OCR Text |
Show and velocity gradients associated with this higher-velocity fixed-fuel-mass-flux flame are the cause of its increased fuel consumption rate. Furthermore, the net NO formation rate is also lower; thus, both increased fuel consumption and reduced NO formation factors combine to yield a lower NO emission index for the fixed-fuel-mass-flux flame. Specifically, the emission index for this flame (EIND = 0.086 g!kg) is 3.4 times lower than the fixed-velocity fuel-dilution flame and 2.9 times lower than the corresponding air-dilution flame. It is interesting to point out that, although this reduced net NO formation rate is dominated by residence time considerations (cf. Table II), the ratio of NO production-to-destruction for the Case 4 flame, compared to the other three cases, is quite large, i.e., 3.50 versus 1.92 and 1.72. This suggests that the relative importance of the various NO-formation pathways changes significantly between the 128 cm/s and 50 cm/s conditions. Further research is required to elucidate the details of such effects. Parametric Studies Calculations were performed for both air- and fuel-dilution with N2 over a range of diluent fractions. Two levels of reactants temperatures were considered, 300 K and 500 K, and the fuel dilution was accomplished for both fixed-velocity conditions (uo = 50 cm/s) and for fixed fuel mass flux (0.326 kg/m2-s at 300K and 0.189 kg/m2-s at 500 K). The previously discussed results were a subset of the calculations perfonned for 300 K reactants. Nitric oxide emission indices are shown as functions of diluent fraction in Fig. 7. In this figure, we observe that fuel dilution is more effective than air dilution when the fuel mass flux is fixed. Fixed fuel mass flux is the condition that is most applicable to practice; however, one must be cautious in extrapolating the laminar flame results to real boilers both because the geometries are different and, more importantly, the practical flow is governed by turbulent mixing. Nevertheless, we gain the insight that local residence times for NO formation in a flame are affected by how the diluent is added to the system. Similar results obtain for the 500 K reactants. EXPERIMENT AL STUDIES As a complement to the numerical simulations, experiments were conducted using a laminar jet flame. Although the jet flame geometry is significantly different from the counterflow studied numerically, they do share common features. For example, at low-velocity conditions for the counterflow, the structure of both flames are dominated by diffusional 9 |