Invalidity of4fcount determination and possibilities for determination of4fhybridization in intermetallics of the light rare earths by core-level spectroscopy

Abstract

We follow up a previous suggestion that x-ray photoelectron spectroscopy line shapes can be used to estimate the degree of screening orbital delocalization. Using compounds $R{\mathrm{Pd}}_3$ ($R$ = La, Ce, Pr, Nd, and Sm) we show that there can be appreciable but decreasing mixing of the $4f$ states with the extended conduction-band states in the series of $R{\mathrm{Pd}}_3$ compounds in going from La to Nd. This mixing is, however, not detectable by photoemission in Sm${\mathrm{Pd}}_3$. We discuss this mixing in both initial and final states. We conclude that the major initial-state mixing involves $f^0-f^1$ configurations in Ce${\mathrm{Pd}}_3$. The major mixing in the final states always involves the $f^n$ and $f^{n+1\mathrm{}}$ configurations, where the $f^{n+1\mathrm{}}$ configuration has a negligible effect for Sm and heavier elements. The implication of this is that core-level spectroscopies can give reliable information on the $f$ count in compounds of samarium and the heavier rare-earth metals.