Transient convective burning of a periodic fuel-droplet array

The transient convective burning of fuel-droplets interacting within 3-D infinite periodic arrays in a hot gas stream is numerically studied for the first time, with considerations of droplet regression, deceleration due to the drag of the droplets, internal liquid motion, variable properties, non-uniform liquid temperature, surface tension, and n-octane one-step oxidation kinetics. Depending upon the initial conditions and other constraints, a flame is established early as either a wake flame or an envelope flame. An initial envelope flame remains an envelope flame, and an initial wake flame has a tendency to develop from a wake flame to an envelope flame. The flame shows no strong tendency to modify significantly the standoff distance during the lifetime of the droplet. For an initial wake flame, the moment of wake-to-envelope transition is advanced as the initial droplet spacing (intermediate) is decreased, but tends to be postponed as the initial droplet spacing is further reduced. The burning rate at smaller initial droplet spacing or smaller initial Reynolds number might be greater for some period during the lifetime because of an earlier wake-to-envelope transition which elevates the average surface temperature. Lower ambient temperature yields a later wake-to-envelope transition time and smaller mass burning rate. At the lower ambient pressure with the same initial relative stream velocity, the average surface temperature is reduced, the wake-to-envelope transition is later, and the mass burning rate is smaller. Validation of our analysis is made by comparing with the results of an isolated droplet Wu and Sirignano [11].     

Vaporization and combustion in three-dimensional droplet arrays

The quasi-steady vaporization and combustion of multiple-droplet arrays is studied numerically. Utilizing the Shvab–Zeldovich formulation, a transformation of the governing equations to a three-dimensional Laplace’s equation is performed, and the solution to Laplace’s equation is obtained numerically to find the effects of droplet interactions in symmetric, multiple-droplet arrays. Vaporization rates, flame surface shapes, and flame locations are found for different droplet array configurations and fuels. The number of droplets, the droplet arrangement within the arrays, and the droplet spacing within the arrays are varied to determine the effects of these parameters. Computations are performed for uniformly spaced three-dimensional arrays of up to 216 droplets, with center-to-center spacing ranging from 3 to 25 droplet radii. As a result of the droplet interactions, the number of droplets and relative droplet spacing significantly affect the vaporization rate of individual droplets within the array, and consequently the flame shape and location. For small droplet spacing, the individual droplet vaporization rate decreases below that obtained for an isolated droplet by several orders of magnitude. A similarity parameter which correlates vaporization rates with array size and spacing is identified. Individual droplet flames, internal group combustion, and external group combustion can be observed depending on the droplet geometry and boundary conditions.     

对流速度对燃料液滴火焰形态及燃烧的影响

考虑气相非稳态及液滴内部环流,建立运动液滴非稳态蒸发燃烧模型.模型采用动网格方法精确追踪液滴表面位置,采用守恒方程组更新液滴表面边界条件.根据单步全局化学反应机理,仿真研究正庚烷燃料液滴在不同对流速度下的火焰形态及燃烧.结果表明:运动液滴内部环流使液滴内部低温区向环流中心移动.当液滴运动速度大于某临界值后,火焰形态由包覆火焰转变为尾迹火焰.包覆火焰的富燃区范围、高温区范围及燃烧速率明显较尾迹火焰大;包覆火焰的液滴表面温度及表面蒸发流率分布也明显不同于尾迹火焰.     

On the sublimit solution branches of the stripe patterns formed in counterflow diffusion flames by diffusional–thermal instability

