Full text of DOI:10.1093/chromsci/41.2.73

Journal of Chromatographic Science, Vol. 41, February 2003
The USACHPPM Gas Chromatographic Procedures for
the Analysis of Waters and Soils for Energetics and
Related Compounds
Richard W. Bishop, Michael A. Hable*, Curtis G. Oliver, and Robert J. Valis
USACHPPM, APG-EA, MD 21010-5422 (Building E2100)
Army Environmental Hygiene Agency (USAEHA) did begin to
publish some of its procedures, however, starting in the 1970s
Abstract
(4–7). T. Jenkins et al. (8–13) at the U.S. Army Corps of Engineers
Cold Regions Research and Engineering Laboratory (CRREL)
(Hanover, NH) have been regularly developing methods for ana-
lyzing munitions and making them public since the 1980s. The
U.S. Environmental Protection Agency (USEPA) has included sev-
eral of these methods into its SW-846 compendium of procedures
(14). SW-846 Method 8330 is a high-performance liquid chro-
matographic (HPLC) technique for analyzing soil and water
extracts for nitroaromatics and nitramines. It has been a popular
procedure for more than a dozen years, especially for analyzing
soil samples. Method 8332 is a variant on Method 8330 and is used
for the analysis of nitroglycerin using a different UV wavelength.
The more recent draft, Method 8095, was proposed in 1998; it
employs gas chromatography (GC) with electron capture detec-
tion (ECD) to analyze sample extracts. Soils are prepared in a sim-
ilar fashion as for Method 8330, but the waters are extracted using
solid-phase extraction techniques.
The
procedures currently used by the U.S. Army Center for Health
Promotion and Preventive Medicine (USACHPPM) for the analysis
of energetics and related compounds in water and soils are
presented. These procedures are based on the use of isoamyl
acetate to extract the analytes of interest from their environmental
matrices with subsequent analysis using gas chromatography with
electron capture detection. The suite of compounds included are
those that have been of environmental significance for years (such
as 2,4,6-trinitrotoluene, hexahydro-1,3,5-trinitro-1,3,5-triazine, and
dinitrotoluenes) and are the subject of several U.S. Environmental
Protection Agency SW-846 methods. The procedures presented in
this study are the product of years of development and refinement
of methods used for the analysis of many real-world samples by the
USACHPPM explosives analysis laboratory. The development,
performance, advantages, and details of these procedures are
described. The extension of these methods to the analysis of other
media is also briefly discussed.
The U.S. Army Center for Health Promotion and Preventive
Medicine (USACHPPM), which was formerly the USAEHA, never
adopted the water analysis portion of Method 8330, preferring its
own water procedures. The very earliest GC procedures that
USAEHA used for the determination of nitroaromatics in water
are primitive by today’s standards. Water samples were extracted
using benzene, and the extracts were analyzed using GC with
packed columns and flame ionization detection. The detection
limit for trinitrotoluene (TNT) at the time was approximately
1 part per million, and nitramines could not be detected at all
with the technique. Over the years, the procedure was modified to
use toluene instead of benzene (for reasons of safety); toluene was
satisfactory for the extraction of nitroaromatics, but not
nitramines. A search for a more suitable solvent eventually led to
isoamyl acetate, which was found to work well for nitroaromatics,
nitramines, and nitroglycerin. Capillary columns (first glass, then
fused silica) replaced packed columns and greatly improved the
chromatography. ECD replaced flame ionization and significantly
reduced detection limits. The suite of analytes was increased to
Introduction
The U.S. Army has been involved with the environmental mon-
itoring of residual energetics and related compounds for over
three decades. Concern for the health and safety of soldiers;
civilian workers; the general population; and the environment
near military bases, ammunition plants, and munitions testing
centers led to this monitoring requirement. Energetic com-
pounds and some of their breakdown products are of sufficient
toxicity and thus are chemicals of concern when present in the
environment (1–3). Methodology that can be used to analyze soil
and water samples for nitroaromatics, nitramines, and nitrate
esters was developed over the years by various army laboratories,
many of them being internal unpublished methods. The U.S.
* Author to whom correspondence should be addressed.
Reproduction (photocopying) of editorial content of this journal is prohibited without publisher’s permission.
73

Journal of Chromatographic Science, Vol. 41, February 2003
mirror the Method 8330 list. Inert injection port liners and seals
permitted the reliable analysis for the nitramine compounds.
