| OCR Text |
Show percent was observed with cyanuric acid injection. Urea exhibited a maximum NOx to N20 conversion of 25 percent at this lower Initial NOx level. The effect of the chemical injection rate on the NOx reduction and N20 emissions are shown In Figures 9 and 10 respectively. As expected, Figure 9 shows increased NOx removals with increasing (:hemical addition rate (N/NO ratio) for all three chemicals and at all temperatures lested. As seen in Figure 10, the amount of chemical Injected, N/NO ratio, has little 11I1pact on the conversion of NOx to N20. At the lower temperature conditions, less than 927°C, there appears to be some increase in N20 production and conversion of NO), to N20 as the N/NO ratio increases. The decrease in NOx to N20 conversion witll increasing N/NO ratio for cyanuric acid injection at 827°C is due to the increased in NOx reduction as N/NO)/ rate increases rather than a decrease in N20 emission levels. For the urea injection data shown previously, the urea was Injected Into the combustor as a pulverized solid. In full scale application of urea injection for NOx reduction, the urea has been typically injected as an aqueous solution (Mansour, et a!, 1987). Tests were conducted during this study to investigate whether injecting a solid or aqueous solution impacted the amount of N20 produced. For the case of injecting an aqueous solution, one might speculate that hydrolysis could occur during evaporization of the drops, perhaps leading to different decomposition products being released to the gas phase to react with NOx ' The performance of pulverized urea and aqueous urea are compared in Figure 1 1. As can be seen in Figure 11, the overall level of NOx removal and N20 emission are comparable for the solid and aqueous injection of urea. This would suggest that there are no large differences In the decomposition products as the solid decomposes and as the drop evaporates and decomposes. However, the liquid urea solution exhibits peak removals of higher injection temperatures than the solid. This is likely due to a thermal effect rather than a chemical effect. In these pilot scale experiments, the Injection temperature Is characterized by a suction pyrometry measurement at the exit of the venturi section (nominally 35 msec) after injection. Since with liquid injection the drops first have to vaporize, the reactive species are released at a lower point In the test section than with solid injection. Since the test section is non-isothermal with a temperature drop of nominally 250°C/sec the liquid solution would have to be Injected at a higher 17 |
