Ivo Filot

Eindhoven University of Technology
Multiscale Modeling of Nanoparticle Catalysis Using Machine Learning and Artificial Neural Networks

Understanding catalytic processes at the nanoscale requires modeling approaches that capture phenomena across multiple length and time scales. Traditional computational techniques, such as density functional theory (DFT) and mean-field microkinetic modeling (MKM), provide detailed insights into reaction mechanisms but often struggle with the complexity and computational cost of large-scale catalytic systems. In this presentation, I discuss strategies for multiscale modeling of nanoparticle catalysis, focusing on how machine learning (ML) and artificial neural networks (ANNs) can enhance predictive capabilities and computational efficiency.

By integrating ML-based surrogate models trained on high-fidelity DFT data with MKM, we develop a framework that links atomistic reaction energetics with mesoscale kinetic simulations. This approach enables a more accurate representation of reaction networks, catalyst dynamics, and structure-activity relationships. I will discuss key challenges in model training, transferability, and validation, as well as strategies to improve the interpretability and accuracy of ML-assisted multiscale models.

The presentation will highlight applications in nanoparticle catalysis, demonstrating how ML-driven modeling can optimize reaction conditions and improve the fundamental understanding of catalytic transformations in complex environments.


Dr. Ivo Filot is a computational chemist specializing in multiscale heterogeneous catalysis. He obtained his Bachelor’s and Master’s degrees with cum laude distinction before completing a PhD in computational chemistry under the supervision of Profs. Rutger van Santen and Emiel Hensen in 2015.

After his PhD, Dr. Filot joined the Multiscale Catalytic Energy Conversion (MCEC) consortium as a tenure-track assistant professor, where he developed computational methods to model catalytic processes across different length and time scales. As part of this work, he contributed to an advanced force field fitting procedure for ReaxFF during a research stay at Penn State University.

In 2018, he was promoted to tenured assistant professor, focusing on reaction mechanisms in transition metal clusters and structure-activity relationships in heterogeneous catalysis. His research aims to improve the understanding and design of catalytic systems through multiscale modeling approaches.