Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

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

Journal Article · · Nature Chemistry
; ; ; ; ; ; ; ; ; ;

Abstract

Atomistic simulation has a broad range of applications from drug design to materials discovery. Machine learning interatomic potentials (MLIPs) have become an efficient alternative to computationally expensive ab initio simulations. For this reason, chemistry and materials science would greatly benefit from a general reactive MLIP, that is, an MLIP that is applicable to a broad range of reactive chemistry without the need for refitting. Here we develop a general reactive MLIP (ANI-1xnr) through automated sampling of condensed-phase reactions. ANI-1xnr is then applied to study five distinct systems: 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-1xnr closely matches experiment (when available) and/or previous studies using traditional model chemistry methods. As such, ANI-1xnr proves to be a highly general reactive MLIP for C, H, N and O elements in the condensed phase, enabling high-throughput in silico reactive chemistry experimentation.

Sponsoring Organization:
USDOE
Grant/Contract Number:
NONE; 89233218CNA000001
OSTI ID:
2320314
Alternate ID(s):
OSTI ID: 2329239
Journal Information:
Nature Chemistry, Journal Name: Nature Chemistry Journal Issue: 5 Vol. 16; ISSN 1755-4330
Publisher:
Nature Publishing GroupCopyright Statement
Country of Publication:
United Kingdom
Language:
English

References (81)

