Showing papers in "Combustion and Flame in 1976"

Flammability of plastics—I. Burning intensity

Abstract: This paper presents the progress of an experimental study to develop a laboratory-scale test to define the burning intensity of various commercial samples of plastics. In the study, a steady state heat balance at the burning surface has been used to derive the burning intensities. An ‘ideal’ burning rate has been defined, which appears to correlate with full-scale fire test data. Conditions for the burning intensities in full-scale fires are postulated. Data for heat of gasification/pyrolysis/depolymerization, heat flux transferred from the flame to the surface, heat flux lost by the surface, and minimum mole fraction of oxygen required for flame extinction are presented for 16 commercial samples of plastics, a wood, and a plywood sample and six organic liquid samples.

Mathematical solutions for explosions in spherical vessels

Abstract: Equations, assumptions and previous solutions for central ignition of premixed gases in closed spherical vessels are reviewed. Three new categories of solution are presented: the Approximate Computer Solution (Case 1), the Dimensionless Universal Expression (Case 2), and the Complete Computer Solution (Case 3). Case 3 is the most accurate and shows the importance of accurate values of burning velocity as the unburnt gas is compressed in the course of the explosion. Recommendations are made concerning such values for methane-air mixtures. Case 2 solutions do not require a computer and they appear to be of good accuracy. During the explosions studied dimensionless pressure, P r , is proportional to dimensioness time, t , raised to the power 4.5.

The probability approach to the modelling of turbulent reacting flows

Abstract: The formulation of a closed set of averaged equations to describe a turbulent reacting flow is hindered by highly nonlinear relations between instantaneous quantities. This problem is overcome by considering the joint probability distribution of the scalars characterising the reaction. The transport equations for the single and joint probability distributions are derived and the unknown terms in the single probability distribution equation are modelled. Solutions of the modelled equation are obtained and the results are compared with previous models. The ability of the modelled equation to account for the influence of both turbulent mixing and finite chemical reaction rates is demonstrated.

Reactions of fuel-nitrogen in rich flame gases

Abstract: Fuel-nitrogen fed as ammonia or as pyridine to rich flames is mainly present in the burnt gas in the forms HCN, NH3, NO and N2. The HCN decays in this region to form more NH3, and the NH3(or species equilibrated with NH3)undergoes two types of reactions: an oxidation to NO at rate R1, reaction with NO to form N2 at rate R2. The total remaining fuel-N not yet converted to NO or N2, i.e. RN = HCN + NH3, therefore decays in accordance with -d[RN] dt =R1 + R2 , and the simultaneous change in NO is -d[NO] dt =R1 − R2 , Empirically, these rates can be epxressed R 1 5×10 12 [ H 2 O ] 2 [ NH 2 ] [ H 2 e −20 kcal /RT mole cc,sec , R 2 = 9×10 12 [ NO ] [ NH 2 . R1 is not readily interpreted in terms of elementary reactions; R2 is the rate of NO + NH2 → N2 + H2O. The findings are not necessarily valid in the primary reaction zone where other processes also occur. To the extent that all the fuel-N added to rich flames attains NO and N2 in accordance with R1 and R2, however, the yield of NO is predictable a priori. The prediction agrees roughly with the yields observed.

Unsteady droplet combustion with droplet heating

Abstract: Unsteady droplet combustion caused by droplet heating is modelled by assuming quasi-steady gas-phase processes and the droplet temperature being spatially uniform but temporally varying. Results show droplet heating is a significant source for the experimentally observed unsteady combustion phenomena of Okajima and Kumagai.

The quenching of deflagration waves

Abstract: A simple model of a low Mach number deflagration wave is studied, with emphasis on the effects of a phenomenological heat loss term. It is shown by explicit construction that two quite different solutions, a fast wave and a slow wave, are possible in the adiabatic limit. In both cases the chemical reaction goes to completion and all the fuel is consumed, but for the fast wave the temperature increases from the ambient value to the usual adiabatic flame temperature, whereas the slow wave is isothermal. An analysis for finite heat loss is carried out in the realistic limit of infinite activation energy. In general this leads to a much simplified numerical problem, but for values of the flame speed of greatest physical interest an analytical description is possible. This predicts quenching when the heat loss is too great or the reaction rate too small, and yields simple explicit quenching criteria.

