In the context of the circular economy, the sorting process during recycling of polymers is essential as regards the efficiency of the process itself and the quality of the so-obtained recycled materials. In this paper, the application of Hyperspectral Imaging (HSI) is proposed for this purpose, providing additional insight into the state of aging and the polymer quality. The underlying study comprises HSI in the wavelength range of 1115-1678 nm considering artificially aged Styrene-Butadiene-Styrene (SBS), where aging is performed for 2, 5, 10, 15 and 20 days in a forced-draft oven at a temperature of 105 °C. The obtained HSI spectra are normalized using the Standard Normal Variate (SNV) method, with the normalized spectra as well as their first and second derivative entering the modeling attempt for SBS aging. For the latter, different partial least squares regression (PLSR) models are evaluated, where the original spectra achieved a correlation of R2=0.94 and a root mean squared error of prediction (RMSEP) of 1.83 days, showing the suitability of HSI for the proper identification of the state of aging of SBS and its potential use for other polymers.
Keywords: Aging; Hyperspectral Imaging; Prediction models; Recycling; SBS; Waste sorting.
Copyright © 2022 The Author(s). Published by Elsevier B.V. All rights reserved.