Bulletin of the American Physical Society
64th Annual Meeting of the APS Division of Fluid Dynamics
Volume 56, Number 18
Sunday–Tuesday, November 20–22, 2011; Baltimore, Maryland
Session D23: Reacting Flows I: Turbulent Combustion |
Hide Abstracts |
Chair: Carlos Pantano, University of Illinois at Urbana-Champaign Room: 326 |
Sunday, November 20, 2011 2:10PM - 2:23PM |
D23.00001: Role of Large Scale Mixing in Soot Evolution in Turbulent Nonpremixed Combustion Michael Mueller, Heinz Pitsch In order for soot to form and grow, long residence times are required at rich mixture fractions. Under these conditions, soot particles grow either by surface reactions with acetylene or by condensation of Polycyclic Aromatic Hydrocarbons (PAH) onto the surface of the particles. In past studies on nonpremixed combustion, condensation of PAH was found to be the dominant growth mechanism, while surface reactions with acetylene played only a minor role. In this work, a recently developed LES model for soot evolution in turbulent nonpremixed combustion is used to examine two configurations in which long residence times under rich conditions are attained in distinctly different ways. In the first configuration, a typical piloted jet flame, soot is formed far downstream from the burner nozzle and grows primarily by PAH condensation. In the second configuration, a bluff body stabilized flame, the dominant soot growth mechanism is found to be surface growth by acetylene. Here, the recirculation zone formed behind the bluff body provides the long residence times needed for soot formation. The combination of very small dissipation rates and a narrow band of mixture fractions in the recirculation zone promotes surface growth by acetylene over PAH condensation. [Preview Abstract] |
Sunday, November 20, 2011 2:23PM - 2:36PM |
D23.00002: Simultaneous soot concentration and strain-rate measurements in turbulent jet flames Venkat Narayanaswamy, Noel Clemens Studies on soot formation are very important both from an environmental standpoint and from a technical perspective. The correlation between the instantaneous soot concentration fields and the corresponding instantaneous strain fields, which would provide important details on the fluid-dynamic aspects of soot formation, relevant for practical combustors, is not understood yet. To address this issue, experiments were performed in our laboratory to study the organization of 2-D soot concentration fields, obtained using LII, and its correlation with corresponding instantaneous 2D strain-rate fields, obtained simultaneously using PIV. The experiments were performed in a turbulent co-flowing jet facility, with an ethylene/N$_{2}$ mixture as the fuel. Different jet-exit Reynolds numbers were obtained by changing the jet-exit velocity, and the corresponding evolution of the soot and strain-field structures is investigated. Our preliminary results show that the instantaneous soot field topology is highly correlated with the instantaneous strain-rate topology. The regions of intense soot concentrations are mainly along the regions of intense strain-rate; furthermore, the soot concentration fields become increasingly convoluted and sparse with increasing Reynolds number. [Preview Abstract] |
Sunday, November 20, 2011 2:36PM - 2:49PM |
D23.00003: Direct numerical simulation of temporally evolving turbulent luminous jet flames with detailed fuel and soot chemistry Vivien Lecoustre, Paul Arias, Somesh Roy, Wei Wang, Zhaoyu Luo, Dan Haworth, Hong Im, Tianfeng Lu, Kwan-Liu Ma, Ramanan Sankaran, Arnaud Trouve Direct numerical simulations of 2D temporally-evolving luminous turbulent ethylene-air jet diffusion flames are performed using a high-order compressible Navier-Stokes solver. The simulations use a reduced mechanism derived from a detailed ethylene-air chemical kinetic mechanism that includes the reaction pathways for the formation of polycyclic aromatic hydrocarbons. The gas-phase chemistry is coupled with a detailed soot particle model based on the method of moments with interpolative closure that accounts for soot nucleation, coagulation, surface growth through HACA mechanism, and oxidation. Radiative heat transfer of CO$_2$, H$_2$O, and soot is treated by solving the radiative transfer equation using the discrete transfer method. This work presents preliminary results of radiation effects on soot dynamics at the tip of a jet diffusion flame with a particular focus on soot formation/oxidation. [Preview Abstract] |
Sunday, November 20, 2011 2:49PM - 3:02PM |
