Validity of the Strutinsky method, studied within the Hartree-Fock framework

The foundation of the Strutinsky method within the Hartree-Fock (HF) framework is discussed. Both the basic energy theorem and the energy averaging method are investigated theoretically and numerically. The equivalence of the averaging method with semiclassical expansions is exhibited. Numerical tests of the basic assumptions of the shell-correction approach are presented, which make use of the results of constrained HF calculations for various nuclei and with different effective interactions. The HF energy is decomposed into a smooth part which is shown to be very close to a liquid drop model energy, a first order shell-correction defined in the usual way, and higher order corrections which usually are neglected in standard Strutunsky calculations. These higher order terms are shown not to contribute more than approximately 1-2 MeV to the total energy in medium and heavy nuclei. New results for the nucleus 40Ca show that this is no longer the case for light nuclei. Alternative ways of defining the smooth part of the energy are also considered. It is particularly found that if the Strutinsky averaging is performed selfconsistently, the sum of the average energy plus the first order shell-correction extracted from the average potential reproduces perfectly well the exact HF energy. (Auth.)