Table 2. Method De-
Table 1. Retention times (min)
Table 3. Mass of TNT and RDX residues recov-
tection Limits (g/L)
for target analytes on analyti-
ered from the extraction of the filtered water
of nitroaromatics
cal and confirmation columns.
versus that from the soot.
and nitramines in
Mass recovered (g)
Retention time (min)
water.*
TNT
RDX
RTX-5 RTX-225
MDL (g/L)
(1.5 m) (0.1 m)
Soot†
Soot†
Analyte
Sample
Water*
Water*
Analyte
S-1E
0.22
0.00
1.03
0.00
NB
0.925
0.337
1,3-DNB
0.004
S-2E
2.00
0.62
0.31
0.00
o-NT
1.409
0.416
2,6-DNT
0.003
m-NT
1.740
0.500
S-3E
16.28
1.15
4.08
0.00
2,4-DNT
0.009
p-NT
1.920
0.564
S-4E
7.65
1.20
2.30
0.00
TNB
0.007
NG
3.761
4.432
2,4,6-TNT
0.010
S-5E
28.32
29.73
3.59
0.00
1,3-DNB
4.746
3.792
RDX
0.004
S-6E
1.70
4.54
0.47
0.00
2,6-DNT
4.909
3.460
4-ADNT
0.003
S-7E
0.47
0.29
0.62
0.00
2,4-DNT
5.736
4.470
2-ADNT
0.003
S-8C
0.88
0.00
27.31
0.00
TNB
7.542
7.987
Tetryl
0.009
S-1W
0.94
0.23
1.62
0.00
2,4,6-TNT
7.690
7.540
S-2W
0.08
0.00
2.31
0.00
* Determined by GC-ECD
RDX
9.335
11.595
S-3W
0.07
0.00
4.29
0.00
using 0.5 L of sample pre-
4-ADNT
9.956
10.582
S-4W
0.95
0.52
1.16
0.00
concentrated to 5 mL of
2-ADNT
10.418
11.058
S-1S
13.95
6.58
0.48
0.00
acetonitrile (Walsh and
Tetryl
11.322
11.595
S-2S
47.74
411
3.07
0.00
Ranney 1998).
HMX
15.629
Crater
0.09
0.79
0.00
0.00
RTX-5: Oven program: 100C for 2
* Mass recovered from the 1520 mL of filtered water
min, to 250C at 10/min and hold 3
multiplied by the ratio of the total snowmelt volume
min; injector 250C; detector 280C.
divided by the volume extracted.
RTX-225: Oven program: 100C for
† Mass recovered from extraction of soot from the total
2 min, to 210C at 10/min and hold
sample of snowmelt.
5 min; injector 200C; detector
210C.
RESULTS AND DISCUSSION
Analytical methodology
therefore may underestimate the amount of explosive
sorbed to the soot. Consequently, the continuous-extrac-
Analysis of the snow samples collected after a deton-
tion Soxhlet procedure was used. This procedure, while
ation resulting from a TNT-containing munition is com-
less convenient than the ultrasonic equilibration proce-
plicated by the presence of a large amount of soot. When
dure, has been shown to be equally effective for extrac-
this snow is melted, the fine soot particles remain in
tion of explosives from soil (Jenkins and Grant 1987),
suspension, coloring the water black. The soot must be
and it does not rely on a favorable partition coefficient.
filtered out prior to extraction of the water using the
This is true because the analyte, once extracted, is removed
standard analytical protocol for analysis of explosives
from interaction with the soot each time the Soxhlet
in water, SW-846 Method 8330 (EPA 1994). The sorp-
device cycles.
tivity of the soot for the nitroaromatic and nitramine
We decided to use the Soxhlet procedure, rather than
organic chemicals that make up the main charge and
investigate the partition coefficient for the various ana-
booster in the PMA-2 land mine is unknown, hence the
lytes between soot and ACN. The mass of analytes re-
soot that was filtered from the water must also be anal-
covered from extraction of the water and from extrac-
yzed to account for all the explosives residues that could
tion of the soot is shown in Table 3. The ratio of mass
be present in the surface snow samples.
of TNT associated with soot versus that associated in
Extraction of the soot to remove the sorbed explo-
the water ranged from 0 to 8.8. There does not appear
sives could be conducted with an equilibrium-based
to be any relationship between this ratio and the total
method such as the soil extraction procedure in SW-
amount of TNT present, probably because a portion of
846 Method 8330 (EPA 1994). In this method 2 g of soil
the TNT present was particulate and only a portion of
is equilibrated with 10 mL of ACN overnight in an ultra-
the total had dissolved. Thus it was important to anal-
sonic bath. This procedure works well for soil because
yze both the water and soot to estimate the total resi-
the equilibrium partition coefficient is favorable for
dues resulting from the detonation of the PMA-2 mine.
extraction of the explosive from soil into ACN. The
For RDX, though, it appears that there is less associa-
equilibrium partition coefficient between ACN and soot
tion with soot; RDX was found only in the water anal-
is unknown and may not be favorable for equilibrium
ysis.
extraction. An equilibrium-based extraction procedure
4
Previous Page