The nonlinear dynamics of striped diffusion flames, formed in a two-dimensional counterflow by diffusional–thermal instability with Lewis numbers sufficiently less than unity, is investigated numerically by examining various two-dimensional flame-structure solutions bifurcating from the one-dimensional steady solution. The Lewis numbers for fuel and oxidizer are identically set to be 0.3, and an overall single-step Arrhenius-type chemical reaction with a Zel'dovich number of 7 is employed as the chemistry model. Particular attention is focused on the flame-stripe solution branches in the sub-extinction regime and on the hysteresis encountered during the transition between different solution branches. In the numerical simulations, a nonlinear solution with eight stripes is first realized from the one-dimensional solution at a Damköhler number slightly greater than the extinction Damköhler number. The eight-stripe solution survives Damköhler numbers much smaller than the extinction Damköhler number until successive bifurcations, leading to the doubling of the pattern wavelength, occur at the subsequent forward-transition conditions. At the first forward-transition Damköhler number occurs the transition to a four-stripe solution, which in turn transits to a two-stripe solution at the second forward-transition Damköhler number, a value somewhat smaller than the first. However, further transition from a two-stripe solution to a one-stripe solution is not always possible even if a one-stripe solution can be accessed independently for particular initial conditions. The Damköhler-number ranges and shapes for the two-stripe and one-stripe solutions are found to be virtually identical, implying that each stripe could be an independent structure if the distance between stripes is sufficiently large. By increasing the Damköhler number, backward transitions can be observed. In comparison with the forward-transition Damköhler numbers, the corresponding backward-transition Damköhler numbers are always much greater, thereby indicating significant hysteresis between the stripe patterns of strained diffusion flames.     

The role of unequal diffusivities in ignition and extinction fronts in strained mixing layers

We have studied flame propagation in a strained mixing layer formed between a fuel stream and an oxidizer stream, which can have different initial temperatures. Allowing the Lewis numbers to deviate from unity, the problem is first formulated within the framework of a thermo-diffusive model and a single irreversible reaction. A compact formulation is then derived in the limit of large activation energy, and solved analytically for high values of the Damköhler number. Simple expressions describing the flame shape and its propagation velocity are obtained. In particular, it is found that the Lewis numbers affect the propagation of the triple flame in a way similar to that obtained in the studies of stretched premixed flames. For example, the flame curvature determined by the transverse enthalpy gradients in the frozen mixing layer leads to flame-front velocities which grow with decreasing values of the Lewis numbers.

The analytical results are complemented by a numerical study which focuses on preferential-diffusion effects on triple flames. The results cover, for different values of the fuel Lewis number, a wide range of values of the Damköhler number leading to propagation speeds which vary from positive values down to large negative values     

Autoignition and early flame behavior of a spherical cluster of 49 monodispersed droplets

Autoignition and early flame behavior of a spherical cluster of 49 monodispersed droplets in a high-temperature air were examined in microgravity. The monodispersed suspended-droplet cluster (MSDC) model with which both droplet spacing and initial droplet diameter were well-controlled was developed, and the solidified-fuel fiber-suspension technique was utilized for making the MSDC model. The tested 3D MSDC models had the HCP (hexagonal closest packing) structure. Individual flames, which enveloped each droplet, or group flame, which enveloped the whole droplet cluster, were formed immediately after ignition. The flame changed from the group flame to a cluster of the individual flames either with increasing the droplet spacing or decreasing the initial droplet diameter. Each of the individual flames merged into the group flame with the lapse of time. Burning sphere diameter decreased at the beginning, and then increased. The transition from the individual flames to the group flame occurred around the time period at which the burning sphere diameter reached its minimum. The time period at which the burning sphere diameter reached its maximum was delayed and the expansion rate of the burning sphere was enhanced with decreasing the droplet spacing or with increasing the initial droplet diameter.     

Flame structure and stabilization of lean-premixed sprays in a counterflow with low-volatility fuel

An experimental study was performed on the combustion of lean-premixed spays in a counterflow. n-Decane was used as a liquid fuel with low volatility. The flame structure and stabilization were discussed based on the flame-spread mechanism of a droplet array with a low-volatility fuel. The spray flame consisted of a blue region and a yellow luminous region. The flame spread among droplets and group-flame formation through the droplet interaction were observed on the premixed spray side, while envelope flames were also observed on the opposing airflow side. The blue-flame region consisted of premixed flames propagating in the mixture layer around each droplet, the envelope diffusion flames around each droplet, the lower parts of the group diffusion flame surrounding each droplet cluster, and the envelope flame around droplets passing through the group flame. The flame was stabilized within a specific range of the mean droplet diameter via a balance between the droplet velocity and the flame-spread rate of the premixed spray.     

