COMBUSTION BEHAVIOR OF SOL-GEL SYNTHESIZED ALUMINUM AND TUNGSTEN TRIOXIDE by DANIEL PRENTICE, B.S.M.E. A THESIS IN MECHANICAL ENGINEERING Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE IN MECHANICAL ENGINEERING Approved Michelle Pantoya Chairperson of the Committee Darryl James Jordan M. Berg Accepted John Borrelli Dean of the Graduate School May, 200
70
Embed
Daniel Prentice- Combustion Behavior of Sol-Gel Synthesized Aluminum and Tungsten Trioxide
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
COMBUSTION BEHAVIOR OF SOL-GEL SYNTHESIZED
ALUMINUM AND TUNGSTEN TRIOXIDE
by
DANIEL PRENTICE, B.S.M.E.
A THESIS
IN
MECHANICAL ENGINEERING
Submitted to the Graduate Faculty of Texas Tech University in
Partial Fulfillment of the Requirements for
the Degree of
MASTER OF SCIENCE
IN
MECHANICAL ENGINEERING
Approved
Michelle Pantoya Chairperson of the Committee
Darryl James
Jordan M. Berg
Accepted
John Borrelli Dean of the Graduate School
May, 200
ii
ACKNOWLEDGEMENTS
I would like to thank my advisor Dr. Michelle Pantoya for introducing me to this
exciting field of nanotechnology research. Thank you for being so generous in your
advice and in allowing me to be a part of the growing literature on nano-composite
energetic materials. I would like to acknowledge Dr. Alex Gash from Lawrence
Livermore National Laboratory for providing the aerogel materials studied as well as the
SEM micrographs and XRD analysis. Thank you to Dr. John Granier and Mr. Keith
Plantier for help in setting up and conducting experiments. Without your help, I would
not have known where to start. I would also like to acknowledge Mr. Dustin Osborne for
conducting the TGA work and Dr. Mark Grimson and Mr. Peter Jung for their help and
advice with the SEM work.
Thank you to my wife Sara for your support during this exciting but stressful time
in our lives. And finally, thank you to my parents for encouraging me to always do my
best and for supporting me in whatever I chose to do.
2. Calculated Andreev numbers and physical parameters for Al+WO3 composites at 31.7-34% TMD ............................................................................... 29
3. Normalized Andreev number for Aerogel 120 and Aerogel 400 in pellet and powder configuration............................................................................. 30
1. X-ray diffraction analysis of Aerogel 120 and Aerogel 400 ......................................7
2. Thermal gravimetric plots of Aerogel 120 and Aerogel 400 .....................................8
3. SEM micrographs of Aerogel 120 and Aerogel 400 .................................................9
4. Photograph of the open channel burn testing apparatus .......................................... 12
5. Schematic of the laser ignition and flame propagation diagnostic system ................................................................................................................... 14
6. Plot of open-channel combustion wave speed as a function of equivalence ratio for Aerogel 120 and Aerogel 400................................................ 19
7. Plot of combustion wave speed as a function of pellet density ............................... 21
8. Plot of ignition delay time as a function of pellet density ....................................... 22
9. Photograph of an Aerogel 400 pellet at 49.6% TMD.............................................. 27
10. Andreev number as a function of pellet oxidizer composition at a TMD range of 31.7-34%........................................................................................ 29
11. Calculated adiabatic flame temperature as a function of mass percent water in the oxidizer .............................................................................................. 32
12. Estimated drop in combustion wave speed as a function of mass percent water in the oxidizer.................................................................................. 33
13. Combustion wave speed as a function of oxidizer surface area and pellet density................................................................................................... 34
14. Semi-Log plot of combustion wave speed as a function of TMD for Aerogel 120 and Aerogel 400 ........................................................................... 37
15. Photograph of sample storage containers ............................................................... 46
16. Schematic of flame propagation apparatus............................................................. 48
17. Photograph of burn channel ................................................................................... 49
ix
18. Plot of combustion wave speed as a function of days aged..................................... 51
19. Thermal gravimetric plot of original and aged Aerogel 120 and Aerogel 400 ........................................................................................................... 56
1
CHAPTER I
INTRODUCTION
1.1 Traditional Thermites
Thermite reactions are characterized by an oxidation-reduction reaction between a
metal and a metallic oxide. The reactions are highly exothermic and typically self-
sustaining. Wang et al. [1] conducted a thorough literature review on typical thermite
applications ranging from welding [2, 3] to the synthesis of metals and alloys [4] or
ceramic and composite materials [5-7]. Other applications include a cost-effective
method for storing radioactive wastes [8], using gas-generating additives to create
thermites that could be used for the demolition of concrete [9] or in self-destruct
electrical components [10].
Because thermites are a mixture, an ideal balance of fuel and oxidizer components
can be attained resulting in high energy densities. The energy density is typically higher
than that of explosives because monomolecular energetic materials often have an
imperfect fuel-oxidizer balance. The reaction rate for a thermite is limited by the mass
transport between the reactants. In monomolecular energetic materials such as explosives,
the reaction rate is governed by chemical kinetics and thus explosives have a faster
energy release rate or greater power than thermites.
