SO2 Retention by CaO-Based Sorbent Spent in CO2 Looping Cycles

Abstract

CaO-based looping cycles are promising processes for CO2 capture from both syngas and flue gas. The technology is based on cyclical carbonation of CaO and regeneration of CaCO3 in a dual fluidized-bed reactor to produce a pure CO2 stream suitable for sequestration. The main limitation of natural sorbents is the loss of carrying capacity with increasing number of reaction cycles, resulting in the need for extra sorbent, and subsequent spent sorbent waste. Use of spent sorbent from CO2 looping cycles for SO2 capture is investigated in this study. Three limestones were investigated: Kelly Rock (Canada), La Blanca (Spain), and Katowice (Poland, Upper Silesia). Carbonation/calcination cycles were performed in a tube furnace with both the original limestones and samples thermally pretreated for different times (i.e., sintered). The spent sorbent samples were sulfated in a thermogravimetric analyzer (TGA). The changes in the resulting sorbent pore structure were then investigated using mercury porosimetry. It has been shown that the sulfation rates of both thermally pretreated and spent sorbent samples are lower in comparison with those of the original samples. However, final conversions of both spent and pretreated sorbents after longer sulfation time were comparable or higher than those observed for the original sorbents under comparable conditions. Maximum sulfation levels strongly depend on sorbent porosity and pore surface area. The shrinkage of sorbent particles during calcination/carbonation cycling resulted in a loss of sorbent porosity on the order of ≤48%, which corresponds to maximum sulfation levels of ∼55% for spent Kelly Rock and Katowice. This is ∼10% higher than that seen with the original samples after 15 h of sulfation. By contrast, La Blanca limestone had more pronounced particle shrinkage during pretreatment and cycling, leading to porosities lower than 35%, which resulted in sulfation conversion of spent samples <30%, which is significantly lower than that for the original sample (45%). These results showed that spent sorbent samples from CO2 looping cycles can be used as sorbents for SO2 retention in cases where significant porosity loss does not occur during CO2 reaction cycles. The higher conversions of spent samples are explained by a shift in pore size distribution toward larger pores that reduce the reaction rate and pore plugging near the particle’s outer surface, with formation of either unreacted core or unreacted network patterns. In the case of spent Kelly Rock and Katowice samples, sorbent particles are practically uniformly sulfated, achieving final conversions that are determined by the total pore volume available for the bulky CaSO4 product.