Nazarov cyclizations of an allenyl vinyl ketone with interception of the oxyallyl cation intermediate for the formation of carbon-carbon bonds

Article abstract of DOI:10.1021/ja909073r

Treatment of an allenyl vinyl ketone with BF₃ ? Et ₂O leads to a cyclic oxyallyl cation by a Nazarov reaction, and when this reaction is conducted in the presence of an acyclic diene, [4 + 3] and [3 + 2] products are obtained efficiently with high regio- and stereoselectivity. The proportion of [4 + 3] to [3 + 2] product depends on the substitution on the diene. Cyclic dienes react with the oxyallyl cation by forming only one carbon-carbon bond, but the site of bond formation can be affected by steric hindrance. Electronrich alkenes intercept the allyl cation by forming one carbon-carbon bond, or two carbon-carbon bonds through [3 + 2] cyclization. In some instances, further treatment of the initial products with BF3 ? Et2O leads to equilibrated products in good yield.

Full text of DOI:10.1021/ja909073r

Published on Web 01/15/2010
Nazarov Cyclizations of an Allenyl Vinyl Ketone with
Interception of the Oxyallyl Cation Intermediate for the
Formation of Carbon-Carbon Bonds
Vanessa M. Marx and D. Jean Burnell*
Department of Chemistry, Dalhousie UniVersity, Halifax, NoVa Scotia B3H 4J3, Canada
Received October 30, 2009; E-mail: jean.burnell@dal.ca
Abstract: Treatment of an allenyl vinyl ketone with BF3 ·Et2O leads to a cyclic oxyallyl cation by a Nazarov
reaction, and when this reaction is conducted in the presence of an acyclic diene, [4 + 3] and [3 + 2]
products are obtained efficiently with high regio- and stereoselectivity. The proportion of [4 + 3] to [3 + 2]
product depends on the substitution on the diene. Cyclic dienes react with the oxyallyl cation by forming
only one carbon-carbon bond, but the site of bond formation can be affected by steric hindrance. Electron-
rich alkenes intercept the allyl cation by forming one carbon-carbon bond, or two carbon-carbon bonds
through [3 + 2] cyclization. In some instances, further treatment of the initial products with BF3 ·Et2O leads
to equilibrated products in good yield.
Scheme 1. Interrupted Nazarov Reaction of AVK 1
Introduction
Oxyallyl cations are key intermediates in the synthesis of
seven-membered carbocycles via [4 + 3] cyclizations with 1,3-
dienes.1 The reaction is known to proceed efficiently with furan
and cyclopentadiene, but the analogous reaction with acyclic
dienes remains underexplored likely because this process can
be complicated by alternative [4 + 2] and [3 + 2] cyclizations
and by competing decomposition pathways.2 Recently, oxyallyl
cations generated by Nazarov cyclization3,4 have been shown
to be effective partners in [4 + 3] cyclizations with furan,
cyclopentadiene, and a few acyclic dienes.5 Furthermore, there
are two reports of Nazarov cyclizations being followed by [3
+ 2] cyclizations.6-8 Recently, we described Nazarov cycliza-
tions of allenyl vinyl ketones (AVK’s) such as 1,9,10 where the
only products isolated were those resulting from interception
of the cationic intermediate 2 by heteroatom nucleophiles, giving
3a by interception at position a, or 3c by interception at position
c (Scheme 1).
There are at least four reasons why the cationic intermediate
2 presents an especially well suited coupling partner for
cyclization reactions: (1) An AVK is a particularly reactive
substrate for the Nazarov reaction,11 cyclizing rapidly under
acidic conditions to generate the oxyallyl cation. (2) The oxyallyl
cation 2 is stabilized by extra conjugation and simple loss of a
(1) (a) Noyori, R. Acc. Chem. Res. 1979, 12, 61. (b) Hoffmann, H. M. R.
