Substituent-controlled reactivity in the Nazarov cyclisation of allenyl vinyl ketones

Article abstract of DOI:10.1002/chem.201100519

Alkyl substitution α to the ketone of an allenyl vinyl ketone enhances Nazarov reactivity by inhibiting alternative pathways involving the allene moiety and through electron donation and/or steric hindrance. This substitution pattern also accelerates Naza

Full text of DOI:10.1002/chem.201100519

DOI: 10.1002/chem.201100519
Substituent-Controlled Reactivity in the Nazarov Cyclisation of Allenyl Vinyl
Ketones
Vanessa M. Marx, Rhonda L. Stoddard, Gavin S. Heverly-Coulson, and
D. Jean Burnell*[a]
Abstract: Alkyl substitution a to the
ketone of an allenyl vinyl ketone en-
hances Nazarov reactivity by inhibiting
alternative pathways involving the
allene moiety and through electron
donation and/or steric hindrance. This
substitution pattern also accelerates
Nazarov cyclisation by increasing the
population of the reactive conformer
and by stabilising the oxyallyl cation
intermediate. Furthermore, a substitu-
tion by an alkyl group does not alter
the regioselectivity of interrupted Naz-
arov reactions when the oxyallyl cation
intermediate is intercepted by addition
of an oxygen nucleophile, or by [4+3]
cyclisation with acyclic dienes. The re-
gioselectivity of the Nazarov process
for allenyl vinyl ketones was deter-
mined to be a result of an electronic
bias in the oxyallyl cation intermediate.
Computational data are consistent with
this observation.
Keywords: allenes · carbocations ·
cyclic compounds · cycloaddition ·
Nazarov reaction
Introduction
The Nazarov reaction[1] of dialkenyl ketones and its more
recently investigated variant, the “interrupted” Nazarov re-
action,[2] are powerful tools for the construction of five-
membered rings. Although strong acids may be required to
induce cyclisation for divinyl ketone substrates, the reaction
proceeds more rapidly when the substrate is activated by
substitution a to the carbonyl group.[3] An alternative means
of activation involves an allene used in the place of one of
the alkene units.[4,5] Allenyl ethers cyclise spontaneously fol-
lowing their generation from propargyl precursors,[4] but al-
lenyl vinyl ketones (AVKs) of general structure 1 are easily
isolable and can be stored for prolonged periods. However,
in the presence of acid, compound 1 undergoes rapid Naza-
rov cyclisation (Scheme 1), and the intermediate oxyallyl
cation 2 has a propensity to be intercepted by nucleophilic
species rather than to follow the traditional elimination
pathway.[6] This interrupted Nazarov reaction has been ex-
ploited to provide a variety of products, such as 3, 4 and 5,
by capture at positions a, c, or a and b of 2 by heteroatom-
or carbon-based nucleophiles.
Scheme 1. Nazarov cyclisation of AVK 1 and representative products
formed by interception of oxyallyl cation 2.
1 represents a single pattern of substitution. The substituent
on the alkene terminus of the AVK (R in 1) has been
varied, and the course of the Nazarov reaction did not
change significantly. However, it was not known if the regio-
chemistry of the trapping process (e.g., in the formation of 3
and 5) might be explained simply by steric hindrance im-
parted by the methyl group at position b in 2 or if there is
an electronic bias in oxyallyl cation 2. Before AVKs can be
exploited fully for the synthesis of complex structures pos-
sessing five-membered rings, it is crucial that the implica-
tions of substitution on the AVK, especially a to the carbon-
yl group, be understood in terms of reactivity and regiose-
lectivity in the trapping process. The results of a computa-
tional and experimental study addressing this important
issue are presented herein.
Although previous studies have demonstrated that some
AVKs with general structure 1 are valuable substrates for a
diverse range of interrupted Nazarov cyclisations, structure
[a] Dr. V. M. Marx, R. L. Stoddard, G. S. Heverly-Coulson,
Prof. D. J. Burnell
Department of Chemistry, Dalhousie University
Nova Scotia, B3H 4R2 (Canada)
Fax : (+1) 902-494-1310
E-mail: jean.burnell@dal.ca
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100519.
