Daniel L. Dietrich

Droplet combustion experiments in spacelab

Individual droplets with diameters ranging from about 2 mm to 5 mm were burned under microgravity conditions in air at 1 bar with an ambient temperature of 300 K. Each droplet was tethered by a silicon carbide fiber of 80 μm or 150 μm diameter to keep it in view of video recording, and, in some tests, a forced air flow was applied in a direction parallel to the fiber axis. Methanol, two methanol-water mixtures, two methanol-dodecanol mixtures, and two heptane-hexadecane mixtures were the fuels. Droplet diameters were measured as functions of time, and they are compared here with existing theoretical predictions. The prediction that methanol droplets extinguish at diameters that increase with increasing initial droplet diameter is verified by these experiments. In addition, the quasi-steady burning-rate constant of the heptane-hexadecane mixtures appears to decrease with increasing droplet diameter; obscuration consistent with very heavy sooting, but without the formation of soot shells, is observed for the largest of these droplets. Forced convective flow around methanol droplets was found to increase the burning rate and to produce a ratio of downstream to upstream flame radius that remained constant as the droplet size decreased, a trend in agreement with earlier results obtained at higher convective velocities for smaller droplets having larger flame standoff ratios. Implications of the experimental results regarding droplet-combustion theory are discussed.

Inverse influence of initial diameter on droplet burning rate in cold and hot ambiences: a thermal action of flame in balance with heat loss

Abstract Isolated droplet burning were conducted in microgravity ambiences of different temperatures to test the initial diameter influence on droplet burning rate that shows a flame scale effect and represents an overall thermal action of flame in balance with heat loss. The coldest ambience examined was room air, which utilized a heater wire to ignite the droplet. All other ambiences hotter than 633 K were acquired through an electrically heated air chamber in a stainless steel can. An inverse influence of initial droplet diameter on burning rate was demonstrated for the cold and hot ambiences. That is, the burning rate respectively decreased and increased in the former and latter cases with raising the initial droplet diameter. The reversion between the two influences appeared gradual. In the hot ambiences the burning rate increase with increasing the initial droplet diameter was larger at higher temperatures. A “net heat” of flame that denotes the difference between “heat gain” by the droplet and “heat loss” to the flame surrounding was suggested responsible for the results. In low-temperature ambiences there is a negative net heat, and it turns gradually positive as the ambience temperature gets higher and the heat loss becomes less. Relating to luminous flame sizes and soot generation of differently sized droplets clarified that the flame radiation, both non-luminous and luminous, is determinative to the net heat in microgravity conditions. In addition, the work identified two peak values of soot generation during burning, which appeared respectively at the room temperature and at about 1000 K. The increase in ambience temperature made also bigger soot shells. The heat contribution of flame by both radiation and conduction was demonstrated hardly over 40% in the total heat required for droplet vaporization during burning in a hot ambience of 773 K.

Pressure effects on combustion of methanol and methanol/dodecanol single droplets and droplet pairs in microgravity

Abstract This paper presents the results of an experimental investigation on the combustion of single droplets and two-droplet arrays of pure methanol and methanol/dodecanol mixtures in air under microgravity conditions. The initial droplet diameters, d 0 , were nominally 0.9 mm. The independent experimental variables were the ambient pressure (0.1–9.0 MPa), fuel mixture ratio (methanol/dodecanol: 100/0–15/85), and interdroplet separation distance l ( l / d 0 = 2.3–8.0). For pure methanol, the results show that the droplet lifetime decreases with increasing interdroplet separation distances at low pressures. At higher pressures (3.0 MPa and above) the droplet lifetime was independent of separation distance. The flame extinguished at a finite droplet size only for pure methanol at 0.1 MPa, in qualitative agreement with theoretical predictions. The extinction droplet diameter was nearly independent of the droplet spacing. Methanol/dodecanol–mixture droplets exhibited microexplosion for both single droplets and droplet arrays. The paper presents maps of the disruption regime for both single droplets and droplet pairs. The difference between the disruptive behavior of single droplets and droplet pairs is explained by differences in liquid-phase circulation induced by the gas-phase asymmetry of the droplet pair. The paper also presents results of the dependence of the onset of disruption (in terms of both volume and time) on the pressure and initial fuel mixture ratio.