Angew. Chem., Int. Ed. Engl. 1984, 23, 1. (c) Mann, J. Tetrahedron
1986, 42, 4611. (d) Rigby, J. H.; Pigge, F. C. Org. React. 1997, 51,
351. (e) Hosomi, A.; Tominagi, Y. in ComprehensiVe Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon: Oxford, 1991;
Vol. 5, pp 593-615. (f) Harmata, M. Acc. Chem. Res. 2001, 34, 595.
(g) Harmata, M.; Rashatasakhon, P. Tetrahedron 2003, 59, 2371, and
references therein.
(2) (a) Chidgey, R.; Hoffmann, H. M. R. Tetrahedron Lett. 1977, 2633.
(b) Hoffmann, H. M. R.; Chidgey, R. Tetrahedron Lett. 1978, 85. (c)
Takaya, H.; Makino, S.; Hayakawa, H.; Noyori, R. J. Am. Chem. Soc.
1978, 100, 1765. (d) Sasaki, T.; Ishibashi, Y.; Ohno, M. Tetrahedron
Lett. 1982, 23, 1693. (e) Blackburn, C.; Childs, R. F.; Kennedy, R. A.
Can. J. Chem. 1983, 61, 1981. (f) Fo¨hlisch, B.; Korfant, H.; Meining,
H.; Frey, W. Eur. J. Org. Chem. 2000, 1335. (g) Fujita, M.; Oshima,
M.; Okuno, S.; Sugimura, T.; Okuyama, T. Org. Lett. 2006, 8, 4113.
(3) (a) Pellissier, H. Tetrahedron 2005, 61, 6479. (b) Frontier, A. J.;
Collison, C. Tetrahedron 2005, 61, 7577. (c) Tius, M. A. Eur. J. Org.
Chem. 2005, 2193.
(8) Regarding [3 + 2] cyclizations: see refs 1f,g, 2g and (a) Hayakawa,
Y.; Yokoyama, K.; Noyori, R. J. Am. Chem. Soc. 1978, 100, 1799.
(b) Lee, T. V.; Richardson, K. A.; Taylor, D. A. Tetrahedron Lett.
1986, 27, 5021. (c) Lee, T. V.; Boucher, R. J.; Porter, J. R.; Taylor,
D. A. Tetrahedron 1988, 44, 4233. (d) Lee, T. V.; Porter, J. R.; Roden,
F. S. Tetrahedron 1991, 47, 139. (e) Hardinger, S. A.; Bayne, C.;
Kantorowski, E.; McClennan, R.; Larres, L.; Nuesse, M.-A. J. Org.
Chem. 1995, 60, 1104. (f) Masuya, K.; Domon, K.; Tanino, K.;
Kuwajima, I. J. Am. Chem. Soc. 1998, 120, 1724. (g) Mizuno, H.;
Domon, K.; Masuya, K.; Tanino, K.; Kuwajima, K. J. Org. Chem.
1999, 64, 2648. (h) Nakamura, E.; Yamago, S. Acc. Chem. Res. 2002,
35, 867.
(4) Grant, T. N.; Rieder, C. J.; West, F. G. Chem. Commun. 2009, 5676,
and references therein.
(5) (a) Wang, Y.; Schill, B. D.; Arif, A. M.; West, F. G. Org. Lett. 2003,
5, 2747. Intramolecular [4 + 3] cyclization: (b) Wang, Y.; Arif, A. M.;
West, F. G. J. Am. Chem. Soc. 1999, 121, 876. (c) Yungai, A.; West,
F. G. Tetrahedron Lett. 2004, 45, 5445.
(6) Giese, S.; Kastrup, L.; Stiens, D.; West, F. G. Angew. Chem., Int. Ed.
2000, 39, 1970.
(9) Marx, V. M.; Burnell, D. J. Org. Lett. 2009, 11, 1229.
(10) Marx, V. M.; Cameron, T. S.; Burnell, D. J. Tetrahedron Lett. 2009,
50, 7213.
(7) Mahmoud, B.; West, F. W. Tetrahedron Lett. 2007, 48, 5091.
