OCR Text |
Show promote the effect of N-agents under fuel-rich conditions. A similar effect of reducing NO levels and broadening the temperature window for effective N-agent injection, was demonstrated earlier under fuel-lean conditions [8]. One can suggest several hypothetical sodium reactions with flue gas components to promote NO reduction: l.Sodium atoms and oxides react with H20, for instance, Na+H20-NaOH+H and NaO+H20-NaOH+OH followed by formation ofNH2 radicals from ammonia, OH+NH3-NH2+H20, and NO-NH2 interaction: NH2+NO-N2+H20. 2. Sodium atoms fonned during N~C03 decomposition react with ammonia [12] followed by dissociation of NaNH2: Na+NH3-NaNH2+H and NaNH2+M-Na+NH2+M. 3. N~C03 dissociates at high temperatures to form oxides, for example, to NaO+Na+C02 or Na20+C02· The oxides then react with ammonia via equations: NaO+NH3-NaOH+NH2 and N~O+NH3-NaOH+NaNH2 followed by the sodium amide decomposition. Although it is unknown which' of the sodium reactions are most important, the effect of all reactions is fonnation of active radicals (NH., or OH) at a faster rate than that of ammonia decomposition without sodium additives. It was assumed for modeling that an interaction of Napromoters with water is responsible for the fonnation of the additional OH radicals. The following empirical reaction was added to the mechanism: [(Na)HOH]+M=>(Na)+OH+H+M. This equation simulates fast interaction of sodium compounds with water molecules with fonnation of a complex, [(Na)HOH], which dissociates into OH and H. The initial concentration of [(Na)HOH] was assumed to be 100 ppm in modeling. The rate constant for [(N a)HOH] decomposition was selected in the form Aexp( -20,000ff) cm3/mol s where the predexponential factor A was varied in the range from 109 to 1012 • These values of the rate constant provided decomposition of the complex with characteristic tirres between approximately 0.1 and 100 s. Injection of the promoter caused slightly higher H and OH concentrations in the reaction media than those in the absence of the promoter. Figure 15 compares the results of modeling NO and NH3 concentrations at t = 1.5 s (the same conditions shown by curve 4 in Figure 14 at T=1255 K) at different values of the predexponential factor A. The promoter [(Na)HOH] was added into the rebuming zone along with ammonia at a 0.1 s delay tirre. The level of TFN without promoter (257 ppm) is shown in Figure 15 for comparison. In a wide range of the promoter decomposition rate, NO and TFN concentrations are substantially lower than their values without promoter. In the presence of the promoter there is a minimum of the final NO concentration at logA= 1 0.$. Thus, there exists an optimum rate constant for fonnation of 8 |