Regioselective lewis acid-mediated [1,3] rearrangement of allylvinyl ethers

Article abstract of DOI:10.1021/ol0505151

(Chemical Equation Presented) Aluminum and copper Lewis acids were implemented to effect a regioselective [1,3] rearrangement of allylvinyl ethers in moderate to good yields. The use of trisubstituted alkenes leads to depressed levels of Claisen products.

Full text of DOI:10.1021/ol0505151

ORGANIC
LETTERS
2005
Vol. 7, No. 11
2173-2176
Regioselective Lewis Acid-Mediated [1,3]
Rearrangement of Allylvinyl Ethers
Christopher G. Nasveschuk and Tomislav Rovis*
Department of Chemistry, Colorado State UniVersity, Fort Collins, Colorado 80523
roVis@lamar.colostate.edu
Received March 9, 2005
ABSTRACT
Aluminum and copper Lewis acids were implemented to effect a regioselective [1,3] rearrangement of allylvinyl ethers in moderate to good
yields. The use of trisubstituted alkenes leads to depressed levels of Claisen products.
The formation of sterically congested carbon-carbon bonds
with control of stereochemistry still represents a formidable
challenge in organic synthesis. One tool that has met some
of these challenges is the Claisen rearrangement, a [3,3]
rearrangement of allylvinyl ethers.1 This thermal sigmatropic
rearrangement has been shown to relay stereochemical
information from the cleaving C-O bond to the forming
C-C bond via a concerted transition state.2 More recently,
the use of Lewis acids has provided rate acceleration for the
Claisen rearrangement.1 However, under the Lewis acid
manifold, ionic intermediates have been proposed, which
diverges from the generally accepted concerted mechanism.3
If true ionic intermediates do exist, then other regioisomeric
products should be observed. Yamamoto and co-workers
reported that pentadienyl vinyl ethers undergo rearrangement
to produce a mixture of [1,3], [3,3], and [5, 3] regioisomers
in the presence of their bulky aluminum bisphenoxy Lewis
acid.3 Grieco and co-workers, under strongly ionizing condi-
tions, were able to selectively produce [1,3] rearrangement
products from sterically hindered allylvinyl ethers.4 Gansauer
has also illustrated a Cu(OTf)2- and B(C6F5)3-catalyzed
regioselective [1,3] rearrangement of tert-butyl alcohol,
benzyl alcohol, and isophorone-derived vinyl ethers.5,6 In
these cases, the Lewis acid-mediated [1,3] rearrangement of
allylvinyl ethers necessarily proceeds through an allyl cation
and metallo-enolate ion pair. As part of our efforts aimed at
developing stereoselective [1,3] rearrangements,7,8 we initi-
ated a study to harness this reactivity, elucidate the factors
that govern regioselectivity, and explore the scope of the
[1,3] rearrangement of allylvinyl ethers (Scheme 1). Herein,
we disclose our results.
Scheme 1. [3,3] vs [1,3] Rearrangement
Experimentation began with a brief Lewis acid screen
using cinnamyl vinyl ether 1. Cu(II), Sn(IV), and Ti(IV)
(5) (a) Gansauer, A.; Fielenbach, D.; Stock, C. AdV. Synth. Catal. 2002,
344, 845-848. (b) Gansauer, A.; Fielenbach, D.; Stock, C.; Geich-Gimbel,
D. AdV. Synth. Catal. 2003, 345, 1017-1030.
(6) Other examples of [1,3] rearrangements include: (a) Shi, G.; Cai,
W. J. Org. Chem. 1995, 60, 6289-6295. (b) Sperling, D.; Reissig, H.-U.;
Fabian, J. Eur. J. Org. Chem. 1999, 1107-1114. (c) Shiina, I.; Nagasue,
H. Tetrahedron Lett. 2002, 43, 5837-5840. (d) Novikov, A. V.; Kennedy,
A. R.; Rainier, J. D. J. Org. Chem. 2003, 68, 993-996. (e) Suzuki, T.;
Inui, M.; Hosokawa, S.; Kobayashi, S. Tetrahedron Lett. 2003, 44, 3713-
3716. (f) Cai, Y.; Roberts, B. P. Tetrahedron Lett. 2003, 44, 4645-4648.
(7) Zhang, Y.; Reynolds, N. T.; Manju, K.; Rovis, T. J. Am. Chem. Soc.
2002, 124, 9720-9721.
(1) (a) Lutz, R. P. Chem. ReV. 1984, 84, 223-247. (b) Wipf, P. In
ComprehensiVe Organic Synthesis; Trost, B. M., Ed.; Pergamon: Oxford,
1991; Vol. 5, pp 827-873. (c) Martin Castro, A. M. Chem. ReV. 2004,
104, 2939-3002.
(2) For a mechanistic discussion, see: Gajewski, J. J. Acc. Chem. Res.
1997, 30, 219-225.
(3) Nonoshita, K.; Banno, H.; Maruoka, K.; Yamamoto, H. J. Am. Chem.
Soc. 1990, 112, 316-322.
(4) Grieco, P. A.; Clark, J. D.; Jagoe, C. T. J. Am. Chem. Soc. 1991,
113, 5488-5489.
(8) Nasveschuk, C. G.; Rovis, T. Angew. Chem., Int. Ed. http://
www3.interscience.wiley.com/cgi-bin/fulltext/110473806/PDFSTART.
10.1021/ol0505151 CCC: $30.25
© 2005 American Chemical Society
Published on Web 05/03/2005

