Unsupervised machine learning for discovery of promising half-Heusler thermoelectric materials

被引：68

作者：

Jia, Xue ^{[1
,2
]}

Deng, Yanshuai ^{[1
,2
]}

Bao, Xin ^{[1
,2
]}

Yao, Honghao ^{[1
,2
]}

Li, Shan ^{[1
,2
]}

Li, Zhou ^{[1
,2
]}

Chen, Chen ^{[1
,2
]}

Wang, Xinyu ^{[1
,2
]}

Mao, Jun ^{[1
,2
]}

Cao, Feng ^{[3
]}

Sui, Jiehe ^{[4
]}

Wu, Junwei ^{[1
,2
]}

Wang, Cuiping ^{[1
,2
,5
]}

Zhang, Qian ^{[1
,2
,4
]}

Liu, Xingjun ^{[1
,2
,4
,5
]}

机构：

[1] Harbin Inst Technol, Sch Mat Sci & Engn, Shenzhen 518055, Peoples R China

[2] Harbin Inst Technol, Inst Mat Genome & Big Data, Shenzhen 518055, Peoples R China

[3] Harbin Inst Technol, Sch Sci, Shenzhen 518055, Peoples R China

[4] Harbin Inst Technol, State Key Lab Adv Welding & Joining, Harbin 150001, Peoples R China

[5] Xiamen Univ, Coll Mat, Dept Mat Sci & Engn, Xiamen 361005, Peoples R China

来源：

NPJ COMPUTATIONAL MATERIALS | 2022年 / 8卷 / 01期

基金：

中国国家自然科学基金; 国家重点研发计划;

关键词：

PERFORMANCE; MECHANISMS; SC; HF;

D O I：

10.1038/s41524-022-00723-9

中图分类号：

O64 [物理化学（理论化学）、化学物理学];

学科分类号：

070304 ; 081704 ;

摘要：

Thermoelectric materials can be potentially applied to waste heat recovery and solid-state cooling because they allow a direct energy conversion between heat and electricity and vice versa. The accelerated materials design based on machine learning has enabled the systematic discovery of promising materials. Herein we proposed a successful strategy to discover and design a series of promising half-Heusler thermoelectric materials through the iterative combination of unsupervised machine learning with the labeled known half-Heusler thermoelectric materials. Subsequently, optimized zT values of similar to 0.5 at 925 K for p-type Sc0.7Y0.3NiSb0.97Sn0.03 and similar to 0.3 at 778 K for n-type Sc0.65Y0.3Ti0.05NiSb were experimentally achieved on the same parent ScNiSb.

引用

页数：9

共 51 条

[1]

Bastanlar Y, 2014, METHODS MOL BIOL, V1107, P105, DOI 10.1007/978-1-62703-748-8_7

[2] Nanograined Half-Heusler Semiconductors as Advanced Thermoelectrics: An Ab Initio High-Throughput Statistical Study [J].