journal article·2021-10Open access

Machine learning for predicting thermal transport properties of solids

Xin Qian, Ronggui Yang

Abstract

Quantitative descriptions of the structure-thermal property correlation have been a bottleneck in designing materials with superb thermal properties. In the past decade, the first-principles phonon calculations using density functional theory and the Boltzmann transport equation have become a common practice for predicting the thermal conductivity of new materials. However, first-principles calculations are too costly for high-throughput material screening and multi-scale structural design. First-principles calculations also face several fundamental challenges in modeling thermal transport properties, e.g., of crystalline materials with defects, of amorphous materials, and for materials at high temperatures. In the past five years, machine learning started to play a role in solving these challenges. This review provides a comprehensive summary and discussion on the state-of-the-art, future opportunities, and the remaining challenges in implementing machine learning for studying thermal conductivity. After an introduction to the working principles of machine learning and descriptors of material structures, recent research using machine learning to study thermal transport is discussed. Three major applications of machine learning for predicting thermal properties are discussed. First, machine learning is applied to solve the challenges in modeling phonon transport of crystals with defects, in amorphous materials, and at high temperatures. Machine learning is used to build high-fidelity interatomic potentials to bridge the gap between first-principles calculations and molecular dynamics simulations. Second, machine learning can be used to study the correlation between thermal conductivity and other properties for high-throughput materials screening. Finally, machine learning is a powerful tool for structural design to achieve target thermal conductance or thermal conductivity.
Source: openaire · Quality: candidate unverified · Source record ↗

Research context

OpenAlex topics: Thermal properties of materials · Machine Learning in Materials Science · Advanced Thermoelectric Materials and Devices

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

References (100+)

