Experimental analysis of an oblique turbulent flame front propagating in a stratified flow

This paper details the experimental study of a turbulent V-shaped flame expanding in a nonhomogeneous premixed flow. Its aim is to characterize the effects of stratification on turbulent flame characteristics. The setup consists of a stationary V-shaped flame stabilized on a rod and expanding freely in a lean premixed methane-air flow. One of the two oblique fronts interacts with a stratified slice, which has an equivalence ratio close to one and a thickness greater than that of the flame front. Several techniques such as PIV and CH^* chemiluminescence are used to investigate the instantaneous fields, while laser Doppler anemometry and thermocouples are combined with a concentration probe to provide information on the mean fields. First, in order to provide a reference, the homogeneous turbulent case is studied. Next, the stratified turbulent premixed flame is investigated. Results show significant modifications of the whole flame and of the velocity field upstream of the flame front. The analysis of the geometric properties of the stratified flame indicates an increase in flame brush thickness, closely related to the local equivalence ratio.     

Formation,growth, and transport of soot in a three-dimensional turbulent non-premixed jet flame

The formation, growth, and transport of soot is investigated via large scale numerical simulation in a three-dimensional turbulent non-premixed n-heptane/air jet flame at a jet Reynolds number of 15,000. For the first time, a detailed chemical mechanism, which includes the soot precursor naphthalene and a high-order method of moments are employed in a three-dimensional simulation of a turbulent sooting flame. The results are used to discuss the interaction of turbulence, chemistry, and the formation of soot. Compared to temperature and other species controlled by oxidation chemistry, naphthalene is found to be affected more significantly by the scalar dissipation rate. While the mixture fraction and temperature fields show fairly smooth spatial and temporal variations, the sensitivity of naphthalene to turbulent mixing causes large inhomogeneities in the precursor fields, which in turn generate even stronger intermittency in the soot fields. A strong correlation is apparent between soot number density and the concentration of naphthalene. On the contrary, while soot mass fraction is usually large where naphthalene is present, pockets of fluid with large soot mass are also frequent in regions with very low naphthalene mass fraction values. From the analysis of Lagrangian statistics, it is shown that soot nucleates and grows mainly in a layer close to the flame and spreads on the rich side of the flame due to the fluctuating mixing field, resulting in more than half of the total soot mass being located at mixture fractions larger than 0.6. Only a small fraction of soot is transported towards the flame and is completely oxidized in the vicinity of the stoichiometric surface. These results show the leading order effects of turbulent mixing in controlling the dynamics of soot in turbulent flames. Finally, given the difficulties in obtaining quantitative data in experiments of turbulent sooting flames, this simulation provides valuable data to guide the development of models for Large Eddy Simulation and Reynolds Average Navier Stokes approaches.     

Flame brush characteristics and burning velocities of premixed turbulent methane/air Bunsen flames

The flame brush characteristics and turbulent burning velocities of premixed turbulent methane/air flames stabilized on a Bunsen-type burner were studied. Particle image velocimetry and Rayleigh scattering techniques were used to measure the instantaneous velocity and temperature fields, respectively. Experiments were performed at various equivalence ratios and bulk flow velocities from 0.7 to 1.0, and 7.7 to 17.0 m/s, respectively. The total turbulence intensity and turbulent integral length scale were controlled by the perforated plate mounted at different positions upstream of the burner exit. The normalized characteristic flame height and centerline flame brush thickness decreased with increasing equivalence ratio, total turbulence intensity, and longitudinal integral length scale, whereas they increased with increasing bulk flow velocity. The normalized horizontal flame brush thickness increased with increasing axial distance from the burner exit and increasing equivalence ratio. The non-dimensional leading edge and half-burning surface turbulent burning velocities increased with increasing non-dimensional turbulence intensity, and they decreased with increasing non-dimensional bulk flow velocity when other turbulence statistics were kept constant. Results show that the non-dimensional leading edge and half-burning surface turbulent burning velocities increased with increasing non-dimensional longitudinal integral length scale. Two correlations to represent the leading edge and half-burning surface turbulent burning velocities were presented as a function of the equivalence ratio, non-dimensional turbulence intensity, non-dimensional bulk flow velocity, and non-dimensional longitudinal integral length scale. Results show that the half-burning surface turbulent burning velocity normalized by the bulk flow velocity decreased as the normalized characteristic flame height increased.     