First Principles Neural Network Potentials for Reactive Simulations of Large Molecular and Condensed Systems journal August 2017
PACKMOL: A package for building initial configurations for molecular dynamics simulations journal October 2009
Constructing high-dimensional neural network potentials: A tutorial review journal March 2015
Gaussian approximation potentials: A brief tutorial introduction journal April 2015
Force Field Development and Nanoreactor Chemistry book February 2019
On the limited memory BFGS method for large scale optimization journal August 1989
Determination of modified embedded atom method parameters for nickel journal September 1997
Graphitization of amorphous carbons: A comparative study of interatomic potentials journal November 2016
A deep learning interatomic potential developed for atomistic simulation of carbon materials journal January 2022
Active learning accelerates ab initio molecular dynamics on reactive energy surfaces journal March 2021
TeaNet: Universal neural network interatomic potential inspired by iterative electronic relaxations journal May 2022
Amp: A modular approach to machine learning in atomistic simulations journal October 2016
LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales journal February 2022
High-pressure ignition delay time measurements of a four-component gasoline surrogate and its high-level blends with ethanol and methyl acetate journal September 2020
Reaction prediction via atomistic simulation: from quantum mechanics to machine learning journal January 2021
Combustion Mechanisms and Kinetics of Fuel Additives: A ReaxFF Molecular Simulation journal October 2018
Extending the Applicability of the ANI Deep Learning Molecular Potential to Sulfur and Halogens journal June 2020
Lifelong Machine Learning Potentials journal June 2023
Parallel Multistream Training of High-Dimensional Neural Network Potentials journal April 2019
Mechanism of Graphene Formation via Detonation Synthesis: A DFTB Nanoreactor Approach journal April 2019
Performance and Cost Assessment of Machine Learning Interatomic Potentials journal October 2019
High-Fidelity Potential Energy Surfaces for Gas-Phase and Gas–Surface Scattering Processes from Machine Learning journal June 2020
The Rise of Neural Networks for Materials and Chemical Dynamics journal July 2021
First-Principles Monte Carlo Simulations of Reaction Equilibria in Compressed Vapors journal April 2016
An empirical valence bond approach for comparing reactions in solutions and in enzymes journal September 1980
ReaxFF:  A Reactive Force Field for Hydrocarbons journal October 2001
Development of a ReaxFF Potential for Carbon Condensed Phases and Its Application to the Thermal Fragmentation of a Large Fullerene journal January 2015
Discovering chemistry with an ab initio nanoreactor journal November 2014
The ReaxFF reactive force-field: development, applications and future directions journal March 2016
Approaching coupled cluster accuracy with a general-purpose neural network potential through transfer learning journal July 2019
Complex reaction processes in combustion unraveled by neural network-based molecular dynamics simulation journal November 2020
A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer journal January 2021
Automated discovery of a robust interatomic potential for aluminum journal February 2021
SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects journal December 2021
E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials journal May 2022
Towards universal neural network potential for material discovery applicable to arbitrary combination of 45 elements journal May 2022
Learning local equivariant representations for large-scale atomistic dynamics journal February 2023
Machine-learned interatomic potentials by active learning: amorphous and liquid hafnium dioxide journal July 2020
Neural network reactive force field for C, H, N, and O systems journal January 2021
The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for molecules journal May 2020
A benchmark dataset for Hydrogen Combustion journal May 2022
Transition1x - a dataset for building generalizable reactive machine learning potentials journal December 2022
A universal graph deep learning interatomic potential for the periodic table journal November 2022
ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost journal January 2017
The TensorMol-0.1 model chemistry: a neural network augmented with long-range physics journal January 2018
A critical comparison of neural network potentials for molecular reaction dynamics with exact permutation symmetry journal January 2019
A transferable active-learning strategy for reactive molecular force fields journal January 2021
Reaction dynamics of Diels–Alder reactions from machine learned potentials journal January 2022
NewtonNet: A Newtonian message passing network for deep learning of interatomic potentials and forces journal January 2022
Exploring chemical and conformational spaces by batch mode deep active learning journal January 2022
A neural network potential with rigorous treatment of long-range dispersion journal January 2023
Gaussian basis sets for accurate calculations on molecular systems in gas and condensed phases journal September 2007
A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu journal April 2010
Quantum and quasi-classical dynamics of the OH + CO → H + CO 2 reaction on a new permutationally invariant neural network potential energy surface journal January 2014
Permutation invariant potential energy surfaces for polyatomic reactions using atomistic neural networks journal June 2016
Potential energy surface interpolation with neural networks for instanton rate calculations journal March 2018
Less is more: Sampling chemical space with active learning journal June 2018
CP2K: An electronic structure and molecular dynamics software package - Quickstep: Efficient and accurate electronic structure calculations journal May 2020
Fitting potential energy surfaces with fundamental invariant neural network. II. Generating fundamental invariants for molecular systems with up to ten atoms journal May 2020
First principles reactive simulation for equation of state prediction journal June 2021
Lightweight and effective tensor sensitivity for atomistic neural networks journal May 2023
Miller experiments in atomistic computer simulations journal September 2014
A second-generation reactive empirical bond order (REBO) potential energy expression for hydrocarbons journal January 2002
Visualization and analysis of atomistic simulation data with OVITO–the Open Visualization Tool journal December 2009
The atomic simulation environment—a Python library for working with atoms journal June 2017
Reactive molecular dynamics for the [Cl–CH 3 –Br] − reaction in the gas phase and in solution: a comparative study using empirical and neural network force fields journal May 2019
PubChem 2023 update journal October 2022
Self-Consistent Equations Including Exchange and Correlation Effects journal November 1965
Density-functional exchange-energy approximation with correct asymptotic behavior journal September 1988
Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density journal January 1988
Separable dual-space Gaussian pseudopotentials journal July 1996
Improved long-range reactive bond-order potential for carbon. I. Construction journal December 2005
Self-assembly of s p 2 -bonded carbon nanostructures from amorphous precursors journal February 2009
Machine learning based interatomic potential for amorphous carbon journal March 2017
Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces journal April 2007
Accurate and transferable multitask prediction of chemical properties with an atoms-in-molecules neural network journal August 2019
Organic Compound Synthes on the Primitive Eart: Several questions about the origin of life have been answered, but much remains to be studied journal July 1959
Query by committee conference January 1992
Simple and efficient algorithms for training machine learning potentials to force data preprint January 2020
On the Opportunities and Risks of Foundation Models preprint January 2021
Learning Together: Towards foundational models for machine learning interatomic potentials with meta-learning preprint January 2023

Similar Records

Improving Bond Dissociations of Reactive Machine Learning Potentials through Physics-Constrained Data Augmentation
Journal Article · Mon Jan 27 23:00:00 EST 2025 · Journal of Chemical Information and Modeling · OSTI ID:2538113
Chemistry-mediated Ostwald ripening in carbon-rich C/O systems at extreme conditions
Journal Article · Thu Mar 17 00:00:00 EDT 2022 · Nature Communications · OSTI ID:1855522

Related Subjects