IMPORTANCE OF CHEMICAL DELAY TIME IN UNDERSTANDING HYPERGOLIC IGNITION BEHAVIORS
C. S. Hampton*, K. K. Ramesh*, and J. E. Smith Jr.†
* Graduate Rearch Assistants,† Professor and Corresponding Author
University of Alabama in Huntsville
Department of Chemical and Materials Engineering
Huntsville, AL, 35899.
Abstract
The time delay for hypergolics to react from initial contact to the appearance of flame is classically termed the ignition delay time. A lar bad diagnostic technique that measures ignition delay times, identifies veral distinct regions within the classical ignition delay time for hypergolic reactions. This technique will be described and compared to previous techniques ud to determine ignition delay time. This technique, becau of its resolution, was the first optical drop test laboratory
technique to accurately measure the chemical delay time imbedded within the ignition delay time. When considering the design of hypergolic rocket motors, the ignition delay time is typically ud. The ignition delay time includes both the transport and mixing of the fuel and oxidizer portion of the process that is controlled by the motor’s design and operating conditions, as well as the chemical delay time. The chemical delay time permits direct comparisons of chemical performance of hypergolic reacts independent of the mixing technique ud. Unsymmetrical dimethylhydrazine (UDMH), hydrazine and mono-methylhydrazine (MMH) were individually reacted with red fuming nitric acid, as well as mixtures of UDMH and hydrazine. Results are prented and discusd along with new analysis approaches that will enhance data reduction and accuracy.
Introduction
Hypergolic bipropellants are defined as fuel and oxidizer combinations that, upon contact, chemically react and relea enough heat to spontaneously ignite. Nitric acid and oxides of nitrogen are ud as the oxidizer. The fuels are organic compounds including the following: amines, heterocyclic compounds, and polyatomic phenols. The discovery of hypergolicity occurred in Germany around 1937. After World War II rearch on hypergolic bipropellants spread to other countries. During the 1950s, interest in hypergolics bipropellants grew as knowledge of their high density, high performanc
e, and long term storability spread. The Titan, Gemini, and Apollo programs all ud hypergolic bipropellants. The rvice module that first orbited the moon and later returned the astronauts to the earth ud a single hypergolic engine. The lunar excursion module that landed the astronauts on the moon ud one engine for descent to the surface, and one engine to return to the rvice module. Currently the Ariane, Long March, Space Shuttle and International Space Station programs are among the urs of hypergolic bipropellants.1
A major problem with hypergols is the toxicity of the components. To develop hypergolic fuels that are less toxic to humans and the environment requires an enhanced understanding of the kinetic mechanisms taking place during hypergolic combustion. Eventually, an understanding of chemical kinetics and free radical generation of the classical hypergolic systems can be ud to develop alternate organics or mixtures of organics that produce similar or enhanced performance over current hypergolics at reduced toxic levels.
One important measure of the performance for a fuel/oxidizer combination is the length of time between reactant contact and appearance of the flame. This ignition delay, measured in milliconds, is important becau longer than desired delays can lead to lower performance or cau catastrophic failure of the engine.
This paper reports on a method to measure and examine ignition delays for hypergolic bipropellants. Using this lar diagnostic technique, measurements can now be accomplished quickly, using very small quantities of fuel and oxidizer to minimize both toxic and explosive hazards of fuel such as UDMH, MMH, and hydrazine with oxidizers such as red fuming nitric acid. As an advanced technique, it required independent confirmation. Anderson (1999) of Talley Defen Inc. accomplished this for the TRW Corporation using a high-pressure impinging jets technique to rapidly measure the reaction between MMH with IRFNA that benched marked an ignition delay time of 0.4 mc for equal volume of MMH and IRFNA.2, 3This ignition delay time is consistent with the lar diagnostic technique that measured a chemical delay of 0.4 mc at an oxidizer to fuel ratio of 1 using MMH and RFNA.4 The high-pressure reactors were later ud to test the delay times of veral new azides being developed by AMCOM at the Redstone Arnal. Since the approach ud by Talley Defen Inc. is expensive, tests at other fuel to oxidizer ratios and mixed azides were not possible. This lar diagnostic approach will be discusd and results for both pure components such as hydrazine and UDMA as well as mixtures of the and
41st Aerospace Sciences Meeting and Exhibit 6-9 January 2003, Reno, Nevada
AIAA 2003-1359 Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2008
For Evaluation Only.
