Photoemission in strongly correlated crystalline f-electron systems: A need for a new approach

The unusual properties of heavy fermion (or heavy electron) materials have sparked an avalanche of research over the last two decades in order to understand the basic phenomena responsible for these properties. Photoelectron spectroscopy (often referred to as PES in the following sections), the most direct measurement of the electronic structure of a material, should in principle be able to shed considerable light on this matter. In general the distinction between a localized and a band-like state is trivially observed in band dispersion. Much of the past work was performed on poly-crystalline samples, scraped in-situ to expose a clean surface for PES. There have since been considerable advances both in the quality of specimens as well as experimental resolution, which raise questions regarding these conclusions. Much of the past work on poly-crystalline samples has been reported in several review articles, most notably Allen et al., and it is not necessary here to review those efforts again, with the exception of subsequent work performed at high resolution. The primary focus of the present review will be on new measurements obtained on single crystals, cleaved or prepared in situ and measured at high resolution, which seem to suggest that agreement with the GS and NCA approximations is less than perfect, and that perhaps the starting models need to be modified, or that even an entirely new approach is called for. Of the promising new models the Periodic Anderson Model is most closely related to the SIM. Indeed, at high temperatures it reverts to the SIM. However, the charge polaron model of Liu (1997) as well as the two-electron band model of Sheng and Cooper (1995) cannot yet be ruled out. Inasmuch as the bulk of the single crystal work was performed by the Los Alamos group, this review will draw heavily on those results. Moreover, since the GS and NCA approximations represent the most comprehensive and widely accepted treatment of heavy fermion PES, it is only natural that the authors primarily concern themselves with analysis of PES data in terms of these models, in order to thoroughly test their validity in light of the new data