Computational Study on the Effects of Turbulence Intensity and Pulse Frequency in Soot and Nox Emissions in Gaseous Diffusion Flames

The main objective of the work presented in this paper will be to present and discuss the development and implementation of a numerical model to simulate soot formation and depletion in turbulent diffusion flames. The relevance of such model lays on the importance of tying the formation of soot to the ongoing reaction mechanism so that it is fully integrated into the combustion process. The model presented is capable of taking into account the direct effects of turbulence on the amount of soot that is produced in non-premixed flames. This research focuses on the study of an axi-symmetric C2H2 – air turbulent diffusion flame issued from a 3mm round jet with Reynolds number values between 8000 and 16500. The trend observed in the net production of soot with respect to the turbulence intensity is in good agreement with the empirical results found in the literature. These reveal a decrease in soot formation with increasing turbulence. This interaction between particulates and turbulence was then exploited in order to develop a mechanical technique by which a simultaneous reduction in soot and NOx was achieved. The level of turbulence was increased locally by application of a sinusoidal pulse frequency to the fuel stream. Such technique reduced the size of the soot-prone, fuel-rich region, with respect to the equivalent steady state flame, by means of enhancing the mixing between the fuel and the oxidizer. The Realizable k-ε model was employed to solve the turbulence transport, whereas the reaction was simulated with a 1-reaction step mechanism and the turbulence-chemistry interaction was solved using the Eddy Dissipation Model. The size of the time steps employed in the unsteady configuration for pulsed flames was 1/20th of the pulse period. The soot model employed in this work observed two different stages in the soot formation process: nuclei inception and particle growth. As a result two transport equations are solved mass fractions of nuclei and soot respectively. The implementation of this model is achieved through user-defined functions that supersede the source terms in the default soot transport equations. Furthermore, the production of NOx was simulated using a classic Zel’dovich mechanism with partial equilibrium assumption for release of atomic oxygen, O, and hydroxyl groups, OH. COMPUTATIONAL STUDY ON THE EFFECTS OF TURBULENCE INTENSITY AND PULSE FREQUENCY... 101