Validity of automated x-ray photoelectron spectroscopy algorithm to determine the amount of substance and the depth distribution of atoms

The author reports a systematic study of the range of validity of a previously developed algorithm for automated x-ray photoelectron spectroscopy analysis, which takes into account the variation in both peak intensity and the intensity in the background of inelastically scattered electrons. This test was done by first simulating spectra for the Au4d peak with gold atoms distributed in the form of a wide range of nanostructures, which includes overlayers with varying thickness, a 5 Å layer of atoms buried at varying depths and a substrate covered with an overlayer of varying thickness. Next, the algorithm was applied to analyze these spectra. The algorithm determines the number of atoms within the outermost 3 $λ$ of the surface. This amount of substance is denoted $AOS3λ$ (where $λ$ is the electron inelastic mean free path). In general the determined $AOS3λ$ is found to be accurate to within ∼10–20% depending on the depth distribution of the atoms. The algorithm also determines a characteristic length L, which was found to give unambiguous information on the depth distribution of the atoms for practically all studied cases. A set of rules for this parameter, which relates the value of L to the depths where the atoms are distributed, was tested, and these rules were found to be generally valid with only a few exceptions. The results were found to be rather independent of the spectral energy range (from 20 to 40 eV below the peak energy) used in the analysis.