Table 2. Characteristics and performance of test kits.
Initial
Range
Kit
Cost/day
% false
Accuracy
(g/L)
Kit
cost ($)
Format
cost ($)
(10 samples)
neg/pos %<RPD = 50
Quantix
6300
96-well
0.0520
840
210
0/0
100
(4000 for govt.)
strips
EnviroGard
2130
96-well
0.550
387
97
0/6
86
strips
Ohmicron
4435
Test
0.075.0
210
168
0/0
85
tubes
DTECH TNT
0
Cups
5-45
100
250
0/30
58
DTECH RDX
0
Cups
5-45
100
250
24/18
32
well had reached a sufficient intensity to match
against the standards by programs within the
the reference color on the card. The time required
Quantix-supplied reader.
for development depends on temperature and
The RaPID kit (Ohmicron, Newtown, Pa.) is a
was predicted to be around 10 minutes. Alterna-
quantitative laboratory assay that can be utilized
tively, a differential reflectometer supplied by
in the field with a battery-powered spectropho-
DTECH could be used to quantify the inhibition
tometer. Antibodies were immobilized on plastic
due to TNT or RDX in the sample compared with
beads containing a ferrous metal particle. Dupli-
the reference. The resulting number is then con-
cate samples or standards, TNT conjugate, anti-
verted to a concentration range based on a calibra-
body beads, and diluent were incubated in 12-
mm 75-mm plastic test tubes for 15 minutes. The
tion table supplied with the kit.
tubes were then placed in a rack that contained
strong magnets. The particles were drawn to the
RESULTS AND DISCUSSION
sides of the tube where they remained when the
Results from the RP-HPLC analyses showed
liquid contents of the tube were poured off and
that 19 of the 33 Crane wells were contaminated
the particles rinsed. Substrate and chromagen was
with nitramine and nitroaromatic explosives and
then added and the color developed for 20 min-
their environmental transformation products
utes. Absorbances were measured and concentra-
(Tables 3 and 4). The samples that were stored at
tions calculated against the standards by the
CRREL for one month were only analyzed for TNT
Ohmicron-supplied spectrophotometer.
and its transformation products by direct injec-
The DTECH TNT and RDX kits (EM Science,
tion. Some of the samples had concentrations that
Gibbstown, N.J.) are semiquantitative field tests
dropped below the detection limit of this method.
that require no electronic equipment. Antibodies
Other samples showed significant transformation
were immobilized on plastic beads contained in
of TNT. All of the Umatilla and Bangor wells had
small vials. For each test, a sample diluted in assay
detectable levels of nitramines and nitroaromatics.
buffer containing TNT conjugate was added to
There are two ways of evaluating the TNT
one vial and buffer containing only the TNT conju-
results from each kit. One way assesses the ability
gate was added to a second vial. These incubated
of the kits to determine correctly if there is con-
for 2 minutes. Then the vials were swirled to sus-
tamination above the EPA's health advisory limit
pend the particles and the contents were poured
of 2 g/L (EPA 1989). The DTECH kit has a detec-
into side-by-side wells in the top of a cup. The bot-
tion limit above that and could not be assessed by
tom of each well was constructed of porous mate-
this criterion. All of the other TNT kits were
rial that allowed the liquid contents of the vial to
successful in indicating the presence of TNT when
drain into absorbent material in the cup while re-
it was there at greater than 2 g/L. There were no
taining the antibody-coated beads. The beads
false negatives. They could be used in remediation
were rinsed in place and substrate was added
projects to indicate when contamination levels
along with a chromogen that produced a blue pre-
dropped below the detection limit of the kit.
cipitate upon activation by the conjugated en-
Another way to assess kit performance is to
zyme. Concentration ranges were determined by
measure accuracy using the relative percent differ-
comparing the color of the sample well to the col-
ence (RPD), where
or on a test card after the color of the reference
4
Previous Page
