The burning rate’s effect on the flame length of weak fire whirls

This paper discusses why the visibly-determined flame length of a weak fire whirl increases as compared with the corresponding pool fire without spin. Here, a fire whirl is called weak when the pure aerodynamic effect of flow circulation has a negligible influence on the flame length. Split cylinders were used to apply a flow circulation to a 3-cm-diameter methane burner flame and a 3-cm-diameter ethanol pool fire. After applying the flow circulation, the flame length of the ethanol pool fire increased about three times, while little change was observed in the flame length of the methane burner flame. The difference is explained by the fact that the burning rate of the methane burner flame was fixed constant, whereas that of the ethanol pool fire increased due to the increased heat input to the fuel surface caused by a change in flame shape pushed toward the fuel surface. The experimental observations thus demonstrate that the burning-rate effect can significantly increase the flame length even under a weak circulation condition. Computational fluid dynamics (CFD) simulations were conducted to understand the detailed flow structure of a fire whirl. An analytical model was then developed based on the experimental observations and CFD calculations; the predicted relationship between the flame height and the burning rate agreed with experimental data.     

Effect of flow circulation on combustion dynamics of fire whirl

In this paper, the effect of flow circulation on the combustion dynamics of fire whirl is systematically investigated by experiments. New correlations for the burning rate, flame height, radial temperature and mass flow rate are established for fire whirl. It is clarified that flow circulation helps increase both the fuel-flame contact area and the actual fuel surface area, which in turn increases both the heat feedback to the fuel surface and the radial velocity in the ground boundary layer, leading to increase of burning rate. A novel idea for correlation of fire whirl flame height is proposed by assuming that the ratio of the fire whirl flame height to the flame height without circulation solely characterizes the effect of circulation. This idea is fully verified, thereby a new formulation for flame height is established, which successfully decouples the burning rate and the circulation. It is indicated that the fuel-rich core in the flame body of fire whirl significantly affects the radial temperature distribution in the continuous flame region, and the flame body can be described by the combination of a cylinder and a cone. The flow circulation significantly suppresses fire plume radius and thus decreases its increasing rate with vertical distance. It is also demonstrated that the fire whirl flame involves laminarized regions in its lower section, coexisting with turbulent regions in the upper portion. The flow circulation enhances the air entrainment in the ground layer by altering the radial velocity profile and increasing the radial velocity. In the low section of flaming region, the significant decrease of mixture between the combustion products and surrounding air dominates the pure aerodynamic effect of flow circulation on the flame height. Finally, it is clarified that fire whirls maintain higher centerline excess temperature than general pool fires due to the effect of less air entrainment.     

Development of rotational CARS for combustion diagnostics using a polarization approach

Rotational Coherent anti-Stokes Raman spectroscopy (CARS) has the last decades been developed into a useful tool for thermometry and concentration measurements in combustion. In this paper, we present a novel polarization approach of the technique, which will enhance its potential and widen the range of conditions at which it can be utilized. The theory of the polarization approach is described in detail. It is shown that by specific arrangement of the polarizations of the laser beams, total suppression of the non-resonant background signal can be obtained, and thus by probing only the resonant CARS signal the diagnostic utility of the technique increases. The main benefit of the approach is in situations where the non-resonant background signal is relatively high in comparison with the resonant signal. The high potential of polarization rotational CARS for thermometry is demonstrated in some illustrative examples, for example, nitrogen thermometry on the fuel side of diffusion flames, and carbon monoxide thermometry in the product gas of ethylene/oxygen/argon-flames.     

