Physical Review B, volume 99, issue 6, publication number 064114

Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning

Publication typeJournal Article
Publication date2019-02-27
Quartile SCImago
Q1
Quartile WOS
Q2
Impact factor3.7
ISSN24699950, 24699969, 10980121, 1550235X
Abstract
We propose a methodology for crystal structure prediction that is based on the evolutionary algorithm USPEX and the machine-learning interatomic potentials actively learning on-the-fly. Our methodology allows for an automated construction of an interatomic interaction model from scratch, replacing the expensive density functional theory (DFT) and giving a speedup of several orders of magnitude. Predicted low-energy structures are then tested on DFT, ensuring that our machine-learning model does not introduce any prediction error. We tested our methodology on prediction of crystal structures of carbon, high-pressure phases of sodium, and boron allotropes, including those that have more than 100 atoms in the primitive cell. All the the main allotropes have been reproduced, and a hitherto unknown 54-atom structure of boron has been predicted with very modest computational effort.

Citations by journals

5
10
15
20
25
Journal of Chemical Physics
Journal of Chemical Physics, 21, 9.46%
Journal of Chemical Physics
21 publications, 9.46%
npj Computational Materials
npj Computational Materials, 11, 4.95%
npj Computational Materials
11 publications, 4.95%
Physical Review B
Physical Review B, 9, 4.05%
Physical Review B
9 publications, 4.05%
Physical Review Materials
Physical Review Materials, 7, 3.15%
Physical Review Materials
7 publications, 3.15%
Journal of Physical Chemistry Letters
Journal of Physical Chemistry Letters, 7, 3.15%
Journal of Physical Chemistry Letters
7 publications, 3.15%
Machine Learning Science and Technology
Machine Learning Science and Technology, 6, 2.7%
Machine Learning Science and Technology
6 publications, 2.7%
Computational Materials Science
Computational Materials Science, 5, 2.25%
Computational Materials Science
5 publications, 2.25%
Journal of Physics Condensed Matter
Journal of Physics Condensed Matter, 4, 1.8%
Journal of Physics Condensed Matter
4 publications, 1.8%
Nature Communications
Nature Communications, 4, 1.8%
Nature Communications
4 publications, 1.8%
Chemical Reviews
Chemical Reviews, 4, 1.8%
Chemical Reviews
4 publications, 1.8%
Physical Chemistry Chemical Physics
Physical Chemistry Chemical Physics, 4, 1.8%
Physical Chemistry Chemical Physics
4 publications, 1.8%
Chemical Science
Chemical Science, 4, 1.8%
Chemical Science
4 publications, 1.8%
Journal of Physical Chemistry A
Journal of Physical Chemistry A, 3, 1.35%
Journal of Physical Chemistry A
3 publications, 1.35%
Journal of Chemical Theory and Computation
Journal of Chemical Theory and Computation, 3, 1.35%
Journal of Chemical Theory and Computation
3 publications, 1.35%
Crystals
Crystals, 3, 1.35%
Crystals
3 publications, 1.35%
Scientific Reports
Scientific Reports, 3, 1.35%
Scientific Reports
3 publications, 1.35%
MRS Bulletin
MRS Bulletin, 3, 1.35%
MRS Bulletin
3 publications, 1.35%
Acta Materialia
Acta Materialia, 3, 1.35%
Acta Materialia
3 publications, 1.35%
Journal of the American Chemical Society
Journal of the American Chemical Society, 3, 1.35%
Journal of the American Chemical Society
3 publications, 1.35%
Journal of Physical Chemistry C
Journal of Physical Chemistry C, 3, 1.35%
Journal of Physical Chemistry C
3 publications, 1.35%
Journal of Chemical Information and Modeling
Journal of Chemical Information and Modeling, 3, 1.35%
Journal of Chemical Information and Modeling
3 publications, 1.35%
Science and Technology of Advanced Materials Methods
Science and Technology of Advanced Materials Methods, 3, 1.35%
Science and Technology of Advanced Materials Methods
3 publications, 1.35%
Nature Reviews Materials
Nature Reviews Materials, 2, 0.9%
Nature Reviews Materials
2 publications, 0.9%
Journal of Applied Physics
Journal of Applied Physics, 2, 0.9%
Journal of Applied Physics
2 publications, 0.9%
Physical Review Letters
Physical Review Letters, 2, 0.9%
Physical Review Letters
2 publications, 0.9%
Molecules
Molecules, 2, 0.9%
Molecules
2 publications, 0.9%
ACS Catalysis
ACS Catalysis, 2, 0.9%
ACS Catalysis
2 publications, 0.9%
Materials Horizons
Materials Horizons, 2, 0.9%
Materials Horizons
2 publications, 0.9%
Applied Surface Science
Applied Surface Science, 2, 0.9%
Applied Surface Science
2 publications, 0.9%
5
10
15
20
25

