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

Title: Accelerated computation of lattice thermal conductivity using neural network interatomic potentials

Abstract

We report with the development of the density functional theory (DFT) and ever-increasing computational capacity, an accurate prediction of lattice thermal conductivity based on the Boltzmann transport theory becomes computationally feasible, contributing to a fundamental understanding of thermal conductivity as well as a choice of the optimal materials for specific applications. However, steep costs in evaluating third-order force constants limit the theoretical investigation to crystals with high symmetry and few atoms in the unit cell. Currently, machine learning potentials are garnering attention as a computationally efficient high-fidelity model of DFT, and several studies have demonstrated that the lattice thermal conductivity could be computed accurately via the machine learning potentials. However, test materials were mostly crystals with high symmetries, and the applicability of machine learning potentials to a wide range of materials has yet to be demonstrated. Furthermore, establishing a standard training set that provides consistent accuracy and computational efficiencies across a wide range of materials would be useful. To address these issues, herein we compute lattice thermal conductivities at 300 K using neural network interatomic potentials. As test materials, we select 25 materials with diverse symmetries and a wide range of lattice thermal conductivities between 10-1 and 103 Wm-1K-1. Amongmore » various choices of training sets, we find that molecular dynamics trajectories at 50–700 K consistently provide results at par with DFT for the test materials. In contrast to pure DFT approaches, the computational cost in the present approach is uniform over the test materials, yielding a speed gain of 2–10 folds. When a smaller reduced training set is used, the relative efficiency increases by up to ~50 folds without sacrificing accuracy significantly. The current work will broaden the application scope of machine learning potentials by establishing a robust framework for accurately computing lattice thermal conductivity with machine learning potentials.« less

Authors:
 [1];  [1];  [1];  [1];  [2];  [1]
+ Show Author Affiliations
  1. Seoul National University (Korea, Republic of)
  2. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Publication Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA); Korea Institute of Ceramic Engineering and Technology (KICET); National Research Foundation of Korea (NRF); Ministry of Science and ICT
OSTI Identifier:
1892139
Alternate Identifier(s):
OSTI ID: 1869471
Report Number(s):
LLNL-JRNL-831802
Journal ID: ISSN 0927-0256; 1048754
Grant/Contract Number:  
AC52-07NA27344; N0002599; 2017M3D1A1040689
Resource Type:
Accepted Manuscript
Journal Name:
Computational Materials Science
Additional Journal Information:
Journal Volume: 211; Journal Issue: N/A; Journal ID: ISSN 0927-0256
Publisher:
Elsevier
Country of Publication:
United States
Language:
English
Subject:
36 MATERIALS SCIENCE; machine learning potential; lattice thermal conductivity; phonon Boltzmann transport equation; accelerated computation

Citation Formats

Choi, Jeong Min, Lee, Kyeongpung, Kim, Sangtae, Moon, Minseok, Jeong, Wonseok, and Han, Seungwu. Accelerated computation of lattice thermal conductivity using neural network interatomic potentials. United States: N. p., 2022. Web. doi:10.1016/j.commatsci.2022.111472.
Copy to clipboard
Choi, Jeong Min, Lee, Kyeongpung, Kim, Sangtae, Moon, Minseok, Jeong, Wonseok, & Han, Seungwu. Accelerated computation of lattice thermal conductivity using neural network interatomic potentials. United States. https://doi.org/10.1016/j.commatsci.2022.111472
Copy to clipboard
Choi, Jeong Min, Lee, Kyeongpung, Kim, Sangtae, Moon, Minseok, Jeong, Wonseok, and Han, Seungwu. Fri . "Accelerated computation of lattice thermal conductivity using neural network interatomic potentials". United States. https://doi.org/10.1016/j.commatsci.2022.111472. https://www.osti.gov/servlets/purl/1892139.
Copy to clipboard
@article{osti_1892139,
title = {Accelerated computation of lattice thermal conductivity using neural network interatomic potentials},
author = {Choi, Jeong Min and Lee, Kyeongpung and Kim, Sangtae and Moon, Minseok and Jeong, Wonseok and Han, Seungwu},
abstractNote = {We report with the development of the density functional theory (DFT) and ever-increasing computational capacity, an accurate prediction of lattice thermal conductivity based on the Boltzmann transport theory becomes computationally feasible, contributing to a fundamental understanding of thermal conductivity as well as a choice of the optimal materials for specific applications. However, steep costs in evaluating third-order force constants limit the theoretical investigation to crystals with high symmetry and few atoms in the unit cell. Currently, machine learning potentials are garnering attention as a computationally efficient high-fidelity model of DFT, and several studies have demonstrated that the lattice thermal conductivity could be computed accurately via the machine learning potentials. However, test materials were mostly crystals with high symmetries, and the applicability of machine learning potentials to a wide range of materials has yet to be demonstrated. Furthermore, establishing a standard training set that provides consistent accuracy and computational efficiencies across a wide range of materials would be useful. To address these issues, herein we compute lattice thermal conductivities at 300 K using neural network interatomic potentials. As test materials, we select 25 materials with diverse symmetries and a wide range of lattice thermal conductivities between 10-1 and 103 Wm-1K-1. Among various choices of training sets, we find that molecular dynamics trajectories at 50–700 K consistently provide results at par with DFT for the test materials. In contrast to pure DFT approaches, the computational cost in the present approach is uniform over the test materials, yielding a speed gain of 2–10 folds. When a smaller reduced training set is used, the relative efficiency increases by up to ~50 folds without sacrificing accuracy significantly. The current work will broaden the application scope of machine learning potentials by establishing a robust framework for accurately computing lattice thermal conductivity with machine learning potentials.},
doi = {10.1016/j.commatsci.2022.111472},
journal = {Computational Materials Science},
number = N/A,
volume = 211,
place = {United States},
year = {Fri May 13 00:00:00 EDT 2022},
month = {Fri May 13 00:00:00 EDT 2022}
}
Copy to clipboard
Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record
https://doi.org/10.1016/j.commatsci.2022.111472
Copyright Statement
Other availability
Search WorldCat Search WorldCat to find libraries that may hold this journal
Save / Share:
Export Metadata
Save to My Library
You must Sign In or Create an Account in order to save documents to your library.

Works referenced in this record:

A deep neural network interatomic potential for studying thermal conductivity of β -Ga 2 O 3
journal, October 2020

  • Li, Ruiyang; Liu, Zeyu; Rohskopf, Andrew
  • Applied Physics Letters, Vol. 117, Issue 15
  • DOI: 10.1063/5.0025051

Combining phonon accuracy with high transferability in Gaussian approximation potential models
journal, July 2020

  • George, Janine; Hautier, Geoffroy; Bartók, Albert P.
  • The Journal of Chemical Physics, Vol. 153, Issue 4
  • DOI: 10.1063/5.0013826

Review of current high-ZT thermoelectric materials
journal, June 2020

Generalized Gradient Approximation Made Simple
journal, October 1996

  • Perdew, John P.; Burke, Kieron; Ernzerhof, Matthias
  • Physical Review Letters, Vol. 77, Issue 18, p. 3865-3868
  • DOI: 10.1103/PhysRevLett.77.3865

Unusual high thermal conductivity in boron arsenide bulk crystals
journal, July 2018

Gallium phosphide as a new material for anodically bonded atomic sensors
journal, August 2014

  • Dural, Nezih; Romalis, Michael V.
  • APL Materials, Vol. 2, Issue 8
  • DOI: 10.1063/1.4891375

Accelerating first-principles estimation of thermal conductivity by machine-learning interatomic potentials: A MTP/ShengBTE solution
journal, January 2021

  • Mortazavi, Bohayra; Podryabinkin, Evgeny V.; Novikov, Ivan S.
  • Computer Physics Communications, Vol. 258
  • DOI: 10.1016/j.cpc.2020.107583