Thus, the method that is presented in this study is one that has
evolved over several decades and has proven to be sensitive and
dependable for the analysis of water samples. USEPA Region 3 has
approved its use for the analysis of samples from specific sites
within its region.
until the water reached room temperature. The excess water was
withdrawn to bring the water level to the mark. Forty microliters
of 3,4-dinitrotoluene (DNT) at 0.30 µg/mL in acetonitrile was
added to the flask as a surrogate compound. One milliliter of
isoamyl acetate (anhydrous, 99+%) (Aldrich, Milwaukee, WI) was
then added to the flask and the flask capped and placed on a rotary
shaker for 30 min. The sample was shaken at a speed of approxi-
mately 15 rpm. At the end of that time, the sample was removed
from the shaker and allowed to stand until the isoamyl acetate
and water layers separated. The isoamyl acetate portion was trans-
ferred with a Pasteur pipette to an autosampler vial for GC anal-
ysis. Usually a standard 2-mL autosampler vial is satisfactory, but
if a sample has a severe emulsion then a limited insert vial may be
required. A laboratory deionized water blank, laboratory control
sample(s), matrix spikes, and the standards to be analyzed were
prepared using this same procedure.
The standards were made by injecting varying amounts of a
mixed component spiking solution into six flasks, each con-
taining 100 mL deionized water. The spiking solution was pre-
pared in acetonitrile by the dilution of 1.0-mg/mL individual
standards of Method 8330 compounds from AccuStandard (New
Haven, CT), 3,4-DNT (Aldrich), and nitroglycerin from Cerilliant
(Austin, TX) up to 50 mL. The spiking solution contained
nitrobenzene, 2,4,6-TNT, and 1,3,5-trinitrobenzene at 0.60 µg/mL
(using 30 µL); 1,3-dinitrobenzene, nitroglycerin, and the nitro-
toluene isomers at 1.8 µg/mL (using 90 µL); 2,4-DNT at 0.40
µg/mL (using 20 µL); 2,6-DNT at 0.20 µg/mL (using 10 µL); 3,4-
DNT at 0.30 µg/mL (using 15 µL); 4-amino-2,6-DNT, 2-amino-
4,6-DNT, and tetryl at 3.0 µg/mL (using 150 µL);
hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) at 2.4 µg/mL
(using 120 µL); and octahydro-1,3,5,7-tetranitro-1,3,5,7-tetra-
zocine (HMX) at 48.0 µg/mL (using 2.4 mL). This solution
was added to the 6 flasks using 2.5, 5.0, 10.0, 25.0, 50.0, and 100
µL, respectively. The low standard was equivalent to one-half of
the reporting limit for most of the analytes. Finally, the same
spiking solution was used for matrix spiking in water samples
(using 40 µL).
USACHPPM has used Method 8330 for many years for the anal-
ysis of soil samples and used its internal procedure equivalent to
what has become draft Method 8095 to perform confirmatory
analyses on the soil extracts and to analyze for nitroglycerin. The
soil procedure that USACHPPM is now employing was developed
out of the requirement to achieve lower detection limits than
Method 8330 provides and the desire to use GC rather than
HPLC. There have always been some difficulties with Method
8330 in terms of the detection of false positives with the UV
detector used with the procedure because this detector is not very
specific. Another problem we have faced with the HPLC proce-
dure is the variability between columns in their ability to obtain
separation of all the analytes. We have found that we must revise
the operating conditions for each new column or whenever we
store a column for any duration. These problems, coupled with
our requirement to set up and perform rapid analyses for envi-
ronmental contaminants in areas containing deployed soldiers,
led us to pursue GC as the instrumental technique of choice for
analyzing soil extracts. The GC draft Method 8095 has its advan-
tages of sensitivity and selectivity over Method 8330. However, its
use of acetonitrile as the solvent can sometimes be a problem
with GC (as will be discussed). The USACHPPM soil procedure
described in this study is, in many ways, a modification of the
water procedure. It employs a similar chromatographic approach
and it uses isoamyl acetate as an extraction solvent. It has been
tested with various soil types and directly compared with Method
8095. The results have been favorable (as will be shown).
It should also be noted that GC–mass spectrometry (MS) has
been proposed as an alternative to GC–ECD (15). Our laboratory
does use it as a tool for confirming positive sample results, espe-
cially in samples containing significant interferences. For this
study, however, GC–MS (or GC–MS–MS) has not yet been proven
to be as economical or practical to use for the routine quantitation
of the suite of analytes that are analyzed by the other methods.
The chromatographic methodologies used for water and soil
have also been incorporated into the USACHPPM procedures for
atmospheric sampling, as described elsewhere (16). They have
also been used for the analysis of a variety of munitions destruc-
tion process waste samples. This particular application is briefly
described in this study.
