Slegt et al.
methanol or iodobenzene is available as base in the solvent
cage.1 However, if a vinyl cation of higher energy is involved
15a, and their iodine substituted counterparts, further substanti-
ates this conclusion.
2b
or if it is the triplet state of I2a that is involved, the propensity
for proton loss may be enhanced.1
9,37
Anyhow, the preferred
Experimental Section
reaction for structurally unbiased linear allenes is reprotonation
at the terminal allene carbons atoms,38 here yielding I2a and
Materials. The iodonium salts 4-tert-butylcyclohex-1-enyl-
(phenyl)iodonium tetrafluoroborate (1), cyclopent-1-enyl(phenyl)-
not I2b. This reaction sequence is not productive toward allylic
ether formation and is therefore disregarded.
Summarizing, we propose that I2b is produced from I2a by a
,3-H shift. As far as we know, this mode of rearrangement is
iodonium tetrafluoroborate (2), and cyclopent-1-enyl(4-methylphenyl)-
iodonium tetrafluoroborate were synthesized as described in refs 9
and 43. Methanol (HPLC grade) was checked to be UV transparent
and purged with Argon prior to use. n-Hexane and n-hexadecane
were used as received as were all available reference compounds
1
39
unprecedented in vinyl cation chemistry.
(5, 6, 7, 9b, 13, 14b, and 16). Nonavailable reference compounds
The data presented here constitute the first unambiguous case
for the direct formation of C5-ring vinyl cation I2a.40 Attempts
were synthesized following literature procedures with use of
commercially available starting materials.
to thermally generate vinyl cation I2a from cyclopent-1-enyl
triflate or nonaflate did yield cyclopentanone, but in both cases
Photochemistry. All reaction mixtures were purged with Argon
prior to irradiation. The solutions are irradiated in a quartz reaction
tube that was sealed with a rubber septum (to allow sampling) and
placed in a merry-go-round apparatus. A Hanau TNN-15/32 low-
pressure mercury lamp placed in a water-cooled quartz tube is used
to supply light with a main emission at λ ) 254 nm.
In a typical experiment 10 mL of a 5 mM solution of the
iodonium salt, containing 10 µL n-hexadecane as internal standard,
was irradiated. At appropriate time intervals 50 µL samples were
taken, using a syringe piercing through the septum. In kinetic runs,
samples are taken every 5 min. The samples were injected in a test
tube containing ∼1 mL demineralized water and 100 µL of
n-hexane or diethyl ether. Extraction was ensured by shaking the
stoppered test tube. After settling of the layers, the organic layer
was removed with a rinsed syringe and analyzed on GC and GC-
MS. At the end of the irradiation, the remainder of the reaction
mixture was reduced in volume by purging with nitrogen and, if
necessary, redissolved in 1 mL of diethyl ether. This sample is also
analyzed on GC and GC-MS. Experiments were carried out in
triplicate.
the reaction occurred through S-O bond fission rather than
vinyl-O bond fission.8
b,41
I2a has been proposed as an inter-
mediate in the photolysis of 1-iodocyclopentene in methanol at
25 °C, which yielded 1,1-dimethoxycyclopentane (12b) next
to cyclopentene (13). It may, however, well be that 12b, in
analogy with 4b, is a radical cation-derived rather than a vinyl
cation-derived product. The nonformation of vinyl ether 14a
-
19
(
and hence no cyclopentanone 14b), or allyl ether 16, is in line
with this idea.
Relative Photoefficiencies. Comparison of the rates of
formation of the leaving groups from 1 and 2 (at low conversion)
showed that photolysis of 2 is 1.4 times more efficient than
that of 1. Thus 2 is more photolabile than 1, while 2 is far less
thermolabile than 1. Presumably, the cyclopentenyl ligand
contributes less electron density to the carbon-iodine bond in
the reactive excited state than the cyclohexenyl moiety, which
lowers the photostability.
In the irradiation of 1, almost 5 times more iodobenzene (7)
is cleaved off (B-bond fission) than vinyl iodide 3 (A-bond
fission); in the photoreaction of 2 only 1.8 times more 7 than
vinyl iodide 11 is produced. These efficiencies reflect the
instability of the intermediates that are generated in these light-
induced reactions, the phenyl cation, the cyclohex-1-enyl cation,
and the cyclopent-1-enyl cation. This is in agreement with a
report about thienyl(aryl)iodonium salts that the cationic iodine
center allows π electron communication between the two ligands
via its d-orbitals.42 There is a substantial effect of one ligand at
the iodonium center on the bond to the other ligand.
