Determinants of Laplacians on surfaces of finite volume

Determinants of the Laplace and other elliptic operators on compact manifolds have been an object of study for many years (see [MP, RS, Vor]). Up until now, however, the theory of determinants has not been extended to non-compact situations, since these typically involve a mixture of discrete and continuous spectra. Recent advances in this theory, which are partially motivated by developments in mathematical physics, have led to a connection, in the compact Riemann surface case, between determinants of Laplacians on spinors and the Selberg zeta function of the underlying surface (see [DP, Kie, Sar, Vor]).Our purpose in this paper is to introduce a notion of determinants on non-compact (finite volume) Riemann surfaces. These will be associated to the Laplacian shifted by a parameters(1–s), and will be defined in terms of a Dirichlet series (w, s) which is a sum that represents the discrete as well as the continuous spectrum. It will be seen to be regular atw=0, and our main theorem (see Sect. 1) will express exp as the Selberg zeta function of the surface times the appropriate -factor.A Sloan Fellow and partially supported by NSF grant DMS-8701865     

Analytic extrapolation inL ∞-norm: An alternative approach to “QCD Sum Rules”

In view of physical applications (especially to QCD Sum Rules), the following problem, pertaining to analytic extrapolation techniques, is studied. We are considering amplitudes, which are (real) analytic functions in the complex plane cut along=[s ₀, ). A modelF ₀(s) of the amplitude is given through the values ofF ₀(s) on some interval=[s ₂,s ₁] (withs ₁<s ₀) and the values of its discontinuity on. These values are approximate, and are supplemented by prescribed error channels, measured inL -norm (both on and). Investigating the compatibility between these data leads to an extremum problem which is solved up to a point where numerical methods can be implemented.Unité Associée au CNRS n^o040768     

The electronic structure of Rhodium: Angle-resolved studies of photoelectron and secondary electron emission

We report electron spectroscopic studies of the Rh(111) surface, with the aim to obtain bulk band-structure information. We have measured normal photoemision using tunable synchrotron radiation in the range of photon energies between 11 eV and 55 eV, and angle-dependent photoemission along the LUX and LKL azimuths using the He resonance lines (=21.2 eV, 40.8 eV). To complement these data, we studied angleresolved secondary electron emission after excitation with electrons and photons. We derive parts of the one-electron energy dispersionE(k) along L, and determine the energies of several bulk critical points (in eV):E(> ₇₊/₈₊)=–2.75±0.10,E(> ₈₊=–0.85±0.10,E(> _7–=16.1±0.5,E(> _6–/> _8–)=20.5±0.5,E(X ₇₊)=–5.0±0.1,E(L ₆₊)=–5.6±0.5,E(L ₆₊/L ₄₊₊₅₊)=–2.65±0.10,E(L ₆₊)=9.0±0.5 eV. Our results are compared to several available band structure calculations.     

Solvable isotherms for a two-component system of charged rods on a line

The Coulomb system consisting of an equal number of positive and negative charged rods confined to a one-dimensional lattice is studied. The grand partition function can be calculated exactly at two values of the coupling constant=q ²/k _B T (q denoting the magnitude of the charges). The exact results lead to the conjecture that in the complex scaled fugacity plane, all the zeros of the grand partition function lie on the negative real axis for<2, on the point=–1 for=2, and on the unit circle for>2. In addition, for>4, we conjecture in general and prove at=4 that the zeros pinch the real axis in the thermodynamic limit, with an essential singularity in the pressure at the reduced density 1/2.     

Monte Carlo simulation of phase separation and clustering in the ABV model

As a model for a binary alloy undergoing an unmixing phase transition, we consider a square lattice where each site can be either taken by an A atom, a B atom, or a vacancy (V), and there exists a repulsive interaction between AB nearest neighbor pairs. Starting from a random initial configuration, unmixing proceeds via random jumps of A atoms or B atoms to nearest neighbor vacant sites. In the absence of any interaction, these jumps occur at jump rates _A and _B, respectively. For a small concentration of vacancies (c _v=0.04) the dynamics of the structure factorS(k,t) and its first two momentsk ₁(t),k ₂ ² (t) is studied during the early stages of phase separation, for several choices of concentrationc _B of B atoms. Forc _B=0.18 also the time evolution of the cluster size distribution is studied. Apart from very early times, the mean cluster sizel(t) as well as the moments of the structure function depend on timet and the ratio of the jump rates (= _B/ _A) only via a scaled timet/(). Qualitatively, the behavior is very similar to the direct exchange model containing no vacancies. Consequences for phase separation of real alloys are briefly discussed.     

