Gregory T. Linteris

Numerical study of the inhibition of premixed and diffusion flames by iron pentacarbonyl

Iron pentacarbonyl (Fe(CO){sub 5}) is an extremely efficient flame inhibitor, yet its inhibition mechanism has not been described. The flame-inhibition mechanism at Fe(CO){sub 5} in premixed and counterflow diffusion flames of methane, oxygen, and nitrogen is investigated. A gas-phase inhibition mechanism involving catalytic removal of H atoms by iron-containing species is presented. For premixed flames, numerical predictions of burning velocity are compared with experimental measurements at three equivalence ratios (0.9, 1.0, and 1.1) and three oxidizer compositions (0.20, 0.21, and 0.24 oxygen mole fraction in nitrogen). For counterflow diffusion flames, numerical predictions of extinction strain rate are compared with experimental results for addition of inhibitor to the air and fuel stream. The numerical predictions agree reasonably well with experimental measurements at low inhibitor mole fraction, but at higher Fe(CO){sub 5} mole fractions the simulations overpredict inhibition. The overprediction is suggested to be due to condensation of iron-containing compounds since calculated supersaturation is suggested to be due to condensation of iron-containing compounds since calculated supersaturation ratios for Fe and FeO are significantly higher than unity in some regions of the flames. The results lead to the conclusion that inhibition occurs primarily by homogeneous gas-phase chemistry.

Inhibition of Premixed Methane-Air Flames by Fluoromethanes

Abstract This paper presents the first calculations and measurements of the burning velocity of premixed hydrocarbon flames inhibited by the three one-carbon fluorinated species CH 2 F 2 , CF 3 H, and CF 4 . The chemistry of these agents is expected to be similar to that of some agents that may be used as replacements for CF 3 Br, so that studying their behavior in methane flames provides an important first step towards understanding the suppression mechanism of hydrocarbon fires by fluorinated compounds. The burning velocity of premixed methane-air flames stabilized on a Mache-Hebra nozzle burner is determined using the total area method from a schlieren image of the flame. The inhibitors are tested over a range of concentration and fuel-air equivalence ratio, φ. The measured burning velocity reduction caused by addition of the inhibitor is compared with that predicted by numerical solution of the species and energy conservation equations employing a detailed chemical kinetic mechanism recently developed at the National Institute of Standards and Technology (NIST). Even in this first test of the kinetic mechanism on inhibited hydrocarbon flames, the numerically predicted burning velocity reductions for methane-air flames with values of φ of 0.9, 1.0, and 1.1 and inhibitor mole fractions in the unburned gases up to 0.08, are in excellent agreement for CH 2 F 2 and CF 4 and within 35% for CF 3 H. The numerical results indicate that the agents CF 3 H and CH 2 F 2 are totally consumed in the flame and the burning velocity is reduced primarily by a reduction in the H-atom concentration through reactions leading to HF formation. In contrast, only about 10% of the CF 4 is consumed in the main reaction zone and it reduces the burning velocity primarily by lowering the final temperature of the burned gases.

Chemical limits to flame inhibition

Abstract This paper deals with the ultimate limits of chemical contributions to flame inhibition. Particular attention is focussed on the inhibition cycles which regenerate the inhibitor. This leads to the definition of an idealized “perfect” inhibition cycle. It is demonstrated that for such an inhibitor in a stoichiometric methane/air flame, additive levels in the 0.001–0.01 mole percent range will lead to a decrease in flame velocity of approximately 30%. This efficiency corresponds roughly to the observed behavior of metallic inhibitors such as iron pentacarbonyl which is known to be as much as 2 orders of magnitude more effective than currently used suppressants. This correspondence between the behavior of a “perfect inhibitor” and iron carbonyl leads to the conclusion that only gas-phase processes can account for its inhibitive power.

