Pilot Test of a Nanoporous, Super-hydrophobic Membrane Contactor Process for Post-combustion CO<sub>2</sub> Capture

Abstract

GTI Energy and Air Liquide Advanced Separations (ALaS) have been developing a novel hollow fiber membrane contactor (HFMC) technology for post-combustion CO2 capture. The process combines advantageous features of both absorption and membrane-based separation processes to separate CO2 from flue gas cost-effectively. The key component of the HFMC technology is the super-hydrophobic, porous hollow fiber, which is made from polyether ether ketone (PEEK). Compared to conventional absorption/desorption technologies, the critical advantage of the HFMC process is the high contact surface area provided by the hollow fibers enabling an increased volumetric mass-transfer rate. In the PEEK HFMC process, the specific surface area has been increased by an order of magnitude over structurally packed or trayed columns, resulting in compact systems with small footprints. A pilot-scale demonstration of the HFMC process on coal-fired power plant flue gas has been performed in Wilsonville, AL at the National Carbon Capture Center (NCCC) treating flue gas from pulverized coal-fired Alabama Power’s Gaston Power Station. A 90% CO2 removal rate was achieved by the HFMC using a 50 wt.% aMDEA solvent during the initial tests with 4 modules and actual coal-fired flue gas at NCCC. The stripped stream from the two-stage flash desorber had a CO2 concentration of >98.6 vol%. Further tests indicated an issue of liquid-side concentration polarization – higher CO2 concentration in the fluid boundary layer (next to the fiber) relative to the bulk flow stream. This issue was resolved by decreasing the aMDEA concentration from 50 wt.% to 35 wt.%. Continuous testing with 28 membrane modules, however, did not match the single module results; the CO2 capture performance declined with time. Quantitative analysis as well as inspection and measurements of the spent modules were conducted to investigate the potential causes. The major issue identified was the tubesheet leaking from patch points and potentially from fiber/epoxy separation. Future steps would include: 1) resolving technical hurdles in materials and manufacturing; 2) increasing the inner diameter of the hollow fibers to achieve a low pressure drop (when flue gas flows through the hollow fibers); and 3) consideration of inclusion of multiple membrane cartridges in one housing. Overall, this project has advanced the HFMC technology to a high TRL level and resolved a number of technical issues (e.g. concentration polarization) that other researchers have not dealt with to date. The project results and publications is a significant contribution to the literature and for other technology developers.