Observation of droplet motion during flame spread on three-fuel-droplet array with a pendulum suspender

Flame spread on a fuel droplet array has been studied as a simple model of spray combustion. A three-fuel-droplet array with a pendulum suspender was employed to investigate interactions between flame spread and droplet motion in the axial direction. Initial droplet diameter was 0.8 mm, and fuel was n-heptane. A silicon carbide pendulum suspender of 15 μm in diameter and 30 mm in length was used for the third droplet. The first fixed droplet was ignited by electric spark. Behavior of the flame and the third droplet was observed using a high-speed video camera with an image intensifier. Particle tracking velocimetry (PTV) measurements were performed to explain the behavior of the third movable droplet. The dimensionless droplet span, which is the average of droplet-to-droplet distances divided by the average initial diameter of the three droplets, was varied from 2.5 to 8 for observing flame spread, and fixed at 5.5 for PTV measurements. It was observed that the third droplet moved away from the second droplet before the flame spread to the third droplet. The displacement of the third droplet is remarkable when the dimensionless droplet span is close to the limit of flame spread. This implies that the movement of the droplet decreases the dimensionless span of the flame spread limit and the flame spread speed near the flame spread limit. Results of PTV measurements suggest that the heat expansion wave, caused by ignition of the premixture which was accumulated around the second droplet, and the burned gas flow from the second droplet pushed away the third droplet; then natural convection, induced by the flames of the first and second droplets, drew the third droplet to the second droplet. The heat expansion wave and the burned gas flow of the second droplet reached nearly 12 in dimensionless span.     

Bifurcation characteristics and flame dynamics of a ducted non-premixed flame with finite rate chemistry

The influence of system parameters such as the flame location, Peclet number and Damköhler number on the bifurcation characteristics and flame dynamics of a ducted non-premixed flame with finite rate chemistry is presented in this paper. In the bifurcation plot with flame location as the bifurcation parameter, subcritical Hopf bifurcation is found for lower values of flame location and supercritical Hopf bifurcation for higher values of flame location, for all the Damköhler numbers used in this study. The flame shapes are captured at eight different phases of a cycle of time series data of acoustic velocity at both the fold and Hopf points for bifurcation with flame location as the parameter. We find that the range of flame height variations at the Hopf point is more than the range of flame height variations obtained at the fold point. We also find that the flame oscillates in the same phase as pressure fluctuation but in a phase different from both velocity and heat release rate fluctuations in the region of hysteresis for bifurcation with flame location. The non-dimensional hysteresis width is plotted as a function of Damköhler number for variation of flame location in the subcritical region. An inverse power law relation is found between the non-dimensional hysteresis width and the Damköhler number. The bifurcation plot with Peclet number as parameter shows a subcritical Hopf bifurcation.     

Numerical investigation of laminar cross-flow non-premixed flames in the presence of a bluff-body

A knowledge of flame stability regimes in the presence of cylindrical bluff-bodies of various dimensions is essential to design non-premixed burners. The reacting flow field in such cases is reported to be three-dimensional and unsteady. In the literature, only a few experimental investigations with limited measurements are available. Therefore, in this work, a detailed numerical study of laminar cross-flow non-premixed methane–air flames in the presence of a square cylinder is presented. The flow, temperature, species and reaction fields have been predicted using a comprehensive transient three-dimensional reacting flow model with detailed chemical kinetics and variable thermo-physical properties, in order to get a good insight into the flame stabilisation phenomena. Further, analyses of quantities such as local equivalence ratio, cell Damköhler number, species velocity, net consumption rate of methane, which are not easily obtained through experiments even with detailed diagnostics, have been carried out. The influence of the flow field due to varying inlet velocity of the oxidiser, in the presence of the bluff-body, on flame anchoring location has been analysed in detail. Local equivalence ratio contours obtained from non-reacting flow calculations are seen to be quite useful in analysing the mixing process and in the prediction of flame anchoring locations when the flames are not separated. Cell Damköhler number has been calculated using cell size, species velocity of the fuel, which is a derived quantity, and the net reaction rate of the fuel. The flame zone, which is customarily inferred from the contours of temperature, CO and OH, is also shown to be predicted well by the contour line corresponding to a Damköhler number equal to unity. The net reaction rate of CH₄ and the net rates of two dominant reactions, which consume methane, show clearly the variation in the flame anchoring locations in these three cases. Further, the three-dimensionality of these flames are analysed by plotting the mean temperature contours in y–z planes. Finally, the unsteadiness in the separated flame case is analysed.     

