the EnSys reagent to these yellowish solutions
Colorimetric on-site
resulted in variable and unstable color changes.
analysis for TNT
Calibration for quantitation was achieved by
and 2,4-DNT
The 20-g soil samples at all three installations
reacting a known standard of TNT in acetone
were extracted in 150-mL plastic extraction bottles
(containing 3% water) with the EnSys reagent for
by adding 100 mL of acetone, and shaking vigor-
samples from locations 1, 3, 4, 5, 7, 7R, 8 and 9.
ously (Jenkins and Walsh 1992). Soil extracts from
Absorbance was measured at 540 nm with a bat-
all locations were analyzed using the EnSys TNT
tery operated spectrophotometer (Hach Model
method (EPA 1995b). The acetone contained 3%
DR/2000). Likewise, a standard with a known
water to ensure that adequate water was present
concentration of DNT was used to calibrate soil
for the chemical reaction that produces color de-
extracts from sampling location 2 and absorbance
velopment. An extraction rate study was con-
was measured at 570 nm. Correction for back-
ducted on the soil from each site to determine the
ground color in the extracts was obtained by mea-
appropriate extraction time. For soils from the
suring the absorbance of each extract prior to
Monite site and Hawthorne AAP, a 3-minute ex-
addition of the EnSys reagent, doubling the value,
traction time was adequate. For samples from
and subtracting it from the final absorbance after
Volunteer AAP, the 3-minute extraction time was
addition of the reagent. Doubling the initial ab-
not adequate, so soils were extracted using 3 min-
sorbance prior to subtraction takes into account
utes of shaking, a 30-minute rest time, and a final
the increased absorbance caused by reaction of
3-minute shaking period. After allowing the soil
humic organics in the extract with base, as dis-
to settle for at least 15 minutes, we removed an
cussed elsewhere (Jenkins and Walsh 1992).
aliquot of each extract using a Plastipak syringe
and filtered it through a Millex SR membrane.
On-site analysis method
Extracts were diluted as appropriate, such that
for ammonium picrate
absorbances after reaction with the EnSys reagent
The on-site analysis method used for ammo-
were less than 1.0.
nium picrate was reported by Thorne and Jenkins
For samples containing mainly TNT, the inten-
(1995). We extracted 20-g subsamples of soil from
sity of color of the extract prior to reaction with
sampling location 6 with 100 mL of acetone con-
the EnSys reagent often indicated the TNT con-
taining 3% (V/V) deionized water by manually
centration and served as a rough guide for sample
shaking for 3 minutes. A 4-mL aliquot was re-
dilution. For extracts containing DNT, this was
moved and the absorbance measured at 400 nm.
not true and the degree of dilution needed for
If the absorbance was above 1.0, the extract was
each sample was obtained by on-site experimen-
diluted with deionized water until the absorbance
tation. Because soil concentration varied by such
was below 1.0. This dilution factor was used to
a large amount, with concentrations in excess of
calculate how much of the original acetone ex-
100,000 g/g, acetone extracts had to be diluted
tract could be applied to a 3-mL SPE-ALUMINA-
by ratios as high as 1:5000 to provide analyte
A (Supelco) cartridge.
concentrations in the linear range of the method
The volumes used for analysis of the duplicate
(04 mg/L). In the field these dilutions were made
subsamples of the discrete samples and for com-
using glass L syringes and graduated cylinders.
posites were as follows: for discrete samples from
When this dilution process was assessed, relative
positions 2, 3 and 7--20 mL; from positions 1 and
standard deviations were always less than 3%
4--10 mL; from position 6--2 mL; from compos-
(Jenkins et al. 1996).
ites--1 mL; from position 5--0.4 mL. These quan-
For seven of the nine sampling locations, ex-
tities were diluted one-to-one with deionized wa-
tracts became reddish when reacted with the
ter and added to the cartridges.
EnSys reagent, meaning that TNT was likely
Picrate ions were retained on the alumina. Most
present. For sampling location 2 at the Monite
interferences were removed by passage of a 5-mL
site, extracts became blue-purple when reacted
aliquot of methanol followed by a 3-mL aliquot of
with the EnSys reagent, showing that DNT was
acetone. Picric acid was eluted from the cartridges
the likely contaminant rather than TNT. At sam-
with 10 mL of acetone, which had been acidified
pling location 6, acetone extracts were fluores-
with four drops of concentrated sulfuric acid. The
cent yellow, denoting the presence of ammonium
initial absorbance at 400 nm was recorded and
picrate as the primary contaminant. Addition of
used as a background correction. After adding an
6
Previous Page
