journal article·2026-06

Universal and Task‐Specific Machine‐Learning Interatomic Potentials With Eleven‐Element Database for Thermal‐Transport Modeling of Semiconductors and Interfaces

Yanxiao Hu, Yongsen Chen, Jing Huang, Xiaoxin Xu, Yabei Wu, Caichao Ye, Jiong Yang, Wenqing Zhang

Advanced Functional Materials· Volume 36· Issue 48
Publisher page

Abstract

ABSTRACT The continuous miniaturization and integration of electronic devices have established thermal transport of semiconductors and crossing interfaces as a critical bottleneck for heat dissipation. The reliable machine‐learning interatomic potentials (MLIPs), together with tractable and dedicated training datasets, offer an accurate and efficient computational approach to simulate these challenging systems. Here, we construct a Semiconductor and Interfaces Database (SemiCID), covering most of the III‐IV‐V semiconductors and their heterointerfaces. Using the ultrasmall and superlinear SUS 2 ‐MLIP framework ( Proc. Natl. Acad. Sci. U.S.A . 122 , e2503439122 (2025) ), we strategically developed a family of ready‐to‐use MLIPs comprising one universal model (SUS 2 ‐SemiCID‐U) for general‐purpose applications and 60 task‐specific models (SUS 2 ‐SemiCID‐TS) for domain‐targeted studies. With these potentials, we accurately reproduce the thermal conductivity of 19 bulk semiconductors, showing very good agreement with experiments. Simulations of the BAs‐based heterostructures—including BAs/BP, BAs/BSb, BAs/AlN, BAs/GaN, and BAs/InN—together with the GaN/AlN reveal a positive correlation between interfacial thermal conductance (ITC) and acoustic‐phonon spectral matching, consistent with physical understanding and measured data for some of them. Taking AlAs/GaAs superlattices as examples, we quantitatively reproduce the observed suppression of thermal conductivity due to interfacial roughness. This work establishes a scalable platform for advancing thermal transport studies of semiconductors and interfaces in realistic electronic devices.
Source: crossref · Quality: baseline

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.

Affiliations

State Key Laboratory of Quantum Functional Materials & Department of Materials Science and Engineering Southern University of Science and Technology Shenzhen Guangdong China

Materials Genome Institute Shanghai Engineering Research Center for Integrated Circuits and Advanced Display Materials Shanghai University Shanghai China

Institute of Innovative Materials & Guangdong Provincial Key Lab for Computational Science and Materials Design Southern University of Science and Technology Shenzhen Guangdong China

Funding

National Natural Science Foundation of China

References (83)