Preparation of soil samples
Soil (and sediment) samples were extracted within 14 days of
their collection and kept refrigerated at 4ºC 2ºC until time of
processing. An aliquot of sample (at least 10 g) was placed in a dis-
posable aluminum weighing pan and air dried to a constant
weight (generally overnight). The dried soil was carefully ground
(if necessary) in a clean mortar, homogenized, and sieved to pass
through a 40-mesh sieve. A subsample of this sieved fraction was
placed in a 40-mL amber glass vial with a Teflon-lined screw cap.
Generally, a 2.0-g portion of sample was taken for analysis, but up
to a 5.0-g sample may be used when lower reporting limits are
desired. Twenty milliliters of deionized water were added to the
vial and 10 µL of 3,4-DNT at 0.50 µg/µL in isoamyl acetate were
injected into the soil as a surrogate compound. Finally, 5.0 mL of
isoamyl acetate was added to the vial. The vial was capped and set
into an ultrasonic bath for a minimum of 12 h and then on a
rotary shaker for two more hours. The sample was removed from
the shaker and allowed to stand until the isoamyl acetate and
water layers separated. It has been found that placing the vial in a
refrigerator for several hours usually helps to produce a very clear
Experimental
Preparation of water samples and standards
Water samples were extracted within seven days of their collec-
tion, and kept refrigerated at 4ºC 2ºC until time of extraction.
The extraction was conducted in a 100-mL screw-top volumetric
flask with a Teflon-lined screw cap. The flask was filled with
sample beyond the meniscus mark and then allowed to stand
74

Journal of Chromatographic Science, Vol. 41, February 2003
isoamyl acetate phase, requiring no filtration or centrifugation.
A portion of the isoamyl acetate extract was transferred with a
Pasteur pipette to an autosampler vial for GC analysis. A labora-
tory blank soil, laboratory spiked control soil, and matrix spikes
were prepared by using the same procedure with the spiking done
directly into the soils prior to the addition of water.
Standards for the soil’s analysis, such as were used for these
studies, were prepared by the serial dilution in isoamyl acetate of
a 1.0-mg/mL mix of Method 8330 compounds, a 1.0-mg/mL indi-
vidual HMX solution, a 1.0-mg/mL standard of 3,4-DNT from
Aldrich, and a 1.0-mg/mL sample of nitroglycerin from Cerilliant.
Standards were made at 2.0, 1.0, 0.5, 0.2, 0.1, 0.02, and 0.01
µg/mL of all the compounds except HMX, which was twice the
concentration of the others. The Method 8330 mix and nitroglyc-
erin were also used for matrix spiking; they were first diluted
together 1/10 in isoamyl acetate and 50 µL used per spike.
Chromatography
Analyses for the Method 8330 target list of compounds, nitro-
glycerin, and the surrogate compound have been conducted
using Agilent Technologies (Palo Alto, CA) (formerly Hewlett
Packard) Model 6890 GCs equipped with ECDs. Although the
standards concentrations differed between the water and soil
analyses, the chromatography was similar between the two
methods. It will generally vary according to the individual GC
used and the nature of the samples, but the basic column types
and conditions used in our laboratory are as follows.
Figure 1. Chromatogram for energetics analysis in water on a 7-m, 0.53-mm-
i.d., 1.0-µm-film dimethylpolysiloxane column with ECD. Temperature pro-
grammed in steps from 80°C to 200°C, and He carrier pressure programmed
from 2.0 to 4.0 psig. The respective components peaks had the following
retention time values: nitrobenzene (1), 1.57; 2-nitrotoluene (2), 2.03; 3-nitro-
toluene (3), 2.29; 4-nitrotoluene (4), 2.41; nitroglycerin (5), 3.26; 1,3-dini-
trobenzene (6), 3.97; 2,6-dinitrotoluene (7), 4.13; 2,4-dinitrotoluene (8), 4.84;
3,4-dinitrotoluene (9), 5.37; 1,3,5-trinitrobenzene (10), 6.32; 2,4,6-trinitro-
toluene (11), 6.88; RDX (12), 8.62; 4-amino-2,6-dinitrotoluene (13), 11.04; 2-
amino-4,6-dinitrotoluene (14), 12.14; tetryl (15), 13.80; and HMX (16),
19.56.