2
1.2 Nano-composite Thermites
Because thermite reactions are mass diffusion limited, decreasing the reactant
particle size decreases the diffusion distances and increases the reaction rate [11]. By
reducing the size of the reactants from micron-scale to nano-scale, flame propagation
rates (i.e. speeds) have been measured up to 1000 times faster than traditional thermites
[12]. Other studies have similarly confirmed that nanocomposite thermites exhibit greater
speeds than their micron scale counterparts [13-16]. In light of the increased speeds found
in nanocomposite thermites, researchers are beginning to develop materials that have
very high energy densities like traditional thermites but also have high energy release
rates (power) like explosives or other monomolecular energetic materials.
Researchers at Lawrence Livermore National Laboratory (LLNL) created a new
class of energetic materials using sol-gel chemistry [17-23]. Their method allows for not
only nanocomposite but actually nanostructured thermite production. They are able to
create a nano-scale structure with a skeletal sol-gel matrix composed of the oxidizer with
the fuel imbedded in the voids between the matrix [22]. The nanostructure allows for a
level of homogeneity and intimate mixing that is otherwise not easily achieved by simply
mixing nano particles. The sol-gel method offers the possibility to control the
composition, morphology and reactivity of the thermite at the nanometer scale [17].
LLNL researchers first synthesized Fe2O3 aerogel and xerogels [17]; then
synthesized composites such as Fe2O3+Al aerogel and xerogel thermites [18]; and further
expanded the technique to synthesize other oxides (Cr2O3, Al2O3, In2O3, Ga2O3, SnO2,
ZrO2, HfO2, Nb2O5, and WO3) [20, 21] and included silicon gas generators into the
3
thermites [24-26]. One result was that water or hydroxyl groups are an inherent
component in the sol-gel synthesis of these thermites [17, 20]. Plantier et al. conducted
combustion experiments using LLNL sol-gel derived Fe2O3 aerogel and xerogel and
compared it to commercially available nanoscale Fe2O3 [27]. They measured the
combustion wave speeds in open and confined burning experiments and showed
relatively poor burning performance of the aerogel and xerogel Fe2O3 compared to the
commercially available nanopowder. This was attributed to the high amount of impurities
(hydroxyl groups and organic contamination) found in the solids. New aerogel and
xerogel powders were then calcined to 410° C to remove impurities and results showed
the largest improvement in combustion wave speed from ~10 m/s to over 900 m/s in
confined burning conditions.
1.3 Advantages of Tungsten Trioxide
For ordnance applications, desirable properties for the mixture include a high
adiabatic flame temperature and large enthalpy of reaction [28]. Tungsten trioxide (WO3)
combined with aluminum has a high adiabatic flame temperature (5544 K [41]) and large
enthalpy of reaction (3801 cal/cc [41]). Another important property when high energy
density is desired is high material density. Tungsten trioxide has high material density
(7.20 g/cc compared with 5.25 g/cc for Fe2O3). Furthermore, oxidizers such as WO3 and
MoO3 sublime at relatively low temperatures unlike Fe2O3 which remains in the solid
phase. The gas generated during sublimation could facilitate a diffusion limited reaction.
Also, Balakir et al. found that oxidizers with high vapor pressure such as WO3 gave the
4
fastest rates of combustion [30]. For these reasons, this study focused on characterizing
the combustion behavior of Al and WO3.
There were three objectives in this study of nano-Al and WO3 combustion. The
first objective was to determine the optimum stoichiometry. Because it has been shown
that varying the bulk density of the thermite has a significant affect on the combustion
behavior of Al and MoO3 [29], the second objective was to determine how the bulk
density affects the combustion wave speed of nano-Al and WO3. The final objective was
to identify the dominant mechanisms controlling the combustion behaviors corresponding
to the different thermite mixes in loose powder and pressed form.
Optimum stoichiometry was determined by varying the fuel/oxidizer ratio of the
mixtures where the ratio that exhibits the fastest combustion wave speed is deemed
optimum. The thermites were composed of nano-scale aluminum powder mixed with four
types of WO3 oxidizers. The oxidizers were calcined and non-calcined WO3 aerogels
prepared at LLNL, and commercially available nano and micron-scale WO3 powders.
These experiments were performed by burning loose powders of various fuel/oxidizer
ratios in an open channel and measuring the combustion wave speed and flame
propagation behavior via high-speed photography and imaging diagnostics. Bulk density
combustion behaviors were determined by preparing samples of various densities and
examining flame propagation behavior also using high-speed imaging diagnostics.
Identifying the dominant mechanisms controlling the combustion behaviors was
accomplished by using an Andreev numerical analysis and by analyzing the visual data
recorded for the combustion wave speed experiments.
5
CHAPTER II
EXPERIMENTAL
2.1 Sample Preparation
The WO3 oxidizers vary according to manufacturing methods and physical
properties described in Table 1. The Al fuel particles were 80 nm average diameter
supplied by Nanotechnologies Inc. (Austin, Texas) and consistently used for all mixtures
in this study. An alumina shell that is roughly 4 nm thick encapsulates the core Al
particle such that this powder has an active Al content of 84 %. Active Al is defined as
the portion of powder not in the form of Al2O3.
Table 1. Oxidizer Properties Oxidizer Name Oxidizer Description (WO3) Particle Surface Area % WO3
The most interesting and important finding from this study is the observed
transition from convection based heat transfer to conduction based heat transfer as
density increases for the aerogel samples. Figure 14 is a semi-logarithmic plot of the
combustion wave speed of the Aerogel 120 and Aerogel 400 as a function of theoretical
maximum density. It was formed by combining the open burn data found in Fig. 6 and
the pellet burn data for the aerogels found in Fig. 7.
