Report on the state-of-the-art and novel solutions in sorting of post-consumer plastic packaging waste

The purpose of this report is to review the state-of-the-art in post-consumer plastic packaging waste sorting in material and plastic recovery facilities, focusing on sensor-based sorting techniques. The report outlines the current situation in Finland with regard to recycling of this waste, as well as the different routes the waste fraction takes in terms of collection, treatment, separation and sorting. Example countries with different collection and sorting schemes are also listed. Statistics on recycling performance on the EU and Finnish levels are given.
The most common sensor-based tool for plastic identification in material and plastic recovery facilities is near-infrared spectroscopy (NIR), which is integrated to sensor-based sorting units. In short, NIR utilizes the reflectance spectrum of objects to infer their chemical composition. The NIR sensor works in tandem with a separation unit, usually pneumatic air nozzles, which use compressed air to separate objects based on the spectrum. The material is presented to the sorting unit usually via a conveyor belt when sorting whole containers. NIR has several limitations that hinder its all-around performance. Moreover, the actuation by air nozzles is imperfect, and is highly affected by the input rate of the material. Novel technological solutions leveraging different sensors with modern data-driven tools such as artificial neural networks have been studied to overcome some of these barriers. These solutions are shortly reviewed, and the levels of their technological maturity in the framework of post-consumer plastic packaging sorting evaluated.