A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer (2020-09-14T00:00:00.000000Z)

TL;DR

A fourth-generation high-dimensional neural network potential including non-local information of charge populations that is able to provide forces, charges and energies in excellent agreement with DFT data is introduced.

Abstract

Machine learning potentials have become an important tool for atomistic simulations in many fields, from chemistry via molecular biology to materials science. Most of the established methods, however, rely on local properties and are thus unable to take global changes in the electronic structure into account, which result from long-range charge transfer or different charge states. In this work we overcome this limitation by introducing a fourth-generation high-dimensional neural network potential that combines a charge equilibration scheme employing environment-dependent atomic electronegativities with accurate atomic energies. The method, which is able to correctly describe global charge distributions in arbitrary systems, yields much improved energies and substantially extends the applicability of modern machine learning potentials. This is demonstrated for a series of systems representing typical scenarios in chemistry and materials science that are incorrectly described by current methods, while the fourth-generation neural network potential is in excellent agreement with electronic structure calculations. Machine learning potentials do not account for long-range charge transfer. Here the authors introduce a fourth-generation high-dimensional neural network potential including non-local information of charge populations that is able to provide forces, charges and energies in excellent agreement with DFT data.

Authors

J. Behler

2 papers

Tsz Wai Ko

2 papers

Jonas A. Finkler

1 papers

TL;DR

Abstract

Authors

References79 items

Convergence Acceleration in Machine Learning Potentials for Atomistic Simulations

Modell

Detecting non-local effects in the electronic structure of a simple covalent system with machine learning methods

Teaching a neural network to attach and detach electrons from molecules

Global optimization of copper clusters at the ZnO(101¯0) surface using a DFT-based neural network potential and genetic algorithms.

Incorporating Electronic Information into Machine Learning Potential Energy Surfaces via Approaching the Ground-State Electronic Energy as a Function of Atom-Based Electronic Populations.

FLAME: A library of atomistic modeling environments

Machine learning for molecular simulation

Machine Learning Interatomic Potentials as Emerging Tools for Materials Science

Surface reconstructions and premelting of the (100) CaF2 surface.

PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments, and Partial Charges.

An electrostatic spectral neighbor analysis potential for lithium nitride

Atomic cluster expansion for accurate and transferable interatomic potentials

Inducing wetting morphologies and increased reactivities of small Au clusters on doped oxide supports.

Discovering a Transferable Charge Assignment Model Using Machine Learning.

A neural-network potential through charge equilibration for WS2: From clusters to sheets.

SchNet - A deep learning architecture for molecules and materials.

The TensorMol-0.1 model chemistry: a neural network augmented with long-range physics

First Principles Neural Network Potentials for Reactive Simulations of Large Molecular and Condensed Systems.

Energy landscape of ZnO clusters and low-density polymorphs

High accuracy and transferability of a neural network potential through charge equilibration for calcium fluoride

Perspective: Machine learning potentials for atomistic simulations.

ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost

Machine Learning Force Fields: Construction, Validation, and Outlook

Bypassing the Kohn-Sham equations with machine learning

Geometries, Binding Energies, Ionization Potentials, and Electron Affinities of Metal Clusters: Mgn0, ± 1, n = 1–7

Elucidating the Nature of the Active Phase in Copper/Ceria Catalysts for CO Oxidation

Moment Tensor Potentials: A Class of Systematically Improvable Interatomic Potentials

Constructing high‐dimensional neural network potentials: A tutorial review

Transferable Atomic Multipole Machine Learning Models for Small Organic Molecules.

Interatomic potentials for ionic systems with density functional accuracy based on charge densities obtained by a neural network

Spectral neighbor analysis method for automated generation of quantum-accurate interatomic potentials

Neural network potentials for metals and oxides – First applications to copper clusters at zinc oxide

A neural network potential-energy surface for the water dimer based on environment-dependent atomic energies and charges.

Density functional studies of LiN and LiN+ (N = 2–30) clusters: Structure, binding and charge distribution

Development of a Transferable Variable Charge Potential for the Study of Energy Conversion Materials FeF2 and FeF3

VESTA 3 for three-dimensional visualization of crystal, volumetric and morphology data

Tailoring the shape of metal ad-particles by doping the oxide support.

Neural network potential-energy surfaces in chemistry: a tool for large-scale simulations.

Support vector machine regression (LS-SVM)--an alternative to artificial neural networks (ANNs) for the analysis of quantum chemistry data?

High-dimensional neural-network potentials for multicomponent systems: Applications to zinc oxide

Atom-centered symmetry functions for constructing high-dimensional neural network potentials.

Tuning the morphology of gold clusters by substrate doping.

The effect of ionization on the global minima of small and medium sized silicon and magnesium clusters.

Potential energy surfaces fitted by artificial neural networks.

Ab initio molecular simulations with numeric atom-centered orbitals

Gaussian approximation potentials: the accuracy of quantum mechanics, without the electrons.

Trends in Energies and Geometric Structures of Neutral and Charged Aluminum Clusters.

Generalized neural-network representation of high-dimensional potential-energy surfaces.

A charge transfer ionic–embedded atom method potential for the O–Al–Ni–Co–Fe system

Nearsightedness of electronic matter.

H-substituted anionic carbon clusters CnH− (n⩽10): Density functional studies and experimental observations

Odd carbon long linear chains HC2n+1H (n = 4-11): properties of the neutrals and radical anions.

Advances in the catalysis of Au nanoparticles

ReaxFF: A Reactive Force Field for Hydrocarbons

First-principles calculations for vacancy formation energies in Cu and Al; non-local effect beyond the LSDA and lattice distortion

Generalized Gradient Approximation Made Simple.

Neural network models of potential energy surfaces

Born-Oppenheimer molecular-dynamics simulations of finite systems: Structure and dynamics of (H2O)2.

Charge equilibration for molecular dynamics simulations

Theoretical study of linear Cn (n=6−10) and HCnH (n=2−10) molecules

Monte Carlo simulation of differences in free energies of hydration

A unified formulation of the constant temperature molecular dynamics methods

Dynamics of folded proteins

Bonded-atom fragments for describing molecular charge densities

and s

Nature Comm

Angew

Multiscale Model

ACS Catalysis 6

Mg 0;± 1 n , n= 1–7

Crystallogr

Modelling and Simulation in Materials Science and Engineering Visualization and analysis of atomistic simulation data with OVITO – the Open Visualization Tool

Comput

Condens

TUTORIAL REVIEW