37
Figure 14 Semi-Log plot of combustion wave speed as a function of TMD for Aerogel 120 and Aerogel 400
The Aerogel 120 produced a higher combustion wave speed than the Aerogel 400
in the low-density, open channel configuration. Conversely, in high-density, pressed
pellet configuration, the Aerogel 400 consistently produced higher combustion wave
speeds than the Aerogel 120. In the open channel configuration, convection heat transfer
is the dominant mechanism as confirmed by Andreev analysis as seen in Table 3. Low
normalized Andreev numbers correspond to conduction while high numbers correspond
to convection. When convection is the dominant mechanism, the most important
parameter affecting the combustion wave speed is the particle diameter. Aerogel 120 is
composed of smaller particles than Aerogel 400, therefore the Aerogel 120 produces
faster combustion wave speeds.
As density is increased, the air pockets found in the material are removed which
lowers the convective effects. Andreev analysis confirms that conduction is the dominant
heat transfer mechanism for the thermites in pellet form (Table 3). The bonded water
38
found in Aerogel 120 impedes the energy transport through conductive mechanisms. For
this reason, the Aerogel 400, which does not contain bonded water impurities,
outperforms the Aerogel 120 in pellet form.
The transition from convection to conduction as density is increased causes the
important material parameters to be different depending on the material density. In
essence, at the low densities found in powders, convection is dominant and the particle
diameter is the most important parameter. At high densities found in pellets, the purity of
the oxidizer outweighs the benefits of smaller particle diameter because conduction is the
dominant mechanism.
Unfortunately, Fig. 14 merely suggests the previous trends. The original set of
experiments did not test the aerogels at densities other than the natural density produced
through mixing. It would be desirable to find the combustion wave speed for several
densities slightly above the natural powder density of each aerogel Also, repeatability
was not conducted due to the limited amount of Aerogel 120 and Aerogel 400 available
and the uncertainty in the total number of experiments that would eventually be
conducted. At the end of this study, when it was apparent that more data would be
beneficial for low-density combustion data, new experiments were conducted. Even
though new nano-Al was used, the aerogels had aged over a period of 22 months which
greatly decreased the performance of the Aerogel 120 and Aerogel 400 (see Chapter VII).
Therefore, new data was unable to be obtained for low-density combustion of Aerogel
120 and Aerogel 400. This causes the transition phenomenon to be suggested but not
proven.
39
CHAPTER VI
CONCLUSIONS
An optimum equivalence ratio of 1.4 was found for thermites composed of nano-
Al and Aerogel 120 or Aerogel 400 oxidizer. The slightly fuel rich composition is
explained by the increased thermal transport properties found when more nano-Al is
used. This pre-heats the reactants and increases the combustion wave speed. Because
thermite reactions are diffusion limited, both the Aerogel 120 and Aerogel 400 exhibit
higher combustion wave speeds than µm WO3 in an open-channel burning configuration.
Convection was determined to be the dominant mechanism for the advance of the
combustion wave in the loose powder open-channel burn experiments. This was
explained by the fact that WO3 sublimes below the adiabatic flame temperature of the
combustion reaction. This causes gas to be expelled during the reaction creating
convection in the form of the violent ejection of burning particulate in all directions.
Andreev analysis further suggests convection dominance. This convection causes the
Aerogel 120 to burn faster than the Aerogel 400 even though the Aerogel 400 has fewer
impurities. The large-scale effect of convection combined with the smaller particle
diameter of Aerogel 120 outweighs the purity benefits of Aerogel 400 in loose powder
form.
The increase in Aerogel 400 combustion wave speed observed as density
increases was attributed to the associated increase in thermal conductivity. Aerogel 400
composites exhibited faster combustion wave speeds at all pellet densities than the
40
Aerogel 120, nm WO3 or µm WO3. Andreev analysis showed that conduction is the
dominant mechanism for combustion in pellet form. This explains why the Aerogel 400
outperformed the Aerogel 120 which is the opposite trend seen in loose powder form.
Bonded water impurities act as a heat sink and impede thermal transport through
conduction. Therefore, even though the Aerogel 400 had larger a particle diameter than
Aerogel 120, its lack of bonded water causes it to perform better. The Aerogel 400
outperformed the nm WO3 and µm WO3 because of its smaller particle diameter and
therefore reduced diffusion distance over the other oxidizers. It was noted that radiation
effects might play a role in the ignition sensitivity and combustion wave speed of these
materials. To date, little research has focused on the radiation effects.
41
CHAPTER VII
ALUMINUM AND TUNGSTEN TRIOXIDE AGING EFFECTS
A preliminary study was conducted to quantify aging effects (i.e. decreased
combustion wave speeds) that had been observed qualitatively. It was know that
aluminum oxidizes in air over time but the length of time and associated environmental
factors were unknown or not quantified. The goal was to obtain a length of time for
which a mixed thermite could be stored before its performance was significantly reduced.
7.1 Aluminum and Tungsten Trioxide Aging Study
7.1.1 Introduction
One of the unique properties of aluminum is that its surface will naturally oxidize
with air to form a layer of alumina (Al2O3). There are a few studies that investigate the
oxidation of nanosized aluminum powders [12 and 37]. Aumann et al. [12] showed that
nano-Al powder consists of an aluminum core with a shell of amorphous alumina. They
said that the shell of alumina is what keeps the mixed thermite from reacting until an
external energy source stimulates the reaction. They exposed the nano-Al to a controlled
amount of oxygen gas at various temperatures for a short time. This showed that the
oxide growth was temperature dependent for time periods on the scale of hours. They
observed no growth in the alumina shell for temperatures less than 350°C. Pesiri et al.
