Stretched Flame Speed Research Articles

Modeling the boundary-layer flashback of premixed hydrogen-enriched swirling flames at high pressures

We model the boundary-layer flashback (BLF) of CH4/H2/air swirling flames via large-eddy simulations with the flame-surface-density method (LES-FSD), in particular, at high pressures. A local displacement speed model tabulating the stretched flame speed is employed to account for the thermo-diffusive effects, flame surface curvature, and heat loss in LES-FSD. The LES-FSD well captures the propagation characteristics during the BLF of swirling flames. In the LES-FSD for lean CH4/H2/air flames at 2.5 bar, the critical equivalence ratio for flashback decreases with the increasing hydrogen volume fraction, consistent with the experiments. This is due to the improved modeling of effects of the flame stretch and heat loss on the local displacement speed. We also develop a simple model to predict the BLF limits of swirling flames. The model estimates the critical bulk velocity for given reactants and swirl number, via the balance between the flame-induced pressure rise and adverse pressure for boundary-layer separation. We validate the model against 14 datasets of CH4/H2/air swirling flame experiments, with the hydrogen volume fractions in fuel from 50% to 100%. The present model well estimates the flashback limits in various operating conditions.

Effect of hydrogen addition on the consumption speed of lean premixed laminar methane flames exposed to combined strain and heat loss

This study presents a numerical analysis of the impact of hydrogen addition on the consumption speed of premixed lean methane-air laminar flames exposed to combined strain and heat loss. Equivalence ratios of 0.9, 0.7, and 0.5 with fuel mixture composition ranging from pure methane to pure hydrogen are considered to cover a wide range of conditions in the lean region. The 1-D asymmetric counter-flow premixed laminar flame (aCFPF) with heat loss on the product side is considered as a flamelet configuration that represents an elementary unit of a turbulent flame and the consumption speed is used to characterise the effect of strain and heat loss. Due to the ambiguity in the definition of the consumption speed of multi-component mixtures, two definitions are compared. The first definition is based on a weighted combination of the consumption rate of the fuel species and the second one is based in the global heat release rate. The definition of the consumption speed based on the heat release results in lower values of the stretched flame speed and even an opposite response to strain rate for some methane-hydrogen-air mixtures compared to the definition based on the fuel consumption. Strain rate leads to an increase of the flame speed for the lean methane-hydrogen mixtures, reaching a maximum value after which the flame speed decreases with strain rate. Heat loss decreases the stretched flame speed and leads to a sooner extinction of the flamelet due to combined strain and heat loss. Hydrogen addition and equivalence ratio significantly impact the maximum consumption speed and the flame response to combined strain rate and heat loss. The effect of hydrogen on the thermo-diffusive properties of the mixture, characterised by the Zeldovich number and the effective Lewis number, are also analyzed and related to the effect on the consumption speed. Two definitions of the Lewis number of the multi-component fuel mixture are evaluated against the results from the aCFPF.

A study of propagation of spherically expanding flames at low pressure using direct measurements and numerical simulations

The outwardly propagating spherical flame (OPF) method is widely used to measure the laminar burning velocity (LBV). However, there are still numerous challenges to perform unambiguous measurements for extreme conditions, mainly for high and low (sub-atmospheric) pressures. In this work, the density weighted displacement speeds relative to burned (Sd,b˜) and fresh gases (Sd,u˜) are considered to report LBV for a large range of sub-atmospheric pressure (from 1 to 0.3 bar), equivalence ratio and fuels (methane and n-decane). These stretched flame speeds are obtained from experiments by using advanced optical diagnostics and the spherical flame propagation is simulated by the code A-SURF with different kinetic schemes to perform a direct comparison with the experimental results. It is shown that Sd,b˜ is significantly affected by the pressure decrease and becomes strongly non-linear with flame stretch. This is explained by the finite flame thickness structure and qualitatively modeled by the finite thickness expression (FTE) model used for extrapolation at zero stretch. However, assumptions (equilibrium and static burned gases) made to report Sd,b˜ are no longer valid and leads to strong under-prediction of LBV from extrapolation especially at low pressures. The shape of Sd,u˜ is less impacted by the pressure decreases and a linear behavior with flame stretch is globally found even for strong sub-atmospheric conditions. However, the temperature of the fresh gas isotherm where Sd,u˜ is extracted is not rigorously equal to the fresh gas’s temperature, and a correction factor is required to get accurate LBV. Therefore, a direct comparison of Sd,u˜ instead of extrapolated LBV between experimental and numerical data seems necessary to validate kinetic schemes at low pressures. A brief discussion is conducted on the validation of FFCM-1 and GRI3.0 for methane and Dryer mechanisms for n-decane at low pressures.

