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

TL;DR

This paper introduces a fast and accurate optical modeling method for thin-film solar cells with complex surface textures, validated against experimental data.

Contribution

The first systematic validation of ray optics for modeling multiscale-textured solar cells, showing high accuracy and significant time savings.

Findings

Ray optics model achieves 2%–6% deviation compared to experimental data, matching RCWA accuracy.

Computation time is reduced from 1 week to under 30 minutes using ray optics.

The method successfully predicts performance for both single-junction and tandem solar cell architectures.

Abstract

Accurate prediction of optical performance in solar cells with multiscale‐textured interfaces is essential for optimizing light management in next‐generation photovoltaics. For the first time, a systematic validation of two complementary modeling approaches is carried out on experimentally fabricated thin‐film silicon (TF Si) solar cells: rigorous coupled‐wave analysis (RCWA), offering a full electromagnetic solution but constrained by boundary conditions, and a ray optics model, operating in the refractive regime. The study involves two device architectures: an a‐Si:H single‐junction cell on commercial Asahi VU‐type glass with random nanotextures, and an nc‐Si:H single‐junction cell on novel micro‐periodic honeycomb‐textured glass developed in‐house. Simulated and measured external quantum efficiency (EQE) and total front reflection losses (1‐R) are benchmarked using the root mean squared error (RMSE). The ray model shows deviations of only 2%–6%, comparable to RCWA, while reducing computation time from 1 week to less than 30 min. Applied to an a‐Si:H/nc‐Si:H tandem device on honeycomb‐textured glass, ray optics reproduced the optical response with spectral deviations below 6% and photocurrent mismatch under 0.2 mA/cm2. These findings uniquely establish ray optics, when combined with accurate optical constants and realistic interface morphologies, as a reliable and computationally efficient predictive tool broadly transferable to thin‐film technologies, including perovskites.

A time‐efficient and accurate optical modeling framework combines ray tracing with measured refractive index (n) and extinction coefficient (k) data, together with atomic force microscopy (AFM) surface profiles, to capture multiscale textures in thin‐film solar cells. The method reproduces reflection losses and predicts optical device performance across single‐junction and tandem architectures, and is readily applicable to next‐generation photovoltaic devices.

Figures