[37] also showed that the oxide shell thickness can be controlled by carefully introducing
42
oxygen gas to the aluminum nanopowder during production. They showed that the nano-
Al powder would still react with the oxygen in the atmosphere even with the protective
alumina layer. This reaction was claimed to increase the thickness of the alumina shell
for 3 months of continuous exposure at which point it stabilizes. Finally, they showed
that the nano-Al powder is very sensitive to any water vapor in the air and will absorb 5%
wt. water from the atmosphere within minutes. Therefore, they recommend handling and
storing the nano-Al powder under dry Nitrogen gas.
Because of the nano-Al’s sensitivity to air and water exposure, aging studies have
been conducted to quantify the effects of exposure to the atmosphere [38-40]. Walter et
al. [38] not only studied the effect of aging on nano-Al, but also on the oxidizers nano-
MoO3 and nano-CuO. They showed that the surface area of nano-MoO3 decreased
twofold from 10-12 days exposure to light and air. The nano-CuO showed only a slight
decrease in particle size with exposure to light and air. The nano-Al metal content
decreased by 50% wt. from two years of exposure to air. This showed that for a thermite
of Al and MoO3, the aging effects of the MoO3 would have a greater impact sooner than
the aging effects of the aluminum. Bulian et al. [39] and Puszynski [40] investigated the
aging of nano-Al powder exposed to air with high moisture content. Bulian not only
looked at the physical properties of the aluminum, but also the effect on the open burn
rate of the exposed aluminum mixed with CuO. Bulian showed very large speed
decreases when the aluminum was exposed to 97% relative humidity air for less than 3
days. After 3 days, the thermite wouldn’t react at all. The major decrease in speed was
even observed when the decreasing reactive aluminum content of the exposed nano-Al
43
was accounted for by increasing the nano-Al mixed with CuO. Bulian attributed the
decrease in speed to two factors: the increase in the Al2O3 thickness and the increase in
the water content in the aluminum. Puszynski further showed that a large reduction in the
percent reactive aluminum content was seen in nano-Al exposed to air with as low as
43% R.H. The time for aluminum content reduction of 50% or greater was on the order of
5-40 days. Finally, both Bulian and Puszynski showed that the effect of water vapor on
nano-Al powder could be stopped by coating the aluminum with oleic acid or Dow
Corning Z-6124 silane. Additionally, the coatings increased the open burn speeds when
the coating weight percent was five or less due to the coating increasing the mixing of the
aluminum and the oxidizer.
Walter et al. [38] only studied the effect of aging on the nano-Al, nano-MoO3 and
nano-CuO separately. The effect of aging on an already mixed thermite was not
investigated. Furthermore, there are other relevant oxidizers, such as WO3, which need to
be investigated. Bulian et al. [39] and Puszynski [40] also chose to study the aging of the
nano-Al and not the thermite itself. Bulian only investigated the effect of very high
relative humidity air exposure and Puszynski didn’t conduct burn speed measurements on
the thermites containing aluminum exposed to lower relative humidity air.
Often times, in a lab environment, it is impractical to test a thermite immediately
after it has been mixed. The thermites are stored in non-airtight plastic or glass
containers. Because of the unknown effect of the exposure of the thermite to the
atmosphere, there is no fixed rule for how long the thermite is good after it has been
mixed. This study will focus specifically on the effect of atmospheric air exposure on
44
thermites consisting of nano-Al mixed with nano-WO3. Separate mixes will be exposed
to the air in an indoor lab environment and sealed in a container for a period of four days.
Combustion wave speed measurements of the loose powders will be made from samples
of the mixes at various times of exposure by using a high-speed camera. This will extend
previous studies by being the first study, to the author’s knowledge, to investigate the
effect of aging on the mixed thermite itself. Furthermore, it will investigate mixes that
include the commonly used oxidizer WO3. Finally, it will quantify the change in
performance of thermites exposed to atmospheric conditions common to a lab
environment and explain the relevant phenomena using previously published studies.
7.1.2 Experimental Setup
The aluminum powder was obtained from Nanotechnologies (Austin, TX) and has
a 120 nm average particle diameter. The average particle diameter is calculated by the
manufacturer using a gas adsorption analyzer using Brauner, Emmett, and Teller (BET)
theory to measure the surface area. The diameter is calculated by assuming the particles
are spherical. These Al particles are passivated with an alumina (Al2O3) shell that is
roughly 4 nm thick and encapsulates the core Al particle. Based on thermal analysis
(Base Hydrolysis Method), the manufacturer determined that this powder has an active
Al content of 81%. Active Al is defined as the portion of powder not in the form of
Al2O3. The previous material specifications were measured on June 25, 2004 for the
particular batch of Al used in this study. Also, the Al used was originally opened on July
14, 2004 after being received from the manufacturer packed in argon gas. After being
45
opened, it was sealed in an airtight container until mixed with the oxidizer on October 18,
2004.
