BIGDML-Towards accurate quantum machine learning force fields for materials.
Journal
Nature communications
ISSN: 2041-1723
Titre abrégé: Nat Commun
Pays: England
ID NLM: 101528555
Informations de publication
Date de publication:
29 Jun 2022
29 Jun 2022
Historique:
received:
17
06
2021
accepted:
01
06
2022
entrez:
29
6
2022
pubmed:
30
6
2022
medline:
30
6
2022
Statut:
epublish
Résumé
Machine-learning force fields (MLFF) should be accurate, computationally and data efficient, and applicable to molecules, materials, and interfaces thereof. Currently, MLFFs often introduce tradeoffs that restrict their practical applicability to small subsets of chemical space or require exhaustive datasets for training. Here, we introduce the Bravais-Inspired Gradient-Domain Machine Learning (BIGDML) approach and demonstrate its ability to construct reliable force fields using a training set with just 10-200 geometries for materials including pristine and defect-containing 2D and 3D semiconductors and metals, as well as chemisorbed and physisorbed atomic and molecular adsorbates on surfaces. The BIGDML model employs the full relevant symmetry group for a given material, does not assume artificial atom types or localization of atomic interactions and exhibits high data efficiency and state-of-the-art energy accuracies (errors substantially below 1 meV per atom) for an extended set of materials. Extensive path-integral molecular dynamics carried out with BIGDML models demonstrate the counterintuitive localization of benzene-graphene dynamics induced by nuclear quantum effects and their strong contributions to the hydrogen diffusion coefficient in a Pd crystal for a wide range of temperatures.
Identifiants
pubmed: 35768400
doi: 10.1038/s41467-022-31093-x
pii: 10.1038/s41467-022-31093-x
pmc: PMC9243122
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
3733Informations de copyright
© 2022. The Author(s).
Références
Angew Chem Int Ed Engl. 2017 Oct 9;56(42):12828-12840
pubmed: 28520235
Phys Rev Lett. 2020 Apr 10;124(14):146401
pubmed: 32338971
J Chem Phys. 2018 Jun 28;148(24):241722
pubmed: 29960322
Nat Commun. 2020 Oct 29;11(1):5461
pubmed: 33122630
J Phys Condens Matter. 2009 Sep 30;21(39):395502
pubmed: 21832390
Phys Rev Lett. 2012 Feb 3;108(5):058301
pubmed: 22400967
Chem Rev. 2021 Aug 25;121(16):10142-10186
pubmed: 33705118
Chem Sci. 2017 Apr 1;8(4):3192-3203
pubmed: 28507695
J Chem Theory Comput. 2018 Jun 12;14(6):2991-3003
pubmed: 29750522
J Chem Theory Comput. 2019 Apr 9;15(4):2574-2586
pubmed: 30794393
Nat Commun. 2020 Aug 17;11(1):4125
pubmed: 32807794
J Chem Phys. 2020 Jul 28;153(4):044104
pubmed: 32752705
J Chem Phys. 2020 Jul 21;153(3):034702
pubmed: 32716159
Annu Rev Phys Chem. 2020 Apr 20;71:361-390
pubmed: 32092281
J Chem Theory Comput. 2019 Jun 11;15(6):3678-3693
pubmed: 31042390
