SKM 2023 – wissenschaftliches Programm

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

O: Fachverband Oberflächenphysik

O 86: Solid-Liquid Interfaces II: Reactions and Electrochemistry I

O 86.6: Vortrag

Donnerstag, 30. März 2023, 16:30–16:45, TRE Phy

Dynamics of the Iridium-Oxide/Water Interface from Machine Learning Potential Simulations — •Nikhil Bapat, Simon Wengert, Hendrik H. Heenen, and Karsten Reuter — Fritz-Haber-Institut der MPG, Berlin, Germany

The interaction of liquid water with a solid substrate at an electrified solid/liquid interface plays a crucial role in the activity and stability for electrocatalysts such as IrO2. It is therefore important to obtain detailed insights about the underlying processes on these interfaces. Here, resolving the dynamical evolution at the atomic scale requires highly accurate yet efficient models that overcome the length and time scale restrictions imposed by traditional ab initio molecular dynamics simulations. To that end, we use the Gaussian approximation potential framework to construct a machine-learned model trained on density-functional theory data for the IrO2(110)/water interface. The established training protocol uses an automated and iterative procedure to ensure that the training data covers the a priori unknown geometric and compositional motifs of the evolving working interface.

The model is then used to run reactive molecular dynamics simulations with varying surface compositions. Crucial for electrolysis, the reduced computational cost allows to efficiently explore the diverse configuration space of adsorbed intermediates in the oxygen evolution reaction. Furthermore, we investigate the mobility of water in dependence of the underlying surface composition in which, long-range effects are reflected in a strong variation of water mobility as a function of distance to the surface.