Determination of spatially averaged consumption speed from spherical expanding flame: A new experimental methodology

Facilities using spherically expanding flame in closed vessels are generally used to report laminar burning velocities from extrapolated stretched flame speed. Considering the 1D planar reference case, flame displacement speed and consumption speed are rigorously identical. This is not true considering spherically expanding flames. While the flame displacement speeds are defined from a kinematic point of view, the consumption speed is linked to the integral of the reaction rate across the flame front. The latter is defined from a kinetic reference frame. Even though spherically expanding flames have been studied for decades, it is still challenging to experimentally determine the consumption speed. Recent developments define analytical expressions for the consumption speed and suggest experimental determination. One of the major issues consists in measuring the temporal evolution of the fresh gas density while the flame expands. In this work, an optical PIV-based method is developed to directly determine the fresh gas density ahead of the flame front and then the spatially averaged consumption speed. For the first time, experimental measurements of flame displacement speeds relative to fresh gases and burned gases, and spatially averaged flame consumption speeds are then reported for methane/air flames at different equivalent ratios. Experimental measurements are systematically compared with DNS which simulates the experimental geometry. A very good agreement is observed between experimental and numerical data irrespective to the equivalence ratio, underlying the robustness of the experimental methodologies. Finally, the sensitivity of the different flame speeds to flow confinement and extrapolation models is discussed.

Laminar burning velocities of C2H4/N2O flames: Experimental study and its chemical kinetics mechanism

Nitrous oxide fuel blend propellants have great potential to be used in rocket engine, and investigations on the fundamental combustion characteristics of such propellants are therefore necessary to control flashback and develop their chemical kinetics mechanism. Laminar burning velocities (LBVs) of C2H4/N2O flames are measured by using spherical expansion flames in this paper at 0.5–2.0 atm and 280 K. The present method for upper and lower limits of effective flame radius is reasonable for the experimental stretched flame speeds can be fitted very well with the flame stretch rates. The LBVs of C2H4/N2O flames are smaller than those of C2H4/N2/O2 flames (same N/O ratio as N2O) at conditions near stoichiometric ratio, while they are larger than those at other conditions especially at fuel-rich side. The LBVs of C2H4/N2O flames are not sensitive to the pressure in the measured range. Two kinds of sub-mechanisms are applied for detailed chemical kinetics mechanisms of C2H4/N2O reactions, which are USC Mech II-2 for hydrocarbon reactions as well as GRI 3.0 mechanism and San Diego mechanism for nitrogen oxide reactions respectively. Eight key elementary reactions are chosen based on the sensitivity analysis, and the effects of their available rate constants from literatures on the LBVs are tested. Modified mechanisms for C2H4/N2O reactions are therefore proposed by replacing the rate constants of these key elementary reactions, which predicts well for LBVs of hydrocarbon/N2O flames. Sensitivity analyses are performed for C2H4/N2O flames at different equivalence ratios, it is found that the N2O decomposition is a dominant reaction in conditions near stoichiometric ratio, while it is the codominant and nondominant reaction in fuel-lean and fuel-rich conditions respectively. Furthermore, the reaction pathways of oxidizers consumption in C2H4/N2O flames and C2H4/N2/O2 flames are analyzed, and the observations on the LBVs of these two flames in the experiment can be well explained through the reaction pathways and their relative changes under fuel-lean, stoichiometric and fuel-rich conditions.

