arXiv subjects
Publications and source records attributed to Bogdan Mihaila.
We review the basics of the coupled-cluster expansion formalism for numerical solutions of the many-body problem, and we outline the principles of an approach directed towards an adequate inclusion of continuum effects in the associated single-energy spectrum. We illustrate our findings by considering the simple case of a single-particle quantum mechanics problem.
We study the non-equilibrium dynamics of the O(N) model in classical and quantum field theory in 1+1 dimensions, for N > 1. We compare numerical results obtained using the Hartree approximation and two next to leading order approximations, the bare vertex approximation and the 2PI-1/N expansion. The later approximations differ through terms of order g^2, where g is the scaled coupling constant, g=λ/N. In this paper we investigate the statement regarding the convergence with respect to g. We find that the differences between these twoapproximation schemes diminish for larger values of N, when λis fixed.
We perform a detailed numerical investigation of the dynamics of broken symmetry $λϕ^4$ field theory in 1+1 dimensions using a Schwinger-Dyson equation truncation scheme based on ignoring vertex corrections. In an earlier paper, we called this the bare vertex approximation (BVA). We assume the initial state is described by a Gaussian density matrix peaked around some non-zero value of $<ϕ(0)>$, and characterized by a single particle Bose-Einstein distribution function at a given temperature. We compute the evolution of the system using three different approximations: Hartree, BVA and a related 2PI-1/N expansion, as a function of coupling strength and initial temperature. In the Hartree approximation, the static phase diagram shows that there is a first order phase transition for this system. As we change the initial starting temperature of the system, we find that the BVA relaxes to a new final temperature and exhibits a second order phase transition. We find that the average fields thermalize for arbitrary initial conditions in the BVA, unlike the behavior exhibited by the Hartree approximation, and we illustrate how $<ϕ(t)>$ and $<χ(t)>$ depend on the initial temperature and on the coupling constant. We find that the 2PI-1/N expansion gives dramatically different results for $<ϕ(t)>$.
We study the domain of validity of a Schwinger-Dyson (SD) approach to non-equilibrium dynamics when there is broken symmetry. We perform exact numerical simulations of the one- and two-point functions of lambda phi^4 field theory in 1+1 dimensions in the classical domain for initial conditions where < phi(x) > not equal to 0. We compare these results to two self-consistent truncations of the SD equations which ignore three-point vertex function corrections. The first approximation, which sets the three-point function to one (the bare vertex approximation (BVA)) gives an excellent description for < phi(x) > = phi(t). The second approximation which ignores higher in 1/N corrections to the 2-PI generating functional (2PI -1/N expansion) is not as accurate for phi(t). Both approximations have serious deficiencies in describing the two-point function when phi(0) > .4.
To model the decay of a quasibound state we use the modified two-potential approach introduced by Gurvitz and Kalbermann. This method has proved itself useful in the past for calculating the decay width and the energy shift of an isolated quasistationary state. We follow the same approach in order to propagate the wave-packet in time with the ultimate goal of extracting the momentum-distribution of emitted particles. The advantage of the method is that it provides the time-dependent wave function in a simple semi-analytic form. We intend to apply this method to the modeling of metastable states for which no direct integration of the time-dependent Schroedinger equation is available today.
We discuss a numerical algorithm for solving nonlinear integro-differential equations, and illustrate our findings for the particular case of Volterra type equations. The algorithm combines a perturbation approach meant to render a linearized version of the problem and a spectral method where unknown functions are expanded in terms of Chebyshev polynomials (El-gendi's method). This approach is shown to be suitable for the calculation of two-point Green functions required in next to leading order studies of time-dependent quantum field theory.
We present numerical solutions for differential equations by expanding the unknown function in terms of Chebyshev polynomials and solving a system of linear equations directly for the values of the function at the extrema (or zeros) of the Chebyshev polynomial of order N (El-gendi's method). The solutions are exact at these points, apart from round-off computer errors and the convergence of other numerical methods used in connection to solving the linear system of equations. Applications to initial value problems in time-dependent quantum field theory, and second order boundary value problems in fluid dynamics are presented.
We compare the momentum space with the standard periodic lattice approach to Monte Carlo calculations in classical $ϕ^4$ field theory. We show that the mismatch in the initial value of $ϕ^2_{\text{cl}}(t)$, results in a shift in the ``thermalized'' value, at large times. The two approaches converge to the same result in the continuum limit.
In this paper we discuss a Schwinger-Dyson [SD] approach for determining the time evolution of the unequal time correlation functions of a non-equilibrium classical field theory, where the classical system is described by an initial density matrix at time $t=0$. We focus on $λϕ^4$ field theory in 1+1 space time dimensions where we can perform exact numerical simulations by sampling an ensemble of initial conditions specified by the initial density matrix. We discuss two approaches. The first, the bare vertex approximation [BVA], is based on ignoring vertex corrections to the SD equations in the auxiliary field formalism relevant for 1/N expansions. The second approximation is a related approximation made to the SD equations of the original formulation in terms of $ϕ$ alone. We compare these SD approximations as well as a Hartree approximation with exact numerical simulations. We find that both approximations based on the SD equations yield good agreement with exact numerical simulations and cure the late time oscillation problem of the Hartree approximation. We also discuss the relationship between the quantum and classical SD equations.