The magnetic behavior of CeB6: Comparison between elastic and inelastic neutron scattering,initial susceptibility and high-field magnetization

We report measurements of the elastic and inelastic neutron scattering, initial susceptibility and high-field magnetization on thoroughly prepared poly- and single crystalline samples of CeB₆. Part of these experiments have been performed at temperatures down to 60 mK and magnetic fields up to 70 kØe. Our neutron-diffraction data provide the first proof that CeB₆ is an antiferromagnet belowT _N2K as has been suggested by previous bulk experiments. The reduced value of the low-temperature magnetic moment both below and aboveT _N points to the existence of a Kondo effect of the ₇ crystal-field (CF) ground state of Ce³⁺. From the low-temperature width of the quasielastic neutron line, the Kondo temperature is inferred to beT _K3 K. The thermal variation of the initial susceptibility (forT>20K) is semiquantitatively explained invoking, besides the Kondo effect, a ₇- ₈ CF splitting of 70 K and magnetic interactions, which are about 10 times stronger between ₈ states than those between ₇ states. This large ₈- ₈ exchange interaction is also assumed to account for the most striking result of this work, i.e. the lack of any CF-transition peak up to 44 meV in our inelastic neutron-scattering spectra.     

Dynamics of the magnetization in the heavy fermion system CeCu6

We have studied CeCu₆ by inelastic neutron scattering. We found both quasielastic and also inelastic transitions, which we interpret as residual crystal field transitions. The quasielastic linewidth is a strongly nonlinear function of temperature, with approximately _QE=5.0 meV at 300 K, with a crossover _QE=kT at about 13 K and with a residual value of _QE=(0)=0.50 meV forT=0 K. Below 5 K the quasielastic intensityI _QE decreases linearly with temperature. _stI _QE/T is in good agreement with direct measurements of the static susceptibility. The data are fully consistent with a nonordering groundstate of CeCu₆.     

From monotonic to oscillatory decay of correlations: Analytical approximation for the two-dimensional,one-component plasma

An approximate evaluation of the pair distribution and the structure factor is performed analytically for the two-dimensional, one-component plasma at any value of the coupling constant. The approximate distribution remains positive and satisfies three sum rules, including the compressibility one. When 0 or 2, exact results are found. At=2 the transition from monotonie (<2) to oscillatory (>2) decay of correlations takes place. Comparison with the Monte Carlo simulations shows good agreement for 0<<4.     

Differential algebras for quantum groups of type B, C, and D

We study higher order bicovariant differential calculi on the quantum groups O_q(N) and Sp _q (N). We show that the second antisymmetrizer exterior algebra _u is the quotient of the universal exterior algebra _u by the principal ideal generated by . Here denotes the unique up to scalars biinvariant 1-form. Moreover is central in _u and _u is an inner differential calculus. We show that the quadratic dual to the left-invariant algebra _{s L} is isomorphic to the reflection equation algebra. Let be an arbitrary left-covariant first order differential calculus. We show that the dimension of the space of left-invariant 2-forms in the universal exterior algebra equals the number of linearly independent quadratic-linear relations in the quantum tangent space.     

On the maximization of the central gravitational red-shift of a schwarzschild sphere with a given surface gravitational red-shift

Following Bondi static, spherically symmetric equilibrium configurations with a core and an envelope have been considered. It has been shown that for any configurations with nonnegative pressure and density and with a surface red-shiftz _s 4.77 arbitrarily large central red-shiftsz _c are possible in the limiting case of arbitrarily large radius. The effects of imposition of further constraints in the form of a real speed of sound not exceeding the speed of light are also examined. It is seen that for a given limiting sound-to-light-speed ratio . (i) There exists a limiting surface red-shiftz _s() 1.71. (ii) A configuration withz _s >z _s() is not possible, (iii) A configuration withz _s=z _s() has a unique and finitez _c=z _c(). (iv) Forz _s<z _s() arbitrarily large central red-shifts can be obtained for configurations with arbitrarily large radii.     

