .
recent combined experimental and theoretical studies have established that the local tube diameter and entanglement length in pure semiflexible polymer solutions actually exhibit a skewed leptocurtic distribution with broad tails @xcite . to include this information in the above analysis is not straightforward , since the elastic interactions between regions of different entanglement length are not known , _ a priori_. for the sake of a first qualitative estimate , the simplest procedure seems to be to average the above results over a distribution of @xmath22 , corresponding to a parallel connection of independent entanglement elements .
qualitatively , this renders the abrupt transition from stiffening to softening somewhat smoother ( c.f .
figure [ fig : stress_strain]b , dashed curve ) .
moreover , the initial stiffening is slightly more pronounced .
this is due to the contributions of @xmath95 , which exhibit both stronger stiffening and fluidization
. nevertheless , all qualitative features like stiffening and remodelling effects and a `` flow state '' indicating fluidization or inelastic shakedown can still be well discerned .
this is consistent with the assumption that pure semiflexible polymer solutions are at least qualitatively well described by a single entanglement length @xcite .
we have presented a theoretical framework for a polymer - based description of the inelastic nonlinear mechanical behaviour of sticky biopolymer networks .
we represented the polymer network on a mean - field level by a test polymer described by the phenomenologically highly successful model . beyond the equilibrium statistical fluctuations captured by the original model
, we additionally allowed for a dynamically evolving thermodynamic state variable @xmath31 characterizing the network of thermo - reversible sticky bonds , which is updated dynamically upon bond breaking according to an appropriate rate equation . within our approximate treatment
we could obtain a number of robust and qualitatively interesting results , which are not sensitive to the precise parameter values chosen , nor to the choice of the transverse rather than the longitudinal susceptibility in ( [ eq : sus_wlc ] ) .
for sufficiently strong deformations , changes in the mean fraction of closed bonds , reflected in @xmath31 , can influence material properties to a degree at which the material behaves qualitatively different from the usual viscoelastic paradigm .
in particular , our theory predicts a pronounced fluidization response , which develops upon strong deformations on top of the intrinsic viscoelastic stiffening response provided by the individual polymers .
after cessation of loading , the system slowly recovers its initial state .
these observations are in qualitative agreement with recently published experimental results obtained for living cells @xcite , and a more quantitative comparison with dedicated measurement results for cells and _ in vitro _
biopolymer networks will be the subject of future work .
we found the yield force for the onset of the fluidization to be sensitive to the deformation rate .
moreover , the fluidization response was shown to be accompanied by a cyclic softening or shake - down effect .
taking into account the spatial heterogeneity of biopolymer solutions by a distribution of entanglement lengths leads to a smoothing of the force - displacement curves without affecting their qualitative characteristics .
one may still expect qualitatively new effects in situations with unusually broad distributions of entanglement lengths , such as for strongly heterogeneous ( e.g. phase separating ) networks .
a parameter that was found to be very important for cells @xcite but has not been discussed much in this contribution is the _ prestressing force _
@xmath96 ( see appendix [ sec : technical ] ) , which is present in adhering cells even in the absence of external driving . experimentally , it was established that increasing prestress is correlated with a higher stiffness of adhering cells @xcite which in turn is correlated with a higher stiffness of the substrate @xcite .
when naively treating the prestressing force just as an additive contribution to the overall force , our model predicts the opposite : the additional force breaks more bonds and the network becomes softer and more fluid - like . to reconcile these apparently contradicting trends , one can appeal to the notion of _ homeostasis _ , which essentially amounts to postulating that the cell actively adapts its structure such that a certain set of thermodynamic state variables remain in a physiologically sensible range @xcite .
this basically implies that the cell will respond to stiff substrates or persistent external stresses by a biological remodelling that corresponds to an increase of @xmath12 and @xmath55 , and possibly to a decrease of @xmath22 . by adapting also the internal prestress @xmath96 ,
the cell may then avoid , apart from the structural collapse , an equally undesirable loss of flexibility .
we note , however , that internal stresses are indeed observed to disrupt the cytoskeleton , as implied by the simple physical theory presented here , if they are not permanently balanced by a substrate @xcite .
even though the nonlinear effects presented in this contribution depend crucially on the bond kinetics , it should not be overlooked that the is an equally indispensable ingredient of the complete theory .
first , it provides the constitutive relation that gives the force in response to an infinitesimal displacement , which in turn governs the bond kinetics .
secondly , the time - dependent linear susceptibility strongly filters the dynamic remodelling of the bond network in practical applications , in which one rarely has access to the microstate of the underlying bond network . in summary ,
the extended model thus integrates experimentally confirmed features of the response such as slow relaxation , power - law rheology , viscoelastic hysteresis , and shear stiffening with a simple bond kinetics scheme , which allows less intuitive complex nonlinear physical remodelling effects like fluidization and inelastic shake - down to be addressed .
finally , we want to point out that it is straightforward to extend the above analysis in a natural way to account for irreversible plastic deformations @xcite . to this end
, one has to allow for the possibility that the transiently broken bonds reform somewhere else than at their original sites ( as always assumed in the foregoing discussion ) after strong deformations with finite residual displacement
. this can be included by accommodating an additional term acting as a `` slip rate '' in ( [ eq : rate ] ) @xcite .
we acknowledge many fruitful discussions with j. glaser , s. sturm , and k. hallatschek , and financial support from the deutsche forschungsgemeinschaft ( dfg ) through for 877 and , within the german excellence initiative , the leipzig school of natural sciences building with molecules and nano - objects .
the deformation pulse protocol used in this paper is @xmath97 to obtain the full nonlinear response of the ( extended ) to the displacement protocol ( [ eq : strain_gauss ] ) , we make use of the superposition principle : we know that the fourier transform of @xmath98 is the linear response to a small displacement step .
after decomposing a finite displacement into infinitesimal displacement steps and a partial integration , we write @xmath99 note that the right hand side of ( [ eq : force ] ) depends on @xmath100 , rendering it a highly nonlinear implicit equation . using ( [ eq : strain_gauss ] ) and integrating by parts , we obtain @xmath101 the next step is to identify a connection between the force in ( [ eq : rate ] ) and @xmath102(t ) .
a finite prestressing force @xmath96 is introduced mainly for technical reasons , namely to avoid the unphysical region of negative ( i.e. compressive ) backbone stress , which would buckle the polymers .
while the physics of the prestressed network can essentially be mapped back to that of a network without prestress by a renormalization of the parameters @xmath12 and @xmath55 , prestress may also be seen as an important feature : indeed , the cytoskeleton of adhered cells is well known to be under permanent prestress , and suspended cells seem prone to spontaneous shape oscillations that could be indicative of a propensity of the cells to set themselves under prestress @xcite . in the context and on the longer time scales of the biological remodelling of the cytoskeleton
it would therefore make sense to think of @xmath96 as a dynamic force generated by molecular motors and polymerization forces . for the following , however , we take the prestress to be constant so that the force entering ( [ eq : rate ] ) is @xmath103 we now face the problem that ( [ eq : force ] ) is an implicit equation for @xmath100 , which depends on @xmath30 , which in turn depends via @xmath32 on @xmath104 .
we therefore use a two - step euler scheme : as initial values for @xmath6 and @xmath23 , we choose the prestressing force , @xmath105 , and the steady - state value under the prestressing force , @xmath106 , respectively .
we then choose a sufficiently small time step @xmath107 and apply the following iterative rule : @xmath108 where @xmath109 .
in the limit @xmath110 , this procedure converges to the exact solution .
ke kasza , f nakamura , s hu , p kollmannsberger , n bonakdar , b fabry , tp stossel , n wang and da weitz 2009 filamin a is essential for active cell stiffening but not passive stiffening under external force * 96 43264335 . * eh zhou , x trepat , cy park , g lenormand , mn oliver , sm mijailovich , c hardin , da weitz , jp butler and jj fredberg 2009 universal behavior of the osmotically compressed cell and its analogy to the colloidal glass transition * 106 10632 . *
n wang , i m tolic - norrelykke , jx chen , sm mijailovich , jp butler , jj fredberg and d stamenovic 2002 cell prestress .
i. stiffness and prestress are closely associated in adherent contractile cells * 282 c606c616 . * | we propose a physical model for the nonlinear inelastic mechanics of sticky biopolymer networks with potential applications to inelastic cell mechanics .
it consists in a minimal extension of the glassy wormlike chain ( ) model , which has recently been highly successful as a quantitative mathematical description of the viscoelastic properties of biopolymer networks and cells . to extend its scope to nonequilibrium situations , where the thermodynamic state variables may evolve dynamically ,
the is furnished with an explicit representation of the kinetics of breaking and reforming sticky bonds . in spite of its simplicity
the model exhibits many experimentally established non - trivial features such as power - law rheology , stress stiffening , fluidization , and cyclic softening effects . |
several articles have recently focused on the electrocaloric effect ( ece ) in ferroelectrics and related materials,@xcite which bears analogy with the well known magnetocaloric effect ( mce).@xcite here we investigate the mechanisms of ece in relaxor ferroelectrics , to be referred to as _ relaxors _ and normal ferroelectrics ( or _ ferroelectrics _ ) , and discuss the specific features of these two groups of materials . in particular , we will discuss the possibility of achieving a giant ece in bulk inorganic relaxors and ferroelectric materials as well as in organic polymers .
these systems offer the prospect of practical applications , such as miniaturized and energy efficient cooling devices , without the need for large electric currents commonly associated with the mce .
a crucial physical quantity in ece is the change of entropy of a polar material under the application and/or removal of an external electric field .
for example , when the electric field is turned on isothermally , the elementary dipolar entities in the system will become partially ordered and the entropy will be lowered .
the entropy lowering of the dipolar subsystem is then compensated by an increase of the temperature of the total system , which characterizes the ece .
the degree of lowering depends on the number of statistically significant configurations in the initial and final states of the system , as well as on the size of the average dipole moment and the volume density of dipolar entities .
other factors may also play a role : if the system undergoes a first order phase transition under the action of external electric field , the entropy will be enhanced on crossing the borderline between the two phases , resulting in a larger ece .
the line of first order transition points terminates at a critical point where the transition becomes continuous,@xcite and it will be of special interest to investigate the behavior of ece in the vicinity of the critical point .
estimates of the ece can be made on the basis of thermodynamic maxwell relations using the measured heat capacity and the field and temperature dependence of the dielectric polarization . from the theoretical point of view ,
a central problem is how to make predictions about the temperature and field dependence of ece . as a first step
, one needs to develop an appropriate phenomenological and/or mesoscopic model , which incorporates the specific physical features of the systems . here
we will make use of the standard landau phenomenological model , which can be applied to both relaxors and ferroelectrics with the corresponding choice of landau coefficients .
these in turn can be derived from the mesoscopic model of the material under study . in the case of relaxors ,
the mesoscopic model of choice is the spherical random bond
random field ( srbrf ) model , which is based on the concept of reorientable polar nanoregions ( pnrs).@xcite thus we should be able to compare the ece in relaxors and ferroelectrics , and determine the parameters , which control the ece in these systems .
finally , using general principles of statistical thermodynamics we will discuss the existence of a theoretical upper bound on the ece and argue that it satisfies a universal relation , which is , in principle , also applicable to mce .
the temperature change of a polar system under adiabatic electric field variation from the initial value @xmath3 to final value @xmath4 can be written in the form @xcite @xmath5 which follows from the thermodynamic maxwell relation @xmath6 involving the entropy density @xmath7 and the physical dielectric polarization @xmath8 ( in units of c / m@xmath9 ) . the volume specific heat at constant field is given by @xmath10 . in deriving eq .
( [ dt1 ] ) , one tacitly assumes that the fluctuations of polarization @xmath11 can be ignored and that @xmath8 represents a thermodynamic variable given by the macroscopic average of @xmath11 .
furthermore , it is implied that the system is ergodic , i.e. , its response time much shorter than the experimental time scale . if the field and temperature dependence of @xmath12 is known from experiments , the integral in eq .
