Reduced chemical kinetics mechanism

Contribution: Faculty of Mechanical Engineering, Banjaluka (MEF-BL)

Summary

One of the main interest of the project partner MEF-BL were reduction of the chemical reaction mechanisms, which could be further efficiently implemented into the commercial CFD code. Several available chemical reaction mechanisms, available from the open literature, are tested and their reduction was discussed.

Introduction

Design of a flexible burner must include a proper understanding of chemical reactions that take place, how they affect the flame behaviour and how the flame interacts with the flow field. The commercially available chemical reactions and flow codes are usually of general character, thus not of big utility for modelling practical combustion systems. Therefore, reduced chemical kinetics mechanisms need to be developed to enable reliable and fast numerical analysis of practical burner.

1.1. Detailed Chemistry

Full, detailed kinetics at the moment is impractical. Efficient CFD calculations have to be based on reduced chemistry, with small number of species and reactions. There are different reducing techniques, from fully automatic algorithms to purpose developed heuristic techniques. The reduction technique is based on analysis of species concentrations, rates of production, and sensitivity coefficients. Quasi-steady-state species are selected and independent elementary reaction steps are chosen in order to eliminate the quasi-steady-state species.

There are reduced mechanisms for methane and propane which give excellent results compared to full mechanisms. When NOx formation analysis is needed the reduced mechanism must include nitrogen chemistry.

The following proven archived detailed mechanisms were applied to modelling of emissions and flame structure:

  1. GRI-Mech 1.2
  2. GRI-Mech 2.11

    Version 2.11 expands GRI-Mech 1.2 by including nitrogen chemistry relevant to natural gas chemistry and reburning. It contains 277 elementary chemical reactions of 49 species.

  3. GRI-Mech 3.0

    GRI-Mech 3.0 is an optimized mechanism designed to model natural gas combustion, including NO formation and reburn chemistry. The new mechanism contains 325 reactions and 53 species (including argon). As available literature studies show, chemical mechanism described in GRI-Mech 3.0 gives results that can be compared with results obtained in laboratories during real measurement process.

  4. Chemical Kinetic Mechanisms for Combustion Applications, developed by the Combustion Division of the Center for Energy Research at the University of California-San Diego
  5. Reaction mechanism is from early work at Sandia National Laboratories

Mechanisms GRI 2.1 and 3.0 were the main basis for development of detailed, reduced skeletal chemical reaction mechanism.

1.2. Reduced Chemistry

MEF-BL chose three reduced chemistry mechanisms, developed by J-Y Chen [1], constructed for methane/air combustion, as follows:

  • 12-step reduced chemistry developed from GRI Mech 2.11

  • 13-step reduced chemistry developed from GRI Mech 3.0

  • 15 step reduced chemistry developed from GRI Mech 3.0

All the reduced mechanisms include NOx chemistry beside the fact that they are simplified and reduced from more then few hundreds reaction to only 12-15. Twelve-step reduced chemical kinetics mechanism includes the following elements: O, H, C and N and their appropriate species: H2, H, O2, OH, H2O, CH3, CH4, CO, CO2, CH2O, C2H2, C2H4, C2H6, NO, HCN, N2. Thirteen-step reduced chemical kinetics mechanism includes following elements: O, H, C and N and their appropriate species the same as for the twelve-step reduction mechanism, with one additional species, i.e. NH3. Fifteen-step reduced chemical kinetics mechanism includes following elements: O, H, C and N and their appropriate species the same as for the thirteen-step reduction mechanism, with two additional species, i.e. HO2 and H2O2.

2. Numerical results

Flame structure and emissions have been modelled using the commercial code, where appropriate applications for simulations of burner-stabilized and freely propagating, laminar, lean methane/air premixed flames exist. Reactor models are one-dimensional, allowing the calculation of temperature profile, concentrations of main and intermediates species. Flame propagation velocity can be calculated as a function of distance.

Since GRI-Mech 3.0 gives results that can be compared with results obtained in laboratories during real measurement process. Thus, this mechanism will be the representative as estimation detailed, skeletal chemical mechanisms.

Research of MEF-BL was concentrated on calculating the temperature profiles, axial velocity and concentrations of main species. The flame structure and emissions of burner-stabilized, lean methane-air premixed laminar flames, at initial temperature of 298 K, and pressure of 101kPa, were calculated. Figure 1 shows the comparison of the calculated temperature profile when GRI 3.0 and three different reduced mechanisms were used. Figure 2 shows the comparison of the calculated NOx profile when GRI 3.0 and 15-step reduced chemistry mechanisms are used.

Figure 1. Comparison of temperature distribution, obtained by GRI Mech 3.0 and reduced mechanisms for the excess air ratio equal 1 (MEF-BL).

Figure 2. Comparison of NO distribution, obtained by GRI Mech 3.0 and the 15-step reduced mechanism for the stoichiometric air ratio equal 1(MEF-BL).

Conclusions

General conclusions about the modelling of chemical kinetics of combustion can be drawn:

  • Detailed chemical reaction mechanism for methane-air combustion is practically unknown and impractical

  • Chemical reaction mechanism described in GRI-Mech 3.0 gives the most appropriate results regarding methane-air combustion

  • Due the limits in CFD codes concerning number of chemical reactions and species, it is not possible practically use reduced chemical reaction mechanism described in GRI-Mech 3.0

Literature

[1] C.J. Sung, C.K. Law, and J.-Y. Chen, "Augmented Reduced Mechanisms for NO Emission in Methane Oxidation", Combustion & Flame 125:906-919 (2001).