J Chem Phys. 2019 Apr 21;150(15):154110
pubmed: 31005079
Sci Adv. 2017 Dec 13;3(12):e1701816
pubmed: 29242828
Chem Commun (Camb). 2016 Aug 16;52(68):10385-8
pubmed: 27480254
Nat Commun. 2021 Jan 15;12(1):398
pubmed: 33452239
Sci Adv. 2020 Sep 2;6(36):
pubmed: 32917594
Acc Chem Res. 2013 Aug 20;46(8):1740-8
pubmed: 23815772
Chemphyschem. 2009 Jan 12;10(1):206-10
pubmed: 18814150
Phys Rev Lett. 2015 Jul 17;115(3):036102
pubmed: 26230805
Nature. 2021 Jan;589(7840):59-64
pubmed: 33408379
Phys Rev Lett. 2007 Apr 6;98(14):146401
pubmed: 17501293
Phys Rev Lett. 2012 Apr 6;108(14):146103
pubmed: 22540809
J Phys Condens Matter. 2017 Oct 24;29(46):465901
pubmed: 29064822
Curr Opin Struct Biol. 2020 Feb;60:77-84
pubmed: 31881449
Nano Lett. 2011 Aug 10;11(8):3227-31
pubmed: 21728349
Nat Commun. 2021 Jan 19;12(1):442
pubmed: 33469007
J Chem Theory Comput. 2020 Jul 14;16(7):4192-4202
pubmed: 32543858
Phys Rev Lett. 2014 Jul 11;113(2):025504
pubmed: 25062206
Phys Chem Chem Phys. 2017 Oct 18;19(40):27374-27383
pubmed: 28972620
ACS Nano. 2020 Jul 28;14(7):7987-7998
pubmed: 32491826
Phys Rev Lett. 2015 Mar 6;114(9):096405
pubmed: 25793835
Nature. 2020 Sep;585(7824):217-220
pubmed: 32908269
J Phys Chem Lett. 2016 Jun 2;7(11):2125-31
pubmed: 27195654
Patterns (N Y). 2020 Nov 12;1(9):100142
pubmed: 33336200
J Phys Chem Lett. 2015 Nov 5;6(21):4233-8
pubmed: 26722963
Proc Natl Acad Sci U S A. 2018 Feb 20;115(8):1724-1729
pubmed: 29432177
Phys Rev B Condens Matter. 1991 Mar 15;43(9):6968-6976
pubmed: 9998159
J Chem Phys. 2018 Jun 28;148(24):241717
pubmed: 29960351
Sci Adv. 2017 May 05;3(5):e1603015
pubmed: 28508076
Science. 2016 Jan 1;351(6268):68-70
pubmed: 26721995
Angew Chem Int Ed Engl. 2018 Apr 9;57(16):4164-4169
pubmed: 29216413
Phys Rev B Condens Matter. 1993 Jul 1;48(1):22-33
pubmed: 10006745
Chem Rev. 2021 Aug 25;121(16):9759-9815
pubmed: 34310133
Phys Rev Lett. 2009 Feb 20;102(7):073005
pubmed: 19257665
Chem Rev. 2021 Aug 25;121(16):9816-9872
pubmed: 34232033
J Mol Model. 2019 Sep 5;25(10):302
pubmed: 31486895
J Chem Phys. 2018 May 28;148(20):204707
pubmed: 29865849
J Chem Phys. 2014 May 14;140(18):18A508
pubmed: 24832316
J Chem Theory Comput. 2021 Dec 14;17(12):7696-7711
pubmed: 34735161
IEEE Trans Neural Netw. 2001;12(2):181-201
pubmed: 18244377
Nat Commun. 2020 Sep 29;11(1):4895
pubmed: 32994393
Phys Rev Lett. 2012 Jun 8;108(23):236402
pubmed: 23003978
J Chem Phys. 2018 Jun 28;148(24):241711
pubmed: 29960321
Nat Commun. 2018 Sep 24;9(1):3887
pubmed: 30250077
J Chem Phys. 2019 Mar 21;150(11):114102
pubmed: 30901990
Phys Rev Lett. 1996 Oct 28;77(18):3865-3868
pubmed: 10062328
Nature. 2018 Jul;559(7715):547-555
pubmed: 30046072
Phys Rev Lett. 2010 Apr 2;104(13):136403
pubmed: 20481899
Nat Commun. 2021 Dec 14;12(1):7273
pubmed: 34907176
Nature. 2013 Jan 17;493(7432):365-70
pubmed: 23254929