journal article·2023-10-26

Quantum-Accurate Modeling of Ferroelectric Phase Transition in Perovskites from Message-Passing Neural Networks

Xinjian Ouyang, Yuan Zhuang, Jiale Zhang, Feng Zhang, Xiao Jie, Weijia Chen, Yanxing Zhang, Laijun Liu, Dawei Wang

The Journal of Physical Chemistry C· Volume 127· Issue 42
Publisher page

Abstract

Graph-based message-passing neural networks (MPNNs) have been proposed to facilitate computational research on materials at the atomic scale, which represent the chemical structure as an indirect graph and incorporate the message-passing scheme to learn the interaction between atoms. Here, we employ the MPNN framework to investigate the temperature-dependent structural phase transitions of perovskites. We take two prototypical ferroelectric perovskites, BaTiO 3 and PbTiO 3, as examples to demonstrate the application of this approach. Our results show that well-trained MPNN models achieve a similar level of accuracy as density functional theory (DFT) calculations in terms of both energy and force, with an accuracy of a few meV per atom. This level of accuracy fulfills the requirement for investigating structural changes. By integrating MPNN models as calculators in molecular dynamics simulations, we investigated the structural phase transitions of the two compounds and reproduced their phase transition sequences. The simulated phase transition temperatures and lattice parameters are comparable to the experimental or DFT results. Moreover, we examined the influence of exchange–correlation functionals on the trained MPNN models. This study demonstrates that MPNN, which can serve as a universal energy model, presents an appealing approach to treating the structural properties of perovskites.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Machine Learning in Materials Science · Electronic and Structural Properties of Oxides · Ferroelectric and Negative Capacitance Devices

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

School of Microelectronics & State Key Laboratory for Mechanical Behavior of Materials, Xi’an Jiaotong University, Xi’an 710049, China

Key Laboratory of Micro-Nano Electronics and System Integration of Xi’an City, Xi’an Jiaotong University, Xi’an 710049, China

School of Physics, Henan Normal University, Xinxiang 453007, China

Key Laboratory of Nonferrous Materials and New Processing Technology, Ministry of Education, College of Materials Science and Engineering, Guilin University of Technology, Guilin 541004, China

Funding

Natural Science Foundation of Guangxi Province · National Natural Science Foundation of China

References (74)

