United States
Environmental Protection
Agency
Municipal Environmental
Research Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-84-051 Apr. 1984
©EPA Project Summary
Gaseous Emissions from Excess
Air Combustion of Propellants
and Explosives
J. Mahannah, D. Schubert, C. Gulp, andT. Schomer
The purpose of this short-term project
was to determine the levels of nitric
oxide (NO), nitrogen dioxide (NCh). and
carbon monoxide (CO) in the off-gases
from the open burning of explosives in
excess air. The ultimate goal is to
reduce the level of NOX, CO, and
participates emitted during the destruc-
tion of surplus, waste, and off-spec
explosives. Previous work (DOE) showed
that a gravel/sand filter in the roof of a
bunker reduced the level of particulates
emitted during excess air combustion
of propellants; few NOX or CO measure-
ments were reported. In this project,
two HMX-(C4H8O8Ne)-based propellants
(Chaparral* 6678, 200 to 538 g, and
Arcadene 311B, 65 tj 162 g) were
burned in a 1.3-m3 steel chamber fitted
with a 30-cm-deep sand and gravel
filter. Air (1 to 6 mVmin) was blown
into the chamber to ensure combustion
and to force the gases through the filter,
which included a 2-cm layer of damp
activated carbon. Of the components
measured, the NO-concentration was
the predominant and most reproducible.
Little NO2 was observed. CO production
fluctuated widely, probably because of
inadequate mixing of the gases within
the chamber. The NO and CO concen-
trations decreased across the filter. NO
reductions were 25% to 67% for
Arcadene and 10% to 57% for Chaparral;
CO reductions were from 38% to 81%
for Arcadene and 33% to 91% for
Chaparral. The project demonstrated
that the filter is effective in partially
eliminating NO and CO emissions, but
'Mention of trade names or commercial products
does not constitute endorsement or recommenda-
tion for use.
that additional work, including the
incorporation of catalysts in the bed and
the introduction of NO control gases
(e.g., NH3), should be undertaken.
This Project Summary was developed
by EPA's Municipal Environmental
Research Laboratory, Cincinnati. OH.
to announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction and Background
When surplus, waste, over-age or out-
of-specification explosives and propellants
must be disposed of, the conventional
procedure is to destroy them by open
buring or by detonation. Other processes,
including incineration, are also used, but
less frequently. Both open burning and
detonation, although low in capital and
operating cost and relatively safe, generate
undesirable air pollutants, including NO,
CO, NO2, and, occasionally, N2O, HCN,
and free radicals (H, OH, N, O). The nature
of these two destructive processes is also
such that control of emissions has not
been practiced or is impractical.
Thermal destruction with excess air in
an incineration system makes it possible
to control the emissions arising from
decomposition of these high energy
materials. The air pollution control
equipment for such systems is, however,
usually complex and costly. Further, the
incinerator must be specially designed to
handle fortuitous blast incidents. Inex-
pensive alternatives for the control of air
pollution during the destruction of
explosives make novel processes attrac-
tive.
-------
The goals of this project were to
evaluate an improved technique for open
air burning of explosives and to recom-
mend innovations. The improved destruc-
tion process for explosives (including
propellants) was designed to lower the
level of particulates and noxious gases
released into the air space, when the
level is compared with that attained
during destruction by detonation or
conventional open air burning. Data are
generally available on the composition of
gases and vapors released from explosives
in actual-use situations (blasting, rocket
propulsion) in which all (or most) of the
oxygen is provided by the explosive; but
few data are available on combustion
products formed during burning (or
incineration) in excess air. Thus, in this
project, data had to be generated on the
levels and proportions of nitric oxide (NO),
nitrogen dioxide (N02>, and carbon
monoxide (CO) in the off-gases from the
open burning of high explosives, including
propellants, in excess air.
Previous work (DOE) showed that a
gravel and sand filter installed in the roof
of a bunker significantly reduced the level
of particulates emitted during the excess
air combustion of high explosives, but
few NOx or CO measurements were
reported.