多学other classical hypergolic fuels contacted with RFNA will be prented and discusd. This technique, becau of its resolution, is the first to measure the chemical reaction time just prior to ignition. The
chemical delay portion of the ignition delay time defines the time during which free radicals generated by the reaction should be analyzed. The following ction will describe previous techniques found in the literature to measure the ignition delay time for comparison with the lar diagnostic technique reported here.5, 6
Literature Review
Diagnostic techniques that measure ignition delay times of hypergolic bipropellants are normally classified into a few distinct types. Drop tests are techniques that drop one reactant from a t height into a stationary quantity of the cond reactant. Mixing tests are techniques that u a method to enhance mixing of the reactant combinations. Impinging jet techniques are tests that u parate fuel and oxidizer injectors to enhance the mixing rate and simulate engine conditions. Also, a few small scale rocket engines have been equipped to measure ignition delay times.
A typical example of a drop test technique was performed by Broatch (1950) and is shown in Figure 1.7In this tup, a light beam was focud on a
photocell a t distance above the organic fuel located in a crucible. The oxidizer, in a stream of droplets of varying size, broke the light beam as it fell into the fuel. The droplets contacted the fuel
at varying times, with the initial droplet being the reference for the ignition delay measurement. A photocell ended the measurement when it nd the appearance of a flame. The time between the two events defined the ignition delay for the hypergolic bipropellant. Since the droplets entered the oxidizer at varying times, the ignition delay was not well defined. In addition, this meant that the oxidizer to fuel ratio could not be defined by Broatch’s technique.
熬夜怎么补救A technique ud by Gunn (1952) was defined as a mixing test by Paushkin (1962) in his review of jet fuels.8, 9 The only significant difference from Broatch’s technique described above was that the quantity of fuel and oxidizer ud by Gunn was veral times larger. Several milliliters of one reactant were decanted into a similar quantity of the other reactant. The start of timing was due to the completion of an electrical circuit. This was achieved by contact between the fuel and oxidizer. The end of timing was obtained by the respon of a photocell to the appearance of flame, similar to Broatch’s technique. The photocell ud by Gunn to detect the flame, and the one ud in Broatch’s technique above, did not indicate the strength of the reaction. The photocells only indicated the end of an ignition delay measurement. The technique prented here, does provide insight into the combustion rate.
The device constructed by Pino (1955) caud the pressurized injection of oxidizer through 4 ports
directed into a stationary quantity of fuel.10 The pressurized injection enhanced mixing of the reactants. The combined quantity of fuel and oxidizer ud was approximately 4 milliliters for one test. Again, the start of timing was due to the completion of an electrical circuit between the reactants. The end of timing was nd electrically due to the ionizing effect of the flame. The results of Pino’s testing show no indication of the oxidizer to fuel ratio. The measuring technique only captures the ignition delay time. No additional phenomena were noted.
Ladanyi and Miller (1956) placed a small glass ampoule containing approximately one milliliter of fuel under the surface of veral milliliters of oxidizer.11 The glass ampoule was crushed by a steel rod to enhance mixing. The initial time began at the moment of ampoule fractured. This was detected by the completion of an electrical circuit between a weight ud to supply the downward force to the steel rod and the steel rod itlf. The final measurement was made by a photocell nsing the appearance of a flame. The authors stated in the paper that mixing rate varied from test to test due to differences in the ampoules. Oxidizer to fuel ratio was not defined and varied during testing. Ignition delay times are a function of both mixing rate and oxidizer to fuel ratio.
Kilpatrick and Baker (1955) ud a device that forced both fuel and oxidizer together using high-pressure gas hydraulics.12 The reactants were initially located in parate chambers below pistons
which forced the propellants together immediately prior to injection into the combustion chamber. The initiation of the timing measurement was through monitoring of piston movement above the fuel and oxidizer. The end of an ignition delay measurement corresponded to an increa of pressure in the test chamber, which had a
Figure 1: Broatch’s Drop Test
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2008For Evaluation Only.
volume of 339 cm3. This was corrected for the respon time of a pressure gage located 0.5 meters from the chamber. The test results using the lar diagnostic approach show that significant vapor volume is created locally by the bipropellants just prior to ignition. Their test did not indicate any attempt to obtain pressure values that consider this phenomenon for the fuels tested. In addition, this pressure ri could originate anywhere in the test chamber affecting pressure gage respon time. The two factors limit the accuracy of the measurements made using this equipment. This technique did control oxidizer to fuel ratio, but provided no additional information other than the ignition delay time.
Saad and Goldwasr (1969) ud impinging fuel and oxidizer jets in their technique.13 They initiated the ignition delay measurement at the moment the valves for fuel and oxidizer were relead. Photocell detection of the flame was again ud to end the measurement. The resolution of their oscilloscope was 100 milliconds/division. As ignition delay values are generally less than 100 milliconds, the resolution of their technique does not provide the accuracy needed for reactions of this speed. They initiated the ignition delay measurement from the moment the valves for fuel and oxidizer were relead. The technique did not measure from the moment of impingement. The measurements are not ignition delays, as defined in the literature.
