A Tale of Two Flames

I haven’t posted in awhile, so here’s a lit review.

The decomposition and oxidation of hydrazine has been a topic of scientific interest for at least eighty years. Early studies, such as those by Askey or Bamford, focused on the vapor phase. Bamford noted that the chemical exploded when sparked or heated.
Audrieth noted much interest in low-concentration (~30%) hydrazine as a fuel in the years after World War II, writing that, “The hydrogen peroxide-hydrazine combination was first utilized by the Germans as a rocket fuel and represents one  of the most promising bi-fuels for long-range high-altitude missiles.” He also documented that this combination appeared to be “self-starting”. In the same year, 1951, a paper by Murray and Hall recorded the observation of a 93% hydrazine flame. They described “possibly two inner cones, separated from one another by a very small distance.” This is extremely interesting to note in light of the dual-flame phenomena for droplets, although the authors attributed the second cone to “radiation from reaction products”.
In 1955, Donald Kiley, a student at Caltech, addressed monopropellant droplet combustion in his thesis. He was unable to ignite, by spark or hotwire, any monopropellant droplet initially in a pure nitrogen atmosphere. Using a porous sphere fed with liquid hydrazine, he then tried to replace the air in the chamber with nitrogen while burning the fuel, but witnessed extinction of the flame before the atmosphere was completely replaced. He did not record at what oxygen concentration this occurred. He also did not mention the presence of a dual flame when hydrazine is burned in air. Kiley noted that hydrazine is probably not a simple diffusion flame, and that decomposition probably occurs somewhere within the flame.
Both F.A. Williams and B.R. Lawver reference a 1957 paper by W. A. Rosser, in which he studied decomposition burning at Caltech’s Jet Propulsion Laboratory (JPL). This paper is the first record of the observation of a dual flame about a droplet of burning hydrazine. In a 1966 paper, Rosser described the JPL experiment. A steady-state burning droplet was simulated by supplying an appropriate mass flow rate to a porous sphere; the hydrazine was ignited by spark. At one atmosphere of ambient pressure, with the O2 content between 0 and 10% by volume (the rest was N2), a decomposition flame was reliably seen. When the O2 slightly exceeded 10%, Rosser recorded the presence of an orange decomposition flame surrounded by a larger, orange oxidation flame. He associated the outer flame with the reaction of ammonia and hydrogen with oxygen. In air, he noted that the oxidation flame is yellow, smaller, and so bright as to render the inner flame indistinct.
In 1960, Dykema and Green also performed hydrazine experiments with air and oxygen. They confirmed the double flame observation.
Lawver added to these studies, choosing to document the combustion of actual droplets. In 1966, he described hanging hydrazine droplets from a thermocouple, in a test apparatus that could be filled with gaseous nitrogen tetroxide. The subsequent combustion was filmed with a 64 frame/s camera. He described a yellow, inner decomposition flame, and an orange, outer oxidation flame. The ambient NTO was at a pressure of one atmosphere and a composition of approximately 62% NO2. A graph of flame thickness history was presented . The inner and outer flame thicknesses were roughly constant throughout the burn, only dropping off toward the very end of the droplet lifetime.
He also conducted an experiment in which he filmed a hydrazine droplet burning in the combustion products of another hydrazine droplet. NTO was still present, but the burning rate and flame thickness were markedly different, the inner and outer flame thicknesses being roughly equal, and increasing as combustion progressed.
In 1966, Rosser expanded his earlier work and was able to sustain a hydrazine decomposition flame in the absence of oxygen, but not at ambient pressures below one atmosphere. The composition of the hydrazine used  – namely the amount of aniline  present – was significant here.