Flame Inhibition by Ferrocene and Blends of Inert and Catalytic Agents

The production of the fire suppressant CF3Br has been banned, and finding a replacement with all of its desirable properties is proving difficult. Iron pentacarbonyl has been found to be up to several orders of magnitude more effective than CF3Br, but it is flammable and highly toxic. Ferrocene [Fe(C5H5)2], which is much less toxic and flammable than Fe(CO)5, can also be used to introduce iron into a flame. We present the first experimental data and numerical modeling for flame inhibition by ferrocene and find it to behave similarly to Fe(CO)5. A ferrocene mole fraction of 200 ppm reduced the burning velocity of slightly preheated premixed methane/air flames by a factor of two, and the effectiveness dropped off sharply at higher mole fractions. For air with a higher oxygen mole fraction, the burning velocity reduction was less. We also present experimental data and modeling for flames with ferrocene blended with CO2 or CF3H. The combination of the thermally acting agent CO2 with ferrocene mitigated the loss of effectiveness experienced by ferrocene alone at higher mole fractions. An agent consisting of 1.5% ferrocene in 98.5% CO2 performed as effectively as CF3Br in achieving a 50% reduction in burning velocity. Likewise, four times less CO2 was required to achieve the 50% reduction if 0.35% ferrocene was added to the CO2. In contrast, addition of 0.35% ferrocene to the hydrofluorocarbon CF3H reduced the CF3H required to achieve the 50% reduction in burning velocity by only about 25%. Thermodynamic equilibrium calculations predict that the formation of iron/fluoride compounds can reduce the concentrations of the iron-species oxide and hydroxide intermediates which are believed to be responsible for the catalytic radical recombination cycles.

Structure and Soot Properties of Nonbuoyant Ethylene/Air Laminar Jet Diffusion Flames

The structure and soot properties of round, soot-emitting, nonbuoyant, laminar jet diffusion flames are described, based on long-duration (175-230-s) experiments at microgravity carried out on orbit in the Space Shuttle Columbia. Experimental conditions included ethylene-fueled flames burning in still air at nominal pressures of 50 and 100 kPa and an ambient temperature of 300 K with luminous flame lengths of 49-64 mm. Measurements included luminous flame shapes using color video imaging, soot concentration (volume fraction) distributions using deconvoluted laser extinction imaging, soot temperature distributions using deconvoluted multiline emission imaging, gas temperature distributions at fuel-lean (plume) conditions using thermocouple probes, soot structure distributions using thermophoretic sampling and analysis by transmission electron microscopy, and flame radiation using a radiometer. The present flames were larger, and emitted soot more readily, than comparable flames observed during ground-based microgravity experiments due to closer approach to steady conditions resulting from the longer test times and the reduced gravitational disturbances of the space-based experiments.

Inhibition of premixed methane-air flames by fluoroethanes and fluoropropanes

This paper presents experimental and modeling results for laminar premixed methane-air flames inhibited by the fluoroethanes C{sub 2}F{sub 6}, C{sub 2}HF{sub 5}, and C{sub 2}H{sub 2}F{sub 4}, and experimental results for the fluoropropanes C{sub 3}F{sub 8} and C{sub 3}HF{sub 7}. The modeling results are in good agreement with the measurements with respect to reproducing flame speeds. For the fluoroethanes, calculated flame structures are used to determine the reaction pathways for inhibitor decomposition and the mechanisms of inhibition, as well as to explain the enhanced soot formation observed for the inhibitors C{sub 2}HF{sub 5}, C{sub 2}H{sub 2}F{sub 4}, and C{sub 3}HF{sub 7}. The agents reduce the burning velocity of rich and stoichiometric flames primarily by raising the effective equivalence ratio and lowering the adiabatic flame temperature. For lean flames, the inhibition is primarily kinetic, since inhibitor reactions help to maintain the final temperature. The peak radical concentrations are reduced beyond that due to the temperature effect through reactions of fluorinated species with radicals.

EXTINCTION CONDITIONS OF NON-PREMIXED FLAMES WITH FINE DROPLETS OF WATER AND WATER/NaOH SOLUTIONS

Interactions of fine droplets of water and water/NaOH solutions with a steady, laminar counterflow methane/air nonpremixed flame are investigated experimentally and numerically. A water atomizer generating a polydisperse distribution of droplet sizes with a median diameter of 20 lm is used in experiments with steady feed rate. Comparisons of the measured flame extinction condition as a function of droplet mass fraction in the air stream indicate a trend similar to that predicted previously using 20 lm monodisperse water droplets. The hybrid Eulerian-Lagrangian numerical model previously developed is generalized to include polydisperse distribution of drop sizes; however, the differences seen between experiments and the numerical predictions at high water mass fractions could not be attributed to variation in size distribution alone. Present experiments support the conclusions of an earlier modeling work that on a mass basis, fine water mist can be as effective as the now-banned gaseous fire suppressant halon 1301. Inclusion of NaOH in water (up to 17.5% by mass) is shown to significantly enhance the fire suppression ability of water by complementing its thermal effects with chemical catalytic radical recombination effects of NaOH.