Flame acceleration in the early stages of burning in tubes

Acceleration of premixed laminar flames in the early stages of burning in long tubes is considered. The acceleration mechanism was suggested earlier by Clanet and Searby [Combust. Flame 105 (1996) 225]. Acceleration happens due to the initial ignition geometry at the tube axis when a flame develops to a finger-shaped front, with surface area growing exponentially in time. Flame surface area grows quite fast but only for a short time. The analytical theory of flame acceleration is developed, which determines the growth rate, the total acceleration time, and the maximal increase of the flame surface area. Direct numerical simulations of the process are performed for the complete set of combustion equations. The simulations results and the theory are in good agreement with the previous experiments. The numerical simulations also demonstrate flame deceleration, which follows acceleration, and the so-called “tulip flames.”     

Direct numerical simulation of hydrogen addition in turbulent premixed Bunsen flames using flamelet-generated manifold reduction

Direct numerical simulations of a lean premixed turbulent Bunsen flame with hydrogen addition have been performed. We show the results for a case with equivalence ratio of 0.7 and a molar fractional distribution of 40% H₂ and 60% CH₄. The flamelet-generated manifold technique is used to reduce the chemistry; flamelets with different equivalence ratios and inflow temperature are used to account for stretch effects that are enhanced by preferential diffusion. The three-dimensional simulation clearly shows enhanced burning velocity in regions convex toward the reactants and reduced burning velocity with possible extinction in regions concave toward the reactants. To obtain these effects it was found to be necessary to include two three-dimensional transport equations with essentially different diffusivities. This point is illustrated by comparison of the results with cases in which either a single transport equation was used or two transport equations with minor differences in diffusivities were used. These latter cases incorporated preferential diffusion in the 1D flamelets (and thus in the manifold), but not in the three-dimensional transport. Thus the three-dimensional preferential diffusion effects are shown to enhance curvature and thereby to increase the turbulent burning velocity and reduce the mean flame height. In addition the turbulent burning velocity increases because hydrogen addition leads to a larger laminar flamelet consumption speed. To demonstrate this second effect, results of the cases mentioned above are compared to the results of simulations of the Bunsen flame with 0% hydrogen added to the fuel.     

Flame-sound interaction in tubes with nonslip walls

Flame interaction with sound is studied for a premixed flame propagating to the closed end of a tube with nonslip walls. The flow geometry is similar to that in the classical Searby experiments on flame-acoustic interaction [Combust. Sci. Technol. 81 (1992) 221]. The problem is solved by direct numerical simulations of the combustion equations. The flame-sound interaction strongly influences oscillations of the flame front. Particularly, sound noticeably increases the oscillation amplitude in comparison with that in an open tube with nonreflecting boundary conditions at the ends studied previously. Oscillations become especially strong in the second part of the tube, where flame pulsations are in resonance with the acoustic wave. Parameters of the flame oscillations are investigated for different values of the tube width and length. It is demonstrated that the oscillations are stronger in wider tubes, though the investigated tube width is limited by the computational facilities. In sufficiently wide tubes, violent folding of a flame front is observed because of the flame-acoustic resonance. By increasing the Lewis number, one also increases the oscillation amplitude.     

Effects of Lewis number on flame surface density transport in turbulent premixed combustion

The transport of flame surface density (FSD) in turbulent premixed flames has been studied using a database obtained from Direct Numerical Simulation (DNS). Three-dimensional freely propagating developing statistically planar turbulent premixed flames have been examined over a range of global Lewis numbers from 0.6 to 1.2. Simplified chemistry has been used and the emphasis is on the effects of Lewis number on FSD transport in the context of Reynolds-averaged closure modelling. Under the same initial conditions of turbulence, flames with low Lewis numbers are found to exhibit counter-gradient transport of FSD, whereas flames with higher Lewis numbers tend to exhibit gradient transport of FSD. Stronger heat release effects for lower Lewis number flames are found to lead to an increase in the positive (negative) value of the dilatation rate (normal strain rate) term in the FSD transport equation with decreasing Lewis number. The contribution of flame curvature to FSD transport is found to be influenced significantly by the effects of Lewis number on the curvature dependence of the magnitude of the reaction progress variable gradient, and on the combined reaction and normal diffusion components of displacement speed. The modelling of the various terms of the FSD transport equation has been analysed in detail and the performance of existing models is assessed with respect to the terms assembled from corresponding quantities extracted from DNS data. Based on this assessment, suitable models are identified which are able to address the effects of non-unity Lewis number on FSD transport, and new or modified models are suggested wherever necessary.     

