DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Deep neural network for the dielectric response of insulators

Abstract

In this work, we introduce a deep neural network to model in a symmetry preserving way the environmental dependence of the centers of the electronic charge. The model learns from ab initio density functional theory, wherein the electronic centers are uniquely assigned by the maximally localized Wannier functions. When combined with the deep potential model of the atomic potential energy surface, the scheme predicts the dielectric response of insulators for trajectories inaccessible to direct ab initio simulation. The scheme is nonperturbative and can capture the response of a mutating chemical environment. We demonstrate the approach by calculating the infrared spectra of liquid water at standard conditions, and of ice under extreme pressure, when it transforms from a molecular to an ionic crystal.

Authors:
ORCiD logo [1];  [2]; ORCiD logo [3]; ORCiD logo [4];  [5]; ORCiD logo [1]
  1. Princeton Univ., NJ (United States)
  2. Peking Univ., Beijing (China)
  3. Temple Univ., Philadelphia, PA (United States)
  4. Institute of Applied Physics and Computational Mathematics, Beijing (China)
  5. Princeton Univ., NJ (United States); Beijing Institute of Big Data Research (China)
Publication Date:
Research Org.:
Princeton Univ., NJ (United States); Univ. of California, Oakland, CA (United States); Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States). National Energy Research Scientific Computing Center (NERSC)
Sponsoring Org.:
USDOE Office of Science (SC); US Department of the Navy, Office of Naval Research (ONR); National Natural Science Foundation of China (NSFC); National Key Research and Development Program of China
OSTI Identifier:
1803785
Grant/Contract Number:  
SC0019394; AC02-05CH11231; N00014-13-1-0338; 11871110; 2016YFB0201200; 2016YFB0201203
Resource Type:
Accepted Manuscript
Journal Name:
Physical Review. B
Additional Journal Information:
Journal Volume: 102; Journal Issue: 4; Journal ID: ISSN 2469-9950
Publisher:
American Physical Society (APS)
Country of Publication:
United States
Language:
English
Subject:
75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND SUPERFLUIDITY; Materials Science; Physics

Citation Formats

Zhang, Linfeng, Chen, Mohan, Wu, Xifan, Wang, Han, E, Weinan, and Car, Roberto. Deep neural network for the dielectric response of insulators. United States: N. p., 2020. Web. doi:10.1103/physrevb.102.041121.
Zhang, Linfeng, Chen, Mohan, Wu, Xifan, Wang, Han, E, Weinan, & Car, Roberto. Deep neural network for the dielectric response of insulators. United States. https://doi.org/10.1103/physrevb.102.041121
Zhang, Linfeng, Chen, Mohan, Wu, Xifan, Wang, Han, E, Weinan, and Car, Roberto. Wed . "Deep neural network for the dielectric response of insulators". United States. https://doi.org/10.1103/physrevb.102.041121. https://www.osti.gov/servlets/purl/1803785.
@article{osti_1803785,
title = {Deep neural network for the dielectric response of insulators},
author = {Zhang, Linfeng and Chen, Mohan and Wu, Xifan and Wang, Han and E, Weinan and Car, Roberto},
abstractNote = {In this work, we introduce a deep neural network to model in a symmetry preserving way the environmental dependence of the centers of the electronic charge. The model learns from ab initio density functional theory, wherein the electronic centers are uniquely assigned by the maximally localized Wannier functions. When combined with the deep potential model of the atomic potential energy surface, the scheme predicts the dielectric response of insulators for trajectories inaccessible to direct ab initio simulation. The scheme is nonperturbative and can capture the response of a mutating chemical environment. We demonstrate the approach by calculating the infrared spectra of liquid water at standard conditions, and of ice under extreme pressure, when it transforms from a molecular to an ionic crystal.},
doi = {10.1103/physrevb.102.041121},
journal = {Physical Review. B},
number = 4,
volume = 102,
place = {United States},
year = {Wed Jul 22 00:00:00 EDT 2020},
month = {Wed Jul 22 00:00:00 EDT 2020}
}

Works referenced in this record:

Melting of ice under pressure
journal, September 2008

  • Schwegler, E.; Sharma, M.; Gygi, F.
  • Proceedings of the National Academy of Sciences, Vol. 105, Issue 39
  • DOI: 10.1073/pnas.0808137105

Isotope effects in liquid water via deep potential molecular dynamics
journal, October 2019


Infrared intensities of liquids. 5. Optical and dielectric constants, integrated intensities, and dipole moment derivatives of water and water-d2 at 22.degree.C
journal, March 1989

  • Bertie, John E.; Ahmed, M. Khalique; Eysel, Hans H.
  • The Journal of Physical Chemistry, Vol. 93, Issue 6
  • DOI: 10.1021/j100343a008

First-principles theory of structural phase transitions for perovskites: Competing instabilities
journal, February 1998


Silicon Liquid Structure and Crystal Nucleation from Ab Initio Deep Metadynamics
journal, December 2018


Maximally localized Wannier functions: Theory and applications
journal, October 2012

  • Marzari, Nicola; Mostofi, Arash A.; Yates, Jonathan R.
  • Reviews of Modern Physics, Vol. 84, Issue 4
  • DOI: 10.1103/RevModPhys.84.1419

Raman Spectra of Liquid Water from Ab Initio Molecular Dynamics: Vibrational Signatures of Charge Fluctuations in the Hydrogen Bond Network
journal, August 2013

  • Wan, Quan; Spanu, Leonardo; Galli, Giulia A.
  • Journal of Chemical Theory and Computation, Vol. 9, Issue 9
  • DOI: 10.1021/ct4005307

Quantum Dynamics and Spectroscopy of Ab Initio Liquid Water: The Interplay of Nuclear and Electronic Quantum Effects
journal, March 2017


Inexpensive modeling of quantum dynamics using path integral generalized Langevin equation thermostats
journal, March 2020

  • Kapil, Venkat; Wilkins, David M.; Lan, Jinggang
  • The Journal of Chemical Physics, Vol. 152, Issue 12
  • DOI: 10.1063/1.5141950

The Structure and Entropy of Ice and of Other Crystals with Some Randomness of Atomic Arrangement
journal, December 1935

  • Pauling, Linus
  • Journal of the American Chemical Society, Vol. 57, Issue 12
  • DOI: 10.1021/ja01315a102

Development of a “First Principles” Water Potential with Flexible Monomers: Dimer Potential Energy Surface, VRT Spectrum, and Second Virial Coefficient
journal, November 2013

  • Babin, Volodymyr; Leforestier, Claude; Paesani, Francesco
  • Journal of Chemical Theory and Computation, Vol. 9, Issue 12
  • DOI: 10.1021/ct400863t

Maximally localized generalized Wannier functions for composite energy bands
journal, November 1997


A Theory of Water and Ionic Solution, with Particular Reference to Hydrogen and Hydroxyl Ions
journal, August 1933

  • Bernal, J. D.; Fowler, R. H.
  • The Journal of Chemical Physics, Vol. 1, Issue 8
  • DOI: 10.1063/1.1749327

Quantum calculations of the IR spectrum of liquid water using ab initio and model potential and dipole moment surfaces and comparison with experiment
journal, May 2015

  • Liu, Hanchao; Wang, Yimin; Bowman, Joel M.
  • The Journal of Chemical Physics, Vol. 142, Issue 19
  • DOI: 10.1063/1.4921045

Enhanced Proton Conductivity in Y-Doped BaZrO 3 via Strain Engineering
journal, October 2017

  • Fluri, Aline; Marcolongo, Aris; Roddatis, Vladimir
  • Advanced Science, Vol. 4, Issue 12
  • DOI: 10.1002/advs.201700467

The phase diagram of high-pressure superionic ice
journal, August 2015

  • Sun, Jiming; Clark, Bryan K.; Torquato, Salvatore
  • Nature Communications, Vol. 6, Issue 1
  • DOI: 10.1038/ncomms9156

Machine learning of molecular electronic properties in chemical compound space
journal, September 2013


Generalized variational density functional perturbation theory
journal, November 2000

  • Putrino, Anna; Sebastiani, Daniel; Parrinello, Michele
  • The Journal of Chemical Physics, Vol. 113, Issue 17
  • DOI: 10.1063/1.1312830

Nearsightedness of electronic matter
journal, August 2005

  • Prodan, E.; Kohn, W.
  • Proceedings of the National Academy of Sciences, Vol. 102, Issue 33
  • DOI: 10.1073/pnas.0505436102

DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics
journal, July 2018


Transferable Machine-Learning Model of the Electron Density
journal, December 2018


Hydroxide diffuses slower than hydronium in water because its solvated structure inhibits correlated proton transfer
journal, March 2018


Using Gaussian process regression to simulate the vibrational Raman spectra of molecular crystals
journal, October 2019

  • Raimbault, Nathaniel; Grisafi, Andrea; Ceriotti, Michele
  • New Journal of Physics, Vol. 21, Issue 10
  • DOI: 10.1088/1367-2630/ab4509

Unified Approach for Molecular Dynamics and Density-Functional Theory
journal, November 1985


Solving the electronic structure problem with machine learning
journal, February 2019

  • Chandrasekaran, Anand; Kamal, Deepak; Batra, Rohit
  • npj Computational Materials, Vol. 5, Issue 1
  • DOI: 10.1038/s41524-019-0162-7

Ab initio spectroscopy and ionic conductivity of water under Earth mantle conditions
journal, June 2018

  • Rozsa, Viktor; Pan, Ding; Giberti, Federico
  • Proceedings of the National Academy of Sciences, Vol. 115, Issue 27
  • DOI: 10.1073/pnas.1800123115

