| OCR Text |
Show Figure 5 shows an alternate arrangement where the oxidant and fuel injection points are located close to each other on the same furnace wall. In this case it is particularly important to inject both oxidant and fuel streams at high velocities away from each other so that the oxidant mixing zone and the fuel reaction zone do not overlap. Since the dominant jet with a high momentum flux tends to overwhelm and entrain the weaker jet with a substantially lower momentum flux, a proper ratio of the fuel and oxidant momentum fluxes and sufficient spacing between any two jets must be maintained. EXPERIMENTAL RESULTS A series of tests were conducted in a refractory-lined cylindrical test furnace. The internal dimensions of the furnace are 3 feet in diameter and 7 feet 8 inches long. A primary burner port is located in the center of the end wall. A flue port is located on the axis of the furnace on the opposite end wall. Several access holes are placed on the flue end wall and the cylindrical side wall for injection of different gases and for insertion of heat sinks. Tests were conducted at a constant firing rate of about 700 SCFH of natural gas and at a constant furnace wall temperature of 23000F, measured at the mid point of the cylindrical wall by a thermocouple. Two to four water-cooled heat sink pipes were inserted through the access holes to maintain the furnace temperature at the constant level. About 150 to 5000 SCFH of nitrogen was introduced into the furnace through the three middle view ports to simulate an actual furnace with various nitrogen concentrations. Oxygen concentration in the flue gas was monitored continuously by an in-situ sensor located at the flue port and kept at 2 to 2.5% on a wet basis by adjusting the flow rate of oxygen or natural gas. NOz was measured by a chemiluminescent type analyzer which was properly calibrated for the effects of the background gases (N2 and CO2). The result of the NOz measurements are plotted in Figure 6. As the baseline of NOz emissions for comparison a low NOz oxygen aspirator burner (Linde Type A Burner) was used (Ref. 2). This burner was optimized for NOz reduction with small fuel and oxygen nozzles. NOz emissions ranged from 0.0027 Ib/lO' Btu at a nitrogen concentration of 6.8% to 0.0746 Ib/ 1(1 Btu at 64.9% N2 and denoted as line B in the figure. Since NOz emission was shown to increase approximately linearly with N2 concentration for different burners tested in the initial series of the tests, only limited number of data with respect to nitrogen concentration were taken at each configuration in the subsequent tests. The results from the various DOC arrangements are denoted in figure 6 as C, D, E, F, G and H. In example C, D and E fuel was injected from the burner port and oxygen was injected from two oxygen lances through the lance holes located 180 degrees apart in the flue end wall. The oxygen jet velocities used for these tests were about 760 to 850 ft/sec. Different fuel and oxygen nozzles were used for each case. NOz emission was reduced by about 50% compared to the base case B. In examples F and G both fuel and oxygen were 4 |
