Bridging the Gap between Direct Dynamics and Globally Accurate Reactive Potential Energy Surfaces Using Neural Networks

Abstract

Direct dynamics simulations become increasingly popular in studying reaction dynamics for complex systems where analytical potential energy surfaces (PESs) are unavailable. Yet, the number and/or the propagation time of trajectories are often limited by high computational costs, and numerous energies and forces generated on-the-fly become wasted after simulations. We demonstrate here an example of reusing only a very small portion of existing direct dynamics data to reconstruct a 90-dimensional globally accurate reactive PES describing the interaction of CO 2 with a movable Ni(100) surface based on a machine learning approach. In addition to reproducing previous results with much better statistics, we predict scattering probabilities of CO 2 at the state-tostate level, which is extremely demanding for direct dynamics. We propose this unified way to investigate gaseous and gas−surface reactions of medium size, initiating with hundreds of preliminary direct dynamics trajectories, followed by low-cost and high-quality simulations on full-dimensional analytical PESs.

Effect of surface temperature on quantum dynamics of H2 on Cu(111) using a chemically accurate potential energy surface

“…In the last few years, construction of chemically accurate high-dimensional neural network potentials (HD-NNPs) [42][43][44][45][46][47] for various important gas-metal collisional processes (e.g., CO 2 -Ni(100)/Pt(111) [48][49][50] , NO-Au(111) 51 ) has been progressed extensively to overcome the bottleneck of expensive AIMD method. Such neural network based approach allows accurate calculation of reaction probabilities even with very low magnitude (10 −5 − 10 −4 ) for highly activated chemisorption reactions, N 2 + Ru(0001) 52 and CHD 3 + Cu(111) 53 .…”

Complex reaction processes in combustion unraveled by neural network-based molecular dynamics simulation

“…NN potentials have also been employed to study the reaction mechanisms of chemical systems. By combining high-precision NN PESs and quantum collision theory, Zhang and Jiang's group have studied a series of elementary reactions in the gas phase and on the surface [32][33][34][35] . Liu and co-workers developed the LASP program to study the heterogeneous catalysis with NN PESs 36 and built stochastic surface walking (SSW)-NN to explore reaction pathways from glucose to 5-hydroxymethylfurfural 37 .…”

Accurate Probabilities for Highly Activated Reaction of Polyatomic Molecules on Surfaces Using a High-Dimensional Neural Network Potential: CHD3 + Cu(111)

“…13−18 So far, these studies either involved diatomic molecules 15−17,19 or they neglected the movement of surface atoms. 18,20−23 Very recently, a high-dimensional neural network potential (HD-NNP) has been developed for a system in which a linear triatomic molecule interacts with a metal surface, i.e., CO 2 + Ni(100), 24 also including surface atom motion. The neglect of surface motion can limit the accuracy of these studies due to the neglect of energy exchange between the molecule and the surface.…”