In this paper we discuss two methods of resumming the leading and next to leading order in 1/N diagrams for the quartic O(N) model. These two approaches have the property that they preserve both boundedness and positivity for expectation values of operators in our numerical simulations. These approximations can be understood either in terms of a truncation to the infinitely coupled Schwinger-Dyson hierarchy of equations, or by choosing a particular two-particle irreducible vacuum energy graph in the effective action of the Cornwall-Jackiw-Tomboulis formalism. We confine our discussion to the case of quantum mechanics where the Lagrangian is $L(x,\dot{x}) = (1/2) \sum_{i=1}^{N} \dot{x}_i^2 - (g/8N) [ \sum_{i=1}^{N} x_i^2 - r_0^2 ]^{2}$. The key to these approximations is to treat both the $x$ propagator and the $x^2$ propagator on similar footing which leads to a theory whose graphs have the same topology as QED with the $x^2$ propagator playing the role of the photon. The bare vertex approximation is obtained by replacing the exact vertex function by the bare one in the exact Schwinger-Dyson equations for the one and two point functions. The second approximation, which we call the dynamic Debye screening approximation, makes the further approximation of replacing the exact $x^2$ propagator by its value at leading order in the 1/N expansion. These two approximations are compared with exact numerical simulations for the quantum roll problem. The bare vertex approximation captures the physics at large and modest $N$ better than the dynamic Debye screening approximation.
We study a quantum dynamical system of N, O(N) symmetric, nonlinear oscillators as a toy model to investigate the systematics of a 1/N expansion. The closed time path (CTP) formalism melded with an expansion in 1/N is used to derive time evolution equations valid to order 1/N (next-to-leading order). The effective potential is also obtained to this order and its properties areelucidated. In order to compare theoretical predictions against numerical solutions of the time-dependent Schrodinger equation, we consider two initial conditions consistent with O(N) symmetry, one of them a quantum roll, the other a wave packet initially to one side of the potential minimum, whose center has all coordinates equal. For the case of the quantum roll we map out the domain of validity of the large-N expansion. We discuss unitarity violation in the 1/N expansion; a well-known problem faced by moment truncation techniques. The 1/N results, both static and dynamic, are also compared to those given by the Hartree variational ansatz at given values of N. We conclude that late-time behavior, where nonlinear effects are significant, is not well-described by either approximation.
In this document we present our approach to the computation of three-body matrix elements, based on the Urbana family of three-nucleon potentials. The calculations refer only to the necessary matrix elements needed to include the three-nucleon interaction in the manner presented in nucl-th/9912023.
We continue the investigations of the $^{16}$O ground state using the coupled-cluster expansion [$\exp({\bf S})$] method with realistic nuclear interaction. In this stage of the project, we take into account the three nucleon interaction, and examine in some detail the definition of the internal Hamiltonian, thus trying to correct for the center-of-mass motion. We show that this may result in a better separation of the internal and center-of-mass degrees of freedom in the many-body nuclear wave function. The resulting ground state wave function is used to calculate the "theoretical" charge form factor and charge density. Using the "theoretical" charge density, we generate the charge form factor in the DWBA picture, which is then compared with the available experimental data. The longitudinal response function in inclusive electron scattering for $^{16}$O is also computed.
We calculate the charge form factor and the longitudinal structure function for $^{16}$O and compare with the available experimental data, up to a momentum transfer of 4 fm$^{-1}$. The ground state correlations are generated using the coupled cluster [exp(S}] method, together with the realistic v-18 NN interaction and the Urbana IX three-nucleon interaction. Center-of-mass corrections are dealt with by adding a center-of-mass Hamiltonian to the usual internal Hamiltonian, and by means of a many-body expansion for the computation of the observables measured in the center-of-mass system.
We investigate the quantum roll for a particle in a $d$-dimensional ``Mexican hat'' potential in quantum mechanics, comparing numerical simulations in $d$-dimensions with the results of a large-$d$ expansion, up to order $1/d$, of the coupled closed time path (CTP) Green's function equations, as well as to a post-Gaussian variational approximation in $d$-dimensions. The quantum roll problem for a set of $N$ coupled oscillators is equivalent to a $(d=N)$-dimensional spherically symmetric quantum mechanics problem. For this problem the large-N expansion is equivalent to an expansion in $1/d$ where $d$ is the number of dimensions. We use the Schwinger-Mahanthappa-Keldysh CTP formalism to determine the causal update equations to order $1/d$. We also study the quantum fluctuations $ $ as a function of time and find that the $1/d$ corrections improve the agreement with numerical simulations at short times (over one or two oscillations) but beyond two oscillations, the approximation fails to correspond to a positive probability function. Using numerical methods, we also study how the long time behavior of the motion changes from its asymptotic ($d \to \infty$) harmonic behavior as we reduce $d$.
This document gives a detailed account of the terms used in the computation of the ground state mean field and the ground state correlations. While the general approach to this description is given in a separate paper (nucl-th/9802029) we give here the explicite expressions used.
A many-body expansion for the computation of the charge form factor in the center-of-mass system is proposed. For convergence testing purposes, we apply our formalism to the case of the harmonic oscillator shell model, where an exact solution exists. We also work out the details of the calculation involving realistic nuclear wave functions. Results obtained for the Argonne $v$18 two-nucleon and Urbana-IX three-nucleon interactions are reported. No corrections due to the meson-exchange charge density are taken into account.
We discuss the computation of two-body matrix elements from the Argonne $v_{18}$ interaction. The matrix elements calculation is presented both in particle-particle and in particle-hole angular momentum coupling. The procedures developed here can be applied to the case of other NN potentials, provided that they have a similar operator format.