| OCR Text |
Show calculations. The initial modelling was conducted using a thermodynamic equilibrium mixture of either CH4 and C2H2 products at the expected exit temperature of the plasma torch (assumed to be about 3000 C) premixed with flue gases with mixture temperatures ranging from 350 F to 1000 F. The thermodynamic equilibrium calculations indicated that a mixture of CH, CH2, C2H, C, H and H2 would be expected. The kinetic modelling results indicate that two major reaction paths would be expected. The first reaction path is the reaction; CH + NO = HCN + 0 (1) This reaction provides the HCN molecules necessary for the desired next reaction step in the Pyronox process namely the reaction of HCN with OH to generate NH2 (Reaction 2) which is then available to react again with NO to form N2 (Reaction 3). The kinetic model does indicate, however, that the formation step of HCN is faster than the reaction of HCN with OH to generate NH2 and that unreacted HCN will remain in the flue gases. The kinetic model also indicated, particularly at the lower range of flue gas temperatures (300 F to 400 F) another significant reaction path exists namely; CH + O2 = HCO + 0 (4) HCO + O2 = H02 + CO (5) H02 + NO = N02 + OH (6) This reaction sequence has the potential to generate unwanted CO and N02. These initial kinetic modelling calculations suggest that HeN, CO and N02 emissions could all be possible byproducts in the PYRONOx process when using hydrocarbon gases as the plasma gas. During experimental phases of the project these species are being carefully monitored. The kinetic model has also been used to screen other gases as possible radical sources for use in the plasma NOx reduction process. Several possible candidates were identified that should give high conversion levels of NO to N~ without unwanted byproducts. EXPERIMENTAL RESULTS MIXING TESTS Cold flow mIXIng tests, using a NO tracer gas injected into the plasma gas, were conducted to quantify the mixing rate between the plasma and flue gases. The mixing tests were conducted in the firetube boiler used for the actual plasma injection tests. Several configurations were assessed in the cold flow tests, the co-rotating tangential injection case with the 50 degrees swirl vanes in the flue gas, radial injection of the plasma gas into the swirling flow, axial injection of the plasma gases into the center of the swirling flow, and axial injection of the plasma gas using an orifice one inch downstream of the injection point. When the orifice plate 10 |
