OCR Text |
Show effect of turbulence on prompt-NO is not significant partly because the prompt-NO production route involves chemical reactions which are much faster than the extended Zeldovich mechanism. However it is interesting to note that by cons i deri ng the effect of turbul ence on the prompt-NO production rate using joint pdf of temperature and reaction progress variable (in this case fuel), the rate of production reaches its maximum value around the fuel profiles. These data can be of significant importance for design engineers by highlighting the areas where thermal or prompt-NO is produced within the combustion system, and ways of reducing NO emission by changing its geometry and other parameters. As demonstrated in Fig. 10(a-b), the most probable values for the joint pdf are concentrated near the flame front. The joint pdf between the fuel fragmentat i on and temperature for prompt -NO is shown in Fig. 10(c-d). It is apparent that the prompt NO exists mainly on the fuel-rich side of the reaction zone at temperatures above 1150 K. All these results indicate that the affect of turbulence on chemistry cannot be neglected. The rate of NO product ion is affected by turbul ent combust ion characteristics. The present theoretical studies indicated that the therma 1 and prompt-NO concentrat i on "depends strongl y on mi xture fract ion and temperature. It is believed therefore that a joint pdf plays an important role for accurate predictions of NO in stationary combustion systems, and the analytical model must account for the influence of temperature and mixture fraction fluctuations. Finally, Fig. 11(a-c) shows the effect of oxygen-enrichment on NO concentration. In this study the combustion air is oxygen-enriched by 8~ and air temperature is reduced to 293 K. In order to keep the oxygen concentration on the flow gas to 3.1%, the primary air velocity is also reduced. Predicted results for oxygen-enriched combustion (Fig. S(a-c)) i ndi cate that fl arne becomes shorter and the maximum fl arne temperature increases by 4%. The predicted NOx concentration is increased only by 3% indicating the dependency of extended Zeldovich mechanism on the temperature and oxygen concentration. However it is bel ieved that at higher oxygen-enrichment with higher flame temperature the reduction in 0 and OH radical overshoot will compensate the increase in flame temperature and therefore, with careful burner design, the NO emission during oxygen enriched combustion can be controlled. x Comparison of the present model calculations and experimental measurements points out some of the shortcomings of the experiments, such as lack of data in the near burner zone or use of a gas sampling probe, where continuous probe sampling leads to poor spatial resolution. Obvi ous 1 y, more experi mental data wi th in the combust i on system would be useful for validation purposes (21). However the present model-data comparisons, shown in Table 1, support the validity of the model calculation. The uniform NO concentration inside the furnace is in good agreement with available data, to within 47 ppm. This is well within the combined uncertainties of the model and experiments. Uncertainties in the kinetic model are more easily quantifiable and coupling the CH4/O detailed chemical mechanism with the extended Zeldovich mechanism wilT give even better agreement but there will be a sharp increase in computer processing time (up to 3 order of magnitude). Uncertainties in the rate 10 |