The tungsten trioxide (WO3) powder was purchased from Aldrich (St. Louis, MO)
and has an average particle diameter of 49.6 nm. The manufacturer supplied the particle
diameter data using an unspecified method. Also, because no purity data was provided, it
was assumed that the WO3 was 100% pure. This assumption was valid because the WO3,
unlike the Al, doesn’t have a passivation shell or any other characteristic which would
cause a loss of purity. The WO3 used in this study was originally opened from the
manufacturer on September 9, 2004 and again stored in an airtight container until mixed
with the fuel on October 18, 2004.
It was calculated that approximately 6 grams of Al+WO3 mixture was needed.
Due to safety considerations, three separate but chemically equivalent batches of 2 grams
each were prepared as described below. Each batch was prepared by combining the
powders in controlled proportions in a suspension solution of isopropanol. The powders
and solution were mechanically mixed using an ultra-sonification process. This process
allowed for a more homogeneous mixture by breaking up large agglomerates. The
mixture and solution was heated at a low temperature in a glass dish to evaporate the
isopropanol solution. After all of the isopropanol had evaporated, the composite powder
was scraped from the dish and placed into a vial to be immediately combined with the
other batches. The three batches were mechanically mixed and then separated into two
equal amounts. One half was placed into a 4 oz. polypropylene vial with a lid from Tyco
46
Healthcare Group LP. (Mansfield, MA) and the other half was placed into a glass dish
open to the ambient laboratory environment.
Both of the thermite samples, one open to the air and one enclosed in the vial,
were stored next to each other in the ambient lab environment as seen in Fig. 15. The
temperature and relative humidity was measured and recorded every 24 hours using a
digital hygro-thermometer calibrated by Control Company (Friendswood, TX) on June 4,
2001. The temperature and relative humidity measurements were averaged for the five
days of aging to get an overall average temperature of 23.6°C, and an overall average
relative humidity of 38%.
Figure 15 Photograph of sample storage containers
47
The performance of the thermite was quantified by measuring the combustion
wave speed of loose powder burning in a channel exposed to an ambient air environment.
The schematic of the full apparatus used to measure the combustion wave speed is found
in Fig. 16. A square channel 0.3175 cm wide and 10 cm long was used to burn the loose
powders in as seen in Fig. 17. The channel was made of acrylic so that the light from the
thermite combustion could pass through. The powder was ignited by a spark gap that
passed through the channel. A trigger fired the spark gap ignition and triggered the
camera at the same time. A Phantom IV high-speed camera from Vision Research
(Wayne, NJ) was used to record the flame propagation. The camera interfaced with a
computer that used Vision Research software to post process the videos. By establishing a
reference length, the software determines speed based on a distance between sequential
time frames.
48
Spark Gap
Ignition
Open Burn Channel
Trigger
High Speed Camera
PC with Imaging
Software
Figure 16 Schematic of flame propagation apparatus
49
Figure 17 Photograph of burn channel
Every day for five days, two samples from the vial and two samples from the tray
were burned and the combustion wave speeds were recorded. Each sample was 250 mg.
Two samples were burned daily to obtain repeatability data. It was observed that the
combustion wave speeds were not constant throughout the burn. Because the speed was
not constant, either the measurement time or the measurement length was required to be
kept constant to achieve comparable speed measurements. It was impractical to keep the
length viewed in the videos constant because the camera was moved daily for other
experiments so the time was held constant. The average speed was measured between the
frame where first light was visible and 40 frames later. Because the camera ran at 32000
50
frames/sec, 40 frames is equal to 1.25 ms. The standard deviation was calculated for all
of the repeated tests.
7.1.3 Results
Table 4 presents the combustion wave speeds recorded over the four-day period.
It shows the individual and average speeds starting on day 0 until day 4. It can be seen
that one repeatability test was conducted for each sample.
Table 4 Combustion Wave Speeds Velocities (m/s)
Days Aged 0 1 2 3 4
Vial Burn 1 36.176 33.591 30.843 31.168 31.446
Vial Burn 2 31.734 30.305 33.260 31.751 31.445
Avg. 33.955 31.948 32.052 31.460 31.446
Std. Dev. 3.141 2.324 1.709 0.412 0.001
Tray Burn 1 37.446 33.042 30.842 29.951 27.936
Tray Burn 2 33.878 36.073 33.829 33.242 32.893
Avg. 35.662 34.558 32.336 31.597 30.415
Std. Dev. 2.523 2.143 2.112 2.327 3.505
In Fig. 18, the average combustion wave speed for the tray and vial mixes is
plotted as a function of the number of days the mix has aged. The error bars represent a
standard deviation between the repeated tests for each sample. The points on the plot
were slightly separated for each day to show the error bars more clearly. All tests were
actually conducted within 30 minutes of each other on each day so the daily points would
truly be on top of one another.
51
26
28
30
32
34
36
38
40
-0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
Days Aged
Co
mb
usti
on
Wave S
peed
(m
/s)
Vial Contained
Tray Contained
Figure 18 Plot of combustion wave speed as a function of days aged
7.1.4 Discussion
Figure 18 shows that there is a decrease in the combustion wave speed for
thermites that have aged for a period of four days. This result was expected because it
was already known that Al particles form thicker alumina shells the longer they are
exposed to air [12 and 37] which reduces the amount of active aluminum or fuel. Using
the data from Table 4, it was calculated that over a period of 4 days, the sample stored in
the vial had a reduction in combustion wave speed of 7.4% and the sample stored in the
tray open to the air had a reduction in combustion wave speed of 14.7%. This shows that
it is certainly better to place a mixed thermite in some type of container than to allow it to
52
be exposed to air freely. Even though the vial was not advertised as air tight, it did have
enough benefit to reduce the speed degradation by half.