Prediction of propagating methane-air flames

Abstract: The kinetics and propagation of laminar methane-air flames were studied using a one-dimensional, flame propagation model. The model is based on a numerical, unsteady-state solution of transformed species and energy conservation equations using explicit techniques for diffusion terms and linearized, implicit techniques for kinetic terms. A methaneoxygen kinetic mechanism consisting of 28 elementary reactions was postulated and used in the flame model. Flame velocity, flame thickness, temperature profile and concentration profiles of 13 species were predicted for a series of methane-air flames. The effects of pressure, methane concentration, initial temperature, rate constants, and transport coefficients were investigated. Many of the model predictions were compared with experimental data, and agreement was generally very good. The concentrations of the radicals H, OH, and O were major factors in the propagation of methane-air flames. The relative importance of each of the 28 reactions was examined; five were found to be negligible, while several were shown to be important in determination of propagation velocity and flame characteristics.

Multicomponent droplet combustion with rapid internal mixing

Abstract: Two models are proposed to describe the gas-phase diffusion-controlled, unsteady combustion of a multicomponent droplet in a stagnant, unbounded atmosphere. The first, termed the Ideal-Mixture Model, assumes that the mixture behaves as an ideal mixture in its phase change characteristics, and that the composition and temperature within the droplet are spatially uniform but temporally varying. Expressions are obtained for the droplet vaporization rate and other quantities of interest. Sample solutions indicate that the components vaporize approximately sequentially in the order of their relative volatilities, and that the vaporization rate is insensitive to the mixture composition during combustion as well as during pure vaporization in hot environments. Available experimental evidence supports the theoretical model. The second model, termed the Shell Model, assumes a shelled distribution of the components such that quasi-steady, single-component vaporization prevails for each shell. Simplified solutions are derived and are shown to closely approximate the bulk vaporization behavior described by the more detailed Ideal-Mixture Model, particularly for the prediction of the total vaporization time.

Asymptotic theory of diffusion-flame extinction in the stagnation-point boundary layer☆

Abstract: An analysis is developed for predicting extinction of the diffusion flame that is established when an oxidizing gas flows about the nose of a vaporizing fuel body. Use is made of the limit of a large ratio of the activation energy to the thermal energy at the flame for the overall combustion process, since this limit encompasses all cases of practical interest. By revealing a correspondence with the asymptotic flame structure of a counterflow diffusion flame analyzed earlier, the theory makes available explicit formulas, in term of a Damkohler number, for study of gas-phase extinction in the present geometry. From these results a simplified but reasonably accurate method is developed for obtaining, from experimental data on extinction, kinetic information concerning the overall oxidation process occurring in the vicinity of extinction. Curves calculated from a parametric study are presented to facilitate application of the technique, and the procedure is illustrated for methanol burning in oxygennitrogen mixtures.

The formation and destruction of hydrogen cyanide from atmospheric and fuel nitrogen in rich atmospheric-pressure flames

Abstract: A novel method involving direct sampling of natural flame ions into a mass spectrometer has been used to measure the concentration of HCN, NH 3 and NO in atmospheric-pressure flames It is found that HCN is formed from N 2 in the reaction zone of rich hydrocarbon flames and is subsequently completely oxidized within the flame, in part to NO This corresponds to Fenimore's ‘prompt NO’ The amount of HCN formed is proportional to the N 2 concentration, but is not very dependent on the nature of the fuel (CH 4 , C 2 H 4 and C 2 H 2 ) or on the temperature (between 2000 and 2560 K) It varies strongly with the equivalence ratio and is greater, the richer the flame In contrast to N 2 , nitrogen compounds in the fuel, whatever their nature (NO, NH 3 , CH 3 CN or pyridine), are converted quantitatively to HCN in the reaction zone of rich hydrocarbon flames The subsequent disappearance of HCN within the flame has been studied, and between 2300 K and 2560 K the rate is consistent with the reaction CN + OH → NCO + H being the rate-limiting step with a rate constant of approximately 1 × 10 −10 cm 3 molecule −1 s −1 , although other mechanisms may be important at lower temperatures

