| OCR Text |
Show on temperature measurements (presented below) showed corresponding increases of 20% and 10%. Why the NOx emission indices are lower with air dilution than with fuel dilution can be explained by differences in both residence times and flame temperatures. The following arguments concerning residence time are speculative, since direct measurements of velocities within the flames were not perfonned; however, temperature effects are based on measured temperature profiles. Results of the numerical studies clearly demonstrate the importance of residence times in explaining any differential effectiveness between air- or fuel-dilution in reducing NOx emissions. With regard to the experiments, it is likely that the physics of the fuel- and air-dilution experiments resemble that of the fixed initial-velocity simulations, rather than the fixed mass flux. This may seem paradoxical, since the experiments were conducted with a fixed fuel mass flux, conditions under which the numerical simulations showed greater NOx-reduction effectiveness for fuel dilution. This can be explained by the fact that the experimental jet flames are dominated by buoyancy, hence, the velocity field of the flame is only weakly related to the initial velocity of the jet. Velocity measurements in similar flames by Santoro et al. [15] show that the jet rapidly accelerates as the buoyant force increases as more and more hot products are fonned. Near the flame tip, the velocities are many times greater than the exit velocity; for example, at 4.5 cm downstream from the nozzle, a position near the tip of the flame, Santoro et al. [15] measured a velocity of approximately 165 cm/s, a value some 33 times larger than the initial jet velocity. Based on these ideas, we would anticipate that the longer the flame, the larger the mean velocity, since buoyancy causes the velocity to increase monotonically from the jet exit to well beyond the flame tip. As a consequence, mean flame-zone residence times should be somewhat smaller with the longer flame, with NOx emissions tending to be lower, all other factors equal. Figure 9 shows that, indeed, the flames with air dilution are longer than those with fuel dilution. Typically, NOx production rates increase very rapidly with temperature. Measurements were perfonned to see what differences in temperature exist between the air- and fuel-diluted flames. One might expect that air-diluted flames are somewhat cooler than the fuel-diluted flames based on the observation that the luminosity of the air-diluted flames is greater than that of the fuel-diluted flames at equal diluent fractions. Here we assume that a greater luminosity implies greater heat losses and, hence, lower flame temperatures. Figure 10 shows the fraction of the total flame length that is blue, i.e., ostensibly soot-free. In this figure, we see that the soot- 12 |
