A R T I C L E S
Adam et al.
Scheme 5
proposed as the logical intermediate in the reaction of sultene
A with cis- or trans-cyclooctene, except that the thiirenium
1
ion 3Ab has been directly observed. The following experimental
facts establish unequivocally the existence of the thiirenium ion
3
Ab: First, the reaction is reversible upon base treatment (K2-
CO3) when a Brønsted acid (TFA) is used for generating 3Ab;
second, the diene 6Ab is formed in slightly acidic or neutral
media. Expectedly, the reversal of the reaction, that is, the
quantitative recovery of the starting materials 1A and 2b, is
not possible on base treatment when the Lewis acid Sn(tpp)-
Figure 1. X-ray structure of episulfide 3Bb.
Scheme 4
(ClO4)2 is employed. This implies that the oxygen atom of
thiirenium ion 3Ab is protonated or ligated to the metal, which
makes 3Ab persist for many hours under acidic conditions.
The reversible generation of the persistent thiirenium ion 3Ab
is truly remarkable, since the few thiirenium ions that are known
3
,5
to date are irreversibly formed. In fact, most of the reported
i
thiirenium ions bear sterically demanding alkyl substituents ( -
t
Pr, Bu) and have nonnucleophilic counterions such as SbF6s
or BF4s to persist sufficiently for spectral observation.3 Only
,6
the di-tert-butyl-S-methyl thiirenium ion was persistent enough
5
c
at room temperature for X-ray analysis. The most common
transformations of such thiirenium ions are ring-opening reac-
tions, but Modena and Pasquato have also reported an unusual
t
ring expansion of a Bu-substituted thiirenium ion to its four-
membered thietium ion. Such ring opening of the thiirenium
ion 3Ab by the trifluoroacetic anion does not take place because
endoperoxide 1B was generated by photooxygenation of the
corresponding thiophene at -30 °C.1 Addition of the cyclooc-
tyne 2b to the photooxygenated mixture and warm-up to room
3
c
the nucleophile has to approach the vinyl carbon by backside
1
temperature (ca. 20 °C) revealed by H NMR spectroscopy that
5
f
attack (a SN2-Vin mechanism ) which is encumbered by steric
hindrance of the eight-membered ring. Rearrangement to a
thietium ion is also not feasible, since an unstabilized primary
carbocation would be formed.
after 3 h all of the thiophene endoperoxide 1B was consumed
and the episulfide 3Bb was obtained in 70% yield (Scheme 4).
Its structure was assigned by various NMR-spectral methods
(H,H-COSY, C,H-COSY, INADEQUATE), mass spectrometry,
The mechanism for the formation of the thiirenium ion 3Ab
is analogous to the sulfur transfer from the sultene 1A to the
cis/trans-2a cyclooctenes, in which the saturated thiiranium ion
and X-ray analysis (Figure 1). Addition of cyclooctyne 2b to
the thermolysate of the thiophene endoperoxide 1B did not lead
to the episulfide 3Bb. On treatment with triphenylphosphine at
room temperature (Scheme 4), expectedly this episulfide was
quickly (15 min) and quantitatively desulfurized to the pyran
intervenes as a short-lived intermediate to afford the cis/trans-
1
3
a thiiranes as the final products (Scheme 5). The reaction
entails nucleophilic attack by the strained triple or double bond
of the olefin or alkyne on the acid-activated sultene 1A. Usually,
a double bond is more reactive in its nucleophilicity than a triple
bond; however, a competition experiment revealed that the
reaction of sultene 1A with trans-cyclooctene (trans-2a) is
6
Bb. Also, heating of this episulfide in CDCl3 at 70 °C for 17
h caused 70% sulfur extrusion. When the thermolysis was
carried out in the presence of trans-cyclooctene (trans-2a), no
sulfur transfer was observed to afford the episulfide trans-3a.
The thiophene endoperoxide 1B did not effect sulfuration of
the dithiacyclononyne 2c. On thermolysis, only the desulfurized
products 4B and 5B of the endoperoxide 1B were detected,
along with elemental sulfur and unreacted alkyne 2c.
(
5) (a) Lucchini, V.; Modena, G.; Valle, G.; Capozzi, G. J. Org. Chem. 1981,
46, 4720-4724. (b) Capozzi, G.; DaCol, L.; Lucchini, V.; Modena, G.;
Valle, G. J. Chem. Soc., Perkin Trans. 2 1980, 68-73. (c) Capozzi, G.;
Lucchini, V.; Modena, G.; Scrimin, P. Tetrahedron Lett. 1977, 911-912.
(
d) Benati, L.; Montevecchi, P. C.; Spagnolo, P. Gazz. Chim. Ital. 1991,
Discussion
121, 387-391. (e) Destro, R.; Lucchini, V.; Modena, G.; Pasquato, L. J.
Org. Chem. 2000, 65, 3367-3370. (f) Modena, G.; Pasquato, L.; Lucchini,
V. Phosphorus, Silicon Sulfur 1994, 95-96, 265-282. (g) Lucchini, V.;
Modena, G.; Pasquato, L. Gazz. Chim. Ital. 1997, 127, 177-188.
6) Destro, R.; Pilati, T.; Simonetta, M. J. Chem. Soc., Chem. Commun. 1977,
576.
The formation of the thiirenium ion 3Ab in the reaction of
the sultene 1A with cyclooctyne 2b and stoichiometric amounts
of acid (Scheme 5) is analogous to that of the thiiranium ion,
(
8318 J. AM. CHEM. SOC.
9
VOL. 124, NO. 28, 2002