On the effective energy for direct initiation of gaseous detonations

Abstract: The present paper demonstrates that the effective and hence the true critical energy Ec for direct initiation of gaseous detonations using electrical sparks corresponds to the total energy deposited into the gas up to the time tf of the peak averaged power, i.e., (E(t) t) max , of the spark. The energy subsequent to this time is found to have no noticeable influence on the initiation processes. The method for demonstrating this experimentally is via the “crowbarred” discharge which was used to initiate cylindrically expanding detonation waves. Almost all previous investigations had implied that the direct initiation process can be characterized by a unique critical value of the source energy where the source energy was invariably taken as the total energy initially stored in the source or its equivalent. The present results indicate that the critical energy Ec is non unique but depends on its rate of deposition. It is found that Ec increases very rapidly with increasing time of energy deposition tf. Howerver, a minimum limiting value of the critical energy is found to exist as tf→0. The present results, in fact, suggest that the direct initiation process should be characterized by two parameters, namely, the peak power of the source and the energy release up to the peak power. The critical peak averaged power of the source, i.e., P c = E c (t f ) t f , also exhibits a minimum value which corresponds to shock strengths of the order of the autoignition limit for the explosive mixture.

Temperature and density in a hydrogenair flame from Rayleigh scattering

Abstract: A technique has been developed for obtaining spatially resolved measurements of temperature and density under combustion conditions. The spectral structure of Rayleigh scattered laser light is determined using a Fabry-Perot interferometer. Spectra in the post-combustion zone exhibit simple Doppler broadening with a Gaussian line profile; however, spectra in the higher density, precombustion zone have appreciable Brillouin components. Radial and axial surveys of temperature and intensity in the premixed hydrogenair flame issuing from a 3.2 mm diameter welding torch tip are obtained. Measured temperatures in the post-combustion zone are in agreement with the adiabatic flame temperature.

An investigation of the shock initiation of liquid nitromethane

Abstract: An experimental investigation of the shock initiation of liquid nitromethane is described. Shock pressures and temperatures in the unreacted liquid ranged from 7.5 to 9.5 GPa and 900 to 1100/sup 0/K. Diagnostic information was obtained from ferroelectric pins immersed in the liquid and from simultaneous streak camera and velocity interferometric records. The data support the thermal ignition model of one-dimensional shock initiation of nitromethane. The interferometer initiation time data are analyzed to obtain kinetic parameters in a first order overall rate law for the initiation-decomposition reaction in nitromethane. The rate constant for this overall reaction is suggested as k/sub 1/ = 2.6 X 10/sup 9/ exp(--23000/RT) sec/sup -1/, which is appreciably different from the rate constant for the unimolecular bond-scission reaction. The data support the conclusion that radical reactions significantly influence the overall initiation reaction in liquid nitromethane.

Near-limit flame spread over paper samples

Abstract: In this study the near-limit characteristics of a spreading flame are considered. Flame spreading rates and temperature profiles are measured as extinction conditions are approached. The flame is extinguished by increasing the heat loss, reducing the total pressure, or reducing the oxygen mole fraction in the environment. The gas phase temperature profiles are obtained with fine-wire thermocouple probes. The flame spreading results show that the power-law correlations of McAlevy and Magee [3] do not remain valid near the extinction limit. In all cases the slope of the Log (flame spread rate) vs. Log (total pressure) curves increase and approach vertical at extinction. Differences in vertical and horizontal flame spreading are discussed. The flame temperature profiles are examined for a near-limit flame, but the total pressure level is the only parameter changed. In the near-limit flame the maximum flame temperature is reduced slightly, but the flame is enlarged in physical size greatly. It is observed that near the pyrolysis front, heat transfer forward in the gas phase and normal to the fuel surface are of the same order of magnitude.

The effect of steam on flame temperature, burning velocity and carbon formation in hydrocarbon flames

Abstract: The effect on flame temmperature, burning velocity and carbon limit of adding water vapor to a premixed flame has been investigated using a Bunsen-type burner operated at atmospheric pressure and employing propane and ethylene as fuels. The results indicate that water vapor does not act as an inert diluent but instead inhibits carbon formation and gives rise to a greater heat release than in its absence. The additional heat release, in turn, partly counteracts the cooling effect of the added steam, so that the flame temperature and burning velocity do not decrease with steam addition to the extent they would if water vapor were an inert heat sink, provided the equilibrium state is attained. The shift of the carbon limit towards higher C O ratios with steam addition gives evidence of the major role of OH radicals in inhibiting carbon formation.