Flame dynamics

This lecture describes recent theoretical developments associated with the dynamics of flames, obtained primarily by exploiting the various temporal and length scales involved in the combustion process. In premixed flames the focus is on flame-flow interactions that occur during the nonlinear development of hydrodynamically unstable large-scale flames, or during the propagation of curved flames in two-dimensional channels. The second part of the paper deals with non-premixed and partially premixed flames, where the focus is on understanding the nature of diffusive-thermal instabilities including the effect of thermal expansion, and on stabilization mechanisms of edge flames, which possess characteristics of both premixed and diffusion flames. The results presented in this talk illustrate how simplified models, when analyzed to their extreme, yield predictions of qualitative nature with physical insight that have advanced our understanding of combustion. This insight can be used to guide the experimental efforts, explain observations and validate large-scale numerical simulations.     

Determination of aluminum-air burning velocities using PIV and Laser sheet tomography

In a context of growing concerns over climate change, aluminum has the potential to serve as a dense energy carrier in order to replace fossil fuels and reduce greenhouse gases emissions. Indeed, its combustion in air may provide carbon-free energy for applications in which a high-energy storage capacity is required. However, attempts of designing a metal-fueled combustor will conflict with a relatively large dispersion of the burning velocity values reported in the literature, even when similar powders are used. This uncertainty is partially due to the range of experimental conditions and techniques used on those previous studies. In the present work, an experimental Bunsen-type aluminum-air burner is introduced. It is shown that the setup is capable of generating stable dust suspensions under well-controlled conditions. The stabilized aluminum-air flames are studied using emission spectroscopy, Particle Image Velocimetry, laser sheet tomography, and direct visualization of the AlO(g) emissions. The measured burning velocities are then compared to previous results obtained for similar powders as a function of dust concentration. A reasonable agreement is obtained, and it is shown that metal flame tomography can yield a more precise indicator of the flame front position than AlO(g) emissions, helping to reduce the data scatter regarding dust-air burning velocities.     

Burning velocity of triple flames with gentle concentration gradient

We investigated the local flame speed of a two-dimensional, methane-air triple flame in a rectangular burner. The velocity fields and the concentration profiles were measured with particle image velocimetry and the Rayleigh scattering method, respectively. There was a requisite combination of initial velocity and initial concentration gradient for consistency of the local concentration gradient at the leading edge of the flame. In these cases, the flame curvatures were also consistent. Accordingly, the burning velocity, defined as local flow velocity at the triple point, was determined by the flame curvature. The burning velocity increased with increasing flame curvature, when the curvature was near zero. After that, the burning velocity decreased with increasing curvature. The peak value thus exceeded the adiabatic one-dimensional laminar burning velocity. Comparing the effects of the measured flame stretch rate on the flow strain κ_s and flame curvature κ_c, κ_s is larger and increases more rapidly than κ_c for flame curvatures satisfying 1/R_f < 250 m⁻¹ and then becomes constant while κ_c still increases for 250 m⁻¹ < 1/R_f, so that κ_c becomes much larger than κ_s. There is also a peak in burning velocity at roughly the transition in flame curvature specified above. Therefore, the burning velocity for a low concentration gradient correlates with the flame stretch rate.     

An analytical study of droplet combustion under microgravity: Quasi-steady transient approach

An analytical model based on an assumption of combined quasi-steady and transient behavior of the process is presented to exemplify the unsteady, sphero-symmetric single droplet combustion under microgravity. The model used in the present study includes an alternative approach of describing the droplet combustion as a process where the diffusion of fuel vapor residing inside the region between the droplet surface and the flame interface experiences quasi-steadiness while the diffusion of oxidizer inside the region between the flame interface and the ambient surrounding experiences unsteadiness. The modeling approach especially focuses on predicting; the variations of droplet and flame diameters with burning time, the effect of vaporization enthalpy on burning behavior, the average burning rates and the effect of change in ambient oxygen concentration on flame structure. The modeling results are compared with a wide range of experimental data available in the literature. It is shown that this simplified quasi-steady transient approach towards droplet combustion yields behavior similar to the classical droplet theory.     

