Accelerated Discovery of Two-Dimensional Optoelectronic Octahedral Oxyhalides via High-Throughput Ab Initio Calculations and Machine Learning

Xing-Yu Ma¹, James P. Lewis^2,3,4, Qing-Bo Yan^5*, Gang Su^1,6*

¹School of Physical Sciences, University of Chinese Academy of Sciences, Beijing 100049, China.

²Department of Physics and Astronomy, West Virginia University, Morgantown, West Virginia 26506-6315, United States

³State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan, Shanxi 030001, China

⁴Beijing Advanced Innovation Center for Materials Genome Engineering, Beijing Information S&T University, Beijing 101400, China

⁵Center of Materials Science and Optoelectronics Engineering, College of Materials Science and Optoelectronic Technology, University of Chinese Academy of Sciences, Beijing 100049, China.

⁶Kavli Institute for Theoretical Sciences, and CAS Center of Excellence in Topological Quantum Computation, University of Chinese Academy of Sciences, Beijing 100190, China

ABSTRACT: Traditional trial-and-error methods are obstacles for large-scale searching of new optoelectronic materials. Here, we introduce a method combining high-throughput ab initio calculations and machine-learning approaches to predict two-dimensional octahedral oxyhalides with improved optoelectronic properties. We develop an effective machine-learning model based on an expansive dataset generated from density functional calculations including the geometric and electronic properties of 300 two-dimensional octahedral oxyhalides. Our model accelerates the screening of potential optoelectronic materials of 5,000 two-dimensional octahedral oxyhalides. The distorted stacked octahedral factors proposed in our model play essential roles in the machine-learning prediction. Several potential two-dimensional optoelectronic octahedral oxyhalides with moderate band gaps, high electron mobilities, and ultrahigh absorbance coefficients are successfully hypothesized.