Related Experiment Video
Updated: Feb 17, 2026
11:38
Influence of Hybrid Perovskite Fabrication Methods on Film Formation, Electronic Structure, and Solar Cell Performance
Published on: February 27, 2017
19.1K
Performance optimization and machine learning-guided parameter sensitivity analysis of lead-free KGeCl3 perovskite
Tanzir Ahamed1,2, Md Mehedi Hasan Bappy1,2, Mohammad Rahimul Islam1
1Department of Electrical and Electronic Engineering, CCN University of Science and Technology Cumilla-3503 Bangladesh tanzir.eee2k15@gmail.com.
RSC Advances
|February 16, 2026
Summary
This study explores lead-free Germanium-based perovskite solar cells (PSCs) with KGeCl3 absorbers. The WS2 electron transport layer (ETL) configuration achieved the highest power conversion efficiency (PCE) of 21.39%.
Area of Science:
- Materials Science
- Renewable Energy
- Photovoltaics
Background:
- Perovskite solar cells (PSCs) offer promising photovoltaic performance.
- Developing lead-free PSCs is crucial for environmental sustainability.
- Germanium-based perovskites are emerging as a viable alternative to lead-based materials.
Purpose of the Study:
- To investigate the performance of lead-free Ge-based PSCs.
- To evaluate different electron transport layers (ETLs) for KGeCl3 absorber-based devices.
- To optimize PSC parameters using simulation and machine learning.
Main Methods:
- Device simulation using SCAPS-1D to optimize material parameters (thickness, doping, defects, resistance, temperature, work function).
- Fabrication and characterization of PSCs with KGeCl3 absorber and various ETLs (WS2, ZnSe, PC60BM, SnS2).
- Application of machine learning (Random Forest, XGBoost, CatBoost, Decision Tree) for feature importance analysis.
Main Results:
- The FTO/CFTS/KGeCl3/WS2/Au solar cell configuration demonstrated the highest power conversion efficiency (PCE) of 21.39%.
- Other ETLs yielded PCEs of 21.38% (ZnSe), 21.05% (PC60BM), and 20.43% (SnS2).
- CatBoost machine learning model achieved high accuracy (99.344%) and R² (0.984) in evaluating material feature importance.
Conclusions:
- Lead-free Ge-based PSCs with KGeCl3 absorbers are highly effective.
- WS2 is an optimal ETL for enhancing the performance of these solar cells.
- Machine learning provides valuable insights into material parameter optimization for PSCs.
