Furthermore, the amount of norbornadiene (present as an
impurity in the quadricyclane) in the soil extracts did not
change significantly, indicating that no isomerization of
quadricyclane into norbornadiene had occurred during the
The abiotic breakdown of quadricyclane in natural soils
is slow compared to the time scales associated with its
migration in the vadose zone of porous media. Consequently,
after a spill, quadricyclane can be expected to migrate as an
LNAPL. In low pH soils, the quadricyclane will then gradually
react to form compounds that are soluble in water. Most of
the migration in the second stage after a spill in these soils
will therefore take place as organic solutes in the aqueous
phase. In soils with near-neutral pH, however, the conversion
to reaction products and their subsequent movement will be
of little importance.
9
months of the experiment in any of the samples.
Overall the pH of the soil is by far the most important
factor in determining the stability of quadricyclane in the
porous media. Significant amounts of the quadricyclane
reacted to give exo-5-norbornen-2-ol and nortricyclyl alcohol
with the ALOLPL soil (Figure 4a), which has the lowest level
of organic matter of any of the soils used, while even greater
amounts of products were observed in the ALOHPL soil
Quadricyclane has a propensity to form microemulsions
that are long-lived due to its density being very close to that
of water. This could play a role in the aqueous migration of
quadricyclane, especially in areas with significant ground-
water flow. Chemical reactions play an important role in
predicting the fate of quadricyclane when exposed to a
subsurface environment. The reaction rates are strongly soil
dependent. This will highly complicate any efforts to simulate
the environmental behavior of quadricyclane.
(
Figure 4b). The reactions were much slower in soil, however,
than in aqueous solutions, and the ratio of nortricyclyl alcohol
to exo-5-norbornen-2-ol was much smaller than for the
aqueous samples, being in the region of 1.5 for the ALOLPL
soil and 0.9 for the ALOHPL soil. The slower reactions in
soils thus appear to favor the formation of exo-5-norbornen-
2
-ol over the formation of nortricyclyl alcohol. There was a
small dip in the total amount of reactants and products
recovered from the 180-day ALOLPL sample due to extraction
problems caused by gel formation in the rolled sample. After
Acknowledgments
6
months, the batch experiment with the ALOHPL soil was
The support of the United States Air Force Civil Engineering
Support Agency, Environics Division, Tyndall Air Force Base,
FL, under Contract F08635-93-C-0071 is gratefully acknowl-
edged. We also wish to thank Dr. Ben Hajek for selecting and
analyzing the four soils.
aborted because the concentration of the two alcohols became
very high. In the less acidic TNOLPH and TNOHPH soils
(
Figure 4c,d, respectively), the quadricyclane appeared stable.
After 9 months, only trace amounts of nortricyclyl alcohol
were detected.
Quadricyclane and norbornadiene can be extracted from
the soils very efficiently. For the t ) 0 samples, close to 100%
was recovered for all the soils using the techniques described.
The actual amounts of reactants and products recovered in
each case are shown in Figure 4. The extraction was not so
effective, however, for the two alcohol products. The results
from the three soils that were doped with the alcohols showed
that the amount of exo-5-norbornen-2-ol recovered ranges
from 56.6% in the ALOHPL soil to 68.8% in the TNOLPH soil.
The extraction of nortricyclyl alcohol was more efficient,
ranging from 71.4% in the ALOHPL soil to 84.7% in the
TNOLPL soil.
Literature Cited
(1) Yaws, C. L.; Yang, H. C.; Hopper, J. R.; Hansen, K. C. Chem. Eng.
1990, 177.
(2) Schwille, F. In Pollutants in Porous Media; Yaron, B., Dagan, G.,
Goldschmid, J., Eds.; Springer-Verlag: Berlin, 1984; pp 27-49.
(3) Mackay, D. M.; Roberts, P. V.; Cherry, J. A. Environ. Sci. Technol.
1985, 19, 384.