( [ dt1 ] ) can be evaluated , yielding an estimate for @xmath13.@xcite in model calculations , it seems convenient to change the integration variable in eq .
( [ dt1 ] ) from @xmath14 to @xmath15 .
this is readily done by applying the thermodynamic identity@xcite @xmath16 with the result @xmath17 this expression is fully equivalent to eq .
( [ dt1 ] ) , with the new integration limits given by @xmath18 and @xmath19
. the partial derivative @xmath20 can be obtained from the free energy density functional @xmath21 . ignoring fluctuations of the order parameter @xmath8
, we write @xmath22 as a power series @xmath23 this has the standard form of a mean field free energy expansion with temperature dependent coefficients @xmath24 , etc . applying the equilibrium condition @xmath25
, we obtain the equation of state @xmath26 and the temperature derivative in eq .
( [ dt2 ] ) becomes @xmath27 where @xmath28 , @xmath29 etc .
it should noted be that @xmath19 in eq .
( [ dt2 ] ) is that solution of eq .
( [ es1 ] ) , which simultaneously minimizes the free energy ( [ f ] ) . the integration in eq .
( [ dt2 ] ) can now be carried out , yielding @xmath30 . \label{dt3}\ ] ] in passing , we note that @xmath31 , in general , depends on the temperature ; however , in writing down eqs .
( [ dt1 ] ) and ( [ dt2 ] ) the temperature dependence of the heat capacity had already been ignored .
the expression in brackets is related to the change of the entropy density @xmath32 . using eq .
( [ f ] ) we can write @xmath33 the first term @xmath34 is the entropy at @xmath35 .
it contains the configuration entropy of dipolar entities , which depends on the number of equilibrium orientations @xmath36 , say , @xmath37 for the @xmath38 equilibrium case.@xcite thus we may expect that @xmath39 , @xmath40 being the total number of dipolar entities such as pnrs in relaxors .
the second term is given by @xmath41 back in eq .
( [ dt3 ] ) , @xmath42 cancels out and the ece temperature change can be rewritten in the familiar form @xcite @xmath43 with @xmath44 .
it should be noted that the values of all temperature - dependent quantities @xmath45 , @xmath8 , @xmath46 , @xmath47 , etc .
, on the r.h.s .
( [ dt3 ] ) are taken at the initial temperature @xmath48 , and @xmath31 at the final field value @xmath49 .
the coefficients @xmath50 can be expressed in terms of linear and nonlinear susceptibilities by formally inverting the relation ( [ es1 ] ) and writing @xmath51 as a power series in @xmath49.@xcite in landau theory , close to a second order phase transition one sets @xmath52 , while @xmath53 are constants .
thus , @xmath54 const . , and @xmath55 .
this leaves only one nonzero term of the order @xmath56 in eq .
( [ dt3 ] ) . on the other hand , @xmath57 and the nonlinear susceptibilities
are also found to diverge when @xmath58 .
thus a formal inversion @xmath51 of eq .
( [ es1 ] ) in powers of @xmath49 would lead to a poorly converging series . in the following
we will apply eq .
( [ dt3 ] ) in order to discuss the predictions of landau theory in two characteristic cases , namely , normal ferroelectrics and relaxors .
as already mentioned , in landau theory of phase transitions in ferroelectrics , the coefficients @xmath53 in eq .
( [ f ] ) are assumed temperature independent and @xmath59 , where @xmath60 is the curie - weiss temperature .
when @xmath61 and @xmath62 , a second order transition occurs at @xmath63 . for @xmath2 and @xmath64 , a first order transition appears at a temperature @xmath65 , given by the relation @xmath66 .
writing @xmath67 , where @xmath68 is the curie constant , we find : @xmath69 for @xmath70 , a critical point will be located at @xmath71 , where @xcite @xmath72 turning next to relaxors , we assume that the relevant elementary dipolar entities at temperatures around the dielectric maximum are polar nanoregions or pnrs . according to the srbrf model,@xcite these pnrs are coupled through gaussian random interactions @xmath73 ( `` random bonds '' ) and are subject to gaussian random fields @xmath74 .
in zero applied field , spontaneous long range order is suppressed @xmath75 .
this means that for a relaxor we can still use the free energy ( [ f ] ) , however , the coefficient @xmath76 must remain positive at all temperatures .
thus , for @xmath61 and @xmath62 there can be no second order phase transition .
the explicit form of @xmath76 follows from eq .
( [ es1 ] ) , namely , @xmath77 where @xmath78 represents the ( quasi)static linear field - cooled dielectric susceptibility .
this can be derived from the srbrf model of relaxors:@xcite @xmath79 with @xmath80 the effective curie constant @xmath68 is given explicitly in terms of the average squared dipole moments @xmath81 of pnrs , namely , @xmath82 , and the average volume @xmath83 associated with a pnr .
later , we will also introduce the saturation polarization @xmath84 , where @xmath85 . for simplicity , we will henceforth neglect the difference between @xmath86 and @xmath87 , and write @xmath88 .
the parameter @xmath68 can be determined experimentally from the asymptotic high temperature behavior of @xmath89.@xcite the parameter @xmath90 is defined _ via _ the average over the infinitely ranged random interaction @xmath91_{av}=j_0/n$ ] .
the spherical glass order parameter , @xmath92 , is a measure of the degree of disorder .
for @xmath35 it is determined by the cubic equation @xmath93 here , @xmath94 is proportional to the variance of the random bond distribution according to @xmath95_{av}-([j_{ij}]_{av})^2=j^2/n$ ] , while @xmath96 measures the correlations of quenched random fields , i.e. , @xmath97_{av}=0 $ ] and @xmath98_{av}=\delta_{ij}\delta$ ] . if @xmath99 , long range order will be suppressed ( @xmath100 ) .
thus the relaxor state is characterized by three physical parameters : @xmath101 , @xmath102 , and @xmath96 .
typically , in relaxors one finds @xmath103 , implying that random bonds are effectively much stronger than random fields .
this allows pnrs to reorient collectively under the action of external fields and relax towards equilibrium . in the opposite case , @xmath104 , pnrs are would be trapped in a frozen static configuration of random fields and no characteristic low - frequency relaxor response due to pnr flipping could be observed .
here we will consider only the case @xmath103 . from eqs .
( [ chi1 ] ) and ( [ achi ] ) we derive the coefficient @xmath76 : @xmath105 the linear susceptibility @xmath89 has been fitted to experimental data for the static field - cooled response in a variety of relaxor systems , from which the parameters of the model have been obtained.@xcite formally , the srbrf model also yields explicit expressions for nonlinear susceptibilities , from which the coefficients @xmath1 and @xmath106 in the free energy ( [ f ] ) can be determined.@xcite however , it has been shown earlier@xcite that realistic values of these coefficients can only be obtained if the coupling between pnrs and lattice strain fluctuations is included .
several mechanisms for such a coupling have been investigated both at a mesoscopic @xcite and phenomenological level.@xcite it has also been shown@xcite that in real three dimensional systems strain coupling gives rise to anisotropy of the anharmonic terms in the free energy.@xcite specifically , strain coupling may change the sign of the coefficient @xmath1 and hence of the corresponding nonlinear susceptibility @xmath107 for a given direction of the applied field . in the following
, we will simply consider @xmath1 and @xmath106 as free parameters and discuss separately the cases @xmath61 and @xmath2 , while keeping @xmath64 . for cubic systems , eq .
( [ f ] ) can be rewritten in a general form @xcite @xmath108 where @xmath109 are so - called dielectric stiffness coefficients with @xmath110 . for a system with orthorhombic symmetry in a field along
@xmath111 $ ] we recover eq .
( [ f ] ) , where @xmath8 is the total polarization , @xmath112 , @xmath113}=(4/3)(\alpha_{11}+\alpha_{12})$ ] , and @xmath114}=(2/9)(3\alpha_{111}+6\alpha_{112}+\alpha_{123})$].@xcite experiments in pmn @xcite indicate that @xmath115 for @xmath116 $ ] , implying @xmath113}<0 $ ] . on the other hand , for tetragonal symmetry with
@xmath117 $ ] one has @xmath118}=4\alpha_{11}$ ] and @xmath119 .
for example , in pmn @xcite one finds in this case that @xmath120 or @xmath118}>0 $ ] , indicating that @xmath121 and @xmath122 .
if @xmath123 at all temperatures , there is no first order phase transition for @xmath62 . if @xmath124 , however , a first order phase transition with a jump of polarization @xmath125 occurs for @xmath49 exceeding some threshold value @xmath126
this can readily be seen by numerical minimization of the free energy ( [ f ] ) for any pair @xmath127 . as the temperature increases
, @xmath128 decreases and vanishes at an isolated critical point @xmath71,@xcite where the derivative @xmath129 diverges.@xcite the critical temperature is determined from the equation @xmath130 and @xmath131 from the second of eqs .
( [ cp ] ) .
it should be noted that , generally speaking , @xmath1 and @xmath106 can also be functions of @xmath132.@xcite for @xmath133 the system is in a supercritical regime with continuous temperature and field dependence of @xmath12 .
to illustrate the temperature and field dependence of the ece in relaxors and ferroelectrics , we calculate @xmath134 from eq .
( [ dt3 ] ) for a selected set of parameter values .
first , we introduce rescaled , dimensionless quantities @xmath22 and @xmath8 according to @xmath135 and @xmath136 , where @xmath88 is the saturation polarization occurring at high field values and/or low temperatures .
this requires a rescaling of the remaining parameters according to @xmath137 , @xmath138 , @xmath139 , etc . also , we redefine @xmath140 and @xmath141 . from eq .
( [ at1 ] ) we see that in relaxors the rescaled parameter @xmath76 behaves as @xmath142 , and @xmath143 becomes @xmath144 here and until the end of this section , the symbols @xmath50 and @xmath145 refer to dimensionless , rescaled parameters , but elsewhere in this paper the same symbols denote the true , physical values of these quantities . in ferroelectrics , @xmath146 and thus @xmath147 , and @xmath148 . in relaxors , in the high temperature limit @xmath149 , @xmath150 tends asymptotically to zero , thus @xmath151 and @xmath152 , i.e. , the same as in ferroelectrics . in fig . 1 ,
@xmath76 and @xmath143 are plotted for a relaxor ferroelectric with @xmath153 and @xmath154 .
also shown is the behavior of a normal ferroelectric .
it should be noted that the essential difference between ferroelectrics and relaxors is the behavior of the corresponding coefficients @xmath76 and @xmath143 . in the following
we will choose @xmath155 const .
@xmath156 and @xmath157 .
the corresponding ece temperature change is obtained from eq .
( [ dt3 ] ) . using the fact that @xmath158 etc .
, we find @xmath159 . \label{dt5}\ ] ] the polarization @xmath12 will be calculated numerically by simultaneously solving eq .
( [ es1 ] ) and minimizing the free energy ( [ f ] ) .
we will do that separately for the two cases @xmath61 and @xmath2 , assuming the denominator @xmath160 to be a constant amplitude factor .
\(i ) _ case @xmath61_. as already stated , @xmath161 in a relaxor , but in ferroelectrics @xmath162 for @xmath163 .
the spontaneous polarization @xmath164 is obtained by minimization of @xmath165 . in real systems
, @xmath45 may not be spatially uniform due to domains .
here we assume that @xmath166 has the same value in all domains regardless of their orientation , and that the contribution of domain walls to the entropy can be neglected.@xcite in fig .
2(a ) we show the calculated values of @xmath13 for a relaxor as function of temperature for various values of @xmath167 , where @xmath131 is formally given by eq .
( [ cp ] ) , although the critical point does not exist for @xmath61 . also shown in fig .
2(b ) is @xmath13 for a ferroelectric with the same parameters @xmath168 , but with different @xmath169 and @xmath46 .