Figure 2. Chromatogram for energetics verification analysis on a 7–8-m, 0.53-
mm-i.d., 1.0-µm-film 50% trifluoropropyl–methylpolysiloxane column with
ECD. Temperature programmed from 80°C to 225°C, and H2 carrier pressure
programmed from 3.0 to 9.0 psig. The respective components peaks had the
following retention time values: nitrobenzene (1), 1.58; 2-nitrotoluene (2),
1.86; 3-nitrotoluene (3), 2.12; 4-nitrotoluene (4), 2.25; 2,6-dinitrotoluene (5),
4.17; nitroglycerin (6), 4.31; 1,3-dinitrobenzene (7), 4.39; 2,4-dinitrotoluene
(8), 4.89; 3,4-dinitrotoluene (9), 5.66; 2,4,6-trinitrotoluene (10), 6.53; 1,3,5-
trinitrobenzene (11), 6.75; 4-amino-2,6-dinitrotoluene (12), 7.21; 2-amino-
4,6-dinitrotoluene (13), 7.65; RDX (14), 7.86; tetryl (15), 8.68; and HMX (16),
12.67.
75

Journal of Chromatographic Science, Vol. 41, February 2003
Primary runs were typically made using a dimethylpolysiloxane
column (0.53-mm i.d., 1.0-µm film thickness) cut to 7 m in
length (see Figure 1 for an example of the chromatography). The
GC oven was temperature programmed from 80ºC at 15º/min to
140ºC, then at 3ºC/min to 170ºC, and finally at 5.0ºC/min to 200ºC
(held 3.0 min). The helium carrier gas was programmed from 2.0
psig (held 13.0 min) to 4.0 psig at a rate of 150 psig/min and then
held. The dilution gas was nitrogen set at 30 mL/min. The injec-
tion port temperature was set at 225ºC; the injection port liner
was a silanized glass 4-mm liner or a Silcosleeve (Restek,
Bellefonte, PA) with a Silcosteel seal used in splitless mode. The
Ni-63 ECD temperature was 250ºC. Data processing was done
using TotalChrom (PE Nelson, Shelton, CT). An Agilent Model
7673A autosampler was used to make the injections; the injection
volume was 1.0 µL for the soil extracts and 2.0 µL for water
extracts.
Confirmation column analyses were done on all positive sam-
ples. Columns that have proven useful for this contain 50%
trifluoropropyl–methylpolysiloxane or 50% phenyl–methylpoly-
siloxane liquid phases. Figures 2 and 3 provide chromatographic
examples. The 50% trifluoropropyl–methylpolysiloxane column
was 0.53-mm i.d. and 1.0-µm film thickness cut to 7–8 m in
length; it was temperature programmed from 80ºC at 15º/min to
225ºC and held for 7 min. It used hydrogen carrier pressure pro-
grammed from 3.0 to 9.0 psig at a rate of 10.0 psig/min and held
Table I. Method Detection and Reporting Limits for the
USACHPPM Water and Soils Procedures
Water (µg/L)
Soil* (µg/g)
Compound
MDL
MRL
MDL
MRL
2,6-DNT
2,4-DNT
2,4,6-TNT
RDX
0.001
0.003
0.004
0.016
0.58
0.016
0.044
0.035
0.004
0.011
0.004
0.023
0.026
0.023
0.015
0.01
0.02
0.03
0.10
3.0
0.09
0.09
0.09
0.03
0.09
0.03
0.10
0.10
0.50
0.09
0.004
0.006
0.007
0.006
0.023
0.006
0.005
0.006
0.005
0.005
0.006
0.026
0.009
0.012
0.025
0.01
0.02
0.01
0.01
0.05
0.02
0.02
0.02
0.02
0.02
0.02
0.05
0.02
0.02
0.05
HMX
2-Nitrotoluene
3-Nitrotoluene
4-Nitrotoluene
Nitrobenzene
1,3-Dinitrobenzene
1,3,5-TNT
4-Amino-2,6-DNT
2-Amino-4,6-DNT
Tetryl
Nitroglycerin
* Soil data based on a 5.0-g sample.
Table II. Precision and Accuracy Data for the
USACHPPM Water Method*
Concentration
range (µg/L)
Compound
%Recovery
%RSD†
2,6-DNT
2,4-DNT
2,4,6-TNT
RDX
98.8
99.4
88.5
96.2
97.0
97.7
100.0
99.8
97.4
98.5
98.4
90.5
95.5
96.5
90.3
3.49
3.21
9.60
4.67
5.18
6.08
9.97
8.91
5.94
4.17
9.88
18.7
12.2
8.82
10.2
0.03–0.08
0.06–0.16
0.09–0.24
0.36–0.96
6.0–20
0.27–0.72
0.27–0.72
0.27–0.72
0.09–0.24
0.27–0.72
0.09–0.24
0.3–0.8
HMX
2-Nitrotoluene
3-Nitrotoluene
4-Nitrotoluene
Nitrobenzene
1,3-Dinitrobenzene
1,3,5-Trinitrobenzene
4-Amino-2,6-DNT
2-Amino-4,6-DNT
Tetryl
Figure 3. Chromatogram for energetics verification analysis on a 10-m, 0.53-
mm-i.d., 1.0-µm-film 50% phenyl–methylpolysiloxane column with ECD.
