Sixteenth International Conference on Environmental, Cultural, Economic & Social Sustainability

Abstract

This project involved constructing a prototype of polymer resin identification system based on mid-infrared (MIR) reflection spectroscopy. The MIR reflectance spectrum contains the chemical information of the material. This fingerprint, in contrast to the popular near-infrared (NIR) spectroscopy, contains much more molecular vibrational resonance information, which we use to construct a multi-spectral and multi-dimensional library of all plastics commonly encountered in the municipal solid waste stream. The main component of the system is the spectroscopic optical reading system. With this element, the MIR reflection spectrum is measured to retrieve the chemical information of the sample. This important component must be carefully designed to ensure the acquisition of a high signal to noise ratio needed to accurately identify plastics under field-type working conditions. We design and customize a commercial MIR spectrometer and a high power IR source to integrate them into a hand-held unit. Once the first prototype version was built, characterization experiments were performed to evaluate and optimize the system performance in several aspects including its fidelity of the measured optical and detection algorithm. The objectives of the research are to: examine the effect of plastic contamination and composition on the efficiency of the proposed infrared plastics detection and differentiation; adapt the laboratory setup to a field compatible configuration, and; identify an algorithm upgrade to minimize the detection uncertainties.