Dennis P. Stocker

SOOTING LIMITS OF MICROGRAVITY SPHERICAL DIFFUSION FLAMES IN OXYGEN-ENRICHED AIR AND DILUTED FUEL

Limiting conditions for soot-particle inception were observed in microgravity spherical diffusion flames burning ethylene at 0.98 bar. Nitrogen was supplied to the ethylene and/or oxygen to obtain the broadest available range of stoichiometric mixture fraction, Zst. Both normal flames (surrounded by oxidizer) and inverse flames (surrounded by fuel) were considered. Soot-free conditions were found to be favored at increased Zst and there was no observed effect of convection direction on the sooting limits. The sooting limits follow a linear relationship between adiabatic flame temperature and Zst, with Zst accounting for a variation of about 700 K in the sooting-limit adiabatic flame temperature. This relationship is in qualitative agreement with a simple theory that assumes soot inception requires the local C/O atom ratio and temperature to be above threshold values, (C/O)c and Tc, respectively. The theory indicates that different mechanisms are responsible for sooting limits at low and high Zst. When inert is added to a fuel/air flame, a sooting limit is obtained when temperature becomes so low that the kinetics of soot inception are too slow to produce soot. On the other hand, a flame with a high Zst has low C/O ratios far into the fuel side of the flame. For such a flame, soot-free conditions can be attained at much higher temperatures because there is sufficient oxygen on the fuel side to favor oxidation of light hydrocarbons over formation of soot precursors.

Simulating Gravity in Microgravity Combustion Using Electric Fields

A small number of intermediate reactions during the combustion of hydrocarbon fuels produces chemi-ions. When acted upon by an electric field, the resulting ion-driven wind modifies the flame by producing a complex coupling that can affect flame shape, soot formation, and stability limits. Nearly all work to date involving electric fields and combustion has been performed under the influence of gravity, which produces a similar and potentially confounding influence. The following work compares the effect of an electric field on a jet flame in 1 g and µg. All tests are performed at the NASA Glenn 2.2-s Drop Tower. Testing shows that microgravity flames can be manipulated to resemble 1 g flames by using an electric field alone.

Numerical and experimental observations of spherical diffusion flames

Spherical diffusion flames supported on a porous sphere were studied numerically and experimentally. Experiments were performed in 2.2 s and 5.2 s microgravity facilities. Numerical results were obtained from a Chemkin-based programme. The programme simulates flow from a porous sphere into a quiescent environment, yields both steady state and transient results and accounts for optically thick gas-phase radiation. The low flow velocities and long residence times in these diffusion flames lead to enhanced radiative and diffusive effects. Despite similar adiabatic flame temperatures, the measured and predicted temperatures varied by as much as 700 K. The temperature reduction correlates with flame size but characteristic flow times and Lewis number also influence temperature. The numerical results show that the ambient gas Lewis number would have a strong effect on flame temperature if the flames were steady and nonradiating. For example, a 10% decrease in Lewis number would increase the steady state flame temperature by 200 K. However, for these transient, radiating flames the effect of Lewis number is small. It was also observed that when hydrocarbon fuel is supplied from the ambient the large diffusion distances associated with these flames can lead to unusual steady state compositions near the outer boundary because decomposition products can diffuse to the outer boundary. This results in a loss of chemical enthalpy from the system but the effect on flame temperature is small. Transient predictions of flame sizes are larger than those observed in microgravity experiments. Close agreement could not be obtained without either increasing the models thermal and mass diffusion properties by 30% or reducing mass flowrate by 25%.

Small-scale smoldering combustion experiments in microgravity

Results from small-scale experiments of the smolder characteristics of a porous combustible material (flexible polyurethane foam) in microgravity and normal gravity are presented. The microgravity experiments were conducted in the Spacelab Glovebox on the USML-1 mission of the Space Shuttle Columbia, June/July 1992, and represent the first smolder experiments ever conducted under extended periods of microgravity. The use of the Glovebox limited the size of the fuel sample that could be tested and the power available for ignition but provided the opportunity to conduct such experiments in space. Four tests were conducted, varying the igniter geometry (axial and plate) and the convective environment (quiescent and forced). A series of comparative tests was also conducted in normal gravity. Measurements conducted included temperature histories at several locations along the fuel sample, video recording of the progress of the smolder, and postcombustion char and gas composition analyses. The results of the tests showed that smolder did not propagate without the assistance of the igniter, primarily because of heat losses from the reaction to the surrounding environment. In microgravity, the reduced heat losses caused by the absence of natural convection resulted in only slightly higher temperatures in the quiescent microgravity test than in normal gravity but a dramatically larger production of combustion products in all microgravity tests. Particularly significant is the proportionately larger amount of carbon monoxide and light organic compounds produced in microgravity, despite comparable temperatures and similar char patterns. This excessive production of fuel-rich combustion products may be a generic characteristic of smoldering polyurethane in microgravity, with an associated increase in the toxic hazard of smolder in spacecraft.

