Time-Efficient, Accurate, and Experimentally Grounded Optical Modeling of Multiscale-Textured Thin-Film Solar Cells

Area of Science:

• Photovoltaics and Renewable Energy
• Optoelectronics and Nanophotonics
• Computational Materials Science

Background:

• Accurate optical performance prediction is crucial for optimizing light management in advanced solar cells.
• Thin-film silicon (TF Si) solar cells require efficient modeling for next-generation photovoltaic development.
• Multiscale-textured interfaces present challenges for precise optical simulations.

Purpose of the Study:

• To systematically validate and compare rigorous coupled-wave analysis (RCWA) and ray optics modeling for TF Si solar cells.
• To assess the accuracy and computational efficiency of ray optics against RCWA for various device architectures.
• To establish ray optics as a reliable predictive tool for thin-film solar cell optical performance.

Main Methods:

• Experimental fabrication of thin-film silicon solar cells with random nanotextures and micro-periodic honeycomb textures.
• Application of rigorous coupled-wave analysis (RCWA) for full electromagnetic simulation.
• Implementation of a ray optics model operating in the refractive regime.
• Benchmarking simulated versus measured external quantum efficiency (EQE) and reflection losses using root mean squared error (RMSE).

Main Results:

• Ray optics model demonstrated accuracy comparable to RCWA, with deviations of only 2%-6% in optical performance prediction.
• Ray optics significantly reduced computation time from one week to under 30 minutes.
• Ray optics accurately predicted the optical response of a tandem TF Si solar cell, with spectral deviations below 6% and photocurrent mismatch under 0.2 mA/cm².

Conclusions:

• Ray optics modeling is a reliable and computationally efficient predictive tool for thin-film solar cell optical performance.
• The ray optics approach is broadly transferable to various thin-film technologies, including perovskite solar cells.
• Accurate optical constants and realistic interface morphologies are key for the success of ray optics simulations.