Candle Flames in Non-Buoyant Atmospheres

This paper addresses the behavior of a candle flame in a long-duration, quiescent microgravity environment both on the space Shuttle and the Mir Orbiting Station. On the Shuttle, the flames became dim blue after an initial transient where there was significant yellow (presumably soot) in the flame. The flame lifetimes were typically less than 60 seconds. The safety-mandated candlebox that contained the candle flame inhibited oxygen transport to the flame and thus limited the flame lifetime. The flames on the Mir were similar, except that the yellow luminosity persisted longer into the flame lifetime because of a higher initial oxygen concentration. The Mir flames bumed for as long as 45 minutes. The difference in the flame lifetime between the Shuttle and Mir flames was primarily the redesigned candlebox that did not inhibit oxygen transport to the flame. In both environments, the flame intensity and the height-to-width ratio gradually decreased as the ambient oxygen content in the sealed chamber slowly decreased. Both sets of experiments showed spontaneous, axisymmetric flame oscillations just prior to extinction. The paper also presents a numerical model of a candle flame. The formulation is two-dimensional and time-dependent in the gas phase with constant specific heats, thermal conductivity and Lewis number (although different species can have different Lewis numbers), one-step finite-rate kinetics, and gas-phase radiative losses from CO2 and H2O. The treatment of the liquid/wick phase assumes that the fuel evaporates from a constant diameter sphere connected to an inert cone. The model predicts a steady flame with a shape and size quantitatively similar to the Shuttle and Mir flames. The computation predicts that the flame size will increase slightly with increasing ambient oxygen mole fraction. The model also predicts pre-extinction flame oscillations if the rate of decrease in ambient oxygen is small enough, such as that which would occur for a flame burning in a sealed ambient.

Combustion of Miscible Binary-Fuel Droplets at High Pressure Under Microgravity

Abstract The objective of this research is to study near-critical and super-critical combustion of droplets consisting of binary fuel mixtures. Experimental results are reported on the burning of fiber-supported droplets of mixtures of n-heptane and n-hexadecane, initially about 1 mm in diameter, under free-fall microgravity conditions. The ambient pressures range up to 3.0 MPa, extending above the critical pressure of both fuels, in room-temperature nitrogen-oxygen atmospheres having oxygen mole fractions of 0.12 and 0.13. Three-stage burning of the binary fuel droplets is observed, and the onset lime of the second stage is compared with the predictions of an existing theory. Experimental evidence of thermo-capillary and/or diffuso-capillary convection during the droplet burning is obtained. The results contribute to improving understanding of binary-fuel droplet-combustion processes at high pressures.

Single droplet combustion of decane in microgravity: experiments and numerical modelling

This paper presents experimental data on single droplet combustion of decane in microgravity and compares the results to a numerical model. The primary independent experiment variables are the ambient pressure and oxygen mole fraction, pressure, droplet size (over a relatively small range) and ignition energy. The droplet history (D2 history) is non-linear with the burning rate constant increasing throughout the test. The average burning rate constant, consistent with classical theory, increased with increasing ambient oxygen mole fraction and was nearly independent of pressure, initial droplet size and ignition energy. The flame typically increased in size initially, and then decreased in size, in response to the shrinking droplet. The flame standoff increased linearly for the majority of the droplet lifetime. The flame surrounding the droplet extinguished at a finite droplet size at lower ambient pressures and an oxygen mole fraction of 0.15. The extinction droplet size increased with decreasing pressure. The model is transient and assumes spherical symmetry, constant thermo-physical properties (specific heat, thermal conductivity and species Lewis number) and single step chemistry. The model includes gas-phase radiative loss and a spherically symmetric, transient liquid phase. The model accurately predicts the droplet and flame histories of the experiments. Good agreement requires that the ignition in the experiment be reasonably approximated in the model and that the model accurately predict the pre-ignition vaporization of the droplet. The model does not accurately predict the dependence of extinction droplet diameter on pressure, a result of the simplified chemistry in the model. The transient flame behaviour suggests the potential importance of fuel vapour accumulation. The model results, however, show that the fractional mass consumption rate of fuel in the flame relative to the fuel vaporized is close to 1.0 for all but the lowest ambient oxygen mole fractions.

