(447c) Computational Prediction of Polymer Solubilities and Its Application in Multilayer Plastic Film Recycling

Billions of pounds of multilayer plastic films are produced annually as packaging materials. These films are made of distinct polymer layers, which makes it a challenging task to recycle them. Recently, we developed a strategy called solvent-targeted recovery and precipitation (STRAP) that can deconstruct and recycle such films through a series of solvent washes. It utilizes the solubility difference of the polymers to selectively dissolve them in carefully chosen solvent systems. One key step in designing a STRAP process for a target film is to identify selective solvent systems and operating temperatures, which could be prohibitively time-consuming using experimental screening alone. Here, we report a computational scheme for polymer solubility prediction and solvent selection that employs Hansen solubility parameters, molecular dynamics simulations, and the COnductor-like Screening Model for Realistic Solvents. These tools take molecular interactions and conformational information into consideration to yield quantitative, temperature-dependent solubility predictions of polymers in pure solvents and solvent mixtures. We demonstrate that this computational scheme can efficiently screen appropriate solvents to guide the design of solvent-mediated processes for multilayer plastic film recycling.