@xmath13 has a peak at @xmath170 and is in general larger than in the relaxor case . at higher temperatures
, however , the difference gradually disappears .
\(ii ) _ case @xmath2_. eqs .
( [ chi1]-[q ] ) imply @xmath171 .
thus , @xmath172 and the first order phase transition in relaxors at @xmath62 is suppressed . in a ferroelectric , however , a first order transition in zero field occurs at @xmath173 .
the critical point is located at @xmath174 in relaxors , and at @xmath175 in ferroelectrics , while @xmath176 in both cases . in general , @xmath13 is found to increase with increasing field and exhibits a peak as a function of temperature . in relaxors ,
the peak position moves to higher temperatures with increasing field values , whereas in ferroelectrics the maximum is located at @xmath177 , where a jump of the spontaneous polarization occurs . in fig . 3 ,
@xmath178 is plotted for four values of the field @xmath179 , as indicated . at @xmath180 in relaxors
, there is a jump of @xmath13 due to the field - induced first - order transition . on the other hand , in the case of a ferroelectric
, there is a jump of @xmath13 at the zero - field first order transition temperature @xmath173 . at the critical point
, @xmath13 is continuous but with an infinite slope in both cases , while at higher temperatures the difference between relaxors and ferroelectrics tends to disappear . in fig . 4 , the ece efficiency @xmath181 is plotted as a function of @xmath167 for four values of temperature close to @xmath182 . as expected , the maximum efficiency is obtained at the corresponding critical points @xmath183 and @xmath180 .
larger values of @xmath13 in ferroelectrics rather than in relaxors are mainly due to the sharp decrease of @xmath143 in relaxors at @xmath184 ( cf .
fig . 1 ) .
however , this does not mean that ferroelectrics are better candidates for achieving giant ece .
namely , one should bear in mind that there are other parameters , such as @xmath68 and @xmath31 , which also have a strong impact on the ece .
moreover , the above comparison between relaxors and ferroelectrics makes only sense if the coefficients @xmath185 are indeed the same in both cases .
therefore , in discussing the ece in specific systems one should carefully consider the actual physical values of all the relevant model parameters , as discussed in the following section .
it has been observed experimentally in a variety of systems that the entropy change @xmath186 in eq .
( [ dt4 ] ) is proportional to @xmath187,@xcite suggesting that the terms of order @xmath188 and higher in the expansion ( [ s1 ] ) make no contribution .
assuming for simplicity that @xmath100 , we recover from eq .
( [ s1 ] ) the empirical quadratic relation @xmath189 where the coefficient @xmath190 can be expressed through eqs .
( [ s1 ] ) and ( [ at1 ] ) , i.e. , @xmath191 according to the landau model , in ferroelectrics the partial derivative is equal to @xmath192 or @xmath193 at all temperatures , whereas in relaxors it approaches the value @xmath194 at high temperatures , but is in general a function of temperature .
thus , in relaxors , @xmath190 is expected to be a function of temperature with @xmath195 . using the relations ( [ dt4 ] ) and ( [ ds ] ) , the ece temperature change @xmath13 can be written as @xmath196 a quadratic relation of the same form is predicted by eq .
( [ dt5 ] ) if the terms @xmath47 , @xmath197 , ... etc . are neglected .
it is exactly true in the case @xmath55 discussed in section iii .
clearly , the above relation is applicable only in the _ quadratic _ regime where the empirical eq .
( [ ds ] ) is valid .
the parameter @xmath190 can be determined directly from the measured ece temperature change @xmath13 , with @xmath12 extracted from dielectric experiments . alternatively , and especially in ferroelectrics or relaxors at temperatures above the freezing temperature
, we can obtain both @xmath12 and the curie constant @xmath68 from the dielectric data , and then deduce @xmath190 from eq .
( [ beta ] ) .
we can then predict @xmath13 from eq .
( [ dt6 ] ) .
it turns out , however , that the values of @xmath13 measured directly may differ from the ones deduced from dielectric data .
for example , in the case of relaxor ferroelectric terpolymer p(vdf - trfe - cfe ) at @xmath198 k and @xmath199 mv / m , the ece measured directly is @xmath200 k , leading to @xmath201 v m c@xmath202k@xmath202,@xcite whereas the value deduced from dielectric data is @xmath203 k.@xcite a tentative explanation of this discrepancy is that even far above the freezing temperature @xmath204 k the system may still be nonergodic , so that the maxwell relations , based on equilibrium thermodynamics , are not applicable.@xcite as a consequence , the value of @xmath12 measured on a short time scale is smaller than its thermodynamic long - time limit .
another possibility is that the empirical relation eq .
( [ ds ] ) is an effective quasi - linear relation observed in a broad range of large field values , while its derivation based on the landau expansion is by assumption restricted to small fields .
we can obtain an estimate for the maximum ece temperature change @xmath205 by assuming that in a sufficiently strong electric field the polarization reaches its saturation value @xmath88 .
relation ( [ ds ] ) is not expected to be valid in this _ saturation _ regime , and we must return to the general expression ( [ dt4 ] ) .
obviously , the expansion ( [ s1 ] ) can not be applied due to convergence problems . on the other hand ,
it is well known @xcite that in the saturation regime the excess entropy of the dipolar subsystem tends to zero .
therefore , according to eq .
( [ s ] ) , @xmath206 should approach the negative value of the configuration entropy @xmath42 , i.e. , @xmath207 .
thus , in the saturation regime eq .
( [ dt4 ] ) leads to @xmath208 this relation gives the theoretical upper bound on ece in terms of just three physical quantities , @xmath209 , @xmath31 , and the configuration number @xmath36 .
interestingly , this result does not depend explicitly on the dipole moment @xmath87 .
moreover , it does not contain any information about possible phase transition occurring in the quadratic regime . for a given value of electric field @xmath49
, the borderline between the two regimes is expected to occur at some temperature @xmath210 where the dipolar energy becomes equal to the thermal fluctuation energy , i.e. , @xmath211 . for @xmath212
the system is in the quadratic , and for @xmath213 in the saturation regime . for the above terpolymer p(vdf - trfe - cfe ) , we can estimate the dipole moment @xmath87 from the relations @xmath214 c / m@xmath9 and @xmath215 . using the value of @xmath190 determined directly from @xmath13
, we have @xmath216 c m , and for @xmath217 mv / m we thus find @xmath218 k. the average volume associated with a pnr in relaxors , @xmath83 , in eq .
( [ dtm1 ] ) is not _ a priory _ known and depends on the total number of pnrs .
we can estimate @xmath209 from the measured values of @xmath68 and @xmath219 using the relation @xmath220 , and rewrite eq .
( [ dtm1 ] ) in the form @xmath221 the value of @xmath219 can be extracted , for example , from hysteresis loops in the saturation regime , and @xmath68 from the asymptotic behavior of @xmath89 . as already discussed above , this value of @xmath68 may differ from the one derived from the experimental value of parameter @xmath222 , observed in ece experiments in the effective quadratic regime . in table
i , the predicted values of @xmath205 for a set of selected systems are listed using the values of @xmath68 deduced from dielectric experiments .
it should be noted that eqs .
( [ dtm1 ] ) and ( [ dtm2 ] ) provide just a theoretical upper bound for the ece in the systems listed . in practice ,
the limit of a fully polarized dipolar subsystem might not be accessible because dielectric breakdown could occur before complete saturation is reached .
nonetheless , the predicted values of @xmath205 permit a comparison between various sytems and might be useful in the search for a giant ece .
( [ dtm1 ] ) indicates that a giant ece is expected to occur in systems with a small value of @xmath209 , or equivalently a large number @xmath40 of dipolar entities at fixed volume @xmath223 . for illustration ,
let us consider a specific example , i.e. , the ferroelectric copolymer p(vdf - trfe ) .
this system consists of microscopic crystalline layers of polarized material embedded in an amorphous environment.@xcite electron irradiation breaks up the layered structure into smaller dipolar units , and turns the polymer into a relaxor . since the number of these new entities is now larger than the number of the original microcrystallites , and assuming the same value of saturation polarization , one expects a stronger ece to occur in the irradiated relaxor copolymer than in the original ferroelectric copolymer .
this is corroborated by the experimental values of the coefficient @xmath190 for the irradiated relaxor copolymer p(vdf - trfe),@xcite @xmath224 v m c@xmath202k@xmath202 , and for the original ferroelectric copolymer in the paraelectric phase,@xcite @xmath225 v m c@xmath202k@xmath202 . in the case of inorganic relaxor 8/65/35 plzt thin films , the value of the coefficient @xmath190 is@xcite @xmath226 v m c@xmath202k@xmath202 , whereas for the ferroelectric pzt one finds@xcite @xmath227 v m c@xmath202k@xmath202 .
again , @xmath190 is larger for the relaxor ; however , this comparison seems less conclusive since the difference in composition between the two systems is much greater than in the above organic case .
relation ( [ dtm1 ] ) has been derived here in the framework of landau theory , however , its validity is essentially based on thermodynamic and statistical principles and is hence quite general . in particular , it is independent of any mesoscopic models such as the srbrf model . moreover , it can be easily generalized to magnetic systems , where @xmath13 represents the mce temperature change .
the smallest physical limit for @xmath209 in ferroelectrics is the volume of the unit cell @xmath228 , yielding the ultimate upper bound on @xmath205 in eq .
( [ dtm1 ] ) for ferroelectric materials . to emphasize this point ,
let us multiply the numerator and denominator by the avogadro number @xmath229 to obtain @xmath230 here , @xmath231 is the gas constant and @xmath232 the molar specific heat .
this simple result does not explicitly contain any information on the microscopic nature of the system , i.e. , whether it is dielectric , magnetic , etc .
, since the corresponding electric or magnetic dipole moment does not appear in eq .
( [ dtm3 ] ) .
therefore , we can regard the above relation as a universal law for the theoretical upper bound on the ece and/or mce in electrically or magnetically polarizable solids . of course
, eq . ( [ dtm3 ] ) implies that the polarization or magnetization must have reached complete saturation . in the dielectric case
, this requires large electric fields with the already mentioned possibility of dielectric breakdown . in the magnetic case , however , extremely large fields of the order of @xmath233 t may be necessary .
( [ dtm3 ] ) is valid in the saturation regime @xmath234 , typically at low temperatures , where @xmath235 is expected to be temperature dependent . in ferroelectrics
, the formation of domains should be taken into account , and the entropy limit @xmath236 could only be reached in very high fields . in relaxors , there is an additional difficulty that the relaxation of pnrs at low temperatures is extremely slow and the above limit can only be reached after very long times . at high temperatures , @xmath235 normally approaches a certain limit .
in metals one has @xmath237 according to the dulong - petit law , which is generally not valid for complex solids .
for example , in perovskites , @xmath238.@xcite thus , the maximum ece in perovskites at high temperatures is of the order @xmath239 .
this limit applies to a large group of perovskites , i.e. , paraelectrics , ferroelectrics , relaxors , etc .
specifically , for @xmath37 ( see table i ) , we find @xmath240 .
for example , if the system can support large electric fields such that @xmath241 k , this would lead to a giant ece temperature change @xmath242 k at room temperature .
this exceeds roughly by a factor of @xmath243 the estimated values for perovskites in table i , which were derived from the physical values of the parameters @xmath219 , @xmath68 , and @xmath31 .
in this paper , we studied the mechanism of ece in relaxor ferroelectrics ( or _ relaxors _ ) and in normal ferroelectrics ( referred to as _ ferroelectrics _ ) .
starting from the widely accepted result for the ece temperature change @xmath134 , based on the thermodynamic maxwell relation , we derived an alternative expression which could directly be applied in theoretical model calculations .
the results for @xmath134 have been obtained in two physical regimes of field - temperature variables @xmath127 .
the first of these is the so - called _ quadratic _ regime , where @xmath13 is proportional to the square of the dielectric polarization @xmath12 , and the second the _ saturation _ regime , where @xmath12 is allowed to reach its maximum value @xmath219 . in the quadratic regime
, the system can be described by the landau - type free energy model , in which the harmonic landau coefficient @xmath76 depends on the physical nature of the system , i.e. , the behavior of @xmath76 in relaxors differs from that of ferroelectrics .
the anharmonic coefficients @xmath53 , which are common to both cases , then determine the critical behavior of the system in a field @xmath49 .
the case @xmath61 does not show any pronounced anomalies . for @xmath2 , the polarization @xmath8 and
hence @xmath13 rises steeply in a relaxor near the isolated critical point @xmath244 , whereas in a ferroelectric , the largest effect occurs near the first order phase transition in zero field , which is absent in the relaxor case .
the ece efficiency @xmath181 shows a similar behavior in both cases , and at higher fields the same asymptotic values are found . of course
, these conclusions only apply provided that all the remaining physical features , such as the number of equilibrium orientations @xmath36 of the elementary dipolar entities , etc .