9
10.1021/ja909073r  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 1685–1689 1685

A R T I C L E S
Marx and Burnell
Table 1. Reactions of AVK 1 with Substituted Butadienes (2 equiv)
Mediated by BF3 ·Et2O
Scheme 2. Geometries of Cycloadditions that would Lead to
[4 + 3] and [3 + 2] Products
nucleophile to trap the oxyallyl cation derived from 1, had been
seen to be generated from this Lewis acid.9 AVK 1 was treated
with BF3 ·Et2O in the presence of various methyl-substituted
butadienes (Table 1). Reactions were rapid (5 min at -78 °C)
and yielded products of [4 + 3] and [3 + 2] cyclizations onto
positions a and b in 2. Whereas BF3 ·Et2O elicited the quantita-
tive formation of 812 (entry 4) in 5 min, InCl3 and Cu(OTf)2
gave 8 more slowly (2 h) and in yields of only 59 and 68%,
respectively. InCl3 is known to give the product of trapping 2
with chloride at position c,10 but the reaction of the diene with
2 must be much faster because chlorinated products were not
observed when the diene was added. This was true of other
experiments involving alkenes with InCl3 (Vide infra). It is
interesting to note that the much more electron-rich trans-1-
methoxy-3-trimethylsilyloxy-1,3-butadiene did not yield any
Nazarov product with 1.
a 7a R3 ) Me, R4 ) H; 7b R3 ) H, R4 ) Me. b Intractable material.
c 11a R1 endo, 11b R1 exo, ratio 1:1. d 12a R1 endo, 12b R1 exo, ratio
5:1, respectively.
proton from 2 to generate a double bond might be unfavorable
as the product would be a fulvene. Thus, the oxyallyl cation
would be more likely to be trapped. (3) The products resulting
from either [4 + 3] or [3 + 2] cyclizations would contain an
exocyclic methylene unit, providing a good synthetic handle
for further transformations. (4) The oxyallyl cation 2 has two
unencumbered positions, a and c, to which a nucleophile might
add readily, which would allow for the production of diverse
cyclopentenone-containing ring systems.
We present here the first survey of Nazarov reactions of an
AVK accompanied by tandem carbon-carbon bond formation
that leads to a diversity of product types with high regio- and
stereoselectivity and often in good to excellent yield. The
product types depend on the nature of the carbon nucleophile,
and the range of examples will now allow generalizations to
be made regarding the outcomes of the reactions.
The regioselectivity of all cyclizations,13 with the exception
of entry 3, was high, and this could be rationalized as arising
from the most reactive cationic site in 2 being a and significant
amounts of charge in the diene moieties at the transition states.
NOE measurements showed that every product had the phenyl
group in an exo position, indicating a high degree of facial
selectivity in the additions to 2. The only [4 + 3] product using
trans-piperylene was 4a (entry 1), but calculations14 showed
that its epimer 4b is 3.0 kJ.mol-1 lower in energy. These findings
would be consistent with a concerted, but asynchronous,
cycloaddition mechanism via an extended geometry similar to
13 in Scheme 2 for the formation of the [4 + 3] products. This
was consistent with a computational study involving the reaction
of butadiene with a metal-bound acyclic oxyallyl species, which
had suggested a concerted pathway to be lower in energy than
a stepwise one for adducts resulting from extended transition
states.15
Results and Discussion
BF3 ·Et2O was chosen as the acid for most reactions because
no nucleophilic species, which might compete with a carbon-
(11) (a) Tius, M. A.; Kwok, C.-K.; Gu, X.; Zhao, C. Synth. Commun. 1994,
24, 871. (b) Hashmi, A. S. K.; Bats, J. W.; Choi, J.-H.; Schwarz, L.
Tetrahedron Lett. 1998, 39, 7491. (c) Tius, M. A. Acc. Chem. Res.
2003, 36, 284–290, and references therein. (d) Dhoro, F.; Tius, M. A.
