(154b) Computational High-Throughput Screening of Glycerol-Derived Solvents for Plastic Waste Recycling

Although plastics are essential in modern life, their use results in significant waste that ends up in landfills or in the natural environment. To combat the effects of plastic pollution, efforts are being amplified on the regulatory and technological levels to increase plastic waste recycling. Solvent-based recycling, whereby a solvent selectively dissolves plastic material which can then be recovered and reprocessed into upcycled products, represents a promising method to scale up plastic recycling. Most known applications of solvent-based recycling disclosed in industrial patents and in academic literature, however, make use common solvents which may pose potential health and environmental risks. Therefore, it is crucial to develop ‘green’ solvents for solvent-based plastic waste recycling. For this purpose, we propose the functionalization of glycerol, a natural product from biomass as well as a byproduct of biodiesel production, that can produce new and potentially less harmful solvents capable of dissolving plastic waste. In this work, we have developed a computational methodology based on a combination of Density Functional Theory, machine learning and Hansen theory which allows for the high-throughput screening of thousands of solvent candidates for plastics commonly found in waste streams. Consequently, we have developed a database of ~10,000 glycerol-based solvents and calculated their physical properties such as the dipole moment, density, melting point, LogS, LogP, Hansen Solubility Parameters (HSPs), and based on the HSPs of the solvents and selected plastics in addition to the melting point of the solvents, we give recommendations for solvents that can carry out selective plastic dissolution for complex plastic waste feedstock.

Reference: Ind. Eng. Chem. Res., 2023. accepted DOI: 10.1021/acs.iecr.2c04567.