Molecular Dynamics Studies of the Water-Copper Interface Using Neural Network Potentials

点击翻译按钮获取中文摘要

The water-metal interface plays an important role in many fields of research like, e.g., electrochemistry, corrosion and electrocatalysis. These topics have gained a central role in recent years due to their importance for energy storage, fuel cell technology and selective chemical reactions. While a large number of experimental studies has been carried out to date, a theoretical description of these systems is still hampered by the high level of complexity. In principle, computer simulations could provide valuable information about the underlying processes at the atomic scale, but a serious problem is the lack of accurate and at the same time efficient potentials to describe the atomic interactions. The goal of this project is the development of consistent neural network potentials for water and the water-metal interface based on density-functional theory using copper as model system. The potential will be applied to large-scale molecular dynamics simulations of the structural and dynamical properties of the water-copper interface without an applied external bias. Understanding these properties is an important prerequisite for future studies of any electrochemical process. Central topics of the project are the mobility of the water molecules at the surface, the role of defects, and the structure of the surface in contact with the solvent.