Development of a Carbon Dioxide Continuous Scrubber (CDOCS) System for Alkaline Fuel Cells
Thesis DisciplineMechanical Engineering
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
Alkaline fuel cells (AFC's) using renewable fuels are a developing technology capable of meeting market niches in standby, standalone and distributed power generation. AFC's generate electricity, heat and water using hydrogen and oxygen as fuels. While AFC's have been known and the principles demonstrated for over sixty years, their use has been restricted primarily to space applications. Recent technological developments have seen the cost of AFC stacks fall considerably; this together with several other advantages over competing fuel cell technology, has rekindled interest in commercial systems. The main deterrent to wide spread commercialisation of AFC systems is susceptibility to carbon dioxide (CO2) in atmospheric air used as the oxygen supply. AFC's require a low cost, low energy, continuous scrubbing device to reduce CO2 in air from approximately 380 parts per million (ppm) atmospheric concentration to below 50 ppm. Current technology to overcome this problem, a solid expendable absorbent called soda lime, is not viable for commercial systems. The project scope included concept generation of a device to remove CO2 from air, the development of a CO2 measurement technique, investigation of chemistry and flow phenomena to determine design relations, and product design and embodiment. The scrubber system conceived specifically for AFC systems uses the temperature swing chemistry of a liquid chemical absorbent, monoethanolamine, and a packed bubble column apparatus to provide intimate gas-liquid interaction. Prototype development proved the Carbon Dioxide Continuous Scrubber (CDOCS) concept and a Patent Cooperation Treaty (PCT) patent was granted, followed by a full American patent. A gas chromatographic measurement technique was developed to measure low ppm concentration CO2 in air, enabling regular monitoring of scrubbed gas. Carbon dioxide was separated from a small sample of scrubbed air by chromatographic columns, and the gases analysed with a thermal conductivity detector. The GC system was capable of measuring to 10 ppm with good resolution and accuracy. Experimental studies were carried out to characterise the flow dynamics and absorption phenomena in the packed bubble column absorber. The relationship between absorption performance and gas-liquid contact time, an important operating parameter for use with AFC's, was theoretically determined and later confirmed by experiment. The regeneration process was studied and the optimal regenerator design determined to be second, smaller packed bubble column. Experiments were conducted to establish design relations for regeneration temperature, flush gas flow rate and the effect of multiple regeneration cycles. A prototype CDOCS system was built to enable experimental characterisation of scrubbing performance as a function of primary design and operating parameters including liquid depth, regenerator operating temperature and solution composition. This resulted in a good understanding of the system, and an optimised experimental run was performed for cost and performance comparison to existing scrubbing technology. The CDOCS was capable of reducing CO2 in air from 380 to 80 ppm for thirty days, providing low cost, low maintenance scrubbing compared to soda lime. The capital cost of the CDOCS is considerably more than for soda lime scrubbers, and the penalty for extended operation is parasitic power consumption by the CDOCS system totalling less than 7% of fuel cell output. It is suggested that a combination of the two technologies be used initially to provide effective, low cost scrubbing for AFC and CDOCS co-development. Future work on the CDOCS project should include reduction of chemical vapour carry over to the fuel cell, followed by integration with an AFC system. This would allow further development, refinement and design for production to reduce capital cost.