8098
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8098 – 8104

FULL PAPER
Figure 1. a) Selected atomic charges and b) delocalisation indices for oxyallyl cations 6–8, in their lowest-energy conformations, as calculated by QTAIM
methods (RI-MP2/cc-pvdz).[9]
Results and Discussion
important differences were revealed in the delocalisation in-
dices (Figure 1b),[8] which are estimates of bond population
by pairs of electrons. Regardless of the presence or position
of the methyl group, the bond from the oxygen-bearing
An increased amount of positive charge borne largely at po-
sition a in 6 (Figure 1), rather than at positions b or c, might
play a significant role in the high regioselectivity observed
for position a in the interrupted Nazarov reactions of AVKs
of general structure 1. Thus, a computational examination of
oxyallyl cation 6, derived from Nazarov cyclisation of AVK
9 with BF3 as the acid, was carried out with the intent of lo-
À
carbon to b and the carbon carbon bond to c had larger in-
dices than the bond from the oxygen-bearing carbon to a,
which points to position a as the most electrophilic carbon
in all three oxyallyl cations. Thus, if steric differences are
negligible, AVKs 9–11 should exhibit similar regioselectivi-
ties in interrupted Nazarov processes.
AVKs 10–12 were synthesised to compare experimentally
with AVK 9 (Scheme 2). The zinc-mediated Barbier-type
coupling of propargyl bromide with the corresponding a,b-
unsaturated aldehydes[10] was more satisfactory than a proce-
dure that uses more costly indium.[6a] The use of indium in
cating any significant electronic bias in the oxyallyl cation.
In addition, to specifically probe the electronic role of sub-
stitution a to the carbonyl group, oxyallyl cation 7, bearing
no a-substituent, and oxyallyl cation 8, derived from a com-
pound with an a-methyl substituent on the alkenyl side of
the carbonyl group, were also examined computationally
(Figure 1). Species 7 and 8 would be derived from the BF3-
mediated Nazarov cyclisations of AVKs 10 and 11, respec-
tively.
By using the quantum theory of atoms in molecules
(QTAIM),[7] the atomic charges on all of the atoms in 6, 7
and 8 were determined computationally (Figure 1a). There
were no significant differences between 6, 7 and 8 in the
lengths of the bonds between adjacent sp2 carbon atoms, but
Scheme 2. Preparation of AVKs 10–12.
Chem. Eur. J. 2011, 17, 8098 – 8104
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8099

J. Burnell et al.
lieu of zinc gave a 1:1 mixture of the propargyl alcohol and
the allenyl alcohol and attempts to prepare the AVKs by ox-
idation of the allenyl alcohols were unsuccessful with a vari-
ety of oxidants, including (COCl)2/Et3N, iodoxybenzoic acid,
BaMnO4, MnO2, the Dess–Martin periodinane (DMP) and
a DMP/NaHCO3 combination.[11] Oxidation of propargyl al-
cohols 13–15, however, was accomplished successfully by
using DMP/NaHCO3. Other reagents, such as CrO3/H2SO4,
(COCl)2/Et3N and even DMP alone, resulted in decomposed
material, and BaMnO4 and MnO2 did not react with the
starting material. Oxidation of 13 was accompanied through
isomerisation of 16 to give AVK 10 directly, whereas AVKs
11 and 12 required a separate base-mediated isomerisation
step with 17 and 18 as the intermediate productsYES. Al-
though both K2CO3 and triethylamine were equally effective
for the base-induced isomerisation of 17 to 11, Na2CO3 and
NaHCO3 failed to produce 11 in acceptable yield.
Scheme 3. Reactions of AVKs 10–12 with trifluoroacetic acid.
