was slowly decanted. The volume of the decanted
Extraction
Archived field-contaminated soils were chosen
solvent was measured (3040 mL), then the sol-
based on previous HPLC analysis that indicated
vent was mixed with 500 mL of reagent-grade
the presence of several of the Method 8330 analytes
water. Previous work has shown that the analytes
over a wide concentration range (Table A1). Fol-
of interest will be retained by the solid phase if
lowing the soil extraction procedure specified by
the aqueous concentration of acetonitrile is less
Method 8330, 2-g soil subsamples were extracted
than 10%.* For each sample, an Empore styrene-
with 10 mL of acetonitrile (with no internal stan-
divinyl benzene reverse phase sulfonated resin
dard) for 18 hours in a cooled sonic bath. To com-
(SDB-RPS) disk was placed on a vacuum filter ap-
pare concentration estimates obtained by GC to
paratus and preconditioned according to
those obtained by HPLC, the extracts were split.
manufacturer's directions: rinsing with 10 mL each
For GC analysis, a portion of each acetonitrile ex-
of acetone, isopropanol, and methanol. For the fi-
tract was filtered through a Millex SR filter unit
nal organic solvent rinse, a 20-mL aliquot of ac-
(Millipore). For HPLC analysis, an aliquot of each
etonitrile was used. The acetonitrile was pulled
acetonitrile extract was mixed with an equal vol-
through each disk, followed by reagent-grade
ume of aqueous calcium chloride prior to filtra-
water, and then the 500-mL AcN/water sample.
tion through a Millex SR filter unit.
The disk was rinsed with a 20-mL aliquot of dis-
Additional archived soils that had trace
tilled water to remove salts. Air was then pulled
analytes, based on previous HPLC analysis, were
through each membrane for 20 minutes to remove
extracted using a higher soil-to-solvent ratio (25 g
excess water. Finally, the disks were eluted using
soil:50 mL acetonitrile). For each soil, duplicate 25-
5 mL of acetonitrile.
g subsamples were extracted with 50 mL of aceto-
nitrile in a cooled sonic bath for 18 hours. If enough
Salting-in preconcentration
soil was available, matrix spikes (MS) and matrix
A limited number of soil extracts were
spike duplicates (MSD) were also prepared and
preconcentrated by mixing 40 mL of acetonitrile
extracted. Soils were spiked with 5 mL of a spike
soil extract with aqueous sodium chloride (65 g
solution (50 g/L nitroaromatics and 200 g/L
NaCl:200 mL reagent-grade water) (Jenkins and
RDX) and left to air-dry for 24 hours in a fume
Miyares 1991). After vigorous mixing and phase
hood. The spike solution contained the analytes
separation, approximately 3.5 mL of acetonitrile
of interest for mine detection (Table 1). The spiked
extract was recovered, yielding a concentration
concentration in soil was 10 g/kg for
factor of 11.
nitroaromatics and 40 g/kg for RDX. All samples
were extracted with 50 mL of acetonitrile contain-
Method detection limits and spike recovery
ing 3,4-DNT (25 g/L) as an internal standard.
Method detection limits (MDL) and spike re-
These samples were also extracted for 18 hours in
coveries were determined using the two
a cooled sonic bath. Prior to GC analysis, extracts
soil:solvent ratios used for the extraction of field
were filtered through Millex SR filter units. Prior
samples. Following the Method 8330 protocol,
to HPLC analysis, 0.5 mL of each filtered acetoni-
seven replicate 2-g soil samples (either Ottawa
trile extract was mixed with 2 mL of reagent-grade
sand or AEC standard soil) were spiked with 1
mL of 10- or 100-g/L spike solutions to yield 5-
water (MilliQ).
and 50-g/kg samples containing the analytes of
Soils collected from an experimental minefield
at Sandia were extracted without air-drying
interest for hazardous waste site characterization
using 2 g soil:10 mL acetonitrile and 25 g soil:50
(Table 1). After one hour, 9 mL of acetonitrile was
mL acetonitrile. Soils from Fort Leonard Wood
added and the samples extracted for 18 hours in a
were extracted without air-drying using 2 g soil:5
cooled sonic bath. At two hours into the sonica-
mL acetonitrile or 20 g soil:50 mL acetonitrile.
tion period, a small aliquot of the extract was taken
for analysis to determine the stability of NG and
PETN in a cooled sonic bath.
Preconcentration
A second set of Ottawa sand samples was
Solid-phase extraction
A limited number of acetonitrile extracts were
preconcentrated using solid-phase extraction.
*Personal communication, Philip G. Thorne, Geologi-
From some of the soils extracted with 50 mL of
cal Sciences Division, CRREL, Hanover, New Hamp-
acetonitrile, the solvent remaining above the soil
shire, 1998.
3
Previous Page
