Chemical state quantification of iron and chromium oxides using XPS: the effect of the background subtraction method

Quantitative chemical analysis based on X-ray photoelectron spectroscopy (XPS) includes elemental identification and, in many cases, quantification of the chemical states of active surface atoms. The two main parts of the analysis are subtracting the inelastic background and fitting a synthetic lineshape to the data. The analysis of transition metal oxides is typically based on their core level peaks. In the photoelectron spectrum, these regions are usually complicated and therefore the choice of the analysis methods can significantly affect the quantification results. In this work, we studied the chemical state quantification of iron and chromium oxides in the case of two chemical states in both samples; the analysis was based on the 2p region of the photoelectron spectrum. In particular, we focused on the effect of the background subtraction method on the analysis results by using (i) the Tougaard background with Tougaard's universal parameters, (ii) the Tougaard background with Seah's element-specific parameters, and (iii) the traditional Shirley background. The lineshape used to represent the 2p peaks was a Gaussian-Lorentzian product function with a constant-exponential tail. We also characterized typical sources of uncertainty related to XPS-based chemical analysis and compared them to the effect of the background. The chemical state proportions obtained with the studied backgrounds varied by ±5% units for iron oxide and ±3% units in the case of chromium oxide. This background-induced variation was found to be comparable, e.g., to the typical uncertainty in the chemical shift between the states.