Membrane based separations of carbon dioxide and phenol under supercritical conditions

Polyimide membranes have been implemented in a variety of applications due to their attractive productivities, selectivities, and processing characteristics. Their use in high pressure carbon dioxide environments has not been well investigated; however, increasing industrial interest in using supercritical CO2 in place of organic solvents as a processing agent makes this study important. Many potential end uses of CO2 as an industrial solvent will require purification of contaminated high pressure CO2 streams with further recycling of the CO2 to prevent significant emission into the atmosphere. Well designed membranes, which selectively partition the organic solutes into the material and transfer it out of the effluent stream, leaving the CO2 at high pressure, offer a low cost and high productivity alternative to de-pressurization. The focus of this project is to study the effects of high pressure carbon dioxide on the properties of polyimide materials and to successfully separate the organic solute, phenol, from CO2. Investigation of the transport properties of 6FDA-based polyimide materials in the presence of pure carbon dioxide at high pressures showed interesting permeability trends, in which the permeabilities exhibited a maximum after which they declined with further increases in pressure. These ìconditionedî polymers showed decreased chain spacing and swelling effects as compared to untested samples, indicating a possible lattice collapse of the material. It seems that the competition between the CO2 induced swelling tendency is at some point balanced by the tendency for a non-equilibrium high free volume polymer to relax to its equilibrium state. The permeability maximum occurs when the two effects are equal, after the volume contraction is dominant. By modeling transport through the membrane starting with Fickís law, and taking into account convective contributions to transport, an unconventional approach, termed ìSorp-Vectionî, to describing mixed gas transport through a membrane was developed. This model predicted a high separation efficiency of 79 for the organic solute, phenol, in a glassy polymer, 6FDA-DAM. Experimental separation efficiencies of 40 ñ 44 were obtained for this polymer, which represent 59% of the efficiency predicted via modeling. These experimental results provide a proof of concept of bulk flow based separations of phenol from CO2. These experiments verify that (1) the separation of phenol from carbon dioxide is possible using glassy materials, (2) these separation factors are large enough for industrial viability, and (3) bulk flow effects are working to enhance the separation effectiveness.