Regensburg 2025 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

O: Fachverband Oberflächenphysik

O 9: Surface Reactions

O 9.3: Talk

Monday, March 17, 2025, 11:00–11:15, H25

Realistic Representations of IrO2 Catalyst Surfaces through Extensive Sampling — •Hao Wan, Hendrik H. Heenen, Christoph Scheurer, and Karsten Reuter — Fritz-Haber-Institut der MPG, Berlin

Iridium oxides catalyze the oxygen evolution reaction with unparalleled activity and stability, even under harsh acidic conditions. However, this performance is sensitively correlated to strong structural, compositional and morphological changes of the working catalyst. At the atomic level little is presently known about the true active state, aside from the unlikelihood of it being ideal rutile IrO2.

This situation spans a vast configurational space, the extensive sampling of which (e.g. via parallel tempering) would be intractable with predictive-quality first-principles calculations. Training a machine-learning interatomic potential (MLIP) as an efficient surrogate is challenged by an unprecedented diversity of training structures, as even the bulk structure and composition is unknown. To this end, we create a comprehensive training set by first assembling prototype bulk structures for various IrOx stoichiometries from existing databases. In an active learning loop, this set is then augmented through extensive sampling of diverse surface structures created from the prototypes. The resulting trained MLIP identifies hexagonal ring structures on the rutile (110), (100), (111) facets as most stable configurations under operating potential, aligning with experimental indications. Activity evaluations on these structures using established descriptors effectively capture trends consistent with experimental observations.

Keywords: Machine learning interatomic potentials; Surface evolution; water electrolysis; IrO2