Factors Affecting Solvent Retention due to Gel Formation during Dissolution-Based Plastic Recycling

Abstract

Dissolution-based recycling provides a way to recover polymers from mixed plastic waste, but a large amount of energy is required to remove residual solvent after precipitation.1,2 In this study, we measured the solvent retention of 34 polymer-solvent systems consisting of eight polymers and nine solvents to understand what controls solvent retention. Solvent retention ranged from 20.16 to 91.65%, showing that different polymer-solvent pairs behave very differently. Cooling rate, precipitation method, and filtration time did not significantly change solvent retention. Instead, solvent volatility, polymer molecular weight, and polymer-solvent affinity were the main variables that have influence on the amount of solvent remaining in polymer-solvent gels. Volatile solvents such as xylene consistently exhibits low solvent retention, while solvents with lower vapor pressure resulted in higher retention.3 A support vector regression model was developed to predict the solvent retention using multiple features. A leave-one-out technique was used to quantify the contribution of each feature. Initially, a six-feature model was developed and further feature reduction to five-, and four-feature models was performed based on leave-one-out results. Among them, five-feature model shows the highest R2 of 0.78 and the lowest Akaike information criterion and Bayesian information criterion values, suggesting that five-feature without the Flory-Huggins interaction parameter is the ideal for predicting the solvent retention. Vapor pressure was the most important variable, followed by polymer molecular weight.

This work can contribute to framework for solvent selection and for predicting solvent retention in more complex feed stocks such as multilayer waste. This approach may also guide more efficient STRAPTM process design.