Machine Learning-Aided Exploration of Ultrahard Materials
Source abstract
Abstract
Ultrahard materials are an essential component in a wide range of industrial applications. In this work, we introduce novel machine learning (ML) features for the prediction of the elastic moduli of materials, from which the Vickers hardness can be calculated. By applying the trained ML models on a space of ∼110,000 materials, these features successfully predict the elastic moduli for a range of materials. This enables the identification of materials with high Vickers hardness, as validated by comparing the predictions against the density functional theory calculations of the moduli. We further explored the predicted moduli by examining several classes of materials with interesting mechanical properties, including binary and ternary alloys, aluminum and magnesium alloys, metal borides, carbides and nitrides, and metal hydrides. Based on our ML models, we identify a number of ultrahard compounds in the B–C and B–C–N chemical spaces and ultrahard ultralight-weight magnesium alloys Mg 3 Zn and Mg 3 Cd. We also observe the inverse of the hydrogen embrittlement effect in a number of metal carbides, where the introduction of hydrogen into metal carbides increases their hardness, and find that substitutional doping of Al in transition-metal borides can yield lighter materials without compromising the thermodynamic stability or the hardness of the material.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Metal and Thin Film Mechanics · Boron and Carbon Nanomaterials Research · MXene and MAX Phase Materials
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Institute for Frontier Materials, Deakin University, Geelong, Victoria 3216, Australia
ARC Centre of Excellence in Exciton Science, School of Science, RMIT University, Melbourne, Victoria 3001 Australia
Applied Artificial Intelligence Institute (A2I2), Deakin University, Geelong, Victoria 3216, Australia
Funding
Deakin University
References (63)
Show source-provided references
1. 10.1038/s41524-019-0221-010.1038/s41524-019-0221-0 ↗
2. 10.1038/s41586-018-0337-210.1038/s41586-018-0337-2 ↗
3. 10.1002/eom2.1219410.1002/eom2.12194 ↗
4. 10.1021/acs.jcim.0c0000310.1021/acs.jcim.0c00003 ↗
5. 10.1063/1.481232310.1063/1.4812323 ↗
6. 10.1016/j.commatsci.2012.02.00210.1016/j.commatsci.2012.02.002 ↗
7. 10.1088/2053-1583/aacfc110.1088/2053-1583/aacfc1 ↗
8. 10.1063/1.240466310.1063/1.2404663 ↗
9. 10.1103/physrevb.34.539010.1103/physrevb.34.5390 ↗
10. 10.1103/physrevlett.80.451010.1103/physrevlett.80.4510 ↗
11. 10.1021/acsami.9b0119610.1021/acsami.9b01196 ↗
12. 10.1021/acs.jpclett.8b0352710.1021/acs.jpclett.8b03527 ↗
13. 10.1021/cm503507h10.1021/cm503507h ↗
14. 10.1038/s41524-019-0226-810.1038/s41524-019-0226-8 ↗
15. Engel E. · 201110.1007/978-3-642-14090-7 ↗
16. 10.1038/srep3425610.1038/srep34256 ↗
17. 10.1038/ncomms1567910.1038/ncomms15679 ↗
