by having all of the sample distribution activities
bulk soil samples for this field exercise. With the
inside a trailer, while the technology developer was set
exception of the background sample, groundwater was
up outside under a canopy. Samples were delivered to
obtained using a sipper push point attached to the end
the technology developer in a vessel labeled with only
of a Geoprobe push rod. The sipper point consists of a
a sample number, and, in the case of soil, its weight. At
4-in. (10.1-cm) section screened with stainless steel wire
the same time that samples were collected and prepared
mesh, to which a new piece of polyethylene tubing was
for on-site analysis, 90 co-located samples or sample
attached for each sampling point (depth of 10 to 19 ft
splits were taken for off-site analysis by the reference
[3 to 6 m]). The groundwater was pumped from the
laboratory. Two additional samples were taken as trip
depth of interest using a peristaltic pump set at a rate of
blanks, which consisted of the MeOH extraction sol-
approximately 500 mL/min, when possible. The back-
vent that was used to prepare all of the GRO soil samples
ground groundwater sample was collected via the peri-
sent off-site. In addition, the State of New Mexico
staltic pump system from a permanent monitoring well.
Environmental Laboratory, which served as the QA
The following sections describe the subsampling and
laboratory for this technology demonstration, received
handling protocols used by the sample distribution team
23 samples, 17 of soil and 6 of groundwater. This QA
for both types of environmental matrices, and the prepa-
laboratory received field, PE, and matrix blank samples
ration of the matrix spikes and PE samples (Tables 2
contaminated with GRO compounds, and field samples
and 3). With the exception of two sets of PE samples,
contaminated with the DRO/bunker C and RRO com-
all of the other QA samples were fortified (spiked) on-
pounds. In addition to the samples that were sent to
site using reference standards purchased in sealed glass
these two laboratories, an entire set of soil and ground-
ampoules containing 1-mL quantities. Once these
water samples contaminated with GRO compounds was
ampoules were opened, aliquots were transferred, in
sent to the CRREL for analysis. The sample identity
every case but one, with glass microliter syringes
was known by both the QA and CRREL laboratories
(Hamilton). With exception of the water taken from the
prior to analysis. All of the samples sent off-site were
background monitoring well, all of the containers used
refrigerated during storage and transportation.
for sample collection and distribution were clean glass
The technology demonstration plan (U.S. Navy
bottles with Teflon-lined septum caps. Samples of the
1999) gives a detailed description of the techniques used
background water were initially held in plastic 4-L jugs.
to collect bulk samples of soil and water, along with
GRO compounds in soil
the sampling locations, and the historical background
information concerning the use of petroleum products
The high vapor pressures (i.e., low boiling points)
on this site. For both the GRO and DRO/bunker C
of many of the hydrocarbons in gasoline make the
plumes, a Geoprobe (Geoprobe Systems, Inc., Sulina,
matrices contaminated with this product, particularly
Kansas) sampler was used to obtain the soil and ground-
soils, susceptible to volatilization losses (Hewitt et al.
water samples. Bulk soil samples were obtained using
1995). In addition, several GRO compounds are sus-
a closed-piston sampler, with a plastic core barrel liner,
ceptible to biological degradation if not properly pre-
that had a 1.75-in. (4.45-cm) diameter and a 3-ft (0.9-
served between collection and analysis (Hewitt 1997).
m) length. A Geoprobe sampler was also used to obtain
Because of these concerns, soil samples taken from the
a sample of background material (uncontaminated soil
locations contaminated by GRO compounds were
matrix). For the locations where RRO contamination
handled with a different procedure than the samples of
exists, bulk soil samples were obtained using a drop
the less volatile and less biological labile DRO/bunker
hammer sampler after hand auguring to the depth of
C and RRO compounds.
interest. The drop hammer sampler had a core barrel
Subsamples of GRO-contaminated soils were placed
liner, consisting of a brass sleeve, with a 2-in. (5.1-cm)
directly into VOA vials that contained either a binary
diameter and 6-in. (15.2-cm) length. After each end of
solvent mixture of carbon tetrachloride and n-heptane
these sampling vessels (plastic tubes and brass sleeves)
or methanol (MeOH). Special precautions to limit
was retrieved, their ends were sealed with plastic caps
exposure were taken that were consistent with guid-
(the tops of the plastic tubes were trimmed leaving no
ance given in Method 5035 and D 4547-98 (EPA 1986,
headspace), they were labeled, and then they were
ASTM 1998). The VOA vials containing the binary
promptly delivered to the trailer for processing.
solvent were prepared by the technology developer,
Groundwater samples were collected from within
while those containing MeOH were prepared at
the GRO and DRO/bunker C contaminant plumes, and
CRREL. In addition, all of the VOA vials containing
uncontaminated groundwater samples were taken from
MeOH had been spiked with two surrogate compounds,
a background location. All of the groundwater samples
p-Bromofluorobenzene and trifluorotoluene, each at a
concentration of 2 g/mL.
were co-located with the borehole used to obtain the
4
Previous Page