Temperature programmed from 80°C to 200°C, and He carrier pressure pro-
grammed from 2.0 to 6.0 psig. The respective components peaks had the fol-
lowing retention time values: nitrobenzene (1), 0.67; 2-nitrotoluene (2), 0.84;
3-nitrotoluene (3), 1.11; 4-nitrotoluene (4), 1.25; nitroglycerin (5), 2.25; 1,3-
dinitrobenzene (6), 3.01; 2,6-dinitrotoluene (7), 3.09; 2,4-dinitrotoluene (8),
3.56; 3,4-dinitrotoluene (9), 4.07; 2,4,6-trinitrotoluene (10), 5.12; 1,3,5-trini-
trobenzene (11), 5.12; RDX (12), 7.30; 4-amino-2,6-dinitrotoluene (13), 7.50;
2-amino-4,6-dinitrotoluene (14), 7.62; tetryl (15), 8.18; and HMX (16), 18.50.
0.3–0.8
0.47–1.2
0.27–0.72
Nitroglycerin
* Data compiled from 27 controls analyzed during 2000–2002.
† RSD, relative standard deviation.
76

Journal of Chromatographic Science, Vol. 41, February 2003
and nitrogen diluted at 40 mL/min. Other conditions were the
sameasdescribedpreviously. The50%phenyl–methylpolysiloxane
column was 0.53-mm i.d., 1.0-µm film thickness, and cut to 10 m
inlength;itwastemperatureprogrammedfrom80ºCat20º/minto
140ºC and then at 6ºC/min to 200ºC and held for 10 min. It used
helium carrier pressure programmed from 2.0 to 6.0 psig at a rate
of 150 psig/min and held and nitrogen diluted set to 30 mL/min.
Other conditions were the same as described previously.
lowest standard run with the method because these can reliably
be obtained with most instrumentation and for most samples.
Precision and accuracy data for both methods are shown in Tables
II and III; these data were compiled from numerous spiked con-
trols analyzed over the past year. Finally, Tables IV and V provide
some comparison data on actual contaminated soils analyzed by
both the USACHPPM and draft Method 8095 or Method 8330 soil
procedures.
Method detection limit (MDL) studies have been conducted on
both methods using the 40CFR part 136 EPA procedure (17).
Results for recent MDL determinations are provided in Table I, as
well as the method reporting limits (MRLs) USACHPPM uses. The
MRLs were conservatively set to be the equivalent of or above the
Results and Discussion
The data shown in Tables I–III indicate that the USACHPPM
water and soil methods were both sensitive and capable of good
precision and accuracy. The precision and accuracy data were the
results of controls run with samples over the time these partic-
ular procedures have been used. The comparison data between
the USACHPPM soils and draft Method 8095 showed good agree-
ment. Similarly, the data for the field samples analyzed by the
USACHPPM and HPLC soil procedures compared favorably.
One of the advantages of the USACHPPM procedures over the
EPA SW846 energetics procedure lies in the preparation proce-
dures used. The water extraction technique is very simple and
quick to perform because there are no solvent concentration
steps involved; a person can routinely process dozens of samples
a day. There are infrequent problems with emulsions forming at
the water–isoamyl acetate interface, but when they do occur they
Table III. Precision and Accuracy Data for the
USACHPPM Soil Method*
Concentration
range (µg/g)
Compound
%Recovery
%RSD†
2,6-DNT
2,4-DNT
2,4,6-TNT
RDX
101.7
89.5
86.5
7.4
10.2
12.8
7.0
19.1
8.3
8.2
8.6
6.8
10.4
14.0
10.2
8.0
11.4
16.2
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
1.2–6.0
94.8
HMX
105.6
104.1
102.1
101.4
97.6
94.9
76.9
83.8
92.4
2-Nitrotoluene
3-Nitrotoluene
4-Nitrotoluene
Nitrobenzene
1,3-Dinitrobenzene
1,3,5-Trinitrobenzene
4-Amino-2,6-DNT
2-Amino-4,6-DNT
Tetryl
Table V. Comparison Data for Soils Analyzed by the
USACHPPM and HPLC Methods*
Site
Compound
USACHPPM
HPLC data
103.4
97.4
Nitroglycerin‡
WHG1
HMX
0.49
0.91
0.04
0.02
0.82
2.7
0.20
0.04
3.9
19
2.6
0.32
1.1
4.0
0.38
0.78
0.02
0.02
0.73
2.4
0.13
0.03
3.5
23
2.2
0.41
1.2
4.9
RDX
2,4,6-TNT
TNB
* Data compiled from 21 controls analyzed during 2001–2002.
