Table 6. Comparison of on-site analytical methods for TNT, RDX, and HMX to EPA Method 8330.
Regression
Regression
Correlation Mean RPD
Number
Method
intercept
slope
coefficient (r) (absol. value)
samples
Reference
MDL < TNT ≤ 100 mg/kg
CRREL
10
0.84
0.74*
72
86
EPA 1997
EnSys RISc
19
0.81
0.45*
90
123
EPA 1997
DTECH
2.9
0.79
0.76*
63
103
EPA 1997
Idetek Quantix
13
0.62
0.46*
84
124
EPA 1997
Ohmicron RaPID Assay
16
1.2
0.51*
97
115
EPA 1997
DTECH†
-17
6.7
0.81*
110
37
Haas and Simmons 1995
one outlier deleted†
3.7
2.4
0.91*
36
EnviroGard plate†
13
1.3
0.79*
122
36
Haas and Simmons 1995
EnviroGard tube†
6.3
0.99
0.90*
95
21
Haas and Simmons 1995
Idetek Quantix†
36
2.1
0.39**
131
37
Haas and Simmons 1995
Ohmicron RaPID Assay†
18
1.8
0.83*
127
37
Haas and Simmons 1995
EnSys RISc†
3.8
0.72
0.91*
56
12
Myers et al. 1994
DTECH†
5.4
0.94
0.30
88
10/11
Myers et al. 1994
100 < TNT < 1000 mg/kg
-25
CRREL
1.4
0.67*
33
15
EPA 1997
EnSys RISc
50
1.1
0.59*
57
21
EPA 1997
-250
DTECH
2.2
0.59**
60
17
EPA 1997
Idetek Quantix
210
0.09
0.30
65
22
EPA 1997
Ohmicron RaPID Assay
680
0.50
0.12
51
16
EPA 1997
TNT > 1000 mg/kg
0
EnSys RISc
0.995
0.76
23
25
EPA 1997
MDL < RDX ≤ 100 mg/kg
-1.2
CRREL
0.56
0.89*
74
64
EPA 1997
EnSys RISc
6.4
0.57
0.50*
61
114
EPA 1997
DTECH
2.7
0.20
0.49*
103
94
EPA 1997
DTECH†
-0.35
0.77
0.95*
66
27
Haas and Simmons 1995
100 < RDX < 1000 mg/kg
-9.9
EnSys RISc
0.68
0.50*
83
32
EPA 1997
DTECH
21
0.15
0.49**
127
25
EPA 1997
RDX > 1000 mg/kg
0
EnSys RISc
0.38
0.64
75
19
EPA 1997
MDL < HMX < 2200 mg/kg
CRREL
0
0.988
0.971*
--
76
Jenkins et al. 1997
EnSys RISc
0
0.988
0.971*
--
76
Jenkins et al. 1997
*Statistically significant at the 99% probability level.
†Statistics calculated from cited reference.
**Statistically significant at the 95% probability level.
HPLC was poorer. Both methods were susceptible
requiring good quantitative agreement with the
to interferences. Although both methods showed
standard laboratory method and that the DTECH
strong tendencies to cross-react with other
kit was better suited for quick, on-site screening
nitroaromatics, sometimes resulting in false posi-
in situations where all samples above a certain
tives, neither method produced a false negative
range will be sent forward to a laboratory for con-
in a sampling of 99 soils. The study concluded that
firmation by the standard method.
the EnSys RISc kit was well suited for analyses
The ENVIROL test kit for TNT has only recently
22
Previous Page
