Machine Learning Driven ReaxFF Optimization for Combustion

Accurate simulation of combus-tion reactions is crucial for un-derstanding combustion mecha-nisms.Reactive force fields(ReaxFF)offer a computation-ally efficient approach to simu-lating complex combustion pro-cesses,but their accuracy de-pends critically on parameteri-zation.This work presents a comprehensive optimization of ReaxFF parameters for gas-phase combustion reactions using a machine learning driven ap-proach.We constructed a dataset of 33 reactions,encompassing key reaction types in combus-tion.High-level double hybrid DFT calculations served as a benchmark to evaluate the per-formance of various density functionals,the semi-empirical PM7 method,and existing ReaxFF parameter sets.We then employed the JAX-ReaxFF framework to optimize the CHO2008 parameters,leveraging its efficient local gradient-based optimization algorithms.The optimized ReaxFF significantly improved the accuracy of potential energy and atomic force predictions,with the mean absolute error(MAE)for energy approaching that of PM7.Analysis of reaction pathways and potential energy surfaces further demonstrated the en-hanced performance of the optimized force field,particularly near transition states.This opti-mized ReaxFF provides a good tool for simulating a wide range of combustion systems,and the presented methodology offers a general strategy for developing system-specific ReaxFF parameters.