Related Experiment Video
Updated: May 22, 2025
11:38
Influence of Hybrid Perovskite Fabrication Methods on Film Formation, Electronic Structure, and Solar Cell Performance
Published on: February 27, 2017
18.4K
Multidimensional high-throughput screening for mixed perovskite materials with machine learning.
Chengbing Chen1, Jianrong Xiao1,2, Zhiyong Wang1,2
1College of Physics and Electronic Information Engineering, Guilin University of Technology, Guilin 541008, China.
The Journal of Chemical Physics
|March 17, 2025
Summary
Researchers developed a machine learning strategy to efficiently screen mixed halide inorganic perovskites (MHIPs) for solar cell applications. This approach identified promising MHIPs with optimal bandgap and light absorption properties, accelerating materials discovery.
Area of Science:
- Materials Science
- Photovoltaics
- Computational Chemistry
Background:
- Mixed halide inorganic perovskites (MHIPs) show great promise for photovoltaic applications due to their stability and performance.
- The vast chemical space of MHIPs makes identifying optimal materials challenging through traditional experimental or theoretical methods alone.
Purpose of the Study:
- To develop and validate a multidimensional high-throughput screening strategy combining machine learning and first-principles calculations.
- To identify MHIPs with optimal bandgap and light absorption properties for enhanced photovoltaic performance.
Main Methods:
- A machine learning approach integrated with first-principles calculations was employed for high-throughput screening.
- Models for bandgap and light absorption were developed and validated, achieving high accuracy (r2 > 0.98).
- Density functional theory (DFT) was used to validate a subset of the screened materials.
Main Results:
- Generated over 306,000 candidate MHIP materials by varying B-site elements.
- Successfully screened 295 materials with desirable characteristics for MHIP applications.
- Identified Cs2AgBi0.5Sb0.25Ir0.25I6 and CsSn0.75Ge0.25I3 as highly promising candidates after DFT validation.
Conclusions:
- The combined machine learning and first-principles screening strategy is effective and scientifically valid for discovering high-performance MHIPs.
- The identified materials provide a strong foundation for further research and optimization of perovskite solar cells.
- This approach significantly accelerates the discovery process for advanced photovoltaic materials.