The λexc ) 248 nm irradiations of 3 (R ) tert-butyl) are carried
out in a different setup. A high-pressure Hg/Xe arc, from which
the IR output was removed by a water filter, was used as the
irradiation source. The light beam was guided through a 77250
model Oriel monochromator to select the desired wavelength, and
aimed at a 3 mL quartz cell, equipped with a glass stopper with
Teflon septum. Per experiment 3 mL of 5 mM solutions of 3 and
1
mM n-hexadecane were used. The λexc ) 350 nm 9,10-
dicyanoanthracene-sensitized irradiation of 3 is carried out in a
Rayonet Photochemical Reactor, RPR 200, equipped with 350 nm
lamps, which was placed in a cool room (4 °C). Reaction mixtures
of 6 mM sensitizer and 50 mM starting material (in methanol, 10
mL) in Pyrex were used.
In summary, we have demonstrated that not only the C6-ring
vinyl cation, but also the C5-ring vinyl cation can be generated
photochemically, under mild reaction conditions. Compelling
evidence for the formation of the C5-ring vinyl cation I2a is the
production, next to the direct trapping products 14a (and 14b),
of the allylic ether 16, which is the trapping product of the
hydride shifted intermediate I2b. The formation of 17a next to
Products. All reaction products were characterized by compari-
son of their retention times on analytical GC and of their mass
spectra (by GC-MS) with those of authentic samples, using
equipment described in ref 12b. In many cases the product mixture
was co-injected with the alleged product. Benzene (5), anisole (6),
iodobenzene (7), 4-tert-butylcyclohexanone (9b), cyclopentene (13),
cyclopentanone (14b), and 3-methoxycyclopentene (16) are com-
mercially available. Most other products were synthesized according
1
9b
to literature procedures: 4-tert-butyl-1-iodocyclohexene (3),
44
(
37) Huck, L. A.; Wan, P. Org. Lett 2004, 6, 1797-1799.
4-tert-butyl-1,1-dimethoxycyclohexane (4b), 4-tert-butylcyclo-
(38) (a) Fornarini, S.; Speranza, M.; Attina, M.; Cacace, F.; Giacomello,
45
46,47
44
hexene (8), 4-tert-butylcyclohex-1-enylbenzene (10a),
1-io-
P. J. Am. Chem. Soc. 1984, 106, 2498-2501. (b) Cramer, P.; Tidwell, T.
T. J. Org. Chem. 1981, 46, 2683-2686.
19b,48
docyclopentene (11),
1,1-dimethoxycyclopentane (12b), 1-meth-
(
39) (a) Newman, M. S.; Beard, C. D. J. Am. Chem. Soc. 1970, 92, 7564-
7
5
567. (b) Shchegolev, A. A.; Kanishchev, M. I. Russ. Chem. ReV. 1981,
(43) Ochiai, M.; Shu, T.; Nagaoka, T.; Kitagawa, Y. J. Org. Chem. 1997,
62, 2130-2138.
(44) Eliel, E. L.; Badding, V. G.; Rerick, M. N. J. Am. Chem. Soc. 1962,
84, 2371-2377.
(45) Stork, G.; White, W. N. J. Am. Chem. Soc. 1956, 78, 4604-4608.
(46) (a) Garbisch, E. W. J. Org. Chem. 1961, 26, 4165-4166. (b) Larock,
R. C.; Baker, B. E. Tetrahedron Lett. 1988, 29, 905-908.
(47) Littke, A. F.; Dai, C.; Fu, G. C. J. Am. Chem. Soc. 2000, 122, 4020-
4028.
0, 553-564.
(40) Indirect formation has been reported (via π-participation or by using
release of ring strain as the driving force): (a) Mayr, H.; Seitz, B.;
Halberstadt-Kausch, I. K. J. Org. Chem. 1981, 46, 1041-1044. (b)
Reference 8c. (c) Reference 19.
(
41) Hanack, M. Angew. Chem. 1978, 90, 346-359.
(42) Bykowski, D.; McDonald, R.; Hinkle, R. J.; Tykwinski, R. R. J.
Org. Chem. 2002, 67, 2798-2804.
2234 J. Org. Chem., Vol. 71, No. 6, 2006