Bulian et al. [39] showed a reduction in the combustion wave speed of over 50%
within two days exposure to air at 40°C and 97% relative humidity. This study shows that
samples exposed to air at a lower temperature and a lower relative humidity experience a
much smaller reduction in the combustion wave speed performance. Bulian et al. [39]
also theorized that the decrease in combustion wave speed was due to the increase in the
alumina barrier on the aluminum particles and increase in the physisorbed and
chemisorbed water. This study shows that the increase in the water content in the
aluminum plays a greater role than the increase in the alumina shell. Far less reduction in
wave speed was seen at double the exposure time at similar dry air exposures but reduced
water vapor exposures.
Of course, Bulian et al. [39] also used copper oxide as the oxidizer and mixed the
thermite after the aluminum exposure. This suggests that a thermite that is already mixed
is less susceptible to environmental exposure. This could be due to the fact that the
oxidizer blocks some of the aluminum particles contact with the environment. Tungsten
trioxide doesn’t experience an effect like the alumina shell formation when exposed to air
so the air exposure by itself would not affect the WO3 as severely as the aluminum.
Further study is needed to prove this theory. Chemical analysis to determine the chemical
composition of the thermite at different stages of exposure would show whether the
alumina formation is hindered in a premixed thermite.
53
There were several problems encountered with this study. The greatest problem
encountered was in measuring an accurate combustion wave speed. Because the wave
speeds did not vary by a large degree, very accurate measurements were needed to find a
trend in the data. Because the combustion wave was identified by visual methods,
problems arose from the brightness of the thermite when burnt. The brightness can be
reduced mechanically and electronically by filters and image processing but reflections
off of the burning channel and the smoke from the combustion made finding the
combustion wave difficult. Other difficulties came from the combustion waves being so
fast in a Al+WO3 thermite. There was only a limited amount of data that could be
recorded with the equipment on hand. The camera recorded at 32,000 frames/sec which
was not ideally fast enough for these very high-speed thermites. Finally, the Al+WO3
thermite exhibited non-constant combustion wave speeds. Other thermites such as
Al+Fe2O3 exhibit much more constant combustion wave speeds when they are burnt in
loose powder form. All of these factors, which would be minor if examining greater
speed variations, become large sources of error when such accuracy is needed. This is
represented by the relatively large error bars seen in Fig. 18.
This study showed that there is a reduction in the degradation of a thermites
performance when it is mixed before being exposed to the environment and when the
ambient conditions have low relative humidity. This produces a need for further
investigation into these phenomena. Besides combustion wave speeds, data about the
thermites surface area is also needed. This would more directly compare to the work that
Bulian et al. [39] and Puszynski [40] have completed. Also, conducting chemical analysis
54
on the thermites at different levels of environmental exposure would show whether or not
the oxidizer impedes the alumina growth. Scanning electron microscope (SEM) pictures
might also provide some insight to the oxidizer protecting factor. Finally, studies which
are conducted for longer time periods with more accurate measurement techniques such
as photo diodes could be conducted.
7.1.4 Conclusions
It has been shown that Al+WO3 thermites exposed to ambient laboratory air
found in an arid climate will reduce the combustion wave speed of the loose powder by
15% within four days. If the thermite is stored in a standard laboratory vial, the
combustion wave speed is reduced by 7% within four days. This shows that even an
imperfect, or not airtight, vial should be used to store mixed thermites. Also, if the study
being conducted with the thermite requires high accuracy, it is necessary to burn the
thermite on the same day as it is mixed.
It has also been shown that a mixed thermite exposed to relatively dry air
experiences a far smaller amount of performance degradation compared to an unmixed
thermite exposed to high humidity air. Therefore, the amount of water vapor in the air
plays a very important role in the rate of decreasing performance. Further study is needed
to show if the mixed thermite may resist environmental exposure to a greater degree than
an unmixed thermite.
55
7.2 Long Term Aerogel Aging
In an effort to obtain new data to reinforce the trends seen in Fig. 14, a new batch
of Aerogel 120 and Aerogel 400 was mixed 22 months after the initial batch. New nano-
Al was used in the mixes but it was required to use the original supply of both aerogel
oxidizers. Open-channel combustion burns for the new Aerogel 120 and Aerogel 400
were conducted using the same procedures described in Section 2.1.3 and 2.3. Both
thermites were mixed at Φ=1.4 as before. The original Aerogel 120 had a combustion
wave speed of 490 m/s. The new Aerogel 120 sample had a combustion wave speed of 8
m/s, a decrease of 98%. The original Aerogel 400 sample had a combustion wave speed
of 310 m/s. The new Aerogel 400 sample had a combustion wave speed of 29 m/s, a
decrease of 91%. Because the aluminum was new, this shows that the aerogels had
almost totally degraded over a period of less than two years.
Considering the preliminary aging study mentioned previously, the most likely
degrading factor was the absorption of water through water vapor in the atmosphere. An
aging study of Al and MoO3 showed that the aging effects of MoO3 were greater than the
aging effects of Al [38]. Thermal gravimetric analysis was conducted on the aged aerogel
samples as shown in Fig. 19. The original TG analysis is plotted along with the new
analysis. It can be clearly seen that the aged oxidizers experience a greater reduction in
mass as they are heated. This confirms the presence of additional water in the aged
samples.
