(566d) Effects of Moisture Sorption on Multiscale Gas Transport in Peg-Based CO2 Separation Membranes

Efficiently removing CO₂ from dilute and concentrated gas mixtures remains a present-day separation challenge. Poly(ethylene glycol) diacrylate (PEGDA) based membranes have shown great utility for CO₂ separations due to their high CO₂ sorption selectivity arising from a high density of ether oxygens. However, many CO₂ containing process streams contain significant amounts of moisture which affect membrane separation performance. Sorbed water vapor often acts as a plasticizing penetrant which increases membrane effective fractional free volume (FFV) and can also alter gas-polymer interactions. Despite the importance of sorbed water on the separation performance of rubbery polymer membranes, the molecular origins of mixed gas/vapor separation properties of PEGDA membranes have not yet been studied in detail.

This study aims to quantify changes to PEGDA copolymer structures under controlled humidity conditions and the resulting effects on both microscopic and macroscopic gas transport. Water vapor was found to act in some cases as an anti-plasticizing agent, decreasing gas permeabilities at moderate relative humidity (<50%), and a plasticizer at higher relative humidities (>50%), increasing gas permeabilities compared to dry conditions. Changes to membrane effective FFV due to water uptake were quantified using water vapor sorption and dilatometry experiments. These measurements yield further insights into how both water and gas diffusion are affected by relative humidity. High field Pulsed Field Gradient (PFG) NMR was used to directly measure CO₂ self-diffusion coefficients for the first time in PEG-based membranes, which are compared with macroscopic gas transport diffusivities. These unique insights across multiple length and timescales lay the groundwork for a more fundamental understanding of multicomponent transport in sorption-selective membranes for CO₂ separations.

This material is based upon work supported by the National Science Foundation under Grant No. 2427603.