Fire dynamics simulation of a turbulent buoyant flame using a mixture-fraction-based combustion model

Fire dynamics simulations of a 7.1-cm buoyant turbulent diffusion flame were performed using a mixture-fraction-based combustion model. In our previous work, good agreement between the measured and the calculated fire flow field was achieved with carefully selected domain and grid sizes using a Lagrangian thermal-element combustion model. The Lagrangian thermal-element model exhibits qualitative as well as quantitative differences in the measured and calculated temperature profiles in the flame zone. The number of Lagrangian thermal elements must be carefully selected and the model is not designed to provide insights into the species distributions in the fire. To address these issues, a mixture-fraction-based combustion model was used in the present work. The domain and grid size dependence using this model are documented. Comparisons between the measured and the calculated velocities, mixture fractions and temperatures show that the mixture-fraction-based combustion model captures the qualitative and quantitative fire behavior very well.     

Effects of position and frequency of obstacles on turbulent premixed propagating flames

This paper studies the effects of the number and location of solid obstacles on the rate of propagation of turbulent premixed flames. A vented explosion chamber is constructed where controlled premixed flames are ignited from rest to propagate past grids or baffles plates as well as other solid obstacles strategically positioned in the chamber. Laser Induced Fluorescence (LIF) is used to image OH which is used as an indicator of the reaction zone while pressure transducers are used to obtain pressure-time traces. Single grids or baffle plates located at different distances from the ignition source are tested. Two as well as three baffle plates are also investigated in varying configurations. It is found that while the peak overpressure increases with increasing number of grids or baffle plates, a limit is reached where the pressure starts to decrease. The location of the obstacles is found to have a significant effect on the overpressure and the flame structure. Higher overpressures are obtained when the baffle plates and obstacles are stacked closer together hence not allowing turbulence to decay. LIF images for OH show that the reaction zones become more contorted with increasing number of baffle plates in the flame path.     

Effects of fresh gas velocity and thermal expansion on the structure of a Bunsen flame tip

Numerical computations and order-of-magnitude estimates are used to describe the tip region of a Bunsen flame where the flame departs from a planar flame at an angle to the incoming fresh gas flow. A single irreversible Arrhenius reaction with high activation energy is assumed. The well-known linear relation between flame velocity and curvature is recovered in the thermodiffusive limit, when the thermal expansion of the gas is left out, for velocities of the fresh gas (U₀) only slightly larger than the velocity of a planar flame (U_L), provided this flame is stable. For large values of the velocity ratio U₀/U_L, the tip region becomes slender and the curvature of the reaction sheet at the tip increases proportionally to U₀/U_L. The thermal expansion of the gas across the flame reduces the aspect ratio of the tip region. A qualitative analysis of the structure of the tip region for very exothermic reactions shows that this region ceases to be slender when the burnt-to-fresh gas temperature ratio becomes of the order of the velocity ratio U₀/U_L. For even larger values of the temperature ratio, the tip region becomes a cap of characteristic size not very different from the thickness of a planar flame.     

Effect of turbulent motions at different length scales on turbulent premixed swirl-stabilised flame topology

Three-dimensional direct numerical simulation data of H₂-air turbulent swirling premixed combustion at two different swirl numbers are analysed to investigate the local reaction zone morphology and its relation with local turbulent motions at different length scales. The effect of small scale turbulent mixing on local flames is investigated, and the results have shown that the contribution of microscale turbulent diffusivity on the local flamelet is insignificant, although there is some evidence of flame thinning for the higher swirl number case. The flame morphology such as high-level convolution and interacting flames, on the other hand, shows greater influence on local flamelets, suggesting the importance of local reaction zone topology on overall combustion processes. The local reaction zones are analysed by using the shapefinders to quantify their topology. Although the shapefinders showed various local reaction zone shapes consisting of “pancakes” and “tubes” and intermissive intense reaction zone distributions, the smallest characteristic length scale shows that the local reaction zones are thin. Finally, the relationship between these local reaction zone topology and turbulent motions at different sizes were discussed. The local reaction zone topology has a direct relation with Taylor microscale, integral length scale and their associated velocity scale, whereas almost no correlation is observed with Kolmogorov length scale, in the presence of inhomogeneous turbulence and strong mean shears. The present results suggest the importance of Taylor microscale on flame surface topology, which is often understated in turbulent combustion modelling frameworks.     