D23.00004: Turbulent propagation of premixed flames: the effects of thermal expansion and integral scale Francesco Creta, Moshe Matalon We study the propagation of a premixed flame in a turbulent environment within the wrinkled flamelet regime. We adopt a hybrid Navier-Stokes/front capturing technique whereby the flame is assimilated to a surface of vanishing thickness separating burnt and unburned regions and subjected to an incident two-dimensional turbulent flow. Among the many observable phenomena, two separate effects on the turbulent propagation speed are investigated, namely that of the thermal expansion (unburned to burnt gas density ratio) and of turbulence integral scale. Turbulent speed is observed to increase and later plateau as thermal expansion increases, this being due to an increase in flame brush thickness with no perceivable curvature variation. The effect of the integral scale of the incoming turbulent field reveals the existence of a particular intermediate value of such scale at which the flame experiences a maximum propagation speed. The same qualitative tendencies are confirmed by a weakly nonlinear Michelson Sivashinsky (MS) model forced with spatially correlated noise that mimics the incoming turbulent flow. Both effects can be combined, together with the effect of turbulence intensity, in a general scaling law for the turbulent propagation speed valid for hydrodynamically stable flames. [Preview Abstract] |
Sunday, November 20, 2011 3:02PM - 3:15PM |
D23.00005: Turbulent Flame Speed and Self Similarity of Expanding Premixed Flames Swetaprovo Chaudhuri, Fujia Wu, Delin Zhu, Chung Law In this study we present experimental turbulent flame speed data measured in constant-pressure expanding turbulent flames, propagating in nearly homogenous isotropic turbulence, in a dual-chamber, fan-stirred vessel. The cold flow is characterized by high speed particle image velocimetry while the flame propagation rate is obtained by tracking high speed Schlieren images of unity Lewis number methane-air flames over wide ranges of pressure and turbulence intensity. It is found that the normalized turbulent flame speed as a function of the average radius scales as a turbulent Reynolds number to the one-half power, where the average radius is the length scale and thermal diffusivity is the transport property, thus showing self-similar propagation. Utilizing this dependence it is found that the turbulent flame speeds from expanding flames and those from Bunsen geometries can be scaled by a single parameter: the turbulent Reynolds number utilizing recent theoretical results obtained by spectral closure of the G equation, after correcting for gas expansion effects. [Preview Abstract] |
Sunday, November 20, 2011 3:15PM - 3:28PM |
D23.00006: An extinction/reignition dynamic method for turbulent combustion Robert Knaus, Carlos Pantano Quasi-randomly distributed locations of high strain in turbulent combustion can cause a nonpremixed or partially premixed flame to develop local regions of extinction called ``flame holes". The presence and extent of these holes can increase certain pollutants and reduce the amount of fuel burned. Accurately modeling the dynamics of these interacting regions can improve the accuracy of combustion simulations by effectively incorporating finite-rate chemistry effects. In the proposed method, the flame hole state is characterized by a progress variable that nominally exists on the stoichiometric surface. The evolution of this field is governed by a partial-differential equation embedded in the time-dependent two-manifold of the flame surface. This equation includes advection, propagation, and flame hole formation (flame hole healing or collapse is accounted by propagation naturally). We present a computational algorithm that solves this equation by embedding it in the usual three-dimensional space. A piece-wise parabolic WENO scheme combined with a compression algorithm are used to evolve the flame hole progress variable. A key aspect of the method is the extension of the surface data to the three-dimensional space in an efficient manner. We present results of this method applied to canonical turbulent combusting flows where the flame holes interact and describe their statistics. [Preview Abstract] |
Sunday, November 20, 2011 3:28PM - 3:41PM |
