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.

-------