An activity of 0.05 was selected because, accord-
horizontally on an orbital shaking table for ap-
ing to Vonk (1985, 1986), diffusion of an organic
proximately 48 hours. After mixing, the vials were
solute (that is a solvent or good swelling agent of
opened and previously weighed pieces of PVC
PVC) with an activity that is less than 0.1 should
were added to the test solutions. When necessary,
be Fickian and thus concentration independent
a few drops of water were added to ensure that
and slowest. At these activities the rate of diffu-
there was no headspace, and the vials were re-
sion is considered so slow that diffusion of an
capped.
organic solute through PVC pipe would take thou-
sands of years (Berens 1985, Vonk 1985, 1986, Olson
Long-term
et al. 1987). Therefore, we would not anticipate
miscible solvent study
In this study, pieces of PVC were exposed to a
any softening to occur at an activity of 0.05 for
range of concentrations of the same solvents as
any (single) organic solute. (So far in our own
the previous study, with the exception of n-
tests we also have not observed any softening at
butylamine. The concentrations of acetone were
an activity of 0.05.) A combination of 18 solvents
40, 20, 10 and 5% (w/v). The concentrations of
and swelling agents of PVC was used in this study.
dimethylformamide and pyridine were 80, 60, 40
and 20%. The concentrations of tetrahydrofuran
(the most aggressive solvent in the previous study)
MATERIALS AND METHODS
were 10, 1.0, 0.1 and 0.01%. The contact times were
2, 4, 8, 12, 16 and 20 weeks. For each sample, time
Small pieces of PVC were cut from 2-in- (5.1-
and concentration, there were four replicate
cm-) diameter schedule-40 well casing. The overall
dimensions of the pieces were approximately 9
samples, except for acetone, where there were five
9 0.8 mm (L W H), giving a total surface
replicates. The test solutions were made up as
area of approximately 1.9 cm2 and an approxi-
described for the short-term miscible solvent study,
except for the acetone solutions.
mate weight of 0.09 g. Special care was taken to
The acetone solutions were made up by add-
eliminate contamination from grease or oil dur-
ing acetone to deionized water in 1-L volumetric
ing the cutting procedure. The pieces of PVC were
flasks, each containing a glass-covered magnetic
washed by placing them in a glass beaker con-
stirring bar. The flasks were then filled with deion-
taining a 2% solution of Liquinox detergent and
ized water, leaving approximately 1 mL of
deionized (Millipore) water and stirring them with
headspace so that the test solution would not be
a glass rod for approximately five minutes. The
lost when the glass stopper was inserted. The stop-
pieces were then rinsed with copious amounts of
pers were then wrapped tightly with parafilm to
deionized water until there were no signs of suds-
reduce vapor loss, and the solutions were mixed
ing, then rinsed with several additional volumes
for approximately two days on magnetic stirring
of deionized water to assure complete removal
plates. When the solutions were mixed, preweigh-
of the detergent. The pieces were then spread on
ed PVC pieces were added to vials, and the vials
paper towels, blotted and left to air dry.
were filled to capacity with solution to eliminate
any headspace and capped.
Short-term
miscible solvent study
In this study, pieces of PVC were exposed to
Mixed-solvent study
In this study, pieces of PVC were exposed to
four concentrations (20, 40, 60 and 80% w/v) of
an aqueous solution of 18 organic chemicals shown
five organic solvents that are totally miscible in
to be solvents or good swelling agents of PVC
water. The five solvents used were: acetone, n-
(Table 3). The activity of each analyte was 0.05,
butylamine, dimethylformamide, pyridine and tet-
giving a total activity of 0.9. Samples were taken
rahydrofuran. The contact times were three and
after 1, 3, 7, 14, 21 and 42 days. Five replicate
seven days. For each time, solvent and concen-
samples were taken on each sampling day. The
tration, there were five replicate samples. The test
mixed-solvent solution was made by adding a
solutions were made up in 22-mL borosilicate glass
volume of each analyte using a microliter syringe
vials with Teflon-lined plastic caps. The vials were
to deionized water in a 1-L volumetric flask con-
filled almost to the top with Milli-Q water, sol-
taining a glass-covered magnetic stirring bar. The
vent was added using a microliter syringe, and
flask was then filled with deionized water, leav-
the vials were then filled to capacity with water
ing approximately 1 mL of headspace. The stop-
to eliminate any headspace, capped and placed
5
Previous Page