In the work now being reported, two
HMX-C4H8O8N8)-based propellants (Cha-
parral 6678 and Arcadene 311B, i.e., C-4)
were burned in a specially built steel
chamber fitted with internal baffles and a
top-side, layered, sand-on-gravel bed
that also contained damp, granulated
activated carbon. A blower furnished
excess air to ensure smooth combustion
and to force the off-gases, vapors, and
particulates through the filter bed.
Thermocouples and gas analysis probes
were installed above and below the filter
to monitor gas/vapor temperatures and
compositions.
Explosives can be broadly classified as:
(1) low explosives (black powder), (2)
high explosives (dynamite, a mixture of
nitroglycerine (CaHsOgNa) and diatoma-
ceous earth; RDX (CaHeOeNe), cyclotri-
methylenetrinitramine; HMX (C^eOaNa),
cyclotetramethylenetetranitramine; AMFO,
(94% ammonium nitrite + 6% fuel oil), and
(3) detonators (lead azide or styphnate,
mercuric fulminate).
Detonators are used to set off high (and
low) explosives for: blasting, setting off
explosive military devices (bombs, mines,
grenades) or propelling munitions (shells,
bullets, flares). High explosives (propel-
lants) are also used as thrusters to power
rockets, space ships, and missiles.
Most explosives (excepting detonators)
are usually formulated so that considerable
initiating energy—from blasting caps,
electric or powder squibs, sparks—must
be concentrated on the material to initiate
an explosion.
The explosive is usually in the form of a
solid or liquid; on explosion, the material
converts to hot gases (and possibly some
paniculate material), which exert great
force on the container originally filled
with the explosive. A gram-mole of HMX
(MW 296) has volume of about 1/5 L
(.052 gal), but on explosion quickly
produces 270 L @ STP (at standard
temperature and pressure) of very hot gas
(equimolar N2, CO, and H20 (steam)).
Many explosives are designed to require
no extra oxygen. Note that the HMX has
sufficient chemically bound oxygen to
produce the gaseous products (so-called
zero-oxygen balance).
It is not surprising that materials other
than CO, N2, and H20 mayalsoform: N20,
NO, N02, HCN, and even free radicals.
The conditions existing during explosion
or burning define the applicable kinetics
and thermodynamic constraints, which
determine the products and by-products
formed.
In unconfined (open) burning, one
would expect that 02 from the air would
combine with CO to form CO2, for
example. The production of NOX may be
enhanced or decreased, depending on
the temperature and evenness of the
combustion, the access to air, and other
factors. (The interested reader may wish
to predict the combustion products of
black powder, where the oxygen-balanced
reaction is 2 KNO3 + S + C = K2S04 + C02 + N2.
As the proportion of C is increased, CO,
sulfites, and sulfides will be formed.)
Explosives are generally formulated to
make them suitable for a particular use
(controlled burn rate, decreased shock
sensitivity). Consider a military shell: the
round must not explode during propulsion
but only on reaching the target area.
Additives include metal powders and
inorganic compounds, which produce
particulates (some soot may also form).
It should be clearly understood that
open burning or detonation of waste
explosives serves a very useful purpose in
protecting human life and limb and will
continue to be used. The work reported
here, as well as any future related effort,
is expected to provide a better understand-
ing of the options for destroying explosives.
We need approaches applicable not only
to explosives but also to other hazardous
materials—approaches maintaining not
only the safety and low cost features of
open burning but also improving air
pollution control.
System Design and Results
The thermal destruction chamber used
for these tests is about 1.3 m3 (45 ft3) in
volume and is constructed of 3.2-mm
(1 /8-in.) steel plate. The overall design is
shown in Figure 1. The filter bed is
supported on an expanded metal grating
overlaid with wire cloth 30 cm from the
top of the chamber. The bed is assembled
from the following materials (bottom-to-
top): 10 cm (4 in.) of road gravel (2.5-cm
(1 -in.(diameter); 5 cm (2 in.) of pea gravel;
2 cm (1 in.) of activated carbon; and 13 cm
(5 in.) of sand. An air blower and
restricted air inlet were incorporated in
the design to provide the proper volume of
air for the combustion and to force the
gases through the filter bed. Gas analyzer
(CO, NO, NO2) probes were placed in the
chamber above and below the filter bed to
monitor components of the combustion
gases. The probes were calibrated with
EPA "Protocol" gases. The temperature
in the chamber was measured with a
Type K (Chromel/Alumel) thermocouple.