, are the same in both cases .
experimentally , the ece entropy change in the _ quadratic _ regime is found to behave as @xmath245 , which is trivially reproduced by the landau model .
the coefficient @xmath190 tends to have larger values in relaxors , leading to a stronger ece . in irradiated organic polymer relaxors ,
this can be explained by the larger number of polar nanoregions ( pnrs ) and thus a smaller average pnr volume @xmath209 . in the _ saturation _ regime
, the entropy of the dipolar subsystem generally approaches the negative value of the configuration entropy , @xmath246 , and the maximum ece value @xmath205 thus crucially depends on the orientational degeneracy @xmath36 .
this work was supported by the slovenian research agency through grants p1 - 0044 , p1 - 0125 , p2 - 0105 , j1 - 0155 , and j1 - 2015 .
research at the pennsylvania state university was supported by the us doe , office of basic energy sciences , division of materials science and engineering under award no . de - fg02 - 07er46410 .
a. s. mischenko , q. zhang , j. f. scott , r. w. whatmore , n. d. mathur , science * 311 * , 1270 ( 2006 ) .
a. s. mischenko , q. zhang , r. w. whatmore , j. f. scott , n. d. mathur , appl .
. lett . * 89 * , 242912 ( 2006 ) .
t. m. correia , j. s. young , r. w. whatmore , j. f. scott , n. d. mathur , q. zhang , appl .
lett . * 95 * , 182904 ( 2009 ) .
b. neese , b. chu , s. g. lu , y. wang , e. furman , q. m. zhang , science * 321 * , 821 ( 2008 ) .
a. m. tishin and y. i. spichkin , _ the magnetocaloric effect and its applications _ ( iop publishing ltd . , bristol , uk , 2003 ) . k. a. gschneidner jr . , v. k. pecharsky , and a. o. tsokol , rep .
phys . * 68 * , 1479 ( 2005 ) .
z. kutnjak , j. petzelt , r. blinc , nature * 441 * , 956 ( 2006 ) .
r. pirc and r. blinc , phys . rev .
b * 60 * , 13470 ( 1999 ) .
h. e. stanley , _ phase transitions and critical phenomena _
( oxford university press , oxford , uk , 1971 ) , p. 23 . y. moriya , h. kawaji , t. tojo , t. atake , phys .
lett . * 90 * , 205901 ( 2003 ) .
s. ikeda , h. kominami , k. koyama , y. wada , j. appl .
* 62 * 3342 ( 1987 ) .
m. iwata , z. kutnjak , y. ishibashi , r. blinc , j. phys .
japan * 77 * , 034703 ( 2008 ) .
k. uchino .
, s. nomura , l. e. cross , s. j. jang , r. e. newnham , j. appl . phys . * 51 * , 1142 ( 1980 ) .
d. viehland , s. j. jang , l. e. cross , m. wuttig , phys .
b * 46 * , 8003 ( 1992 ) .
z. kutnjak , r. pirc , and r. blinc , appl .
lett . * 80 * , 3162 ( 2002 ) .
r. pirc , r. blinc , z. kutnjak , phys .
b * 65 * , 214101 ( 2002 ) .
v. bobnar , b. vodopivec , a. levstik , z .- y .
cheng , q. m. zhang , phys .
b * 67 * , 094205 ( 2003 ) .
r. pirc , r. blinc , v. s. vikhnin , phys .
b * 74 * , 054108 ( 2006 ) .
r. pirc , r. blinc , j. f. scott , phys .
b * 79 * , 214114 ( 2009 )
. a. amin , r. e. newnham , l. e. cross , phys .
b * 34 * , 1595 ( 1986 ) .
j. h. qiu , q. jiang , phys .
a * 372 * , 7191 ( 2008 ) . z. kutnjak , b. vodopivec , r. blinc , phys . rev .
b * 77 * , 054102 ( 2008 ) . m. lines , a. glass , _ principles and applications of ferroelectrics and related materials _
( clarendon press , oxford , 1977 ) , p. 148 . s. g. lu and q. m. zhang , adv .
* 21 * , 1983 ( 2009 ) .
b. neese , s. g. lu , b. chu , q. m. zhang , appl . phys . lett . *
94 * , 042910 ( 2009 ) .
z. kutnjak , c. filipi .
a. levstik , j. non - cryst .
solids * 305 * , 398 ( 2002 ) .
s. sugiyama , a. takagi , k. tsuzuki , jap .
* 30 * , 2170 ( 1991 ) .
s. kamba , v. bovtun , j. petzelt , i. rychetsky , r. mizaras , a. brilingas , j. banys , j. grigas , m. kosec , j. phys . :
matter * 12 * , 497 ( 2000 ) .
s. g. lu , b. roi , q. m. zhang , z. kutnjak , x. li , e. furman , l. j. gorny , m. lin , b. mali , m. kosec , r. blinc , r. pirc , j. appl .
( in the press ) . s. g. lu , b. roi , q. m. zhang , z. kutnjak , x. li , r. pirc , l. j. gorny , m. lin , j. appl . phys .
( in the press ) . q. m. zhang , z .- y .
cheng , v. bharti , t .- b .
xu , h. xu , t. x. mai , s. j. gross , in _
smart structures and materials 2000 : electroactive polymer actuators and devices ( eapad ) _ , edited by y. bar - cohen , [ proc .
spie * 3987 * , 34 ( 2000 ) ] . | the electrocaloric effect ( ece ) in normal and relaxor ferroelectrics is investigated in the framework of a thermodynamic approach based on the maxwell relation and a landau - type free energy model .
the static dielectric response of relaxors is described by the spherical random bond random field model , yielding the first landau coefficient @xmath0 , which differs from the usual expression for ferroelectrics .
the fourth - order coefficient @xmath1 is treated as a phenomenological parameter , which is either positive or negative due to the anisotropy of the stress - mediated coupling between the polar nanoregions . when @xmath2 , the maximum ece in a relaxor is predicted near the critical point in the temperature - field phase diagram ,
whereas in a ferroelectric it occurs at the first order phase transition .
the theoretical upper bound on the ece temperature change is estimated from the values of saturated polarization , effective curie constant , and specific heat of the material . |
observations of our and outer galaxies allowed to identify various galactic populations : the halo , the thick disk , the thin disk , and the bulge . a model for the evolution of galaxies should explain the origin and properties of these populations , as well as other basic observations like e.g. the relation of hubble types with local environment , in a unifying scheme .
current models for galaxy formation broadly divide into two families : those considering a dissipational collapse ( eggen et al . 1962 ; larson 1974 ) ; and those which consider galaxies as the results of the accretion of individual fragments undergoing ( some ) indipendent chemical and dynamical evolution ( toomre & toomre 1972 ; searle & zinn 1978 ) .
the transition between the halo and disk phases is continuous in smooth _ dissipational collapse _ models , while disk formation is a secondary mechanism in _ accretion _ ones .
separation between these two classes of models may be quite artificial : in fact various properties of galaxies , like e.g. the light distribution of ellipticals , are well reproduced by inhomogenous collapses leading to some kind of violent relaxation ( lynden - bell 1967 ) ; on the other side , simulations based on cosmologies dominated by cold dark matter predict that in high density regions galaxies form hierchically by merging of smaller subunits , while in low density ones they form more gradually by infall of diffuse matter ( frenk et al .
1985 ) . within this framework
, the mechanisms of formation of our own galaxy ( the milky way ) could be determined by examining fossil remnants of the early phases represented by the old ( and often metal - poor ) stars .
the interpretation of the large amount of data gathered in the last years on dynamics and metallicities ( as defined by the most easily observed element , fe ) of field stars is however still controversial , and while e.g. some authors consider the thick disk and the bulge ( gilmore et al .
1989 ) as distinct galactic components , others ( norris 1993 ) think they are simply the outer ( and oldest ) part of the disk and central part of the halo respectively .
scenarios of galactic evolution including a hiatus between the formation of the halo and of a secondary disk ( ostriker & thuan 1975 ) , that were introduced to justify the rarity of metal - poor stars in the solar neighbourhood ( schmidt 1963 ) , are widely applied e.g. to explain the hot , metal - rich intergalactic gas seen in clusters ( berman & suchkov 1991 ) ; however , up to now the observational basis for this hiatus ( based on the age gap between open and globular clusters : demarque et al .
1992 , carraro et al .
1999 ; the white dwarf cooling sequence : wonget et al .
1987 , knox et al .
1999 ; and the th / nd ratio nucleo - chronometer for disk and halo stars : malaney & fowler 1989 , cowan et al .
1999 ) are rather weak and controversial .
relative abundances of o and fe in stars of different overall metal abundance provide further constraints to the early evolution of the halo and the formation of the galactic disk ( wheeler et al .
o is the main product of hydrostatic he - burning : hence the ejecta of core - collapse supernovae ( sne ) resulting from the evolution of massive stars , usually identified with type ii sne , are expected to be very rich in o ( woosley & weaver 1986 ; thielemann et al 1990 ) . on the other side , while a fraction of the fe presently observed in the interstellar medium was synthesized in massive stars ( thielemann et al 1990 ) , a large fraction of it was likely produced in explosive burning under degenerate conditions in type ia sne ( nomoto et al . 1984 ) .
typical lifetimes of the progenitors of type ia sne ( @xmath4 yr ) are much longer than those of the progenitors of type ii sne ( @xmath5 yr ) , and they are actually longer than , or of the same order of , the free fall time in the galaxy ( @xmath6 yr ) ; for these reasons the production of the bulk of fe is expected to be delayed with respect to that of o ( matteucci & greggio 1986 ) . a clear break in the run of o abundances with overall metallicity [ fe / h]=log@xmath7(x)@xmath8log@xmath7(x)@xmath9 for any abundance ratio x. ] should signal the onset of the contribution by type ia sne , and the location of this break provides an independent estimate for the timescale of star formation during the early stages of galactic evolution ( matteucci & franois 1992 : hereinafter mf ) .
it should be added that other @xmath10elements ( like mg , si , and ca ) are expected to behave similarly to o , although for si and ca a small contribution by type ia sne is also expected . in the last years
various investigations have been devoted to the study of the run of [ o / fe ] with [ fe / h ] in halo and disk stars ( wheeler et al .
1989 , king 1994 , nissen & schuster 1997 , fuhrman 1998 , 1999 , israelian et al .
1998 , boesgaard et al . 1999 ) .
however , a variety of basic questions still lacks of a clearcut answer .
the [ o / fe ] ratio in the halo and the location of the change of slope in the run [ o / fe ] vs [ fe / h ] have been addressed by king ( 1994 ) , who concluded that this change may occur at any value in the range @xmath11[fe / h]@xmath12 , corresponding to timescales for the halo formation between @xmath13 and @xmath14 yr ( mf ) ; this range is large enough to accomodate both a fast , ordered dissipational collapse ( eggen et al .
1962 ) , or a much slower , accretion scenario ( searle & zinn 1978 ) .
edvardsson et al .
