Table 5. Comparison of pairs of composited
Collection vessel experiment
samples added to VOA vials with and without
From the data in the last table (Table 5) we can
solvent present.
assess the effect of adding a soil sample to a VOA
vial containing water vs. adding soil to an empty
Time (min.) lapse before addition
vial and then adding water. These experiments
of water/temporary cap/depth
Immersed
Solvent added
used composited samples to minimize effects of
(g TCE/g)
(g TCE/g)
(cm) of samples collected
spatial variability. The results of these experiments
0.1/None/45
1.7/1.6
1.3/1.1
show that even if the water is added immediately
2/Parafilm/38
1.7/1.6
0.50/0.50
after the soil sample, measurable TCE was lost.
20/VOA septum/40
1.7/1.5
0.53/0.69
When the period between addition of soil and
200/VOA septum/43
1.9/2.1
0.60/0.59
water was increased, and temporary covers were
Note: Four samples collected at a depth of 35 cm with a 5-
used, greater than 60% loss of TCE resulted. Simi-
cm3 syringe established mean and standard deviation of
lar findings have also been reported for laboratory
2.00.4 g TCE/g.
experiments comparing treated and field-contami-
nated soils transferred to vials with and without
solvent (Jenkins and Schumacher 1987, Hewitt et
is to avoid the potential problem that could occur
al. 1995). Therefore, vials into which soils are trans-
when samples that require purge and trap gas
ferred for in-vial handling and analysis methods
chromatography mass spectrometry (PT/GC/MS)
cannot have their hermetic seals broken to intro-
analysis have concentrations of 1 ppm or greater,
duce a solvent, or for any other reason, without
as that would damage the detection system.
significant VOC loss. An exception to this state-
Samples with these high VOC concentrations will
ment is when the soil sample is immersed into a
become a common occurrence on those sites where
solvent in which the VOCs are highly soluble (i.e.,
VOCs can be easily detected (smelled) by field
methanol, etc.), then the VOA vial can be re-
personnel. To address this issue, Methods 5021 and
peatedly opened and aliquots withdrawn without
5035 recommend that two or more colocated
significant VOC losses (Hewitt 1995b).
samples be collected, thus allowing for a screen-
The results in Table 5 also show that the amount
ing analysis to be performed prior to PT/GC/MS.
of TCE lost did not appear to change when the
One logical solution would be to always collect
addition of water was varied between 2 to 200 min-
one colocated sample for equilibrium HS analysis
utes. One explanation for this is that the TCE in
(Hewitt et al. 1995). The concentrations established
the soil plugs rapidly achieved an equilibrium con-
by HS would inform the analyst how to prepare a
colocated sample, or which colocated sample
dition with the air inside the vial. Therefore, when
should be then taken for PT/GC/MS analysis.
the water was introduced, TCE was lost due to dis-
placement (10 mL of water/10 cm3 of air) and other
However, as pointed out earlier, the sample is
likely to be compromised with regard to VOC con-
air exchange processes. The rapid attainment of
equilibration for these soil plugs (3.6 cm2/g, sur-
centration, unless a nondisruptive and limited
exposure transfer is made to a VOA vial that is
face area to mass ratio) agrees with the previous
either prepared for in-vial handling and analysis
experiment, showing that 90% of the TCE was lost
or that contains an appropriate solvent.
in less than 2 minutes when similar plugs of soil
If PT/GC/MS analysis is required, VOC con-
were temporarily stored in plastic bags.
centrations exceed 1 ppm, and an organic solvent
such as methanol cannot be used on-site. At
present there is no effective way of storing an in-
DISCUSSION
tact soil sample so that it can be placed in a sol-
Once in-vial methods are used for site investi-
vent off-site, without allowing for uncontrolled
gations, samples with VOC concentration more
volatilization losses. Collection of samples in core
representative of their environment will be
liners that are then covered with either sheets of
handled and analyzed by off-site laboratories. Re-
TFE or aluminum foil fail to maintain a VOC con-
ducing volatilization losses by using in-vial meth-
centration representative of the field condition, as
ods, samples that better represent the site condi-
would VOA vials requiring their hermetic seal be
tions will often equate to higher concentrations.
broken so as to introduce a solvent after sample
Therefore, laboratories will have to be better in-
collection.
formed of the site conditions and the sampling and
However, a device is under development that
analysis plan. The reason for increased awareness
shows promise for allowing colocated samples to
8
Previous Page