The two propellants in this study
(Chaparral 6678 and C-4) both contain
HMX as the primary explosive (60% and
84.8%, respectively). Both propellants
are soft solids that can be easily cut to
size. The propellant was ignited by
electrically heating a nichrome wire
embedded in a block of the propellant
charge.
A series of experiments was first
carried out in the chamber with black
powder to observe gas flow and to adjust
internal baffles. The next tests established
the minimum weight of propellant
needed to generate measurable volumes
of gases. Chaparral samples of 200 to
500 g and C-4 samples of 100 to 200 g
were required to give measurable volumes
of NO and CO (Table 1). The first 27 trial
burns were also used to optimize condi-
tions for the tests in terms of the
minimum volume of air needed to ensure
combustion of the propellant. When the
air blower with restricted intake was
used, the minimum air flow was 2.4
mVmin (85 ftVmin). Lower flows resulted
in increased CO levels, indicating incom-
plete combustion to CO2.
Though several runs were carried out
during the study with each propellant
(runs 28 through 42), considerable
variations occurred in the volumes of
gases generated and in the concentration
of NO, NO2, and CO in the gas, both before
and after passage through the filter.
These fluctuations did not correlate
-------
Table 1. Combustion of Chaparral and Arcadene 31 IB !C-4)
Burn
#
7
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
27
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Propellant
Chaparral
Chaparral
Chaparral
Chaparral
Chaparral
C-4
C-4
C-4
C-4
Chaparral
Chaparral
C-4
C-4
C-4
C-4
C-4
Chaparral
Chaparral
Chaparral
Chaparral
Chaparral
Chaparral
Chaparral
Chaparral
Chaparral
Chaparral
Chaparral
Chaparral
C-4
C-4
C-4
Chaparral
Chaparral
Chaparral
C-4
C-4
C-4
C-4
C-4
Chaparral
Chaparral
Chaparral
Propellant
Weight
(9>
450
340
450
450
225
225
225
225
225
630
180
180
180
180
180
180
545
545
480
446
474
211
498
493
248
435
476
522
162
65
70
503
493
528
129
114
114
103
84
200
500
500
Air Flow
Rate
(m3/minl
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
6.4
3.9
2.9
1.0
1.0
1.0
1.0
1.0
1.0
1.5
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
2.4
Chamber
Temperature
<°C)
_
180
299
136
147
166
173
155
130
311
342
370
366
326
282
355
420
278
370
318
370
194
92
172
376
394
416
131
139
153
190
380
Maximum Gas Concentrations (ppm)
Bottom Top
NO
-
250
250
250
250
250
250
360
370
125
375
320
360
200
200
260
140
500
500
500
500
210
500
500
1250
950
525
525
150
1300
600
350
300
700
450
400
1100
1150
200
200
200
CO
3000
-
175
3OOO
0
0
0
0
3000
0
0
0
0
0
0
0
0
0
0
0
0
3000
600
450
2100
975
1050
150
-
0
0
0
0
600
0
0
30
450
9
30
45
NOZ
-
5
5
10
30
125
-
50
15
-
50
40
10
10
15
15
10
5
0
0
0
0
0
0
0
0
5
5
0
0
70
0
0
0
0
2
20
0
0
0
NO
-
205
250
250
200
250
250
-
200
190
80
210
170
130
130
120
120
150
180
200
130
110
230
175
175
175
225
275
250
125
60O
400
300
300
250
100
100
250
825
100
150
150
CO
3000
0
O
3OOO
0
0
0
-
0
0
0
1200
0
0
0
0
0
0
0
0
0
0
3000
900
300
750
3OO
225
75
-
600
O
0
0
150
0
0
0
270
0
15
15
/V02
-
0
7O
10
5
10
0
25
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
-
55
10
0
0
20
0
0
5
15
0
0
0
with the sample size, but may have arisen
from incomplete mixing of the gases or
uneven burning of the propellant. Of the
gases measured, NO was usually the
predominant product of combustion.
Nitrogen dioxide (NO2) was not detected
during the burning of Chaparral and was
found only in very small concentrations
during the combustion of C-4. Results of
the gas analyses and estimates of the
removal efficiency of the filter are
summarized in Tables 2 and 3 for the two
propellents.
