J Forensic Sci, Jul. 1975, Vol. 20, No.

A. D. Beveridge,' Ph.D.; S. F. Payton, ~ B.Sc.;

R. J. Audette, 2 Ph.D.; A. J. Lambertus, ~ B.Sc.; and
R. C. Shaddick, ~ M.Sc.

Systematic Analysis of Explosive Residues

Examination of bomb scene evidence is an area of increasing involvement for the

forensic scientist in which incidents may vary from student mischief to mass murder.
To undertake explosive casework responsibilities, the scientist must have personal
experience of the type of physical evidence left after explosions, including device remains, characteristic damage, and chemical residues.
Recognition of device remains is aided by law enforcement agency publications [1-6].
Explosive damage to metal has been studied [7] and the significance of damage in bomb
incident investigations has been discussed in general terms [8]. Chemical residues may
contain unreacted explosive components [9-13] and also condensed reaction products
[12,14]. Solid products have been studied less than gaseous products, but have been
reviewed for black powder [14,15] and chlorate and perchlorate mixtures [16]. Smokeless
powders yield nitrites which form the basis of propellent powder pattern tests [17,18].
Since solid reaction products from dynamite do no work, their chemical composition has
been deduced more from thermodynamic calculations than analysis [12,19].
Published methods for explosive residue analysis include microscopic examination for
unreacted explosive [10,11], chemical spot tests [10,13,15,20,21], thin-layer chromatography (TLC) [9,10,22,23], and infrared spectroscopy [24,25].
This study presents a scheme for systematic analysis of explosive residues isolated by
physical removal or solvent extraction. The techniques are a combination of infrared
spectroscopy, TLC, X-ray diffraction, emission spectrography, microscopy, and
chemical spot tests. Application to residue analysis is demonstrated by test explosions
using black powder, single- and double-base smokeless powder, chlorate/sugar,
dynamite, C4 plastic explosive, and PETN-based detonating cord (a glossary of terms
is given in Appendix A).

Experimental
Analysis
The analytical scheme in Fig. 1 was applied to all of the explosives and their
residues. Figures 2 and 3 expand on the solvent extracts.
Received for publication 11 Nov. 1974; accepted for publication 23 Jan. 1975.
1Section Head, Chemistry Section, Royal Canadian Mounted Police, Crime Detection Laboratory, Edmonton, Alberta, Canada.
2Forensic Scientists, Chemistry Section, Royal Canadian Mounted Police, Crime Detection
Laboratory, Edmonton, Alta., Canada.
431

JOURNALO F FORENSIC SCIENCES

432

DEBRIS

STEREOMICROSCOPE
l

CHE~ICAL~ESIDUE
UNREACTEEXPLOSI
D VE

DEVICEPARISEXPLOSI
,
VE
WRAPPERS
ETC.

.1

IR

ETHER

ACETONE

WATER~

IR

'
LR

TLC

TLC

INSOLUBLE

POLARIZING
MICROSCOPE

X-RAYDIFF. IR

TLC

ES

SPOTTESTS

FIG. 1--General scheme for analysis of explosive residues.

ACETONE

ETHER

I
IR

RNO2
Plasticizer

-IR

TLC

I
or

NaOHIGRIESS

NG

EGND

TLC

NR4N03

Wax
P e t r o l eum

RNO 2

Benzene/n-Hexane(l:l)
Xylene/n-He

TiC13/DMAB

CNS

Plasticizer
F
Benzene/n-Hexa~e(1:l)
or
Xylene/n-Hexane{3:2)
[ - EtOH
i
NESSLERS
NaOHIGRIESS

EtOH

I
CHCl3/Acetone
(1:1)

I%

NaOH/ J
GR~ESS[
amine/H2S04 I Thymoll
I
H2SO4
Diphenyl-

TiCi3/DMAB
TIIT
DNT

NC
NG
EGDN

TNT
DNT

,o~

pET.
RDX

RDX

FIG. 2--Analysis of organic extracts.

Explosives and Explosions

Table 1 summarizes the explosives and conditions used in this study. Charges were
made up and initiated by Explosive Ordnance Disposal (EOD) trained personnel using
safety fuse, nonelectric blasting caps, and detonating cord boosters as required. Low
explosives were confined in l/tin, wall steel pipes with threaded end caps. Explosive
residue recovery procedures were concentrated on recognizable device remains, the
crater, and adjacent surfaces.

BEVERIDGE ET AL ON EXPLOSIVE RESIDUES

433

INSOLUBLE

WATER

I
POLARIZING MICROSCOPE

IR

ES

NO~
SO/~
CO~

K
Na

SPOTTESTS

X-RAY DIFF.

TLC

'

PULP

I~g, e t c .

UNEXPLODED

PRODUCTS

NaNO3

K2SO4

KNO 3

N~2SO 4

NH4NO 3

Na2CO 3

NH~
CI-

EtOH

I
1%
DIPNENYLANNEIH2SO 4

No~

I
EtOH
NESSLERS
REAGENT

NH~

FIG. 3---Analysis of aqueous extracts.