18. 10.1021/acs.jpcc.9b0104510.1021/acs.jpcc.9b01045 ↗
19. 10.1021/acs.jcim.0c0046410.1021/acs.jcim.0c00464 ↗
20. 10.1021/jacs.8b0271710.1021/jacs.8b02717 ↗
21. 10.1016/j.commatsci.2021.11067110.1016/j.commatsci.2021.110671 ↗
22. 10.1016/j.matdes.2008.05.01910.1016/j.matdes.2008.05.019 ↗
23. 10.1021/acscentsci.8b0022910.1021/acscentsci.8b00229 ↗
24. 10.1021/ci300482n10.1021/ci300482n ↗
25. 10.1038/s42256-019-0048-x10.1038/s42256-019-0048-x ↗
26. Tawfik, S. A.; Russo, S. P. Naturally-Meaningful and Efficient Descriptors: Machine Learning of Material Properties Based on Robust One-Shot Ab Initio Descriptors. 2022, https://arxiv.org/abs/2203.03392.10.1186/s13321-022-00658-9 ↗
27. 10.1038/sdata.2015.910.1038/sdata.2015.9 ↗
28. 10.1016/j.ijrmhm.2012.02.02110.1016/j.ijrmhm.2012.02.021 ↗
29. 10.1016/j.intermet.2011.03.02610.1016/j.intermet.2011.03.026 ↗
30. Chen, T.; Guestrin, C. XGBoost. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; Association for Computing Machinery: New York, NY, USA, 2016; pp 785–794.10.1145/2939672.2939785 ↗
31. 10.1016/j.cpc.2021.10803310.1016/j.cpc.2021.108033 ↗
32. 10.1103/physrevb.48.1311510.1103/physrevb.48.13115 ↗
33. 10.1103/physrevlett.77.386510.1103/physrevlett.77.3865 ↗
34. 10.1103/physrevb.13.518810.1103/physrevb.13.5188 ↗
35. 10.1016/S0257-8972(99)00498-310.1016/S0257-8972(99)00498-3 ↗
36. 10.1103/physrevlett.86.334810.1103/physrevlett.86.3348 ↗
37. 10.1016/j.jallcom.2008.04.08610.1016/j.jallcom.2008.04.086 ↗
38. 10.1063/1.504076310.1063/1.5040763 ↗
39. 10.1038/nature1644510.1038/nature16445 ↗
40. 10.1007/s11661-012-1217-210.1007/s11661-012-1217-2 ↗
41. 10.1126/science.118284810.1126/science.1182848 ↗
42. 10.1016/j.jnucmat.2015.09.05610.1016/j.jnucmat.2015.09.056 ↗
43. 10.1016/j.jallcom.2006.02.03610.1016/j.jallcom.2006.02.036 ↗
44. 10.1016/j.nme.2017.05.01210.1016/j.nme.2017.05.012 ↗
45. 10.1007/s11837-018-2850-z10.1007/s11837-018-2850-z ↗
46. 10.1016/j.jallcom.2004.04.07310.1016/j.jallcom.2004.04.073 ↗
47. 10.1016/j.commatsci.2018.03.07510.1016/j.commatsci.2018.03.075 ↗
48. van der Maaten L. · J. Mach. Learn. Res. · 2008 · 9 · 2579
49. 10.1002/pssb.20155203110.1002/pssb.201552031 ↗
50. 10.1063/1.482992610.1063/1.4829926 ↗
51. 10.1063/1.297776010.1063/1.2977760 ↗
52. 10.1073/pnas.110263610810.1073/pnas.1102636108 ↗
53. 10.1016/j.jeurceramsoc.2015.08.01610.1016/j.jeurceramsoc.2015.08.016 ↗
54. 10.1021/ic302040410.1021/ic3020404 ↗
55. 10.1080/23311940.2015.107670210.1080/23311940.2015.1076702 ↗
56. 10.1016/j.jppr.2018.04.00110.1016/j.jppr.2018.04.001 ↗
57. 10.1016/s1471-5317(02)00013-510.1016/s1471-5317(02)00013-5 ↗
58. 10.1126/sciadv.aaq014810.1126/sciadv.aaq0148 ↗
59. 10.1007/s11661-010-0275-610.1007/s11661-010-0275-6 ↗
60. 10.1021/ic50131a04210.1021/ic50131a042 ↗
61. 10.1039/d1ma00101a10.1039/d1ma00101a ↗
62. 10.1021/jacs.2c0059210.1021/jacs.2c00592 ↗
63. 10.1016/j.actamat.2017.06.01610.1016/j.actamat.2017.06.016 ↗