Extrapolation and DNS-mapping in determining laminar flame speeds of syngas/air mixtures

The laminar flame speeds of atmospheric lean syngas/air flames were measured using the expanding spherical flame under isobaric conditions. Two methods, namely extrapolation and DNS-mapping, were utilized to determine the un-stretched laminar flame speed with quantified uncertainty from the experimentally measured instantaneous stretched flame speeds. Results from four extrapolation expressions show that the extrapolation uncertainty can exceed 5% for equivalence ratio of 0.5 and hydrogen mole fraction of 0.6. Two detailed chemical mechanisms, the Li-mech and the HP-mech, were used in the DNS-mapping. It is found that the uncertainty of DNS-mapping stays small, typically less than 2% for all the experimental conditions. The relative merits of extrapolation and DNS-mapping in determining the laminar flame speeds were then compared. Sensitivity analysis was also conducted and results show that H and OH radicals play crucial roles in the combustion chemistry of H2/CO flames. Furthermore, near linearity between the mass burning rate and the maximum mole fraction/concentration of the (H + OH) radicals was empirically observed for the equivalence ratios of 0.5–0.9, H2 content of 30–70% (by volume) at different pressures (1.0–50.0 atm) and initial temperatures (298–448 K).

Effects of endothermic chain-branching reaction on spherical flame initiation and propagation

A theoretical model is developed to describe the spherical flame initiation and propagation. It considers endothermic chain-branching reaction and exothermic recombination reaction. Based on this model, the effects of endothermic chain-branching reaction on spherical flame initiation and propagation are assessed. First, the analytical solutions for the distributions of fuel and radical mass fraction as well as temperature are obtained within the framework of large activation energy and quasi-steady assumption. Then, a correlation describing spherical flame initiation and propagation is derived. Based on this correlation, different factors affecting spherical flame propagation and initiation are examined. It is found that endothermicity of the chain-branching reaction suppresses radical accumulation at the flame front and thus reduces flame intensity. With the increase of endothermicity, the unstretched flame speed decreases while both flame ball radius and Markstein length increases. Endothermicity has a stronger effect on the stretched flame speed with larger fuel Lewis number. The Markstein length is found to increase monotonically with endothermicity. Furthermore, the endothermicity of the chain-branching reaction is shown to affect the transition among different flame regimes including ignition kernel, flame ball, propagating spherical flame, and planar flame. The critical ignition power radius increases with endothermicity, indicating that endothermicity inhibits the ignition process. The influence of endothermicity on ignition becomes relatively stronger at higher crossover temperature or higher fuel Lewis number. Moreover, one-dimensional transient simulations are conducted to validate the theoretical results. It is shown that the quasi-steady-state assumption used in theoretical analysis is reasonable and that the same conclusion on the effects of endothermic chain-branching reaction can be drawn from simulation and theoretical analysis.

Experimental study on the flame propagation behaviors of R245fa(C3H3F5)/CH4/O2/N2 mixtures in a constant volume combustion chamber

This experimental study was aimed at identifying the near-flammable limit and the unstretched flame speed of outwardly propagating spherical flames according to changes in the mole fraction of oxygen in the oxidizer (RO) and mole fraction of R245fa in the fuel (RR) for R245fa/CH4/O2/N2 mixtures in a constant volume combustion chamber. The flames observed according to changes in RO and RR of the R245fa mixtures could be divided into three regimes, with typical, transition, and buoyancy-induced behaviors. These representative flame behaviors were similar to those observed in previously reported experiments on R134a mixtures. The R245fa mixtures particularly showed a more expanded near-flammable limit than R134a. The rising velocity Ur and the rising length Lr in the buoyancy-induced regime with the near-flammable limit were used to identify the buoyancy effect on the flame that rose and propagated outward under normal gravity conditions. In addition, as the slope of the linear relationship between the stretch rates (k) and stretched flame speeds (Sd), the Markstein length (Lb) of the R245fa mixture tended to be positive in the typical regime and negative in the transition and buoyancy-induced regimes. Such characteristics are similar to those of R134a. However, in the typical regime, the results from a regression analysis on the unstretched flame speeds (Su) derived from the same linear relationship showed that the flame inhibition effect from mixing R245fa was smaller than that of R134a.