Show source-provided references
1. 10.1103/PhysRevB.52.630110.1103/PhysRevB.52.6301 ↗
2. 10.1103/PhysRevLett.73.186110.1103/PhysRevLett.73.1861 ↗
3. 10.1080/0015019049045488210.1080/00150190490454882 ↗
4. 10.1103/PhysRevB.96.05411610.1103/PhysRevB.96.054116 ↗
5. 10.1080/0015019060069575010.1080/00150190600695750 ↗
6. 10.1063/1.364637710.1063/1.3646377 ↗
7. 10.1103/PhysRevB.100.17410910.1103/PhysRevB.100.174109 ↗
8. 10.1103/PhysRevB.88.10410210.1103/PhysRevB.88.104102 ↗
9. 10.1103/PhysRevB.94.13430810.1103/PhysRevB.94.134308 ↗
10. 10.1039/C9CP02955A10.1039/C9CP02955A ↗
11. 10.1103/PhysRevLett.107.17550210.1103/PhysRevLett.107.175502 ↗
12. 10.1103/PhysRevLett.109.06720310.1103/PhysRevLett.109.067203 ↗
13. 10.1038/ncomms610010.1038/ncomms6100 ↗
14. 10.1038/ncomms1101410.1038/ncomms11014 ↗
15. 10.1063/1.505170310.1063/1.5051703 ↗
16. 10.1088/1674-1056/27/12/12770210.1088/1674-1056/27/12/127702 ↗
17. 10.7498/aps.65.23710110.7498/aps.65.237101 ↗
18. 10.1103/PhysRevLett.104.13640310.1103/PhysRevLett.104.136403 ↗
19. 10.1140/epjb/s10051-021-00156-110.1140/epjb/s10051-021-00156-1 ↗
20. 10.1038/s41467-020-20427-210.1038/s41467-020-20427-2 ↗
21. 10.1103/PhysRevB.105.09411610.1103/PhysRevB.105.094116 ↗
22. 10.1063/1.496619210.1063/1.4966192 ↗
23. Batatia, I.; Batzner, S.; Kovács, D. P.; Musaelian, A.; Simm, G. N. C.; Drautz, R.; Ortner, C.; Kozinsky, B.; Csányi, G. Design space of E(3)-equivariant atom-centered interatomic potentials, arXiv:2205.06643, 2022.
24. 10.1063/1.512633610.1063/1.5126336 ↗
25. 10.1038/s41467-022-29939-510.1038/s41467-022-29939-5 ↗
26. Gilmer, J.; Schoenholz, S. S.; Riley, P. F.; Vinyals, O.; Dahl, G. E.In Neural Message Passing for Quantum Chemistry, Proceedings of the 34th International Conference on Machine Learning, 2017; pp 1263–1272.
27. 10.1063/1.501977910.1063/1.5019779 ↗
28. 10.1103/PhysRevLett.120.14530110.1103/PhysRevLett.120.145301 ↗
29. 10.1103/PhysRevLett.120.14300110.1103/PhysRevLett.120.143001 ↗
30. 10.1103/PhysRevB.107.14410210.1103/PhysRevB.107.144102 ↗
31. Gasteiger, J.; Groß, J.; Günnemann, S. Directional Message Passing for Molecular Graphs, International Conference on Learning Representations, 2020.
32. Gasteiger, J.; Giri, S.; Margraf, J. T.; Günnemann, S. In Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Molecules, Machine Learning for Molecules Workshop, 2020.
33. 10.1038/s41524-022-00845-010.1038/s41524-022-00845-0 ↗
34. 10.1021/acs.jctc.9b0018110.1021/acs.jctc.9b00181 ↗
35. He, K.; Zhang, X.; Ren, S.; Sun, J.In Deep Residual Learning for Image Recognition, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016; pp 770–778.10.1109/CVPR.2016.90 ↗
36. Schütt, K.; Unke, O.; Gastegger, M.In Equivariant Message Passing for the Prediction of Tensorial Properties and Molecular Spectra, Proceedings of the 38th International Conference on Machine Learning, 2021; pp 9377–9388.
37. Martin R. M. · 202010.1017/9781108555586 ↗
38. 10.1103/PhysRevLett.77.386510.1103/PhysRevLett.77.3865 ↗
39. 10.1088/0953-8984/22/25/25320210.1088/0953-8984/22/25/253202 ↗
40. 10.1088/1361-648X/aa680e10.1088/1361-648X/aa680e ↗
41. Resta, R.; Vanderbilt, D. Physics of Ferroelectrics: A Modern Perspective; Springer: Berlin, 2007; pp 31–68.
42. 10.1103/PhysRevB.47.165110.1103/PhysRevB.47.1651 ↗
43. 10.1103/PhysRev.145.56110.1103/PhysRev.145.561 ↗
44. 10.1103/PhysRevLett.100.13640610.1103/PhysRevLett.100.136406 ↗
45. 10.1038/s42254-021-00314-510.1038/s42254-021-00314-5 ↗
46. 10.1103/PhysRevB.32.378010.1103/PhysRevB.32.3780 ↗
47. 10.1063/1.324530310.1063/1.3245303 ↗
48. 10.1103/PhysRevB.92.09430110.1103/PhysRevB.92.094301 ↗
49. Loshchilov, I.; Hutter, F. Decoupled Weight Decay Regularization, 7th International Conference on Learning Representations, 2019.
50. 10.1103/PhysRevA.33.425310.1103/PhysRevA.33.4253 ↗
51. 10.1080/0026897930010037110.1080/00268979300100371 ↗
52. Martienssen W. · 2005 · 110.1007/3-540-30437-1 ↗
53. 10.1103/PhysRevB.82.13410610.1103/PhysRevB.82.134106 ↗
54. 10.1103/PhysRevB.73.23511610.1103/PhysRevB.73.235116 ↗
55. 10.1103/PhysRevB.95.05411110.1103/PhysRevB.95.054111 ↗
56. 10.1103/PhysRevLett.115.03640210.1103/PhysRevLett.115.036402 ↗
57. 10.1107/S0365110X5600030910.1107/S0365110X56000309 ↗
58. 10.1107/S002188987901175410.1107/S0021889879011754 ↗
59. 10.1103/PhysRevB.77.16510710.1103/PhysRevB.77.165107 ↗
60. 10.1038/srep4348210.1038/srep43482 ↗
61. 10.1103/PhysRevB.96.03514310.1103/PhysRevB.96.035143 ↗
62. 10.1103/PhysRevB.105.02410610.1103/PhysRevB.105.024106 ↗
63. 10.1143/JPSJ.81.12470210.1143/JPSJ.81.124702 ↗
64. 10.1103/PhysRevB.88.06430610.1103/PhysRevB.88.064306 ↗
65. 10.1021/acs.jpcc.5b0320710.1021/acs.jpcc.5b03207 ↗
66. Xie, P.; Chen, Y.; E, W.; Car, R. Ab initio multi-scale modeling of ferroelectrics: The case of PbTiO3. arXiv:2205.11839, 2022.
68. Gong, S.; Xie, T.; Shao-Horn, Y.; Gomez-Bombarelli, R.; Grossman, J. C. Examining graph neural networks for crystal structures: limitations and opportunities for capturing periodicity, arXiv:2208.05039, 2022.10.21203/rs.3.rs-2042719/v1 ↗
69. Paszke, A.; Gross, S.; Massa, F.; Lerer, A.; Bradbury, J.; Chanan, G.; Killeen, T.; Lin, Z.; Gimelshein, N.; Antiga, L. In Pytorch: An Imperative Style, High-Performance Deep Learning Library, 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), 2019.
70. 10.1080/0015019980800917310.1080/00150199808009173 ↗
71. 10.1016/0031-9163(66)91127-910.1016/0031-9163(66)91127-9 ↗
72. 10.1103/PhysRevB.70.10411210.1103/PhysRevB.70.104112 ↗
73. 10.1103/PhysRevB.108.L02010310.1103/PhysRevB.108.L020103 ↗
74. 10.1103/PhysRevB.65.10411110.1103/PhysRevB.65.104111 ↗