Radiative Loss from Non-Premixed Flames in Reduced-Gravity Environments

This paper presents the results of an experimental and theoretical study of radiation from luminous jet diffusion flames in partial-gravity environments. Tests were conducted for laminar, non-premixed methane flames burning in quiescent air on-board the NASA KC-135 research aircraft for a range of gravity levels. Flame radiation and gravitational acceleration were measured, and flame imaging was performed. The radiation data are compared with those obtained from normal-gravity and microgravity tests, conducted in a drop facility. Effects of g-jitter on radiation measurements are discussed. With the aid of predictions from a numerical model of jet diffusion flames, the influence of gravity on radiation through its effects on the temperature, species, and velocity fields is analyzed. Good agreement between predictions and measurements is obtained.

Turbulent Structure Dynamics of Buoyant and Non-Buoyant Pulsed Jet Diffusion Flames

The flame structure dynamics of strongly pulsed, turbulent diffusion flames were examined experimentally in a co-flow combustor. High-speed visual imaging and thermocouple measurements were performed to determine celerity, defined as as being the bulk velocity of a given flame puff structure in the large-scale, turbulent flame structures. Tests were conducted in normal gravity and microgravity with a fixed fuel injection velocity with a Reynolds number of 5,000 and also with a constant fueling rate where the Reynolds number ranged from 5,000 to 12,500. The celerity of strongly interacting flame puffs is as much as two times greater than for the case of isolated flame puffs. The amount of decrease in celerity at the visible flame tip due to the removal of buoyancy ranges from 7% to 11% in most cases, to as much as 36% for both fixed jet injection velocity and constant fueling rate. At the same time, the flame length is modestly affected by the removal of positive buoyancy, amounting to a decrease of as much as 20%. These observations hold for both fixed injection velocity and constant fueling rate cases. The observed increases in the flame puff celerity and the mean flame length with decreasing jet-off time, for a given injection time and gravity level, are consistent with a decreased rate of oxidizer entrainment into each flame puff structure due to increased flame puff interactions. A scaling argument accounts for the decrease of the flame puff celerity with downstream distance when both quantities are normalized by the appropriate injection conditions. The celerity, as characterized by the temperature measurement method, appears to be essentially unaffected by buoyancy at any given downstream location when appropriately scaled. The visual tracking method suggests a modest buoyancy effect at a given downstream distance, suggesting a subtle impact of buoyancy on the flame puff structures that does not impact the bulk motion.

Effects of C/O Ratio and Temperature on Sooting Limits of Spherical Diffusion Flames

‡‡ Limiting conditions for soot particle inception in spherical diffusion flames were investigated numerically. The flames were modeled using a one-dimensional, time accurate diffusion flame code with detailed chemistry and transport and an optically thick radiation model. Seventeen normal and inverse flames were considered, covering a wide range of stoichiometric mixture fraction, adiabatic flame temperature, residence time and scalar dissipation rate. These flames were previously observed to reach their sooting limits after 2 s of microgravity. Sooting-limit diffusion flames with scalar dissipation rate lower than 2 s -1 were found to have temperatures near 1400 K where C/O = 0.51, whereas flames with greater scalar dissipation rate required increased temperatures. This finding was valid across a broad range of fuel and oxidizer compositions and convection directions.

Effects of Gravity on Swirl-Stabilized Fluidized Beds

A fluidized bed reactor for use under microgravity and partial gravity conditions has been designed and tested at NASA Glenns 2.2 Second Drop Tower. Injection of a swirling flow is utilized to stabilize the bed. Under microgravity conditions, the fluidized bed is formed under a balance of radial drag and centrifugal forces acting on the bed particles. In laboratory tests, gravitational effects, primarily related to settling effects of the particles under normal gravity conditions, are clearly observed and impact the fluidized bed formation. In microgravity, there is a lack of gravitational settling of the particles. Analysis of the particle distribution is carried out by imaging with orthogonal cameras.