Techniques
lnfrared Spectroscopy--A Perkin Elmer Model 457A infrared spectrophotometer was
used. Ether and acetone liquid extracts were sandwiched between 13-mm KBr disks
and recovered by washing. Solids were analyzed as KBr pellets or nondestructively on a
diamond cell (High Pressure Diamond Optics, McLean, Va.).
Infrared (IR) analysis may identify explosive components [24,25], inorganic ions [26],
and contaminants, and may aid TLC selection and interpretation.
Thin-Layer Chromatography--TLC was run on 0.25-mm, precoated silica gel G glass
plates (Mackery-Nagel and Co., Duren, Germany). Principal solvent systems were
benzene [23], benzene/hexane [10], xylene/hexane [10], ethanol [9], and chloroform/
acetone [9]. Developers were NaOH/Griess reagent [23], titanium trichloride/dimethylaminobenzaldehyde (TiCI~/DMAB) [9], diphenylamine/H2SO,, and ethanol/Nessler
reagent. The developers are applied as follows.
NaOH/Griess: Spray with 1N NaOH, hold in oven at 100~ for 10 min, cool, and
spray with Griess reagent (0.25% sulfanilic acid and 0.1% a-naphthylamine in 1:1
aqueous acetic acid). Red spot for EGDN, NG, NC, RDX, PETN. High concentrations
appear as a yellow spot with red edges [9]. TNT gives brown spots.
TiClJDMAB: Spray with 15% TiCI~/HC1, air-dry, and spray with DMAB reagent
(1 g DMAB in 30 ml EtOH, 3 ml HC1 1:19, 180 ml BuOH). TNT and DNT give yellow
spots [9].
Ph2NH/H2SO,: Spray with 1% Ph2NH in conc. H2SO,. Blue spot for nitrates.
EtOH/Nessler: Spray with EtOH followed by Nessler reagent. Orange spot for NH,

X-Ray Powder Diffraction--Eqtflpment was a Philips X-ray generator (PW 1130/00/60)

with a copper diffraction tube (Type RD 60/2 PW 2103/00), operated at 50 mA,
40 kV, with 10 to 25-min exposure. A nickel filter removed Cu K8 radiation. Samples
were contained in 0.3-mm capillaries in Debye Scherrer l14.8-mm powder cameras.
Ilford Industrial Type G X-ray film was used.

Canadian
Canadian
Canadian
Canadian

75% Forcite |
60% Geogel | (gelatin dynamite)
Cd plastic explosive
PETN detonating cord

Industries Ltd.
Industries Ltd.
Armed Forces
Armed Forces

Canadian Industries Ltd.

60% Forcite|

box

Canadian Industries Ltd.

unconfined
unconfined
unconfined
unconfined

unconfined

pipe
unconfined

pipe
pipe
pipe
pipe
unconfined

Curtis-Harvey
Du Pont
Hercules
homemade
Canadian Industries Ltd.

Black powder
Smokeless, single base
Smokeless, double base
Chlorate/sugar
20% Polar | stumping powder
(ammonia dynamite)
40% Forcite|
(ammonia-gelatin dynamite)

Confinement

Source

Explosive

TABLE 1--Explosives and conditions.

bunker, soil, 9 by 12-ft room

bunker, 9 by 12-ft room
bunker, 9 by 12-ft room
bunker

sandbagged bunker
sandbagged bunker, 9 by 12-ft room
sponge rubber, carpet, clothing, ceiling tile,
wood, 9 by 12-ft room
wood, gypsum wallboard, plastic base
molding
steel
steel
steel, wood, sandbagged bunker
aluminum, steel, wood, soil

sandbagged
sandbagged
sandbagged
sandbagged
steel, wood

Surface

r'n
z
C)
pn
O~

rn
z
03
C~
Co

4x
GO

BEVERIDGE ET AL ON EXPLOSIVE RESIDUES

435

X-ray powder diffraction permits nondestructive identification of crystalline materials

by comparison to standards or by use of the American Society for Testing and
Materials (ASTM) Powder Diffraction File [27].
Emission Spectrograph--Equipment was a Jarrell-Ash 3.4-m Ebert with dc arc
excitation at 10 A for 10 s using a 10-/Jm slit. Samples were contained in 1A-in. outside
diameter carbon cups. The emission spectrograph was employed to identify explosive
metallic content such as alkali metals, mineral contaminants such as boron in Chile
saltpeter, and manufacturer additives.
Polarizing Microscope--Equipment was a Leitz Orthoplan | microscope equipped with
an interference wedge (Leitz S-100) and a Mettler FP5-FP52 hot stage. This instrument
permits nondestructive classification of crystalline materials by morphological analysis
[28] and ready identification of isotropic substances lay the Becke line technique for
refractive index riD.
Spot Tests--Spot tests were mainly those described by Amas and Yallop [13,20].
Chemical spot tests were used to check components such as NO~-, which in low yield
(< 10~ might be missed by instrumental analysis, and for rapid screening of a high
yield of residue.
Results

Tables 2, 3, and 4 summarize the analysis of residues from the explosives listed in
Table 1, by use of the scheme in Fig. 1.
Discussion

L o w Explosives
Low explosives confined in steel pipes with threaded end caps exploded on flame
initiation. Figure 4 illustrates damage to two series of pipes by charges of black powder,
single-base smokeless powder (rifle propellant), double-base smokeless powder (pistol
propellant), and (for comparison) dynamite. Fragmentation by chlorate/sugar was
similar to single-base powder. The degree of fragmentation is seen to be much higher for
double-base powder than for the other low explosives.
Table 2 summarizes the analysis of the low explosives and their residues using the
scheme in Fig. 1.
Black powder gave pipe fragments with fresh rusty areas and a high yield of
grayish residue. Unreacted granules were commonly recovered in pipe threads and sometimes in the body of the pipe. The surroundings were usually blackened. Figure 5 shows
the effect of a 0.75-1b charge in a corner of a 9 by 12-ft room. Pipe fragments are
shown in the bottom row of Fig. 4. Analysis of debris from this explosion is summarized
in Table 2, No. 4. Thiocyanate was identified by 1R spectroscopy in an acetone extract.
Black powder residue consisted primarily of potassium sulfate identified by X-ray
powder diffraction. An 1R spectrum of typical residue (Fig. 6) showed strong sulfate and
weak nitrate. Weak carbonate peaks and absence of nitrate peaks were also observed in
different explosions; that is, composition and distribution of explosive residue was not
uniform. Reviews [14,15] suggest a higher percentage of potassium carbonate, but since
it was not observed in X-ray powder diffraction patterns, this indicates that its yield is
less than 10%. The safety fuse also gave potassium sulfate as principal residue from the
black powder core.
The value of product identification is illustrated in Table 2, No. 3, since if the few