† RSD, relative standard deviation.
‡ The data for nitroglycerin was calculated from 17 controls.
WHG2
WHG5
LMW5
HMX
RDX
2,4,6-TNT
TNB
HMX
RDX
2,4,6-TNT
TNB
Table IV. Comparison Data for Soils Analyzed by the
USACHPPM and 8095 Methods*
Site
Compound
USACHPPM
Draft 8095
4935005
HMX
RDX
2,4,6-TNT
HMX
RDX
2,4,6-TNT
HMX
RDX
2,4,6-TNT
HMX
RDX
1.3
0.86
0.29
1.3
0.66
0.53
0.28
0.61
0.24
0.024
< 0.1
0.38
0.038
< 0.1
0.28
0.018
< 0.1
0.15
HMX
RDX
2,4,6-TNT
TNB
0.94
0.27
0.1
0.02
0.3
1.9
0.24
0.1
0.03
0.3
4935007
5005034
5005035
5005037
0.18
0.063
0.32
0.40
0.065
0.28
0.23
0.014
< 0.05
0.15
L33+2
L34+2
2,4,6-TNT
RDX
2,4,6-TNT
TNB
0.02
0.01
0.04
380
2300
110
0.01
0.02
0.05
341
2400
120
L35+2
C-Line
RDX
2,4,6-TNT
HMX
2,4,6-TNT
HMX
RDX
RDX
2,4,6-TNT
* Field samples and HPLC data provided by CRREL (µg/g).
* Field samples collected during 2002 and analyzed by USACHPPM (µg/g).
77

Journal of Chromatographic Science, Vol. 41, February 2003
can easily be overcome using techniques such as the addition of a
little sodium sulfate or centrifugation. This problem sometimes
occurs during the analysis of surface waters. The standards for the
water analyses are extracted from water rather than prepared
directly in isoamyl acetate. This was instituted many years ago
when there was a question as to the extraction efficiencies for
some of the compounds at the varying concentrations. In reality,
nonextracted standards would probably be acceptable in most
cases, but we have not pursued this. The varying amounts of the
compounds in the standards were based on the variability in
required reporting levels of the compounds in water plus the vari-
ability in the ECD response.
The preparation of the soils was also simple to perform. The
ultrasonic bath was used to produce good soil–water interaction;
it may not be necessary to do it for a full 12 h, but that was the
time selected out of convenience. The shaking step accomplished
the actual extraction of the analytes into the isoamyl acetate.
Unlike water, the standards for the soil analyses were not
extracted because it was not practical being that the soils were so
variable in composition. The compounds, with the exception of
HMX, were present at equal concentrations because there have
been no required reporting limits in soil based on health concerns
(unlike the situation with water).
The soil data presented in this study were all gathered from
soils extracted using the described procedure. Recent tests were
made on some of the same soils whereby the extraction vials were
laid on their sides on a platform shaker and shaken for 3 h. The
results were comparable to the ultrasonic bath–rotary shaker
technique, indicating that this may be a further simplified proce-
dure for soil extractions.
Several GC issues must be addressed in order to obtain good
chromatography with these procedures. The nonpolar (dimethyl-
polysiloxane) primary column used for energetics analysis was
capable of separating all the analytes of interest in a relatively
short time. However, there were occasionally background inter-
ferences with the peaks for the early eluting nitrotoluene isomers
depending on the lot of isoamyl acetate used. The isoamyl acetate
must be of high purity. A column containing a different liquid
phase was used to quantitate compounds that could not be deter-
mined with the primary column. Secondary column analysis was
also routinely done to verify positive detections on the primary
column. Polar (50% trifluoropropyl–methylpolysiloxane) or
intermediate polarity (50% phenyl–methylpolysiloxane) columns
were useful for this purpose. Chromatographic conditions can be
varied, but a short column is required if HMX verification is
required. HMX is very reactive; a fast flow rate and temperature
program is required to get it through the polar column before
peak degradation begins to occur.
