Modelling solubility in semi-crystalline polymers: a critical comparative review

The prediction of gases and vapors solubility isotherms in semi-crystalline polymers is still an unsolved problem of great technological relevance. This work provides an overview of the existing models developed for the assessment of solubility in semi-crystalline materials and presents a comparative discussion of their strengths and limitations. Experimental evidence suggests that the presence of impermeable crystalline domains is responsible for a reduced sorption capacity of the amorphous network compared to the unconstrained amorphous state. Three different modeling categories are distinguished. A set of models ascribes the reduced sorption capacity exhibited by the amorphous phase to an increased density state induced by the confinement provided by the crystals, which is represented with the application of a hydrostatic constraint pressure term. Using a different perspective, other modeling strategies assume the fraction of elastically effective chains, or 'tie-chains' linking crystal domains, to be responsible for increased activity of the penetrating molecules in the matrix and consequently for their lower solubility. Finally, the last category of models proposes the application of the Non-Equilibrium Lattice Fluid theory to the amorphous phase, which is regarded as in a pseudo glassy state, due to the presence of the surrounding crystallites, which are responsible for a hindered structure mobility. The theoretical foundations of the listed approaches are presented, and the relative results are collected and discussed. In conclusion, some prospective work guidelines for modeling strategies improvement are provided. (C) 2022 Elsevier B.V. All rights reserved.

Automated summary

This study provides an overview and comparative discussion of existing models for predicting the solubility of gases and vapors in semi-crystalline polymers. The presence of impermeable crystalline domains is found to reduce the sorption capacity of the amorphous network. Three different modeling categories are distinguished, and their theoretical foundations and relative results are presented and discussed. Prospective work guidelines for improving modeling strategies are also provided.