Minimum ignition energies in flowing kerosine-air mixtures

Abstract: Measurements of minimum ignition energy were carried out on kerosine sprays injected into a flowing air stream under conditions of atmospheric pressure and temperature. All the measurements were performed using optimum values of spark duration and with the spark gap adjusted to slightly exceed the quenching distance. Tests were carried out at air velocities up to 50 m/sec on five simplex swirl atomizers designed to produce fuel drops ranging in Sauter Mean Diameter (SMD) from 30 to 100 μm. Ignition was accomplished using capacitance sparks whose energy level could be varied in steps from 3 to 130 mJ. The results obtained clearly demonstrate that by far the most important factor influencing minimum energy is fuel drop size. Ignition performance is improved by increase in spark energy, but very large additional amounts of energy are needed to compensate for even a slight deterioration in atomization quality. It is shown that, for any given atomizer, an increase in velocity can actually improve the weak ignition limit, due to the beneficial effect of velocity on atomization quality. If, however, the fuel drop size is maintained constant, then increase in air velocity has a detrimental effect on the weak ignition limit, in agreement with results obtained on gaseous mixtures.

Radiative energy transfer from turbulent diffusion flames

Abstract: Transmittance and radiance measurements have been performed on turbulent ethane and propane flames by a flame array method that had been applied previously to arrays of laminar flames. The results showed that turbulent flames cannot be regarded simply as laminar flames of increased optical depth. Graphs of absorptance vs. radiance for the turbulent flames were shifted towards increased absorptance in comparison with the corresponding laminar-flame data. As one consequence of this shift, radiance values extrapolated to infinite number of flames, N x , were reduced by a factor of 0.71 for ethane and 0.79 for propane with respect to those for laminar flames. Moreover, at fuel flow rates of 120 cm 3 /s for each turbulent flame and of 4.5 cm 3 /s for each laminar flames, the transmittance of an array of n turbulent ethane flames equaled that of about 2.6 n laminar ethane flames, while their radiance equaled that of only about 1.4 n laminar flames. The corresponding ratios were 2.1 and 1.2, respectively, with propane flames. Measurements of total radiative power of single turbulent propane flames showed that within the fuel flow range of 70 to 316 cm 3 /s a constant fraction χ of the total rate of heat release is emitted as radiation. The value of χ was about .21 for several burner nozzle geometries and about .25 for a 20 mm i.d. straight-tube burner. Effects of fuel flow rate and of flame spacing on radiative properties have been explored, and power-law relationships linking fuel flow rate, flame dimensions and radiative properties are proposed.

On the concept of the critical size of a detonation kernel

Abstract: This paper introduces the concept of the detonation kernel in analogy to the concept of the flame kernel in ignition theory. The critical size of a detonation kernel is specified by the balance condition between the source energy density and the chemical energy density of the gas enclosed by the shock wave generated by the initiation source. It is found that the critical size of the detonation kernel is insensitive to the particular characteristics of the initiation source and thus depends on the chemical properties of the mixture itself. The critical detonation kernel size can be used to obtain the critical energy for direct initiation and the transverse wave spacing of a multiheaded detonation. The agreement with experimental data is excellent. The present paper provides a quantitative theory linking the characteristics of a detonation wave to the chemical properties of the explosive mixture via the concept of the detonation kernel.

Flame spread over an inclined thin fuel surface

Abstract: Downward flame spread velocity over a thin α-cellulose sheet was measured from the vertical to the horizontal positions under external radiant fluxes of 0, 0.9, 1.4 and 2.0W/cm 2 . The flame spread velocity had little dependency on the angle of inclination of the sheet with the stable lower flame. When the lower flame became unstable, wavy flames, cellular flames and flame rolls were observed below the lower surface. With these unstable lower flames, the flame spread velocity increased significantly reaching several cm/s. Qualitative analysis based on the Rayleigh instability mechanism describes well the effects of the inclination angle and the external flux on the instability of the lower flame.