pipe and building material

room wall

black powder

black powder
smokeless, single-base
smokeless, single-base

smokeless, single-base

smokeless, double-base
smokeless, double-base

chlorate/sugar
chlorate/sugar

5
6

8
9

10
11
pipe and sand

pipe and building material

pipe and building material

pipe and sand

pipe and sand

black powder

pipe and soil

Residue-Bearing Surface

black powder
black powder

Explosive

1
2

Number
black granules
gray crystalline
black deposit
gray crystalline
b l a c k granules
gray crystalline
black granules
black deposit
black cylinders
resinous
black cylinders
resinous
black cylinders
black platelets
resinous
black platelets
white granular
resinous
white granular

Microscopic Appearance

TABLE 2--Analysis o f low explosive residues.

unreacted
NO3-, K2SO,, CO32unreacted
NO3-, K2SO,, C O 3 2 - , CNSNC, DNT, K2SO,
NC, K2SO4, NO2unreacted
NC, DNT, (CaSO4-0.5H20)
unreacted
NG, NC, K2SO,
NG, NC, SO42-, NO2-, (Na)
unreacted
NaC10~, sucrose
NaC103, sucrose, NaCI
unreacted

K2SO,, CO3 2-

KNO3, S, C
NO3-, K2SO,, CO32-

Components Identified

4=,

O
m
z
t')
m

m
Z
co

"11

o"13

Go
o)

FIG. 4 - - P i p e b o m b f r a g m e n t a t i o n b y (left to right) black powder,

and 40% dynamite: (top)12 by 1 89
p i p e a n d (bottom) 6 b y 2 89
pipe.

single-base smokeless powder,

double-base smokeless powder,

4~
,.,4

r-"
m
~o

m
co

<
m

Ill
x
"13

oJ
m
<
m
ID
m
m

moist, compact
moist, compact
white crystalline
white crystalline
white crystalline
unreacted 40070
white crystalline
paper
white crystalline
unreacted 40~
white crystalline

2 by 6-in wood
3A-in. plywood
3-in. foam rubber and
89
plywood
carpet and 89
plywood

cloth and 89

ceiling tile and

89
plywood

60

75

40
40
40

40

40

40

16
17
18

19

20

21

steel

plywood

moist, compact

15

89

crater

6 by 3-in. crater

9 by 3-in. crater

9 by 2 89

shattered
9 by 4-in. crater
no crater

folded to U-shape

Components Identified

NG, NaNO3, Na2SO4 (I)

NG, NC, N a N O , Na~SO, (I)

EGDN, NG, NC, NH4NO3, NaNO3, S,

pulp
EGDN, NG, NC, NH4NO3, NaNO3, S,
pulp
EGDN, NG, NC, NH4NO3, NaNO3, S,
pulp
NG, NaNO~, SO4z
NG, NC, NaNO3, Na2SO, (I)
EGDN, NG, NC, NH4NO3, NaNO3,
Na2SO4 (I)
NG, NaNO3, Na2SO4 (I)

granular
EGDN, NG, NH,NO3, NaNO3, S, pulp
white crystalline
NG, NH,NO3, NaNO3, Na2SO4 (I), NO2paper, black deposit
white crystalline
NaNO. (Na2SO,)t-n, (Na2CO3)nf CNS-

20

crater

ammonia gelatin
40

8 by 2 89

14

3A-in. plywood

Microscopic
Appearance

12
13

Damage

ammonia
20
20

Residue-Bearing
Surface

Explosive,
070

Number

TABLE 3IAnalysis of dynamite residues.

f.O
C)
I'N
Z
O
m

1"13
Z
f/)
O

C
~CJ
Z

0o

co

40

40 b

40

60

60

60

75

75 c
gelatin
60
60 d

23

24

25

26

27

28

29

30

Y8-in. steel

~/~-in. steel

NaNO3, (Na2SO,)I-n, (Na2CO3)n a

NG, (Na2SO4h
EGDN, NG, NC,
SO2-

white crystalline
white crystalline
unreacted 60%
paper
gypsum

EGDN, NG, NC, NaNO3, pulp, BaSO~

NO~-, NO~-, Na~CO~. HzO, Ba

white crystalline

8 by 1-in. dent

moist, compact
white crystalline

white crystalline

8 by 1-in. dent

8 by 5-in. crater

gypsum

NG, NC, iNH4NO3, NaNO~, SO, 2-,

(gypsum)
NG, NC, NH~NO3, NaNO3, Na2SO4 (I),
CNS , (gypsum)
NG, NH,NO3, NaNO3, Na2SO4 (I),
CNS NaNO3, Na2SO, (I)

NH,NO3,

NaNO~,

NaNO~, (Na~SO,h-n, (Na2CO~)n a

white crystalline
black deposit
white crystalline
n, ( N a 2 C O 3 ) n a

NH,N03, NaNO3, Na2SO4 (I), (plaster)

plaster only

shattered

shattered

gypsum wallboard and

2 by 6-in. wood
gypsum wallboard and
plastic molding
%-in. steel

aA-in, plywood

pipe and 9 by 12-ft room

fragmentation
shattered
fragmentation
shattered
9 by 5 V~-in. crater

9 by 6-in. crater
shattered
shattered

pipe and 9 by 12-ft room

wooden box and batteries

9 by 12-ft room

" n = 0.2 to0.5.

~
sticks.
c No wrapper paper.
dNo wrapper paper, 1 lb shaped.

