News

Simulating XPS Spectra with the Help of Artificial Intelligence: New Theoretical Protocol and Online Application

07/07/2025

banner
banner

A team of researchers has developed a new method to accurately simulate X-ray Photoelectron Spectroscopy (XPS) spectra, a powerful technique used to analyze the chemical composition of materials at the atomic level. Thanks to an advanced theoretical approach based on Density Functional Theory (DFT) and a specific computational method called ΔSCF, it is now possible to predict the binding energies of electrons in carbon, nitrogen, and oxygen atoms found in organic molecules, functionalized materials, and thin films.

The protocol was validated by comparing theoretical results with experimental data, showing high accuracy across a wide range of systems including aromatic compounds, pharmaceuticals, and biomolecules. The generated data also served as the training set for a machine learning (ML) model capable of quickly and reliably predicting XPS spectra.

Building on this theoretical work, the team at CNR-ISM (CNR - Istituto di Struttura della Materia) developed XPS-ML-Predictor, a web-based application powered by artificial intelligence. The app allows users to predict the XPS spectrum for carbon atoms (C1s) directly from a molecular structure file (in .xyz format), also displaying atom-specific contributions to the overall spectral shape. The tool is intuitive, freely accessible online, and supports molecules containing carbon, nitrogen, oxygen, sulfur, and halogens (fluorine, chlorine, bromine, iodine).

All data, input files, and materials necessary to reproduce the results are available in a public repository to promote transparency and data sharing within the scientific community.

The project was funded by ICSC – National Research Center in High Performance Computing, Big Data and Quantum Computing (as part of the EU’s NextGenerationEU program) and by the Italian Ministry of University and Research (MUR) through the PRIN 2022 program (Projects of Relevant National Interest), project “NIR+”.

Per informazioni:
Francesco Porcelli
CNR - Istituto di struttura della materia
francesco.porcelli@ism.cnr.it
Comunicazione Cnr-Ism, email: comunicazione@ism.cnr.it

Vedi anche: