Sample structure prediction from measured XPS data using Bayesian estimation and SESSA simulator

Hiroshi Shinotsuka; Kenji Nagata; Malinda Siriwardana; Hideki Yoshikawa; Hayaru Shouno; Masato Okada

doi:10.1016/j.elspec.2023.147370

Article Sample structure prediction from measured XPS data using Bayesian estimation and SESSA simulator

Hiroshi Shinotsuka (Research and Services Division of Materials Data and Integrated System, National Institute for Materials Science) ; Kenji Nagata (Research and Services Division of Materials Data and Integrated System, National Institute for Materials Science) ; Malinda Siriwardana (Research and Services Division of Materials Data and Integrated System, National Institute for Materials Science) ; Hideki Yoshikawa (Research and Services Division of Materials Data and Integrated System, National Institute for Materials Science) ; Hayaru Shouno (Graduate School of Informatics and Engineering, The University of Electro-Communications) ; Masato Okada (Graduate School of Frontier Science, The University of Tokyo)

Download file (1.86 MB)
Download as zip (1.56 MB)

Collection

Citation

Hiroshi Shinotsuka, Kenji Nagata, Malinda Siriwardana, Hideki Yoshikawa, Hayaru Shouno, Masato Okada. Sample structure prediction from measured XPS data using Bayesian estimation and SESSA simulator. JOURNAL OF ELECTRON SPECTROSCOPY AND RELATED PHENOMENA. 2023, 267 (), 147370. https://doi.org/10.1016/j.elspec.2023.147370

(BibTeX)

Description:

(abstract)

We have developed a framework for solving the inverse problem of X-ray photoelectron spectroscopy (XPS) by incorporating an XPS simulator, Simulation of Electron Spectra for Surface Analysis (SESSA), into Bayesian estimation to obtain an overall picture of the distribution of plausible sample structures from the measured XPS data. The Bayesian estimation framework automated the very tedious task of adjusting the sample structure parameters manually in the simulator. As an example, we performed virtual experiments of angle-resolved XPS on a four-layered sample, and we estimated the sample structures based on the XPS intensity data obtained from experiments. We succeeded in not only obtaining an optimal solution, but also visualizing the distribution of the solution through the Bayesian posterior probability distribution.

Rights: