High Passivation Performance of Cat-CVD i-a-Si:H Derived from Bayesian Optimization with Practical Constraints.

Area of Science:

• Materials Science
• Renewable Energy
• Semiconductor Physics

Background:

• High-efficiency silicon heterojunction (SHJ) solar cells rely on intrinsic hydrogenated amorphous silicon (i-a-Si:H) for effective passivation.
• Achieving optimal i-a-Si:H quality necessitates precise control over hydrogen content and film density.

Purpose of the Study:

• To discover i-a-Si:H deposition conditions that maximize passivation performance using Bayesian optimization (BO).
• To establish a practical BO scheme incorporating prediction models and experimental constraints for materials science applications.

Main Methods:

• Utilized catalytic chemical vapor deposition combined with Bayesian optimization (BO) to explore deposition parameters.
• Developed a practical BO scheme with multiple prediction models to handle experimental constraints.
• Analyzed optimal conditions using contour plots to understand parameter interactions.

Main Results:

• Achieved a high effective minority carrier lifetime (τ_eff) exceeding 2.6 ms within 8 BO cycles.
• Identified a critical parameter space between substrate temperature and SiH₄ flow rate for high τ_eff.
• Demonstrated that controlling these parameters inhibits high void fraction and epitaxial growth, ensuring high passivation quality.

Conclusions:

• Bayesian optimization is an effective tool for discovering optimal deposition conditions for i-a-Si:H passivation in SHJ solar cells.
• The substrate temperature and SiH₄ flow rate are crucial parameters for achieving high passivation quality by preventing defects.
• The developed BO scheme and analysis methods are transferable to optimizing other semiconductor deposition processes, including plasma-enhanced chemical vapor deposition for SHJ solar cells.