Abstract
Analogies and differences between the random phase approximation (RPA) and the Gaussian approximation at finite temperature are discussed. A comparison is made for Hartree and Hartree-Fock temperature dependent mean fields. Some ambiguities, in the definition of the Hartree mean field and in the relative correction, are analyzed. In a solvable model with a repulsive potential, the Gaussian approximation and the RPA, both built from the Hartree mean field, are found more reliable than the RPA built from a Hartree-Fock mean field.
- Received 12 January 1993
DOI:https://doi.org/10.1103/PhysRevC.49.131
©1994 American Physical Society