Laminar burning characteristics of 2-MTHF compared with ethanol and isooctane

2-Methyl tetrahydrofuran (2-MTHF) is one of the promising second-generation biofuel candidates, raising increasing interest because of the recent breakthrough in producing 2-MTHF from biomass. As a potential gasoline extender and renewable oxygenate, its high energy density and avoidance of competing with food make it attractive compared with ethanol. In the current work, the laminar burning characteristics of 2-MTHF-air mixtures are studied with varying equivalent ratios (0.88–1.43) and initial temperatures (333K, 363K and 393K) at ambient pressure in a constant-volume vessel. The results are compared with ethanol and isooctane, a representative component in gasoline. The stretched flame speed, un-stretched flame speed, Markstein length, flame thickness, Markstein number, laminar burning velocity and burning flux are calculated and analyzed. The result shows that for most tests, the ranking of un-stretched flame propagation speed is ethanol, 2-MTHF and isooctane. The peaks of un-stretched flame speeds against varying equivalent ratios appear at Φ=1.1–1.2 for three tested fuels. The Markstein length indicates that generally isooctane has the most diffusion-thermal stable flame than 2-MTHF and ethanol when Φ<1.2 at 393K. Ethanol shows the smallest flame thickness in most of the test points. The laminar burning velocity of 2-MTHF is much faster than isooctane and is comparative with ethanol, indicating its fast-burning property and potential of improving engine thermal efficiency.

Extrapolation of laminar flame speeds from stretched flames: Role of finite flame thickness

Recognizing the persistent inaccuracy of existing linear and nonlinear expressions to extract laminar flame speeds from experimental data of stretched flame speeds, and the essential and independent role of the flame thickness, in addition to the laminar flame speed, in describing the coupled diffusive and reactive response of the laminar flame structure and propagation, we have adopted the finite flame thickness expression of Frankel and Sivashinsky in the extrapolation. Using numerically generated stretched flame data of the outwardly expanding spherical flame for both lean and rich hydrogen/air and n-heptane/air mixtures, it is demonstrated that the finite thickness expression yields not only improved extraction of the laminar flame speed, it also collaterally yields the laminar flame thickness as well as the Markstein length. Additionally, we have examined the cause for the inaccuracy of the pervious linear and nonlinear extrapolation expressions, the effects of pressure on the extrapolation results, and the robustness of the various expressions in accounting for the experimental errors.

Self-acceleration of cellular flames and laminar flame speed of syngas/air mixtures at elevated pressures

Experimental study on the self-acceleration characteristics and laminar flame speed of CO/H2/air mixtures was conducted at elevated pressures up to 0.6 MPa with spherical outwardly expanding flames. Experimental conditions for the CO/H2/air mixtures of hydrogen fraction in syngas from 0.2 to 0.8, the pressures from 0.4 to 0.6 MPa and equivalence ratios from 0.5 to 1.0. At elevated pressures, the cellular structure occurs on the early stage of the flame development due to the significant influences of thermal diffusive and hydrodynamic instabilities and flame front was accelerated. Critical radius after which flame front becomes unstable decreases with H2 content in the fuel mixtures. For syngas mixtures with higher H2 content, critical radius increases with the increase of the equivalence ratio. Critical radius decreases with the increase of the equivalence ratio for the mixture with lower H2 content. Critical Peclet number, which is defined as the ratio of critical radius to flame thickness increases firstly and then decreases with H2 content for the mixtures with higher equivalence ratios and decreases all the time for the mixture with lower ones. In addition, the acceleration exponent which indicates the acceleration characteristics when flame front becomes unstable increases with the flame front propagation and is not the same with that of the turbulent flame. At last, an updated method, which excludes the acceleration effect of the cellular structure on the stretched flame speed at various flame radii has been proposed. It will help to obtain the laminar flame speed for fuel/air mixtures at elevated pressures with small time region between the end of the ignition spark and the onset of flame instability. This updated model replaces the experimental determined exponential term of the fractal structure and the updated intrinsic flame instability model. The measured laminar flame speed data are compared and analyzed with those predicted by Davis and Li mechanisms of syngas.

