Semi-classical calculation of the nuclear spin-isospin response functions

Semi-classical theory has been proven to be a very fruitful tool for calculating the response functions in the quasielastic peak. In particular a Thomas-Fermi RPA calculation (TF-RPA) can give a good description of the response functions measured in electron scattering. This may be no longer the case for surface responses measured with hadronic probes (p,p') for which the (TF-RPA) calculation predicts a sizeable contrast between transverse and longitudinal spin responses in contradiction with the Los Alamos (p vector,p vector') experiment. We show that ℎ2 corrections to the (TF-RPA) theory significantly improve the agreement with the Los Alamos data. (orig.)