In addition to the drop test technique reported earlier, Broatch also ud the technique of impinging jets.7He ud high-speed photography to capture pictures of the combined jets and flame. The ignition delay was calculated from the length of the combined jets to the fully developed flame front. No consideration was obrved in the paper for the possibility of flame propagation upstream or downstream. No attempt was made to record the moment of ignition. Again, the measurements do not reprent ignition delays. The benefits of the photography technique were to capture the strength of the flame as Broatch varied oxidizer to fuel ratio and temperature, and the ability to photograph phenomena such as pre-ignition boiling of the liquid pha.
Spengler and Bauer (1966) ud impinging jets to test the influence of pressure and varying chemical composition on ignition delay measurements.14This technique consisted of starting the timing by contacting the fuel and oxidizer, which completed an electrical circuit. The timing measurement was ended by nsing the flame with a photocell located between the two injectors. This technique for impinging jets emed to be the only true measure of ignition delay among the three discusd. However, once again, the technique for measuring the ignition delay provided no additional information.
In review, veral types of ignition delay techniques have been examined. They vary in complexity and ability to measure ignition delay. The design of an ignition delay apparatus is heavily dependent upon the transport and mixing portion of the ignition delay time measurement and not on the chemical delay time, which is directly tied to the chemical reaction rate.
If the intent of the ignition delay technique is to screen hypergolic bipropellants, then the test should provide reliable information on the relative chemical performance. The ignition delay diagnostic technique prented in this paper was designed as a rearch tool to provide such data. As will be shown, this technique is the first to provide a measure of the chemical performance of the reactant combinations. The chemical delay time is independent of the mixing technique by not the fuel to oxid
izer ratio.
Equipment
The entire operating system is designed to study the reaction rates and mechanisms of hypergolic bipropellants. The equipment us a variety of techniques, including lar timing and visible and near infrared Raman spectroscopic measurements, to meet the objectives. The combustion chamber, supporting systems, and diagnostics have been designed and asmbled as detailed in recent publications.4, 5, 6, 15 This equipment has recently been updated to permit rapid analysis and multiple triggering circuits for chemical analysis and is shown schematically in Figure 2. The lar source currently employed for the ignition delay measurements and visible Raman spectroscopy is a Lexel Model 95-4 Argon-Ion lar. A Continuum Model SL 10Hz Nd:Yag lar is ud for near-infrared Raman spectroscopic measurements. Two spectrometers, each equipped with an optical multi-channel analyzer (OMA), can monitor the visible and near-infrared regions. The spectrometer ud for the near-infrared region is an Acton Rearch Corporation SpectraPro-275 mm equipped with a Princeton Instruments INGAAS OMA. For the visible region an SPEX 1877B 3.0 m Triple Spectrometer equipped with an EG&G Princeton OMA is ud. A Tektronix TDS 5104 Digital Oscilloscope is ud measure the ignition and chemical delay times, and trigger the Nd:Yag l
ar
and the two optical multichannel analyzers to make time resolved measurements. The outputs of all system components are fully interfaced to a very powerful digital computer.
The combustion chamber is shown in Figure 3. The body and flanges were machined from 304 stainless steel. The chamber itlf can be aled and pressurized to hold a flowing, inert gas or vacuum. All test results prented in this paper were performed at atmospheric
pressure in air.
The chamber contains eight side ports. One port supports a linear positioner that is ud to align the combustor. This combustor is a passivated stainless steel receptacle for the hypergolic bipropellant. It is designed with a slight conical indention on the top surface to center the liquid pha properly and quickly dissipates heat resulting from the combustion process. Another port allows feed through of hypodermic needles connected to Cole-Parmer digital syringe pumps
to inject the fuels and oxidizers. There is also a port for temperature and pressure monitoring of the chamber. The remaining ports hold high quality optics for lar diagnostics at the appropriate wavelength.
Currently, the combustion system supports ignition time delay studies between liquid fuel and oxidizer under various atmospheres but can be modified to support a host of fuel/oxidizer combinations.
Experimental Procedure
The experimental procedure involves three main steps, namely lar and phototransistor alignment,
determining the proper oxidizer to fuel ratio, and measuring the respon from the combustion chamber. To measure the ignition delay with the system, an Argon-Ion lar source is pasd directly above a droplet of oxidizer within the combustor. This is accomplished by raising the combustor or combustion chamber until the bottom edge of the Gaussian lar source is clipped by the top edge of the combustor. This also aligns the lar beam with the top surface of the oxidizer droplet. This is similar to the work of Broatch. However, Broatch’s technique could not monitor the top surface of the liquid pha in the crucible as our technique does.7
As shown in Figure 4, two phototransistors are required to monitor the chamber. In operation, phototransistor #1 monitors the lar beam intensity through a pinhole 200 µm in diameter. This asmbly is positioned by a three axis Daedal micropositioner. For a nominal experiment, the horizontal position of the
detector is adjusted to maximize the signal on channel 1
27英寸显示器的长宽高
of a digital storage oscilloscope (DSO). Once aligned horizontally, the vertical position is lowered until the output voltage from phototransistor #1 begins to decrea. This indicates that the pinhole is being attenuated by the oxidizer surface. The pinhole is then raid until the signal is again maximiz
ed. This is accomplished to achieve maximum nsitivity and to align the lower edge of the pinhole with the oxidizer surface. The pinhole/diode geometry restricts the view
Figure 4: Phototransistor Placement
Figure 3: Top View of the Combustion Chamber Figure 2: Overview of the Entire System
of the phototransistor to approximately 200 µm above the surface.