Burning velocity and flame structure of CH4/NH3/air turbulent premixed flames at high pressure

Ammonia is one of the most promising alternative fuels. In particular, ammonia combustion for gas turbine combustors for power generation is expected. To shift the fuel for a gas turbine combustor to ammonia step-by-step, the partial replacement of natural gas by ammonia is considered. To reveal the turbulent combustion characteristics, CH₄/NH₃/air turbulent premixed flame at 0.5 MPa was experimentally investigated. The ammonia ratio based on the mole fraction and lower heating value was varied from 0 to 0.2. The results showed that the ratio of the turbulent burning velocity and unstretched laminar burning velocity decreased with an increase in the ammonia ratio. The reason for this variation is that the flame area decreased with an increase in the ammonia ratio as the flame surface density decreased and the fractal inner cutoff increased. The volume fractions in the turbulent flame region were almost the same with ammonia addition, indicating that combustion oscillation can be handled in a manner similar to that for the case of natural gas for CH₄/NH₃/air flames.     

Numerical study on laminar burning velocity of ammonia flame with methanol addition

The combustion characteristics of ammonia/methanol mixtures were investigated numerically in this study. Methanol has a dramatic promotive effect on the laminar burning velocity (LBV) of ammonia. Three mechanisms from literature and another four self-developed mechanisms constructed in this study were evaluated using the measured laminar burning velocities of ammonia/methanol mixtures from Wang et al. (Combust.Flame. 2021). Generally, none of the selected mechanisms can precisely predict the measured laminar burning velocities at all conditions. Aiming to develop a simplified and reliable mechanism for ammonia/methanol mixtures, the constructed mechanism utilized NUI Galway mechanism (Combust.Flame. 2016) as methanol sub-mechanism and the Otomo mechanism (Int. J. Hydrogen. Energy. 2018) as ammonia sub-mechanism was optimized and reduced. The reduced mechanism entitled ‘DNO-NH3’, can accurately reproduce the measured laminar burning velocities of ammonia/methanol mixtures under all conditions. A reaction path analysis of the ammonia/methanol mixtures based on the DNO-NH3 mechanism shows that methanol is not directly involved in ammonia oxidation, instead, the produced methyl radicals from methanol oxidization contribute to the dehydrogenation of ammonia. Besides, NO_x emission analysis demonstrates that 60% methanol addition results in the highest NO_x emissions. The most important reactions dominating the NO_x consumption and production are identified in this study.     

Measurements of the heat release rate integral in turbulent premixed stagnation flames with particle image velocimetry

A new definition of turbulent consumption speed is proposed in this work that is based on the heat release rate integral, rather than the mass burning rate integral. Its detailed derivation and the assumptions involved are discussed in a general context that applies to all properly defined reaction progress variables. The major advantage of the proposed definition is that it does not require the thin-flame assumption, in contrast to previous definitions. Experimental determination of the local turbulent displacement speed, SD, and the local turbulent consumption speed, SC, is also demonstrated with the particle image velocimetry technique in three turbulent premixed stagnation flames. The turbulence intensity of these flames is of the same order of the laminar burning velocity. Based on the current data, a model equation for the local mean heat release rate is proposed. The relationship between SD and SC is discussed along with a possible modeling approach for the turbulent displacement speed.     

DNS of a premixed turbulent V flame and LES of a ducted flame using a FSD-PDF subgrid scale closure with FPI-tabulated chemistry