Rotational CARS for simultaneous measurements of temperature and concentrations of N2, O2, CO, and CO2 demonstrated in a CO/air diffusion flame

Rotational coherent anti-Stokes Raman spectroscopy (CARS) has over the years demonstrated its strong potential to measure temperature and relative concentrations of major species in combustion. A recent work is the development and experimental validation of a CO₂ model for thermometry, in addition to our previous rotational CARS models for other molecules. In the present work, additional calibration measurements for relative CO₂/N₂ concentrations have been made in the temperature range 294-1246 K in standardized CO₂/N₂ mixtures. Following these calibration measurements, rotational CARS measurements were performed in a laminar CO/air diffusion flame stabilized on a Wolfhard-Parker burner. High-quality spectra were recorded from the fuel-rich region to the surrounding hot air in a lateral cross section of the flame. The spectra were evaluated to obtain simultaneous profiles of temperature and concentrations of all major species; N₂, O₂, CO, and CO₂. The potential for rotational CARS as a multi-species detection technique is discussed in relation to corresponding strategies for vibrational CARS.     

Burning characteristics of candle flames in sub-atmospheric pressures: An experimental study and scaling analysis

Burning characteristics (mass burning rate, natural convection boundary layer thickness, flame height and dark zone height) of laminar diffusion flames produced by a candle at sub-atmospheric pressures in the range of P_∞?=?50–100?kPa were experimentally studied in a reduced-pressure chamber; such data are not reported to date. Scaling analysis was performed to interpret the pressure dependence. The new experimental findings for candle flames in the sub-atmospheric pressures were well interpreted by the proposed scaling laws: (1) the mass burning rate was higher for a candle with larger wick length, and it increased with increasing ambient pressure, a stagnant layer B-number model based on natural convection boundary (flame boundary layer thickness) was developed to scale the mass burning rate of candle flames at various pressures; (2) the flame boundary layer thickness was wider in lower pressure and can be well represented by a natural convection boundary layer solution; (3) flame height was higher for a candle with larger wick length, meanwhile the ratio of flame height to burning rate was independent of pressure; (4) the flame dark zone height representing a soot formation length scale changes little with pressure, meanwhile its ratio to the total flame height is scaled with pressure by $P_{\infty }^{?1/2}/L_{w,e}^{3/4}$ (L_w,e is effective wick length inside flame). This work provided new experimental data and scaling laws of candle flame behaviors in sub-atmospheric pressures, which provided information for future characterization and soot modeling for diffusion flames associated with melting and evaporation processes of solid fuels.     

A theoretical study of premixed turbulent flame development

Flame development in a statistically stationary and uniform, planar, one-dimensional turbulent flow is theoretically studied. A generalized balance equation for the mean combustion progress variable, which includes turbulent diffusion and pressure-driven transport terms, as well as the mean rate of product creation, is introduced and analyzed by invoking the sole assumption of a self-similar flame structure, well-supported by numerous experiments. The assumption offers the opportunity to simplify the problem by splitting the aforementioned partial differential equation into two ordinary differential equations, which separately model spatial variations of the progress variable and time variations of flame speed and thickness. The self-similar profile of the progress variable, obtained in numerous experiments, is theoretically predicted. Closures of the normalized pressure-driven transport term and mean rate of product creation are obtained. The closed balance equation shows that turbulent diffusion dominates during the initial stage of flame development, followed by the transition to counter-gradient transport in a sufficiently developed flame. A criterion of the transition is derived. The transition is promoted by the heat release and pressure-driven transport. Fully developed mean flame brush thickness and speed are shown to decrease when either density ratio or pressure-driven transport increases. Solutions for the development of the thickness are obtained. The development is accelerated by the pressure-driven transport and heat release.     

不同宽度和角度下木材表面火蔓延特性的实验研究

选取宽2～11 cm、厚3 mm的樟木、杉木和白木试样进行了一系列实验,测量了火蔓延速度、火焰高度、火焰温度和试样表面温度。结果表明,火焰高度随试样宽度增加而增大,并在宽度增加至某个值后不再增大;同宽度下,密度大的试样火焰高度大。放置角度为负时,不同密度试样的火蔓延速度差较小,角度为正时,火蔓延速度随角度增加而增大,且不同密度试样的火蔓延速度差变大.     