Flame-quench distance measurements in a CFR engine

Abstract: A method has been devised which uses ion probe detection of flames burning through a gap inside the combustion chamber to provide rapid and convenient measurement of two-surface flame-quench distances in an operating engine. This method has been used to establish that such quench distances range from about 0.01 cm to > 0.2 cm, depending in decreasing order on equivalence ratio, manifold pressure, wall temperature, spark timing, and fuel type. A comparison between these experimental findings and the results from one- and two-surface measurements performed by others indicates that two-surface quench values are 2.5 to 12 times greater than one-surface values. The steep dependence on lean equivalence ratio of the measured two-surface quench values imitates to some extent the known behavior of lean hydrocarbon emissions (HCE), but more detailed single-surface quenching data is now needed to fully appreciate the mechanism involved in exhaust HCE.

Shock tube study of ignition delay times in methaneoxygennitrogenargon mixtures

Abstract: Ignition delay times have been measured in homogeneous CH/sub 4/-O/sub 2/ and CH/sub 4/-O/sub 2/-N/sub 2/ mixtures dilute in argon, in order to investigate the influence of molecular nitrogen. The mixtures were heated behind reflected waves in a conventional shock tube to temperatures between 1640 and 2150/sup 0/K at pressures from 1 to 6 atm. Equivalence ratios of 1.0 and 2.0 were investigated and the methane concentration ranged from 1 x 10/sup -7/ to 9 x 10/sup -7/ mole/cc. Ignition was identified from photometric observations of OH (3064 A), H/sub 2/O (2.7 ..mu..) and CO/sub 2/(4.3 ..mu..) emission. For both the CH/sub 4/-O/sub 2/ and CH/sub 4/-O/sub 2/-N/sub 2/ mixtures, the measured ignition delays (tau/sub i/) were correlated by the empirical expression tau/sub i/ = 4.4 x 10/sup -15/ exp (52,300/RT)(CH/sub 4/)/sup 0/./sup 33/(O/sub 2/)/sup -1/./sup 03/, in mole, cal, cm and sec units. It was concluded that molecular nitrogen has little or no influence upon the ignition delay times of methane-oxygen mixtures. This conclusion is supported by several earlier investigations of methane ignition but conflicts with the recent study by Zallen and Wittig.

Fire growth rates in wood cribs

Abstract: The burning history of a wood crib ignited at the center of its base has been investigated theoretically and experimentally. A simple energy-balance model has proven successful in predicting the radial fire-spread rates and mass burning rates for varying crib geometries with accuracies of ± 10%. Exceptions to the validity of the model were only noted for very densely packed cribs for which significant lateral spread occurred simultaneously with vertical fire spread. Cribs consisting of sticks with thicknesses of 0.635 cm, 1.905 cm and 3.17 cm were burned in the present experiments. Analysis of pressure modeling experiments has also shown that pressure modeling cannot, in general, model the fire growth rates in wood cribs.

Molecular beam mass spectrometry applied to determining the kinetics of reactions in flames II. A critique of rate coefficient determinations

Abstract: The microstructure of low-pressure methane-oxygen-argon flames was investigated using modulated molecular beam-mass spectrometry. Profiles of radical and stable species concentration, temperature, and area expansion ratio were used to calculate rate coefficients as a function of temperature for certain elementary reactions occurring in flames, namely, H + O/sub 2/ ..-->.. OH + O, H + CH/sub 4/ ..-->.. CH/sub 3/ + H/sub 2/, CO + OH ..-->.. CO/sub 2/ + H, CH/sub 3/ + O ..-->.. H/sub 2/CO + H, and H + CF/sub 3/Br ..-->.. HBr + CF/sub 3/. The profiles were modified (computationally) to simulate the effect of various perturbations and errors possible in sampling and analysis, and the effect on the rate coefficients is discussed. Detailed consideration is given to data reduction techniques, temperature profile-composition profile alignment, and the possible temperature dependence of mass spectral fragmentation. The rate coefficients are not dramatically sensitive to the imposed perturbations, although the results depend upon the nature of the reaction in question. Rate coefficients determined for high activation energy reactions and for reactions singularly responsible for the chemical behavior of a given stable species are in agreement with values determined by other techniques. Flame structure studies in which all significant radicalmore » and stable species are measured by a single technique are judged to be viable sources of high-temperature rate data for elementary reactions, where such reactions have been identified.« less

