# Development of a New Parameter Optimization Scheme for a Reactive Force Field (ReaxFF) Based on a Machine Learning Approach

## TL;DR

This paper introduces a machine learning-based method to optimize ReaxFF parameters for reactive molecular dynamics, enabling accurate high-temperature simulations and insights into chemical vapor deposition processes.

## Contribution

A novel parameter optimization scheme combining k-nearest neighbors and random forest algorithms for ReaxFF, improving efficiency and accuracy in high-temperature reactive MD simulations.

## Key findings

- Optimized ReaxFF parameters accurately predict high-temperature properties.
- The method successfully simulated crystal growth in CVD of α-Al₂O₃.
- Surface growth behaviors were elucidated using the new parameters.

## Abstract

Reactive molecular dynamics (MD) simulation is performed using a reactive force field (ReaxFF). To this end, we developed a new method to optimize the ReaxFF parameters based on a machine learning approach. This approach combines the $k$-nearest neighbor and random forest regressor algorithm to efficiently locate several possible ReaxFF parameter sets, thereby the optimized ReaxFF parameter can predict physical properties even in a high-temperature condition within a small effort of parameter refinement. As a pilot test of the developed approach, the optimized ReaxFF parameter set was applied to perform chemical vapor deposition (CVD) of an $\alpha$-Al$_2$O$_3$ crystal. The crystal structure of $\alpha$-Al$_2$O$_3$ was reasonably reproduced even at a relatively high temperature (2000 K). The reactive MD simulation suggests that the (11$\overline{2}$0) surface grows faster than the (0001) surface, indicating that the developed parameter optimization technique could be used for understanding the chemical reaction in the CVD process.