“…It is evident that the present model does not entirely cover the kinetics of the relaxation treated. The increasing difference between experimental and theoretical curves for smaller U and the analysis of the relations (4), ( 5), (12) showed that the reason for this lies in the description of the Fermi-Dirac distribution functions f 0 , f 1 . Adopting the condition f p (x, t p ) = 1 for a particular t p , an increase of the error would be observed if a spatial spreading of the f 0 , f 1 exists.…”