The analysis of tunneling characteristics measured on Bi2Sr2CaCu2O8 surface

Cleaved in air a-b surface of Bi₂Sr₂CaCu₂O₈ (BSCCO-2212) was measured by means of STM and STS at 4.2 K in liquid hellium bath. From fitting experimental conductivity curves by Dynes' function two superconductivity parameters (gap value) and (smearing parameter) were obtained. The shape of gap structure superimposed on dI/dV characteristics depends on tip-sample distance, what is expressed by the increase of and decrease of with shortening of s. The phenomenon of becoming gap structure more distinct when approaching the tunneling tip to the surface is explained by us as the non-vacuum tunneling, where the surface contamination layer on non-metallic BiO top-surface layer strongly influences the tunneling process. Only for s short enough tunneling electrons penetrate to deeper situated CuO layers and reflect their superconducting behaviour. Non-vacuum STM images are therefore sensitive to the tip-sample distance adjustment. The dependence of gap parameters on lateral position of the tip above the sample can also occur. In such cases STS enables to state which elements of the image belong to the topography of the surface and which to its electron density of states.     

Two-body correlations and pair formation in the two-dimensional Coulomb gas

We investigate pair correlations in the two-dimensional Coulomb gas made up of two species of point ions carrying electric charges Z₁ e(>0) and Z₂ e(<0), and interaction by the logarithmic Coulomb potential. This system is known to be classically stable for couplings=e ²/k _BTc=2/¦Z₁Z₂¦ (whereT is the temperature). Correlations between equally charged ions are shown to be greatly modified at short distances, in the range _c/2<< _c, due to gradual ion condensation. The usual integral equations for the pair correlation functions admit no solutions in that range. Preliminary Monte Carlo simulations for the symmetric case (Z₁=–Z₂) reveal a striking chemical equilibrium between tightly bound ion pairs and free ions, which is reasonably well described by a simple Bjerrum model.     

A Sum Rule for the Two-Dimensional Two-Component Plasma

In a two-dimensional, two-component plasma, the second moment of the density correlation function has the simple value {12[1–(/4)]²}^–1, where is the dimensionless coupling constant. This result is derived by using analogies with critical systems.     

Charge Fluctuations in the Two-Dimensional One-Component Plasma

We study, via computer simulations, the fluctuations in the net electric charge in a two-dimensional, one component plasma (OCP) with uniform background charge density –e in a region inside a much larger overall neutral system. Setting e=1, this is the same as the fluctuations in N , the number of mobile particles of charge e. As expected, the distribution of N has, for large , a Gaussian form with a variance which grows only as ^||, where || is the length of the perimeter of . The properties of this system depend only on the coupling parameter =kT, which is the same as the reciprocal temperature in our units. Our simulations show that when the coupling parameter increases, ^() decreases to an asymptotic value ^()^(2)/2 which is equal (or very close) to that obtained for the corresponding variance of particles on a rigid triangular lattice. Thus, for large , the characteristic length _L=2^/ associated with charge fluctuations behaves very differently from that of the Debye length, _D1/ , which it approaches as 0. The pair correlation function of the OCP is also studied.     

Estimates of general Mayer graphs. II. Long-range behavior of graphs with two root points occurring in the theory of ionized systems

We find the asymptotic behavior of general Mayer 2-graphs (Mayer graphs with two root points), which occur in the theory of ionized systems. This problem arises when one wants to compute corrections to the Debye length for large values of the plasma parameter. For a given 2-graph (r) with Debye-Hückel linese ^– /r, we prove the inequalitiesC _m r ^– e ^– (r) (r ₀)C_Mr^3k–l e ^– , for anyrr ₀, and whereC _m andC _M are positive and finite constants which depend only on . These bounds are finite whenever (r) is not infinite everywhere. The integersl, k, and denote, respectively, the number of lines of the graph , its number of field points, and its local line connectivity (the maximum number of chains linking the root points, which have no line in common). From this result, we deduce that the simple irreducible 2-graphs dominant at large distances decay exponentially likee ^– and have an isthmus between the root points (an isthmus is a line whose deletion separates the graph into two disjoint components, each one containing a root point). We prove also that 2-graphs that have a number of linesl > 3k+ are infinite. We exhibit simple, irreducible prototypes satisfying this condition, for anyk 6. This implies that the Abe-Meeron theory of ionized gases as applied to a classical plasma is not free from divergences. Finally, we extend the preceding results to 2-graphs with lines F_L=(e ^– /r)^k L, withk _L real positive. We prove that they still decay exponentially likee ^– , where is now the maximal flow in a network associated to by assigning the capacityk _L to each lineL.     

