CombiClean™: Facilitating Contaminant Removal in Recycled Plastics
The objective of the project is to develop a hyperspectral data base to enable more effective sorting and cleaning of secondary plastics feedstocks. The project will produce several tangible outcomes. An open source database, CombiClean™, will be developed, disseminated and archived in a publicly available repository. Hyperspectral characterization (combined FTIR, Raman, and LIBS) for model systems in virgin, contaminated, and cleaned conditions will be collected. Generated data will be used to train machine learning algorithms and demonstrate improved sorting. High throughput methods will be used to develop customized cleaning solutions based on specific contaminants incorporating enzymes. A process model will be populated by the cleaning data. Process economics and life-cycle impacts will be calculated to compare the new optimized processes against the present baseline of simple caustic/surfactants at high temperatures.
Project Team:
Michigan State University, Sealed Air
19-01-MM-02