( 1993 ) studied the [ o / fe ] run in disk stars ; they suggested that this ratio is constant for [ fe / h]@xmath15 , and argued that the spread in [ fe / h ] values at any age is an evidence for infall of metal - poor material .
even less understood is the [ o / fe ] run at intermediate metallicities , corresponding to the thick disk phase ( gilmore et al .
1989 ; nissen & schuster 1997 ) .
the main concerns in previous investigations on o abundances relate ( i ) to the paucity of samples of significant sizes studied in a homogeneous way , and then to the possible existence of systematic offsets between different sets of data ; and ( ii ) to the discrepancy between o abundances determined using high excitation permitted and low excitation forbidden lines ( the first usually observed in dwarfs , the second in giants ) .
most of this discrepancy can be removed by adopting higher temperatures in the analysis of dwarfs ( king 1993 ) ; furthermore , the effects of departures from the local thermodynamic equilibrium ( lte ) assumption when considering the formation of high excitation permitted o i lines should also be considered , in order to provide abundances at the level of accuracy required for the present purposes .
recent results based on the oh band at the extreme uv edge of ground - based observations have further complicated this issue , suggesting the presence of a quite strong slope in the [ fe / o ] run with [ fe / h ] amongst metal - poor stars ( israelian et al .
1998 , boesgaard et al . 1999 ) both a high temperature scale , and consideration of departures from lte were included in the new homogeneous determinations of abundances of light elements and fe for a large sample of stars we are presenting in this series of papers .
a new , hopefully improved temperature scale based on irfm temperatures for population i stars and the new model atmospheres by kurucz ( 1993 , cd - rom 13 ) was obtained in gratton et al .
( 1996a , paper i ) .
an extensive discussion of the effects of departures from the assumption of lte in line formation in the stellar atmospheres was given in gratton et al .
( 1999 , paper ii ) ; in that discussion , we exploited an empirical calibration of the poorly known cross sections for collisions with h i atoms drawn from a parallel analysis of the spectra of rr lyare variables at minimum light , where non - lte effects are much larger than in the stars here considered ( clementini et al .
our final abundances for about 300 stars were presented and discussed in carretta et al .
( 2000 , paper iii ) .
we found that most discrepancies present in earlier works have been removed , our results showing a high degree of internal consistency , at least for stars with effective temperature @xmath16 k , although our results are still not easy to be reconciled with the o abundances from the uv oh bands .
however , in this paper we will show that once combined with stellar kinematics and compared with models of galactic chemical evolution , our results allow to throw new light into some of above mentioned questions : we find that in the framework of homogeneous models , the collapse of the halo and the formation of the thick disk occurred on a short timescale ( a few @xmath17 yr ) , although star formation in these environments likely lasted for @xmath18 gyr ; and that there was a sudden decrease in star formation between the thick and thin disk phases , which are then clearly distinct galactic components .
it is worth noticing that a decrease in the star formation rate between the formation of the spheroidal components and the disc in disc dominated galaxies , had already been suggested by larson ( larson et al .
he suggested that such a decrease in the star formation could be due either to the action of tidal forces inhibiting star formation during the later stages of the collapse or to a two - phase structure of the gas , with dense clouds forming rapidly in a spheroidal component and less dense intercloud gas not forming stars and settling to a disc .
we argue that our results fit in a scenario in which both collapse and accretion were important in the formation of the milky way , these mechanisms having similar timescales ; the relative weights of the two contributions in other galaxies might explain the hubble sequence .
an early , short presentation of the content of this paper was given as a talk at the conference on formation of the galactic halo ( gratton et al .
1996b ) . here
we give a more complete presentation of our arguments . in the meanwhile , fuhrmann ( 1998 )
reached quite similar conclusions , based on an independent careful analysis of a smaller smaple of nearby stars .
our results for fe , o and mg are displayed in fig .
[ fig:1 ] and [ fig:2 ] . to further improve homogeneity , we only plotted data for dwarfs ( edvardsson et al .
1993 ; tomkin et al .
1992 ; nissen & edvardsson 1992 ; zhao & magain 1990 ) ; all these stars have @xmath16 k. inclusion of cooler and/or lower gravity stars in our sample does not change the present discussion . the original data ( equivalent widths ) used in these papers
are very consistent with each other , and were measured on high signal - to - noise , high resolution spectra ; typical errors of individual equivalent widths determined from independent estimates for the same stars , are @xmath19 m , yielding errors of @xmath20 dex in abundances derived from individual lines .
possible errors in the atmospheric parameters required in the analysis of individual stars ( @xmath21 k in the effective temperatures ; @xmath22 dex in the surface gravity ; @xmath23 in the microturbulent velocity ) cause uncertainties of @xmath24 dex in [ fe / h ] , @xmath25 dex in [ o / h ] , and @xmath25 dex in [ fe / o ] ; these are internal errors .
analogous values for mg are @xmath26 dex in [ mg / h ] , and @xmath27 dex in [ fe / mg ] .
systematic errors should be small since the present analysis is differential with respect to the sun . in total , we determined [ o / fe ] ratios for about 160 stars and [ mg / fe ] for 197 stars . following wheeler et al .
( 1989 ) we have chosen to use o and mg as reference elements , because they are almost uniquely produced in massive stars .
the scatter of data for individual stars is small and compatible with observational errors for stars with [ o / h]@xmath28 and [ mg / h]@xmath29 , although we can not exclude a gentle ( @xmath30 ) slope in the run of [ fe / o ] with [ o / h ] ; the scatter we obtain for more metal - poor stars ( 0.10 dex in both [ fe / o ] and [ fe / mg ] ratios ; for o the peculiar n - rich dwarf hd 74000 was omitted ) may indicate that errors are larger in this abundance range ( perhaps due to the apparent star faintness ) . however , the composition of the interstellar matter could have been slightly inhomogeneous during the early phases of galactic evolution , as suggested by the large spread in the abundances of n - capture elements found by mcwilliam et al .
( 1995 ) for [ fe / h]@xmath31 ( i.e. [ o / h@xmath32 ) , and the scatter in the fe / o ratios amongst halo stars found in the careful analysis by nissen & schuster ( 1997 ) .
the upper limit for intrinsic star - to - star variations derived from the spread in our data is 0.07 dex , once observational errors are taken into account .
[ fig:1]a indicates that the run of [ fe / o ] with [ o / h ] is quite flat from [ o / h][email protected] , [ fe / o][email protected] ( these are the most metal - poor stars in our sample ) to [ o / h][email protected] , [ fe / o][email protected] , where it is possible to locate ( with an uncertainty of about @xmath34 dex ) the change of slope due to the onset of the contribution to nucleosynthesis by the bulk of type ia sne . a similar result is provided by fig . [ fig:2 ] for the run of [ fe / mg ] with [ fe / h ] .
we remark that the values of [ o / h ] and [ mg / h ] at which the changes of slope occur is large ( [ o / h]@xmath35[mg / h]@xmath36 ) .
a direct comparison with existing galactic evolution model is possible for o , for which the predictions of mf models are available .
we overposed on fig .
[ fig:1]a lines representing the predictions given by models 1 and 2 of mf ; these models were computed with an @xmath37folding time of 1 gyr for the infall ( halo collapse ) , and two different laws of star formation : the times required for [ fe / h ] to raise at [ fe / h]=@xmath38 ( roughly corresponding to the halo - disk transitions ) are @xmath39 and @xmath13 yr respectively .
both models were arbitrarily scaled to match the [ fe / o ] ratio for halo stars ( [ fe / h]@xmath40 ) .
these scalings only imply small changes in the adopted yields of fe from type ii sne , which on turn depend on the cut - off mass for the remnants for core - collapse sne , an ill - defined quantity at present ( timmes et al .
while these scalings do not affect our conclusions , they help to see the main features we like to point out . insofar metal - poor stars are considered ( [ fe / h]@xmath41 ) , the r.m.s .
values of the residuals of points for individual stars around the lines representing the mf models are 0.158 and 0.127 dex for model 1 and 2 respectively : data for metal - poor stars are then better represented by model 2 , which considers the production of fe from massive stars alone during this phase ( the shallow slope of [ fe / o ] with [ o / h ] in this model is due to the dependence of the o / fe abundance ratio in the ejecta of type ii sne with progenitor mass ) ; while model 1 is in clear disagreement with observations . the conclusion of this ( and other comparisons not shown in fig . [ fig:1 ] ) , is that the raise of o abundances up to [ o / h]=@xmath42 occurred on a timescale which is not much longer than the lifetime of type ia sne .
this timescale will be better quantified later .
while model 2 is able to better reproduce observations at low metallicities , it fails in the metal - rich range .
in fact , @xmath43-tests show that a linear dependence of [ fe / o ] on [ o / h ] ( roughly similar to that given by both model 1 and 2 ) is not a good representation of the observed [ fe / o ] s for [ o / h]@xmath28 , since the scatter of [ fe / o ] values for [ o / h]@xmath44 is much larger than observational errors .
a very similar result might be obtained for mg .
the simplest interpretation is that the fe content suddenly increased at [ o / h]@xmath35[mg / h]@xmath36 ; the implication is that there was a phase in which the production of o and mg ( i.e. the formation of type ii sne ) felt down to small values , leaving only the fe producers in activity .
this obviously means a sudden decrease in the formation of massive stars and , since we are moving in the framework of a constant initial mass function ( mf ) , in the star formation _ tout court_. we found this same feature when considering si and ca abundances rather than o ones (
these abundances are not discussed in paper iii , but they display trends similar to those found for o and mg : gratton et al .
, in preparation ) .
the synthesis of these elements is also likely related to massive stars ( thielemann et al .
1990 ) .
the phase of low star formation must have lasted enough to allow the explosion of a large fraction of halo and thick disk type ia sne , since [ o / h ] and [ mg / h ] start increasing again from values of [ fe / o ] and [ fe / mg ] @xmath1 dex higher than that achieved in the previous phase .
the simultaneous increase of both [ o / h ] and [ fe / o ] in stars with [ o / h]@xmath28 , [ o / fe]@xmath45 seems to require that both type ia and type ii sne contribute to the chemical enrichment during this phase .
the regression line through data in this region is : @xmath46}=(0.34\pm 0.07){\rm [ o / h ] } + ( 0.04\pm 0.10),\ ] ] based on 71 stars .
however , results for mg in the metal - rich regime are quite different ; the regression line through data in this region is : @xmath47}=(-0.055\pm 0.025){\rm [ mg / h ] } - ( 0.089\pm 0.076),\ ] ] based on 164 stars .
if real , this result indicates that mg abundances do not increase at the same rate as o ones , suggesting that the o / mg ratio in the ejecta of type ii sne is a function metal abundance ; this could be understood if severe mass loss reduce production of mg in massive , metal rich stars .
lrclrclrcl parameter & & & + @xmath48 ( ) & 21 & 48 & 100 & 21 & 144 & 52 & 164 & 206 & 24 + @xmath49 ( kpc ) & 21 & 2.1 & 2.3 & 21 & 0.6 & 1.1 & 164 & 0.18 & 0.20 + @xmath50 & 21 & 0.69 & 0.28 & 21 & 0.38 & 0.19 & 164 & 0.14 & 0.07 + @xmath51 ( gyr ) & 13 & 1.16 & 0.08 & 21 & 1.12 & 0.06 & 164 & 0.69 & 0.23 + @xmath52fe / h@xmath53 $ ] & 29&@xmath331.68 & 0.41 & 21 & @xmath330.63 & 0.15 & 164 & @xmath330.19 & 0.26 + @xmath52o / h@xmath53 $ ] & 29&@xmath331.23 & 0.40 & 19 & @xmath330.29 & 0.12 & 68 & @xmath330.18 & 0.18 + @xmath52fe / o@xmath53^a$ ] & 28&@xmath330.46 & 0.10 & 19 & @xmath330.36 & 0.06 & 67 & @xmath330.01 & 0.11 + @xmath52mg / h@xmath53 $ ] & 22&@xmath331.42 & 0.41 & 21 & @xmath330.29 & 0.13 & 164 & @xmath330.10 & 0.24 + @xmath52fe / mg@xmath53 $ ] & 22&@xmath330.30 & 0.10 & 21 & @xmath330.32 & 0.05 & 164 & @xmath330.08 & 0.08 +
a further basic step can be done by combining information provided by the ( [ o / h ] vs [ fe / o ] ) and ( [ mg / h ] vs [ fe / mg ] ) diagrams with those obtained from the kinematics of stars in our sample . a caveat should be done here , since stars analyzed in the present paper were collected from various sources , and the ( sometimes not well defined ) selection criteria introduce important biases : the distribution of stars with o and mg abundances in our sample is very different from that obtained from a volume limited sample in the solar neighbourhood , and high velocity stars are likely overrepresented amongst the most metal poor ones . in this section
we will use this comparison simply to identify stellar populations defined on chemical grounds with those defined from dynamics .
the basic data for this comparison are the o , mg and fe abundances drawn from our analysis , and dynamical data determined by edvardsson et al .
