Simple Structural Descriptor Obtained from Symbolic Classification for Predicting the Oxygen Vacancy Defect Formation of Perovskites

Siyu Liu; Jing Wang; Zhongtao Duan; Kongxiang Wang; Wanlu Zhang; Ruiqian Guo; Fengxian Xie

doi:10.1021/acsami.1c24003

Simple Structural Descriptor Obtained from Symbolic Classification for Predicting the Oxygen Vacancy Defect Formation of Perovskites

ACS Appl Mater Interfaces. 2022 Mar 9;14(9):11758-11767. doi: 10.1021/acsami.1c24003. Epub 2022 Feb 23.

Authors

Siyu Liu¹, Jing Wang², Zhongtao Duan², Kongxiang Wang², Wanlu Zhang², Ruiqian Guo^{1

2

3

4}, Fengxian Xie^{1

2

3}

Affiliations

¹ Institute of Future Lighting, Academy for Engineering and Technology, Fudan University, Shanghai 200433, China.
² Institute for Electric Light Sources, School of Information Science and Technology, Fudan University, Shanghai 200433, China.
³ Zhongshan-Fudan Joint Innovation Center, Zhongshan 528437, China.
⁴ Yiwu Research Institute of Fudan University, Zhejiang 322000, China.

PMID: 35196010
DOI: 10.1021/acsami.1c24003

Abstract

Symbolic classification is an approach of interpretable machine learning for building mathematical formulas that fit certain data sets. In this work, symbolic classification is used to establish the relationship between oxygen vacancy defect formation energy and structural features. We find a structural descriptor n_a(r_a/E_na - r_b), where n_a is the valence of the a-site ion, r_a is the radius of the a-site ion, E_na is the electronegativity of the a-site ion, and r_b is the radius of the b-site ion. It accelerates the screening of defect-free oxide perovskites in advance of density functional theory (DFT) calculations and experimental characterization. Our results demonstrate the potential of symbolic classification for accelerating the data-driven design and discovery of materials with improved properties.

Keywords: machine learning; perovskite; structure descriptor; symbolic classification; vacancy defects.