attempt to cyclopropanate the allyl alcohol 3 to (()-cis-
sabinene hydrate 4 (Scheme 1).7 The tertiary alcohol 3 was
Scheme 2
Scheme 1
ment with ice11) was subjected to the Simmons-Smith
reaction. The product 712 was immediately subjected to a
catalytic amount of HCl, leading to the ring expansion13
product 8. Thus, in this case the conventional allylic hydroxy-
directed cyclopropanation is unavailable and the lithium
oxyanionic version is a considerable advance.
reported to be “extremely unstable” and “when subjected to
a Simmons-Smith reaction, decomposed rapidly and no
bicyclic product could be isolated.”7 The allylic carbocation
derived by acid-induced dehydroxylation of 3 would be
exceptionally stable in view of the tertiary nature of both of
its termini. Since we required a sample of 4 for comparison
with one that we had synthesized as a demonstration of the
use of a new synthetic procedure based on the lithium-ene
cyclization,8 we sought to surmount this obstacle.
Since it is well established, as indicated above, that allylic
hydroxyl groups facilitate Simmons-Smith cyclopropana-
tions via complexation of the organozinc intermediate with
the hydroxyl group, it seemed possible that the lithium
oxyanionic group could be even more activating. Two tests
of this concept were applied. In the first, we attempted to
cyclopropanate the conjugate base of 9, which had been
reported14 to be unreactive toward 1.1 equiv of Simmons-
Smith reagent. Indeed, the inseparable mixture of product
and unreacted starting material contained 44% of the
cyclopropane (NMR, Scheme 3). In the second test, a large
We now present an operationally extremely simple and
apparently general solution to the instability of certain
alcohols under Simmons-Smith cyclopropanation condi-
tions. The procedure is illustrated with the most efficient
synthesis of (()-cis-sabinene hydrate 4 (Scheme 1). The
ketone 17 is treated with methyllithium in dry ether at -78
°C. After the reaction mixture is warmed to 0 °C, the
resulting lithium alkoxide 2 is subjected directly to the
Simmons-Smith conditions9 to provide the racemic natural
product in 64% yield.10 Thus, not only is the reaction
successful in the presence of the allylic oxyanionic group,
but even had the conversion of 3 to 4 been successful, the
route from 1 is shorter than that which would proceed
through 3.
Scheme 3
excess of an equimolar amount of the of the allyl methyl
ether 11 (R ) CH3)15 and the corresponding allylic lithium
The lithium oxyanion mediated procedure also allows one
to cyclopropanate allylic alcohols that cannot be isolated.
This is demonstrated by the alkylative ring expansion shown
in Scheme 2. The butyllithium adduct 6, the corresponding
alcohol of which is expected to be too unstable to isolate
(rearranging to a 3-substituted cyclohexen-2-one upon treat-
Scheme 4
(7) Fanta, W. I.; Erman, W. F. J. Org. Chem. 1968, 33, 1656-1659.
(8) Cheng, D.; Knox, K. R.; Cohen, T. J. Am. Chem. Soc. 2000, 122,
412-413.
oxyanion 11 (R ) Li) was subjected to the standard
(9) In the experiments described in this paper, the zinc-copper couple
is prepared freshly according to Shank, R. S.; Shechter, H. J. Org. Chem.
1959, 24, 1825-1826. The following general cyclopropanation procedure
is based on that of Dauben.4b MeLi (0.80 mL, 1.40 M in ether, 1.12 mmol)
was added under argon to a solution of enone 1 (1.0 mmol) dissolved in
ether (5 mL) at -78 °C, and the reaction mixture was warmed to 0 °C
before it was cannulated into a suspension of Zn-Cu couple(180 mg, 2.75
mmol), I2 (1 mg), and CH2I2 (576 mg, 2.15 mmol) in ether (20 mL). The
reaction mixture was heated at reflux for 2 h. It was then diluted with ether
(3 × 25 mL), and the organic phase was decanted. The latter was washed
with 5% K2CO3 (2 × 30 mL), brine (50 mL), and water (50 mL) and dried
(K2CO3). The solvent was removed to give an oily residue that was subjected
to column chromatography to give 99 mg (0.64 mmol, 64% yield) of the
cyclopropanation product 4.
cyclopropanation conditions.9 The cyclopropanation products
(11) Woods, G. F.; Griswold, P. H., Jr.; Armbrecht, B. H.; Blumenthal,
D. I.; Plapinger, R. J. Am. Chem. Soc. 1950, 72, 1645-1648.
(12) Bicyclic 7 can be isolated by careful column chromatography with
1
1% NEt3 in the eluent (hexanes/ethyl acetate 4:1). A reasonable H NMR
spectrum was obtained, but it was not further characterized because of its
instability. The 45% yield of 8 was obtained by treating the crude 7 with
a catalytic amount of HCl.
(13) Wenkert, E.; Buckwalter, B. L.; Sathe, S. S. Synth. Commun. 1973,
3, 261-264.
(14) Morikawa, T.; Sasaki, H.; Mori, K.; Shiro, M.; Taguchi, T. Chem.
Pharm. Bull. 1992, 40, 3189-3193.
(10) For previous syntheses of (()-cis-sabinene hydrate 4, see refs 7
and 8 and Gaoni, Y. Tetrahedron 1972, 28, 5525-5531.
(15) Shono, T.; Ikeda, A. J. Am. Chem. Soc. 1972, 94, 7892-7898.
Damico, R. J. Org. Chem. 1968, 33, 1550-1556.
2122
Org. Lett., Vol. 3, No. 13, 2001