Further studies of the reaction kernel structure and stabilization of jet diffusion flames

The structure of axisymmetric laminar jet diffusion flames of ethane, ethylene, acetylene, and propane in quasi-quiescent air has been studied numerically in normal earth gravity (1g) and zero gravity (0g). The time-dependent full Navier–Stokes equations with buoyancy were solved using an implicit, third-order accurate numerical scheme, including a C₃-chemistry model and an optically thin-media radiation model for heat losses. Observations of the flames were also made at the NASA Glenn 2.2-Second Drop Tower. For all cases of the fuels and gravity levels investigated, a peak reactivity spot, i.e., reaction kernel, was formed in the flame base, thereby holding a trailing diffusion flame. The location of the reaction kernel with respect to the burner rim depended inversely on the reaction-kernel reactivity or velocity. In the C₂ and C₃ hydrocarbon flames, the H₂–O₂ chain reactions were important at the reaction kernel, yet the CH₃ + O → CH₂O + H reaction, a dominant contributor to the heat-release rate in methane flames studied previously, did not outweigh other exothermic reactions. Instead of the C₁-route oxidation pathway in methane flames, the C₂ and C₃ hydrocarbon fuels dehydrogenated on the fuel side and acetylene was a major hydrocarbon fragment burning at the reaction kernel. The reaction-kernel correlations between the reactivity (the heat-release or oxygen-consumption rate) and the velocity, obtained previously for methane, were developed further for various fuels in more universal forms using variables related to local Damköhler numbers and Peclet numbers.     

Behaviour of a confined fire located in an unventilated zone

The behaviour of a fire in an enclosure is studied for a configuration where the fuel source is located in the upper hot unventilated zone trapped by a soffit. The experimental study, undertaken in a laboratory-scale compartment with a fuel source above the level of a soffit, included the determination of the parameters (ventilation factor, rate of fuel supply) controlling the combustion or leading to extinction. Measurements (PIV, thermocouples, gas sampling and analysis) were performed to propose a hypothesis on the structure of the flame (flame stabilisation mechanisms, premixed or diffusion types). Video photography is used to determine the area covered by the flames. This information is used as a criterion to identify the combustion regimes. The results show that the gaseous fuel is diluted in the combustion products (CP) in the upper layer and that a recirculatory motion is formed, driven by buoyancy forces, which enhances the mixing of fuel and CP. These then travel horizontally towards the vent along the interface between the lower fresh air and upper zones, and are premixed with the convected air in the enclosure, before entering the reaction zone and being burnt. The flame stabilises at the interface between the upper hot and lower ventilated layers in the compartment. The observed “ghosting flame” is stabilised by a triple flame if the flame speed of the premixed flame is higher than the natural convection velocity induced in the compartment. The flame stability is quantified by a criterion based on the area of the horizontal flame. It has been observed that the combustion is controlled by the available mass fuel flux at the reaction zone if the ventilation is sufficient. This information is essential for the modelling of the phenomena involved in fires with such an underventilated fuel source.     

Modes of particle combustion in iron dust flames

The so-called argon/helium test is proposed to identify the combustion mode of particles in iron dust flames. Iron powders of different particle sizes varying from 3 to 34 μm were dispersed in simulated air compositions where nitrogen was replaced by argon and helium. Due to the independence of the particle burning rate on the oxygen diffusivity in the kinetic mode, the ratio between the flame speeds in helium and argon mixtures is expected to be smaller if the particle burning rate is controlled by reaction kinetics rather than oxygen diffusion. Experiments were performed in a reduced-gravity environment on a parabolic flight aircraft to prevent particle settling and buoyancy-driven disruption of the flame. Uniform suspensions of the iron powders were produced inside glass tubes and a flame was initiated at the open end of the tube. Quenching plate assemblies of various channel widths were installed inside the tube and pass or quench events were used to measure the quenching distance. Flame propagation was recorded by a high-speed digital camera and spectral measurements were used to determine the temperature of the condensed emitters in the flame. The measured flame speeds and quenching distances were in good agreement with previously developed one-dimensional, dust flame model where the particles are assumed to burn in a diffusive mode and heat losses are described on a volumetric basis. However, a significant drop of the ratio of flame speeds in helium and argon mixtures was observed for finer 3 μm particles and was attributed to a transition from the combustion controlled by diffusion for larger particles to kinetically controlled burning of micron-size particles. In helium mixtures, the lower flame temperatures measured in suspensions of fine particles in comparison to larger particles reinforces this assumption.     