Hersh, in work with Lawver and others, released a 1967 technical report in which he described the experimental probing of the flame surrounding a hydrazine droplet. Thermocouples were used to gather the temperature profiles of droplets burning in NTO and O2/N2 mixtures; a concentration sampling probe was used in an attempt to determine the species composition of hydrazine burning in NTO. Hersh concluded that the decomposition flame was located very close to the surface of the droplet, as the gas inside the inner flame appeared to consist of hydrazine decomposition products. The outer flame contained NO2 decomposing into NO and O2. Temperature profiles supported this; temperature close to the droplet surface exceeded the hydrazine boiling point and was more consistent with the hydrazine decomposition reaction. Peak temperature was reached in the area between the observed inner and outer flames, consistent with the decomposition reaction of NO2. The observed inner and outer flame locations did not correspond to significant temperature changes, but seemed to be areas of highly visible radiation. This temperature/flame observation also held true for combustion in O2/N2 mixtures. Speaking of NTO, Hersh wrote, “These results suggest that the two ‘visible flames’ are associated with the oxidizer while two real flames are found only for hydrazine. The combustion of hydrazine with any oxidizer will be a two flame structure consisting of decomposition at the droplet surface and the oxidation of the decomposition products.”
A 1971 NASA contractor report, by C.B. Allison of Pennsylvania State University, provides a thorough review of experimental and theoretical efforts in the study of hydrazine flames up to that year. He did not mention the work by Hersh, but did reference a later paper by Lawver that might include the issue of visible vs. actual flames. Allison’s experimental work included the burning of hydrazine, MMH, and UDMH  in both droplets and porous spheres. These fuels were burned in combustion products, consisting mainly of CO2, N2, and O2. O2 content of this oxidizer was varied. Special attention was given to high temperature conditions, approximating those in a rocket combustion chamber. Allison categorized three types of combustion: monopropellant, where the fuel can support a flame in the absence of an oxidizer; bipropellant, in which the diffusion flame  consists of the reaction between evaporated fuel and oxidizer; and “hybrid”, in which a monopropellant is burned in an oxidizing medium. This is characterized by an inner decomposition flame and an outer oxidation flame and applies to hydrazine. Allison produced a model of burning rate for the three hydrazine fuels which accounted for dual flames and agreed with his own and other experimentalist’s data to within 20%. He also conjectured that, for the hydrazine/NTO reaction, “the visible fronts may or may not indicate the position of the monopropellant and bipropellant flames” in light of Lawver’s (Hersh’s?)  work.
Most droplet combustion studies since the 1970s seem to focus on numerical models of the chemical kinetics involved in hypergolic ignition.
A gel is defined by Brinker and Scherer5 as “a substance that contains a continuous solid skeleton enclosing a continuous liquid phase. The continuity of the solid structure gives elasticity to the gel.” Arnold adds that the solid-like characteristics of gels greatly increase their safety in regards to handling and leakage. When shear stress is applied, the gel begins to flow and exhibit more liquid characteristics. Gels may be inorganic, such as silica derivatives, which means that the gellant does not combust. Organic gellants, like hydroxypropylcellulose (HPC) or hydroxyethylcellulose (HEC), burn with the propellant. Significantly, gels also enable the addition of metal particles such as aluminum, which increase the energy density of the propellant. Natan and Rahimi also noted that gelled droplets tend to have a lower burning rate than their liquid counterparts, but that they generally follow the conventional d-squared burning rate law.
According  to Natan, the investigation of gelled hypergolic fuels began in the 1950s. Various versions of hydrazine and NTO gels remained of interest throughout the 1970s and 1990s, primarily in engine performance evaluations. A missile utilizing carbon-loaded gelled MMH and gelled inhibited red fuming nitric acid (IRFNA) completed a successful flight test in 1999, including a midcourse re-targeting maneuver.
In a 2006 paper , Solomon and Natan described the droplet combustion of JP-8 in Thixatrol 289, an organic gellant. As burning progresses, a layer or shell of slightly elastic dried gel is formed on the droplet surface. Product vapor then accumulates within the shell, increasing the droplet volume. Eventually the pressure ruptures the gel and releases a gaseous jet. This phenomenon may be repeated several times over the lifetime of the droplet.
Arnold et al. presented a 2010 paper which detailed the droplet burning of MMH/HPC, RP-1/SiO2, and JP-8/SiO2 gels. Each droplet was suspended from a thermocouple, ignited in air, and observed with a high-speed camera. Gelled MMH behaves similarly to liquid MMH over the first 75% of the burning time, although it combusts at a slightly higher temperature. Arnold also observed “swelling and jetting” in the MMH/HPC gel. When HPC remains after combustion is finished, the structure is elastic rather than stiff.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s