Computational Heterogeneous Catalysis

Contact Information

Address

Het Kranenveld 14
Helix
5612 AE Eindhoven
Netherlands
View on map, opens in new tab

Phone

+31 402 472078

Email

[email protected]

Organisational Structure

•
Materials for Electrochemistry, Catalysis and Separation
- •
  Catalysis
  - •
    Computational Heterogeneous Catalysis
  - •
    Chemistry of Heterogeneous Catalysts

See full hierarchy

Profile

<div class="OutlineElement Ltr SCXW180330705 BCX0"> Our group develops and applies computational methods to understand heterogeneous catalysis at the molecular level. Using quantum chemistry, microkinetic modelling, and machine learning, we unravel how the structure of a catalyst determines its activity, selectivity, and stability, and how this structure evolves under reaction conditions. Our research focuses on the conversion of CO, CO₂, and hydrogen into sustainable fuels and chemicals, with Fischer-Tropsch synthesis and CO₂ hydrogenation as central examples. Working closely with the experimental groups of Emiel Hensen and Nikolay Kosinov, we calibrate our models with kinetic measurements and operando spectroscopy. </div> <div class="OutlineElement Ltr SCXW180330705 BCX0"> A central theme is the nature of the active site. Steps, corners, terraces, support interfaces, oxide defects, and isolated metal atoms each create a distinct chemical environment. By systematically enumerating the sites that a nanoparticle exposes, we explain how particle size, shape, and the support determine catalytic performance, for example the structure sensitivity of cobalt in Fischer-Tropsch synthesis and of nickel in CO₂ conversion. Electronic-structure analysis further shows why particular site configurations activate molecules such as CO, and how reducible oxides such as ceria stabilise metal clusters and single atoms. </div> <div class="OutlineElement Ltr SCXW180330705 BCX0"> To connect these atomic-scale insights to the laboratory and the reactor, we build multiscale models. First-principles microkinetic models identify rate-controlling steps and competing reaction pathways, while neural networks trained on these models allow the kinetics to be coupled directly to reactor simulations. Together with experimental colleagues, we also use isotope effects to discriminate between pathways in branched reaction networks such as Fischer-Tropsch synthesis. </div> <div class="OutlineElement Ltr SCXW180330705 BCX0"> Increasingly, our work addresses the dynamic nature of working catalysts. Under reaction conditions, surfaces reconstruct, adsorbate layers reorganise, and particles sinter or lose active sites. We aim to describe this evolution with the same kinetic rigour as activity and selectivity, using machine learning and automated computational workflows to reach the much longer length and time scales involved. This is particularly relevant for catalysts that must operate with intermittent supplies of renewable hydrogen. </div>