Coupled fire dynamics and thermal response of complex building structures

Simulation of the effects of severe fires on the structural integrity of buildings requires a close coupling between the gas phase energy release and transport phenomena, and the stress analysis in the load-bearing materials. The connection between the two is established primarily through the interaction of the radiative heat transfer between the solid and gas phases with the conduction of heat through the structural elements. This process is made difficult in large, geometrically complex buildings by the wide disparity in length and time scales that must be accounted for in the simulations. A procedure for overcoming these difficulties used in the analysis of the collapse of the World Trade Center towers is presented. The large scale temperature and other thermophysical properties in the gas phase are predicted using the NIST Fire Dynamics Simulator. Heat transfer to subgrid scale structural elements is calculated using a simple radiative transport model that assumes the compartment is locally divided into a hot, soot laden upper layer and a cool relatively clear lower layer. The properties of the two layers are extracted from temporal averages of the results obtained from the Fire Dynamics Simulator. Explicit formulae for the heat flux are obtained as a function of temperature, hot layer depth, soot concentration, and orientation of each structural element. These formulae are used to generate realistic thermal boundary conditions for a coupled transient three-dimensional finite element code. This code is used to generate solutions for the heating of complex structural assemblies.     

Modelling of lifted turbulent diffusion flames in a channel mixing layer by the flame hole dynamics

The partial quenching structure of turbulent diffusion flames in a turbulent mixing layer is investigated by the method of flame hole dynamics as an effort to develop a prediction model for the turbulent flame lift off. The essence of the flame hole dynamics is derivation of the random walk mapping, from the flame-edge theory, which governs expansion or contraction of the quenching holes initially created by the local quenching events. The numerical simulation for the flame hole dynamics is carried out in two stages. First, a direct numerical simulation is performed for a constant-density fuel–air channel mixing layer to obtain the background turbulent flow and mixing fields, from which a time series of two-dimensional scalar-dissipation-rate array is extracted. Subsequently, a Lagrangian simulation of the flame hole random walk mapping, projected to the scalar dissipation rate array, yields a temporally evolving turbulent extinction process and its statistics on partial quenching characteristics. In particular, the probability of encountering the reacting state, while conditioned with the instantaneous scalar dissipation rate, is examined to reveal that the conditional probability has a sharp transition across the crossover scalar dissipation rate, at which the flame edge changes its direction of propagation. This statistical characteristic implies that the flame edge propagation instead of the local quenching event is the main mechanism controlling the partial quenching events in turbulent flames. In addition, the conditional probability can be approximated by a heavyside function across the crossover scalar dissipation rate.     

Experimental study of air dilution in oxy-liquid fuel flames

The influence of oxidizer dilution in oxy-liquid ethanol flames is experimentally investigated by using a coaxial air-assisted injector positioned in a vertical combustion chamber. This study accounts for the influence of a two-phase mode since two different injector geometries are used: for the first configuration, a vaporization mode is observed at nominal power in oxy conditions, while for the second one, a brush mode is observed. Dilution with air is applied by keeping oxidizer velocity constant. Flame structure is observed through CH emission: dilution leads to an increase in the flame diameter, and collective effects of two-phase combustion are encouraged. The effect of dilution on oxy flame stability is also studied: for a given oxygen mass fraction in the oxidizer, the oxidizer flow rate is increased until extinction occurs. Dilution leads to a less stable flame, which may be essentially explained by the decrease in laminar flame speed with dilution. For high oxidizer dilution levels, the change in flame structure might be another parameter to consider. Finally, species concentrations are measured using a standard gas sampling technique. NO and CO evolutions with dilution are different between both two-phase combustion regimes. An empirical approach based on thermal NO mechanism and CO oxidation reaction enables one to explain the evolutions for brush mode. For vaporization mode, the residence time in burned gases is also to be considered.     