Citations by publishers

5
10
15
20
25
30
35
40
Elsevier
Elsevier, 39, 17.57%
Elsevier
39 publications, 17.57%
American Chemical Society (ACS)
American Chemical Society (ACS), 35, 15.77%
American Chemical Society (ACS)
35 publications, 15.77%
Springer Nature
Springer Nature, 30, 13.51%
Springer Nature
30 publications, 13.51%
American Institute of Physics (AIP)
American Institute of Physics (AIP), 26, 11.71%
American Institute of Physics (AIP)
26 publications, 11.71%
American Physical Society (APS)
American Physical Society (APS), 18, 8.11%
American Physical Society (APS)
18 publications, 8.11%
IOP Publishing
IOP Publishing, 16, 7.21%
IOP Publishing
16 publications, 7.21%
Royal Society of Chemistry (RSC)
Royal Society of Chemistry (RSC), 13, 5.86%
Royal Society of Chemistry (RSC)
13 publications, 5.86%
Wiley
Wiley, 10, 4.5%
Wiley
10 publications, 4.5%
Multidisciplinary Digital Publishing Institute (MDPI)
Multidisciplinary Digital Publishing Institute (MDPI), 9, 4.05%
Multidisciplinary Digital Publishing Institute (MDPI)
9 publications, 4.05%
Cambridge University Press
Cambridge University Press, 6, 2.7%
Cambridge University Press
6 publications, 2.7%
Taylor & Francis
Taylor & Francis, 3, 1.35%
Taylor & Francis
3 publications, 1.35%
Ceramic Society of Japan
Ceramic Society of Japan, 2, 0.9%
Ceramic Society of Japan
2 publications, 0.9%
World Scientific
World Scientific, 1, 0.45%
World Scientific
1 publication, 0.45%
Frontiers Media S.A.
Frontiers Media S.A., 1, 0.45%
Frontiers Media S.A.
1 publication, 0.45%
Tech Science Press, 1, 0.45%
Tech Science Press
1 publication, 0.45%
Materials Research Society
Materials Research Society, 1, 0.45%
Materials Research Society
1 publication, 0.45%
Public Library of Science (PLoS)
Public Library of Science (PLoS), 1, 0.45%
Public Library of Science (PLoS)
1 publication, 0.45%
IEEE
IEEE, 1, 0.45%
IEEE
1 publication, 0.45%
Hindawi Limited
Hindawi Limited, 1, 0.45%
Hindawi Limited
1 publication, 0.45%
Oxford University Press
Oxford University Press, 1, 0.45%
Oxford University Press
1 publication, 0.45%
Proceedings of the National Academy of Sciences (PNAS)
Proceedings of the National Academy of Sciences (PNAS), 1, 0.45%
Proceedings of the National Academy of Sciences (PNAS)
1 publication, 0.45%
5
10
15
20
25
30
35
40
  • We do not take into account publications that without a DOI.
  • Statistics recalculated only for publications connected to researchers, organizations and labs registered on the platform.
  • Statistics recalculated weekly.
Metrics
Share
Cite this
GOST |
Cite this
GOST Copy
Podryabinkin E. V. et al. Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning // Physical Review B. 2019. Vol. 99. No. 6. 064114
GOST all authors (up to 50) Copy
Podryabinkin E. V., Tikhonov E. V., Shapeev A. V., Oganov A. R. Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning // Physical Review B. 2019. Vol. 99. No. 6. 064114
RIS |
Cite this
RIS Copy
TY - JOUR
DO - 10.1103/PhysRevB.99.064114
UR - https://doi.org/10.1103%2FPhysRevB.99.064114
TI - Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning
T2 - Physical Review B
AU - Podryabinkin, Evgeny V.
AU - Tikhonov, Evgeny V.
AU - Shapeev, Alexander V.
AU - Oganov, A. R.
PY - 2019
DA - 2019/02/27 00:00:00
PB - American Physical Society (APS)
IS - 6
VL - 99
SN - 2469-9950
SN - 2469-9969
SN - 1098-0121
SN - 1550-235X
ER -
BibTex
Cite this
BibTex Copy
@article{2019_Podryabinkin
author = {Evgeny V. Podryabinkin and Evgeny V. Tikhonov and Alexander V. Shapeev and A. R. Oganov},
title = {Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning},
journal = {Physical Review B},
year = {2019},
volume = {99},
publisher = {American Physical Society (APS)},
month = {feb},
url = {https://doi.org/10.1103%2FPhysRevB.99.064114},
number = {6},
doi = {10.1103/PhysRevB.99.064114}
}
Found error?