31
32

40 b

22

60
r

m
o~

_w

1"1

S
w_

Z
m
X

zo
Q

133

I"11

detonating cord
detonating cord
0.45-in. diameter
detonating cord
0.2-in. diameter

37
38

39

C4
C4
C4
C4, PETN booster

Explosive

33
34
35
36

Number

aluminum-sheathed
wood

soil

3A-in. plywood
3/8-in. steel
3/8-in. steel
89
plywood placed
12 in. from charge

Residue-Bearing Surface

cut metal and wood

trench

10 by 9-in. crater
7 by 3 89
crater
5 by 8-in. crater
shattered

Damage

TABLE 4--Analysis o f plastic explosive residues.

plastic
deposit
deposit
deposit
deposit

white fibers from cord

white plastic
orange plastic wrapper

white
black
black
black
black

Microscopic Appearance

PETN

PETN
negative

plasticizer, RDX
negative
negative
RDX
RDX, PETN

Components Identified

m
z

"n

4x
o

BEVERIDGE ET AL ON EXPLOSIVE RESIDUES

441

FIG. 5----Blackpowder pipe bomb debris.

granules of unreacted black powder had not been recognized, then identification of the
explosive would have to have been based on the reaction product only.
The single-base smokeless powder consisted of NC cylinders coated with graphite and
DNT to control burning. It also contained a small amount of potassium sulfate additive.
Numbers 5 through 7 in Table 2 summarize the analyses. Figure 7 illustrates the effect
of a cap-initiated pipe bomb in a room corner. No wall push was observed, unreacted
explosive was thrown around the room, and a devastating fire was started within the
wall.
The double-base smokeless powder consisted of graphite-coated NG and NC, plus a
small amount of potassium sulfate additive. Numbers 8 and 9 in Table 2 summarize the
analyses. A flame-initiated, 0.5-1b charge in a room showed wall push, blew out the
windows, but left intact a light bulb in the ceiling. Of twelve fragments recovered from
the crater and walls, five bore unreacted explosive. All fragments showed fresh rusty
areas and traces of resinous residue. NG and NC were identified by IR and TLC.
Aqueous extraction gave a low yield of contaminants in which sulfate was identified by

FIG. 6 - - - B l a c k p o w d e r residue, I R s p e c t r u m o f a q u e o u s e x t r a c t ( K 2 S 0 4 , K N 0 3 ) .

m
z
0
Ill

m
z

-n
9

C
-n

BEVERIDGE ET AL ON EXPLOSIVE RESIDUES

443

FIG. 7--Single-base smokeless powder pipe bomb debris.

IR and sodium by spectrography. Spot tests showed nitrate, nitrite, ammonium, and
chloride. The X-ray powder pattern was not identified. The danger of reliance on
chemical tests is apparent, since if analysis consisted only of TLC and of spot tests for
sodium, nitrate, and ammonium, then in the absence of unreacted explosive the
examiner could be misled toward dynamite; that is, the analysis would appear to be NG,
NC, NH, Na, and NO3-, of which three are the result of contamination from the
surroundings. The range of instrumental analysis applied in Fig. 1 should prevent this
type of error, since all dynamite aqueous extracts were identified by X-ray powder
diffraction.
Chlorate/sugar produced a fireball and explosion on flame initiation. Unreacted
explosive was recovered. Pipe fragments bore brown gummy residue and fresh rust
spots. Analyses are summarized in Table 2, Nos. 10 and 11. Extraction yielded
unreacted sugar and sodium chlorate identified by IR, and the sodium chloride product
identified by polarizing microscopy and X-ray powder diffraction. Since sodium
chlorate is also isotropic, the polarizing microscope was well suited to this analysis.
Potassium chlorate is anisotrooic.

444

JOURNALOF FORENSIC SCIENCES

High Explosives
Dynamite--Dynamite is the explosive class most commonly encountered in bomb
incidents. The classes of dynamite used in this study were gelatin (NG, NC), ammonia
(NG, NH,NO3), and ammonia gelatin (NG, NC, NH,NO3). In our service area and elsewhere, "nitroglycerine" dynamites contain up to 85o7o ethylene glycol dinitrate (EGDN)
with the nitroglycerine. The EGDN depresses the freezing point of the NG, which lessens
the risk of dynamite becoming unstable through freezing and thawing. The EGDN is
more volatile than NG and is the basis of "dynamite sniffers" [1,10]. The dynamite in
this study contained sodium nitrate and sulfur for oxygen balance, except for the
gelatin which did not contain elemental sulfur. None of these dynamites contained DNT,
but it is a common additive.
Cook [12] predicts that dynamite with sodium nitrate should yield sodium carbonate
and that in the presence o f sulfur, coproducts may include sodium sulfide, sulfite, and
sulfate, depending on the oxygen balance. For the ammonia gelatins, calculated
products are primarily sodium sulfate and sodium carbonate in a ratio in which the
sodium carbonate percentage increases as the explosive percentage strength decreases [19].
These data point to TLC and X-ray diffraction as prime identification techniques.
Following the TLC work of Jenkins and Yallop [9], this study commenced with benzene
eluent, but this did not cleanly separate EGDN and NG. Good separation was achieved
by benzene/hexane eluent [29] and this system superseded benzene.
The results of a comparative study of TLC solvent systems on 250-~m silica gel G
plates are shown in Table 5. Rf values are similar to those reported by Jenkins and
Yallop [9] on 250-/am silica gel G, but are about two times less than Rf values reported
by Hoffmann and Byall [10] on 200-/am silica gel G. This discrepancy is presumably due
to the difference in thickness of the silica gel layer and underlines the necessity for
running fresh standards with every TLC. Table 5 indicates that chloroform/acetone
(1/1) is the best system for initial TLC analysis of an unknown. Ammonium and nitrate
TABLE 5--TLC systems employed for explosive analysis.
TLC Systems

Explosive

Benzene

Benzene:
Hexane
1: 1

EGDN a
NG a
NC a
PETN a
RDX a
DNT b
TNT b
NO3 - c.
NH+ d

0.38
0
0.37
0
0.37
0.40

0.23
0.17
0
0.15
0
0.17
0.20

a NaOH/Griess, 250-~m silica gel G.

bTiCIJDMAB, 250-~tm silica gel G.
c Ph2NH/H~SO,, 250-1~msitica gel G.
dEtOH/Nessler, 250-:~m silica gel G.
* NaNO3, KNO~, NH~NO~.