56
Figure 19 Thermal gravimetric plot of original and aged Aerogel 120 and Aerogel 400
57
REFERENCES
[1] Wang, L. L.; Munir, Z. A.; Maximov, Y. M.; “Thermite reactions: their utilization in the synthesis and processing of materials,” J. of Materials Science, v. 28, pp. 3693-3708 (1993)
[2] Jha, B.; Doctoral Dissertation, Illinois Institute of Technology, December (1989) [3] Key, A. J.; Aust. Weld. J., v. 15, Autumn (1985) [4] Goldschmidt, H.; Vautin, C.; J. Soc. Chem. Ind. v. 19, pp. 543 (1898) [5] Merzhanov, A. G.; “Combustion and Plasma Synthesis of High-Temperature
Materials,” edited by Z. A. Munir and J. B. Holt, VCH, New York, pp. 204-10 (1990)
[6] Munir, Z. A.; Anselmi-Tamburini, U.; Mater. Sci. Repts., v. 3, pp. 277 (1989) [7] Odawara, O.; Sosei to Kako, v. 28 (322), pp. 1109 (1987) [8] Spector, M. L.; Suriani, E.; Stuken-Brocker, G. L.; Ind. Eng. Chem. Process Des.
Dev., v. 7[1], pp. 117 (1968) [9] Mohler, J. H.; Halcomb, D. L.; Begeal, D. R.; “Proceedings of the 15th
International Pyrotechnics Seminar,” I.I.T Research Inst., Chicago, IL, pp. 711-20 (1990)
[10] Kelster, F. Z.; Smelker, G. S.; US Pat. 3742 129, 26 June (1976) [11] Dubrovin, A. S.; Kuznetsov, V. L.; Ezikov, V. I.; Chirkov, N. A.; Rusakov, L. N.;
Russ. Metall. (Engl. Trans.), v. 5, pp. 56 (1968) [12] Aumann, C. E.; Skofronick, G. L.; Martin, J. A.; J. Vac. Sci. Technol. B, v. 13, pp.
1178 (1995) [13] Granier, J; Pantoya, M; “Laser Ignition of Nanocomposite Thermites,”
Combustion and Flame, v. 138, pp. 373-383 (2004) [14] Bockmon, B. S.; Pantoya, M. L.; Son, S. F.; Asay, B. W.; Mang, J. T.;
“Combustion wave speeds and Propagation Mechanisms of Meta-Stable Intermolecular Composites,” Journal of Applied Physics, v. 96 (2004)
58
[15] Ivanov, G. V.; Tepper, F.; “ ‘Activated’ Aluminum as a Stored Energy Source of Propellants,” Challenges in Propellants and Combustion 100 Years after Nobel, edited by K. K. Kuo et al., Begell house, New York, pp. 636-645 (1997)
[16] Moore, K; “Combustion Behaviors of Bimodal Aluminum Size Distributions in
Thermites,” Master’s Thesis, Texas Tech University, May (2005) [17] Gash, A.E.; Simpson, R.L.; Tillotson, T.M.; Satcher, J.H.; and Hrubesh, L.W.;
“Making Nanostructured Pyrotechnics In A Beaker,” Proceedings of the 27th International Pyrotechnics Seminar, pp.41-53, Grand Junction, Colorado (2000)
[18] Tillotson, T. M.; Gash, A. E.; Simpson, R.L.; Hrubesh, L. W.; Satcher Jr., J. H.;
Poco, J. F.; “Nanostructured energetic materials using sol-gel methodologies,” Journal of Non-Crystalline Solids (2001)
[19] Simpson, R. L.; Tillotson, T. M.; Satcher, J. H.; Hrubesh, L. W.; Gash, A. E.;
“Nanostructured Energetic Materials Derived from Sol-Gel Chemistry,” Int. Annu. Conf. ICT (31st Energetic Materials), Karlsruhe, Germany, June 27-30 (2000)
[20] Gash, A. E.; Tillotson, T. M.; Satcher, Jr., J. H.; Poco, J. F.; Hrubesh, L. W.;
Simpson, R.L; “Use of Epoxides in the Sol-Gel Synthesis of Porous Iron(III) Oxide Monoliths from Fe(III) Salts”, Chemistry of Materials (2001)
[21] Gash, A. E.; Tillotson, T. M.; Satcher, Jr., J. H.; Hrubesh, L. W.; Simpson, R. L.;
“New Sol-Gel Synthetic Route to Transition and Main-Group Metal Oxide Aerogels Using Inorganic Salt Precursors”, Journal of Non-Crystalline Solids (2001)
[22] Gash, A. E.; Satcher, Jr., J. H.; Simpson, R. L.; Clapsaddle, B. J.; “Nanostructured
Energetic Materials with Sol-gel Methods,” Mat. Res. Soc. Symp. Proc., v. 800 (2004)
[23] Tillotson, T. M.; Simpson, R. L.; Hrubesh, L. W.; Gash, A. E.; Thomas, I. M.;
Poco, J. F.; “Nanostructured Energetic Materials Using Sol-Gel Methodologies,” International Symposium on Aerogels, Albuquerque, NM., October 8-11 (2000)
[24] Clapsaddle, B. J.; Zhao, L.; Gash, A. E.; Satcher, Jr., J. H.; Shea, K. J.; Pantoya,
M. L.; Simpson, R. L.; “Synthesis and Characterization of Mixed Metal Oxide Nanocomposite Energetic Materials”, Mat. Res. Soc. Symp. Proc., v. 800 (2004)
59
[25] Clapsaddle, B. J.; Gash, A. E.; Satcher Jr., J. H.; Simpson, R. L.; “Silicon oxide in an iron(III) oxide matrix: the sol-gel synthesis and characterization of Fe-Si mixed oxide nanocomposites that contain iron oxide as the major phase,” Journal of Non-Crystalline Solids, v. 331, pp. 190-201 (2003)
[26] Clapsaddle, B. J.; Sprehn, D. W.; Gash, A. E.; Satcher Jr., J. H.; Simpson, R. L.;
“A versatile sol-gel synthesis route to metal-silicon mixed oxide nanocomposites that contain metal oxides as the major phase,” Journal of Non-Crystalline Solids (2004)
[27] Plantier, K. B.; Pantoya, M. L.; Gash, A. E.; “Combustion Wave Speeds of
Nanocomposite Al/Fe2O3: The Effects of Fe2O3 Particle Synthesis Technique,” Combustion and Flame, Jan. (2005)
[28] Prakash, A.; McCormick, A. V.; Zachariah, M. R.; “Synthesis and Reactivity of a
Super-Reactive Metastable Intermolecular Composite (MIC) Formulation of Al/KMnO4”, Advanced Materials, v. 17, pp. 900 ( 2005)
[29] Pantoya, M. L.; Granier, J. J.; “Combustion Behavior of Highly Energetic
Thermites: Nano versus Micron Composites,” Propellants, Explosives, Pyrotechnics, v. 30, no. 1, pp. 53-62 (2005)
[30] Balakir, E. A., Bushuev, Yu. G., Bareskov, N. A., Kosyakin, A. E., Kudryavtsev,
Yu. V., Fedorova, O. N., ibid. v. 11, no. 36 (1975) [31] Yang, Y., Wang, S., Zhaoyong, S., Dlott, D., J. Appl. Phys. V. 95, no. 7, pp.
3667-3676 (2004) [32] Bobolev, V. K., Margolin, A. D., Chuiko, S. V., “Stability of Normal Burning of
Porous Systems at Constant pressure,” Combustion, Explosion and Shock Waves, v. 2, no. 4, p. 24-32 (1966)
[33] Kaviany, M., Principles of Heat Transfer in Porous Media, Springer-Verlag, New
York (1991) [34] Son, S. F., Asay, B. W., Busse, J. R., Jorgensen, B. S., Bockmon, B., Pantoya, M.,
“Reaction Propagation Physics of Al/MoO3 Nanocomposite Thermites,” Proceedings of the 28thInternational Pyrotechnics Symposium (2001)
[35] Dubrovin, A. S., Slepova, L. V., Kuznetsov, V. L, Combust. Explos. Shock Waves
(Engl. Trans.), v. 6, pp. 60 (1970) [36] Brinker, C. J., Sherer, G. W., Sol-Gel Science, Academic Press, Inc., Boston
(1990)
60
[37] Pesiri, D., Aumann, C. E., Bilger, L., Booth, D., Carpenter, R. D., Dye, R., O’Neill, E., Shelton, D., Walter, K. C., “Industrial Scale Nano-Aluminum Powder Manufacturing,” Journal of Pyrotechnics, 19 (2004)
[38] Walter, K. C., Aumann, C. E., Carpenter, D., O’Neill, E. H., Pesiri, D. R.,
“Energetic Materials Development at Technanogy Materials Development,” Proc. Mat. Res. Soc. Symp., v. 800 (2004)
[39] Bulian, C. J., Kerr, T. T., Puszynski, J. A., “Ignition Studies of Aluminum and
Metal Oxide Nanopowders,” Proc. International Pyrotechnics Seminars, 31 (2004)
[40] Puszynski, J. A., “Reactivity of Nanosize Aluminum with Metal Oxides and
Water Vapor,” Proc. Mat. Res. Soc. Symp., v. 800 (2004) [41] Fischer, S. H., Grubelich, M. C., “Theoretical Energy Release of Thermites,
Intermetallics and Combustible Metals,” Proceedings of the 24th International Pyrotechnics Seminar, (1998)
[42] Armstrong, R. “Models for gasless combustion in layered materials and random
media,” Combust. Sci. Technol. v. 71, no. 155, pp 74 (1990) [43] Sun, J., Pantoya, M. L., Simon, S. L., “Dependence of Size and Size Distribution
on Reactivity of Aluminum Nanoparticles in Reactions with Oxygen and MoO3,” Journal of Thermal Analysis and Calorimetry, In Press
PERMISSION TO COPY
In presenting this thesis in partial fulfillment of the requirements for a master’s
degree at Texas Tech University or Texas Tech University Health Sciences Center, I
agree that the Library and my major department shall make it freely available for research
purposes. Permission to copy this thesis for scholarly purposes may be granted by the
Director of the Library or my major professor. It is understood that any copying or
publication of this thesis for financial gain shall not be allowed without my further
written permission and that any user may be liable for copyright infringement.
Agree (Permission is granted.)
___Daniel Prentice_______________________________ 3-31-2006______ Student Signature Date Disagree (Permission is not granted.) _______________________________________________ _________________ Student Signature Date