Numerical investigation of the effects of polydisperse water sprays on extinction conditions of counterflow methane non-premixed flames

Water, sprayed in the form of tiny droplets, has emerged as a potential fire suppressant after the halon compounds such as trifluorobromomethane (CF₃Br, Halon 1301) were banned by the Montreal protocol. The size distribution of the water droplet plays a crucial role in the effectiveness of the water spray in fire suppression. A numerical investigation of the influence of size distribution of a polydisperse water spray on extinction of counterflow diffusion flames is presented in this paper. This study uses laminar finite rate model with reduced CHEMKIN chemistry for numerical simulations. The discrete phase, namely the water spray, is simulated using Lagrangian Discrete Phase Modelling approach. In this work, the polydispersity of water spray is taken into account in the numerical simulation by a suitable Rosin–Rammler distribution. Results obtained from numerical simulation are validated with the experimental results reported in the literature. This study demonstrates that the representation of the polydisperse spray by a monodisperse spray (with droplet diameter same as the SMD of the polydisperse spray) in numerical simulations is not always justified and it leads to deviation from the experimental results. The effects of number mean diameter and spread parameter on the efficacy of flame suppression are investigated for polydisperse sprays. A comprehensive comparison is done between the effectiveness of monodisperse and polydisperse water sprays. An optimum droplet diameter is obtained for monodisperse sprays for which the effectiveness of the spray is maximum. The effects of evaporation Damköhler number and Stokes number of water droplets on flame suppression have also been explained.     

An elementary model for the validation of flamelet approximations in non-premixed turbulent combustion

The fundamental soundness of three flamelet models for non-premixed turbulent combustion is examined on the basis of their performance in an idealized model problem that merges ideas from the laminar asymptotic theory for non-premixed flames and rigorous homogenization theory for the diffusion of a passive scalar. The overall flame configuration is stabilized by a mean gradient in the passive scalar: large Damköhler number asymptotics results are available for the laminar case to quantify the finite-rate effects that cause the flame to depart from its equilibrium state; the same results can also be used to incorporate higher-order corrections in the approximation of the reactive variables in terms of the passive scalar. The use of such flamelet approximations has been extended well beyond the laminar regime as they lie at the core of practical strategies to simulate non-premixed flames in the turbulent regime: the flamelet representation avoids the problem of turbulence closure for the reactive variables by replacing it by the presumably much simpler closure problem for a passive scalar. It is precisely the validity of this substitution outside the laminar regime that is addressed here in the idealized context of a class of small-scale periodic flows for which extensive rigorous results are available for the passive scalar statistics. Results for this simplified problem are reported here for significant wide ranges of Peclet and Damköhler numbers. Asymptotic convergence is observed in terms of the Damköhler number, with a convergence rate that is found to match the laminar predictions and appears relatively insensitive to the Peclet number. The passive scalar dissipation plays a key role in achieving higher-order corrections for the finite-rate case: replacing its pointwise value by an averaged value is convenient practically and can be rigorously motivated for the class of flows studied here, but while it does achieve an overall improvement over the lower-order equilibrium model, the simplification compromises the higher asymptotic convergence observed with the original finite-rate flamelet model with exact local dissipation.(Some figures in this article are in colour only in the electronic version; see www.iop.org)     

Structure and dynamics of edge flames in the near wake of unequal merging shear flows

We examine in this study the structure and dynamic properties of an edge flame formed in the near-wake of two initially separated shear flows, one containing fuel and the other oxidiser. A comprehensive study is carried out within the diffusive-thermal framework where the flow field, computed a-priori, is used for the determination of the combustion field. Our focus is on the effects of three controlling parameters: the Damköhler number controlling the overall flow rate, the oxidiser-to-fuel strain rate ratio of the supply streams that determines the extent of oxidiser entrainment towards the mixing zone, and the Lewis number, assumed equal for the fuel and oxidiser, that depends on the mixture composition. Response curves, representing the edge flame standoff distance as a function of Damköhler number, exhibit two distinct shapes: C-shaped and U-shaped curves characterising the response of low and high Lewis number flames, respectively. Stability considerations show that the upper solution branch of the C-shaped response curve is unstable and hence corresponds to physically unrealistic states, but due to heat conduction toward the cold plate the lower solution branch is always stable. The states forming this solution branch correspond to flame attachment, where the edge flame remains practically attached to the tip of the plate until it is blown off by the flow when the velocity exceeds a critical value. The U-shaped response, on the other hand, consists of equilibrium states that are globally stable. Thus, high Lewis number flames can be always stabilised near the splitter plate, with the edge held stationary or undergoing a back and forth motion, or lifted and stabilised downstream by the flow. Insight into the distinct stabilisation characteristics, exhibited by the different Lewis number cases, is given by examining the relationship between the local flow velocity and the edge propagation speed.     