Xylene:
Hexane
3:2
0.31
0.21
0
0.21
0
0.22
0.30

CCI,:
Acetone:
C2H4C12 CHCL
4:1
1: 1
0.23
0.16
0
0.12
0
0.20
0.15

0.65
0
0.71
0.46
0.63
0.67

Ethanol
0.64
0
0.60
0.49
0.65
0.61
0.57
0.13

BEVERIDGE ET AL ON EXPLOSIVE RESIDUES

445

ions migrate at different rates and sodium, potassium, and ammonium nitrates are not
distinguishable by developing nitrate only. This has been overlooked in previous
publications on explosive residue analysis. Ammonium ion was conveniently developed
using the Nessler reagent [13].
Thus, to characterize dynamite by TLC, a minimum of three systems was employed:
benzene/hexane with NaOH/Griess for EGDN, NG, and NC; ethanol with Ph2NH/
H~SO, for NO3-; and ethanol/Nessler for NH, +.
Table 3 summarizes the results of 18 dynamite explosions. Unless otherwise noted,
the charge was one 8 by 1 88
stick. The results are discussed by class.
Ammonia Dynamite--Explosions were conducted with one 8 by 1 88
stick on wood
and one deteriorated 8 by 1 88
stick on steel. The results are summarized in Table 3,
Nos. 13 and 14, and clearly demonstrate the complexity of explosive residue analysis.
In both cases a white crystalline residue was observed, but this was not recognizable as
unreacted dynamite. Acetone extraction and TLC identified NG in the wood crater
fragments but neither on the wood crater surroundings nor on the steel. Aqueous
extraction yielded white crystals, the IR spectrum of which showed the presence of
sulfate and carbonate products in addition to unreacted nitrate. The relative proportions
in the IR varied as shown below, where s is strong, w is weak, and m is medium.

wood crater fragments

wood crater surroundings
steel

8042- (S)
SO, 2- (m)
SO, 2- ( s )

CO32- (w)
CO32- (w)
CO32- (m)

NO3- (w)
NO3- ( s )
NO3- (m)

X-ray powder diffraction of the mixture [30] identified the product from the wood as
Na~SO,(I) and that from steel as sodium carbonate sulfate (see Table 6). The IR
spectrum of sodium carbonate sulfate is illustrated in Fig. 8 (steel residue) and the
morphology in Fig. 9. These products are further discussed in Appendix B. The nitrate
in each case was identified as the unreacted sodium salt by X-ray powder diffraction
and the polarizing microscope. No ammonium ion was found in the wood debris and
was noted only by spot test screening in the steel extract.
That the same explosive gives different products on unconfined detonation is best
explained by the deterioration of one stick. The product composition is apparently
determined by the mole ratio of CO32- :SO, 2-. These results show the major drawback
to solvent extraction in that sodium sulfate and sodium carbonate might have been
present independently, but are isolated from a complex aqueous system as a solid
solution (see Appendix B).
Ammonia Gelatin Dynamite--The velocity of 40~ Forcite | is around 7000 feet per
second (fps) unconfined and 13,000 fps confined. Velocities are proportionately higher
for 60~ and 75070. The variation of velocity with confinement is typical of gelatin
dynamites [31]. Figure 10 shows the increased shattering effect of an 8 by 1 88
stick
of explosive on 90
plywood as the velocity increases from 6000 to 25,000 fps. The
explosives illustrated were 20, 40, and 60070 dynamite, and C4, respectively (Table 3,
Nos. 13, 17, 26; Table 4, No. 34). Tests in residential premises showed a marked contrast
to the low explosives. Walls were shattered, as opposed to pushed, and the reverse blast
brought window glass back into the rooms. Figure 11 shows damage caused by two
sticks of unconfined 40070 Forcite | in a 9 by 12-ft room (Table 3, No. 22). When
confined in pipes, some pipe fragments passed through several wails and doors.

FIG. 8--20% dynamite, IR spectrum of aqueous extract from steel (sodium carbonate sulfate, NAN03).

m
z
0
m

0
co

m
z

o"11

BEVERIDGE ET AL ON EXPLOSIVE RESIDUES 4 4 7

FIG. 9--Sodium carbonate sulfate in aqueous extract of explosive residue (crossed polars,
original magnification x250).

FIG. lO--Damage by a unit charge o f 20, 40, and 60% dynamite and C4.
Numbers 16 through 30 in Table 3 summarize the test explosions of this class. In
only three instances (Nos. 18, 20, 26) was unreacted explosive recognized in the debris.
The confined explosives, Nos. 23 through 25, were distinguished from the unconfined
explosives by the composition of the reaction product as determined by X-ray powder
diffraction. The confined explosives gave sodium carbonate sulfate, whereas the unconfined gave Na~SO,(I). Unlike 20070 stumping powder, unconfined Forcite| residue

448

JOURNAL OF FORENSIC SCIENCES

TABLE 6--X-ray powder diffraction data of sodium carbonatesulfates.