( 1993 ) , for stars with [ o / h]@xmath28 ; for stars more metal - poor than this limit in the tomkin et al .
( 1992 ) and zhao & magain ( 1990 ) samples , similar data were obtained using parallaxes and proper motion from hipparcos ( perryman et al . 1997 ) , radial velocities from the literature ( carney et al .
1994 whenever possible ; else data were taken from the simbad database ) , and the code for galactic orbit calculations by aarseth as modified by carraro ( 1994 ) .
both these samples only include dwarfs in the solar neighbourhood ; the edvardsson et al .
sample is essentially a magnitude - limited sample ( although the magnitude limit is a - known - function of metallicity ) , with no kinematical bias ; while tomkin et al .
( 1992 ) and zhao & magain ( 1990 ) selected high proper motion stars , so that a bias toward high velocity stars is certainly present amongst the most metal - poor stars .
the location of the stars with known dynamical data in the ( [ o / h ] vs [ fe / o ] ) diagram is shown in fig .
[ fig:3 ] , where we plotted with different symbols stars in different regions of the diagram : metal - poor stars ( [ o / h]@xmath41 : group a ) ; fe - poor o - rich stars ( [ o / h]@xmath28 , [ fe / o]@xmath54 : group b ) ; and fe - rich o - rich stars ( [ o / h]@xmath28 , [ fe / o]@xmath45 : group c ) . the analogous diagram for mg is shown in fig .
[ fig:4 ] , but in this case group a are stars with [ mg / h]@xmath55 ; group b are stars with @xmath56[mg / h]@xmath57and [ mg / fe]@xmath54 ; and group c are the remaining metal rich stars .
a star - by - star comparison shows that stars are attributed to the same groups when using o and mg . in the four panels of fig .
[ fig:5 ] and [ fig:6 ] [ fe / o ] and [ fe / mg ] ratios for these stars are plotted against the rotational velocity around the galactic center , the orbital eccentricity , the maximum height of the orbit over the galactic plane @xmath58 , and the age @xmath59 respectively .
the age values are averages of two independent values : * those determined by edvardsson et al .
( 1993 ) and schuster & nissen ( 1989 ) derived using an homogenous procedure from the @xmath60 diagram , calibrated against age using standard isochrones ( _ i.e. _ not o - enhanced ) by vandenberg and bell ( 1985 ) * and those derived using absolute magnitudes from hipparcos parallaxes ( perryman et al .
1997 ) , interpolating within the grid of isochrones by padua group ( girardi et al .
2000 ) ; in this case the enhancement of o and the other @xmath10elements was considered by modifying the fe abundance according the procedure suggested by straniero et al .
( 1997 ) the two age scales are somewhat different ; they were homogeneized to an a common ( arbitrary ) scale before averaging them .
absolute ages are affected by various uncertainties related to stellar models , while the relative ranking should be quite reliable , as indicated by the good agreement existing between the two set of age determinations ( see fig .. [ fig:10 ] ) . since basic data for the two age estimates are independent each other , internal uncertainties on the ages may be obtained by the r.m.s scatter of the differences ; in this way , we estimate that typical error bars in log(age ) for individual stars are @xmath61 dex ( i.e. @xmath62% ) .
the internal scatter we get for group a and b is of this same order , indicating that small if any internal scatter exists for these groups .
the average properties for the three chemical groups are listed in table [ tab:1 ] .
the main conclusions we may draw from fig . 1 - 6 and table [ tab:1 ] are : 1 .
though some bias may be present , present data support the identification of group a as the halo , of group b as the thick disk ( see e.g. robin et al .
1996 , norris , 1999 , buser et al .
1999 ) , and of group c as the thin disk .
2 . in the framework of homogeneous models for the galactic evolution ,
the formation of stars in the halo and the thick disk was fast ( a few @xmath17 yr ) , i.e. shorter than the typical timescale of evolution for the progenitors of type ia sne .
3 . given the lack of a clear selection criterion amongst metal - poor stars , our data can not be used to draw any conclusion about clear breaks between the halo and thick disk populations ; the age difference must be small ( i.e. the two populations are virtually coeval ) , else the [ o / fe ] and [ mg / fe ] ratios of the two populations should be different ; the ages derived for both groups are well in excess of 10 gyr . anyway , this discontinuity is supported by other studies ( see norris 1993 ) . as noticed by wyse and gilmore ( 1992 ) ,
the dynamical properties of the thick disk are clearly indicative of a flattened intermediate population supported by rotation ( these characteristics are drawn from the edvardsson et al .
sample , which are not affected by kinematical biases ) ; while the halo space distribution is much wider , and the system appears to be supported by velocity dispersion rather than by rotation . in our analysis ,
this last result is likely enhanced by the selection biases in the carney et al . sample .
further strong supports to the present conclusion that the thick disk is a homogeneous , chemically old population is given by the results by fuhrmann ( 1998 , 1999 ) , and nissen & schuster ( 1997 ) : these authors found that thick disk stars have low [ fe / o ] and [ fe / mg ] ratios ( equal or even lower than those found for halo stars of similar metallicity ) , with very small intrinsic scatter .
4 . there was a sudden decrease in star formation during the transition between the thick and thin disk phases , which are then clearly distinct at least from the chemical point of view .
the phase of low star formation must have lasted at least 1 gyr in order to allow for the explosion of the bulk of type ia sne ; however , the hiatus was not longer than 3 gyr , else there would be obvious break in the [ fe / o ] and [ fe / mg ] vs age diagrams .
we remark that the age difference between the thick disk and the oldest thin disk stars could have been overestimated in fig .
[ fig:5]d and [ fig:6]d , since these ages were derived assuming solar ratios of o and mg to fe .
again , results from fuhrmann ( 1998 , 1999 ) and nissen & schuster ( 1997 ) well support this picture .
both the [ fe / o ] and [ fe / mg ] ratios for thin disk stars increase with time : in fact there is a significative correlation of these ratios with stellar ages ; the linear regression line are : @xmath46 } = -(0.19\pm 0.06 ) \log t
+ ( 0.12\pm 0.11),\ ] ] based on 68 stars for o , and : @xmath47 } = -(0.085\pm 0.026 ) \log t - ( 0.024\pm 0.075),\ ] ] based on 164 stars for mg .
the persson correlation coefficients are 0.35 and 0.25 respectively : the probability of getting such high correlation coefficients by chance are @xmath63 and @xmath64 .
the scatter around the mean lines are compatible with the error bars in ages , [ fe / o ] and [ fe / mg ] values .
if the raise of [ fe / o ] at constant [ o / h ] is due to a sudden decrease in star formation during the transition between the thick and thin disk phases , then there should be a corresponding gap at [ fe / o]@xmath65 in the distribution of stars with [ fe / o ] . a cumulative diagram of the distribution of [ fe / o ] and [ fe / mg ] values for stars in the edvardsson et al .
sample ( fig .
[ fig:7 ] and [ fig:8 ] ) indeed supports this inference . in the rest of this section
, we will evaluate the statistical significance of this gap for o ; similar ( though slightly less significative ) results were obtained for mg .
to this purpose , we performed tests on the distribution of [ fe / o ] values in that sample , which consists of the brightest _ bona fide _ single stars with @xmath66@xmath67 k , having an absolute magnitude from 0.4 to 2 mag smaller than that of the zero age main sequence at the same @xmath68 colour , in equally spaced bins in [ fe / h ] .
generally , populations of each bin were kept similar , but the two most metal - poor bins ( [ fe / h]@xmath55 ) have populations about half those of the others , due to the magnitude limit of the survey .
however , the 85 stars for which o abundances are available are nearly uniformly distributed with [ fe / h ] , since we have these data for 42% of the stars with [ fe / h]@xmath29 , and for 79% of the stars more metal - poor than this limit .
a @xmath43-test confirms that the distribution of stars having [ fe / o ] values can not be distinguished from a uniform distribution in [ fe / h ] ; anyway , we repeated the following analysis using both a uniform distribution , and distributions obtained by summing a random gaussian distributed term ( representing observational errors ) to the observed [ fe / h ] s .
the results are very similar .
we first tested the hypothesis that the [ fe / o ] values are distributed uniformly , as expected if a linear dependence of [ fe / o ] on [ fe / h ] holds .
we found that this hypothesis can be rejected at a high level of confidence , using both a @xmath43-test and a serie of monte carlo simulations .
each monte carlo simulation included 10,000 extractions of 85 [ fe / h ] values ( using both early described approaches ) ; individual [ fe / o ] s were the sum of the values deduced from these [ fe / h ] s using the [ fe / o]-[fe / h ] law , and of a random gaussian distributed term representing the spread due to observational errors and to the intrinsic star - to - star variations .
the simulations were repeated for three values of the standard deviation for this term : 0.05 , 0.07 , and 0.10 dex ; the intermediate value was deduced from our error analysis , and it is equal to the standard deviation from the mean value for thick disk stars ; the last one is consistent with the residuals around the best fit relation for thin disk stars . most of the deviation from a uniform distribution in [ fe / o ] is due to the excess of stars with [ fe / o]@xmath69 ( corresponding to the thick disk population ) , and to the lack of stars with @xmath70[fe / o]@xmath71 ( the expected location of the gap ) .
it can be noticed that a quadratic dependence of [ fe / o ] on [ fe / h ] gives a better fit to observational data .
we then repeated the monte carlo simulations with a similar [ fe / o]-[fe / h ] law .
we found that the probability that the very low number of stars in a bin of 0.1 dex centered at [ fe / o]=@xmath72 is due to chance is about 0.024 for a uniform distribution with [ fe / h ] , and 0.013 for a distribution function equal to the frequency distribution ( the exact values depend on the assumed observational errors ; the above mentioned values refer to the less significant cases obtained assuming that present [ fe / o ] values have errors of 0.10 dex ) . a good significance ( chance probability @xmath73 ) is achieved for 0.1 dex bins centered over the range @xmath74[o / fe]@xmath54
; this suggests that the gap is broader than 0.1 dex .
we conclude that present available data support the hypothesis that there is a gap in the distribution of [ fe / o ] as expected from a sudden decrease in star formation during the transition from the thick to thin disk phases ; however we think this test should be repeated using a larger and properly selected sample .
the overall run of [ fe / o ] with [ fe / h ] is certainly related to the delayed fe synthesis : however , the interplay between star formation rate , progenitor lifetime , and infall can only be cleared out by detailed modelling of galactic chemical evolution .
unfortunately , various aspects of star formation and evolution are still not well understood , so that these models have several free parameters .
furthermore , a fully appropriate comparison between abundances in metal - poor stars and models of galactic evolution should require models which include consistent and detailed treatment of both chemistry and dynamics .
however , this is still beyond current computational capabilities , and at present dynamics has to be introduced into chemical evolution models parametrically , increasing even more the number of free perameters . in order to explore allowed ranges for some of the relevant quantities
, we then computed a large number of single - zone models ( with infall of original unprocessed material ) appropriate for the solar neighborhood ; the code we used is an improved version of that of mf .
the main advantage of this approach is the reduction in the number of free - parameters , although results of our comparisons strictly apply only to homogeneous collapse models , and not e.g. to accretion scenarios .
we will see in the next section that even with these limitations , results of our comparisons are enough to suggest that the best scenario required to explain the star formation history in the solar neighborhood should include both dissipational collapse and accretion . when comparing the predictions of our models with observations , we considered a wide range of constraints , including present abundances in the interstellar medium , current gas density , star formation and sn rates , the distribution of abundances among g - dwarfs , the run of [ fe / h ] with age , imf , etc ( matteucci 1991 ) , and retained only those models which give an overall match to all of them .
we find that in the best cases , r.m.s .
values of the residuals of [ fe / o ] s for individual stars around lines representing our models are @xmath75 dex for halo and thick disk stars , and @xmath76 dex for thin disk stars , in good agreement with observational errors , as shown by monte carlo simulations where both errors in [ fe / o ] and [ o / h ] were taken into account ( similar simulations were also used to reject those models providing poor fits ) .
predictions about the [ fe / o ] ratios provided by some of these models are plotted in fig .