An aerothermochemical analysis of combustion of carbon in the stagnation flow

Abstract: An aerothermochemical analysis is made of the combustion of solid carbon exposed to the stagnation flow of oxidizing gas, and the effects of the surface Damkohler number and the oxygen concentration in the oxidizing flow on the combustion rate of solid carbon are examined. A rather high surface temperature of solid carbon (higher than 1500 °K) and a very high rate of CO oxidation in gas phase are postulated; the heterogeneous reaction is taken as 2 C + CO 2 →2 CO , and the homogeneous reaction as 2 CO + O 2 →2 CO 2 . The analysis is reduced to a problem of solving a nonlinear second-order ordinary differential equation of energy for various combinations of the gas-phase Damkohler number and the surface Damkohler number. The surface reaction is found to have chemically the same effect as the increase of the stoichiometric ratio of gas-phase reaction and gasdynamically the increase of the fuel injection rate, so that the combustion rate of solid carbon is generally expected to be small. For the gas-phase reaction of infinitely high rate, an explicit expression for the relation between the surface Damkohler number, the oxygen concentration, and the combustion rate can be derived.

Local temperature measurements in flames by laser raman spectroscopy

Abstract: The application of laser Raman scattering to temperature measurement in flames was studied using the post-flame region of a propane-oxygen-air flame. A schematic diagram of the equipment is presented and the experimental arrangement is discussed. The temperature distribution as a function of radius and distance from the burner was graphed. (DDA)

Instability of downward flame spread over paper in an air stream

Abstract: An experimental study has been conducted to elucidate the mechanism causing the instability of downward flame spread in an air stream. Following detailed observations of unstable flame spread phenomena, the flow fields near unstably spreading flames and the temperature profiles across the preheat zones were examined using particle tracer techniques and fine wire thermocouples. When the free-stream velocity was above that of the stable flame-spread limit, a series of local blow offs were observed during the spread. In this case, most leading flame edges were inclined or bended, and the blow offs occurred mainly at the locations where both inclined angle of the leading flame edge from the horizontal surface and curvature of the leading flame edge were small. Measured temperature-time diagrams at a given distance from the paper surface were almost similar if flames spread stably across the thermocouple junction. The flame spread phenomena including blow offs at the burning zone with a straight leading flame edge could be considered to depend mainly on the velocity component of the free-stream normal to the leading flame edge but scarcely on the velocity profile in the boundary layer of the approach flow.

Burning characteristics of a spherical particle reacting with ambient oxidizing gas at its surface

Abstract: An analytical model is presented to describe the quasi-steady burning and spontaneous extinction of a spherical particle in a quiescent oxidizing gas. In the special case of an exothermic Arrhenius first-order reaction of the oxidant with the surface of the particle, the model yields burning and extinction domains in terms of four dimensionless numbers. For a high activation energy of this reaction an approximation to the heat release rate is made, and this results in a transcendental algebraic equation having the four dimensionless numbers as the parameters. The analytical results are illustrated by choosing the burning of a carbon particle as an example.

Investigation of combustion phenomena associated with the flow of hot propellant gases. I. Spectroscopic temperature measurements inside the muzzle flash of a rifle

Abstract: The spatial and temporal temperature distribution in the muzzle blast field of a rifle of caliber 7.62 mm was determined by evaluation of the radiation emitted by the muzzle flash. In the primary flash directly adjacent to the muzzle a maximum gas temperature of 1645 K was observed by means of emission- absorption measurements of both continuum and discrete radiation. The influence of the surrounding cool layers was taken into account by applying Abel-inversion techniques. In the intermediate flash about 100 to 300 mm downstream line-reversal-techniques using K- and Na-resonance lines gave temperatures of about 1600 K increasing with time and with muzzle distance to a maximum temperature of 1940 K. This rise of temperature is due to the onset of combustion processes. Further downstream at a distance of 300 to 500 mm in the secondary flash a fast intense combustion of the propellant gases mixed with air leads to gas temperatures of about 2500 K. By firing into inert gases this flash could be suppressed whereas the primary and intermediate flash remained unchanged. When firing into oxygen a maximum gas temperature of 3000 K was observed in the secondary flash.