Data-efficient and physics-inspired machine learning models to treat complex catalyst materials and reaction networks

Mie Andersen

Aarhus Institute of Advanced Studies and
Center of Interstellar Catalysis, Aarhus University, Denmark

Monday, 25 April 2022,16:00 s.t.

The talk will be given in hybrid mode.

You can join at:
Freihaus Hörsaal 7 (HS 7)
TU Freihaus, Yellow Area, 2nd floor
Wiedner Hauptstraße 8, 1040 Vienna

Or you can join the zoom meeting:
https://tuwien.zoom.us/j/92739417554?pwd=MlFkNjJxUjFkUUhPaUJmZ0ZnMjVOZz09
Meeting ID: 927 3941 7554     Passcode: X74b82XE

Data-efficient and physics-inspired machine learning models to treat complex catalyst materials and reaction networks

Recently developed machine-learning methods hold great promise for simultaneously reducing the computational cost and increasing the accuracy in catalysis modeling, allowing us to capture more complexity, make our models more realistic, and perhaps even obtain new physical insights. I will introduce our work using the compressed sensing “SISSO” method to develop physics-inspired and interpretable models for the binding energies of atoms and small molecules at various types of facets and active site motifs present at transition metal alloys and doped transition metal oxides [1, 2, 3]. Additionally, I will discuss recent work aimed at describing more complex adsorbates with bi- or polydentate adsorption motifs at transition metals and their alloys [4]. Further insights into the catalytic function of materials can be obtained by coupling density functional theory calculations (or machine-learning predictions thereof) with kinetic models. I will show examples of varying complexity ranging from materials screening using inexpensive mean-field models [5] to a detailed investigation of CO hydrogenation over Rh catalysts, employing more computationally demanding kinetic Monte Carlo simulations in combination with cluster expansion techniques to treat adsorbate-adsorbate interactions [6].

Mie Andersen

Mie Andersen is a Fellow and Associate Professor at Aarhus Institute of Advanced Studies, Department of Physics and Astronomy, and the Center for Interstellar Catalysis at Aarhus University in Denmark. She obtained her PhD in nanoscience from Aarhus University in 2014 and then moved to the Technical University of Munich to join the group of Prof. Karsten Reuter, first as an Alexander von Humboldt Fellow and subsequently as a group leader. In 2021 she received an AIAS fellowship and a Villum Young Investigator grant and took up her current position in Aarhus. Her research interests include heterogeneous catalysis, surface science, astrochemistry, and machine learning. In her work, she uses computational methods to study the relationship between materials structure and catalytic activity, primarily for industrial heterogeneous catalysis, but lately also for interstellar dust grain catalysis.