J. Am. Chem. Soc. 2005, 127, 12472. (e) delos Santos, D. B.; Banaag,
A. R.; Tius, M. A. Org. Lett. 2006, 8, 2579. (f) Banaag, A. R.; Tius,
M. A. J. Am. Chem. Soc. 2007, 129, 5328. (g) Basak, A. K.; Tius,
M. A. Org. Lett. 2008, 10, 4073. (h) Banaag, A. R.; Tius, M. A. J.
Org. Chem. 2008, 73, 8133. (i) Cao, P.; Sun, X.-L.; Zhu, B.-H.; Shen,
Q.; Xie, Z.; Tang, Y. Org. Lett. 2009, 11, 3048.
(12) The structure of 8 was confirmed by X-ray crystallography.
(13) The regiochemistry of the reactions was determined by analysis of
2D NMR spectra (COSY, HSQC, and HMBC) of the products.
(14) HF/6-31G//HF/6-311G(d, p) using Gaussian 03, Revision C.02.
(15) (a) Cramer, C. J.; Barrows, S. E. J. Phys. Org. Chem. 2000, 13, 176.
See also the earlier paper: (b) Cramer, C. J.; Barrows, S. E. J. Org.
Chem. 1998, 63, 5523.
9
1686 J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010

Formation of Carbon-Carbon Bonds
A R T I C L E S
Scheme 3. Conversion of 10 and 11a,b into 13
Table 2. Reactions of AVK 1 with Cyclic Dienes in CH2Cl2
at -78 °C
A concerted reaction would require the diene to assume an
s-cis conformation. s-Cis conformations would be disfavored
for the dienes in entries 2, 5, and 7, and these dienes gave no
[4 + 3] products, but provided [3 + 2] products, only, in
moderate to good yield. In contrast, the dienes in entries 1, 3,
4, and 6 have more accessible s-cis conformers, and these dienes
did afford [4 + 3] products. It was significant that 5, 6, and 9a
were the only [3 + 2] products in entries 1, 2, and 5, indicating
a high degree of stereoselectivity in this cyclization, also. This
might be expected for a cycloaddition via a geometry similar
to 14 (Scheme 2) or a very rapid stepwise reaction.16 The
product 9a in entry 5 was compared with its (undetected) epimer
9b computationally,14 and it was 9b that was lower in energy
by 2.2 kJ.mol-1, suggesting that 9a forms as the result of a
concerted reaction.
a Five equivalents of furan,
5 equiv of BF3 ·Et2O, 5
min. b In
All products were stable with BF3 ·Et2O at -78 °C, but with
BF3 ·Et2O at room temperature over a period of hours some
compounds equilibrated. One may presume that this much
slower process involved ring cleavage to give the enolate and
the delocalized cation, which then recyclized. The [3 + 2]
product 5 equilibrated to a 2.2:1 mixture of the [4 + 3]
compound 4a and the epimer 4b. On the other hand, compound
6 remained unchanged in the presence of BF3 ·Et2O. Resub-
jecting the mixture of 10 and 11a,b to BF3 ·Et2O provided 15
in 75% yield. This remarkable outcome indicates that each
compound underwent ring-opening, and then reclosure of the
enolate onto position c (Scheme 3). Compound 15 represents
the product of a formal [5 + 4] cyclization of 2 with the diene.
In every instance in Table 1 where the reaction products were
mixtures of different ring-systems, addition of more BF3 ·Et2O
to the mixture resulted in confluence to a single ring-system.
The number of examples of [4 + 3] cyclizations employing
cyclic dienes is large.1 Also, the cyclic cations of Nazarov
reactions have been reported to be trapped with cyclic dienes.5
When some cyclic dienes were mixed with AVK 1 and
BF3 ·Et2O, the Nazarov reactions were interrupted readily by
the cyclic dienes, but it was surprising that no products of [4 +
3] cyclization were detected. Only one carbon-carbon bond
formed (Table 2). Furan, which has been used in a number of
[4 + 3] cyclizations with oxyallyl cations,1g,5 trapped 2 only at
position a to provide 16 when the reaction was mediated by
BF3 ·Et2O or Cu(OTf)2 (entry 1). InCl3 elicited the formation
of a minor amount of 17 via capture at position c (entry 2).
