Neural Network Based in Silico Simulation of Combustion Reactions

TL;DR

This paper introduces neural network potentials trained on reference datasets to perform fast, accurate reactive molecular dynamics simulations of combustion processes, enabling detailed mechanism exploration of large systems.

Contribution

The study develops neural network potentials that replicate DFT-level accuracy for large-scale combustion simulations, significantly improving computational efficiency.

Findings

Neural network potentials predict potential energy and forces with DFT accuracy.

Nanosecond MD trajectories for large systems were generated efficiently.

Detailed combustion mechanisms for hydrogen and methane were elucidated.

Abstract

Understanding and prediction of the chemical reactions are fundamental demanding in the study of many complex chemical systems. Reactive molecular dynamics (MD) simulation has been widely used for this purpose as it can offer atomic details and can help us better interpret chemical reaction mechanisms. In this study, two reference datasets were constructed and corresponding neural network (NN) potentials were trained based on them. For given large-scale reaction systems, the NN potentials can predict the potential energy and atomic forces of DFT precision, while it is orders of magnitude faster than the conventional DFT calculation. With these two models, reactive MD simulations were performed to explore the combustion mechanisms of hydrogen and methane. Benefit from the high efficiency of the NN model, nanosecond MD trajectories for large-scale systems containing hundreds of atoms were produced and detailed combustion mechanism was obtained. Through further development, the algorithms in this study can be used to explore and discovery reaction mechanisms of many complex reaction systems, such as combustion, synthesis, and heterogeneous catalysis without any predefined reaction coordinates and elementary reaction steps.