Mechanistic Modeling of the Thermal Degradation of Hydroxyl-Containing Polymers: Part 2 – Product Distribution, Reaction Pathways, and Rates

In the first part of this series of manuscripts, the focus was on discovering and delineating families of elementary step reactions and describing how the method of moments and the use of deterministically arranged dyad probabilities can be implemented within a population balance model to describe the thermal deconstruction of hydroxy-containing polymers. In this part, the results from the simulation of the pyrolysis of poly(vinyl alcohol) (PVA) and ethylene vinyl alcohol (EVOH) are compared to experimental thermogravimetric analysis and pyrolysis data obtained by carrying out using a micropyrolyzer coupled to two-dimensional chromatography, along with flame ionization and time-of-flight mass spectrometer detectors (Py–GC×GC–FID/TOF–MS). Both the PVA and EVOH models, using a heating rate of 10 °C min–1, were able to replicate experimentally determined mass-loss curves. In doing so, the difference in onset temperatures between the thermal deconstruction of PVA and EVOH was revealed to be the result of ethyl comonomers in EVOH chains inhibiting the rapid end-chain unzipping chemistries observed in PVA. Additionally, at low pyrolysis isothermal temperatures (340 and 360 °C), the EVOH model was able to reproduce the partial conversion yields of four major linear products, ethanal, 2-butenal, propene, and acetone, identifying the underlying chemical reactions that govern the thermal degradations of these polymers. After validating the model, net rate analysis was used to identify the major pathways for degradation in the PVA and EVOH systems, namely, assisted 1,2-dehydration, Grob-type 1,3-diol fragmentation, retro-aldol and (non)-adjacent retro-ene reactions. Lastly, to highlight the capabilities of this novel methodological framework and the reaction families at the elementary step level that it contains, this work further quantified the sensitivity of the degradation systems to backbone impurities and differing comonomer concentrations. With the successful validation of these models, this study has demonstrated a novel method of tracking the evolution of midchain topology probabilistically, expanding the suite of polymer pyrolysis systems that can be modeled using the method of moments, and presents the most detailed models of PVA and EVOH pyrolysis to date.