Microgravity experiments of flame spreading along a fuel droplet array in fuel vapor-air mixture

Combustion experiments of fuel droplet array in fuel vapor-air mixture were performed at microgravities to investigate growth mechanism of group combustion of fuel droplets. A 10-droplet array was inserted into the test section filled with a saturated fuel vapor-air mixture as a simple model of prevaporized sprays. Gas equivalence ratio of the fuel vapor-air mixture was regulated by the test section temperature. n-Decane droplets of 0.8 mm in the initial diameter were suspended at the crossing points of 10 sets of X-shaped suspenders. The first droplet was ignited by a hot wire to initiate flame spread along a fuel droplet array. Flame spread speed was obtained from the history of the leading edge position of a spreading flame. Effects of droplet spacing and gas equivalence ratio on the flame spreading behavior and the flame spread speed were examined. The droplet spacing and the gas equivalence ratio were varied from 1.6 to 10.2 mm and from 0.2 to 0.7, respectively. The gas equivalence ratio has little effect on the relationship between the flame spreading behavior and the droplet spacing. The flame spread speed increases as the increase in the gas equivalence ratio at all droplet spacings. The influence of the gas equivalence ratio on the flame spread speed becomes strong as the increase in the droplet spacings. The flame spread speed increases as the increase in the droplet spacing, and then decreases. The maximum flame spread speed appears in the range from 2.4 to 3 mm at all gas equivalence ratios.     

Critical conditions at liftoff limit of a laminar n-butane-air jet diffusion flame

The stability mechanism of laminar coflow jet diffusion flames in normal gravity has been studied computationally and experimentally. N-butane, the heaviest alkane in a gaseous state at ambient temperature and pressure, is used as the fuel since the reaction mechanism is similar to that of higher (liquid) hydrocarbons. The critical mean n-butane jet and coflowing air velocities at flame stability limits are measured using a small fuel tube burner (0.8 mm inner diameter). The time-dependent, axisymmetric numerical code with a detailed reaction mechanism (58 species and 540 reactions), molecular diffusive transport, and a radiation model, reveals a flame structure. A fuel-lean peak reactivity spot (i.e., reaction kernel), possessing the hybrid nature of diffusion-premixed flame structure at a constant temperature of ≈1560 K, is formed at the flame base and controls the flame stability. In a near-quiescent environment, the flame base resides below the fuel tube exit plane and thereby premixing is limited. As the coflowing air velocity is increased incrementally under a fixed fuel jet velocity, the flame base moves slightly above (≈1 mm) the burner exit and vigorous premixed combustion becomes prevailing. The local heat-release rate at the reaction kernel nearly doubles due to the increased convective oxygen flux (i.e., a blowing effect). The local Damköhler number, newly defined as a ratio of the square root of the local heat-release rate and the local velocity, decreases gradually first and drops abruptly at a critical threshold value and the flame base lifts off from the burner rim. The calculated coflow air velocity at liftoff is ≈0.38 m/s at the fuel jet velocity of 2 m/s, which is consistent with an extrapolated measured value of 0.41 m/s. This work has determined the critical Damköhler number at the stability limit quantitatively, for the first time, for laminar jet diffusion flames.     