Another important factor that must be considered when per-
forming GC analyses for energetic compounds is the use of a clean,
properly silanized injection port liner and inert injection port seal.
Commercially prepared liners and seals are recommended. Peaks
for the more reactive compounds, especially HMX and the amino-
dinitrotoluene isomers, will show distorted peak shapes or disap-
pear entirely if the liner or seal is dirty or not silanized. Oncolumn
injections are not recommended with this analysis because non-
volatile compounds deposited at the head of the column produce
problems with sensitivity and reproducibility.
It should be noted that the compound that was identified as
“tetryl” by GC is actually not tetryl but probably a thermal trans-
formation product, possibly N-methyl picramide (18). Because
the transformation appeared to be complete, it was a satisfactory
measure of the amount of tetryl in a sample.
A major difference between draft Method 8095 and the
USACHPPM soil method involves the extraction solvents used.
Method 8095 uses acetonitrile for extraction and injection into a
GC. Acetonitrile is a good extraction solvent and is quite appro-
priate for HPLC usage (19), but it is not necessarily the best for use
with GC systems. For example, it is not recommended for use with
Agilent GC–MSD systems because it is corrosive. As mentioned
previously, our laboratory has used variants of draft Method 8095
for some time to confirm HPLC soil results, but found some prob-
lems with reproducibility and rapid deterioration of column per-
formance. These difficulties have not been present since we began
to use isoamyl acetate, and we have observed higher recoveries for
some compounds with the use of this solvent.
The information presented in this study concerns the analyses
of environmental samples for the suite of Method 8330 com-
pounds plus nitroglycerin. However, the USACHPPM laboratory
also performed analyses for the same compounds recently in sev-
eral thousand samples of considerably different matrices. This
was done from 1999 until early 2002 in support of the U.S. Army’s
Assembled Chemical Weapons Assessment project, which investi-
gated different techniques for the demilitarization of chemical
munitions. The samples ranged from highly caustic to highly
acidic aqueous solutions, biofeedstock solutions, and sludges to
air samples collected on XAD-2 resin and in impingers containing
water. The basic methodologies described previously were suc-
cessfully modified to permit the analyses of these samples. Liquid
samples were pH adjusted to be slightly acidic and extracted in the
same way as water samples, and sludges were done in much the
same manner as the soils procedure. Air samples were also
extracted with isoamyl acetate. The chromatographic procedure
was virtually the same as described previously, although admit-
tedly much more difficult. Many sample dilutions and alternate
column injections had to be made because of interferences or
high levels of target analytes. The important fact to consider, how-
ever, is that the analytical approach described in this study
worked well, even with these most difficult of matrices. Table VI
displays the synopsis of the results from the matrix spikes run on
caustic hydrolysates during 1999–2001 for the surrogate and the
two key analytes from the energetic feed material.
Table VI. Recovery Data for Spikes into Caustic
Hydrolysate Samples*
RDX
2,4,6-TNT
3,4-DNT
Mean recovery
Standard deviation
No. of spikes
110
10.5
53
103
7.8
61
99
15.8
58
* Data compiled from samples analyzed during 1999–2001. The concentration range
was 0.1–1.0 mg/L for all analytes and analyzed by variation of the USACHPPM water
method.
78

Journal of Chromatographic Science, Vol. 41, February 2003
6. F. Belkin, R.W. Bishop, and M.V. Sheely. Analysis of explosives in
water by capillary gas chromatography. J. Chromatogr. Sci. 24:
532–34 (1985).
7. M. Hable, C. Stern, C. Asowata, and K. Williams. The determination
of nitroaromatics and nitramines in ground and drinking water by
widebore capillary gas chromatography. J. Chromatogr. Sci. 29:
131–35 (1991).
8. T.F. Jenkins, C.F. Bauer, D.C. Leggett, and C.L. Grant. Reverse phase
HPLC method for analysis of TNT, RDX, HMX and 2,4-DNT in muni-
tions wastewater, CRREL Special Report 84-29. U.S. Army Cold
Regions Research and Engineering Laboratory, Hanover, NH, 1984.
9. T.F. Jenkins, M.E. Walsh, P.W. Schumacher, P.H. Miyares, C.F. Bauer,
and C.L. Grant. Liquid chromatographic method for the determina-
tion of extractable nitroaromatic and nitramine residues in soil.
J. AOAC 72: 890–99 (1989).
10. S.-I. Nam. Evaluation of Thin-Layer Chromatography for
Confirmation of Analyte Identity, CRREL Special Report 97-21. U.S.
Army Cold Regions Research and Engineering Laboratory, Hanover,
NH, 1997.