Intractable material was obtained when cyclopentadiene or 6,6-
dimethylfulvene were added, but addition of 1,2,3,4,5-pentam-
ethylcyclopentadiene led to the formation of a minor product
18, by interception of 2 at position a, and a major product 19,
by capture of 2 at the less sterically hindered position c. Using
1,3-cyclohexadiene to trap the oxyallyl cation provided only
addition, there was 22% of a product where one furan added to two of 2
at position a. c Five equivalents of furan, 5 equiv of InCl3, 2 h. d Two
equivalents of diene, 1.1 equiv of BF3 ·Et2O, 5 min. e In addition, there
was 8% of the product of the Diels-Alder reaction of the diene with the
allene of 1.
Scheme 4. Formation of 20
20, as a single diastereomer. The relative stereochemistry of
20 was established by a J-based configuration analysis method.18
The position of the diene moiety in 20 is intriguing because
loss of a proton from a putative intermediate 21 should be
expected to lead to a more substituted diene. However, a [3 + 2]
cyclization involving the enol oxygen would give 22 as an
intermediate,19 and conjugate elimination/ring-opening during
workup would result in the preferential formation of 20 (Scheme
4). It has been suggested by the calculations of Cramer and
Barrows15 that a dipolar [3 + 2] cycloaddition can be comparable,
or even lower, in energy to all-carbon [3 +2] and [4 + 3]
cycloadditions, depending on the nature of the oxyallyl cation and
the diene involved. In the case of AVK 1 with the cyclic dienes,
it is plausible that the direct formation of [4 + 3] and [3 + 2]
products would suffer from considerable steric hindrance in
transition state geometries similar to 13 and 14, disfavoring these
pathways relative to a dipolar [3 + 2] cycloaddition generating
22.
(18) (a) Matsumori, N.; Kaneno, D.; Murata, M.; Nakamura, H.; Tachibana,
K. J. Org. Chem. 1999, 64, 866. Reviews: (b) Bifulco, G.; Dambruoso,
P.; Gomez-Paloma, L.; Riccio, R. Chem. ReV. 2007, 107, 3744. (c)
Kwan, E. E.; Huang, S. G. Eur. J. Org. Chem. 2008, 2671. NOE
measurements were in accord with the stereochemical assignment.
(19) A product of intramolecular cyclization involving an enol oxygen during
Nazarov cyclization has been reported: Bender, J. A.; Blize, A. E.;
Browder, C. C.; Giese, S.; West, F. G. J. Org. Chem. 1998, 63, 2430.
(16) The stepwise formation of [3 + 2] products might be considered a
5-(enolendo)-exo-trig closure, a disfavored process: Baldwin, J. E.;
Lusch, M. J. Tetrahedron 1982, 38, 2939.
(17) Similar observation with furan: Rieder, C. J.; Fradette, R. J.; West,
F. G. Chem. Commun. 2008, 1572.
9
J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1687

A R T I C L E S
Marx and Burnell
Table 3. Reactions of AVK 1 with Allylsilanes Mediated by Various
Lewis Acids
Table 4. Reactions of AVK 1 with Electron-Rich Alkenes (2 equiv)
Mediated by BF3 ·Et2O (1.1 equiv) in CH2Cl2 at -78 °C for 5 min
yields (%)
24a-c
entry
X
Lewis acid
time
23
25
1
2
3
4
5
6
Me
Me
Me
i-Pr
i-Pr
OEt
BF3 ·Et2Oa
5 min
1.5 h
2 h
5 min
2 h
27
20
22
19
27
-
54 (24a-c)
10 (24a)
-
57 (24b)
28 (24b)
15 (24c)
-
42
56
-
13
-
b
Cu(OTf)2
b
InCl3
a Ether group exo/endo 5:1. b Ca. 90% the isomer shown. c Included
an inseparable byproduct from subsequent reaction of 32.
BF3 ·Et2Oa
InCl3Ob
BF3 ·Et2Oa
5 min
Scheme 6. Attempted Interrupted Nazarov Reaction of AVK
1 with 33
a Two equivalents of allylsilane, 1.1 equiv of BF3 ·Et2O. b Five
equivalents of allylsilane, 5 equiv of Lewis acid.