2Na2SO, 9Na2CO3b

Na2SO,(I)a

d, .~

I/Io

d, A

I/Io

4.60
3.87

10
63

4.60
3.85

20
90

3.56

40

3.55

70

2.82
2.66

100
100

2.80
2.66

100
100

2.30
2.19
2.13

2
8
1

2.30
2.18

10
50

1.93

40

1.92

90

1.78

1.77

50

1.57
1.54

10
7

1.56
1.53

40
20

2Na2SO,9Na2CO3c (Na2SO,)I_ n(Na2CO3)nd

d, A

I/Io

3.88
3.78
3.51

70
80
70

2.78
2.64
2.58

100
70
80

2.13
2.10
1.97
1.93
1.89
1.75

30
30
10
40
40
30

1.62

1.50

10

d, A

I/Io

9.20
4.51

10
20

3.798
3.530
3.444
3.076
2.795
2.645
2.587
2.479
2.348
2.302

70
50
10
20
100
60
60
5
10
5

2.149
2.106
1.980
1.932
1.903
1.765
1.740

20
1
10
30
30
20
5

a ASTM Powder Diffraction File No. 1-0990 [27].

b Ramsdell [33].
c ASTM Powder Diffraction File No. 2-0840 [27].
dThis study.
showed no carbonate in the IR (Fig. 12). The presence of some carbonate was
inferred, however, from effervescence with acid. There was insufficient sodium nitrate
in the confined explosive residue for identification in the X-ray powder pattern, but it
was recognizable in the bulk sodium carbonate sulphate matrix by orthoscopic and
conoscopic observation with the polarizing microscope (Nos. 23 and 24).
Sodium nitrate was identified by X-ray powder diffraction in all unconfined explosions
and NG was identified by TLC in approximately three-quarters of the explosions.
However, ammonium nitrate and NC were detected in only 47 and 40% of the tests,
respectively. In the TLC systems tested, NC had an Ry value of 0, in common with many
solvent-extracted contaminants which often develop a yellow color with Griess reagent
and possibly mask NC. Where a TLC system was used which separates E G D N and NG,
(Nos. 17, 18, 22 through 28, 30) EGDN was found in only one-third of the cases in
which NG was identified. This cautions against complete reliance on a dynamite sniffer
to detect residue.
IR prior to TLC of ether and acetone extracts detected explosive components in 50%
of the tests, as opposed to 80% for TLC. In some TLC analyses, contamination reduced
the Rf value of explosive components relative to standards, and in these cases IR aided
interpretation.

BEVERtDGE ET AL ON EXPLOSIVE RESIDUES

449

FIG. 11--40% Forcite | debris (two sticks, unconfined).

No consistent difference in residue composition was found between Forcite| 40, 60, or
75070, notwithstanding their calculated products being in varying proportions. The
complexity of the Na2SO4/Na2CO3 aqueous system complicates this issue. For 60 and
7507o Forcite| the yield of sulfate product was minor compared to sodium nitrate, but
as No. 16 shows, this is not a consistent difference from 4007o Forcite| .
Explosions of 7507o Forcite| with and without wrapper paper (Nos. 29 and 30) showed
little difference in products, although CNS- was detected with paper only. CNS- was
also a product of black powder (No. 4), 20070 dynamite (No. 14), and 6007oForcite| (No.
28).
The effect of contamination was not serious, although sodium chloride of unknown
origin was often noted. Gypsum plaster in particular was noted in some X-ray powder
patterns but did not prevent identification of product sulfate in No. 28. Figure 13 shows
the IR spectrum of the aqueous extract from No. 27, 60~ Forcite| on gypsum wallboard, in which sulfate product is identified by a peak at 615 cm -l. Its low yield
precluded further identification from the X-ray powder pattern.
In summary, out of 15 explosions of ammonia-gelatin dynamite, 5 residues could be
classed as being from ammonia-gelatin dynamite (Nos. 18, 20, 26 through 28), 2 from
ammonia dynamite (Nos. 22, 29), 1 from gelatin dynamite (No. 17), and 7 from
dynamite (Nos. 16, 19, 21, 23-25, 30).
Gelatin--One pound of a 5-1b stick of seismic dynamite was shaped and detonated

unconfined, IR spectrum of aqueous extract from carpet (Na2SO, (1) and NAN03).

FIG. 13--60% Forcite | unconfined, differentiation of sulfates by IR analysis [NAN03; CaS04, 2H20 (contaminant); sodium
sulfate (product)].

FIG. 12--40% Forcite |

m
z
0
m

o_

0
Cr)

:73
m
Z

o- n

BEVERIDGE ET AL ON EXPLOSIVE RESIDUES

451

on steel (Table 3, No. 32). A heavy white crystalline residue was largely sodium
carbonate. This residue could not be classed as being from gelatin dynamite since NC was
not recovered, but the identification of sodium carbonate indicates that the dynamite
contained no elemental sulfur and again stresses the value of product identification by
instrumental analysis. The presence of barium was shown spectrographically in the
insoluble material, which emphasizes the value of systematic analysis.
Plastic C4--The velocity of C4 is around 25,000 fps. C4 was detonated by either
blasting cap or detonating cord booster. No unreacted explosive was observed in the
characteristic black deposit. The analyses are summarized in Table 4, Nos. 34 to 36. In
only one out of three explosions (No. 36) was RDX identified by TLC. On one piece
of debris, PETN from detonating cord booster was also identified.
Detonating Cord--The detonating cord was charged with PETN. In one of two
explosions (Table 4, Nos. 38 and 39) PETN was identified by TLC. In both cases
traces of wrapper were recovered.