AVK 9 was amenable to purification by chromatography
over silica gel.[6a] AVK 10 also proved to be stable with silica
gel, but AVKs 11 and 12 were much less so. In the presence
of silica gel, 11 and 12 produced some of the corresponding
Nazarov elimination products 19 (41%) and 20 (14%). This
observation parallels the reactivity observed by Hashmi et
al.[5a] for AVKs with this substitution pattern, which also un-
derwent rapid Nazarov cyclisation promoted by the weakly
acidic silica gel. Purification of AVKs 11 and 12 was also at-
tempted by chromatography using basic alumina; however,
this resulted in decomposition. Thus, the two-step yields of
11 and 12 were determined without purification, by
1H NMR spectroscopy by using 1,3,5-trimethoxybenzene as
an internal standard. It is important to note that AVKs 11
and 12 decomposed rapidly at room temperature, either
neat or in CH2Cl2. These compounds could only be stored
for less than 24 h in CH2Cl2 (0.1m) at À208C.
Figure 2. X-ray crystal structure of dinitrobenzoate 23a.
Although AVK 9 yielded [3+2] cyclisation products, such
as 5, if Nazarov cyclisation was induced with BF3·OEt2 in
the presence of electron-rich alkenes,[6c] similar products
were not seen with 10–12 and NMR analysis of the product
mixtures suggested that the bulk of the material reacted di-
rectly with the allene moiety of the AVK. Cation 6, derived
from AVK 9, also underwent [4+3] cyclisations[13] with sub-
Treatment of 11 and 12 with BF3·OEt2 at À788C for
10 min led to complex mixtures, although 23% of Nazarov
product 20 was isolated following the reaction of the latter.
On the other hand, compound 10 was much more tolerant
of the acid and after one hour Nazarov product 21 was ob-
tained in 64% yield.
stituted butadiene derivatives to give bicycloACTHUNRTGNEUNG[4.2.1]nonenone
derivatives 24–26 (Scheme 4).
AVK 10 reacted with 2,3-dimethylbutadiene, but the
major product was Diels–Alder adduct 27 (Scheme 5). In
contrast, AVK 11 underwent rapid Nazarov cyclisation fol-
lowed by [4+3] cyclisation in the presence of 2,3-dimethyl-
AVK 10 reacted cleanly in the presence of TFA, but the
product was compound 22, the result of conjugate addition
to the allene (Scheme 3). AVK 11 yielded Nazarov elimina-
tion product 19 in 48% yield, although AVK 12 gave mainly
23, which represents an intercepted Nazarov product in
which the oxygen nucleophile attacked position a. The rela-
tive stereochemistry of 23 was confirmed by X-ray crystallo-
graphic analysis of the 3,5-dinitrobenzoate of 23
(Figure 2).[12]
butadiene to afford bicycloACHTNUGRTNEUNG
[4.2.1]nonenone derivative 28[14]
in good yield (Scheme 6). AVK 12 reacted similarly to pro-
vide 31. Both 11 and 12 gave regioisomeric [4+3] cyclisation
products (29a and b, and 32a and b) upon reaction in the
presence of isoprene.[15] Nazarov reactions in the presence
of trans-piperylene gave 30 and 33 and the regioselectivity
of these [4+3] cyclisations was the same as was seen with
8100
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8098 – 8104

Nazarov Cyclisation of Allenyl Vinyl Ketones
FULL PAPER
Scheme 4. ACHTUNGTRENNUNG
[4+3] Cyclisations of AVK 9.[6c]
Scheme 5. ACHTUNGTRENNUNG[4+2] Cyclisation of AVK 10 with 2,3-dimethylbutadiene.
AVK 9, that is, 26.[6c,14,15] These provide experimental cor-
roboration of the computational result that the electronic
bias in the oxyallyl cations derived from AVKs 9 and 11
(i.e., 6 and 8) is the same, regardless of the position of the
alkyl substituent. That 26, 30 and 33 were obtained as single
stereoisomers is consistent with the [4+3] cyclisation being
through a concerted, though likely asynchronous, reaction
with an extended geometry.
