Table 4. Mean weight gain and relative hardness
cant change in the Barcol readings, but the 40%
(Barcol) values of PVC exposed to miscible solvent
(6.8 M) and 60% (10.2 M) solutions did. The PVC
solutions after three and seven days of exposure.
pieces exposed to the 40% solutions were notice-
ably softened when tested manually, and those
3 days
7 days
Solution (%)
% Wt. gain Barcol
% Wt. gain Barcol
exposed to the 60% solutions were pliable (i.e.,
could be bent in half easily). In contrast, the highest
Control
concentrations tested (60%) of n-butylamine (8.2
0.0
0.07
79.0
0.07
79.0
M), pyridine (7.5 M) and dimethylformamide (8.2
Acetone
M) had no effect on the hardness readings. We do
not have any χ values for these organic solvents.
20
0.07
79.7
0.22
78.3
40
4.29
64.7*
7.76
50.0*
As a result of this study we know that aque-
60
20.39
39.0*
20.14
37.0*
ous solutions of at least two organic solvents that
n-butylamine
are either solvents or good swelling agents of PVC
and are totally miscible in water can soften PVC.
20
0.14
78.3
0.07
79.0
40
0.22
79.3
0.03
78.7
Because this study only ran for seven days, it is
60
0.21
78.0
0.22
78.7
possible that exposure to solutions of n-butyla-
mine, pyridine or dimethylformamide may cause
Dimethylformamide
softening if given more time.
20
0.08
79.3
0.00
79.0
40
0.10
79.3
0.04
78.0
Long-term
60
0.08
79.0
0.04
79.0
miscible solvent study
Pyridine
Since the performance of n-butylamine was be-
20
0.13
79.0
0.16
78.0
tween that of dimethylformamide and pyridine
40
0.20
79.0
0.11
78.7
in the previous study and since n-butylamine is
60
0.24
78.0
0.30
78.3
a dangerous and highly toxic chemical to work
Tetrahydrofuran
with, we decided that further study with this
S†
chemical was neither warranted nor desirable.
20
29.68
35.7*
29.66
40
86.53
86.66
S
S
Therefore, in this study we exposed pieces of PVC
60
D
D
to aqueous solutions of tetrahydrofuran, acetone,
pyridine and dimethylformamide for a much
* Value significantly (95% level) different from controls
† D PVC dissolved and couldn't be tested
longer time than in the previous study, up to 20
S Softened PVC couldn't be tested
weeks. The range in concentrations for these test
solutions was determined based on the results
agents (tetrahydrofurn, acetone, dimethylforma-
from the previous study (Table 4). The mean
mide, pyridine and n-butylamine) for up to seven
changes in hardness readings for this study are
days. Acetone and pyridine are EPA priority pol-
given in Table 5, and the complete set of hard-
lutants and/or target analytes (Montgomery and
ness readings and change in weight values can
Welkom 1989), and the rest are common in the
be found in Appendix B. Again tetrahydrofuran
work environment according to the National In-
was the most aggressive solvent. By the end of
stitute of Occupational Safety and Health (Mont-
this study (20 weeks), even the hardness read-
gomery 1991) and thus potential groundwater con-
ings for the samples exposed to the lowest con-
taminants. The change in weight data and hard-
centration tested (0.01%, or 0.0014 M) were sig-
ness readings are summarized in Table 4, and the
nificantly lower than the controls. None of these
complete data set can be found in Appendix A.
samples were noticeably softened when tested
Tetrahydrofuran is a solvent of PVC and is com-
manually, even the samples exposed to the high-
monly used as a component in PVC glues. The
est concentration tested, a 10% solution (1.4 M).
aqueous solution of this chemical was also the
Although there is a lot of variability in the weight
most aggressive. After seven days the 20% solu-
change data, it appears that the samples exposed
tion (2.8 M) softened PVC, and the 60% solution
to the 1.0, 0.1 and 0.01% solutions may have
(8.4 M) dissolved it! Acetone has a χ value of 0.61
reached equilibrium. If this is the case, then no
and thus is a good swelling agent of PVC. The
further changes would be expected.
aqueous acetone solution was the next most ag-
For acetone there was a significant change in
gressive aqueous solution. The lowest concentra-
the Barcol readings of samples exposed to the 10%
tion tested (20% or 3.4 M) did not cause a signifi-
(1.7 M), 20% (3.4 M) and 40% (6.9 M) solutions
7
Previous Page