Experimental study on laminar flame characteristics of methane-PRF95 dual fuel under lean burn conditions

The effects of methane addition to PRF95 (primary reference fuel with 95% volume of iso-octane and 5% volume of n-heptane) on the fundamental combustion parameters are experimentally investigated in a cylindrical combustion vessel using classical schlieren technique. In this study, methane is added with three energy fractions of 25%, 50% and 75% to PRF95. The laminar flame propagation, Markstein length and flame instability of dual fuels under different initial pressures and with different equivalence ratios, especially under lean burn condition, are well studied. Spherical flames are experimentally investigated at the initial temperature of 373K and under the pressures of 2.5bar, 5bar and 10bar. The equivalence ratios vary with 0.8, 1.0 and 1.2. The stretched flame speeds are determined by outwardly spherical flame method. The results show that at low initial pressures, the addition of methane to PRF95 increases the stretched flame speeds with lean equivalence ratios while decreases it in rich region. Laminar flame of methane-PRF95 mixtures burn faster than those of pure methane and PRF95 with equivalence ratio of 0.8 over the whole range of the initial pressures investigated, and this trend is more obvious at low pressure. Comparing the data of 25% methane dual fuel (DF25) with that of base fuels with the equivalence ratio of 0.8 and under the initial pressure of 2.5bar, it can be seen that the flame speed of DF25 is 57% faster than that of methane and 22% faster than that of PRF95. These results provide important theoretical references to lean burn SI engine with methane-gasoline dual fuels under lean burn conditions.

Laminar burning characteristics of ethyl propionate, ethyl butyrate, ethyl acetate, gasoline and ethanol fuels

Ethyl esters have been considered as promising second-generation biofuel candidates, due to the available production from low grade biomass waste. Furthermore, with desirable energy densities, emissions performance, low solubility and higher Research Octane Number (RON), ethyl esters have proven attractive as fuel additives or alternatives for gasoline. In this study, high-speed schlieren photography was used to investigate the laminar burning characteristics of three ethyl ester fuels: ethyl acetate, ethyl propionate, and ethyl butyrate, in comparison with gasoline and ethanol at different initial temperatures and a variety of equivalence ratios, with an initial pressure of 0.1MPa in a constant-volume vessel. For the five fuels, the stretched flame speeds, the un-stretched flame speeds, Markstein lengths, Markstein number, laminar burning velocities and laminar burning flux were calculated and analysed using the outwardly spherical flame method. The results show that for all examined initial temperatures (60°C, 90°C and 120°C) and equivalence ratios; ethanol had the highest un-stretched flame propagation speeds, whilst ethyl acetate (EA) had the lowest. At high initial temperatures (120°C), it was observed that the un-stretched flame speed trends of ethyl propionate (EP) and ethyl butyrate (EB) proved faster compared to gasoline, especially for rich conditions. The EB and EA flames demonstrated greater stability when compared to ethanol, EP, and gasoline. Analysis showed that ethanol yielded the fastest flame velocities, whilst EA consistently had the lowest among all the five fuels. The laminar burning velocities of the EP fuel were faster compared to EB and EA, whilst slower than ethanol and gasoline at 60°C. Further increase of the initial temperature, up to 120°C, showed the laminar flame speed of EP and EB to be faster than gasoline, indicating a fast-burning property, and potential of improving engine thermal efficiency.