The computational and experimental data presented
herein concerning substituent-controlled reactivity in the
Nazarov cyclisation of allenyl vinyl ketones are in accord
with the following rationalisation. The most inherently elec-
trophilic site on the oxyallyl cation intermediate is position
a (the a-alkenyl position) and alkyl substitution on position
a does not change this. However, substitution by an alkyl
group can significantly change the outcome of Nazarov cyc-
lisations. Although AVK 10 is capable of cyclisation, as
shown by the formation of 21 in the presence of BF3·OEt2,
attempts to elicit an interrupted Nazarov process are pre-
dicted to be pre-empted by the more rapid acid-mediated
reaction of the allene, as in the formation of 22 and 27. It is
proposed that an alkyl group on the allene (as in 9) attenu-
ates the reactivity at the central carbon of the allene, by
electron donation and/or by steric hindrance, allowing the
Nazarov pathway to dominate. Furthermore, it is hypothes-
ised that an alkyl group on the alkene or the allene acceler-
Scheme 6. Nazarov reactions with subsequent [4+3] cyclisations of AVKs
11 and 12.
ates the Nazarov reaction. It can be seen in Figure 1a that
the methyl groups in oxyallyl cations 7 and 8 bear a signifi-
cant atomic charge that should facilitate the formation of
the oxyallyl cation intermediate. In addition, calculations re-
vealed that the alkyl group on the alkene in 11 exerts a sig-
nificant influence on the conformation, and thus the reactiv-
ity, of the AVK (Scheme 7). An s-trans–s-trans arrangement
must resemble the transition state geometry for the Nazarov
cyclisation. An s-trans conformation of the carbonyl and the
allene must be favoured because this feature is seen in the
lowest energy conformers of 9, 10 and 11. However, without
a methyl group on the alkene, the s-cis conformation is pre-
ferred for the carbonyl–alkene moiety leading to s-cis–s-
Chem. Eur. J. 2011, 17, 8098 – 8104
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8101

J. Burnell et al.
rupted Nazarov process is dependent on steric hindrance
about position a, although the electronic bias in favour of
position a in the intermediate oxyallyl cation is responsible
for the regioselectivity. These findings will be invaluable in
the design of feasible synthetic approaches involving Naza-
rov cyclisations of AVKs.
Experimental Section
General considerations: Dichloromethane was distilled from calcium hy-
dride. Ethyl acetate and hexanes were distilled. Flash column chromatog-
raphy used 230–400 mesh silica gel. Melting points were acquired by
using a Fisher–Johns apparatus and are uncorrected. IR spectra (FT)
1
were recorded with an FT instrument by using CsI plates. H NMR spec-
tra were acquired at 500 MHz and 13C NMR spectra at 125 MHz as solu-
tions in CDCl3. All HRMS were collected through positive-ion electro-
spray ionisation (ESI). For computational methods see the Supporting In-
formation.
Scheme 7. Conformations of AVKs calculated at RI-MP2/cc-pvdz.[9]
trans conformations for AVKs 9 and 10 being lower in
energy than the s-trans–s-trans conformation by more than
12 kJmolÀ1. In contrast, the s-trans–s-trans conformation is
the lowest-energy conformation for AVK 11 by 5.5 kJmolÀ1
over the s-cis–s-trans conformation because the methyl
group on the alkene induces an unfavourable through-space
interaction with the allene when the alkene–carbonyl
moiety is in an s-cis conformation. Enhancement of Nazarov
reactivity though the influence of an a substituent on con-
formation has previously been observed for divinyl keto-
nes.[1c]
Interrupted Nazarov processes, however, are expected to
be less efficient with 11 and 12 than with 9 because the
more substituted position a offers more steric hindrance.
