Methodology development for management of recyclable polymers by Fourier transform infrared spectroscopy and PLS-DA

In this study, samples of recyclable polymers representing the seven standard classes were analysed using Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy (ATR-FTIR) combined with a classification technique. The classification models were developed using Partial Least Squares Discriminant Analysis (PLS-DA), and its performance was assessed based on sensitivity, specificity, accuracy, efficiency, and the Matthews's Correlation Coefficient (MCC). The PLS-DA models demonstrated high sensitivity, specificity, accuracy, and efficiency, with values ranging from 97.3 % to 100 %, except for the "Other" class, which showed values between 63.2 % and 97.7 %. The Matthews's Correlation Coefficient values ranged from 0.92 to 1.00, except for the "Other" class, which had a value of 0.61, indicating strong overall performance of the classification models. This work demonstrates that PLS-DA combined with ATR-FTIR spectroscopy is a highly effective analytical tool in the identify of polymers for the reprocessing of recyclable materials.