The macrocyclization reaction of terminal dibromoalkanes with sulfide on alumina. The use of a solid support as an alternative to the high dilution technique

Article abstract of DOI:10.1016/0040-4039(93)88023-C

The reaction of a series of terminal dibromoalkanes with S^-2 on Al₂O₃ has been examined. The use of a solid support for the macrocyclization represents a viable alternate procedure to the more traditional high dilution technique in solution. The cyclization also occur with thioacetamide on unactivated alumina.

Full text of DOI:10.1016/0040-4039(93)88023-C

Vol. 33. No.
Britain
pp.
$5.00 +
Printed
The Macrocyclization Reaction of Terminal Dibromoalkanes
with Sulfide on Alumina. The Use of a Solid Support as
an Alternative to the High Dilution Technique.
Lay Choo Tan,’ Richard M.
George W. Kabalka,*
Marc Hillmyer and Julie Woosley
e, Knoxville, Tennessee
USA.
Department of Chemistry, University of
The
of a series of terminal
represents a
with
alternate procedure to the more traditional hi dilution
alumina.
on
has been examined. The use of
a solid support for the
technique in solution. The cyclizations also occur with thioacetamide on
The synthesis of large rings from acyclic precursors (macrocyclization) has always been difficult to
achieve. The reason for this is that the desired unimolecular cyclization reaction competes with an
analogous bimolecular reaction. To inhibit the bimolecular reaction, the chemistry is normally performed
under high dilution, either by using a large volume of solvent or by very slowly adding the reagents to
a small volume of solvent.* A machine has even been invented to facilitate the slow addition of
Because the high dilution technique does not always work well, alternate procedures are desirable.
The surface of a solid would appear to be a viable medium in which to run the macrocyclization
reactions. The reaction of a series of terminal dibromoalkanes with
represents a test of this concept. In the first step, reacts with
on
in two steps (eq. 1)
to form a surface-bound,
immobile thiolate. If the distance between the sulfur of the thiolate and the closest
the length of the thiolate, it follows that the bimolecular reaction of thiolate with
only cyclization is possible. One can control the average distance between
is greater than
cannot occur and
by controlling the
quantity of
adsorbed on the surface. Thus, the yield of cyclized product should increase as the
concentration of sulfide on the surface is
Br

7710
Another factor may also favor the macrocyclization reaction on solids. There is experimental’ and
theoretical6 evidence that the rate at which two ends of a chain find each other is considerably greater
when one end of the chain is immobile than when both ends of the chain are free to move. This will
favor the unmolecular reaction, at the expense of the bimolecular reaction, on the solid as compared to
solution counterpart.
Regen and co-workers were the first to prepare
close to quantitative yield.’ and
and use it to prepare dialkyl sulfides in
have subsequently used the reagent in the syntheses
of a few cyclic sulfides containing two triple bonds. Our reagent is prepared by adsorption of
from
aqueous solution onto Brockmann Activity 1 neutral alumina and then activating the
resulting solid at a given temperature under vacuum.
chemical and physical properties. Unlike
has interesting, although unexplained,
the reagent is colored and the color
and
depends on the concentration of
The reagent must either be used immediately or stored under argon
because, when exposed to air, the free-flowing powder becomes gummy and gives poor yields of desired
product. Activation temperature is also critical: the yield of product decreases as the activation
temperature of the solid increases beyond 95”. A solid activated at 95” gives the best results.
The reactions were performed by
a solution of
in 10 ml of toluene with
at 95” for 20 hours; longer reaction times did not improve yields. Solvent and alumina are
important because reactions run in the absence of solvent or alumina produced very low yields of
products. The ratio of
reactions run on freshly prepared
to
was kept at approximately
are summarized in Table 1.
in all cases. The results of
Table 1. Cyclic Sulfide Yields in the Reaction of
with
Concentration of
on
x
S i z e
x
1.5
5
6
7
8
44.7
55.1
59.7
38.4
79.2
62.4
47.1
9
26.2
12.6
9.0
15.6
19.1
37.7
15.7
12.4
20.4
28.6
10
11
13
17
a)
of
activated at 95”. b)
activated at
c) Reaction run in a sealed tube at
d) Yield
The first thing to note is that all the cyclization reactions attempted were successful, even with
and but the product yields varied greatly. A plot of yield versus ring size (not

