(173aq) A KMC Based Tool to Understand the Chemical Recycling of Polyurethanes

Polyurethanes (PUs) are the sixth most highly produced polymers globally and find a variety of applications like mattresses, insulation foams, automobile parts, etc. Traditionally, polyurethanes are formed by the addition reaction between diisocyanate and diol monomers. On the one hand, the raw materials used to produce linear and cross-linked PUs pose serious health hazards and environmental risks. On the other hand, conventional PUs offer limited recyclability and are incinerated, releasing toxic chemicals into the atmosphere. Solvolysis techniques like glycolysis and methanolysis have the potential to recover the monomers efficiently. Therefore, it is imperative to understand the chemistry of the PU solvolysis process at various reaction conditions to achieve higher monomer recovery rates.

The existing kinetic studies provide little to no information on the influence of reaction conditions, polymer structure, or molecular sequences on monomer recovery. These models generally track lumped polymeric species as a function of temperature or time. Tracking specific chain sequences and changes in the morphological aspects is essential as PUs consist of two monomers arranged randomly. Kinetic Monte Carlo (KMC) based models can help address this issue by explicitly tracking the sequences of all the polymeric chain species.

The current study aims at developing a KMC framework to unravel the depolymerization pathways of linear PUs. An in-house KMC model developed to generate linear PU chain sequences is used to initialize the polymer chains (hard and soft segments). Further, a set of reaction families are defined to track the depolymerization as single events. The rate parameters for solvolysis were optimized to fit the molecular weight distributions. The recovered monomer yields were validated against the experimental data available in the literature. Finally, a reaction map was plotted to understand the change in reaction dynamics – random vs. chain-end scissions – with solvent concentration, time, and temperature.