High-Throughput Machine-Learning-Driven Synthesis of Full-Heusler Compounds

被引：262

作者：

Oliynyk, Anton O. ^{[1
]}

Antono, Erin ^{[2
]}

Sparks, Taylor D. ^{[3
]}

Ghadbeigi, Leila ^{[3
]}

Gaultois, Michael W. ^{[4
]}

Meredig, Bryce ^{[2
]}

Mar, Arthur ^{[1
]}

机构：

[1] Univ Alberta, Dept Chem, Edmonton, AB T6G 2G2, Canada

[2] Citrine Informat, Redwood City, CA 94063 USA

[3] Univ Utah, Dept Mat Sci & Engn, Salt Lake City, UT 84112 USA

[4] Univ Cambridge, Dept Chem, Lensfield Rd, Cambridge CB2 1EW, England

来源：

CHEMISTRY OF MATERIALS | 2016年 / 28卷 / 20期

基金：

加拿大自然科学与工程研究理事会;

关键词：

CRYSTAL-STRUCTURE; ELECTRONIC-STRUCTURE; MAGNETIC-PROPERTIES; NI-P; PREDICTION; PHASES; SYSTEM;

D O I：

10.1021/acs.chemmater.6b02724

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：

A machine-learning model has been trained to discover Hensler compounds, which are intermetallics exhibiting diverse physical properties attractive for applications in thermoelectric and spintronic materials. Improving these properties requires knowledge of crystal structures, which occur in three subtle variations (Hensler, inverse Hensler, and CsCl-type structures) that are difficult, and at times impossible, to distinguish by diffraction techniques. Compared to alternative approaches, this Hensler discovery engine performs exceptionally well, making fast and reliable predictions of the occurrence of Hensler vs non-Heusler compounds for an arbitrary combination of elements with no structural input on over 400 000 candidates. The model has a true positive rate of 0.94 (and false positive rate of 0.01). It is also valuable for data sanitizing, by flagging questionable entries in crystallographic databases. It was applied to screen candidates with the formula AB(2)C and predict the existence of 12 novel gallides MRu2Ga and RuM2Ga (M = Ti-Co) as Hensler compounds, which were confirmed experimentally. One member, TiRu2Ga, exhibited diagnostic superstructure peaks that confirm the adoption of an ordered Hensler as opposed to a disordered CsCl-type structure.

引用

页码：7324 / 7331

页数：8

共 40 条

[1]

[Anonymous], 2015, PEARSONS CRYSTAL STR

[2]

[Anonymous], 2016, GRAND CHALL 3 DO REM

[3] Anomalous magnetic properties of Heusler superconductor YbPd2Sn [J].

Aoki, Y ;

Sato, HR ;

Sugawara, H ;

Sato, H .

PHYSICA C-SUPERCONDUCTIVITY AND ITS APPLICATIONS, 2000, 333 (3-4) :187-194

[4] The crystal structure of the Heusler alloys. [J].

Bradley, AJ ;

Rodgers, JW .

PROCEEDINGS OF THE ROYAL SOCIETY OF LONDON SERIES A-CONTAINING PAPERS OF A MATHEMATICAL AND PHYSICAL CHARACTER, 1934, 144 (A852) :0340-0359

[5] Random forests [J].

Breiman, L .

MACHINE LEARNING, 2001, 45 (01) :5-32

[6] Informatics derived materials databases for multifunctional properties [J].

Broderick, Scott ;

Rajan, Krishna .

SCIENCE AND TECHNOLOGY OF ADVANCED MATERIALS, 2015, 16 (01) :1-8

[7] Finding Unprecedentedly Low-Thermal-Conductivity Half-Heusler Semiconductors via High-Throughput Materials Modeling [J].

Carrete, Jesus ;

Li, Wu ;

Mingo, Natalio ;

Wang, Shidong ;

Curtarolo, Stefano .

PHYSICAL REVIEW X, 2014, 4 (01)

[8]

Chadov S, 2010, NAT MATER, V9, P541, DOI [10.1038/NMAT2770, 10.1038/nmat2770]

[9]

Curtarolo S, 2013, NAT MATER, V12, P191, DOI [10.1038/NMAT3568, 10.1038/nmat3568]

[10] Origin and properties of the gap in the half-ferromagnetic Heusler alloys [J].

Galanakis, I ;

Dederichs, PH ;

Papanikolaou, N .

PHYSICAL REVIEW B, 2002, 66 (13) :1-10

← 1 2 3 4 →