Exploring the frontiers of chemistry with a general reactive machine learning potential

03 November 2022, Version 1
This content is a preprint and has not undergone peer review at the time of posting.


Reactive chemistry atomistic simulation has a broad range of applications from drug design to energy to materials discovery. Machine learning interatomic potentials (MLIP) have become an efficient alternative to computationally expensive quantum chemistry simulations. In practice, reactive MLIPs require refitting to extensive datasets for each new application, and prior knowledge of reaction networks is required to generate fitting data. In this work, we develop a general reactive MLIP through unbiased active learning with a nanoreactor molecular dynamics inspired sampler. The resulting potential (ANI-nr) is then applied to study five distinct condensed phase reactive chemistry problems: carbon solid-phase nucleation, graphene ring formation from acetylene, biofuel additives, combustion of methane and the spontaneous formation of glycine from early earth small molecules. In all studies, ANI-nr closely matches experiment and/or previous studies using traditional model chemistry methods, without needing to be refit for each application, which enables high-throughput in silico reactive chemistry experimentation.


Reactive molecular dynamics
Reactive force field
Reaction mechanisms
neural network potential
machine learning potential
active learning


Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting Policy [opens in a new tab] - please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here [opens in a new tab] .
This site is protected by reCAPTCHA and the Google Privacy Policy [opens in a new tab] and Terms of Service [opens in a new tab] apply.