Two complementary simulations of premixed turbulent flames are discussed. Low Reynolds number two-dimensional direct numerical simulation of a premixed turbulent V flame is first performed, to further analyze the behavior of various flame quantities and to study key ingredients of premixed turbulent combustion modeling. Flame surface density, subgrid-scale variance of progress variables, and unresolved turbulent fluxes are analyzed. These simulations include fully detailed chemistry from a flame-generated tabulation (FPI) and the analysis focuses on the dynamics of the thin flame front. Then, a novel subgrid scale closure for large eddy simulation of premixed turbulent combustion (FSD-PDF) is proposed. It combines the flame surface density (FSD) approach with a presumed probability density function (PDF) of the progress variable that is used in FPI chemistry tabulation. The FSD is useful for introducing in the presumed PDF the influence of the spatially filtered thin reaction zone evolving within the subgrid. This is achieved via the exact relation between the PDF and the FSD. This relation involves the conditional filtered average of the magnitude of the gradient of the progress variable. In the modeling, this conditional filtered mean is approximated from the filtered gradient of the progress variable of the FPI laminar flame. Balance equations providing mean and variance of the progress variable together with the measure of the filtered gradient are used to presume the PDF. A three-dimensional larger Reynolds number flow configuration (ORACLES experiment) is then computed with FSD-PDF and the results are compared with measurements.     

Hydrogen autoignition in a turbulent jet with preheated co-flow air

The autoignition of hydrogen in a turbulent jet with preheated air is studied computationally using the stand-alone one-dimensional turbulence (ODT) model. The simulations are based on varying the jet Reynolds number and the mixture pressure. Also, computations are carried out for homogeneous autoignition at different mixture fractions and the same two pressure conditions considered for the jet simulations. The simulations show that autoignition is delayed in the jet configuration relative to the earliest autoignition events in homogeneous mixtures. This delay is primarily due to the presence of scalar dissipation associated with the scalar mixing layer in the jet configuration as well as with the presence of turbulent stirring. Turbulence plays additional roles in the subsequent stages of the autoignition process. Pressure effects also are present during the autoignition process and the subsequent high-temperature combustion stages. These effects may be attributed primarily to the sensitivity of the autoignition delay time to the mixture conditions and the role of pressure and air preheating on molecular transport properties. The overall trends are such that turbulence increases autoignition delay times and accordingly the ignition length and pressure further contribute to this delay.     

Large eddy simulation of soot formation in a turbulent non-premixed jet flame

A recently developed subgrid model for soot dynamics [H. El-Asrag, T. Lu, C.K. Law, S. Menon, Combust. Flame 150 (2007) 108-126] is used to study the soot formation in a non-premixed turbulent flame. The model allows coupling between reaction, diffusion and soot (including soot diffusion and thermophoretic forces) processes in the subgrid domain without requiring ad hoc filtering or model parameter adjustments. The combined model includes the entire process, from the initial phase, when the soot nucleus diameter is much smaller than the mean free path, to the final phase, after coagulation and aggregation, where it can be considered in the continuum regime. A relatively detailed but reduced kinetics for ethylene-air is used to simulate an experimentally studied non-premixed ethylene/air jet diffusion flame. Acetylene is used as a soot precursor species. The soot volume fraction order of magnitude, the location of its maxima, and the soot particle size distribution are all captured reasonably. Along the centerline, an initial region dominated by nucleation and surface growth is established followed by an oxidation region. The diffusion effect is found to be most important in the nucleation regime, while the thermophoretic forces become more influential downstream of the potential core in the oxidation zone. The particle size distribution shows a log-normal distribution in the nucleation region, and a more Gaussian like distribution further downstream. Limitations of the current approach and possible solution strategies are also discussed.     

Numerical simulations of premixed flame cellular instability for a simple chain-branching model

Two-dimensional direct numerical simulations are performed to investigate the non-linear dynamics of low Lewis number premixed flames, in the context of a two-step chain-branching chemistry model. This consists of a thermally-neutral, but temperature sensitive, chain-branching step which produces intermediates such as radicals and an exothermic, zero activation energy chain-completion step which converts the intermediates into products. Emphasis is on examining the role of intermediates in the flame structure on the cellular instability and in comparing and contrasting with previous one-step chemistry model solutions. When intermediates are present only in small concentrations in the underlying one-dimensional flame structure, the two-step cellular dynamics are qualitatively similar to those of the one-step model, including cell-splitting and re-merging, symmetry breaking bifurcations and formation of asymmetric cells, localized quenching of the flame front and a significant enhancement of the flame speed. However, a higher peak value of the intermediates concentration, corresponding to a more distributed heat release, is shown to have a significant stabilizing effect, e.g., in a domain of fixed transverse size, the fully developed cellular structure and flame speed remain closer to those of the one-dimensional flame.