General
The analytical scheme in Fig. 1 was derived from practical experience and the chemical
literature and uses the instruments routinely used in Royal Canadian Mounted
Police (RCMP) laboratories. It is not an absolute and is adapted as conditions demand.
For example, ether would not be used to extract residue from an engine block. Similarly,
if volatile liquids such as nitrobenzene or gasoline were suspected, then high vacuum
techniques, vapor phase IR, and gas chromatography would be applied. The test
explosions were part of a series of workshop seminars run by our laboratory for
training purposes, and involved police and fire investigators and IED specialists in
addition to laboratory personnel.
The analytical results show that the scheme permitted correct classification of the
explosives, although subclassification of dynamite was not always achieved. Unreacted
explosive in debris was recognized by microscopic examination for most of the low
explosives, few of the dynamites, and none of the plastic explosives. The value
of explosive product identification has been demonstrated and within the ammoniagelatin series it distinguished confined and unconfined dynamite. Residue composition
and distribution were not uniform and the most comprehensive recoveries were, not
surprisingly, made from the crater area, which receives a potential hemisphere of blast.
The most useful techniques were microscopy, TLC, X-ray powder diffraction, and
infrared spectroscopy. The emission spectrograph replaced spot tests for metals. Spot
tests provided rapid screening of a high yield of residue, but were usually relegated to
checking for components missed by the other techniques. A negative spot test was often
given more weight than a positive test. However, a spot test solution as a TLC developer
was very useful and greatly enhanced specificity.
Calcium sulfate building products, ammonium chloride from batteries, sodium
chloride of unknown origin, and resins were the contaminants most often encountered
by solvent extraction, but they did not preclude residue identification in the test
series.
Table 7 summarizes the components identified in this study and illustrates how
results may be interpreted in analysis of an unknown. The scheme has been
routinely applied to casework with satisfactory results,
Summary

A scheme for systematic analysis of explosive residues is presented and demonstrated

452

JOURNALOF FORENSIC SCIENCES

TABLE 7--Explosive components identified and their sources.

Component
NaNO3
NH4NO3
KNO3
K~SO,
Na2CO3
Na~SO,(I)
(Na2SO4)1- n(Na2CO3)n
NaC103
KC10,
NaC1
KC1
NH,C1
NO3NO2NH, +
CNSCO32SO, zC1Na
K
Ca
B
Ba
Pb
A1
Zn
EGDN
NG
NC
TNT
DNT
PETN
RDX

Probable Origin
dynamite
ammonia dynamite, blasting slurry
black powder
black powder, safety fuse, smokeless powders
dynamite
dynamite
dynamite
chlorate
chlorate
NaC103, "permitted" dynamite, contaminant
KC103
dry cell battery, "permitted" dynamite
dynamite, black powder
smokeless powders, dynamite
ammonia dynamite, NH,NO3 blasting slurry, dry cell battery
black powder, dynamite
dynamite
dynamite, black powder, smokeless powder
dry cell battery, "permitted" dynamite, contaminant
dynamite, contaminant
black powder
contaminant, dynamite
dynamite
dynamite
blasting cap residue
blasting cap casing
dry cell battery casing
dynamite
dynamite, double-base smokeless powder
dynamite, single-base smokeless powder
blasting slurry, military demolition charge
dynamite, single-base smokeless powder
detonating cord
C4 plastic explosive

by test explosions using commercial, military, and homemade explosives. The significance of reaction product identification is demonstrated.
Acknowledgments

The authors gratefully acknowledge the cooperation of the Base Commander and
EOD team of the Canadian Forces Base Edmonton, the Chief Constable and Bomb
Detail of the Edmonton City Police, and the Explosives Division of Canadian Industries
Ltd., Edmonton. We thank the Officer in Charge of the RCMP Crime Detection
Laboratory, Edmonton for his encouragement of this work and the Photography
Section for their technical assistance.

APPENDIX A
Glossary of Explosive Abbreviations and Terminology
Black P o w d e r - - K N O J C / S , usually 75/15/10

BEVERIDGE ET AL ON EXPLOSIVE RESIDUES

453

Blasting Cap--A small high explosive charge in a metal cylinder, used to initiate a bulk
high explosive charge by means of a shock wave; initiated electrically or by safety
fuse
Booster--A small high explosive charge used in conjunction with a blasting cap to ensure
detonation of a bulk high explosive charge
C--Carbon
C4--Military code for plastic explosive based on RDX
Chlorate/Sugar--Mechanical mixture of potassium or sodium chlorate and sugar
Detonating Cord--A linear explosive consisting of a series of cylindrical wrappings with
a central high explosive core, usually PETN
DNT--2,4-dinitrotoluene, CTH6(NO2)2
EGDN--Ethyleneglycol dinitrate, C2H4(ONO2)2
Foreite| --Trade name for dynamite (Canadian Industries Ltd.)
Geogel | --Trade name for dynamite (Canadian Industries Ltd.)
High Explosive--Explosive with a velocity of detonation greater than 3280 ft/s, for
example, dynamite and plastic explosive
IED--Improvised explosive device
KNO3--Potassium nitrate
Low Explosive--Explosive with a velocity of detonation less than 3280 ft/s, for
example, black powder and pistol and rifle smokeless powder propellants
NaNO3--Sodium nitrate
NC--Nitrocellulose, variable formulation
NG--Nitroglycerine, C3Hs(ONO2)3
NH,NO3--Ammonium nitrate
PETN--Pentaerythritol tetranitrate, C~Hs(ONO2),
Plastic Explosive--High explosive plus plasticizer to enable shaping of the charge,
for example, C4
Pola~ --Trade name for dynamite (Canadian Industries Ltd.)
RDX--Research and development explosive, cyclotrimethylenetrinitramine, C3H~N3(NO2)3
S--Sulfur
Safety Fuse--A series of cylindrical wrappings with a black powder core which burns at
a constant rate, usually 40 s/ft
Smokeless Powder (double base)--Propellant based on nitrocellulose and nitroglycerine
Smokeless Powder (single base)--Propellant based on nitrocellulose
TNT--2,4,6-trinitrotoluene, C7H~(NO2)3