This is shown by comparison of the yields of trapped prod-
ucts from AVK 9, which are nearly quantitative in the reac-
tion with TFA and with BF3·OEt2 and 2,3-dimethylbuta-
diene, with the yields of the analogous processes with AVKs
11 and 12. This steric hindrance also explains why AVK 11
gave conventional, eliminated Nazarov product 19 with
TFA, whereas AVK 9 yielded only the interrupted Nazarov
product.[6a]
(E)-2-Methyl-1-phenylhex-1-en-5-yn-3-ol (14): Propargyl bromide (12 g,
0.10 mol) was combined with a-methyl-trans-cinnamaldehyde (10 mL,
0.08 mol) in a mixture of a saturated aqueous solution of ammonium
chloride and methanol (5:1, 200 mL) and stirred in an ice bath. Zinc dust
(6.7 g, 0.10 mol) was added in four portions, over 20 min. Additional zinc
dust (10.3 g, 0.16 mol) was then added in four portions, over 10 min, and
the mixture was stirred vigorously at RT for 1 h. The mixture was filtered
through Celite, and the aqueous layer was extracted thoroughly with di-
ethyl ether. The combined organic layers were washed with brine, dried
over MgSO4 and concentrated under vacuum. Flash column chromatog-
raphy of the residue (10% diethyl ether in pentane) yielded 14 (5.5 g,
37%) as a colourless oil. 1H NMR: d=7.34 (2H, m), 7.23 (2H, m), 7.11
(1H, m), 6.60 (1H, m), 4.37 (1H, m), 2.58 (2H, m), 2.19 (1H, d, J=
3.7 Hz), 2.09 (1H, t, J=2.7 Hz), 1.89 ppm (3H, d, J=1.4 Hz); 13C NMR:
d=138.4, 137.2, 129.0, 128.1, 126.6, 126.5, 80.6, 75.5, 70.9, 26.1, 13.5 ppm;
IR (film): n˜ =3380, 3297, 2128, 1605 cmÀ1; HRMS (ESI): m/z calcd for
C13H14ONa+: 209.0937; found: 209.0941. NMR data match the litera-
ture.[16, 17]
(E)-2-Methyl-1-phenylhex-1-en-5-yn-3-one (17) and (E)-2-methyl-1-phe-
nylhexa-1,4,5-trien-3-one (11): DMP (1.5 g, 3.5 mmol) was added to a vig-
orously stirring solution of alcohol 14 (0.5 g, 2.7 mmol) and sodium bicar-
bonate (2.3 g, 27 mmol) in CH2Cl2 (25 mL) at RT. After 45 min, saturated
aqueous solutions of sodium bicarbonate (25 mL) and sodium thiosul-
phate (25 mL) were added, as well as a small amount of diethyl ether,
and the mixture was stirred until both layers went clear. The mixture was
extracted thoroughly with CH2Cl2 and the organic layer was washed with
brine, dried over MgSO4 and concentrated under vacuum, yielding crude
17 as a pale yellow oil. 1H NMR: d=7.58 (1H, q, J=1.4 Hz), 7.43 (4H,
m), 7.37 (1H, m), 3.74 (2H, d, J=2.8 Hz), 2.29 (1H, t, J=2.8 Hz),
2.10 ppm (3H, d, J=1.4 Hz); 13C NMR: d=194.8, 140.5, 135.8, 135.4,
Conclusion
129.8, 128.8, 128.5, 77.3, 72.9, 29.7, 13.3 ppm; IR (film): n˜ =1641 cmÀ1
;
HRMS (ESI): m/z calcd for C13H12ONa+: 207.0780; found: 207.0781.
Without further purification, propargyl ketone 17 from the initial step
was redissolved in CH2Cl2 (25 mL) and potassium carbonate (0.41 g,
3.0 mmol) was added. After stirring at RT for 20 min, the solution was
filtered and concentrated under vacuum, yielding 11 (0.18 g, 36%) as a
The presence of an alkyl group a to the ketone on either
the alkene or the allene of an AVK has a significant impact
on the efficiency of Nazarov and interrupted Nazarov reac-
tions. AVKs lacking an a substituent can undergo a conven-
tional Nazarov cyclisation, but the interrupted variant of the
reaction is not feasible. Interrupted Nazarov reactions can
be expected with an AVK bearing an a substituent. An
alkyl substituent on the allene is beneficial as it inhibits al-
ternative allene reactions and may accelerate the Nazarov
reaction. An alkyl substituent on the alkene accelerates the
Nazarov reaction, but interception of the intermediate oxy-
allyl cation by a nucleophile is retarded by steric hindrance.