Dynamic behavior of diffusion flame interacting with a large-scale vortex by laser imaging techniques

Instantaneous and simultaneous measurements of two-dimensional temperature and OH-LIF profiles by combining Rayleigh scattering with laser induced fluorescence (LIF) were demonstrated in a nitrogen-diluted hydrogen (H₂ 30% + N₂ 70%) laminar normal diffusion flame interacting with a large-scale vortex by oscillating central fuel flow or in an inverse diffusion flame by oscillating central airflow. The dynamic behavior of the diffusion flame extinction and reignition during the flame–vortex interaction processes was investigated. The results obtained are described as follows. (1) The width of the reaction zone decreases remarkably, and a decrease in flame temperature and OH-LIF is seen with increasing central airflow in an inverse diffusion flame. OH-LIF increases, and temperature does not change with increasing central fuel flow in a normal diffusion flame. The computations predict the experimental results well, and it is revealed that flame temperature characteristics result from the preferential diffusion of heat and species, which induces excess enthalpy or on enthalpy deficit, and an increase or decrease in H₂ mole fraction in the flame. (2) When a large velocity fluctuation is given to the central flow, the temperature and the OH-LIF at the reaction zone become thin at the convex and circumferential part of the vortex where a high temperature layer exists, and the temperature at the reaction zone is lowered in the inverse flame and the normal flame. (3) The width and temperature of the reaction zone interacting with the vortex recover quickly to that of the laminar steady flame after the vortex passing in the normal flame, but the recovery to that of the steady flame after the vortex passing is delayed in the inverse flame. (4) When a remarkably large velocity fluctuation is given to the central airflow in the inverse flame, thinning of temperature and reaction zone starts at the convex and circumferential part of the vortex, resulting in a and flame extinction completely occurs at the tail part of the vortex and makes the pair of edge flames. The outside edge flame reignites and connects with the upstream reaction zone. The inside edge flame finally extinguishes as the supply of fuel is interrupted by the outside edge flame.     

Binary collision of a burning droplet and a non-burning droplet of xylene: Outcome regimes and flame dynamics

The binary collisions of a burning droplet and a non-burning droplet of xylene are experimentally investigated. The experimental parameters span an extensive range of Weber number and impact parameter, covering the collision outcome regimes of coalescence, reflexive separation, and stretching separation. A high-speed camera captures the temporal details of the collision process, involving flame spread, visible radiation, and flame distributions around droplets. For reflexive separation and stretching separation, the flame from the droplet spreads to the ligament, surrounding it during the interaction process, and then spreads around separated droplets and satellite droplets. Highly-interactive flames are formed in-between the droplets, with very sooty flames generated for most collisions. For the coalescence case, a swirling flame forms around the rotating coalesced droplet. For similar Weber numbers, visible flame radiation is compared for different collision regimes. The visible flame radiation changes more significantly for the reflexive and stretching separation cases than it does for the coalescence case. The change of the averaged visible flame radiation for reflexive separation and stretching separation is more than two times higher than that for coalescence. The map of three different collision regimes is plotted in the Weber number versus impact parameter domain and compared with available theoretical model predictions. Although the different outcomes of collision with the presence of flame can be well predicted by the model, using fluid properties determined by the averaged properties of the two droplets, the dynamics of the detailed processes involved in the collisions are very interesting and have strong implications on overall combustion behavior that go well beyond the mapped regimes.