Indentification of plastic waste using spectroscopy and neural networks

In this paper we investigate a new approach for the automated sorting of post-consumer plastic waste. We show that rapid and reliable identification of polymers can be achieved using a combination of fixed-filter near-infrared spectroscopy and neural network data analysis, and we demonstrate the effectiveness of the proposed method for sorting polyethylene terephthalate, high density polyethylene, and poly(vinyl chloride). Finally, we discuss a proposed compact, rugged instrument based on the new sorting method. Owing to the flexibility gained by incorporating neural networks in our system, this method can easily be extended to include additional polymers.