Core level spectroscopies of Ce and its compounds

Screening of core holes by delocalized and localized valence electrons is discussed in the light of a scheme involving configuration interaction in both initial and final states. The similarities of this formulation to the models of Friedel, Toyozawa and Kotani, and Schoenhammer and Gunnarsson is pointed out and used to reach a qualitative understanding of Ce intermetallic compounds. Evidence is presented for the importance of f-level hybridization and a surprisingly large f-level width of the order of 0.5 to 1.5 eV. Reasons are given why it is not possible to deduce ground state f-counts directly from, for instance, peak intensities in core level spectroscopies. The discussions imply a need for revision of current views of the electronic structure of Ce and its compounds. (Auth.)