PNNL

Abstract

The Al K alpha, 1486.6eV, based XPS of Fe 2p and Fe 3p for Fe(III) in Fe2O3 and Fe(II) in FeO are compared with theoretical predictions based on ab initio wavefunctions that accurately treat the final, core-hole, multiplets. The principal objectives of this comparison are to understand the multiplet structure and to evaluate the use of both the 2p and 3p spectra in determining oxidation state. In order to properly interpret the features of these spectra and to use the XPS to provide atomistic insights as well as atomic composition, it is necessary to understand the origin of the multiplet energies and intensities. The theoretical treatment takes into account the ligand field and spin-orbit splittings, the covalent mixing of ligand and Fe 3d orbitals, and the angular momentum coupling of the open shell electrons. These effects lead to the distribution of XPS intensity into a large number of final, ionic, states that are only partly resolved with energies spread over a wide range of binding energies, BEs. For this reason, it is necessary to record the Fe 2p and 3p XPS spectra over a wide energy range which includes all the multiplets in the theoretical treatment as well as additional shake satellites. We also evaluate the effects of differing assumptions concerning the extrinsic background subtraction, to be sure our experimental spectrum may be fairly compared to the theory. We conclude that the Fe 3p XPS provides an additional means for distinguishing Fe(III) and Fe(II) oxidation states beyond just using the Fe 2p spectrum. In particular, with the use of the Fe 3p XPS, the depth of the material probed is about 1.5 times greater than for the Fe 2p XPS. In addition, a new type of atomic many-body effect that involves excitations into orbitals that have Fe f,l=3, symmetry has been shown to be important for the Fe 3p XPS.