Deep Potential: A General Representation of a Many-Body Potential Energy Surface
journal, January 2018

  • Han, Jiequn; Zhang, Linfeng; Car, Roberto
  • Communications in Computational Physics, Vol. 23, Issue 3
  • DOI: 10.4208/cicp.OA-2017-0213

A unified formulation of the constant temperature molecular dynamics methods
journal, July 1984

  • Nosé, Shuichi
  • The Journal of Chemical Physics, Vol. 81, Issue 1
  • DOI: 10.1063/1.447334

Macroscopic polarization in crystalline dielectrics: the geometric phase approach
journal, July 1994


Vibrational Sum-Frequency Generation Spectroscopy at the Water/Lipid Interface: Molecular Dynamics Simulation Study
journal, May 2010

  • Nagata, Yuki; Mukamel, Shaul
  • Journal of the American Chemical Society, Vol. 132, Issue 18
  • DOI: 10.1021/ja100508n

Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
journal, January 2012


Ab Initio Infrared Absorption Study of the Hydrogen-Bond Symmetrization in Ice
journal, August 1998


Infrared and Raman Spectroscopy of Liquid Water through “First-Principles” Many-Body Molecular Dynamics
journal, February 2015

  • Medders, Gregory R.; Paesani, Francesco
  • Journal of Chemical Theory and Computation, Vol. 11, Issue 3
  • DOI: 10.1021/ct501131j

Toward reliable density functional methods without adjustable parameters: The PBE0 model
journal, April 1999

  • Adamo, Carlo; Barone, Vincenzo
  • The Journal of Chemical Physics, Vol. 110, Issue 13
  • DOI: 10.1063/1.478522

Raman spectrum and polarizability of liquid water from deep neural networks
journal, January 2020

  • Sommers, Grace M.; Calegari Andrade, Marcos F.; Zhang, Linfeng
  • Physical Chemistry Chemical Physics, Vol. 22, Issue 19
  • DOI: 10.1039/D0CP01893G

DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models
journal, February 2020


Reinforced dynamics for enhanced sampling in large atomic and molecular systems
journal, March 2018

  • Zhang, Linfeng; Wang, Han; E., Weinan
  • The Journal of Chemical Physics, Vol. 148, Issue 12
  • DOI: 10.1063/1.5019675

Canonical dynamics: Equilibrium phase-space distributions
journal, March 1985


Accurate molecular polarizabilities with coupled cluster theory and machine learning
journal, February 2019

  • Wilkins, David M.; Grisafi, Andrea; Yang, Yang
  • Proceedings of the National Academy of Sciences, Vol. 116, Issue 9
  • DOI: 10.1073/pnas.1816132116

Proton dynamics and the phase diagram of dense water ice
journal, June 2018

  • Hernandez, J. -A.; Caracas, R.
  • The Journal of Chemical Physics, Vol. 148, Issue 21
  • DOI: 10.1063/1.5028389

Machine learning molecular dynamics for the simulation of infrared spectra
journal, January 2017

  • Gastegger, Michael; Behler, Jörg; Marquetand, Philipp
  • Chemical Science, Vol. 8, Issue 10
  • DOI: 10.1039/C7SC02267K

Accurate ab initio and “hybrid” potential energy surfaces, intramolecular vibrational energies, and classical ir spectrum of the water dimer
journal, April 2009

  • Shank, Alex; Wang, Yimin; Kaledin, Alexey
  • The Journal of Chemical Physics, Vol. 130, Issue 14
  • DOI: 10.1063/1.3112403

Compression of Ice to 210 Gigapascals: Infrared Evidence for a Symmetric Hydrogen-Bonded Phase
journal, July 1996


Machine Learning Force Fields: Construction, Validation, and Outlook
journal, December 2016


Density Functional and Density Matrix Method Scaling Linearly with the Number of Atoms
journal, April 1996


Nanosecond X-ray diffraction of shock-compressed superionic water ice
journal, May 2019


First-Principles Framework to Compute Sum-Frequency Generation Vibrational Spectra of Semiconductors and Insulators
journal, December 2015


Machine learning of accurate energy-conserving molecular force fields
journal, May 2017

  • Chmiela, Stefan; Tkatchenko, Alexandre; Sauceda, Huziel E.
  • Science Advances, Vol. 3, Issue 5
  • DOI: 10.1126/sciadv.1603015

Active learning of uniformly accurate interatomic potentials for materials simulation
journal, February 2019


Intermolecular Dynamical Charge Fluctuations in Water: A Signature of the H-Bond Network
journal, October 2005


Active learning of linearly parametrized interatomic potentials
journal, December 2017


Dynamical Structure, Bonding, and Thermodynamics of the Superionic Sublattice in α AgI
journal, October 2006


Works referencing / citing this record:

86 PFLOPS Deep Potential Molecular Dynamics simulation of 100 million atoms with ab initio accuracy
journal, February 2021