Scheme 5. Treatment of the [3 + 2] Products 24a,b with BF3 ·Et2O
at Room Temperature
recyclization of 27b evident, but it led to the triisopropylsilyl
[3.2.1]-compound 26b in very good yield.
Oxygen-substituted alkenes intercepted 2 with high facial
selectivity by nucleophilic attack at position a or by the
formation of [3 + 2] products (Table 4). The silyl enol ether
and the enol acetate (entries 1 and 2) trapped the oxyallyl cation
inefficiently to give low yields of 28 and 29, respectively. The
enol ether, n-propoxyethene, gave two types of products, 28
and 30, in high overall yield (entry 3). However, the [3 + 2]
product 30, in contrast with 24a,b, simply decomposed when
treated with BF3 ·Et2O at room temperature. Further substitution
of the double bond as in entries 4 and 5 led to products of
trapping 2 at position a, and 31 and 32 were not accompanied
by [3 + 2] products. The [3 + 2] product 30 was obtained as an
epimeric mixture, but it is important to note that the formation of
31 was stereoselective. The stereochemistry was determined by
an NMR method.18 This stereochemistry might have been the
consequence of the formation of a transient intermediate, analogous
to 22, produced by a [3 + 2] process involving the enol oxygen.
In light of entries 1 and 4 in Table 4, it was somewhat
surprising that introduction of neither the tert-butyldimethylsilyl
enol ether derived from cyclopentanone nor a cyclic enol ether,
dihydropyran, gave any trapped Nazarov product. Also, the
enamine derived from cyclopentanone and pyrollidine and
N-vinylpyrollidone failed to trap 2 to any significant degree,
and intractable material was obtained. It is probable that if the
alkene or diene is too electron-rich, it will add via a Michael
reaction to the central allenic carbon of the AVK before the AVK
has undergone the Nazarov reaction. Evidence for this comes from
an attempted interception of 2 with an excess of the electron-rich
cyclobutene derivative 33 (Scheme 6). The 1H NMR spectrum of
the initial product mixture showed signals mainly for the Michael
product 34 and residual 33. Although the isolated yield of 34
following flash chromatography was low, it was subsequently
established that 34 decomposed rapidly on silica gel.
When BF3 ·Et2O was added to 1 and allyltrimethylsilane, two
Nazarov products formed very rapidly. One (23) was the product
of addition to position a in 2, but the major product was the [3
+ 2] product 24a (Table 3).13 The relative stereochemistry was
determined by measurement of NOE’s. The effect of different
Lewis acids and different silanes was briefly examined. When
Cu(OTf)2 was employed as the Lewis acid, the reaction was
slower and three products were obtained: 23, 24a, and 25, of
which 25 was the result of capture of 2 at position c. With InCl3,
the reaction had approximately the same rate as with Cu(OTf)2,
25 was the major product, and there was none of the [3 + 2]
product. Utilization of a more robust silane, allyltriisopropyl-
silane, resulted in an increased proportion of the [3 + 2] product
24b in the presence of either BF3 ·Et2O or InCl3. Allyltriethox-
ysilane with BF3 ·Et2O gave a poor yield of the [3 + 2] product
25c, but no other Nazarov product was detected.