Inhibition of Premixed Methane Flames by Manganese and Tin Compounds

Abstract The first experimental measurements of the influence of manganese- and tin-containing compounds (MMT, TMT) on the burning velocity of methane/air flames are presented. Comparisons with Fe(CO) 5 and CF 3 Br demonstrate that manganese and tin-containing compounds are effective inhibitors. The inhibition efficiency of MMT is about a factor of two less than that of iron pentacarbonyl, and that of TMT is about 26 times less effective, although TMT is still about twice as effective as CF 3 Br. There exist conditions for which both MMT and TMT show a loss of effectiveness beyond that expected because of radical depletion, and the cause is believed to be particle formation. Kinetic models describing the inhibition mechanisms of manganese- and tin-containing compounds are suggested. Simulations of MMT- and TMT-inhibited flames show reasonable agreement with experimental data. The decomposition of the parent molecule for the tin and manganese species is found to have a small effect on the inhibition properties for the concentrations in this work. The inhibition effect of TMT is determined mostly by the rate of the association reaction H + SnO + M ↔ SnOH + M, and the catalytic recombination cycle is completed by the reactions SnOH + H ↔ SnO + H 2 and SnOH + OH ↔ SnO + H 2 O. The inhibition mechanism by manganese-containing compounds includes the reactions: MnO + H 2 O ↔ Mn(OH) 2 ; Mn(OH) 2 + H ↔ MnOH + H 2 O, and MnOH + OH (or H) ↔ MnO + H 2 O (or H 2 ), and the burning velocity is most sensitive to the rate of the reaction Mn(OH) 2 + H ↔ MnOH + H 2 O.

Inhibition of Premixed and Non-Premixed Flames With Fine Droplets of Water and Solutions

Inhibition/extinction of premixed and non-premixed methane/air flames with fine droplets of water and solutions containing several chemical agents has been investigated experimentally. While solutions allow delivery of much higher concentrations of chemical agent to the flame front than otherwise possible, the non-premixed flame extinction results indicate saturation (or condensation) of the agent at some effective temperature below the flame temperature. Based on the chemical additives considered, on a molar basis, the following order of effectiveness is observed: KOH>NaCl>NaOH. The inhibition of premixed flames by similar size droplets indicates insensitivity to NaOH mass fraction in the water. This insensitivity was related to the shorter residence time of the droplets (13 μm median diameter) through the premixed flame structure. Detailed comparison of the premixed and non-premixed flame inhibition/extinction with pure water droplets supports the importance of droplet residence time and optimum droplet size in controlling the interaction of droplets with the flame front.

The role of particles in the inhibition of counterflow diffusion flames by iron pentacarbonyl

Abstract Laser light scattering and thermophoretic sampling have been used to investigate particle formation in counterflow diffusion flames inhibited by iron pentacarbonyl Fe(CO) 5 . Three CH 4 -O 2 -N 2 reactant mixtures are investigated, with Fe(CO) 5 added to the fuel or the oxidizer stream in each. Flame calculations that incorporate only gas-phase chemistry are used to assist in interpretation of the experimental results. In flames with the inhibitor added on the flame side of the stagnation plane, the region of particle formation overlaps with the region of high H-atom concentration, and particle formation may interfere with the inhibition chemistry. When the inhibitor is added on the non-flame side of the stagnation plane, significant condensation of metal or metal oxide particles is found, and implies that particles prevent active inhibiting species from reaching the region of high radical concentration. As the inhibitor loading increases, the maximum scattering cross section increases sharply, and the difference between the measured and predicted inhibition effect widens, suggesting that particle formation is the cause of the deviation. Laser-based particle size measurements and thermophoretic sampling in low strain rate flames show that the particles have diameters between 10 nm and 30 nm. Thermophoresis affects the nanoparticle distribution in the flames, in some cases causing particles to cross the stagnation plane. The scattering magnitude in the counterflow diffusion flames appears to be strongly dependent on the residence time, and relatively independent of the peak flame temperature.