Berlin 2024 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

O: Fachverband Oberflächenphysik

O 111: Heterogeneous Catalysis II

O 111.4: Vortrag

Freitag, 22. März 2024, 11:15–11:30, TC 006

High-throughput catalyst screening for CO₂ to methanol conversion with machine-learned force-fields — •Ondřej Krejčí, Prajwal Pisal, and Patrick Rinke — Department of Applied Physics, School of Science, Aalto University, Espoo, Finland

The search for new and better catalysts is one of the key research directions in material science, as heterogeneous catalysis is essential in converting CO₂ to fuel in a closed loop carbon cycle. Approximative predictions of activity, like the Sabatier principle, have been very popular for catalytic screening. However, to take the nanostructure of real thermocatalysts into account, we need to scan the adsorption energies (AEs) for a variety of different facets and binding sites.

In this work, we will present our current workflow for obtaining the relevant AEs in CO₂ thermoreduction to methanol. We employ trained machine learning force-field from the Open Catalyst Project [1], to accelerate the search for ideal catalysts. We have calculated the surface stabilities for various facets with all Miller indices ∈ {−2,−1,0,1,2} and picked the most stable cuts for each facet. Subsequently, we have created all possible high symmetry binding sites on those facets and predicted AEs for the reaction key semi-products: *H, *OH, *OCHO and *OCH₃. The AE distributions are further analysed for material*s activity.

[1] L. Chanussot et al., ACS Catal. 11, 6059-6072 (2021); R. Tran et al., ACS Catal. 13, 3066-3084 (2023); https://opencatalystproject.org/

Keywords: Heterogeneous Catalysis; CO2 reduction; Machine learning force field; High-throughput calculations; Sabatier principle