Design, upscaling and modeling of hollow fiber membrane modules for efficient CO2 separation

Carbon capture using membrane technologies represents a promising alternative to conventional amine-based absorption processes for the decarbonization of industrial emissions and flue gases. This work presents the design, experimental validation, and mathematical modeling of facilitated transport hollow fiber membrane modules for efficient CO₂ separation. Three modules with different shell geometries were designed and tested: two conventional cylindrical configurations (BIG 1 and BIG 2) and a novel conical configuration (BIG 3). Thin-film composite membranes were fabricated by dip coating hollow fibers with a solution containing sterically hindered poly(allylamine) (SHPAA), poly(vinyl alcohol) (PVA), the ionic liquid 1-ethyl-3-methylimidazolium acetate ([Emim][OAc]), and purified graphene oxide nanoplatelets (pGO). The experimental results demonstrated that module geometry significantly affects separation performance when identical membrane materials are used, highlighting the important role of hydrodynamic conditions and flow distribution in determining effective module performance. The conical module BIG 3 exhibited high performance and several distinctive advantages in gas velocity management, pressure gradient optimization, and suppression of back-diffusion. A one-dimensional transport model incorporating mass and momentum balances was developed and validated, showing deviations below 15% compared to experimental data. The analysis also highlighted humidity control as a critical factor for maintaining carrier activity in facilitated transport. This thesis demonstrates that geometric optimization of the module can substantially enhance separation performance without modifying membrane chemistry, providing design guidelines for scale-up toward industrial post-combustion CO₂ capture applications.