Search Menu
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information
Search Scholarly Publications

Title: Thermodynamic integration by neural network potentials based on first-principles dynamic calculations

Journal Article · · Physical Review B
 [1];  [2];  [2];  [2];  [2];  [3];  [4];  [5]; ORCiD logo [5];  [5]
  1. Kumamoto University (Japan); OSTI
  2. Kumamoto University (Japan)
  3. Kobe Univ. (Japan)
  4. Kyushu Sangyo Univ., Fukuoka (Japan)
  5. Univ. of Southern California, Los Angeles, CA (United States)
+ Show Author Affiliations

In this work, simulation-size effect in evaluating the melting temperature of material is studied systematically by combining thermodynamic integration (TI) based on first-principles molecular-dynamics (FPMD) simulations and machine learning. Since the numerical integration to determine the free energies of two different phases as a function of temperature is very time consuming, the FPMD-based TI method has only been applied to small systems, i.e., less than 100 atoms. To accelerate the numerical integration, we here construct an interatomic potential based on the artificial neural-network (ANN) method, which retains the first-principles accuracy at a significantly lower computational cost. The free energies of the solid and liquid phases of rubidium are accurately obtained by the ANN potential, where its weight parameters are optimized to reproduce FPMD results. The ANN results reveal a significant size dependence up to 500 atoms.

Research Organization:
Univ. of Southern California, Los Angeles, CA (United States)
Sponsoring Organization:
Japan Science and Technology Agency (JST); Japan Society for the Promotion of Science (JSPS); USDOE Office of Science (SC), Basic Energy Sciences (BES). Materials Sciences & Engineering Division
Grant/Contract Number:
SC0018195
OSTI ID:
1803287
Journal Information:
Physical Review B, Journal Name: Physical Review B Journal Issue: 21 Vol. 100; ISSN 2469-9950
Publisher:
American Physical Society (APS)Copyright Statement
Country of Publication:
United States
Language:
English

References (58)

Electronic and Structural Transitions in Dense Liquid Sodium. journal December 2007
Adaptive machine learning framework to accelerate ab initio molecular dynamics journal December 2014
Approximation by superpositions of a sigmoidal function journal December 1989
Approximation by superpositions of a sigmoidal function journal December 1992
Multilayer feedforward networks are universal approximators journal January 1989
Linear-scaling density-functional-theory calculations of electronic structure based on real-space grids: design, analysis, and scalability test of parallel algorithms journal November 2001
Grand canonical molecular dynamics simulations of Cu–Au nanoalloys in thermal equilibrium using reactive ANN potentials journal December 2015
An implementation of artificial neural-network potentials for atomistic materials simulations: Performance for TiO2 journal March 2016
Representing high-dimensional potential-energy surfaces for reactions at surfaces by neural networks journal September 2004
First-Principles Investigation of the Li−Fe−F Phase Diagram and Equilibrium and Nonequilibrium Conversion Reactions of Iron Fluorides with Lithium journal August 2008
A Density-Functional Theory-Based Neural Network Potential for Water Clusters Including van der Waals Corrections journal April 2013
Potential Energy Surfaces Fitted by Artificial Neural Networks journal March 2010
Melting curves of lithium, sodium, potassium, and rubidium to 80 kilobars journal April 1968
Electronic and structural transitions in dense liquid sodium journal September 2007
Cold melting and solid structures of dense lithium journal January 2011
Reducing Dzyaloshinskii-Moriya interaction and field-free spin-orbit torque switching in synthetic antiferromagnets journal May 2021
Ab initio vibrational free energies including anharmonicity for multicomponent alloys journal July 2019
Datasets of seed mucilage traits for Arabidopsis thaliana natural accessions with atypical outer mucilage journal March 2021
Electronic structure of AlFeN films exhibiting crystallographic orientation change from c- to a-axis with Fe concentrations and annealing effect journal February 2020
Atom-centered symmetry functions for constructing high-dimensional neural network potentials journal February 2011
New Monte Carlo method to compute the free energy of arbitrary solids. Application to the fcc and hcp phases of hard spheres journal October 1984
Reversible multiple time scale molecular dynamics journal August 1992
The melting points of MgO up to 4 TPa predicted based on ab initio thermodynamic integration molecular dynamics journal February 2018
A method for the solution of certain non-linear problems in least squares journal January 1944
Density-functional energies and forces with Gaussian-broadened fractional occupations journal May 1994
Projector augmented-wave method journal December 1994
Free-energy calculations in solids from first-principles molecular dynamics: Vacancy formation in sodium journal January 1995
First-order phase transitions by first-principles free-energy calculations: The melting of Al journal April 1998
From ultrasoft pseudopotentials to the projector augmented-wave method journal January 1999
Iron under Earth’s core conditions: Liquid-state thermodynamics and high-pressure melting curve from ab initio calculations journal April 2002
Ab initio melting curve of the fcc phase of aluminum journal May 2002
Exchange-correlation energy and the phase diagram of Si journal November 2003
Descriptions of surface chemical reactions using a neural network representation of the potential-energy surface journal March 2006
Properties of graphite at melting from multilayer thermodynamic integration journal October 2009
Ab initio quality neural-network potential for sodium journal May 2010
High-dimensional neural-network potentials for multicomponent systems: Applications to zinc oxide journal April 2011
Constraints on the phase diagram of molybdenum from first-principles free-energy calculations journal February 2012
Neural network interatomic potential for the phase change material GeTe journal May 2012
Theory of melting at high pressures: Amending density functional theory with quantum Monte Carlo journal October 2014
Improved method of calculating ab initio high-temperature thermodynamic properties with application to ZrC journal June 2015
Efficient and accurate machine-learning interpolation of atomic energies in compositions with many species journal July 2017
Efficient approach to compute melting properties fully from ab initio with application to Cu journal December 2017
Melting properties from ab initio free energy calculations: Iron at the Earth's inner-core boundary journal December 2018
Rocky Core Solubility in Jupiter and Giant Exoplanets journal March 2012
Identifying Structural Flow Defects in Disordered Solids Using Machine-Learning Methods journal March 2015
Ab Initio Molecular Dynamics Study of First-Order Phase Transitions: Melting of Silicon journal March 1995
Generalized Gradient Approximation Made Simple journal October 1996
Melting of Dense Sodium journal May 2005
Carbon Phase Diagram from Ab Initio Molecular Dynamics journal October 2005
Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces journal April 2007
Training feedforward networks with the Marquardt algorithm journal January 1994
An Algorithm for Least-Squares Estimation of Nonlinear Parameters journal June 1963
Effects of Inner-Core 2 p States on Melting Curve and Structure of Dense Sodium at High Pressures journal June 2008
Efficient approach to compute melting properties fully from ab initio with application to Cu collection January 2017
Ab initio quality neural-network potential for sodium text January 2010
Constraints on the phase diagram of molybdenum from first-principles free-energy calculations text January 2011
Theory of melting at high pressures: Amending Density Functional Theory with Quantum Monte Carlo text January 2013
Identifying structural flow defects in disordered solids using machine learning methods text January 2014

Similar Records

Related Subjects

75 CONDENSED MATTER PHYSICS
SUPERCONDUCTIVITY AND SUPERFLUIDITY
Materials Science
Physics