Investigating oxygen reduction kinetics at Au-water interfaces via neural network potential accelerated metadynamics Xin Yang , Arghya Bhowmik, Tejs Vegge, Heine Anton Hansen Department of Energy Conversion and Storage Technical University of Denmark, Denmark Explicit modeling of reactions via ab-initio molecular dynamics (AIMD) at the solid-liquid interfaces can provide new understandings towards the reaction mechanisms. However, prohibitive computational cost severely restricts the time- and length-scale of AIMD. Equivariant graph neural network (GNN) based accurate surrogate potentials 1 can accelerate the speed of performing molecular dynamics by several orders after learning on representative structures in a data-efficient manner. we combined uncertainty-aware GNN potentials and enhanced sampling 2,3 to investigate the reactive process of the oxygen reduction reaction (ORR) at Au-water interface. By using a well- established active learning framework 4 , we can evenly sample equilibrium structures from MD simulations and non-equilibrium reaction intermediates that are rarely visited during the reaction. The trained GNNs have shown exceptional performance in terms of force prediction accuracy, the ability to reproduce structural properties, and low uncertainties when performing MD and metadynamics simulations. Furthermore, the collective variables employed in this work enable an automatic search of reaction pathways and provide a detailed understanding towards the ORR reaction mechanism at Au-water interface. Our simulations identify the two-electron and four-electron transfer reaction mechanisms of ORR on different Au facets, which are in agreement with the experimental findings. The methodology employed in this study can pave the way for modeling complex chemical reactions at electrochemical interfaces with an explicit solvent at ambient conditions. References 1. Schütt, K.; Unke, O.; Gastegger, M. Equivariant message passing for the prediction of tensorial properties and molecular spectra. International Conference on Machine Learning. 2021; pp 9377–9388. 2. Laio, A.; Parrinello, M. Escaping free-energy minima. Proceedings of the National Academy of Sciences 2002, 99, 12562– 12566. 3. Laio, A.; Gervasio, F. L. Metadynamics: a method to simulate rare events and reconstruct the free energy in biophysics, chemistry and material science. Reports on Progress in Physics 2008, 71, 126601. 4. Li, C.; Wang, X.; Dong, W.; Yan, J.; Liu, Q.; Zha, H. Joint active learning with feature selection via cur matrix decomposition. IEEE transactions on pattern analysis and machine intelligence 2018, 41, 1382–1396.
DD07
© The Author(s), 2023
Made with FlippingBook Learn more on our blog