the concentrations of HMX predominated in these
Table 7. HMX concentrations (mg/kg) from
soils, the colorimetric method was calibrated with
extraction time study using acetone.
a solution containing a known concentration of
HMX (mg/kg) by RP-HPLC
HMX.
extraction time
Results of these analyses are presented in Table
8. Concentrations of HMX and RDX in these ex-
Sample
3-minutes 10-minutes
30-minutes
tracts, obtained by RP-HPLC analysis, ranged
from 17 to 293 mg/kg, and from ND (not detect-
1-44-1 (015 cm)
260
262
266
1-44-1 (3045 cm)
<1
<1
<1
able) to 1.1 mg/kg, respectively, on a soil dry
1-44-1 (105120 cm)
2.7
2.6
2.7
weight basis. An inspection of the results from the
colorimetric on-site method indicates that they are
quite similar to those for HMX from RP-HPLC.
The DTECH results, on the other hand, do not
opment of this method (Jenkins and Walsh 1992).
seem to correlate with either the HMX or the RDX
The fast extraction kinetics with acetone for soils
results, giving a value that is intermediate between
from the inland ranges will make the use of on-site
the two.
methods for site characterization very convenient,
We examined HMX concentrations from RP-
and analytical turn-around times will be short.
HPLC and those from the on-site colorimetic
In a second experiment, a set of 11 soil samples
method using correlation analysis. Note that this
was selected on the basis of results from analysis
comparison is made on aliquots of the same
by SW846 Method 8330. We selected these samples
extract, i.e., the variability attributable to
to encompass a range of HMX concentrations. A
subsampling the soil is removed. When a linear
20-g portion of each soil was extracted with
model with intercept was fitted, a slope of 0.945
acetone by manually shaking periodically over 30
and intercept of 4.57 were obtained with a corre-
minutes. Aliquots of each acetone extract were
lation coefficient of 0.984 (Fig. 9). This result is
analyzed by three methods: 1) HPLC using the
quite similar to that found at CFB-Valcartier, which
confirmation separation of Method 8330, 2)
had a slope for this relationship of 1.004, with a
DTECH's on-site enzyme immunoassay method,
correlation coefficient of 0.992. These results dem-
and 3) the on-site colorimetric method. Because
onstrate that the colorimetic method could be used
our research at CFB-Valcartier demonstrated that
to estimate HMX concentrations for soils at the
the colorimetric method would respond to both
Fort Ord impact ranges with a confidence approxi-
HMX and RDX (Jenkins et al. 1997a), and because
mately equivalent to that from RP-HPLC analysis.
Table 8. Assessment of on-site methods for RDX and HMX in Fort
Ord soil samples.
Concentration (mg/kg)
HPLC
Depth
Location
(cm)
RDX
HMX
Colorimetric
D TECH
1-44-1
015
0.2
210
200
>6
1-44-2
015
0.1
253
268
>6
1-44-3
015
0.9
282
258
>6
1-44-5
015
ND*
197
176
>6
1-44-6
015
1.0
235
234
4660
1-44-6
1530
ND
18
30
>6
1-44-7
015
0.3
293
244
>6
1-44-8
015
ND
68
47
2.55.0
1-44-8
1530
ND
17
8
1.52.5
1-44-11
015
ND
162
141
4.66.0
1-44-12
015
1.1
110
87
4.66.0
*Not detectable.
18
Previous Page
