in flocculation, and the diluted extract is filtered
indicated that tetryl, another secondary explosive
through a 0.45-m membrane.
often determined in conjunction with TNT and
4. A 100-L portion is injected into an HPLC
RDX, is unstable at the temperatures required for
Soxhlet extraction (Jenkins and Walsh 1994). That,
equipped with a primary analytical column (LC-
combined with the ability to extract many samples
18) and is eluted with methanol/water (1:1) at 1.5
simultaneously using the sonic bath approach,
mL/min; retention times for the 14 target analytes
makes ultrasonication the preferred technique.
range from 2.4 to 14.2 minutes.
Results of extraction studies indicate that even
5. If target analytes are detected, their presence
when acetonitrile is used with ultrasonic extrac-
is confirmed on a confirmation column (LC-CN).
tion, the extraction is kinetically slow for weath-
6. The estimated quantitation limits in soil for
ered field-contaminated soils (Jenkins and Grant
most analytes is about 0.25 mg/kg, with RDX and
1987, Jenkins et al. 1989). For that reason, SW-846
HMX being somewhat higher at 1.0 and 2.2, re-
Method 8330 (EPA 1995) requires acetonitrile ex-
spectively. No limits are provided for the Am-
traction in an ultrasonic bath for 18 hours.
DNTs.
This procedure was subjected to a ruggedness
test (Jenkins et al. 1989), and a full-scale collabo-
RP-HPLC determination (Method 8330)
Generally, detection of the analyte within the
rative test (Bauer et al. 1990) was conducted un-
proper retention-time window on two columns
der the auspices of the Association of Official
with different retention orders is required for con-
Analytical Chemists (AOAC). In addition to ac-
firmation of the presence of these explosives.
ceptance by the EPA Office of Solid Waste as
Method 8330 specifies primary analysis on an LC-
SW-846 Method 8330 (EPA 1995), this procedure
18 (octadecylsilane) column with confirmation on
also has been adopted as Standard Method 991.09
a cyanopropylsilane (LC-CN) column (Jenkins et
by the AOAC (AOAC 1990) and as ASTM Method
al. 1989).
D5143-90 (ASTM 1990). In addition, the procedure
Walsh et al. (1973) were the first to report on
has been used successfully by a large number of
the use of RP-HPLC for the analysis of nitro-
commercial laboratories for several years.
aromatics in munitions waste. Most subsequent
HPLC methods for these compounds rely on ultra-
violet detection because of its sensitivity and rug-
SUMMARY
gedness. Initially, determination was specified at
254 nm because of the availability of fixed wave-
A large number of defense-related sites are con-
length detectors based on the mercury vapor
taminated with elevated levels of secondary ex-
lamps and a significant absorbance of all target
plosives. Levels of contamination range from
analytes at this wavelength. Current instruments
barely detectable to levels over 10% that need spe-
are generally equipped with either variable wave-
cial handling because of the detonation potential.
Characterization of explosives-contaminated sites
wavelengths of maximum absorption can be se-
is particularly difficult because of the very hetero-
lected to optimize detection. However, 254 nm is
geneous distribution of contamination in the en-
still often used because of the low incidence of
vironment and within samples. To improve site
interference at this wavelength.
characterization, several options exist: collecting
more samples, providing on-site analytical data
to help direct the investigation, sample
Method specifications
compositing, improving homogenization of
and validation (Method 8330)
Based on the research described above, SW-846
samples, and extracting larger samples. On-site
Method 8330 (EPA 1995) specifies the following:
analytical methods are essential to more economi-
1. Soil samples are air-dried and ground in a
cal and improved characterization. What they lack
mortar and pestle for homogenization.
in precision and accuracy when used to simulta-
2. A 2-g subsample is placed in an amber vial,
neously identify specific multiple compounds, the
10 mL of acetonitrile is added, and the vial is
on-site methods more than make up for in the in-
placed in a temperature-controlled ultrasonic bath
creased number of samples that can be analyzed.
for 18 hours.
While verification using a standard analytical
3. The vial is removed from the bath and the
method such as EPA Method 8330 should be part
soil is allowed to settle, a 5-mL aliquot is removed
of any quality assurance program, reducing the
and diluted with 5 mL of aqueous CaCl2 to assist
number of samples analyzed by more expensive
25
Previous Page