[ fig:3]b and c , overposed to the same observational points shown in fig .
[ fig:3]a . on the whole
, we found that only models having two distinct infall episodes , the first connected to the halo ( and thick disk ) , and the second to the thin disk , fit observational data ( see also chiappini et al .
1997 for similar models ) ; while models with a single infall episode do not give a good match even when star formation law was changed with time , due to either large accumulation of gas during the phase corresponding to the raise of the [ fe / o ] ratio , or lack of gas at present ( depending on the rate of decay ) .
decay of the halo infall rate should be fast in order to reproduce the almost flat run of [ fe / o ] with [ o / h ] in the halo and thick disk : the @xmath37folding time is shorter than or equal to the adopted time delay for type ia sne , i.e. @xmath77 gyr .
the halo infall should have contributed no more than 20 - 30% of the present density in the solar neighborhood ( and the fraction in stars should be about half that ) .
the decay of the thin disk infall should be rather slow ( @xmath78 gyr , which implies a present infall rate @xmath79pc@xmath80gyr@xmath81 ) , else there would be not enough gas at present , the present interstellar medium should be too metal rich , the observed roughly linear run of [ fe / o ] with [ o / h ] in the thin disk would not be reproduced , and there would be an excess of moderately metal - poor stars ( [ fe / h]@xmath82 ) .
this value for the present infall rate is larger than the upper limit deduced from observations ( @xmath83pc@xmath80gyr@xmath81 : mirabel 1989 ) .
this inconsistency might be removed by either assuming that star formation law changes with time ( the efficiency should then be larger in the halo than in the thin disk ) , or that the initial mass function ( imf ) is flatter at large masses ( slope @xmath84 ) than that adopted in most of our models ( scalo 1986 ) ( note that in order to avoid o overproduction with this flatter imf , an upper limit of @xmath85 has to be adopted for type ii sne ) .
both these alternatives do not contradict basic costraints .
models where both infall episodes start at high values at the beginning and then decay ( fig .
[ fig:3]b ) , as well as models where the disk infall is delayed with respect to the halo one ( fig .
[ fig:3]c ) , fits data quite well .
it should be noticed that in the first class of models the star formation must be arbitrarily lowered at the thick - thin disk transition ( that should occur @xmath0 gyr after beginning ) , and there is no hiatus : the raise of the [ fe / o ] ratio at constant [ o / h ] is due to fe production by the large number of type ia sne from the halo - thick disk , while in the meantime only small amounts of o are produced , due to the decreased star formation rate , which barely compensate for the dilution by metal - poor infalling material ( the [ o / h ] ratio starts increasing again when most gas is consumed and the infall rate decays down to small values ) .
the less appealing aspect of this class of models is the large fraction of metal - poor stars , that only can be reconciled with the small number of metal - poor g - dwarfs observed in the solar neighborhood by assuming that the scale height is a strong function of metallicity . on the other side ,
the second class of models naturally yield to a hiatus if a threshold ( surface ) gas density for star formation is adopted ( kennicutt 1989 ) .
a threshold gas density might be expected if the processes involved in star formation are self - regulating , and for a low enough surface gas density the feedback mechanisms that regulate the star formation rate break down ( gallagher & hunter 1984 ; elmegreen 1992 ; burkert et al
observationally , an approximate value of @xmath86 atoms @xmath87pc@xmath80 has been proposed for this threshold from observation of irregular and hi - rich spiral galaxies ( skillman 1986 ; van der hulst et al .
1987 ) ; this threshold value is close to the current gas density in the solar neighbourhood ( rana & basu 1991 ; kujiken & gilmore 1989 ) , as expected in a self - regulating mechanism . with the above
mentioned infall rates , a star formation rate with threshold produces a hiatus in star formation at the end of the halo and thick disk phase ; in our models , star formation in this early phase lasts for 1.5 - 2.5 gyr ( inversely depending on the adopted threshold density ) , and for given yields the final [ o / h ] value depends on the ratio between the halo infall and the threshold density , because the early chemical evolution is essentially that of a closed box ( phillips et al 1990 ) . in these models ,
type ia sne begin to contribute to nucleosynthesis during the last phases of halo and thick disk evolution , raising the [ fe / o ] ratio ; this contribution continues during the hiatus , which must be at least as long as time delay of sn ia ( @xmath88 gyr ) , in order to raise the [ fe / o ] by at least 0.13 dex , that we estimate as the lower limit given by observations .
however , the upper limit for hiatus duration can not be determined from the [ fe / o ] run due to a saturation effect .
part of the gas might be lost at the thick - thin disk transition ( due e.g. to a galactic wind induced by sn explosions : silk 1985 ) ; if this occurs , this metal - rich gas will be replaced by more metal - poor infalling gas , and the starting value of [ o / h ] in the thin disk is lower than the maximum achieved during the thick disk evolution .
models where up to 75% of the gas is lost at this epoch fit data quite well , but they predict the existence of some very o - poor , fe - rich thin disk stars , which are not present in the observed samples .
we think that the overposition of the thick and thin disk sequences over some range in [ o / h ] can be better explained by noting that stars currently in the solar neighborhood likely formed over a range of galactocentric distances ( see franois and matteucci 1993 ) , where the halo surface density and the largest [ o / h ] values achieved during the halo and thick disk evolution were different .
note that this is also the explanation favoured by other authors ( e.g. edvardsson et al .
what our data tells us about scenarios of galaxy formation ?
the fast rate of star formation for the halo and thick disk ( @xmath89 yr ) is consistent with a smooth dissipational collapse ( eggen et al . 1962 ; larson 1974 ) .
however , in this case the transition from the thick to thin disk phases should be continuous : some heating mechanisms causing the observed sudden decrease in star formation during this transition is missing and should then be introduced .
possible candidates are a high sn rate ( silk 1985 ; berman & suchkov 1991 ) , and merging of smaller galaxies with our own ; these mechanisms might also be acting simultaneously .
our tests with models with a single infall episode ( but variable star formation rate ) suggest that a simple heating as that produced by sne can not reproduce the whole spectrum of observations . also , the discontinuity in specific angular momentum between halo and thick disk suggests that there is not a smooth transitions between these two populations ( wyse and gilmore 1992 ) . on the other side , merging with gas poor satellite(s ) having a mass larger than a few hundredths the disk mass may heat ( and destroys ) any pre - existing stellar thin disk ( quinn et al .
1993 , walker et al . 1996 ) and create a thick disk , although this result is not obtained in all simulations ( see e.g. huang & carlberg 1997 ) , so that it seems to depend on the initial conditions as well as on the properties of the satellite : merging of satellites on prograde orbits more likely produce thick disks , while those on retrograde orbits mainly produce disk tilts ( velazquez and white 1999 ) .
the mass range is fixed by the amount of kinetic energy to be injected into the disk : merging with a companion of comparable mass would have transformed the milky way into an elliptical , while merging with a very small satellite has only minor effects . if the disk or the satellite contained gas ( as indicated by chemical evolution models ) , merging was likely accompanied by a burst in star formation ; a possible support in favour of such a burst is the presence of a numerous population of globular clusters likely connected to the thick disk , distinct from the population connected to the halo ( zinn , 1985 ) .
perhaps this burst exhausted the existing gas , contributing to the present thick disk or it may have been concentrated in the bulge regions , far from the solar neighbourhood ( as suggested by some simulations : mihos & hernquist 1994 ) , or finally the high sn rate might have caused a galactic wind ( berman & suchkov 1991 ) . anyway , the present thin disk should have formed later by a secondary process ( ostriker & thuan 1975 ) ; models of disk evolution ( kennicutt 1989 ; burkert et al . 1992 ) then indicate that some time would be required before the critical density was reached and star formation started again .
thick disk stars are then likely older than the oldest stars with thin - disk kinematics and a discontinuity would be expected between the thick and thin disk phases .
this agrees with our fig .
[ fig:5]d , which also indicates that a similar large merging could not have occurred during the last 10 gyrs , although observations of the sagittarius dwarf galaxy ( ibata et al . 1994 ) , presently merging with the milky way , indicate that minor episodes are still occurring . the n - body simulations ( quinn et al .
1993 ) indicate that if relatively large merging occurred , the present thick disk would be composed of both stars belonging to the merged satellite(s ) and to the original disk ; the last one should dominate due to its larger mass .
a pure accretion scenario readily explains the lack of appreciable kinematic and chemical gradients in the outer halo ( carney 1993 ) , but it fails to explain the gradients observed in the inner regions . in a smooth dissipational collapse scenario , like that of eggen et al .
, the proposed merging episode appears as an _ ad hoc _ hypothesis .
however , more realistic simulations of galaxy formation by dissipational collapse which include dark matter , gas dynamics , star formation and sn feedback ( katz 1992 ) suggest that many stars form in the cores of dark matter clumps that form during the collapse : even within this scheme then a long - living thin disk could likely form only after the end of an early chaotic phase .
the emerging favoured scenario considers then the inhomogeneous dissipational collapse of the protogalaxy with formation of a few secondary fragments ( having of the order of several hundredths or even a few tenths of the total mass ) which are accreted later , as proposed by norris ( 1994 ) . in this scenario ,
a significant fraction of the halo is due to accretion of fragments , explaining its low specific angular momentum ( wyse and gilmore 1992 ) .
our contributions to this scheme is the consideration that the similar [ fe / o ] ratios for thick disk and halo stars can best be understood if timescales for both contraction and accretion are @xmath90 gyr ( although some later accretion of low - mass fragments is possible ) , and the suggestion that this scheme might naturally produce a discontinuity between the thick and thin disk phases . up to now , the proposed scenario refers to the milky way .
however , it can be easily extended to other galaxies . a strong support to a scenario of spiral formation including both dissipational collapse and accretion is given by the observation that spirals without significant bulges do not have thick disks ( van der kruit & searle 1981a , 1981b ; morrison et al .
1994 , 1997 ; fry et al . 1999 ; matthews et al . 1999 ) .
then ( i ) the presence of thick disks and bulges is not an obvious outcome of galactic formation , but rather depend on some mechanism ( e.g. accretion ) that may or may not be active ; and ( ii ) their origins are likely related ( although likely not on an evolutionary sequence , due to the very different specific angular momentum ) .
this last assertion agrees with the low [ fe / mg ] ratios for stars in the bulge of our own galaxy ( mcwilliam & rich 1994 ) , which suggest that the difference between the ages of the bulk of stars in the bulge , halo and thick disk is @xmath90 gyr .
the presence in the galactic bulge of stars much more metal - rich than the thick disk stars in the solar neighborhood might be understood by assuming either that these stars formed from metal - enriched material in the burst induced by the same merging episode(s ) causing the formation of the thick disk ( mihos & hernquist 1994 ) ; or simply by a pre - existing radial metallicity gradient in the early disk , which is predicted by dissipational collapse models ( larson 1974 ) and should not be canceled by later merging episode(s ) ( quinn et al .
