journal article·2025-02-10

Ionic Conductivity Study of Antiperovskite Solid-State Electrolytes Based on Interpretable Machine Learning

Shang Xiang, Shaowen Lu, Jiawei Li, Kai Xie, Rui Zhu, Huanan Wang, Kai Huang, Chaoen Li, Jiang Wu, Shibo Chen, Yuhui Shen, Yuelin Chen, Zhengyang Wen

ACS Applied Energy Materials· Volume 8· Issue 3
Publisher page

Abstract

The development of high-performance all-solid-state ion batteries necessitates the design of solid-state electrolytes (SSEs) with high ionic conductivity and excellent electrochemical stability. Antiperovskite (AP) X 3 BA, as the electronically inverted derivative of perovskite ABX 3, has garnered significant attention in the field of energy storage batteries due to its superior ionic conductivity. However, the relationship between their structure and ion diffusion behavior warrants further investigation. In this work, we constructed a machine learning (ML) framework for predicting and analyzing the ionic conductivity of the AP SSE, which encompasses data collection, feature selection, and training of various ML models. The optimal ML model demonstrated an exceptional classification performance, achieving an accuracy rate as high as 94%. Furthermore, we employed the ion substitution method to expand the sample size from 168 to 150,000 orders of magnitude. Based on this expanded data set, we examined and analyzed the mechanisms underlying high ionic conductivity from a big data perspective. The findings reveal a strong correlation between the ionic conductivity and atomic-scale characteristics at the A-site. The electronegativity, density, and ionic radius at the A-site are identified as the three most critical features influencing ionic conductivity. The interpretable ML model constructed in this study enables high-precision prediction of the ionic conductivity of AP materials, provides insightful design principles, and significantly accelerates the development and application of AP SSEs.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Machine Learning in Materials Science · Advanced Battery Materials and Technologies · Advancements in Battery Materials

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

Affiliations

College of Energy and Mechanical Engineering

Shanghai University of Electric Power

Shanghai Tongtianbixiang Environmental Protection Technology Company

College of Mathematics and Physics

Shanghai Institute of Special Equipment Inspection and Technical Research

School of Civil and Transportation Engineering

Ningbo University of Technology

Funding

Natural Science Foundation of Shanghai Municipality · National Natural Science Foundation of China

References (40)

Show source-provided references
1. 10.1038/nchem.208510.1038/nchem.2085 ↗
2. 10.1039/C9TA10658H10.1039/C9TA10658H ↗
3. 10.1016/j.rser.2024.11478310.1016/j.rser.2024.114783 ↗
4. 10.1142/S179360472130001210.1142/S1793604721300012 ↗
5. 10.1039/D1TA03680G10.1039/D1TA03680G ↗
6. 10.3390/nano1008160610.3390/nano10081606 ↗
7. 10.1039/C9EE03828K10.1039/C9EE03828K ↗
8. 10.1002/inf2.1225210.1002/inf2.12252 ↗
9. 10.1021/acs.chemrev.1c0059410.1021/acs.chemrev.1c00594 ↗
10. 10.1088/1361-6463/ad4daf10.1088/1361-6463/ad4daf ↗
11. 10.1021/acsmaterialsau.1c0002610.1021/acsmaterialsau.1c00026 ↗
12. 10.1016/j.commatsci.2023.11218810.1016/j.commatsci.2023.112188 ↗
13. 10.1103/PhysRevMaterials.4.04460110.1103/PhysRevMaterials.4.044601 ↗
14. 10.1038/s41524-022-00951-z10.1038/s41524-022-00951-z ↗
15. 10.1038/s41524-017-0056-510.1038/s41524-017-0056-5 ↗
16. 10.1038/nature1453910.1038/nature14539 ↗
17. 10.1007/s12525-021-00475-210.1007/s12525-021-00475-2 ↗
18. 10.1109/TCCN.2017.275837010.1109/TCCN.2017.2758370 ↗
19. 10.1016/j.ceramint.2023.05.15610.1016/j.ceramint.2023.05.156 ↗
20. 10.1016/j.commatsci.2022.11147510.1016/j.commatsci.2022.111475 ↗
21. 10.1038/s41524-023-01000-z10.1038/s41524-023-01000-z ↗
22. 10.1039/D2TA03613D10.1039/D2TA03613D ↗
23. 10.1038/s41597-020-0491-x10.1038/s41597-020-0491-x ↗
24. 10.1038/s41597-020-0474-y10.1038/s41597-020-0474-y ↗
25. 10.1038/s41563-020-0777-610.1038/s41563-020-0777-6 ↗
26. 10.1039/D0EE02838J10.1039/D0EE02838J ↗
27. 10.1021/acs.jpcc.3c0290810.1021/acs.jpcc.3c02908 ↗
28. 10.1016/j.nanoms.2023.08.00210.1016/j.nanoms.2023.08.002 ↗
29. James G. · 2023 · 6910.1007/978-3-031-38747-0_3 ↗
30. Ke G. · Adv. Neural Inf. Process. Syst. · 2017 · 30 · 52
31. 10.1016/j.solener.2023.11185310.1016/j.solener.2023.111853 ↗
32. 10.1107/S205252061501397910.1107/S2052520615013979 ↗
33. 10.7717/peerj-cs.62310.7717/peerj-cs.623 ↗
34. 10.1109/LSP.2020.301683710.1109/LSP.2020.3016837 ↗
35. 10.1016/j.solmat.2024.11288110.1016/j.solmat.2024.112881 ↗
36. 10.1007/s00500-020-04911-x10.1007/s00500-020-04911-x ↗
37. 10.1021/jacs.7b0444410.1021/jacs.7b04444 ↗
38. 10.1016/j.ensm.2024.10336310.1016/j.ensm.2024.103363 ↗
39. 10.1021/acsomega.3c0140010.1021/acsomega.3c01400 ↗
40. 10.1007/s13253-021-00479-710.1007/s13253-021-00479-7 ↗