Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
O: Fachverband Oberflächenphysik
O 24: Solid-liquid interfaces: Reactions and electrochemistry – Poster
O 24.5: Poster
Montag, 9. März 2026, 18:00–20:00, P2
High-Dimensional Neural Network Potentials for Molecular Dynamics Simulations of Mineral-Water Interfaces — •Maite Böhm1,2, Bernadeta Prus1,2, and Jörg Behler1,2 — 1Theoretische Chemie II, Ruhr-Universität Bochum, Germany — 2Research Center Chemical Sciences and Sustainability, Research Alliance Ruhr, Germany
In recent years, High-Dimensional Neural Network Potentials (HDNNP) have emerged as a popular tool for atomistic simulations of complex systems such as mineral-water interfaces. In this work, we present the construction of an HDNNP based on density functional theory energies and forces for tricalcium aluminate (C3A)-water interfaces. We discuss the development of a dataset that includes the relevant solid-liquid interface interactions by active learning and propose approaches for an accelerated extension of the dataset. The trained potential is applied in Molecular Dynamics simulations of the (100) surface of C3A in contact with water, which are further investigated with respect to the structure and dynamics of water on the surface.
Keywords: Machine Learning Potentials; Neural Networks; Interfaces; Molecular Dynamics