Properties of single crystalline semiconducting CoSb 3
journal, October 1996

  • Caillat, T.; Borshchevsky, A.; Fleurial, J. ‐P.
  • Journal of Applied Physics, Vol. 80, Issue 8
  • DOI: 10.1063/1.363405

From ultrasoft pseudopotentials to the projector augmented-wave method
journal, January 1999

Compressive sensing lattice dynamics. I. General formalism
journal, November 2019

a-TDEP: Temperature Dependent Effective Potential for Abinit – Lattice dynamic properties including anharmonicity
journal, September 2020

  • Bottin, François; Bieder, Jordan; Bouchet, Johann
  • Computer Physics Communications, Vol. 254
  • DOI: 10.1016/j.cpc.2020.107301

Prediction of Low-Thermal-Conductivity Compounds with First-Principles Anharmonic Lattice-Dynamics Calculations and Bayesian Optimization
journal, November 2015

VESTA 3 for three-dimensional visualization of crystal, volumetric and morphology data
journal, October 2011

Neural network potential for studying the thermal conductivity of Sn
journal, December 2021

Distributions of phonon lifetimes in Brillouin zones
journal, March 2015

Computationally guided discovery of thermoelectric materials
journal, August 2017

Machine learning interatomic potential developed for molecular simulations on thermal properties of β-Ga 2 O 3
journal, October 2020

  • Liu, Yuan-Bin; Yang, Jia-Yue; Xin, Gong-Ming
  • The Journal of Chemical Physics, Vol. 153, Issue 14
  • DOI: 10.1063/5.0027643

Simulating lattice thermal conductivity in semiconducting materials using high-dimensional neural network potential
journal, August 2019

  • Minamitani, Emi; Ogura, Masayoshi; Watanabe, Satoshi
  • Applied Physics Express, Vol. 12, Issue 9
  • DOI: 10.7567/1882-0786/ab36bc

Moment Tensor Potentials: A Class of Systematically Improvable Interatomic Potentials
journal, January 2016

  • Shapeev, Alexander V.
  • Multiscale Modeling & Simulation, Vol. 14, Issue 3
  • DOI: 10.1137/15M1054183

High Performance Thermoelectric Tl 9 BiTe 6 with an Extremely Low Thermal Conductivity
journal, May 2001

Inorganic thermoelectric materials: A review
journal, March 2020

  • Hasan, Md. Nazibul; Wahid, Herman; Nayan, Nafarizal
  • International Journal of Energy Research, Vol. 44, Issue 8
  • DOI: 10.1002/er.5313

Gaussian approximation potential for studying the thermal conductivity of silicene
journal, September 2019

  • Zhang, Cunzhi; Sun, Qiang
  • Journal of Applied Physics, Vol. 126, Issue 10
  • DOI: 10.1063/1.5119281

Electrons and Phonons
journal, November 1961

  • Ziman, J. M.; Levy, Paul W.
  • Physics Today, Vol. 14, Issue 11
  • DOI: 10.1063/1.3057244

The Inorganic Crystal Structure Database (ICSD)—Present and Future
journal, January 2004

A scattering rate model for accelerated evaluation of lattice thermal conductivity bypassing anharmonic force constants
journal, May 2019

  • Xie, Han; Yan, Jiahao; Gu, Xiaokun
  • Journal of Applied Physics, Vol. 125, Issue 20
  • DOI: 10.1063/1.5091504

Thermal barrier coating materials
journal, June 2005

Atom-centered symmetry functions for constructing high-dimensional neural network potentials
journal, February 2011

  • Behler, Jörg
  • The Journal of Chemical Physics, Vol. 134, Issue 7
  • DOI: 10.1063/1.3553717

High Thermal Conductivity of Wurtzite Boron Arsenide Predicted by Including Four-Phonon Scattering with Machine Learning Potential
journal, August 2021

  • Liu, Zhichao; Yang, Xiaolong; Zhang, Bo
  • ACS Applied Materials & Interfaces, Vol. 13, Issue 45
  • DOI: 10.1021/acsami.1c11595

Review of thermal conductivity in composites: Mechanisms, parameters and theory
journal, October 2016

