Evaluating UV-C-induced degradation of polyurethane and polystyrene: Comparing O-PTIR and portable ATR-FTIR

Understanding UV-C-induced polymer degradation is crucial for assessing and predicting the lifespan and performance of polymers in practical applications. Polyurethane (PU) and polystyrene (PS) are two widely used polymer materials that exhibit different degradation behaviours due to their molecular structure. In this study, we employed a portable Attenuated total reflectance - Fourier transform infrared spectroscopy (ATR-FTIR) and Optical Photothermal IR (O-PTIR) microscope combined with machine learning classification models to non-destructively investigate the chemical and structural changes that occur during the early stages of UV degradation. ATR-FTIR effectively captures the overall chemical changes, while O-PTIR provides higher spatial resolution, revealing local surface heterogeneity and degradation depth. In this study three classification approaches were evaluated: Partial Least Squares-Discriminant Analysis (PLS-DA), Support Vector Machine (SVM), and Random Forest classification using the collected spectral data to classify UV degradation stages. For the results, PLS-DA showed the highest accuracy, especially for O-PTIR data (PS:81 % & PU:80 %). SVM also performed well (PS:71 % & PU:82 %) and was able to effectively capture complex spectral relationships. This approach provides a reliable method for identifying UV degradation in polymers at early stages of exposure (from 4 days).