SKM 2023 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

O: Fachverband Oberflächenphysik

O 85: Electronic Structure of Surfaces II

O 85.8: Talk

Thursday, March 30, 2023, 16:45–17:00, REC C 213

Automatic Quantification of Transitional Metal X-ray Photoelectron Spectra using Convolutional Neural Networks — •Lukas Pielsticker, Walid Hetaba, and Mark Greiner — Max Planck Institute for Chemical Energy Conversion, Mülheim an der Ruhr, Germany

In X-ray photoelectron spectroscopy (XPS), quantitative analysis of the nature and composition of surface chemical species is typically performed manually through empirical curve fitting by expert spectroscopists. However, recent advancements in the ease-of-use and reliability of XPS instruments have led to ever more (novice) users creating increasingly large data sets that are becoming harder to analyze by hand. Reflecting this development, more automated analysis techniques are desirable to aid these users with the analysis of big XPS datasets. Here we show that by training convolutional neural networks (CNN) on artificially generated XP spectra with known quantifications (i.e., for each spectrum, the concentration of each chemical species is known), it is possible to obtain models for auto-quantification of transition metal XP spectra. CNNs are shown to be capable of quantitatively determining the presence of metallic and oxide phases, achieving competitive accuracy as more conventional data analysis methods. The proposed networks are flexible enough to accommodate spectra containing multiple chemical elements and measured with different experimental settings. The use of dropout variational inference for the determination of quantification uncertainty is discussed.