Overall, with alkyl substitution, the efficiency of the inter-
1
red oil. H NMR: d=7.50 (1H, m), 7.41 (5H, m), 6.34 (1H, t, J=6.6 Hz),
5.21 (2H, d, J=6.6 Hz), 2.12 ppm (3H, d, J=1.9 Hz); 13C NMR: d=
215.8, 193.0, 139.3, 137.2, 135.7, 133.2, 129.7, 128.4, 92.5, 78.5, 13.9 ppm;
IR (film): n˜ =1967, 1943, 1650, 1601 cmÀ1; HRMS (ESI): m/z calcd for
C13H13ONa+: 185.0961; found: 185.0968. Note: the yield of AVK 11 in
the crude reaction mixture was determined by 1H NMR spectroscopy by
using 1,3,5-trimethoxybenzene as an internal standard and was used with-
out further purification.
2-Methyl-4-methylene-3-phenylcyclopent-2-enone (19): SiO2 (0.12 g,
2.7 mmol) was added to a solution of AVK 11 (0.10 g, 0.54 mmol) in
CH2Cl2 (5 mL) at RT and stirred overnight. The mixture was filtered and
8102
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8098 – 8104

Nazarov Cyclisation of Allenyl Vinyl Ketones
FULL PAPER
concentrated under vacuum and flash column chromatography (10–20%
ethyl acetate in hexanes) of the residue yielded 19 (0.041 g, 41%) as a
yellow oil. 1H NMR: d=7.47 (3H, m), 7.34 (2H, m), 5.29 (1H, s), 5.22
(1H, s), 3.19 (2H, s), 1.88 ppm (3H, s); 13C NMR: d=205.2, 164.5, 143.7,
141.1, 133.2, 128.8, 128.5 (4C), 110.9, 39.7, 9.4 ppm; IR (film): n˜ =1703,
m), 3.57 (1H, t, J=5.7 Hz), 2.74 (1H, d, J=19 Hz), 2.69 (1H, d, J=
19 Hz), 2.63 (1H, brd, J=19 Hz), 1.65 (1H, brd, J=19 Hz), 1.68 (3H, s),
1.62 ppm (3H, s); 13C NMR: d=199.4, 144.8, 142.6, 134.9, 130.6, 129.1
(2C), 128.6 (2C), 124.8, 124.3, 124.1, 110.9, 54.2, 39.6, 34.4, 19.0,
18.9 ppm; IR (film): n˜ =1695, 1614 cmÀ1; HRMS (ESI): m/z calcd for
C18H20ONa+: 275.1406; found: 275.1403.
1612 cmÀ1
185.0962.