7711
shown) follows the usual pattern observed for cyclixation reactions? the yields are low for small rings,
increase steadily for through 7-membered rings, gradually decrease for the larger through
membered rings, and then gradually increase for still larger rings. Our
compare favorably to the
high dilution reactions run in
reactions run with solid
ethanol” or
at
as well as the heterogeneous
in ethanol at room temperature.” Our yields are considerably
higher than those obtained in refluxing ethanol and similar in several cases to those obtained in the other
two media. Even in those cases where our yields are not as good as those obtained in
and
the surface chemistry may still be superior because of the ease which the
reactions are run. Although the data are not yet extensive, the prediction that the yield of product goes
up as the concentration of
more methylene groups.
on
goes down is born out for reactions which form rings with 7 or
A few reactions were also run in solution, but not with the high dilution technique.
Dibromopentane and 0.833 M in a ratio in ethanol and in HMPA at 95”
afforded only trace amounts of pentamethylene sulfide after 20 hours, and in
HMPA gave no cyclic sulfide. Because the average distance between sulfide anions in these solutions
(12.6 A) is considerably greater than it is for 6.32 x mol of alumina the least
concentrated surface used in this study, the difference in product yields between the two methods is
particularly noteworthy.
There is an alternate way in which to generate
affords which has been used classically to precipitate metal cations and, more recently, in the
synthesis of large Al,
in situ on
Hydrolysis of thioacetamide
0 has a surface layer of OH groups (and physisorbed
Indeed, when (1.00 mmol) is treated with
(1.0 g) in refluxing toluene for 10 hours under argon,
which could react with thioacetamide to form
thioacetamide (2.00 mmol) and
the corresponding cyclic sulfide and disulfide are formed in respectable yields (eq. 2). Other
combinations of thioacetamide and
experiments are shown in Table 2.
afforded lower yields of products. The results of these
+
Table 2.
Yields of Cyclic Sulfides and Disulfides in the Reaction of
with
Precursor
Cyclic
Cyclic
28.3
17.9
8.9
31.2
26.3
16.4
28.4
26.4
6
9
12
15.2
25.3
(Silica Gel)

7712
These reactions are interesting because significant amounts of cyclic disulfides are generated in
addition to the expected sulfides. At least with
contains surface OH groups and physisorbed
unactivated
is born out by the fact that no reaction occurred in the absence of unactivated
unactivated silica, which also
gave comparable yields of the two products as
did. That the presence and nature of the
used in the
is important
or the presence of
activated at
groups?4
Why do these reactions afford disulfides when the ones on
chemistry involves anions such as
which removes its physisorbed
and approximately 50% of the surface OH
do not? In the latter case the
and thiolates which remain attached to the surface. It is not
possible to liberate dithiols, which are the precursors to the disulfides, from the surface by extraction of
the solid with toluene or diethyl ether. With thioacetamide, however, the chemistry involves neutral
species such as
react, at another. Here it is possible to generate dithiols in solution which form disulfides on oxidation
with during workup (eqs. 3-4).
and thiols, which are free to leave the surface at one place and return, and perhaps
(3)
Cyclic Disulfide
(4)
Acknowledgment: This work was supported by the National Science Foundation, Research Corporation
and the Department of Energy. MH and JW thank the University of Tennessee Science Alliance Center
of Excellence for support.
REFERENCES AND NOTES
1.
Present address: Department of Chemistry, Cumberland College, Williamsburg, KY 40769.
F. Topics in Current Chemistry
F. Topics in Current Chemistry 1983,113, 1.
F. Chem. Res. 1981, 314.
2.
(a) Knops, P.; Sendhoff, N.; Mekelburger, H.-B.;
1. (b)
L.,
3.
Korbach, S.;
W.;
4.
The absolute amount of product is also important. At very low concentrations of
it is possible
to have a very high yield, but few mmoles of product.
5.
6.
7.
8.
9.
Forbes, M.D.E.; Myers, T.L.; Dukes,
D.; Sumpter, B. unpublished results.
Maynard,
Am. Chem.
Czech, B.;
S.; Regen, S.L. Synthesis 1980, 113.
Gleiter, R.; Ritter, J.; Imgartinger, H.;
Nicolaou, K.C.; Skokotas, G.; Maligres, P.;
J. Tetruhedron
Zuccarello, G.; Schweiger, E.J. Toshima,
1272.
1991, 2883.
Wendebom, S.
Chem. Int. Ed.
10.
11.
12.
13.
14.
See, for example, Nakagaki, R.; Sakuragi, H.; Mutai, K. Phys. Org. Chem.
Friedman, P.; Allen, P., Jr.; Org. Chem. 1965, 30, 780 and references cited therein.
Mandolini, L.; Vontor, T. Synth. 1979, 9, 857.
Singh, A.; Mehrotra, A.; Regen, S.L. Synth. Commun. 1981,
H.; Ratnasamy, P. Rev.-Sci. 1978, 17, 31.
187.
409.
(Received in USA 20 May 1992; accepted 21 September 1992)