Combustion of single droplets and droplet pairs in a vibrating field under microgravity

This paper presents results of an experimental investigation on acoustic effects on combustion of single droplets and droplet pairs in microgravity. The ambient gas was air at atmospheric temperature and pressure, with octane as the fuel. A loudspeaker at the bottom of the chamber produced the acoustic field. Experimental results of single droplets showed that at low frequency and small to moderate acoustic intensities the evaporation rate increases, and the burning rate constant is nearly proportional to the product of frequency, f, and square of displacement, Xa2, fXa2. At higher acoustic intensities, the burning rate constant either remains constant or decreases, and in some cases, flame extinction occurs at a finite droplet diameter. The burning rate constant for a droplet pair is consistently lower than that for a single droplet. At lower frequencies, the burning rate constant reaches a maximum at an intermediate acoustic intensity. At higher frequencies, the burning rate constant increases monotonically with increasing acoustic intensity. The flame size decreases as a result of interactions, as does the critical spacing that indicates a merged flame around the droplet pair versus individual flames surrounding the droplets. The results also show that interactions stabilize the flame, in that droplet pairs burn to completion under conditions in which the flame surrounding a single droplet extinguishes at a finite droplet diameter.

Prevention of Over-Pressurization During Combustion in a Sealed Chamber

The combustion of flammable material in a sealed chamber invariably leads to an initial pressure rise in the volume. The pressure rise is due to the increase in the total number of gaseous moles (condensed fuel plus chamber oxygen combining to form gaseous carbon dioxide and water vapor) and, most importantly, the temperature rise of the gas in the chamber. Though the rise in temperature and pressure would reduce with time after flame extinguishment due to the absorption of heat by the walls and contents of the sealed spacecraft, the initial pressure rise from a fire, if large enough, could lead to a vehicle overpressure and the release of gas through the pressure relief valve. This paper presents a simple lumped-parameter model of the pressure rise in a sealed chamber resulting from the heat release during combustion. The transient model considers the increase in gaseous moles due to combustion, and heat transfer to the chamber walls by convection and radiation and to the fuel-sample holder by conduction, as a function of the burning rate of the material. The results of the model are compared to the pressure rise in an experimental chamber during flame spread tests as well as to the pressure fall-off after flame extinguishment. The experiments involve flame spread over thin solid fuel samples. Estimates of the heat release rate profiles for input to the model come from the assumed stoichiometric burning of the fuel along with the observed flame spread behavior. The sensitivity of the model to predict maximum chamber pressure is determined with respect to the uncertainties in input parameters. Model predictions are also presented for the pressure profile anticipated in the Fire Safety-1 experiment, a material flammability and fire safety experiment proposed for the European Space Agency (ESA) Automated Transfer Vehicle (ATV). Computations are done for a range of scenarios including various initial pressures and sample sizes. Based on these results, various mitigation approaches are suggested to prevent vehicle over-pressurization and help guide the definition of the space experiment. Nomenclature Af = area of the flame over the fuel-sample surface, m 2 Aw = area of the total available surfaces heat is convected to, m 2

Partial-Burning Regime for Quasi-Steady Droplet Combustion Supported by Cool Flames

A simplified model for droplet combustion in the partial-burning regime is applied to the cool-flame regime observed in droplet-burning experiments performed in the International Space Station with normal-alkanes fuels resulting in expressions for the quasi-steady droplet burning rate and for the flame standoff ratio. The simplified predictions are found to produce reasonable agreement with the experimentally measured values of burning-rate constants but not with their apparent dependencies on pressure or on the initial droplet diameter. Good agreement is found, however, with newly measured and numerically calculated flame standoff ratios in this droplet combustion supported by cool flames.

Spark ignition of a bidisperse, n-decane fuel spray

The spark ignition characteristics of a bidisperse aerosol were investigated. The purpose of the investigation was to determine the parameter(s) of the droplet size distribution that best correlates the ignition behavior of a spray. Two monodisperse aeeosols, generated by identical Berglund-Liu monodisperse aerosol generators, were used to create the bidisperse spray. A Phase Doppler Particle Analyzer was used to measure the droplet size distribution, droplet flux, droplet velocity and gas velocity at the spark gap. A capacitive discharge ignition system was used to generate the spark which ignited the mixture. Mixtures of oxygen and nitrogen were used as the oxidizing gas instead of air. The results show that the Sauter mean diameter correlates the ignition of the bidisperse sprays very well. The arithmetic, area, and volume mean diameters do not correlate the ignition behavior at all. In fact, in some cases, the arithmetic, area, and volume mean diameters exhibit a trend where the ignition energy decreases with, increasing droplet size. Comparison with monodisperse data also demonstrated that the Sauter mean diameter adequately describes the ignition of the bidisperse sprays. A semi-empirical characteristic time model was used to model the spark ignition of the sprays. The model was extended to include the effects of finite rate chemistry. The model correlated the ignition behavior of the bidisperse spray quite well. The best correlation was obtained by using the Sauter mean diameter consistent with the results of the experimental program.