11. P.G. Thorne and K.F. Myers. Evaluation of Commercial Enzyme
Immunoassays for the Field Screening of TNT and RDX in Water.
CRREL Special Report 97-32. U.S. Army Cold Regions Research and
Engineering Laboratory, Hanover, NH, 1997.
12. M.E. Walsh and T. Ranney. Determination of nitroaromatic,
nitramine, and nitrate ester explosives in water using solid-phase
extraction and gas chromatography–electron capture detection:
comparison with high-performance liquid chromatography.
J. Chromatogr. Sci. 36: 406–16 (1998).
13. M.E. Walsh and T. Ranney. Determination of Nitroaromatic,
Nitramine, and Nitrate Ester Explosives in Soils Using GC–ECD.
CRREL Special Report 99-12. U.S. Army Cold Regions Research and
Engineering Laboratory, Hanover, NH, 1999.
Conclusion
The primary procedures that the USACHPPM uses for the anal-
yses of energetics and related compounds have been described in
this article. These procedures have been tested and refined over
the past few years to the point in which we believe they are accept-
able alternates to any methods currently available for this pur-
pose. The procedures are relatively easy to perform in both
sample preparation and the chromatographic analytical
approach. The test data supports the conclusion that the method-
ology can provide sensitive and accurate analytical data for water
and soil matrices and can be modified to accommodate other,
more complex types of samples. It is hoped that the information
provided will add to the arsenal of methods for those analysts
tasked with measuring nitroaromatics, nitramines, and nitro-
glycerin in the environment.
Acknowledgments
The authors wish to express their appreciation to both Thomas
F. Jenkins and Marianne E. Walsh of CRREL (Hanover, NH). Their
advice, support, and assistance over the course of our years of
explosives analyses and methods development have been
extremely helpful.
14. U.S. Environmental Protection Agency. Test Methods for Evaluating
Solid Waste, Physical/Chemical Methods, SW-846 Update III. Office
of Solid Waste, Washington, D.C., 1997.
15. J. Yinon. Trace analysis of explosives in water by gas chromatog-
raphy–mass spectrometry with a temperature-programmed injector.
J. Chromatogr. A 742: 205–209 (1996).
16. M.A. Hable, J.B. Sutphin, C.G. Oliver, E.F. Gordon, and R.W. Bishop.
A procedure for sampling and analysis of air for energetics and
related compounds. J. Chromatogr. Sci. 40: 77–82 (2002).
17. Environmental Protection Agency. Appendix B to Part 136–Definition
and Procedure for the Determination of the Method Detection
Limit–Revision 1.11. 40 CFR, 7-1-91 ed., Ch. 1, pp. 554–56.
18. M.E. Walsh. Environmental Transformation Products of Nitroaro-
matics and Nitramines. Literature Review and Recommendations for
Analytical Method Development. CRREL Special Report 90-2. U.S.
Army Cold Regions Research and Engineering Laboratory, Hanover,
NH, 1990.
References
1. W.G. Palmer, M.J. Small, J.C. Dacre, and J.C. Eaton. “Toxicology
and Environmental Hazards.” In Organic Energetic Compounds.
Nova Science Publishers, Hauppauge, NY, 1996, Chapter 5, pp.
289–372. ISBN 1-56072-201-0.
2. D.H. Rosenblatt, E.P. Burrows, W.R. Mitchell, and D.L. Parmer.
“Organic Explosives and Related Compounds”. In The Handbook of
Environmental Chemistry. Springer-Verlag, Berlin, Germany, 1991.
Vol. 3, Part G, pp. 195–234.
3. D.L. Kaplan and A.M. Kaplan. 2,4,6-Trinitrotoluene–surfactant com-
plexes: decomposition, mutagenicity, and soil leaching studies.
Environ. Sci. Technol. 16(9): 566–71 (1982).
4. N.B. Jurinski, G.E. Podolak, and H.L. Hess. Comparison of analytical
methods for trace quantities of 2,4,6-trinitrotoluene. Am. Ind. Hyg.
J. 36: 497–502 (1975).
19. T.F. Jenkins and C.L. Grant. Comparison of extraction techniques for
munition residues in soil. Anal. Chem. 59(9): 1326–31 (1987).
5. R. Bongiovanni, G.E. Podolak, L.D. Clark, and D.T. Scarborough.
Analysis of trace amounts of six selected poly-nitro compounds in
soils. Am. Ind. Hyg. Assoc. J. 45(4): 22–226 (1984).
Manuscript accepted December 30, 2002.
79