Effects of molecular architecture on fluid ingress behavior of glassy polymer networks

Matthew Blaine Jaskson

Abstract

This manuscript demonstrates the synthesis of glassy polymer network isomers to control morphological variations and study solvent ingress behavior independent of chemical affinity. Well-controlled network architectures with varying free volume average hole-sizes have been shown to substantially influence solvent ingress within glassy polymer networks. Bisphenol-A diglycidyl ether (DGEBA), bisphenol-F diglycidyl ether (DGEBF), Triglycidyl p-aminophenol (pAP, MY0510), Triglycidyl maminophenol (mAP, MY0610), and tetraglydicyl-4,4'-diamino-diphenyl methane (TGDDM, MY721) were cured with 3,3'- and 4,4'-diaminodiphenyl sulfone (DDS) at a stoichiometric ratio of 1:1 oxirane to amine active hydrogen to generate a series of network architectures with an average free volume hole-size (V h ) ranging between 54-82 Å3 . Polymer networks were exposed to water and a broad range of organic solvents ranging in van der Waals (vdW) volumes from 18-88 Å3 for up to 10,000h time. A clear relationship between glassy polymer network Vh and fluid penetration has been established. As penetrant vdW volume approached Vh , uptake kinetics significantly decreased, and as penetrant vdW volume exceeded Vh , a blocking mechanism dominated ingress and prevented penetrant transport. These results suggest that reducing the free volume hole-size is a reasonable approach to control solvent properties for glassy polymer networks. New techniques to monitor and predict the diffusion behavior of liquids through glassy networks are also presented. Digital Image Correlation (DIC) was employed to accurately measure the strain developed during case II diffusion. This technique also presented a new theory for a relationship between sample topology and irreversible macroscopic brittle failure induced by solvent absorption. A new modeling technique has been developed which can accurately predict the chemical and physical interactions a solvent may have with a glassy network. This new model can be used as a qualitative tool to screen for relative fluid resistance of new materials.