(4) Maruyama, K.; Tamiaka, H.; Kawabata, S. J. Org. Chem. 1985,
5
0, 4742.
(5) Weissman, S. A.; Baxter, S. G. Tetrahedron Lett. 1987, 28, 603.
(6) Dauben, W. G.; Cargill, R. L. Tetrahedron 1961, 15, 197.
(
(
(
7) Tabushi, I.; Yamamura, K.; Togashi, A. J. Org. Chem. 1976, 41,
169.
No other products were observed in measurable quantities,
even when the pH swing was performed, leading us to
conclude that the mass imbalance in the longer term rolled
Alabama soils is most likely due to to the less efficient
extraction of the alcohol products. The losses in the two
Tennessee soils, where the alcohols were not involved, can
be explained by losses through the caps of the vials where
quadricyclane is the most volatile component present.
Measurements on the two water-filled vials showed a weight
loss of the order of 0.6% over 9 months.
2
8) Rieber, N.; Alberts, J.; Lipsky, J. A.; Lemal, D. M. J. Am. Chem.
Soc. 1969, 91, 5668.
9) Tabushi, I.; Yamamura, K.; Yoshida, Z. J. Am. Chem. Soc. 1972,
9
4, 787.
(
(
(
10) Smith, C. D. J. Am. Chem. Soc. 1966, 88, 4273.
11) Koser, G. F.; Faircloth, J. N. J. Org. Chem. 1976, 41, 583.
12) Patrick, T. B.; Bechtold, D. S. J. Org. Chem. 1984, 49, 1935.
(13) W o¨ hrle, D.; Bohlen, H.; Rothkopf, H. W. Makromol. Chem. 1983,
84, 763.
1
In a homogeneous acidic aqueous solution, dissolved
quadricyclane has a short half-life and reacts rapidly to form
the two reaction products; however, the breakdown rates of
quadricyclane in the four natural soils were observed to be
much slower, and quadricyclane was found to be chemically
stable in soils at pH 6.0 and higher. Since 3 cm3 of
(
(
14) Gassman, P. G.; Yamaguchi, R. Tetrahedron 1982, 38, 1113.
15) Gee, G. W.; Bauder, J. W. In Methods of Soil Analysis, Part I,
Physical and Mineralogical Methods, 2nd ed.; Klute, A., Ed.; ASA:
Madison, WI, 1986; pp 383-411.
(
(
16) Karathanasis, A. D.; Hajek, B. F. Soil Sci. Am. J. 1984, 48, 413.
17) Hajek, B. F.; Adams, F.; Cope, J. T., Jr. Soil Sci. Am. Proc. 1972,
3
quadricyclane and 2 cm of water were added to each vial
3
6, 436
18) Etzweiler, F.; Senn, E.; Schmidt, H. W. H. Anal. Chem. 1995, 67,
55.
19) Cristol, S. J.; Morrill, T. C.; Sanchez, R. A. J. Org. Chem. 1966, 31,
726.
containing the soil, there is a large excess of quadricyclane
over the solubility limit, and we physically observed a separate
non-aqueous phase. The rate of quadricyclane reaction
therefore depends on the surface area of the quadricyclane-
water interface(s). If we assume that these interfacial areas
are the same for all soils, we would expect that the change
of rates of reaction with pH would be similar to those observed
in the homogeneous aqueous solutions. This is not the case,
so some other process must be occurring, although it is not
possible to draw any definite conclusions from the existing
data. The composition of the soil solution, soil particles, and
organic matter did not seem to induce any additional reactions
of quadricyclane or its reaction products.
(
(
(
6
2
20) Cristol, S. J.; Morrill, T. C.; Sanchez, R. A. J. Org. Chem. 1966, 31,
2733.
Received for review January 2, 1996. Revised manuscript
received October 8, 1996. Accepted October 15, 1996.
X
ES960060Y
X
Abstract published in Advance ACS Abstracts, January 1, 1997.
VOL. 31, NO. 3, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6 5 5