Lewis acids provided mixtures of [1,3] and [3,3] adducts
(entries 1-3, Table 1). The reaction worked equally well in
Table 2. Electronic Effect of a Para Substituent
Table 1. Initial Lewis Acid Screen
a Reactions conducted using 1.05 equiv of Lewis acid. b Combined yield.
c Reaction was warmed to 23 °C and starting material isolated.
to secondary Lewis acid coordination to the methoxy group.
Electron-withdrawing substitution about the ring (6) resulted
in recovered starting material, even upon warming the reac-
tion mixture from -78 °C to ambient temperature (entry 4).
To further explore why electronic variation of the substrate
did not afford significant improvements in [1,3] regioselec-
tivity, a control experiment was designed (Scheme 2). The
a Reactions conducted in toluene or CH2Cl2 using 1.05 equiv of Lewis
acid. b Reaction performed at -50 °C. c Combined isolated yield.
CH2Cl2 or toluene. Lewis basic solvents were found to be
inferior.
Bisphenoxy Al complexes seemed to be better suited for
this study due to the electronic and steric perturbations that
could be achieved through substitution of the phenol ring.9
We identified two different Lewis acids that gave comple-
mentary regioselectivity for the rearrangement of cinnama-
ldehyde-derived allylvinyl ether 1 (entries 4 and 5, Table
1). Steric bulk about the Al-metal center proved to be optimal
for the [3,3] product, while halide substitution produced a
stronger Lewis acid that was moderately [1,3] selective.
While this dichotomy is not easy to rationalize, it is evident
that there is a fine line between accelerating the concerted
[3,3] rearrangement versus generating an ionic intermediate.
An increase in the strength of the Lewis acid or in the
stability of the allyl cation should result in increased [1,3]
selectivity.
Scheme 2. Control Experiment
regioisomeric allylvinyl ether 7 was synthesized and sub-
jected to the optimized conditions. In the event both
regioisomeric allylvinyl ethers provided the same major
product. We hypothesize that if an ionic mechanism is
operating, the allyl cation and enolate size determines the
observed regioselectivity.10 Indeed, a crossover experiment
using deuterium-labeled vinyl ether illustrated nearly com-
We next sought better [1,3] selectivity through electronic
variation at the 4-phenyl position of the model substrate. As
expected, on the basis of the above rationale, 4-methyl sub-
stitution (4) increased the [1,3] selectivity to 81:19 (entry 2,
Table 2). However, 4-methoxy substitution (5) resulted in
decreased regioselectivity (entry 3), a situation we ascribe
(9) (a) Yamamoto, H.; Maruoka, K. Pure Appl. Chem. 1988, 60, 21-
26. (b) Yamamoto, H.; Yanagisawa, A.; Ishihara, K.; Saito, S. Pure Appl.
Chem. 1998, 70, 1507-1512.
(10) Jia, Z. S.; Ottosson, H.; Zeng, X.; Thibblin, A. J. Org. Chem. 2002,
67, 182-187.
2174
Org. Lett., Vol. 7, No. 11, 2005

