|
Special Issue:
SPECIAL TOPIC — Smart design of materials and design of smart materials
|
| SPECIAL TOPIC—Smart design of materials and design of smart materials |
Prev
Next
|
|
|
Prediction of lattice thermal conductivity with two-stage interpretable machine learning |
| Jinlong Hu(胡锦龙)1,†, Yuting Zuo(左钰婷)1,†, Yuzhou Hao(郝昱州)1,†, Guoyu Shu(舒国钰)1, Yang Wang(王洋)1, Minxuan Feng(冯敏轩)1, Xuejie Li(李雪洁)1, Xiaoying Wang(王晓莹)1, Jun Sun(孙军)1, Xiangdong Ding(丁向东)1, Zhibin Gao(高志斌)1,‡, Guimei Zhu(朱桂妹)2,§, and Baowen Li(李保文)3,4,5 |
1 State Key Laboratory for Mechanical Behavior of Materials, Xi'an Jiaotong University, Xi'an 710049, China;
2 School of Microelectronics, Southern University of Science and Technology, Shenzhen 518055, China;
3 Department of Materials Science and Engineering, Southern University of Science and Technology, Shenzhen 518055, China;
4 Department of Physics, Southern University of Science and Technology, Shenzhen 518055, China;
5 Paul M. Rady Department of Mechanical Engineering and Department of Physics, University of Colorado, Boulder, Colorado 80305-0427, USA |
|
|
|
|
Abstract Thermoelectric and thermal materials are essential in achieving carbon neutrality. However, the high cost of lattice thermal conductivity calculations and the limited applicability of classical physical models have led to the inefficient development of thermoelectric materials. In this study, we proposed a two-stage machine learning framework with physical interpretability incorporating domain knowledge to calculate high/low thermal conductivity rapidly. Specifically, crystal graph convolutional neural network (CGCNN) is constructed to predict the fundamental physical parameters related to lattice thermal conductivity. Based on the above physical parameters, an interpretable machine learning model-sure independence screening and sparsifying operator (SISSO), is trained to predict the lattice thermal conductivity. We have predicted the lattice thermal conductivity of all available materials in the open quantum materials database (OQMD) (https://www.oqmd.org/). The proposed approach guides the next step of searching for materials with ultra-high or ultra-low lattice thermal conductivity and promotes the development of new thermal insulation materials and thermoelectric materials.
|
Received: 31 December 2022
Revised: 06 February 2023
Accepted manuscript online: 10 February 2023
|
|
PACS:
|
63.20.-e
|
(Phonons in crystal lattices)
|
| |
63.20.Ry
|
(Anharmonic lattice modes)
|
| |
74.25.fg
|
(Thermoelectric effects)
|
| |
72.20.-i
|
(Conductivity phenomena in semiconductors and insulators)
|
|
| Fund: We acknowledge the support of the National Natural Science Foundation of China (Grant Nos. 12104356 and 52250191), China Postdoctoral Science Foundation (Grant No. 2022M712552), the Opening Project of Shanghai Key Laboratory of Special Artificial Microstructure Materials and Technology (Grant No. Ammt2022B-1), and the Fundamental Research Funds for the Central Universities. We also acknowledge the support by HPC Platform, Xi'an Jiaotong University. |
Corresponding Authors:
Zhibin Gao, Guimei Zhu
E-mail: zhibin.gao@xjtu.edu.cn;zhugm@sustech.edu.cn
|
Cite this article:
Jinlong Hu(胡锦龙), Yuting Zuo(左钰婷), Yuzhou Hao(郝昱州), Guoyu Shu(舒国钰), Yang Wang(王洋), Minxuan Feng(冯敏轩), Xuejie Li(李雪洁), Xiaoying Wang(王晓莹), Jun Sun(孙军), Xiangdong Ding(丁向东), Zhibin Gao(高志斌), Guimei Zhu(朱桂妹), and Baowen Li(李保文) Prediction of lattice thermal conductivity with two-stage interpretable machine learning 2023 Chin. Phys. B 32 046301