APPENDIX B
Na2SO,(I) is a metastable form of sodium sulfate which is stabilized by carbonate

[27,32]. Sodium carbonate sulfate is reported as the rare mineral burkeite [27] and as a
synthetic material [33] isolated from the sodium carbonate/sodium sulfate aqueous
system [34]. Sodium carbonate sulfate was synthesized in our laboratory by evaporation
of an aqueous solution of sodium sulfate and sodium carbonate, with a sodium
carbonate mole ratio of 0.2 to 0.5. The X-ray powder pattern in this range showed
neither precursor and, due to isomorphous exchange of carbonate and sulfate [33],
showed no difference in d-spacings or intensities. Comparative d-spacings are given in
Table 6 for variously reported sodium carbonate sulfates. Our synthetic material was
indistinguishable from the carbonate sulfate residues noted in Table 3, Nos. 14 and
23 through 25.

454

JOURNALOF FORENSIC SCIENCES

References
[I] National Bomb Data Center, U.S. Department of Justice, Federal Bureau of Investigation,
Picatinny Arsenal, Dover, N.J.
[2] Canadian Bomb ~ ata Centre, Royal Canadian Mounted iPolice, Ottawa, Ont.
[3] "Blasting Cap Recognition and Identification Manual," Public Security Center, International
Association of Chiefs of Police, Gaithersburg, Md., 1973.
[4] "Introduction to Explosives," Public Security Center, International Association of Chiefs of
Police, Gaithersburg, Md., 1973.
[5] "Bomb Incident Management," Public Security Center, International Association of Chiefs of
Police, Gaithersburg, Md., 1973.
[6] "Bomb Incident Target Analysis," Public Security Center, International Association of Chiefs
of Police, Gaithersburg, Md., 1973.
[7] Tardif, H. P. and Sterling, T. S., Journal of Forensic Sciences, JFSCA, Vol. 12, No. 3, 1967,
pp. 247-272.
[8] Yallop, H. J., Journal of the Forensic Science Society, Vol. 5, No. 1, 1965, pp. 6-10.
[91 Jenkins, R. and Yallop, H. J., Explosivstoffe, No. 6, 1970, pp. 139-141.
[10] Hoffman, C. M. and Byall, E. B., Journal of Forensic Sciences, JFSCA, Vol. 19, No. 1,
Jan. 1974, pp. 54-63.
[lll Washington, W. D. and Midkiff, C. P., Journal of the Association of Official Analytical
Chemists, Vol. 55, No. 4, 1973, pp. 811-822.
[12] Cook, M. A., Science of High Explos&es, Rheinholdt, New York, 1958, Chapter 12;
Appendix 2, p. 389.
[13] Amas, S. A. H. and Yallop, H. J., Journal of the Forensic Science Society, Vol. 6, No.
4, 1966, pp. 185-188.
[14] Urbanski, T., Chemistry and Technology of Explosives, Pergamon Press, New York, 1967,
Vol. 3, Chapter 3.
[15] Lapagesse, E., International Criminal Police Review, No. 239, June-July 1970, pp. 176-183.
[16] Stettbacher, A., Protar, Vol. 10, 1944, pp. 158-164.
[17] Walker, J. T., Journal of Criminal Law and Criminology, Vol. 31, No. 4, 1940, p. 497.
[18] Maiti, P. C., Journal of the Forensic Science Society, Vol. 13, No. 3, 1973, p. 197.
[19] Canadian Industries Ltd., personal communication, Oct. 1970.
[20] Amas, S. A. H. and Yallop, H. J., Analyst, Vol. 94, No. 1122, 1969, p. 828.
[21] Sanger, D. G., Journal of the Forensic Science Society, Vol. 13, No. 3, 1973, p. 177.
[22] Kempe, A. R. and Tannert, W. K., Journal of Forensic Sciences, JFSCA, Vol. 17, No. 2,
April 1972, pp. 323-324.
[231 Lloyd, L B. F., Journal of the Forensic Science Society, Vol. 7, No. 4, 1967, p. 198.
[24] Pristera, J., Halik, M., Castelli, A., and Fredericks, W., Analytical Chemistry, Vol. 32, No.
4, 1960, pp. 495-508.
[25] Chason, D. E. and Norwitz, G., MicrochemicalJournal, Vol. 17, No. 1, 1972, pp. 31-60.
[26] Nyquist, R. A. and Kagel, R. O., Infrared Spectra oflnorganic Compounds, Academic Press,
New York, 1971.
[27] "ASTM Powder Diffraction File," Joint Committee on Powder Diffraction Standards,
Swarthmore, Pa., 1972.
[28] McCrone, W. C. and Delly, J. G., The Particle Atlas, 2nd ed., Ann Arbor Science Publishers,
Ann Arbor, 1973, Vol. 4.
[29] Byall, E. B., personal communication, Sept. 1973.
[30] Azaroff, L. A. and Buerger, M. J., The Powder Method, McGraw-Hill, New York, 1958,
Chapter 13.
[31] "Blasters Handbook," 6th ed., Canadian Industries Ltd., Explosives Division, Montreal,
1968.
[32] Kracek, F. C. and Ksanda, C. J., Journal of Physical Chemistry, Vol. 34, 1930, pp.
1741-1744.
[33] RamsdeU, L. S., American Mineralogy, Vol. 24, No. 2, 1939, pp. 109-115.
[34] Schroeder, W. C., Berk, A. A., and Gabriel, A., Journal of the American Chemical Society,
Vol. 58, No. 5, 1936, pp. 843-849.
Chemistry Section
Crime Detection Laboratory
Royal Canadian Mounted Police
Edmonton, Alta. TSJ 2N1
Canada