Laminar flame speeds of lean high-hydrogen syngas at normal and elevated pressures

The laminar flame speed is one of the most important combustion properties of a combustible mixture. It is an important target for chemical mechanism validation and development, especially at fuel-lean and high pressure conditions. In this study, the laminar flame speeds of two types of lean high-hydrogen syngas/oxygen/helium mixtures were measured at normal and evaluated pressures up to 10atm using a dual-chambered high pressure combustion facility. Similar to experiments, numerical simulations of outwardly spherical flame propagation were conducted. Three chemical mechanisms for syngas available in the literature were considered in simulation and their performance in terms of predicting the stretched flame speeds, laminar flame speeds and burned Markstein lengths was examined through comparison between experimental and simulation results. It was found that at both normal and elevated pressures, the present experimental results agree well with those predicted by simulations using these three chemical mechanisms. Therefore, these chemical mechanisms for syngas can well predict the laminar flame properties of lean high-hydrogen syngas. Besides, the laminar flame speeds measured in the present work were compared with those measured from the heat flux method and large difference was observed.

An Experimental Measurement on Laminar Burning Velocities and Markstein Length of Iso-Butane-Air Mixtures at Ambient Conditions

In the present work, experimental investigation on laminar combustion of iso-butane-air mixtures was conducted in constant volume explosion vessel. The experiments were conducted at wide range of equivalence ratios ranging between Ф = 0.6 and 1.4 and atmospheric pressure of 0.1 MPa and ambient temperature of 303K. Using spherically expanding flame method, flame parameters including stretched, unstretched flame propagation speeds, laminar burning velocities and Markstein length were calculated. For laminar burning velocities the method of error bars of 95% confidence level was applied. In addition, values of Markstein lengths were measured in wide range of equivalence ratios to study the influence of stretch rate on flame instability and burning velocity. It was found that the stretched flame speed and laminar burning velocities increased with equivalence ratios and the peak value was obtained at equivalence ratio of Ф = 1.1. The Markstein length decreased with the increases in equivalence ratios, which indicates that the diffusion thermal flame instability increased at high equivalence ratios in richer mixture side. However, the total deviations in the laminar burning velocities have discrepancies of 1.2-2.9% for all investigated mixtures.

Maximum stretched flame speeds of laminar premixed counter-flow flames at variable Lewis number

This paper examines Lewis-number effects on stretched, laminar, premixed flames near extinction. It presents the experimental measurement of maximum stretched flame speed and extinction limit for the premixed laminar combustion of selected low, unity and high-Lewis number mixtures. Stretched, fuel-lean, laminar flames of methane with Le≅1, propane with Le>1 and hydrogen with Le≪1 are studied experimentally in a counter-flow flame configuration. Flow velocity is measured in these flames by particle tracking velocimetry. Results show that a maximum reference flame speed exists for mixtures with Le≳1 at lower flame-stretch values than the extinction stretch rate. In contrast, a continually-increasing reference flame speed is measured for Le≪1 mixtures until extinction occurs when the flame is constrained by the stagnation point. Laminar flame results are also compared to numerical simulations employing a one-dimensional stagnation flame model. The chemical-kinetic models for each respective mixture capture the important trends of a maximum in su,ref ahead of the extinction stretch rate for methane and propane, and an increasing su,ref to extinction for hydrogen. These results are important to the investigation of the leading edge theory of premixed turbulent combustion, in which maximum stretched flamelet speed is a key parameter.