Scheme 3. Proposal for Regiocontrol
Table 4. Scope of [1,3] Rearrangement
plete scrambling of the label.11 In an attempt to further control
regioselectivity, we turned our attention to allyl systems
bearing trisubstituted alkenes. It is known that elevated
temperatures are required in systems that undergo [3,3]
rearrangement to form quaternary carbon centers.1 We
thought to apply this apparent characteristic of the Claisen
rearrangement to our methodology: under ionic conditions,
[3,3] recombination of the metallo enolate and allyl cation
to form a quaternary carbon center at the tertiary cation
should be slow relative to [1,3] recombination at the less
hindered secondary cation (Scheme 3).
Trisubstituted allylvinyl ethers were easily prepared by
carboalumination of commercially available alkynes,12 trap-
ping with aldehydes, and Hg(II)-catalyzed transvinyl-
ation.13 A variety of Lewis acids afforded the [1,3] adduct
in moderate to good yield and excellent regioselectivity
(Table 3).
a Reactions conducted using 5 mol % Lewis acid at -50 °C for 5 h.
b Reaction conducted at -30 °C for 48 h.
Table 3. Lewis Acid Screen
The reaction seems to be somewhat insulated from
electronic issues, albeit the yield and the selectivity decreased
slightly by switching from aromatic to alkyl substitution of
the double bond (entries 6 and 7). Intriguingly, we observe
diminished selectivities when a tert-butyl group (13) is
present at the 1-position and the reaction faces a choice
between attack at the tertiary position ([3,3]) or the neopentyl
position ([1,3]) (entry 4). Rearrangement of an isopropenyl
vinyl ether produced a 50:50 mixture of regioisomers when
using Cu(OTf)2 as the Lewis acid (eq 1). Increasing the
strength of the Lewis acid restored the regioselectivity. We
believe this is consistent with our hypothesis that a weaker
Lewis acid is capable of accelerating the Claisen via a
concerted pathway, while the stronger Lewis acid results in
bond ionization.
a Reactions conducted in toluene or CH2Cl2 using 1.05 equiv of Lewis
acid. b Reaction performed using 10 mol % Lewis acid.
In conclusion, we have reported a regioselective [1,3]
rearrangement of allylvinyl ethers. Reactivity issues concern-
We were delighted to find that a broad range of trisub-
stituted allylvinyl ethers undergo [1,3] rearrangement in
exceptionally high selectivity (Table 4).
(12) Wipf, P.; Lim, S. Angew. Chem., Int. Ed. Engl. 1993, 32, 1068-
1071.
(13) Maruoka, K.; Banno, H.; Yamamoto, H. Tetrahedron: Asymmetry
1991, 2, 647-662.
(11) See Supporting Information for details.
Org. Lett., Vol. 7, No. 11, 2005
2175

ing regioselective [1,3] rearrangement of allylvinyl ethers
in terms of Lewis acidity and substrate bias have been
highlighted. Work in this area to identify chiral Lewis acids
that effect an enantio- and regioselective [1,3] rearrangement
of allyl vinyl ethers is currently under way.
support. T.R. is a fellow of the Alfred P. Sloan Foundation.
We thank Professors Charles Henry and Alan Kennan for
helpful discussions and Donald Dick for technical assistance.
Supporting Information Available: Experimental pro-
cedures and characterization data of all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Acknowledgment. Financial support has been provided
by the National Institute of General Medical Sciences
(GM65407). We also thank Merck Research Laboratories,
GlaxoSmithKline, Amgen, and Eli Lilly for unrestricted
OL0505151
2176
Org. Lett., Vol. 7, No. 11, 2005