metals in conjunction with explosives at muni-
OVERVIEW OF SAMPLING AND ANALYSIS
tions sites.
FOR EXPLOSIVES IN SOIL
The frequency of occurrence of specific explo-
sives in soils was assessed by Walsh et al. (1993),
The environmental characteristics of munitions
who compiled analytical data on soils collected
compounds in soil indicate that they are extremely
from 44 Army ammunition plants, arsenals, and
heterogeneous in spatial distribution. Concentra-
depots, and two explosive ordnance disposal sites.
tions range from nondetectable levels (<0.5 parts
Of the 1155 samples analyzed by EPA Method
per million [ppm]) to percent levels (>10,000 ppm)
8330, 319 samples (28%) contained detectable
for samples collected within several feet of each
levels of explosives. The frequency of occurrence
other. Some waste disposal practices, such as OB/
and the maximum concentrations detected are
OD, exacerbate the problem at these sites and may
shown in Table 2. TNT was the most commonly
result in conditions ranging from no soil contami-
occurring compound in contaminated samples; it
nation up to solid "chunks" of bulk secondary
was detected in 66% of the contaminated samples
explosives, such as TNT or RDX. Secondary ex-
and in 80% of the samples if the two explosive ord-
plosives concentrations above 10% (>100,000 ppm)
nance disposal sites are excluded. Overall, either
in soil are also of concern from a potential reactiv-
TNT or RDX or both were detected in 72% of the
ity standpoint and may affect sample and materi-
samples containing explosives residues, and 94%
als handling processes during remediation. An
if the ordnance sites are excluded. Thus, by screen-
explosives hazard safety analysis is needed for
ing for TNT and RDX at ammunition plants, arse-
materials handling equipment to prevent initiat-
nals, and depots, 94% of the contaminated areas
ing forces that could propagate a detonation
could be identified (80% if only TNT was
throughout the soil mass.
determined). This demonstrates the feasibility of
Reliance on laboratory analyses only for site
screening for one or two compounds or classes of
characterization may result in a large percentage
compounds to identify the initial extent of con-
of samples (up to 80%, depending upon the site)
tamination at munitions sites. The two ordnance
with nondetectable levels. The remaining samples
sites were predominantly contaminated with
may indicate concentrations within a range of four
DNTs, probably from improper detonation of
orders of magnitude. Analyzing a small number
waste propellant. The table also shows that NB
of samples at an off-site laboratory may result in
and NTs were not detected in these samples; how-
inadequate site characterization for estimating soil
ever, NTs are found in waste produced from the
quantities for remediation and may miss poten-
manufacture of DNT.
tially reactive material. Laboratory analytical costs
vary depending on required turnaround time.
Typical costs for EPA Method 8330 analysis range
Table 2. Occurrence of analytes detected in
from 0 to 0 per sample for 30-day turn-
soil contaminated with explosives.
around, 0 to 0 for 7-day turnaround, and
approximately 00 per sample for 3-day turn-
Sample with
Maximum
around, if it is available.
analyte present
level
(g/g)
Because of the extremely heterogeneous distri-
Compound
(%)
bution of explosives in soils, on-site analytical
Nitroaromatics
methods are a valuable, cost-effective tool to as-
TNT
66
102,000
sess the nature and extent of contamination. Be-
TNB
34
1790
cause costs per sample are lower, more samples
DNB
17
61
can be analyzed and the availability of near-real-
2,4-DNT
45
318
time results permits redesign of the sampling
2,6-DNT
7
4.5
2-AmDNT
17
373
scheme while in the field. On-site screening also
4-AmDNT
7
11
facilitates more effective use of off-site laborato-
Tetryl
9
1260
ries using more robust analytical methods. Even
if only on-site methods are used to determine the
Nitramines
presence or absence of contamination (i.e., all posi-
RDX
27
13,900
HMX
12
5700
tive samples are sent off-site for laboratory analy-
sis), analytical costs can be reduced considerably.
TNT and/or RDX
72
Because on-site methods provide near-real-time
feedback, the results of screening can be used to
Derived from Walsh et al. (1993).
3
Previous Page