Identification of droplet burning modes in lean, partially prevaporized swirl-stabilized spray flames

Experimental and numerical investigations of single droplet burning modes in a lean, partially prevaporized swirl-stabilized spray flame are reported. In the experiment single droplet flames have been visualized by CH-PLIF and simultaneous recording of the Mie signal. Two single droplet burning modes were identified: the envelope flame is a spherical diffusion flame burning at near-stoichiometric conditions. The wake flame is a potentially lean, partially premixed flame located downstream of the droplet. The droplet burning mode is of practical relevance, since it has significant impact on NO formation due to incomplete prevaporization.The droplet burning mode is determined by the ratio of chemical and convective time scales. The convective time scale is related to the droplet slip velocity. The impact of turbulent gas phase velocity fluctuations on droplet mechanics and droplet burning is discussed, based on a previous numerical investigation. In the present study the droplet slip velocity was measured with the 3D Phase Doppler (3D-PD) technique. For the measured slip velocities and ambient conditions in the hot gas region of the spray flame, simulations of single droplet burning were performed utilizing detailed models for chemical reaction, diffusive transport and vaporization. An agreement between the droplet burning modes predicted by the simulation and the droplet burning modes observed in the experiments was found.     

Multiple speeds of flame edge propagation for Lewis numbers above one

Edges of diffusion flames in a counterflow burner are examined numerically for Lewis greater than unity. When the speed of propagation is plotted against Damköhler for a range of Lewis a fold bifurcation is observed. It is shown that there exist stable positively and negatively propagating edges for some Damköhler and Lewis number pairs. It is further shown that changed local conditions can lead to a transition from positive (advancing into the unburnt gasses) to negative (receding) propagation.     

A numerical study on extinction behaviour of laminar micro-diffusion flames

We conducted a numerical study on the fluid dynamic, thermal and chemical structures of laminar methane–air micro flames established under quiescent atmospheric conditions. The micro flame is defined as a flame on the order of one millimetre or less established at the exit of a vertically-aligned straight tube. The numerical model consists of convective–diffusive heat and mass transport with a one-step, irreversible, exothermic reaction with selected kinetics constants validated for near-extinction analyses. Calculations conducted under the burner rim temperature 300 K and the adiabatic burner wall showed that there is the minimum burner diameter for the micro flame to exist. The Damköhler number (the ratio of the diffusive transport time to the chemical time) was used to explain why a flame with a height of less than a few hundred microns is not able to exist under the adiabatic burner wall condition. We also conducted scaling analysis to explain the difference in extinction characteristics caused by different burner wall conditions. This study also discussed the difference in governing mechanisms between micro flames and microgravity flames, both of which exhibit similar spherical flame shape.     

Understanding cool flames and warm flames

Low-temperature flames such as cool flames, warm flames, double flames, and auto-ignition assisted flames play a critical role in the performance of advanced engines and fuel design. In this paper, an overview of the recent progresses in understanding low-temperature flames and dynamics as well as their impacts on combustion, advanced engines, and fuel development will be presented. Specifically, at first, a brief review of the history of cool flames is made. Then, the recent experimental studies and computational modeling of the flame structures, dynamics, and burning limits of non-premixed and premixed cool flames, warm flames, and double flames are presented. The flammability limit diagram and the temperature-dependent chain-branching reaction pathways, respectively, for hot, warm, and cool flames at elevated temperature and pressure will be discussed and analyzed. After that, the effect of low temperature auto-ignition of auto-igniting mixtures at high ignition Damköhler numbers at engine conditions on the propagation of cool flames, warm flames, and double flames as well as turbulent flames will be discussed. Finally, a new platform using low temperature flames for the development and validation of chemical kinetic models of alternative fuels will be presented. Discussions of future research of the dynamics and control of low temperature flames under engine conditions will be made.     

Diffusion flame-holes

Recent models of straight diffusion flame edges are extended to consider the effect of a curved edge forming the perimeter of an axisymmetric ‘hole’, where a burning flame surrounds a quenched inner region. For ‘free’ flame-holes (without a heat sink near the axis), at small Damköhler number (Da), the holes grow if the initial radius is large but shrink if it is small. For large Da, the holes shrink for any initial radius. Thus, free flame-holes are not stable for any Da, which is consistent with experimental observations. When the flame-hole is ‘anchored’ by a heat sink near to the axis, stationary holes of finite radius can exist for sufficiently high Da, but the solutions revert to ‘free’ hole behaviour for radii sufficiently larger than the heat sink radius. Based on these results, it is suggested that quasi-stationary flame-holes are not likely to be a common feature of turbulent diffusion flames, except possibly when large lateral gradients of Da are present due to intense vortices passing through the flame front.