An inquiry concerning the principles of Si 2p core-level photoemission shift assignments at the Si/SiO2 interface

It has long been held that high resolution x-ray photoemission spectra of the Si 2p core level at SiSiO2 interfaces provide important structural information, and that any credible interfacial model must be able to account for these data. To this end it has traditionally been assumed that there existed a one-to-one relationship between chemically shifted spectral features and interfacial silicon atoms in specific formal oxidation states. A series of new measurements of Si 2p core level binding energies for cluster-derived Si/Si oxide interfaces appear to stand in direct contradiction to this critical assumption. In this article we present a critique of various responses to the challenge posed by the new observations. Particular attention is given to the logical consequences of either maintaining or rejecting the one-to-one relationship between shifted features and formal silicon oxidation states and to the challenges each of these responses must meet if they are to prevail. (C) 1996 American Vacuum Society.