| OCR Text |
Show Figure 3 shows typical results for NO reduction with cold and hot NH3 injection over a range of combustor temperatures at 25% excess air conditions similar to those in the high reduction plasma experiments. As shown, hot NH3 injection was moderately more effective than cold injection at all flue gas temperatures investigated, and NO reduction increased with combustion temperature. The difference between the two curves in Figure 3 is an indication of the thennal benefit of heating the ammonia before injection at these conditions. To capture the actual benefit of generating ammonia radicals for injection using plasma technology, data from the previous plasma experiments were compared to hot injection data obtained under similar combustion and ammonia flow conditions. The plasma experiments by Zhou were not used to explore the effect of flue gas temperature on NOx reduction, so the data relevant to such a comparison with hot ammonia injection are sparse. Multiple data points for NOx reduction at the same excess air condition but different combustion temperatures were found for 0% excess air only. Figure 4 shows results for NO reduction at 0% excess air plotted against combustion flue gas temperature. The combustion temperature was not reported in the plasma experiments, and so it is assumed that the temperatures found in the hot injection experiments at the same fuel and air conditions apply. This is consistent with reports by Zhou that the plasma injection process did not affect the temperature below the injection point. The difference between the two curves in Figure 4 indicates that there is a benefit to externally generating ammonia radicals that is due to the presence of the radicals themselves and is beyond the thermal benefit of hot injection. While the lack of plasma experiment data precluded making further comparisons between plasma and hot ammonia injection over . a range of flue gas temperatures holding other conditions constant, direct point comparisons between the two sets of data give insight into the NOx reduction effectiveness of each under identical conditions. In this way, the relative contributions of thennal and plasma effects can be clearly seen in Figure 5. The plasma experiment data provided four points in which all conditions were held constant except for the air flow rate. These conditions were duplicated using both hot and cold injection, leading to direct point comparisons among hot NH3, cold NH3, and radical injection· at four excess air conditions. Since the combustion temperature increased with decreasing excess air (the same amount of fuel was used in all four cases), the difference between the NO reduction values in the columns in Figure 5 include two separate effects. However, between the 0% and 25% excess air conditions, which is a typical range for commercial boilers, the flue gas temperature increase was no more than 16 K, indicating that the excess air condition itself was the dominant variable. The average combustion temperatures at 75%, 50%, 25%, and 00/0 excess air were 922 K, 938 K, 960 K, and 976 K, 5 |
