High Pressure and Temperature Neural Network Reactive Force Field for Energetic Materials

TL;DR

This paper introduces a neural network reactive force field for energetic materials, enabling accurate simulations of their behavior under extreme conditions, surpassing traditional force fields in performance and predictive capability.

Contribution

The authors develop a neural network-based reactive interatomic potential with an iterative training approach, improving accuracy for energetic materials under various conditions.

Findings

Enhanced accuracy over existing force fields in predicting detonation performance

Better modeling of decomposition products and vibrational spectra

Effective simulation of energetic materials under shock loading

Abstract

Reactive force fields for molecular dynamics have enabled a wide range of studies in numerous material classes. These force fields are computationally inexpensive as compared to electronic structure calculations and allow for simulations of millions of atoms. However, the accuracy of traditional force fields is limited by their functional forms, preventing continual refinement and improvement. Therefore, we develop a neural network based reactive interatomic potential for the prediction of the mechanical, thermal, and chemical response of energetic materials at extreme conditions for energetic materials. The training set is expanded in an automatic iterative approach and consists of various CHNO materials and their reactions under ambient and under shock loading conditions. This new potential shows improved accuracy over the current state of the art force fields for a wide range of properties such as detonation performance, decomposition product formation, and vibrational spectra under ambient and shock loading conditions.