Effects of steam on the sulfation of limestone and NOx formation in an air- and oxy-fired pilot-scale circulating fluidized bed combustor

The existing fluidized bed combustion literature on sulfation shows that above 30% conversion, direct sulfation via reaction with CaCO3 is faster than indirect sulfation with CaO. However, while this is true for dry flue gases, it is not the case if steam (H2O(g)) is present at realistic levels for coal combustion, and it has been confirmed by experiments employing thermogravimetric analysis (TGA) and tube furnace (TF) testing that direct sulfation is in fact slower than indirect sulfation for nearly all levels of conversion if steam (H2O(g)) is present. In this work we have also examined the effects of H2O(g) on SO2 capture and NH3 oxidation to NOx over calcium-containing compounds under air-and oxy-fired conditions in a pilot-scale circulating fluidized bed combustor (CFBC) utilizing limestone addition. The results of the pilot-scale tests confirm suggestions from our previous work that sulfur capture from the air firing of low-moisture fuels benefits from steam-sulfation. For petroleum coke, the addition of 8% vol H2O(g) resulted in increased SO2 retention and Ca utilization, as well as decreased NOx emissions by up to 44%. The simultaneous reduction of SO2 and NOx was attributed to enhanced solid-state diffusion (sintering) by H2O(g). Under oxy-fuel-firing conditions, H2O(g) addition also resulted in decreased NOx emissions, but the pilot-scale tests showed poorer sulfur capture performance and calcium utilization as compared to air firing when H2O(g) was present, thereby reconfirming the TGA/TF results. It appears that most bench-scale work on sulfation to date has underestimated the true rate of reaction for sulfation in the presence of H2O(g). This conclusion explains at least in part why indirect sulfation is often faster than direct sulfation in pilot plant studies on oxy-fuel circulating fluidized bed combustion. Moreover, this work stresses the importance of including H2O(g) in bench-scale experiments that attempt to simulate real combustion environments. (C) 2011 Elsevier Ltd. All rights reserved.