Accurate computational prediction of core-electron binding energies in carbon-based materials: A machine-learning model combining density-functional theory and $\boldsymbol{GW}$

TL;DR

This paper introduces a machine-learning model that accurately predicts core-electron binding energies in carbon-based materials by combining DFT and GW methods, enabling quick and precise XPS spectrum predictions.

Contribution

The authors develop a novel ML approach integrating DFT and GW calculations with kernel ridge regression for rapid, accurate core-electron binding energy predictions.

Findings

Achieves spectral predictions within 0.1 eV of experimental data

Provides a fast, user-friendly online tool for XPS predictions

Successfully applied to materials with carbon, hydrogen, and oxygen

Abstract

We present a quantitatively accurate machine-learning (ML) model for the computational prediction of core-electron binding energies, from which x-ray photoelectron spectroscopy (XPS) spectra can be readily obtained. Our model combines density functional theory (DFT) with $GW$ and uses kernel ridge regression for the ML predictions. We apply the new approach to materials and molecules containing carbon, hydrogen and oxygen, and obtain qualitative and quantitative agreement with experiment, resolving spectral features within 0.1 eV of reference experimental spectra. The method only requires the user to provide a structural model for the material under study to obtain an XPS prediction within seconds. Our new tool is freely available online through the XPS Prediction Server.