Precise determination of the active sites for real catalysts is a challenging task. Experimental approaches are often used to gain insight in the coordination of active sites, but pinning down the exact local structure is difficult, and is therefore often achieved with density functional theory (DFT) calculations. Representative low-index surfaces and specific adsorption sites are considered, but often not directly confirmed at atomistic scale by experiment. To establish a solid link between theory and experiment, UHV experiments performed on model systems, where surfaces are well defined and controlled, are combined with DFT calculations. This combination allows to propose reaction mechanisms and determine activation barriers, to achieve an understanding on how certain surface aspects or properties (e.g. specific defects) affect the catalytic characteristics of a given surface. It is then possible to connect these results gathered in UHV to what might occur under more realistic conditions. Magnetite Fe3O4(001) is such a model system. Despite being well defined and studied under UHV, we recently found that its surface termination undergoes changes such as Fe diffusion, some of them already accessible below room temperature and more importantly under typical UHV conditions. The surface can be considered dynamic: for any given transition state along a reaction path, surface atoms can be displaced, if the resulting improvements in adsorption energies of adsorbates or reductions of activation barriers are at least compensating the energetic cost of these displacements. Ground states of adsorbates are not always immediately responsible of the observed reactions they are involved in. They are often first required to reach non-equilibrium configurations, from which further reactions steps and/or desorptions occur. The same applies for surfaces. Then, the combination and interplay of both these aspects increase significantly the complexity of DFT calculations, therefore requiring a strong guidance from experimental UHV techniques. To illustrate the above concepts, I will present 2 cases: Firstly, H2O/Fe3O4(111), where surface Fe atoms are displaced with respect to the pristine surface termination, to enhance water adsorption energies, directly affecting the observed ground state. Secondly, Pt(CO)/Fe3O4(001) where the rate-limiting step of a Mars-Van-Krevelen CO oxidation process is minimized considering adsorbate metastability and surface dynamics.

Achieving quantitative agreement with experiment: The surprising role of cation diffusion in surface reactions on magnetite (Fe3O4)

ACS Spring 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.