(29d) Flue Gas Co2 Capture by Means of a Biomimetic Facilitated Transport Membrane

Trachtenberg, M. C., Carbozyme, Inc.
Smith, D. A., Carbozyme, Inc.
Cowan, R. M., Carbozyme, Inc.
Wang, X., University of Texas at Austin

The importance of carbon dioxide (CO2) capture to control greenhouse gases is plainly evident. Flue gas - a high concentration, high volume, point source ? is the most likely capture target. The vast majority of flue gas (~79%) derives from the utility industry. For utility applications cost of energy (COE) is a critical parameter as it results in de-rating the plant. Additional load capacity is then required to serve current needs and retain contractual integrity.

Several CO2 capture methods are available or under development. However, with very few exceptions they are either too expensive or have not been scaled to the size required, or both. The DOE is supporting a large number of different technological approaches, at various stages along the research, development, engineering, pre-pilot and pilot stages, towards demonstration capabilities as a prelude to full-scale implementation. Carbozyme's enzyme catalyzed, hollow fiber, contained liquid membrane permeators is one such technology that is poised to enter pre-pilot demonstration phase.

The Carbozyme technology uses the enzyme carbonic anhydrase to catalyze (facilitate) the conversion of CO2 to bicarbonate at the feed side of the membrane and then reverses the process on the permeate side allowing the CO2 to move down its concentration gradient. At the same time the salt filled aqueous solution restricts the movement of nitrogen, oxygen and other gases as these must undergo physical absorption, a process that is slow and that suffers from limited volumetric capacity. The consequence is that the CO2 is greatly enriched at the permeate side. For example, a process engineering calculation showed that a 13.8% CO2 feed gas concentration (EPRI case 7C) exits as a 53.5% wet stream from the permeator. After drying the CO2 concentration is calculated to be 98.3%. Measured laboratory data, using simulated flue gas mixtures, showed that a 15% feed stream yielded a 95% dry stream. These data were further supported by demonstration of comparable CO2 capture behavior from feed streams derived by combusting methane or propane. The feed gas CO2 concentrations tested ranged from air to 40%. Thus, any fuel currently combusted can be handled by this design. Long-term studies (1,000h), with deliberate, upset showed the design to be stable and robust. Additional data show that the enzyme is stable in the face of many of the components found in post-scrubbing flue gas. We are currently engaged in determining acceptance criteria to determine if any additional scrubbing may be needed. We have also developed scale-up processes coupled with process engineering studies to project size and cost profiles. We project that the cost of energy for such a system will be less than 13% and the cost per tone of CO2 captured and compressed will be less than $15.