Machine learning substitutional defect formation energies in ABO3 perovskites

被引:26
作者
Sharma, Vinit [1 ,2 ]
Kumar, Pankaj [3 ]
Dev, Pratibha [3 ]
Pilania, Ghanshyam [4 ]
机构
[1] Univ Tennessee, Joint Inst Computat Sci, Knoxville, TN 37996 USA
[2] Oak Ridge Natl Lab, Natl Inst Computat Sci, POB 2009, Oak Ridge, TN 37831 USA
[3] Howard Univ, Dept Phys & Astron, Washington, DC 20059 USA
[4] Los Alamos Natl Lab, Mat Sci & Technol Div, Los Alamos, NM 87545 USA
基金
美国国家科学基金会;
关键词
DENSITY-FUNCTIONAL THEORY; DOPED LAMNO3 PEROVSKITE; COMPUTER-SIMULATION; DIELECTRIC-CONSTANT; MAGNETIC-PROPERTIES; IONIC-RADII; BATIO3; 1ST-PRINCIPLES; TEMPERATURE; TRANSITION;
D O I
10.1063/5.0015538
中图分类号
O59 [应用物理学];
学科分类号
摘要
Perovskite oxides are a promising material platform for use in a wide range of technological applications including electronics, sensors, fuel cells, and catalysis. This is owing to the extraordinary tunability of their physical and chemical properties via defect engineering. The feasibility and the stability of a defect, such as a substitutional dopant, in the host lattice is usually obtained via experiments and/or through detailed quantum mechanical calculations. Both of these conventional routes are expensive and time consuming. An alternative is a data-driven machine learning (ML)-based approach. In this work, we have applied ML techniques to identify the factors that influence defect formation energy, which is an important measure of the stability of the defects, in perovskite oxides. Using 13 elemental properties as features and random forest regression, we demonstrate a systematic approach to down-selecting from the larger set of features to those that are important, establishing a framework for accurate predictions of the defect formation energy. We quantitatively show that the most important factors that control the dopant stability are the dopant ionic size, heat of formation, effective tolerance factor, and oxidation state. Our work reveals previously unknown correlations, chemical trends, and the interplay between stability and underlying chemistries. Hence, these results showcase the efficacy of ML tools in identifying and quantifying different feature-dependencies and provide a promising route toward dopant selection in the perovskites. We have developed a framework that itself is general and can be potentially applied to other material classes.
引用
收藏
页数:10
相关论文
共 74 条
  • [1] COMPUTER-SIMULATION STUDIES OF STRONTIUM-TITANATE
    AKHTAR, MJ
    AKHTAR, ZUN
    JACKSON, RA
    CATLOW, CRA
    [J]. JOURNAL OF THE AMERICAN CERAMIC SOCIETY, 1995, 78 (02) : 421 - 428
  • [2] Multifidelity Information Fusion with Machine Learning: A Case Study of Dopant Formation Energies in Hafnia
    Batra, Rohit
    Pilania, Ghanshyam
    Uberuaga, Blas P.
    Ramprasad, Rampi
    [J]. ACS APPLIED MATERIALS & INTERFACES, 2019, 11 (28) : 24906 - 24918
  • [3] Random forests
    Breiman, L
    [J]. MACHINE LEARNING, 2001, 45 (01) : 5 - 32
  • [4] Influence of foreign ions on the crystal structure of BaTiO3
    Buscaglia, MT
    Buscaglia, V
    Viviani, M
    Nanni, P
    Hanuskova, M
    [J]. JOURNAL OF THE EUROPEAN CERAMIC SOCIETY, 2000, 20 (12) : 1997 - 2007
  • [5] Atomistic simulation of dopant incorporation in barium titanate
    Buscaglia, Maria Teresa
    Buscaglia, Vincenzo
    Viviani, Massimo
    Nanni, Paolo
    [J]. 2001, American Ceramic Soc, Westerville (84)
  • [6] PROCESSING OF LAMINATED BARIUM-TITANATE STRUCTURES FOR STRESS-SENSING APPLICATIONS
    CAPURSO, JS
    ALLES, AB
    SCHULZE, WA
    [J]. JOURNAL OF THE AMERICAN CERAMIC SOCIETY, 1995, 78 (09) : 2476 - 2480
  • [7] Structural phase transition and T-c distribution in Hf-doped LaMnO3 investigated using perturbed-angular-correlation spectroscopy
    Catchen, GL
    Evenson, WE
    Allred, D
    [J]. PHYSICAL REVIEW B, 1996, 54 (06) : R3679 - R3682
  • [8] Electric field control of the LaAlO3/SrTiO3 interface ground state
    Caviglia, A. D.
    Gariglio, S.
    Reyren, N.
    Jaccard, D.
    Schneider, T.
    Gabay, M.
    Thiel, S.
    Hammerl, G.
    Mannhart, J.
    Triscone, J. -M.
    [J]. NATURE, 2008, 456 (7222) : 624 - 627
  • [9] CHAN NH, 1984, J AM CERAM SOC, V67, P285, DOI 10.1111/j.1151-2916.1984.tb18849.x
  • [10] Chang MC, 2001, ADV QUANTUM CHEM, V37, P179