Options
2023
Conference Paper
Title
Characterization of thin-film structures of silicon heterojunction solar cells with inline reflectance spectroscopy
Abstract
In this work, we explore reflectance spectrophotometry as a fast, inline-compatible quality-assurance method to extract thin-film layer thicknesses on silicon heterojunction (SHJ) solar cell precursors. We develop an optical model based on the transfer-matrix method (TMM) and so-called rigorous polarization ray-tracing technique for the model-based fitting of experimental reflectance spectra. First, individual layers of both amorphous silicon (a-Si) and indium-doped tin-oxide (ITO) on planar surfaces are measured with both an inline and an offline spectrophotometer to compare the fit results and to understand the quality of the optical constants that are necessary as input parameters to perform the model-based fitting. We find that using effective optical constants based on spectral ellipsometry (SE) results in reliable fitted a-Si stack and ITO thicknesses in the case of individual layers on planar surfaces. Second, we analyze single layers of ITO on a textured surface, finding a reasonable agreement with a mean relative deviation of 5% between the thickness values based on inline and offline reflectance spectra. We also compare the results of the reflectance fits to thicknesses extracted from scanning electron microscopy (SEM) images, finding a mean relative deviation of 10.7% between the two methods. Third, we perform a reflectance analysis of a-Si/ITO stacks on textured surfaces typical for SHJ solar cells and find that while the ITO thickness can be fitted to an accuracy of 4% in mean relative deviation compared to SEM images, the a-Si stack thicknesses are overestimated by 18-51% compared to reference measurements based on SE. Subsequently, we discuss the reasons for the deviations between the model-based reflectance fitting and the reference methods, and consider possibilities to improve the agreement.
Author(s)