Conclusions and
Recommendations
The results of tests on the excess air
combustion of high explosives in a steel
chamber fitted with a charcoal-containing
sand and gravel filter (1) indicate that the
levels of the evolved nitric oxide (NO) and
carbon monoxide (CO) are effectively
lowered, and (2) verify a prior observation
that particulates are effectively removed.
By the reaction of CO with ©2 in the air,
the level (quantity) of CO is much
reduced over what is common during
confined combustion (propellants) or
during explosions. The filter is effective
in further reducing the level of noxious
gas (CO, NO, etc.) concentration during
passage through the composite bed.
Definitive values for the achievable
reduction in level of each gas cannot be
presented because of the wide variation
in data. Specifically, NO reductions by the
filter ranged from 25% to 67% for C-4 and
10% to 57% for Chaparral; CO reductions
ranged from 38% to 81 % for C-4 and from
33% to 91% for Chaparral.
Little or no NO2 results from the
combustion of either C-4 or Chaparral
6678 under the conditions existing in the
chamber used in this study. Low NO2
generation has also been reported by
other investigators.
The variations in removal of NO and CO
may be due, at least in part, to the design
of the excess air inlet system and the
resulting uneven burning of the propellant
and inadequate mixing of the gases in the
chamber. These problems could be
avoided by modifying the design of the
chamber and by developing more effective
means of providing air to the sample. For
example, better mixing of the gases and a
more even burn might be achieved by
providing underflow and/or tangential
air flow to the chamber. Investigation of
-------
Exhaust
30 cm f12") sq. opening
53cm(21">TC
Probe - 2
Places
30 cm
(12"Bedl
Bottom Sampling Lines
to Monitor
Blower
61 cm
124 ")
30 x 51 cm
fl 2" w x 20" Ig)
Figure 1. Schematic of combustion chamber.
81 cm
(32") sq. opening
89 cm
(35" sq.)
Table 2. Gaseous Products from Chaparral
Incineration*
Tables.
Gaseous Products From C-4 Incin-
eration'
Pollutant
NO
NOZ
CO
Before
Filter
(ppm)
94-295
0
0-434
After
Filter
(ppm)
56-202
0
0-40
%-Fleduction
by Filter^
10-57
33-91
Pollutant
NO
NO2
CO
Before
Filter
(ppm)
182-676
0-7
0-475
After
Filter
(ppm)
69-304
0-8
0-132
%-Fteduction
by Filter^
25-67
38-81
*Time-weighted averages.
Sample weight: 200 to 528 g.
Chamber temperatures: 376° to 416°C.
t The%-reduction was calculated for each burn
and not from the "before filter" and "after
filter" values cited.
''Time-weighted averages.
Sample weight: 70 to 162 g.
Chamber temperature: 131° to 172°C.
t The%-reduction was calculated for each burn
and not from the "before filter" and "after
filter" values cited.
these factors should be a major consider-
ation in any future study.
Future work should be directed towai
applying known or novel technology to
changing the compostion of the gases
emitted to the atmosphere, e.g., adding
catalysts to the filter bed to reduce NO,
N02, and CO emissions. A nickel oxide
catalyst is reported to accelerate the
reaction of NO with CO:
2NO + 2CO-
NiO
-2CO2 + N2.
When nickel catalysts were used in a
fluidized bed incinerator to dispose of
explosive wastes, NO, N02, and CO
emissions were reduced substantially.
Similarly, although NO is a relatively
difficult air pollutant to remove from a gas
stream, NO2 much more readily dissolves
in water (ultimately forming nitric acid).
Though adsorption is poor for N02 (but
better than for CO and NO), it can be
adsorbed on activated carbon. Therefore,
it would be advantageous to explore
methods to rapidly convert NO to N02,
such as by adding copper or other
oxidation catalysts to the filter bed, or to
determine whether catalyzed reduction
(using NH3) can be effectively applied.
The full report was submitted in
fulfillment of EPA Contract 68-03-3069
by IT Corporation under the sponsorsr
of the U.S. Environmental Protectio,.
Agency.
------- |