Effects of Soret diffusion on spherical flame initiation and propagation

Dynamics of spherical flame initiation and propagation with Soret diffusion are investigated using large-activation-energy asymptotic analysis. Under the assumptions of constant density and quasi-steady flame propagation, a general correlation between the flame propagating speed and flame radius considering Soret diffusion and external energy deposition is derived. Emphasis is placed on assessing the effects of Soret diffusion on spherical flame propagation speed, Markstein length, and critical ignition condition. The stretched flame speed is found to be increased and reduced by the Soret diffusion of light and heavy fuels, respectively. For both light and heavy fuels, the absolute value of Markstein length increases after including Soret diffusion, indicating that premixed flames become more sensitive to stretch rate with Soret diffusion. It is found that the Markstein length can be characterized by an effective Lewis number which includes the effects of Soret diffusion. Moreover, Soret diffusion is shown to affect the ignition process since the spherical flame kernel is highly stretched. For large hydrocarbon fuels with high Lewis numbers, the minimum ignition power becomes much larger after considering Soret diffusion.

Comparative study on the laminar flame speed enhancement of methane with ethane and ethylene addition

A systematic analysis of the flame speed enhancement effects of ethane and ethylene addition to methane was performed. Flame speeds of outwardly propagating spherical flames were measured in a high-pressure, cylindrical flame speed vessel using schlieren photography. It was ensured that the measured data were outside the ignition- and chamber-confinement-affected radii. The unstretched, unburnt flame speeds were obtained using nonlinear extrapolation methods from the experimentally determined stretched flame speeds. The AramcoMech 1.3 kinetics mechanism was validated against flame speeds of ethane, ethylene, and acetylene over a wide range of pressures and equivalence ratios. Good agreement between the measured data and those reported in the literature was seen. The model agreed closely with the measurements at fuel-lean conditions, but a slight over prediction was observed for the fuel-rich cases. Flame speeds of 80/20 and 60/40 (by volume) CH4/C2Hx mixtures in air and in air with excess nitrogen were measured over a wide range of fuel concentrations. In all cases, the addition of ethylene increased the flame speed more than did an equivalent addition of ethane. The Arrhenius effect (predominantly kinetic effect) was the principal driver for flame speed enhancement of methane with ethane/ethylene addition. Sensitivity analysis revealed that thermal and diffusive pathways were equally contributing to the flame speed enhancement of CH4/C2H6 mixtures, but the latter played a comparatively minor role in the C2H4-based blends.

Uncertainty in stretch extrapolation of laminar flame speed from expanding spherical flames

The present work investigated the uncertainties associated with the extrapolation of stretched flames to zero stretch in flame speed measurements using expanding spherical flames. Direct numerical simulations of time evolution of expanding spherical flames from a small ignition kernel to a propagating front with sufficiently large radius provide the relations between stretched flame speed and stretch rate that can be used to assess the uncertainty of extrapolation models. It is found that the uncertainties of flame extrapolation largely depend on the mixture Lewis numbers. While the uncertainty is minimized for stoichiometric H2/air and n-heptane/air flames, the uncertainty can be as high as 60% for lean H2/air mixtures, and 10% for lean and rich n-heptane/air mixtures. The present findings show that the weakly stretched flame assumption fails for lean hydrogen mixtures, and give a good explanation to the discrepancies between experiments and model predictions for H2/air as well as the discrepancies between measurements of n-heptane/air using spherical and counterflow flames. A relation between extrapolation uncertainties and the product of Markstein number and Karlovitz number is provided, which can be useful for uncertainty quantification of future and existing measurements.

Flame acceleration in unconfined hydrogen/air deflagrations using infrared photography

Flame behavior and blast waves generated during unconfined hydrogen deflagrations were experimentally studied using infrared photography. Infrared photography enables expanding spherical flame behaviors to be measured and flame acceleration exponents to be evaluated. In the present experiments, hydrogen/air mixtures of various concentrations were filled in a plastic tent of thin vinyl sheet of 1 m3 and ignited by an electric spark. The onset of accelerative dynamics on the flame propagation was analyzed by the time histories of the flame radius and the stretched flame speed. The results demonstrated that the self-acceleration of the flame, which was caused by diffusional-thermal and hydrodynamic instabilities of the blast wave, was influenced by hydrogen deflagrations in unconfined areas. In particular, it was demonstrated that the overpressure rapidly increased with time. The burning velocity acceleration was greatly enhanced with spontaneous-turbulization.