journal article·2024-01-24

Improving ab initio diffusion calculations in materials through Gaussian process regression

Seyyedfaridoddin Fattahpour, Sara Kadkhodaei

Physical Review Materials· Volume 8· Issue 1
Publisher page

Abstract

Saddle point search schemes are widely used to identify the transition state of different processes, like chemical reactions, surface and bulk diffusion, surface adsorption, and many more. In solid-state materials with relatively large numbers of atoms, the minimum mode following schemes such as dimer are commonly used because they alleviate the calculation of the Hessian on the high-dimensional potential energy surface. Here, we show that the dimer search can be further accelerated by leveraging Gaussian process regression (GPR). The GPR serves as a surrogate model to feed the dimer with the required energy and force input. We test the GPR-accelerated dimer method for predicting the diffusion coefficient of vacancy-mediated self-diffusion in body-centered cubic molybdenum and sulfur diffusion in hexagonal molybdenum disulfide. We use a multitask learning approach that utilizes a shared covariance function between energy and force input, and we show that the multitask learning significantly improves the performance of the GPR surrogate model compared to previously used learning approaches. Additionally, we demonstrate that a translation-hop sampling approach is necessary to avoid overfitting the GPR surrogate model to the minimum-mode-following pathway and thus succeeding in locating the saddle point. We show that our method reduces the number of evaluations compared to a conventional dimer method.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Machine Learning in Materials Science · Fuel Cells and Related Materials · Nuclear Materials and Properties

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

Affiliations

Department of Civil, Materials, and Environmental Engineering, University of Illinois Chicago, Chicago, Illinois 60607, USA

Funding

National Science Foundation

References (38)

Show source-provided references
1. 10.1063/1.174960410.1063/1.1749604 ↗
2. 10.1016/0022-3697(57)90059-810.1016/0022-3697(57)90059-8 ↗
3. 10.1063/1.174971310.1063/1.1749713 ↗
4. 10.1063/1.508674610.1063/1.5086746 ↗
5. H. Jönsson · 1998
6. 10.1063/1.479834410.1063/1.4798344 ↗
7. 10.1063/1.132967210.1063/1.1329672 ↗
8. 10.1063/1.48009710.1063/1.480097 ↗
9. 10.1073/pnas.040812710210.1073/pnas.0408127102 ↗
10. 10.1063/1.284194110.1063/1.2841941 ↗
11. 10.1063/1.132322410.1063/1.1323224 ↗
12. 10.1063/1.486241010.1063/1.4862410 ↗
13. 10.1063/1.180957410.1063/1.1809574 ↗
14. 10.1063/1.487343710.1063/1.4873437 ↗
15. 10.1021/acs.jctc.5b0121610.1021/acs.jctc.5b01216 ↗
16. 10.1126/sciadv.160301510.1126/sciadv.1603015 ↗
17. 10.1038/s41467-018-06169-210.1038/s41467-018-06169-2 ↗
18. 10.1016/j.cpc.2019.02.00710.1016/j.cpc.2019.02.007 ↗
19. 10.1038/s41467-022-31093-x10.1038/s41467-022-31093-x ↗
20. 10.1063/1.498678710.1063/1.4986787 ↗
21. 10.1021/acs.jctc.8b0070810.1021/acs.jctc.8b00708 ↗
22. 10.1021/acs.jpca.9b0823910.1021/acs.jpca.9b08239 ↗
23. 10.1021/acs.jctc.9b0103810.1021/acs.jctc.9b01038 ↗
24. 10.1021/acs.jctc.9b0069210.1021/acs.jctc.9b00692 ↗
25. 10.1021/acs.jctc.0c0034810.1021/acs.jctc.0c00348 ↗
26. 10.1002/(SICI)1096-987X(199605)17:7%3C888::AID-JCC12%3E3.0.CO;2-710.1002/(SICI)1096-987X(199605)17:7%3C888::AID-JCC12%3E3.0.CO;2-7 ↗
27. E. Solak · 2002
28. 10.1063/1.501710310.1063/1.5017103 ↗
29. 10.1103/PhysRevB.54.1116910.1103/PhysRevB.54.11169 ↗
30. 10.1103/PhysRevB.50.1795310.1103/PhysRevB.50.17953 ↗
31. 10.1103/PhysRevLett.77.386510.1103/PhysRevLett.77.3865 ↗
32. C. E. Rasmussen · 200510.7551/mitpress/3206.001.0001 ↗
33. E. V. Bonilla · 2007
34. K. Swersky · 2013
35. J. R. Gardner · 2018
36. 10.1103/PhysRevMaterials.6.02380310.1103/PhysRevMaterials.6.023803 ↗
37. C. Herzig · 1987
38. 10.1103/PhysRevB.88.03530110.1103/PhysRevB.88.035301 ↗