ShengBTE: A solver of the Boltzmann transport equation for phonons
journal, June 2014

  • Li, Wu; Carrete, Jesús; A. Katcho, Nebil
  • Computer Physics Communications, Vol. 185, Issue 6
  • DOI: 10.1016/j.cpc.2014.02.015

Crystallization of amorphous GeTe simulated by neural network potential addressing medium-range order
journal, August 2020

High-temperature phonon transport properties of SnSe from machine-learning interatomic potential
journal, July 2021

Thermal conductivity of diamond nanowires from first principles
journal, May 2012

First principles calculation of lattice thermal conductivity of metals considering phonon-phonon and phonon-electron scattering
journal, June 2016

  • Wang, Yan; Lu, Zexi; Ruan, Xiulin
  • Journal of Applied Physics, Vol. 119, Issue 22
  • DOI: 10.1063/1.4953366

Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set
journal, July 1996

Material descriptors for predicting thermoelectric performance
journal, January 2015

  • Yan, Jun; Gorai, Prashun; Ortiz, Brenden
  • Energy & Environmental Science, Vol. 8, Issue 3
  • DOI: 10.1039/C4EE03157A

Ab initio molecular dynamics for liquid metals
journal, December 1995

Accessing thermal conductivity of complex compounds by machine learning interatomic potentials
journal, October 2019

Temperature dependent effective potential method for accurate free energy calculations of solids
journal, March 2013

Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set
journal, October 1996

First principles phonon calculations in materials science
journal, November 2015

First-Principles Calculations of Vibrational Lifetimes and Decay Channels: Hydrogen-Related Modes in Si
journal, March 2006

Transferability of neural network potentials for varying stoichiometry: Phonons and thermal conductivity of Mn x Ge y compounds
journal, June 2020

  • Mangold, Claudia; Chen, Shunda; Barbalinardo, Giuseppe
  • Journal of Applied Physics, Vol. 127, Issue 24
  • DOI: 10.1063/5.0009550

Nonperturbative phonon scatterings and the two-channel thermal transport in Tl 3 VSe 4
journal, June 2021

Capturing Anharmonicity in a Lattice Thermal Conductivity Model for High-Throughput Predictions
journal, November 2016

Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces
journal, April 2007

First-Principles Determination of Ultrahigh Thermal Conductivity of Boron Arsenide: A Competitor for Diamond?
journal, July 2013

Rapid Prediction of Anisotropic Lattice Thermal Conductivity: Application to Layered Materials
journal, February 2019

Thermal diffusivity measurement of rock-forming minerals from 300° to 1100°K
journal, January 1968

  • Kanamori, Hiroo; Fujii, Naoyuki; Mizutani, Hitoshi
  • Journal of Geophysical Research, Vol. 73, Issue 2
  • DOI: 10.1029/JB073i002p00595

Lattice dynamics of anharmonic solids from first principles
journal, November 2011

The Hiphive Package for the Extraction of High‐Order Force Constants by Machine Learning
journal, February 2019

  • Eriksson, Fredrik; Fransson, Erik; Erhart, Paul
  • Advanced Theory and Simulations, Vol. 2, Issue 5
  • DOI: 10.1002/adts.201800184

Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons
journal, April 2010

Anisotropic thermal conductivity in single crystal β-gallium oxide
journal, March 2015

  • Guo, Zhi; Verma, Amit; Wu, Xufei
  • Applied Physics Letters, Vol. 106, Issue 11
  • DOI: 10.1063/1.4916078

Phonon-engineered extreme thermal conductivity materials
journal, March 2021

SIMPLE-NN: An efficient package for training and executing neural-network interatomic potentials
journal, September 2019

Thermal Conductivity of Silicate Liquid Determined by Machine Learning Potentials
journal, September 2021

  • Deng, Jie; Stixrude, Lars
  • Geophysical Research Letters, Vol. 48, Issue 17
  • DOI: 10.1029/2021GL093806
    Sort by:
    [ × clear filter / sort ]

    Similar Records in DOE PAGES and OSTI.GOV collections: