CP620, Shock Compression of Condensed Matter - 2001
edited by M. D. Furnish, N. N. Thadhani, and Y. Horie
2002 American Institute of Physics 0-7354-0068-7
EFFECT OF METAL PARTICLE SIZE ON BLAST PERFORMANCE
OF RDX-BASED EXPLOSIVES
Jeffery J. Davis and Philip J. Miller
Naval Aviation Science and Technology Office, Research and Technology Department, Code 4T4330D,
Naval Air Warfare Center Weapons Division, 1 Administration Circle, China Lake, CA 93555-6100
Abstract. This paper discusses the role that aluminum particle size has on explosives blast performance.
Tests were performed using a small sealed chamber and an open-ended shock tube. Three explosives
were tested and the results presented. The Al particle size examined was 20 microns and 150 nanometers
in a pressed PBXN-109 analog composition. (PBXN-109 was also tested.) A noticeable difference in the
internal blast pressure was observed between the 20 jam and 150 nm Al in the sealed chamber but not in
the shock tube. The chamber results compare favorably with modeling work performed.
This reaction is diffusion-limited if the oxide
coating remains in contact with the unburned Al. As
the particle size is decreased, the speed of that
reaction increases and the ability for complete
conversion to the oxide is increased.
The effect of using smaller sized Al should be
observed in the increased pressure of the initial
blast wave. As the reaction is allowed to continue, it
has been surmised that the larger Al will burn late,
while the smaller Al will have been consumed.
Thus, the total energy is the same while the energy
release profile is different.
INTRODUCTION
The addition of metal to energetic materials has
been shown to be effective in increasing the blast
characteristics of formulations. The role that metals
play in affecting the reaction of energetic material
has been studied for decades. The accepted reaction
mechanism of the metal is that the particle reacts
after the detonation wave has swept past, releasing
its energy in a much longer time frame and thus
creating a late time reaction and a longer pressure
profile. The use of jum-sized Al additive has
become the standard for explosive formulation
throughout the Department of Defense (DOD). The
Navy's standard explosive for blast is PBXN-109,
which contains RDX and 20 jam Al.
While metals have been studied for decades,
the availability of small particle metals (particularly
Al) over the past couple of years has prompted
researchers to examine the effect that particle size
has on energetic materials (1,2,3). The dramatic
results obtained by increasing the burn rate of
propellants have fueled interest from the explosive
community. The main metal reaction pathway is the
oxidation of the metal.
2 Al + 3/2 O2 -> A12O3 + heat
EXPERIMENT
To test blast performance of the explosives, a
closed chamber test was used to evaluate the initial
blast effects from explosives. The steel cylindrical
chamber used was 10 inches (inner diameter) by
16 inches (high) with 1-inch-thick walls. A pressure
gauge (Kistler Model 607-C3 100,000 psi) was
mounted at the top of the chamber in a Teflon
mount (Fig. 1). (Teflon is used to try to isolate the
gauge from the shock traveling in the chamber
wall.) The amount of energetic material used was
950
2-3 grams and were 1/2- x 1/2-inch cylinders. An
explosive pellet and a detonator (Reynolds RP-501)
were held in an aluminum fixture mounted on a
Plexiglas block. The fixture provided for some
confinement as well as ensured that the detonator
and explosive were in contact.
FIGURE 2. Shock tube.
Three sets of samples were tested for this
paper. The baseline formulation was PBXN-109.
The two other formulations were pressed analogs to
PBXN-109. One set contained 20 urn Al and the
other 150 nm Al. Each composition was tested five
times. The explosive compositions are given in
Tables 1 and 2.
MDX-81 Al has an average particle size of
20 um and is spherically shaped. This Al is used in
PBXN-109. The nanosized Al "Alex" was from
Russia (Argonide Inc. Lot # A06-24-25R). It had a
wide particle size distribution with a range of 50 to
500 nm with a typical size of 150 to 200 nm.
A major concern with evaluating nanosized Al
is the oxide layer that exists on the particle. This
layer is typically on the order of 3-nm thick for all
particle sizes. It is due to the diffusion of oxygen,
which leads to passivation of the surface of the
metal. For um-sized particles, the 3 nm shell is not
significant in evaluating the amount of pure metal
in the particle. However, for a nanosized particle,
the oxide shell can take up an appreciable volume
of the particle. For the Al used in this study, the
oxide layer was measured to be approximately 14%
by weight of the particle. Thus, the total amount of
available Al is lower in the nanosized composition.
FIGURE 1. Internal blast chamber.
The chamber has an interior volume of
20,600 cc, this results in approximately 0.2 mole of
O2 present in the chamber under standard
temperature and pressure. Typical samples have
0.02 mole of Al. Thus, we had sufficient O2 present
for complete reaction of the Al into A12O3.
The second experimental setup consisted of a
shock tube sealed at one end and open at the other.
The steel tube was 20 feet long and 4 inches in
diameter. Pressure gauges (PCB 102 A) were
mounted along the side of the tube at various
locations away from the explosive. The gauges
were at 4 inches, 15 inches, and 3, 8, 13, and
18 feet. A picture of the shock tube is given in
Fig. 2.
TABLE 1. Compositions used in internal blast tests.
Explosive
Organic
Metal
Binder
, . J
Process
PBXN-109
68% RDX
12%HTPB
1.69
Cast
RDX/20 um Al/Binder
70% RDX
10%Zeon
1.64
Pressed
RDX/150nmAl/Binder
70% RDX
20% 20 um
-MDX-81
20% 20 um
-MDX-81
20% 150nmAl
-Alex
10%Zeon
1.71
Pressed
951
TABLE 2. Compositions used in shock tube tests.
Explosive
Organic
Metal
Binder
PBXN-109
68%RDX
RDX/20 jim Al/Binder
70% RDX
RDX/1 50 nm Al/Binder
70% RDX
20% 20 nm
-MDX-81
20% 20 jam
-MDX-81
20%150nmAl
-Alex
RESULTS AND DISCUSSIONS
Process
12%HTPB
, . ^
(g/cc)
1.69
10%Viton
1.84
Pressed
10%Viton
1.89
Pressed
Cast
—— Alex - averaged
- - - MDX-81 - averas
2000-
1500-
The results from the two sets of tests are
presented.
a 10001
Internal Blast Chamber
5000-500-
A tvnir.al nres sure -time trare in the internal
70
blast chamber for the 20 (im Al is shown in Fig. 3.
The ringing is due to reflections in the chamber as
the pressure wave cycles between the top and
bottom of the chamber.
75
80
85
90
95
Time [us]
FIGURE 4. Comparison between the 20 jam Al (MDX-81) and
the!50nmAl(Alex).
Shock Tube
2000-n
• MDX81Test#16
1500-
A comparison of the
compositions is shown in Fig. 5.
1000-
RDX/Viton/Al
500-
-Alex
- MDX-81
0-500-
60
70
75
80
Time [us]
85
FIGURE 3. Pressure-time trace for 20 jam Al sample.
In Fig. 4, the averaged results from five 150 nm
Al tests are compared with the averaged results
from five 20 um Al tests. The initial pressure of the
nanosized Al is clearly higher than the jam Al (an
average increase of 63%). If one determines the
impulse by integrating the pressure-time curve from
the initial rise in pressure for a fixed time, the
impulse from the composition containing Alex is
2.75 times higher then the MDX-81 composition.
The results shown in Fig. 4 compare well with
the calculated pressures from DYNA2D (4). The
calculation showed the ringing as the wave is
reflected inside the chamber as seen in the
experiment. The model also predicted a higher peak
pressure and impulse for the nanosized Al, as was
observed.
500
1000
1500
Time [us]
2000
25I
FIGURE 5. Shock tube result comparing Al particle size.
The results from the tube were not as straight
forward as those observed in the chamber. First, the
arrival times of the wave at the gauges were not
consistent in the same test and also varied within
the same material. This made analysis difficult due
to the lack of a consistent time. Also, comparison
was made with PBXN-109 and PBXN-5 (RDX and
no Al). PBXN-5's pressure traces were very
similar in structure to the pressed 109 analogs.
While slight differences were noticed with the
different Al, they were not as dramatic as was
observed in the chamber. The impulse for the
952
nanosized Al was slightly higher but was due, for
the most part, to the mismatch of rise times. At the
gauge located 4 inches from the explosive charge,
the electrical noise from firing the detonator made
determination of initial rise often difficult. The
average from five tests on each material is
presented in Figure 6. Figure 7 shows a comparison
from single set of tests over a longer time base.
CONCLUSIONS
The nanosized Al showed a significantly
greater initial blast pressure in the closed chamber
test but no change was observed in the shock tube.
The blast chamber's experimental work has good
agreement with the calculated blast pressure
obtained from DYNA2D modeling.
Work is
necessary to develop a better test method to observe
the reaction mechanisms in air. Modeling work is
also continuing to develop a reaction kinetics model
to account for differences in particle size of
aluminum.
Alex
MDX-81
ACKNOWLEDGMENTS
The ONR 6.2 Air Weaponry Technology
Program through Tom Lofrus and the ONR 6.1
Marine Corps Research Program sponsored this
work.
Time [us]
FIGURE 6. Pressure-time trace of average of five tests from
gauge located 4 inches from explosive charge.
REFERENCES
1500-
Davis, J.; Miller, P. J.; and Bedford, C. "Effect of
metal particle size on the detonation properties of
various explosives," in the Proceedings of TTCP WTP-4 Technical Workshop, Indian Head, MD,
April 16, 1999.
Woody, D. L.; Davis, J. J.; and Bedford, C. D.
"Comparison of the visible emissions from energetic
materials containing differing particle sized
aluminum," published in Proceedings of JANNAF
Combustion Subcommittee and Propulsion Systems
Hazards Subcommittee Joint Meeting, November
1996.
Davis, J.; Miller, P. J.; Bui, Q.; and Pockrandt, S.,
"Effect of metal particle size on internal blast
explosives," published in Proceedings of JANNAF
Combustion Subcommittee and Propulsion Systems
Hazards Subcommittee Joint Meeting, November
2000.
Miller, P., "A reactive flow model with coupled
reaction kinetics for detonation and combustion of
non-ideal explosives," Decomposition, Combustion
and Detonation Chemistry, eds. T. Brill et.al.,
Vol.418, p. 413, Mater. Res. Soc. Symposium
Proceedings, 1996.
1000-
500-
0-
-500-
-1000'
100
120
140
160
180
200
Time [us]
FIGURE 7. Pressure-time trace of individual tests from gauge
4 inches from charge.
Gauges mounted on the side might have caused
some of the differences between the tube and the
chamber that had its gauge mounted on the top in a
direct path of the oncoming particles. While the
gauge in the chamber was protected, it might not
have been adequate. The nanometer-sized Al would
be able to travel faster than the 20 um Al and thus
might be able to impact a gauge mounted in its
path, increasing the observed pressure. Turbulence
for mixing is known to enhance metal reaction but
neither tube nor chamber should have experienced
much turbulent mixing during the initial pressure
rise.
953