Automated Analysis of Internal Quantum Efficiency Using Chain Order Regression

Spectral analysis of internal quantum efficiency (IQE) measurements of solar cells is a powerful method to identify performance-limiting mechanisms in photovoltaic devices. This analysis is usually performed using complex curve-fitting methods to extract various electrical and optical performance parameters. As these traditional fitting methods are not easy to use and are often sensitive to measurement noise, many users do not utilize the full potential of the IQE measurements to provide the key properties of their solar cells. In this study, we propose a simplified approach to analyze IQE curves of silicon solar cells using machine learning models that are trained to extract valuable information regarding the cell's performance and decoupling the parasitic absorption of the anti-reflection coating. The proposed approach is demonstrated to be a powerful characterization tool for solar cells as machine learning unlocks the full potential of IQE measurements.