An experimental program was conducted to study the burning of laminar gas jet diffusion flames in a zero-gravity environment. The tests were conducted in the Lewis Research Center's 2.2- Second-Zero-Gravity Facility and were a part of a continuing effort investigating the effects of gravity on basic combustion processes. The photographic results indicate that steady state gas jet diffusion flames existed in zero gravity but they were geometrically quite different than their normal-gravity counterparts. Methane-air flames were found to be approximately 50 percent longer and wider in zero gravity than in normal gravity.

An experimental investigation was conducted on methane, laminar-jet, diffusion flames with coaxial, forced-air flow to examine flame shapes in zero-gravity and in situations where buoyancy aids (normal-gravity flames) or hinders (inverted-gravity flames) the flow velocities. Fuel nozzles ranged in size from 0.051 to 0.305 cm inside radius, while the coaxial, convergent, air nozzle had a 1.4 cm inside radius at the fuel exit plane. Fuel flows ranged from 1.55 to 10.3 cu cm/sec and air flows from 0 to 597 cu cm/sec. A computer program developed under a previous government contract was used to calculate the characteristic dimensions of normal and zero-gravity flames only. The results include a comparison between the experimental data and the computed axial flame lengths for normal gravity and zero gravity which showed good agreement. Inverted-gravity flame width was correlated with the ratio of fuel nozzle radius to average fuel velocity. Flame extinguishment upon entry into weightlessness was studied, and it was found that relatively low forced-air velocities (approximately 10 cm/sec) are sufficient to sustain methane flame combustion in zero gravity. Flame color is also discussed. Haggard, J. B., Jr. Glenn Research Center NASA-TP-1841, E-487 RTOP 506-55-22

An experimental program was conducted to study the burning of laminar gas jet diffusion flames in a zero-gravity environment. The tests were conducted in the Lewis Research Center 2.2-Second Zero-Gravity Facility. The photographic results indicated that a sudden decrease in gravity level from 1 to 0 effected an immediate reduction in the length of the flame. Continued time in zero gravity resulted in the flame expanding away from the burner until extinguishment appeared to occur. Nondimensionalization of the governing flow equation yielded the parameters used to correlate the buoyancy effects.

"From the standpoint of industrial safety, explosions present a major hazard, particularly for chemical plants and refineries, where accidental releases of large quantities of flammable material can often occur. When an explosion does take place at one of these facilities, the greatest hazards are often large-scale deflagrations, which can result in significant damage to individual buildings or even widespread damage over a large area, as in the case of unconfined vapor cloud explosions. At these scales, the formation of flame instabilities significantly increase the overall burning rate of the flame and the severity of the event.Without an adequate understanding of the mechanisms which govern the growth of these flame instabilities, and how they interact, existing modeling techniques are unable to accurately predict the consequences of these events. This creates considerable uncertainty in the protection requirements for these facilities. To improve our understanding of the mechanisms behind the development of large-scale flame instabilities, a comprehensive experimental study was performed examining the onset and growth of spherical-flame instabilities for propane-air, methane-air, and hydrogen-air flames at large scale. Across these fuels, a range of concentrations were studied, allowing for the growth of these instabilities to be evaluated for both thermal-diffusively stable and unstable flames. In addition, the interaction between weak initial turbulence and the growth of these instabilities was also examined. From the experimental work, it was shown that the critical radius for onset of instability varies linearly with mixture Markstein length, consistent with smaller-scale studies seen in the literature. For mixtures with positive Markstein length, an oscillatory rate of flame acceleration was observed that had not been previously identified. These oscillations were consistent with the fractal theory of spherical-flame acceleration and the formation of discrete length scales of instability. In the case of thermal-diffusively unstable mixtures, with negative Markstein length, no oscillations were observed; however, the rate of flame acceleration was consistent with a 4/3 power law exponent described in the literature. When weak initial turbulence was introduced, it was found that thermal-diffusively stable flames, with positive Markstein length, behave similar to mixtures with negative Markstein length under quiescent conditions. Specifically, initial turbulence significantly reduced the critical radius for onset of instability and increased the overall rate of spherical-flame acceleration in these mixtures. Using a number of assumptions inferred from the experimental work, an analytical model was developed to describe the growth of flame instabilities on a spherical flame, considering a fractal approach where multiple length scales grow on the flame surface simultaneously. This model shows how the growth of these instabilities can have a significant impact on the flame speed and overall rate of fuel consumption, and why they must be accounted for in the development of models describing large-scale flames. In particular, this work shows how turbulence plays a key role in determining the onset of instability and the rate of spherical- flame acceleration that occurs. This analytical model can be used to provide insight into the development of new engineering correlations or provide a basis for the development of sub-grid CFD models that quantify the effect of flame instabilities at large scale." --

A Hydrogen-Helium mixture was chosen to investigate the structure of a counterflow diffusion flame. Reacting and non reacting conditions were studied at the same Reynolds number. To study the reaction zone structure, high speed tomography based on Mie scattering was employed using a copper vapor laser and a Fastax high speed camera. LDV measurements were also obtained. Different seeding techniques were used to visualize both the turbulent air and fuel jets. The tomographic records were digitized and recorded in a digital computer for statistical treatment. Significant differences in the wrinkle scales between the reacting and non reacting flows were found. A fractal statistical analysis of the tomography records was done to quantify these differences. Seeding of both fuel and air jets provided a mean for the evaluation of the reaction zone thickness. The strain of the reaction zone was obtained from the time resolved tomographic records. Local flame extinction and reignition were observed for different H2/Helium fuel mixtures. Keywords: Turbulent diffusion flames, Rayleigh scattering.