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

Xue Jia; Yanshuai Deng; Xin Bao; Honghao Yao; Shan Li; Zhou Li; Chen Chen; Xinyu Wang; Jun Mao; Feng Cao; Jiehe Sui; Junwei Wu; Cuiping Wang; Qian Zhang; Xingjun Liu

doi:10.1038/s41524-022-00723-9

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

Xue Jia, Yanshuai Deng, Xin Bao, Honghao Yao, Shan Li, Zhou Li, Chen Chen, Xinyu Wang, Jun Mao, Feng Cao, Jiehe Sui, Junwei Wu, Cuiping Wang, Qian Zhang, Xingjun Liu

Research output: Contribution to journal › Article › peer-review

50 Scopus citations

Abstract

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 ~0.5 at 925 K for p-type Sc_0.7Y_0.3NiSb_0.97Sn_0.03 and ~0.3 at 778 K for n-type Sc_0.65Y_0.3Ti_0.05NiSb were experimentally achieved on the same parent ScNiSb.

Original language	English
Article number	34
Journal	npj Computational Materials
Volume	8
Issue number	1
DOIs	https://doi.org/10.1038/s41524-022-00723-9
State	Published - Dec 2022
Externally published	Yes

Access to Document

10.1038/s41524-022-00723-9

Cite this

@article{63580bac90f248b3b12e27f3a9fbb650,

title = "Unsupervised machine learning for discovery of promising half-Heusler thermoelectric materials",

abstract = "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 ~0.5 at 925 K for p-type Sc0.7Y0.3NiSb0.97Sn0.03 and ~0.3 at 778 K for n-type Sc0.65Y0.3Ti0.05NiSb were experimentally achieved on the same parent ScNiSb.",

author = "Xue Jia and Yanshuai Deng and Xin Bao and Honghao Yao and Shan Li and Zhou Li and Chen Chen and Xinyu Wang and Jun Mao and Feng Cao and Jiehe Sui and Junwei Wu and Cuiping Wang and Qian Zhang and Xingjun Liu",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s).",

year = "2022",

month = dec,

doi = "10.1038/s41524-022-00723-9",

language = "英语",

volume = "8",

journal = "npj Computational Materials",

issn = "2057-3960",

publisher = "Nature Publishing Group",

number = "1",

}

TY - JOUR

T1 - Unsupervised machine learning for discovery of promising half-Heusler thermoelectric materials

AU - Jia, Xue

AU - Deng, Yanshuai

AU - Bao, Xin

AU - Yao, Honghao

AU - Li, Shan

AU - Li, Zhou

AU - Chen, Chen

AU - Wang, Xinyu

AU - Mao, Jun

AU - Cao, Feng

AU - Sui, Jiehe

AU - Wu, Junwei

AU - Wang, Cuiping

AU - Zhang, Qian

AU - Liu, Xingjun

PY - 2022/12

Y1 - 2022/12

N2 - 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 ~0.5 at 925 K for p-type Sc0.7Y0.3NiSb0.97Sn0.03 and ~0.3 at 778 K for n-type Sc0.65Y0.3Ti0.05NiSb were experimentally achieved on the same parent ScNiSb.

AB - 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 ~0.5 at 925 K for p-type Sc0.7Y0.3NiSb0.97Sn0.03 and ~0.3 at 778 K for n-type Sc0.65Y0.3Ti0.05NiSb were experimentally achieved on the same parent ScNiSb.

UR - http://www.scopus.com/inward/record.url?scp=85126176829&partnerID=8YFLogxK

U2 - 10.1038/s41524-022-00723-9

DO - 10.1038/s41524-022-00723-9

M3 - 文章

AN - SCOPUS:85126176829

SN - 2057-3960

VL - 8

JO - npj Computational Materials

JF - npj Computational Materials

IS - 1

M1 - 34

ER -

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

Abstract

Access to Document

Other files and links

Fingerprint

Cite this