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

O: Fachverband Oberflächenphysik

O 32: Poster: Solid-Liquid Interfaces

O 32.2: Poster

Tuesday, March 19, 2024, 18:00–20:00, Poster C

Constructing High-Dimensional Neural Network Potentials for Oxide-Water Interfaces — •Jan Elsner and Jörg Behler — Theoretische Chemie II, Ruhr-Universität Bochum, Germany, and Research Center Chemical Sciences and Sustainability, Research Alliance Ruhr, Germany

Oxide-water interfaces are promising systems for catalytic water splitting, a process of significant interest due to its potential in sustainable hydrogen production. However, the complex nature of such interfaces and long time scales associated with dynamical processes presents a substantial theoretical challenge. High-Dimensional Neural Network Potentials (HDNNPs) provide a solution to these challenges by enabling atomistic simulations with DFT-level accuracy at only a fraction of the computational expense. We present the construction of a HDNNP suitable for studying oxide-water interfaces with the overarching goal to utilize these HDNNPs for atomistic simulations of these interfaces.

Keywords: Machine Learning Potentials; Solid-liquid interfaces; Molecular Dynamics