容易的意思Phototransistor #2 monitors emission from the combustor through a band pass filter that eliminates scatter light from the Argon-Ion lar source. The filter is necessary due to the extreme nsitivity of this phototransistor throughout the visible and near infrared spectrum. A ries of concentrating optics are ud to collect the flame emission. The focusing optics restricts the view of the flame by the phototransistor to a region approximately 2 mm in diameter, directly above and centered on the combustor.
Using the appropriate safety equipment, under a chemical fume hood the syringes for the fuel and oxidizer are loaded. They are then taken parately to the testing area and attached to the ends of 1-meter long hypodermic needles. A dilute sodium hydroxide solution is nearby in ca of an accid
ental acid spill. For a single test, the amount of fuel and oxidizer ud is generally less than 100 µL.
埃森大学军训心得800A measurement is made by using a digital syringe pump to place a precalculated amount of oxidizer onto the combustor through a hypodermic needle as shown in Figure 4. This calculation ts the desired oxidizer to fuel ratio by mass. Once the oxidizer to fuel ratio is obtained, a fuel droplet of known volume is discharged from the cond hypodermic needle positioned approximately 1.5 cm above the oxidizer. This droplet pass through the lar beam and attenuates the respon of phototransistor #1, which will be en as a decrea in signal on channel 1 of the DSO. This marks the moment of contact between fuel and oxidizer. Later, a vapor pha above the fuel and oxidizer pool ignites, and the flame nd by phototransistor #2. Channel 2 of the DSO, measures the output from phototransistor #2. The time lag between the two events is a direct measurement of the ignition delay.日屁小说
In addition to the timing measurements discusd above, we are currently integrating the Tektronix TDS 5104 DSO to trigger the puld Nd:Yag lar and the two OMA’s in Figure 2. This will permit time resolved Raman measurements of intermediate vapor pha species formed above the fuel/oxidizer mixture within the chemical delay region but before actual combustion to develop reaction pathways for the reaction.
昨天的故事Results and Discussions
In this ction, the results from the lar diagnostic technique are prented. It will be shown that this technique provides a better understanding of the events taking place during hypergolic combustion.
Figure 5 shows the DSO traces resulting from a typical drop test for (UDMH) contacted with red fuming nitric acid (RFNA). The UDMH had a purity of 98% and was obtained from the Aldrich Company. The RFNA had a nitric acid concentration of 90-95%, with the balance being nitrogen tetroxide. The results in this figure are for an oxidizer to fuel ratio of 2. In Figure 5, the upper trace (channel 1) shows the output from phototransistor #1, which monitors the surface of the oxidizer droplet. The lower trace (channel 2) shows the output from phototransistor #2, which monitors flame emission. For this test, the horizontal time scale was 2
milliconds/division.
Various reference points have been added to this figure to identify key features resulting from this technique. Point A on channel 1 reprents the moment of contact between the fuel and oxidizer. As the fuel droplet falls through this region, a variety of respons can occur. If the droplet is oscillating at some fundamental frequency, the lower surface of the droplet can refract and reflect the lar beam resulting in various respons. Once the lower surface of the fuel droplet contacts the oxidizer, a region of rapid linear decrea follows. To consistently reference the ignition time delay to a definable point, we extrapolate this linear region to the initial reference level. This approach eliminates experimental variations associated with the droplet oscillations and radius of curvature. An identical approach is ud at point B to neglect the shape of the trailing edge of the droplet and any wake that it creates on entry. Thus, the region from point A to point B reprents the droplet of fuel from initial contact to final entry into the oxidizer. When the fuel droplet completely enters the oxidizer, the lar light can again fully reach phototransistor #1 as illustrated by the return of the signal to its initial reference level.
The respon of phototransistor #1 in the region from point B to C is reprentative of a still liquid to air interface. This is the region in which mixing is taking place between fuel and oxidizer by a combination of convection and diffusion between the two reactants. In Figure 5 region B-C reprent
s the time period when mass transfer limited kinetics are occurring. In other Figure 5: Typical Result Drop Test for UMDH/RFNA