D23.00007: Intermittency in Premixed Turbulent Reacting Flows Peter Hamlington, Alexei Poludnenko, Elaine Oran Characterizing the intermittency of velocity gradient and scalar gradient fields in turbulent reacting flows is important for developing a better understanding of the interactions between turbulence and flames. Here we examine intermittency in premixed reacting flows using numerical simulations of stoichiometric hydrogen-air combustion at a range of turbulence intensities. Simulations of homogeneous isotropic turbulence with a nonreacting passive scalar are also carried out in order to allow comparisons with the reacting flow results. We examine intermittency by calculating probability density functions and moments of the local enstrophy, energy dissipation rate, and scalar dissipation rate. Conditional analyses based on local, instantaneous values of the reactant mass fraction are used to study variations in the statistics through the flame. We observe variations in the intermittency depending on the intensity of the turbulence, the location in the flame, and the quantity under consideration. We discuss the implications of these results for the flame structure, and also provide an explanation for the observed results by considering the two-way interactions between turbulence and premixed flames. [Preview Abstract] |
Sunday, November 20, 2011 3:41PM - 3:54PM |
D23.00008: Using LCS to study coherent structures in reacting flows Melissa Green, Peter Hamlington, Alexei Poludnenko, Elaine Oran Previous research has shown that chemical reactions in a compressible fluid flow interact strongly with the surrounding turbulence both in quantitative measures and qualitative character. In the case of a flame propagating through homogeneous isotropic turbulence, the rapid flow expansion generated in the reaction zone causes a significant attenuation in the vorticity. This suppression of the vorticity magnitude complicates the tracking of individual coherent structures using Eulerian methods, therefore we use Lagrangian coherent structures to study the nature of the vortex dynamics, focusing on structure creation, destruction, and reorientation. [Preview Abstract] |
Sunday, November 20, 2011 3:54PM - 4:07PM |
D23.00009: \textit{A Priori} Analysis of Subgrid Mass Flux Vectors from Massively Parallel Direct Numerical Simulations of High Pressure H2/O2 Reacting Shear Layers Justin Foster, Richard Miller Direct Numerical Simulations (DNS) are conducted for temporally developing reacting H2/O2 shear layers at an ambient pressure of 100atm. The compressible form of the governing equations are coupled with the Peng Robinson real gas equation of state and are solved using eighth order central finite differences and fourth order Runge Kutta time integration with resolutions up to $\sim $3/4 billion grid points. The formulation includes a detailed pressure dependent kinetics mechanism having 8 species and 19 steps, detailed property models, and generalized forms of the multicomponent heat and mass diffusion vectors derived from nonequilibrium thermodynamics and fluctuation theory. The DNS is performed over a range of Reynolds numbers up to 4500 based on the free stream velocity difference and initial vorticity thickness. The results are then analyzed in an \textit{a priori} manner to illustrate the role of the subgrid mass flux vector within the filtered form of the governing equations relevant to Large Eddy Simulations. The subgrid mass flux vector is found to be a significant term; particularly within localized regions of the flame. [Preview Abstract] |
Sunday, November 20, 2011 4:07PM - 4:20PM |
D23.00010: Characteristics of Turbulent Premixed Flames under the Pressure Rising Process in a Closed Vessel Naoya Fukushima, Basmil Yenerdag, Masayasu Shimura, Mamoru Tanahashi, Toshio Miyauchi In a closed vessel such as SI engines, the internal pressure increases due to dilatation during the combustion after the ignition. To clarify quantitative characteristics of turbulent premixed flames under the pressure rising process, direct numerical simulation (DNS) of turbulent premixed flames in a closed vessel at relatively high Reynolds number has been conducted. Detailed kinetic mechanism for hydrogen-air mixtures is used. Because of the local pressure rise, turbulence is enhanced at the unburnt side and flame surface is distorted, which results in increase of the flame surface. Heat release rate of each flame element is augmented since the pressure rise makes flame thickness thin. Under this pressure rising process, the flame thickness, the flame front curvature and the local heat release rate can be scaled by laminar flame thickness and the maximum heat release rate obtained from one dimensional DNS of laminar flame propagation by using averaged temperature in the unburnt region of the vessel as the inlet temperature. The tangential strain rate on the flame front can be scaled by Taylor micro scale averaged in the unburnt side. The local heat release rate is positively correlated with the curvature and the tangential strain. The time evolution of the flame surface area is also investigated quantitatively. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700