Integrated COSMO-RS screening and experimental validation of choline chloride-based deep eutectic solvents for sustainable polyurethane depolymerization

Polyurethane (PU) waste has become a serious environmental issue, which has increased the demand for more sustainable and efficient recycling methods. Among these, deep eutectic solvents (DESs) have gained attention as green solvents for PU depolymerization because of their environmentally friendly properties. In this study, a computational screening method based on performance and sustainability indices was used to study the suitability of 10 choline chloride (ChCl)-based DESs for this purpose. This assessment combined the performance index from the Conductor-like Screening Model for Real Solvents (COSMO-RS) with the sustainability index from the Analytical GREENness (AGREE) metric and biodegradability values. Both indices were evaluated for the selected DESs by assigning a 55% weighting to the sustainability index and 45% to the performance index to emphasize the sustainability factor alongside performance. This method allowed for a more balanced assessment of depolymerization efficiency and environmental impact. The computational results showed that amide-based DESs, especially ChCl-Urea, demonstrated better solvation performance, with selectivity and capacity values of 0.741 and 1.718, respectively. Based on the screening results, ChCl-Urea and ChCl-Glycerol were selected as the most promising candidates because both DESs exhibited favorable solvation performance, lower toxicity, and improved biodegradability compared to the other systems studied. Experimental validation performed at 170℃ for 480 min supported the computational predictions: ChCl-Urea achieved complete (100%) PU degradation, whereas ChCl-Glycerol showed only a 29.87% degradation efficiency. The superior performance of ChCl-Urea was associated with its strong cooperative hydrogen-bonding interactions, which promoted the activation and cleavage of PU carbamate linkages. Overall, this study demonstrates that integrating computational prediction with sustainability evaluation can provide a practical and reliable approach for selecting greener solvents for advanced PU recycling applications.