|
[1] He J and Tritt T M 2017 Science 357 eaak9997
[2] Bell L 2008 Science 321 1457
[3] Chang C, Wu M, He D, Pei Y, Wu C F, Wu X, Hulei Y, Zhu F, Wang K, Chen Y, Huang L, Li J, He J and Zhao L 2018 Science 360 778
[4] He W, Wang D, Wu H, Xiao Y, Yang Z, He D, Feng Y, Hao Y J, Dong J, Chetty R, Hao L, Chen D, Qin J, Yang Q, Li X, Song J M, Zhu Y, Xu W, Niu C and Zhao L 2019 Science 365 1418
[5] Zhang Y and Chen L 2018 NPJ Comput. Mater. 4 25
[6] Wen C, Zhang Y, Wang C, Xue D, Bai Y, Antonov S, Dai L, Lookman T and Su Y 2019 Acta Mater. 170 109
[7] Rampi R, Rohit B, Ghanshyam P, Arun M K and Chiho K 2017 NPJ Comput. Mater. 3 54
[8] Liu Y, Wu J, Wang Z, Lu X G, Avdeevd M, Shi S, Wang C and Yu T 2020 Acta Mater. 195 454
[9] Mitchell J B 2014 WIREs Comput. Mol. Sci. 4 468
[10] Bharat M, Anthony G, Hong D, Wei C, Kristin P, Mark A, Andrew C and Maciej H 2016 NPJ Comput. Mater. 2 1
[11] Maarten J, Wei C, Randy N, Kristin P, Gerbrand C, Anubhav J, Mark A and Anthony G 2016 Sci. Rep. 6 34256
[12] Paul R, Katherine E, Philip A, Casey F, Malia W, Aurelio M, Matthias Z, Sorelle F, Joshua S and Alexander N 2016 Nature 533 73
[13] Xue D, Balachandran P, Hogden J, Theiler J, Xue D and Lookman T 2016 Nat. Commun. 7 11241
[14] Edward K, Kevin H, Stefanie J and Elsa O 2017 NPJ Comput. Mater. 3 53
[15] Edward K, Kevin H, Alex T, Sara M, Emma S, Adam S, Andrew M and Elsa O 2017 Sci. Data 4 170127
[16] Wan X, Feng W, Wang Y, Wang H, Zhang X, Deng C and Yang N 2019 Nano Lett. 19 3387
[17] Seko A, Togo A, Hayashi H, Tsuda K, Chaput L and Tanaka I 2015 Phys. Rev. Lett. 115 205901
[18] Faber F A, Lindmaa A, von Lilienfeld O A and Armiento R 2016 Phys. Rev. Lett. 117 135502
[19] Xue D, Balachandran P, Hogden J, Theiler J, Xue D and Lookman T 2016 Nat. Commun. 7 11241
[20] Jaafreh R, Kang Y and Hamad K 2021 ACS Appl. Mater. Interfaces 13 57204
[21] Juneja R, Yumnam G, Satsangi S and Singh A 2019 Chem. Mater. 31 5145
[22] Miyazaki H, Tamura T, Mikami M, Watanabe K, Ide N, Ozkendir O M and Nishino Y 2021 Sci. Rep. 11 13410
[23] Ju S, Yoshida R, Liu C, Wu S, Hongo K, Tadano T and Shiomi J 2021 Phys. Rev. Mater. 5 053801
[24] Loftis C, Yuan K, Zhao Y, Hu M and Hu J 2021 J. Phys. Chem. A 4 435
[25] Carrete J, Li W, Mingo N, Wang S and Curtarolo S 2014 Phys. Rev. X 4 011019
[26] Roekeghem A, Carrete J, Oses C, Curtarolo S and Mingo N 2016 Phys. Rev. X 6 041061
[27] Xie T and Grossman J C 2018 Phys. Rev. Lett. 120 145301
[28] Zhu T, He R, Gong S, Xie T, Gorai P, Nielsch K and Grossman J 2021 Energy Environ. Sci. 14 3559
[29] Rudin C 2019 Nat. Mach. Intell. 1 206
[30] Ouyang R, Curtarolo S, Ahmetcik E, Scheffler M and Ghiringhelli L M 2018 Phys. Rev. Mater. 2 083802
[31] Loftis C, Yuan K, Zhao Y, Hu M and Hu J 2020 J. Phys. Chem. A 125 435
[32] Liu J, Han S, Cao G, Zhou Z, Sheng C and Liu H 2020 J. Phys. D 53 315301
[33] Morelli D T and Slack G A 2006 High Lattice Thermal Conductivity Solids (New York: Springer) pp.37-68
[34] http://aflowlib.org/
[35] Kirklin S, Saal J, Meredig B, Thompson A, Doak J, Aykol M, Rhl S and Wolverton C 2015 NPJ Comput. Mater. 1 15010
[36] Tshitoyan V, Dagdelen J, Weston L, Dunn A, Rong Z, Kononova O, Persson K, Ceder G and Jain A 2019 Nature 571 95 |
| No Suggested Reading articles found! |
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
Altmetric
|
|
blogs
Facebook pages
Wikipedia page
Google+ users
|
Online attention
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.
View more on Altmetrics
|
|
|
