CO2 Separation and Fuel Desulfurization Involving Room-Temperature Ionic Liquids

Imidazolium based room-temperature ionic liquids were used to investigate the technical and economic feasibility of energy-related chemical separations. The separations conducted for this work are sulfur-removal from fuels and carbon dioxide removal from flue gas emissions. In the sulfur-removal experiments, the sulfurcontaining compound dibenzothiophene was removed from octane by liquid-liquid extraction, yielding partition coefficients with ionic liquid around 1.5. Attempts to regenerate the ionic liquid using vacuum filtration were not fully successful in removing dibenzothiophene; extractions with the regenerated ionic liquid had a 50% decreased capacity. For this process to become economically feasible in comparison to the currently used hydrotreating, the regenerability of the liquids will need to be further explored; in addition experiments with more realistic fuel mixtures will need to be conducted. In the carbon dioxide removal experiments, a new quartz crystal microbalance (QCM) technique was developed and membrane studies were used to determine the solubility, diffusivity, and membrane permeability of carbon dioxide in various ionic liquids. The QCM techniques provide a rapid and inexpensive means of determining carbon dioxide solubility and diffusivity in thin ionic liquid layers. Many previously untested ionic liquids and ionic liquid mixtures were tested for carbon dioxide solubility. Of the liquids tested, most of the Henry’s Law values were determined to be in the range of 35 to 50 atm. These values are in good agreement with recently published values for similar room-temperature ionic liquids. An ionic liquid containing an octyl-methylimidazolium cation in which 13 of the hydrogens on the octyl group were substituted with fluorine exhibited significantly greater solubility of carbon dioxide, yielding a Henry’s Law value of 6 atm when combined with the Tf₂N (bis((trifluoromethyl)sulfonyl)amide) anion. CO₂ solubility in ionic liquids was also measured in the presence of water vapor. This test indicated that uptake of water at 40% relative humidity could not be said to have an impact on CO₂ solubility within the uncertainty of the QCM. The membrane studies involved measurement of the single-gas iv permeance of nitrogen and carbon dioxide in a supported ionic liquid membrane (SILM). From these experiments the permeance values of CO₂ were found to be on the order of 4.06×10^-9 mol/(atm.cm².s); for N2 the permeance was on the order of 3.22×10^-11 mol/(atm.cm².s). These permeance values were used to calculate a CO₂:N₂ selectivity of 127. Based on the single-gas permeance values measured for carbon dioxide and nitrogen, an economic comparison was made between amine scrubbing and ionic liquid membrane technology for use in flue gas treatment. The present preliminary values seem to indicate that the process could be competitive if the diffusivity for ionic liquid mixtures is much greater than that observed for the fluorinated liquid. Further studies with more accurate representations of flue gases and conditions are needed for a real feasibility determination; however, the preliminary values show some promise.