1993 ) . within the mixed scenario ,
significant thick disks and bulges are related to accretion of satellites .
currently available statistics ( zaritsky et al . 1993 ) indicate that there is about one satellite with @xmath91 per primary , though we notice that the milky way has a much larger number of faint companions and the statistics is based on regions where galaxy density is lower than in the local group ; furthermore , there is an excess of close satellites ( separation @xmath92 kpc ) near the minor axis of primaries ( holmberg 1969 ; zaritsky et al .
1997 ) , and it has been suggested that the present population of satellites to the milky way represents only a small fraction of the original population ( see e.g. klypin et al .
1999 ) . the excess of close satellite near the minor axis may be explained by assuming that close satellites on low inclination prograde orbits have smaller chance to survive , due to dynamical friction ; it has been suggested that these _ missing _ satellites have merged into the primary ( zaritsky & gonzalez 1999 ) ; their perturbation may have created the thick disk , and their gas may have fueled the bulge .
given the small number of satellites for each primary galaxy , a strong stochastic variation from galaxy to galaxy is expected .
the present scenario is then coherent to a picture where the entire hubble sequence from sc to ellipticals might be reproduced by assuming an increasing importance of accretion , which should be correlated with the density of galaxies in the local environment ( see e.g. schweizer 2000 ) .
berman , b.g . ,
suchkov , a.a .
1991 , ass , 184 , 169 boesgaard , a.m. , king , j.r . , deliyannis , c.p . , vogt , s. 1999 , aj , 117 , 492 burkert , a. , truran , j.w . , hensler , g. 1992 , apj 391 , 651 buser , r. , rong , j. , karaali , s. 1999 , a&a 348 , 98 carney , b.w .
1993 , in the globular cluster - galaxy connection , eds g.h .
smith , j.p .
brodie , asp conf .
48 , p. 234
carney , b.w .
, latham , d.w . ,
laird , j.b . ,
aguilar , l.a .
1994 , aj , 107 , 2240 carney , b.w . ,
wright , j.s .
, sneden , c. , laird , j.b . ,
aguilar , l.a .
, latham , d.w .
1997 , aj , 114 , 363 carraro , g. , 1994 , a&a 288 , 751 carraro , g. , girardi , l. , chiosi , c. 1999 , mnras , 309 , 430 carretta , e. , gratton , r.g . , sneden , c. 1996 , a&a , in press ( paper iii ) chiappini , c. , matteucci , f. , gratton , r.g .
1997 , apj , 477 , 765 clementini , g. , carretta , e. , gratton , r.g . , merighi , r. , mould , j. , mccarthy , j.k .
1995 , aj , 110 , 2319 cowan , j.j . ,
pfeiffer , b. , kratz , k .-
thielemann , f .-
k , sneden , c. , burles , s. , tytler , d. , beers , t.c . 1999 ,
apj , 521 , 194 demarque , p. , green , e.m . , guenther , d.b .
1992 , aj , 103 , 151 edvardsson , b. , andersen , j. , gustafsson , b. , lambert , d.l . ,
nissen , p.e . ,
tomkin , j. 1993 .
a&a , 275 , 101 eggen , o.j . ,
lynden - bell , d. , sandage , a.r .
1962 , apj , 136 , 748 elmegreen , b.g . 1992 , in the galactic interstellar medium , d. pfenniger , p. bartholdi eds , springer - verlag , berlin , p. 157 franois , p. matteucci , f. 1993 , a&a , 280 , 136 frenk , c.s . ,
white , s.d.m . ,
efstathiou , g. , davis , m. 1985 , nature , 317 , 595 fry , a.m. , morrison , h.l .
, harding , p. , borosin , t.a . 1999 ,
aj , 118 , 1209 fuhrmann , k. 1998 , a&a , 338 , 161 fuhrmann , k. 1999 , ap&ss , 265 , 265 gallagher , j.s . , hunter , d.m .
1984 , araa 22 , 37 gilmore , g. , wyse , r.f.g . , kuijken , c. 1989 , araa , 27 , 555 girardi , l. , bressan , a. , bertelli , g. , chiosi , c. 2000 , a&as , 141 , 371 gratton , r.g . , carretta , e. , castelli , f. 1996a , a&a , 314 , 191 ( paper i ) gratton , r.g . ,
carretta , e. , matteucci , f. , sneden , c. 1996b , in formation of the galactic halo ... inside and out , h. morrison and a. sarajedini eds . , asp conference series , 92 , p. 307
gratton , r.g .
, carretta , e. , eriksson , k. , gustafsson , b. 1999 , a&a , 350 , 955 ( paper ii ) holmberg , e. 1969 , ark .
astron . , 5 , 305 huang , s. , carlberg , r.g .
, apj , 480 , 503 ibata , r.a . , gilmore , g. , irwin , m.j .
1994 , nature , 370 , 194 israelian , g. , garca lpez , r.j . , rebolo , r. 1998 , apj , 507 , 805 katz , n. 1992 , apj , 391 , 502 kennicutt , r.c.jr , 1989 , apj , 344 , 685 king , j.r .
1993 , aj , 106 , 1206 king , j.r .
1994 , aj , 107 , 350 king , j.r .
1997 , aj , 113 , 2302 klypin , a. , kractsov , a.v . , valenzuela , o. , prada , f. 1999 , apj , 522 , 82 knox , r.a .
, hawkins , m.r.s . , hambly , n.c . 1999 , mnras , 306 , 736 kujiken , k. , gilmore , g. 1989 , mnras , 239 , 604 kurucz , r.l . 1992 , private communication larson , r.b .
1974 , mnras , 166 , 585 larson , r.b .
1976 , mnras , 176 , 31 lynden - bell , d. 1967 , mnras , 136 , 101 malaney , r.a . , fowler , w.a .
1989 , mnras , 237 , 67 matteucci , f. 1991 , in frontiers of stellar evolution , d. lambert ed .
, paspcs 20 , 539 matteucci , f. , franois , p. 1992 , a&a , 262 , l1 ( mf ) matteucci , f. , greggio , l. 1986 , a&a , 154 , 279 matthews , l.d . ,
gallagher , j.s . , van dreil , w. 1999 , aj , 1188 , 2751 mcwilliam , rich , m.j .
1994 , apjs , 91 , 749 mcwilliam , a. , preston , g.w . ,
sneden , c. , searle , l. 1995 , aj , 109 , 2757 mihos , j.c .
, hernquist , l. 1994 , apjl , 425 , l13 mirabel , i.f .
1989 , in structure and dynamics of the interstellar medium , ed .
m. moles , g. tenorio - tagle , and j. melnick ( berlin : springer verlag ) , p. 396
morrison , h.l . , miller , e.d . ,
harding , p. , stinebring , d.r . , boroson , t.a .
1997 , aj , 113 , 2061 morrison , h.l . ,
boroson , t.a . ,
harding , p. 1994 , aj , 108 , 1191 nissen , p.e . , edvardsson , b. 1992 , a&a , 261 , 255 nissen , p.e . , & schuster , w.j .
1997 , a&a , 326 , 751 nomoto , k. , thielemann , f.k . , yokoi , i. 1984 , apj , 286 , 644 norris , j.e .
1993 , in the globular cluster - galaxy connection , eds g.h .
smith , j.p .
brodie , asp conf .
48 , p. 259 norris , j.e .
1994 , apj 431 , 645 norris , j.e .
1999 , ap&ss 265 , 213 ostriker , j.p . , thuan , t.x .
1975 , apj , 202 , 353 perryman , m.a.c .
, et al . 1997 , a&a 323 , l49 phillipps , s. , edmunds , m.g . ,
davies , j.i .
1990 , mnras , 244 , 168 quinn , p.j . ,
hernquist , l. , fullagar , d.p .
1993 , apj , 403 , 74 rana , n.c . ,
basu , s. 1991 , a&a , 265 , 499 robin a.c . ,
haywood , m. , crz , m. , ojha , d.k . , bienaym , o. 1996 , a&a 305 , 125 scalo , j.m .
1986 , fund .
cosmic phys . , 11 , 1 schmidt , m. 1963 , apj , 137 , 758 schuster , w.j . , nissen , p.e .
1989 , a&a , 222 , 69 schweizer , f. 2000 , astro - ph/0002263 searle , l. , zinn , r. 1978 , apj , 225 , 357 silk , j. 1985 , apj , 297 , 9 skillman , e.d .
1986 , star formation in galaxies , p. 263
lonsdale persson , c.j . ,
nasa , washington straniero , o. , chieffi , a. , limongi , m. 1997 , apj , 490 , 425 thielemann , f.k . , hashimoto , m. , nomoto , k. 1990 , apj , 349 , 222 timmes , f.x . ,
woosley , s.e . , axelrod , t.a .
1995 , apjs , 98 , 617 tomkin , j. lemke , m. , lambert , d.l . , sneden , c. 1992 , aj , 104 , 1568 toomre , a. , toomre , j. 1972 , apj , 178 , 623 tsujimoto , t. , shigeyama , t. , yoshii , y. 1999 , apj , 519 , 63 vandenberg , d.a . ,
bell , r.a .
1985 , apjs , 58 , 561 van der hulst , j.m . ,
skillman , e.d . ,
kennicutt , r.c , bothun , g.d .
1987 , a&a , 177 , 63 van der kruit , p.c . ,
searle , l. 1981a , a&a , 95 , 105 van der kruit , p.c . , searle , l. 1981b , a&a , 95 , 116 velazquez , h. , white , s.d.m .
1999 , mnras , 304 , 254 walker , i.r . ,
mihos , j.c . ,
hernquist , l. 1996 , apj , 460 , 121 wheeler , j.c . , sneden , c. , truran , j.w .
1989 , araa , 27 , 279 woosley , s.e . , weaver , t.a .
1986 , iau coll .
d. mihalas , k.a .
winkler , p. 91 wonget , d.e . ,
hansen , c.j .
, liebert , j. , van horn , h.m . ,
fontaine , g. , nather , r.e .
, kepler , s.o . , lamb , d.q .
1987 , apj , 315 , l77 wyse , r.f.g . ,
gilmore , g. 1992 , aj , 104 , 144 zaritsky , d. , gonzalez , a.h .
1999 , pasp , 111 , 1508 zaritsky , d. , smith , r. , frenk , c. , white , s.d.m .
1993 , apj , 405 , 464 zaritsky , d. , smith , r. , frenk , c.s . ,
white , s.d.m .
1997 , apjl , 478 , l53 zinn , r. , 1985 , apj , 293 , 424 zhao , g. , magain , p. 1990
, a&as , 86 , 65 | the accurate o , mg and fe abundances derived in previous papers of this series from a homogenous reanalysis of high quality data for a large sample of stars are combined with stellar kinematics in order to discuss the history of star formation in the solar neighborhood .
we found that the fe / o and fe / mg abundance ratios are roughly constant in the ( inner ) halo and the thick disk ; this means that the timescale of halo collapse was shorter than or of the same order of typical lifetime of progenitors of type ia sne ( @xmath0 gyr ) , this conclusion being somewhat relaxed ( referring to star formation in the individual fragments ) in an accretion model for the galaxy formation .
both fe / o and fe / mg ratios raised by @xmath1 dex while the o / h and mg / h ratios hold constant during the transition from the thick to thin disk phases , indicating a sudden decrease in star formation in the solar neighbourhood at that epoch .
these results are discussed in the framework of current views of galaxy formation ; they fit in a scenario where both dissipational collapse and accretions were active on a quite similar timescale .
12c13@xmath2c/@xmath3c enth |
Subsets and Splits