(110c) Laboratory, Pilot, and Plant Scale Studies in CO2 Scrubbing

Kawatra, S. K. - Presenter, Michigan Technological University
Prior experimentation with CO2 scrubbing technologies have required caustic alkali hydroxides or toxic amines. To avoid associated corrosion issues and toxicity hazards, the effectiveness of sodium carbonate for capturing CO2 was tested at laboratory scale. Sodium carbonate is a part of the natural mechanisms by which the ocean captures CO2 from the atmosphere. We detail the laboratory scale experiments used to determine the potential efficacy of sodium carbonate capture solutions. After promising initial results, we scaled this up to a 4” diameter by 4ft tall packed column, with capturing solution flowing downwards and a simulated flue gas bubbling upwards. We have found that sodium carbonate solution in distilled water with frother modifiers is very effective at capturing CO2 from a simulated flue gas. At pilot scale using these conditions we are able to capture enough CO2 that we cannot effectively measure the remaining traces, corresponding to over 98% CO2 capture. Following these results and to understand the difficulties involved in capturing CO2 from a real flue gas, we partnered with the Michigan Tech steam plant to install a similar scrubber in their facility. We encountered several distinct problems during this scale up including: the limited availability of distilled water, the presence of CO and other impurities in the flue gas, practical issues involving the appropriate pressurization and filtration of the flue gas, and practical issues involving the influence of pressure and momentum on process kinetics inside the column. We also gained significant insight on the effects of flue gas variability on the capture process. From these issues, we have found that several solution impurities can have a significant negative impact on the performance of the sodium carbonate scrubbing solution – limited time was spent on isolating the specific impurities, but both tap water and softened water were found to have significantly worse capture rates in baseline testing. We have found that the frother modifiers may be deactivated by some of these impurities, requiring additional water preparation or careful frother choice to overcome. We have found that overly high gas flowrates result in gas channels forming inside the column, significantly impeding the CO2 transfer into the aqueous phase. Despite these challenges, the core process has remained effective even at pilot scale. In this document we report the capture efficiencies of this CO2 scrubbing process at numerous conditions at each scale of the experimentation, including (where applicable) using distilled water, tap water, softened water, varying concentrations of sodium carbonate, varying concentrations of CO2 in simulated and real flue gases, and in the presence and absence of various frother chemistries.