Near‐Infrared Spectroscopy and Chemometrics as a Tool for Nondestructive Polymer Degradation Analysis in Photovoltaic Modules

The long‐term reliability of photovoltaic (PV) modules is crucial, and polymeric components play an important role in ensuring it. This makes nondestructive methods for detecting material degradation valuable and essential. This study proposes and validates a methodology that utilizes near‐infrared (NIR) spectroscopy and chemometric techniques to identify degradation effects in PV encapsulants and backsheets. Principal component analysis (PCA) and random decision forest (RDF) algorithms are applied to preprocessed NIR spectra to detect changes and their chemical or physical origin. Accelerated damp heat and UV weathering tests on ethylene‐vinyl acetate (EVA) or polyolefin elastomer (POE) encapsulants, as well as polyethylene terephthalate (PET) backsheets, reveal degradation‐induced spectral changes in specific regions of the NIR spectra. Hydrolysis degradation in the ‐OH combination region (5000–5300 cm −1 ) is compared to changes due to physical water absorption. In UV‐aged PET, photo‐oxidation is identified within the C=O combination region (5200–5310 cm −1 ). The findings are validated through complementary destructive analyses, which confirm the chemical origin of the observed spectral shifts. The results demonstrate that NIR spectroscopy, in combination with chemometric techniques, enables nondestructive identification of severe polymer degradation in PV modules. The developed methodology demonstrates NIR's potential to complement conventional laboratory analyses.