Spectroscopic analysis with piecewise quantification model for accurate PET/cotton blend ratio determination

The qualitative and quantitative analysis of polyethylene terephthalate /cotton blended textiles is critical in textile production, trade and scientific research. In this study, Raman and infrared spectroscopy was innovatively integrated to establish a segmented quantitative model based on Raman characteristic peak ratios (1610 cm^-1/1090 cm^-1 and 1720 cm^-1/1090 cm^-1) and an infrared characteristic peak ratio (1720 cm^-1/1020 cm^-1). The accuracy of blend ratio measurements for self-prepared samples was evaluated using the coefficient of determination (R²), mean relative error (MRE), root mean square error (RMSE) and principal component analysis (PCA) method. Results indicate that the MREs for the three sets of characteristic peak ratios were 2.34 %, 6.37 %, and 22.75 %, respectively, with overestimated MREs of 10.12 %, 9.74 %, and 34.79 %. The root-mean-square errors (RMSEs) corresponded were 3.7647, 6.4006, and 10.5306, respectively. The combined Raman and infrared spectroscopy approach achieved 90 % quantitative accuracy and 100 % qualitative accuracy for commercial PET/cotton blended fabrics. This method demonstrates superior performance in detecting PET/cotton blends, providing a foundation for establishing rapid, non-destructive, and efficient analytical techniques for textile industries.