Popis: |
mass, momentum, energy, and species, are obtained using a computer program based Sheu and Chen ( 1996). A four-step reduced mechanism is extended to include In the present study, the methane oxidation is based reaction kinetics and reactive species for prediction of on the forty-step starting mechanism of Peters (1993), NOx emission from atmospheric methane-air diffusion cf. reactions l-40 in Table 1. This starting mechanism flame. For methane oxidation, a fifty-step starting has been used, for example, by Mauss and Peters (1993) mechanism is used and seventeen species are for methane premixed flame and by Chelliah et al. considered. Among the seventeen species, seven species (1993) for counterflow diffusion flame. Ten additional are treated as independent reactive species. The reactions are added to predict the radical species concentrations of these species are obtained from the necessary for prediction of NOx (thermal and prompt solutions of the transport equations of each respective NO, N20 and N02) in the flame, cf. reactions 41-50 species. Steady-state assumption is invoked for the Table 1. Selection of these ten reactions is based on the remaining ten species and the concentrations are sensitivity analysis of the reactions affecting the NOx determined from algebraic equations. For NOx emission determined by Miller and Bowman (1989), and prediction, the nitrogen chemistry is assumed to be Drake and Blint (1991). independent of methane oxidation. The NOx reduced The nitrogen chemistry model of Glarborg et al. mechanism is derived from a thirty-step starting (1992) includes twenty-seven reactions (e.g., reactions mechanism for eight species with three species being 51-77 in Table 1) and seven species (i.e., NO, HCN, N, considered as independent reactive species. The CN, NCO, NH and N20) to account for the formation of solutions for the reactive species are obtained from the NO, N20. The NO2 formation mechanism is not solutions of the transport equations. The remaining five considered. To account for NO2 formation, three major species are obtained from algebraic equations invoking reactions for NO2 chemistry of Miller and Bowan the steady-state assumption. The prediction is compared (1989) are added. The rate parameters of these three with experimental data in the literature of a ducted, reactions are listed as reactions 78-80 in Table 1. In laminar jet diffusion flame fueled by methane. summary, the present chemistry model adopts a thit-tyReasonable agreement is obtained. The effects of radiant step starting mechanism for NOx formation, which energy transport with participating species of H20, CO2 includes (a) thermal NO formation (reactions 55-57, and soot particles are studied. The radiative transfer is Table I); (b) the initiating steps of prompt NO accounted for by implementing the Discrete-Ordinates formation (reactions 69-70); (c) the major steps of HCN method in the calculation. The soot volume fraction is oxidation (reactions 51-53, 60-67, and 76-77); (d) predicted by a calibrated soot-oxidation model. The destruction of NO through CHi + NO (reactions 71-75) calculation suggests that OH and H20 are the dominant and through FN + NO (reactions 54, 57, 68) where FN oxidant species in the fuel-rich and flame-tip regions, denotes any reactive nitrogen species; (e) the major steps and 02 is the dominant species in the fuel-lean region. of N20 formation (reactions 54, 58, 59, 68). To solve the radiative transfer equation (RTE), soot Introduction volume fraction and number density are needed. The semi-empirical formula of Syed et al. (I 990) is used. Numerical simulation has played an important role in The model considers the surface growth and coagulation the study of NOx formation in flames. It provides us on the formation rate of particle number density, and the insight into the NOx formation pathways; for example, effects of surface growth and nucleation on the on the relative importance of thermal NOx,, prompt formation rate of soot mass fraction. The soot oxidation NOx, and N20 formation in NOx emission from model of Bradely et al. (1984) is used, and the collision combustion flames, as well as into the interaction efficiency for OH radical is recalibrated based on the between the chemical reaction and fluid dynamics. The experimental results of Roth et al. (1990), as discussed present study considers an axisymmetric methane-air herein. Kaplan et al. (1994a) assumes that the radiative diffusion flame. Solutions for the transport equations of transport is independent of the wavelength, and that the ’ Currently with CFD Research Corporation. Huntsville. AL : Professor and Director. Scmor Member AIAA; Comesponding Author Copyright 8 2000 by J.C. Sheu and L.-D. Chm. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission (c)2000 American Institute of Aeronautics & Astronautics or published with permission of author(s) and/or author(s)’ sponsoring organization. total absorption coefficient is a linear combination of the individual absorption coefftcients of C02, H20 and soot particles. This approach is adopted in present study. The absorption coefficient of soot particles is calculated assuming that the Rayleigh limit of small spherical particle -is applicable. Thus, the spectral absorption coefficient can be determined following Tien and Lee (1982), with the real and imaginary parts of the complex refractive index of soot particles determined from experimental results of Habib and Vervisch (1988). The objectives of the present study are to (a) evaluate the use of reduced mechanism for prediction of the NOx formation in methane fueled diffusion flames, (b) evaluate soot oxidation models for calculation of the radiative transport in the flame, and (c) evaluate the effects of radiative transport on the flame structure. The first term of the right hand side of the transport equations of soot number density and soot mass fraction represents the thertnophoretic flux. The thermophoretic axial velocity is assumed to be negligible, and the thermophoretic radial velocity vr is approximated by (Kennedy et al., 1990) |