The trimethylsilyl [2.2.1]-compound 24a could be converted
in 5 h to 23 in moderate yield in the presence of BF3 ·Et2O at
room temperature, but a minor product was the trimethylsilyl
[3.2.1]-compound 26a (Scheme 5). This minor compound would
have been the result of cyclization of intermediate 27a in concert
with a shift of the silyl group. This sort of process had been
anticipated, but not observed, by West6 in their study of
interception of Nazarov reactions with allylsilanes. When the
triisopropylsilyl [2.2.1]-compound 24b was stirred with
BF3 ·Et2O at room temperature, not only was the analogous
9
1688 J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010

Formation of Carbon-Carbon Bonds
A R T I C L E S
Table 5. Reactions of AVK 1 with Styrenes (2 equiv) Mediated by
BF3 ·Et2O (1.1 equiv) in CH2Cl2 at -78 °C for 5 min
Scheme 7. Reactions of AVK’s 39 and 40 Giving [4 + 3] and
[3 + 2] Products
beginning with Lewis acid-mediated Narazov cyclization and
then interception of the intermediate, the oxyallyl cation 2, by
various acyclic dienes, cyclic dienes, electron-rich alkenes, or
styrenes by the formation of an additional ring by a [4 + 3]
and/or [3 + 2] cyclization or by the formation of one additional
carbon-carbon bond. The bicyclic compounds generated in this
way are densely substituted, and would be difficult to access as
succinctly in other ways. The products of these interrupted
Nazarov reactions generally reflect excellent regio- and stereo-
selectivity in the trapping reaction. In some instances, equilibrat-
ing conditions were shown to enhance the proportion of one
product at the expense of another or to provide a different carbon
skeleton. Preliminary studies show that this process is not only
limited to aromatic derivatives, but is also amenable to unsub-
stituted or alkyl-substituted allenyl vinyl ketones. These results
invite exploitation of AVK’s for the synthesis of more complex
carbocyclic compounds.
Styrenes afforded only [3 + 2] products as single diastere-
omers, for which the relative stereochemistry was confirmed
by NOE measurements (Table 5).13 Styrene itself (entry 1) gave
a poor yield of 35, and an electron-poor derivative (entry 2)
did not trap the oxyallyl cation at all. On the other hand,
electron-donation was clearly advantageous (entries 3-5).
Reaction of 1 with the trans-isomer provided 38 in a yield that
was slightly higher (entry 5) than the yield from the less
substituted styrene (entry 3), but the corresponding cis-isomer
failed to trap 2 (entry 6). This curious result may reflect reduced
electron-donation by the methoxyphenyl group into the alkene
moiety in the cis isomer,20 and it is also consistent with the
hypothesis that the kinetic route to the [3 + 2] products must
be concerted, or at least very rapid, at - 78 °C because bond
rotation in an intermediate carbocation could have given 38 from
the cis isomer.
Finally, two reactions previously observed with AVK 1 were
reassessed with an unsubstituted AVK 39 and the isopropyl-
substituted AVK 40 (Scheme 7).13 Once again, in Nazarov
reactions mediated by BF3 ·Et2O, 2,3-dimethylbutadiene inter-
cepted the intermediate oxyallyl cations to provide the [4 + 3]
products 41 and 42, exclusively. Also, addition of 3,4-
dimethoxystyrene led rapidly to the formation of only the [3 +
2] products 43 and 44. In every instance, only one diastereomer
was produced.
Acknowledgment. This work was supported by the Natural
Sciences and Engineering Research Council of Canada and the
Killam Trusts. We thank Dr. T. S. Cameron for the X-ray structure
of 8, Dr. M. Lumsden for assistance with nuclear magnetic
resonance experiments, Mr. Xiao Feng for mass spectrometry, and
Mr. Gavin Heverly-Coulson for the computational data. Also, we
are grateful to Mr. Craig T. M. Stamp and Mr. Rylan J. Lundgren
for helpful discussions.
Conclusions
Supporting Information Available: Experimental details and
characterization data for all new compounds, including the X-ray
crystal structure of 8 and J-HMBC data for compounds 20 and
31. This material is available free of charge via the Internet at
http://pubs.acs.org.
In summary, it has been shown that an AVK is a versatile
source of cyclic molecules via a cascade reaction sequence
(20) (a) Overberger, C. G.; Tanner, D.; Pearce, E. M. J. Am. Chem. Soc.
1958, 4566. (b) Schmid, G. H.; Garratt, D. G. Can. J. Chem. 1974,
52, 1807. (c) Mayr, H.; Fo¨rner, W.; von Rague´ Schleyer, P. J. Am.
Chem. Soc. 1979, 101, 6032.
JA909073R
9
J. AM. CHEM. SOC. VOL. 132, NO. 5, 2010 1689