OCR Text |
Show INTRODUCTION Nitrogen oxides (NOx) are formed in practical combustion systems by: 1) high temperature thermal fixation of atmospheric nitrogen l and 2) by oxidation of nitrogen chemically bound in the fuel. 2 Combustion modification techniques which reduce oxygen availability, particularly durin~ fuel pyrolysis steps, limit NOx formation by both paths. Staged combustion 3-5 and reburning6-8 are capable of reducing emissions from levels found in coal fired combustors by 70 percent if the initial concentration is relatively high. However, emission levels of less than-200 ppm are difficult to achieve solely by combustion modifications without introducing other problems such as reduced carbon burnout or flame impingement. Post combustion gas cleanup is necessary to achieve extremely low NOx levels. Lyon9 has shown that NOx can be reduced selectively to N2 by injecting ammonia under excess air conditions at temperatures from approximately 850 to 11000C. Muzio et al. 10 have reported results with urea and other amines. Under fuel rich-Conditions selective reduction can be achieved, but only at temperatures above 11000C.11,12 Cyanuric acid has also been suggested for use as a selective reducing agent under fuel lean conditions. 13 Perry and Siebers 14 have recently reported dramatic reductions with HNCO at temperatures as low as 6000C, although these high reductions at low temperatures appe~r to be due to catalytic activity involving the stainless steel reactor. I5 This paper summarizes the results of a study which was undertaken to investigate the possibility of positive synergism between the injection of various selective reducing agents, such as ammonia, and more conventional combustion modifications. Parametric screening studies were conducted in a 25 kW refractory lined tunnel furnace; theoretical support was provided by detailed kinetic modeling. The work focused on the importance of reaction temperature, local stoichiometric ratio, and the composition of the selective reducing agent. Cyanuric acid, urea, ammonium sulfate, and ammonia gas were considered as selective reducing agents. Fuel rich and fuel lean conditions were included in a temperature range of 600 to 13000C. EXPERIMENTAL SYSTEMS The combustion experiments were conducted in a 15.2 centimeter diameter by 2.4 meter long refractory lined tunnel furnace. All data reported here are for natural gas firing in a 15 kW premixed burner. The oxident was air and ammonia was added to the natural gas to provide the desired level of post-flame NO. The NO values, along with all other concentrations, are reported on a dry basis, corrected for dilution to 0.0 percent 02. The temperature was declining at a rate of 2250 C/s throughout the region of interest in this study. When the furnace was operated to simulate staged combustion, the gases from the first stage, normally fuel rich, were mixed with burnout air via a radial injector positioned on the furnace axis. The temperatures reported for the staging location are those of the furnace gas immediately after the staging air has been added and mixed. The change in 2 |