Show source-provided references
1. Murshed · 2016
2. Bar-Cohen · J. Heat Transfer · 2012 · 13410.1115/1.4005708 ↗
3. Clarke · Mater. Today · 2005 · 8 · 2210.1016/S1369-7021(05)70934-2 ↗
4. Eang · Energy Policy · 2008 · 36 · 398210.1016/j.enpol.2008.07.014 ↗
5. Abdlrahman · Build. Environ. · 1991 · 26 · 18910.1016/0360-1323(91)90026-8 ↗
6. Qian · Nat. Mater. · 2021
7. Slack · J. Phys. Chem. Solids · 1973 · 34 · 32110.1016/0022-3697(73)90092-9 ↗
8. Broido · Appl. Phys. Lett. · 2007 · 9110.1063/1.2822891 ↗
9. Li · Phys. Rev. B · 2012 · 85
10. Ward · Phys. Rev. B · 2009 · 8010.1103/PhysRevB.80.125203 ↗
11. Esfarjani · Phys. Rev. B · 2011 · 8410.1103/PhysRevB.84.085204 ↗
12. Ward · Phys. Rev. B · 2010 · 8110.1103/PhysRevB.81.085205 ↗
13. Lindsay · Phys. Rev. Lett. · 2012 · 10910.1103/PhysRevLett.109.095901 ↗
14. Santia · Appl. Phys. Lett. · 2015 · 10710.1063/1.4927742 ↗
15. Fugallo · Nano Lett. · 2014 · 14 · 610910.1021/nl502059f ↗
16. Lindsay · Phys. Rev. B · 2010 · 82
17. Lindsay · Phys. Rev. B · 2011 · 8310.1103/PhysRevB.83.235428 ↗
18. Gu · J. Appl. Phys. · 2015 · 117
19. Xie · Appl. Phys. Lett. · 2014 · 104
20. Jain · Sci. Rep. · 2015 · 5 · 850110.1038/srep08501 ↗
21. Qin · Phys. Chem. Chem. Phys. · 2015 · 17 · 485410.1039/C4CP04858J ↗
22. Sun · Adv Mater · 2017 · 29
23. Zhu · Adv. Electron. Mater. · 2016 · 2
24. Gu · J. Appl. Phys. · 2016 · 119
25. Gu · Appl. Phys. Lett. · 2014 · 105
26. Gu · Phys. Rev. B · 2016 · 94
27. Jiang · Adv Mater · 2017 · 29
28. Chen · ACS Nano · 2011 · 5 · 32110.1021/nn102915x ↗
29. Lindsay · Phys. Rev. Lett. · 2013 · 11110.1103/PhysRevLett.111.025901 ↗
30. Feng · Phys. Rev. B · 2017 · 96 · 16120110.1103/PhysRevB.96.161201 ↗
31. Tian · Science · 2018 · 361 · 58210.1126/science.aat7932 ↗
32. Kang · Science · 2018 · 361 · 57510.1126/science.aat5522 ↗
33. Li · Science · 2018 · 361 · 57910.1126/science.aat8982 ↗
34. Chen · Science · 2020 · 367 · 55510.1126/science.aaz6149 ↗
35. Skelton · Phys. Rev. Lett. · 2016 · 11710.1103/PhysRevLett.117.075502 ↗
36. Souvatzis · Phys. Rev. Lett. · 2008 · 10010.1103/PhysRevLett.100.095901 ↗
37. Glensk · Phys. Rev. Lett. · 2019 · 12310.1103/PhysRevLett.123.235501 ↗
38. Murakami · EPL (Europhys. Lett.) · 2013 · 102 · 4600210.1209/0295-5075/102/46002 ↗
39. Katcho · Phys. Rev. B · 2014 · 9010.1103/PhysRevB.90.094117 ↗
40. Allen · Phys. Rev. B · 1993 · 48 · 1258110.1103/PhysRevB.48.12581 ↗
41. Simoncelli · Nat. Phys. · 2019 · 15 · 80910.1038/s41567-019-0520-x ↗
42. Isaeva · Nat. Commun. · 2019 · 10 · 385310.1038/s41467-019-11572-4 ↗
43. Braun · Adv Mater · 2018 · 3010.1002/adma.201805004 ↗
44. Gaultois · Chem. Mater. · 2013 · 25 · 291110.1021/cm400893e ↗
45. Qian · Nano Energy · 2017 · 41 · 39410.1016/j.nanoen.2017.09.047 ↗
46. Rohskopf · Npj Comput. Mater. · 2017 · 3 · 2710.1038/s41524-017-0026-y ↗
47. Howell · J. Chem. Phys. · 2012 · 13710.1063/1.4767516 ↗
48. Qian · J. Phys. Chem. C · 2015 · 119 · 2830010.1021/acs.jpcc.5b09527 ↗
49. Curtarolo · Comput. Mater. Sci. · 2012 · 58 · 22710.1016/j.commatsci.2012.02.002 ↗
50. Jain · APL Mater. · 2013 · 110.1063/1.4812323 ↗
51. Draxl · MRS Bull. · 2018 · 43 · 67610.1557/mrs.2018.208 ↗
52. Hellenbrandt · Crystallogr. Rev. · 2014 · 10 · 1710.1080/08893110410001664882 ↗
53. Saal · Jom · 2013 · 65 · 150110.1007/s11837-013-0755-4 ↗
54. Gorai · Nat. Rev. Mater. · 2017 · 210.1038/natrevmats.2017.53 ↗
55. Carrete · Phys. Rev. X · 2014 · 4
56. Xie · J. Appl. Phys. · 2019 · 125
57. Li · ACS Appl. Mater. Interfaces · 2019 · 11 · 2485910.1021/acsami.9b01196 ↗
58. Miller · Chem. Mater. · 2017 · 29 · 249410.1021/acs.chemmater.6b04179 ↗
59. McKinney · Chem. Mater. · 2019 · 31 · 204810.1021/acs.chemmater.8b05084 ↗
60. Siemens · Nat. Mater. · 2010 · 9 · 2610.1038/nmat2568 ↗
61. Hoogeboom-Po · Proc. Natl. Acad. Sci. · 2015 · 112 · 484610.1073/pnas.1503449112 ↗
62. Hu · Nat. Nanotechnol. · 2015 · 10 · 70110.1038/nnano.2015.109 ↗
63. Poudel · Science · 2008 · 320 · 63410.1126/science.1156446 ↗
64. Hopkins · Isrn Mech. Eng. · 2013 · 2013 · 110.1155/2013/682586 ↗
65. Hopkins · Phys. Rev. B · 2010 · 8210.1103/PhysRevB.82.085307 ↗
66. Ravichandran · Nat. Mater. · 2014 · 13 · 16810.1038/nmat3826 ↗
67. Ge · Phys. Rev. Lett. · 2006 · 9610.1103/PhysRevLett.96.186101 ↗
68. Luckyanova · Science · 2012 · 338 · 93610.1126/science.1225549 ↗
69. Samuel · Ibm J. Res. Dev. · 1959 · 3 · 21010.1147/rd.33.0210 ↗
70. Awad · 2015
71. Nielsen · 2015
72. Wang · Phys. Rev. B · 2016 · 94
73. Silver · Nature · 2016 · 529 · 48410.1038/nature16961 ↗
74. Bellman · 2015
75. Lindsay · J. Phys. Condens. Matter · 2008 · 2010.1088/0953-8984/20/16/165209 ↗
76. Seko · 2018
77. Carrete · Comput. Phys. Commun. · 2017 · 220 · 35110.1016/j.cpc.2017.06.023 ↗
78. Pozdnyakov · Phys. Rev. Lett. · 2020 · 12510.1103/PhysRevLett.125.166001 ↗
79. Shapeev · Multiscale Model. Simul. · 2016 · 14 · 115310.1137/15M1054183 ↗
80. Zuo · J. Phys. Chem. A · 2020 · 124 · 73110.1021/acs.jpca.9b08723 ↗
81. Behler · Phys. Rev. Lett. · 2007 · 9810.1103/PhysRevLett.98.146401 ↗
82. Weyl · 1946
83. Xie · Phys. Rev. Lett. · 2018 · 120
84. Rupp · Phys. Rev. Lett. · 2012 · 10810.1103/PhysRevLett.108.058301 ↗
85. Lam Pham · Sci. Technol. Adv. Mater. · 2017 · 18 · 75610.1080/14686996.2017.1378060 ↗
86. Caro · Phys. Rev. B · 2019 · 10010.1103/PhysRevB.100.024112 ↗
87. Seko · Phys. Rev. B · 2017 · 9510.1103/PhysRevB.95.144110 ↗
88. Schmidt · Npj Comput. Mater. · 2019 · 5 · 8310.1038/s41524-019-0221-0 ↗
89. Murphy · 2012
90. Noble · Nat. Biotechnol. · 2006 · 24 · 156510.1038/nbt1206-1565 ↗
91. Orović · Math. Probl. Eng. · 2016 · 2016 · 110.1155/2016/7616393 ↗
92. Tibshirani · J. R. Stat. Soc. Ser. B · 1996 · 58 · 26710.1111/j.2517-6161.1996.tb02080.x ↗
93. Quinonero-Candela · J. Mach. Learn. Res. · 2005 · 6
94. Rasmussen · 2006
95. Deep sparse rectifier neural networks
96. Clevert · arxiv · 2015 · 1511 · 07289
97. Bishop · 1995
98. Csáji · 2001 · 7
99. Quinlan · Mach. Learn. · 1986 · 1 · 8110.1023/A:1022643204877 ↗
100. Quinlan · Int. J. Man-Mach. Stud. · 1987 · 27 · 22110.1016/S0020-7373(87)80053-6 ↗