Structural and magnetic phase transitions in Pr2NiO4

Neutron powder diffraction and magnetization measurements have been performed in stoichiometric Pr₂NiO₄₊(0), which at room temperature (RT) is orthorhombic (Bmab). Ni²⁺ becomes three dimensionally (3D) antiferromagnetically ordered atT _N 325K, with a propagation vectork=[100], and spins oriented along thea axis (parallel to the propagation vector). The magnetic structure belongs to the _7g (––+)-representation (g _x mode, Shubnikov groupB _pmab') ofBmab, and a magnetic moment of 1.5 _B is measured at 155K. This compound undergoes two different structural phase transitions. First, going down in temperature, Pr₂NiO₄ transforms by a first order phase transition from orthorhombic (Bmab) to tetragonal (P4₂/ncm) atT _c1115K. At high temperature, we can predict a transformation from orthorhombic (Bmab) to tetragonal (I4/mmm) in a continuous way. The extrapolated temperature for this second structural transition isT _c21500±100K. The low temperature structural transformation allows a change in the magnetic structure which forT<T _c1 is better described in the orthorhombic symmetry. Just belowT _c1 the magnetic structure is described by the: _3g (–+)-representation ofPccn(g_xc_yf_z mode, Shubnikov groupPccn), this magnetic phase begins to disappear going down in temperature and at the same time a new magnetic phase grows. This new magnetic structure is described by the _1g ofPccn (c _xg_ya_z mode, Shubnikov groupPccn). Both magnetic structures coexist in a certain temperature range. At 1.5 K thec _xg_ya_z mode represents the total of the magnetic ordering. To reproduce the observed magnetic intensities we are forced to consider that Pr³⁺ is polarized below 40 K, with a magnetic structure which is coupled to the Ni sublattices (i.e. _1g and _3g ). The magnetic moment at low temperature for Pr is about 1.28 _B .     

Resonant tunnelling in asymmetrical double barrier structures – complex coordinate approach

The complex coordinate approach of theoretical atomic spectroscopy is applied to the study of resonant tunnelling in (Ga,In)As/(Al,In)As-based asymmetrical double-barrier heterostructures within applied electric field. The method yields complex eigenvalues E = E _r – i/2 of the Hamiltonian matrix evaluated in a normalizable basis, where E _r is the resonant energy and = / is the tunnelling lifetime. The dependence of E _res on the field and barrier asymmetry is discussed.     

Further studies in aesthetic field theory III: A bounded particle in Γ jk;l i =0,g ij;k =0 theory

In previous papers we showed that certain aesthetic ideas led to a bounded particle. In this paper, we show that a theory based on _jk;l ⁱ =0,g _ij;k =0 withg=0 can also lead to a bounded particle. This theory has the advantage that all tensors constructed fromg _ij , _jk ⁱ , _i are treated in a uniform way. Also, we have sixty-four distinct _jk ⁱ appearing. This was not the case in our previous work.     

Some properties of an “aesthetic” field theory

We continue our study of the Lorentz-invariant field theory based on the equations _jk;l ⁱ =0 and g_ij;k=0. To first order in a perturbation expansion, we find _jk;l ⁱ =0 reduces to the wave equation. In orders higher than the first, we find that _jk;l ⁱ =0 cannot be linearized. We also find that the simple wave-type equation g^ij2g/xⁱx^j=0 is contained in the theory when an appropriate choice is made for the parameters at the origin point.     

Spinor fields on 2 + 1 topologically nontrivial spacetime

Zeta-function regularization is used to evaluate the vacuum energy for a spinor field defined on a space-time of the formR ×H ²/,H ² being the Lobatchevsky plane and a strictly hyperbolic discrete subgroup ofPSL(2,R). The necessary analytic continuation is performed making use of the Selberg trace formula.