OCR Text |
Show 3 Hydrogen is used as the primary fuel for two reasons: 1) the system was designed for this fuel (Thornton et aI., 1987), and 2) hydrocarbon fuels either produce excessive soot or burn unstably in the lower chamber. The lack of carbon in the primary combustion products is not thought to be a serious drawback of the present experiments. The temperatures and flue gas oxygen contents of interest are simulated. The products of the primary combustion pass through the Inconel nozzle block, thereby producing the sonic jet. Nitrous oxide, nitrogen coolant gas, and the afterburning fuel are injected into this jet by the side-stream depicted in Figure 1. Only a schematic representation is shown in Figure 1; the actual nozzle with side stream has a design which pennits the side stream to mix efficiently with the main jet. This was fully explored in the work of Thornton et aI. (1987). The arrangement produces a stirred reactor inlet jet that is nearly premixed. Within the stirred reactor, the N20, afterburning fuel, and primary combustion products react in an essentially uniform, well mixed field. The stirred reactor has a 16cc volume, operates at essentially atmospheric pressure, has four drain holes, and two sample probe ports, including one for the thennocouple and the one for the gas sample probe. During check out experiments the stirred reactor was profiled. The temperature was found to be uniform within ±I% (relative), and the N20 concentration was found to be unifonn within ±IO% (relative). For the production experiments, the probes were placed to sample conditions which agreed with the space-averaged conditions for the reactor. The thennocouple probe is PtJRh material, coated with a ceramic material to prevent temperature increases due to surface reactions. The readings are corrected by calculation for radiation heat loss. This is a small correction, typically 10 to 50K over the range of the experiments. Gas samples are continuously withdrawn from the reactor using a quartz probe. These gases are quenched aerodynamically and by probe wall water cooling, are dried, and routed to gas analyzers for NO, NOx, 02, CO, and C02. The N20 is measured by on-line electron capture detection (ECD) gas chromatography. Separation is by a Hayesep DB porous polymer column operated at 100 degrees C using a 5% C14I95% Ar carrier gas. Daily calibrations are performed in order to monitor small changes in the ECD sensitivity. The lower detection limit of this method is about 0.1 ppmv. No formation (or loss) ofN20 in the gas samples is expected because of: 1) the use of on-line sampling to reduce time for reaction, 2) the absence of S02, and 3) the pre drying of the sample. The experimental conditions for the afterbuming chamber for the production experiments are as follows: I. The temperature of this chamber before the afterburning fuel is added is 1080, 1200-1250, and 1330-1370K. This is essentially the inlet jet total temperature. |