bration range. Because of the limited linear range
to yield more accurate concentration estimates
than an average calibration factor.
Using estimates based on quadratic calibration
models, recovery was 90% or greater for each of
the nitroaromatics and nitrate esters. Recoveries
were lower for the nitramines and amino-nitrotol-
uenes, but well within the acceptable range of SW-
846 methods. Based on the standard deviations of
the means for seven replicates, MDLs were com-
puted (Table 14, Table A2). These MDLs are all be-
low water quality criteria except for 2,6-DNT 106
for increased cancer risk.
We obtained a lower MDL for 2,6-DNT (0.0025
g/L) by extracting a greater volume of water (1
L), eluting with less acetonitrile (4.0 mL), spiking
at lower concentrations (Grant et al. 1991), and
using 10 replicates. We used both cartridge and
disk SPE (Table 15, Table A2), and the MDLs were
generally quite similar for each analyte. If the anal-
yte of most interest is 2,6-DNT, the MDL could be
lowered even more by preconcentrating a greater
volume of water. We limited the volume we pre-
concentrated to prevent breakthrough of HMX
and RDX. 2,6-DNT is well retained on both solid
Table 14. Recovery data and MDLs for various
analytes for 500-mL water samples preconcentrated
to 5.0 mL acetonitrile using Empore SDB-RPS
disks.
Mean*
found
Target
Mean
conc.
conc.
recovery
RSD
MDL
(g/L)
(g/L)
(g/L)
Analyte
(%)
(%)
DNB
0.197
0.20
99
9.7
0.06
2,6-DNT
0.187
0.20
93
7.1
0.04
2,4-DNT
0.208
0.20
104
7.6
0.05
TNB
0.189
0.20
94
7.7
0.05
TNT
0.233
0.20
116
8.0
0.06
RDX
0.176
0.20
88
7.2
0.04
4-Am-2,6-DNT
0.150
0.20
75
11.2
0.05
2-Am-4,6-DNT
0.174
0.20
87
11.6
0.06
Tetryl
0.190
0.20
95
8.3
0.05
DNA
0.148
0.20
74
9.2
0.04
NB
0.969
1.0
97
7.1
0.2
o-NT
0.927
1.0
93
5.1
0.2
m-NT
0.918
1.0
92
4.6
0.1
p-NT
0.897
1.0
90
5.3
0.2
NG
0.918
1.0
92
5.5
0.2
PETN
0.992
1.0
99
4.8
0.2
HMX
1.58
2.0
79
8.1
0.4
* Mean of seven replicate 500-mL water samples preconcen-
trated to 5 mL AcN using Empore SDB-RPS disks. Complete
data set is in Table A1.
16
Previous Page