Show source-provided references
1. 10.1038/s44287-025-00196-010.1038/s44287-025-00196-0 ↗
2. 10.1109/IEDM45625.2022.1001934910.1109/IEDM45625.2022.10019349 ↗
3. 10.1109/VLSITechnologyandCir46783.2024.1063151110.1109/VLSITechnologyandCir46783.2024.10631511 ↗
4. 10.1109/IRPS.2017.793633610.1109/IRPS.2017.7936336 ↗
5. 10.3390/electronics809094210.3390/electronics8090942 ↗
6. 10.1109/IITC/MAM57687.2023.1015476810.1109/IITC/MAM57687.2023.10154768 ↗
7. 10.1002/adfm.20190385710.1002/adfm.201903857 ↗
8. 10.1088/0256-307X/41/10/10630110.1088/0256-307X/41/10/106301 ↗
9. 10.1088/0256-307X/41/1/01630210.1088/0256-307X/41/1/016302 ↗
10. 10.1103/RevModPhys.94.02500210.1103/RevModPhys.94.025002 ↗
11. 10.1016/j.ijheatmasstransfer.2024.12622710.1016/j.ijheatmasstransfer.2024.126227 ↗
12. 10.1063/1.326749610.1063/1.3267496 ↗
13. 10.1021/acs.jpcc.0c1038710.1021/acs.jpcc.0c10387 ↗
14. 10.3390/nano1207104410.3390/nano12071044 ↗
15. 10.1088/1674-1056/ae043210.1088/1674-1056/ae0432 ↗
16. 10.1021/acsami.8b0370910.1021/acsami.8b03709 ↗
17. 10.1021/acs.chemrev.0c0086810.1021/acs.chemrev.0c00868 ↗
18. 10.1002/adma.20190276510.1002/adma.201902765 ↗
19. 10.1016/j.actamat.2021.11698010.1016/j.actamat.2021.116980 ↗
20. 10.1088/1674-1056/ad1b4210.1088/1674-1056/ad1b42 ↗
21. 10.1039/D0MH00787K10.1039/D0MH00787K ↗
22. 10.1021/acsami.3c1690510.1021/acsami.3c16905 ↗
23. 10.1103/w7qp-tl6z10.1103/w7qp-tl6z ↗
24. 10.1063/1.481232310.1063/1.4812323 ↗
25. 10.1016/j.commatsci.2014.05.01410.1016/j.commatsci.2014.05.014 ↗
26. 10.1038/npjcompumats.2015.1010.1038/npjcompumats.2015.10 ↗
27. 10.1002/smll.20170199810.1002/smll.201701998 ↗
28. 10.1073/pnas.250343912210.1073/pnas.2503439122 ↗
29. 10.1137/15M105418310.1137/15M1054183 ↗
30. 10.1103/PhysRevB.104.10430910.1103/PhysRevB.104.104309 ↗
31. 10.1103/PhysRevLett.120.14300110.1103/PhysRevLett.120.143001 ↗
32. 10.1103/PhysRevLett.111.02590110.1103/PhysRevLett.111.025901 ↗
33. 10.1126/science.aaz614910.1126/science.aaz6149 ↗
34. 10.1002/adfm.20180511610.1002/adfm.201805116 ↗
35. 10.1126/science.aat552210.1126/science.aat5522 ↗
36. 10.1126/science.aat898210.1126/science.aat8982 ↗
37. 10.1126/science.aat793210.1126/science.aat7932 ↗
38. Bulk‐Like Intrinsic Phonon Thermal Conductivity of Micrometer‐Thick AlN Films
39. 10.1103/PhysRevMaterials.4.04460210.1103/PhysRevMaterials.4.044602 ↗
40. 10.1063/5.010691610.1063/5.0106916 ↗
41. 10.1063/5.000891910.1063/5.0008919 ↗
42. 10.1063/5.014965110.1063/5.0149651 ↗
43. 10.1063/1.509717210.1063/1.5097172 ↗
44. 10.1016/0022-3697(87)90153-310.1016/0022-3697(87)90153-3 ↗
45. 10.1103/PhysRev.141.76710.1103/PhysRev.141.767 ↗
46. 10.1016/0038-1101(67)90069-X10.1016/0038-1101(67)90069-X ↗
47. 10.1103/PhysRevMaterials.3.08460710.1103/PhysRevMaterials.3.084607 ↗
48. 10.1103/PhysRevMaterials.3.01460110.1103/PhysRevMaterials.3.014601 ↗
49. 10.3390/nano1208128810.3390/nano12081288 ↗
50. 10.1088/0268-1242/18/7/31510.1088/0268-1242/18/7/315 ↗
51. 10.1007/s10973-016-5765-310.1007/s10973-016-5765-3 ↗
52. 10.1088/1361-6528/abb6a310.1088/1361-6528/abb6a3 ↗
53. 10.1063/1.353650710.1063/1.3536507 ↗
54. 10.1016/S0022-3697(97)00222-910.1016/S0022-3697(97)00222-9 ↗
55. 10.1063/1.494844610.1063/1.4948446 ↗
56. 10.1039/C6CE02642G10.1039/C6CE02642G ↗
57. 10.1063/1.173526410.1063/1.1735264 ↗
58. 10.1063/1.177697710.1063/1.1776977 ↗
59. 10.1103/PhysRevB.5.230110.1103/PhysRevB.5.2301 ↗
60. 10.1007/s11665-018-3715-x10.1007/s11665-018-3715-x ↗
61. 10.1016/j.jallcom.2022.16594910.1016/j.jallcom.2022.165949 ↗
62. 10.1103/PhysRev.134.A47110.1103/PhysRev.134.A471 ↗
63. 10.1088/0256-307X/38/9/09720110.1088/0256-307X/38/9/097201 ↗
64. 10.1126/science.abn429010.1126/science.abn4290 ↗
65. 10.1126/science.abn472710.1126/science.abn4727 ↗
66. 10.1063/5.024538110.1063/5.0245381 ↗
67. 10.1063/5.004966210.1063/5.0049662 ↗
68. 10.1016/j.ijheatmasstransfer.2021.12209010.1016/j.ijheatmasstransfer.2021.122090 ↗
69. 10.1063/1.500382710.1063/1.5003827 ↗
70. 10.1103/ym8w-gwr110.1103/ym8w-gwr1 ↗
71. 10.1088/1361-648X/ad5fd710.1088/1361-648X/ad5fd7 ↗
72. 10.1002/adfm.20080098410.1002/adfm.200800984 ↗
73. 10.1021/acs.chemmater.0c0369110.1021/acs.chemmater.0c03691 ↗
74. 10.1021/nl400116210.1021/nl4001162 ↗
75. 10.1016/j.ultramic.2008.08.00110.1016/j.ultramic.2008.08.001 ↗
76. 10.1103/PhysRevLett.113.06590110.1103/PhysRevLett.113.065901 ↗
77. 10.1103/PhysRevB.79.07531610.1103/PhysRevB.79.075316 ↗
78. 10.1103/PhysRevB.88.15531110.1103/PhysRevB.88.155311 ↗
79. 10.1103/PhysRevB.97.08530610.1103/PhysRevB.97.085306 ↗
80. 10.1103/PhysRevB.103.23513110.1103/PhysRevB.103.235131 ↗
81. 10.1103/PhysRevLett.77.386510.1103/PhysRevLett.77.3865 ↗
82. 10.1016/j.cpc.2014.02.01510.1016/j.cpc.2014.02.015 ↗
83. 10.1016/j.cpc.2021.10817110.1016/j.cpc.2021.108171 ↗