Leiden University logo.

nl en

Modeling the catalytically active site in dynamical environments

  • Mie Andersen (Technical University of Munich, Germany)
Thursday 2 April 2020
Hotel NH Noordwijk Conference Centre Leeuwenhorst

Mie Andersen
Theoretical Chemistry, Technische Universität München, Lichtenbergstr. 4, 85747 Garching, Germany
Email: mie.andersen@ch.tum.de

Molecular dynamics snapshot of graphene flake on liquid Cu.

Catalytic materials under operating conditions may present a significant uncertainty regarding the active site carrying out the desired reaction. At the same time, microkinetic models often rely on input data obtained by computationally expensive density functional theory (DFT) calculations, which may limit the study to one or a few chosen (and hopefully representative) active site models.

In my talk I will discuss recent work where we used compressed sensing methods to identify new low-cost and accurate descriptors that allow to predict adsorption energies at various active sites [1]. Combined with Brønsted-Evans-Polyani relations for the activation energies, these constitue the critical input to a microkinetic model. I will discuss active sites at both transition metal, oxide, and carbide catalysts as well as the various facets and adsorption sites these materials may expose. The descriptors are expressed as non-linear functions of intrinsic properties of the clean catalyst surface, e.g. coordination numbers and d-band moments. From a single DFT calculation of these properties, we predict adsorption energies at all potential surface sites, and thereby also the most stable geometry.

Modeling the active site becomes even more challenging when the catalytic material is present in the liquid state, which is the case for the recently discovered liquid Cu catalyst for high-quality graphene synthesis [2]. I will discuss two approches to tackle this challenge. On the one hand we carried out an ab initio thermodynamics study to assess the stability of a wide range of hydrocarbon adsorbates under various reaction conditions (temperature, methane and hydrogen pressures) used in experimental graphene growth protocols at solid and liquid Cu surfaces [3]. This gives insight into the stable intermediates as well as the role of the hydrogen content during synthesis for the relative stability of hydrogen-passivated versus pure carbon clusters. Subsequently, we performed molecular dynamics simulations using both ab initio and semi-empirical methods to assess the role of the liquid Cu surface during graphene growth [4]. We find that the graphene flake significantly influences properties of the underlying liquid substrate and induces a long-range perturbation to the Cu surface density. This could have important implications for the experimentally observed micrometer-range self-alignment of graphene flakes on liquid Cu [2].


  1. M. Andersen, S. Levchenko, M. Scheffler, K. Reuter, ACS Catal. 9, 2752 (2019).
  2. D. Geng et al. PNAS 109, 7992 (2012).
  3. M. Andersen, J.S. Cingolani, K. Reuter, J. Phys. Chem. C 123, 22299 (2019).
  4. J.S. Cingolani, M. Andersen, K. Reuter (in preparation).
This website uses cookies.