Mean field methods for atomic and nuclear reactions: The link between time-dependent and time-independent approaches
J. Uhlig - J.C. Lemm - A. Weiguny
Institut für Theoretische Physik I, Universität Münster, D-48149 Münster, Germany
Received: 7 January 1998 / Revised version: 20 April 1998 Communicated by F. Lenz
Three variants of mean field methods for atomic and nuclear reactions are compared with respect to both conception and applicability: The time-dependent Hartree-Fock method solves the equation of motion for a Hermitian density operator as initial value problem, with the colliding fragments in a continuum state of relative motion. With no specification of the final state, the method is restricted to inclusive reactions. The time-dependent mean field method, as developed by Kerman, Levit and Negele as well as by Reinhardt, calculates the density for specific transitions and thus applies to exclusive reactions. It uses the Hubbard-Stratonovich transformation to express the full time-development operator with two-body interactions as functional integral over one-body densities. In stationary phase approximation and with Slater determinants as initial and final states, it defines non-Hermitian, time-dependent mean field equations to be solved self-consistently as boundary value problem in time. The time-independent mean field method of Giraud and Nagarajan is based on a Schwinger-type variational principle for the resolvent. It leads to a set of inhomogeneous, non-Hermitian equations of Hartree-Fock type to be solved for given total energy. All information about initial and final channels is contained in the inhomogeneities, hence the method is designed for exclusive reactions. A direct link is established between the time-dependent and time-independent versions. Their relation is non-trivial due to the non-linear nature of mean field methods.
03.65.Nk Nonrelativistic scattering theory - 24.10.-i Nuclear-reaction models and methods - 34.10.+x General theories and models of atomic and molecular collisions and interactions (including statistical theories, transition state, stochastic and trajectory models, etc.)