;
HRMS (ESI): m/z calcd for C13H13O+: 185.0961; found:
(1R*,6S*,8S*)-1,3,4-Trimethyl-7-methylene-8-phenylbicycloACTHNUGRTNEUNG[4.2.1]non-3-
en-9-one (28): BF3·OEt2 (1.1 equiv) was added to a solution of AVK 11
(0.10 g, 0.54 mmol) and 2,3-dimethylbutadiene (0.30 mL, 2.7 mol) in
CH2Cl2 (55 mL) at À788C. The solution was stirred for 10 min and
4-Methylene-3-phenylcyclopent-2-enone (21): BF3·OEt2 (0.070 mL,
0.60 mmol) was added to a solution of AVK 10 (0.070 g, 0.41 mmol) in
CH2Cl2 (40 mL) at À788C. The solution was stirred for 1 h and then
poured into
a separatory funnel containing a saturated solution of
poured into
a separatory funnel containing a saturated solution of
sodium bicarbonate. The aqueous layer was extracted thoroughly with
CH2Cl2. The combined organic solutions were dried over Na2SO4 and
concentrated under vacuum. Flash column chromatography (2.5%
EtOAc in hexanes) of the residue yielded 28 (0.11 g, 79%) as a colour-
less oil. 1H NMR: d=7.22 (2H, m), 7.17 (1H, m), 6.92 (2H, m), 5.11
(1H, t, J=1.9 Hz), 4.87 (1H, t, J=1.9 Hz), 3.63 (1H, m), 3.24 (1H, m),
2.54 (1H, brd, J=16 Hz), 2.42 (1H, dd, J=16, 7.7 Hz), 2.25 (1H, d, J=
16 Hz), 2.18 (1H, d, J=16 Hz), 1.81 (3H, s), 1.72 (3H, s), 0.68 ppm (3H,
s); 13C NMR: d=222.9, 152.1, 144.5, 129.1, 128.6, 128.1, 126.7, 126.5,
111.1, 57.9, 54.9, 53.4, 50.6, 41.7, 24.6, 23.9, 20.3 ppm; IR (film): n˜ =1743,
1597 cmÀ1; HRMS (ESI): m/z calcd for C19H22ONa+: 289.1563; found:
289.1556.
sodium bicarbonate. The organic layer was removed and the aqueous
layer was extracted with additional CH2Cl2 (ꢁ2). The combined organic
layers were dried over Na2SO4 and concentrated under vacuum. Flash
column chromatography (10–20% EtOAc in hexanes) of the residue pro-
vided 21 (0.045 g, 64%) as a yellow oil. 1H NMR: d=7.47 (5H, m), 6.37
(1H, m), 5.51 (1H, m), 5.43 (1H, m), 3.22 ppm (2H, m); 13C NMR: d=
204.5, 170.6, 143.6, 133.4, 133.3, 130.1, 128.9 (2C), 128.4 (2C), 113.8,
41.5 ppm; IR (film): n˜ =1721, 1695, 1610 cmÀ1 (w); HRMS (ESI): m/z
calcd for C12H10ONa+: 193.0624; found: 193.0628. 1H NMR data match
the literature.[18]
(3aR*,7aS*)-3a,4,5,6,7,7a-Hexahydro-7a-hydroxy-3-methyl-1H-inden-1-
one (23): TFA (0.20 mL, 2.6 mmol) was added to a solution of AVK 12
(0.080 g, 0.54 mmol) in CH2Cl2 (55 mL) at À788C and this mixture was
stirred for 30 min. The mixture was poured into a separatory funnel con-
taining a saturated solution of sodium bicarbonate, the organic layer was
removed and the aqueous layer was extracted with additional CH2Cl2 (ꢁ
2). The organic layers were combined, dried over Na2SO4 and concentrat-
ed under vacuum. The crude product was then subjected to column chro-
matography (Al2O3, basic, activated, 20% EtOAc in hexanes then
MeOH), which provided 20 (0.009 g, 10%) and 23 (0.052 g, 58%) as
yellow oils. For 20: 1H NMR: d=5.24 (1H, s), 5.11 (1H, s), 2.98 (2H, s),
2.45 (2H, m), 2.26 (2H, m), 1.76 (2H, m), 1.70 ppm (2H, m); 13C NMR:
d=204.9, 165.7, 143.9, 143.1, 107.1, 40.1, 23.1, 22.0, 21.8, 20.6 ppm; IR
(film): n˜ =1704, 1641 cmÀ1; HRMS (ESI): m/z calcd for C10H12ONa+:
171.0780; found: 171.0785. For 23: 1H NMR: d=5.94 (1H, m), 2.76 (1H,
m), 2.64 (1H, brs), 2.10 (3H, t, J=1.3 Hz), 1.89 (1H, m), 1.72 (1H, m),
1.61 (4H, m), 1.46 (1H, m), 1.27 ppm (1H, m); 13C NMR: d=210.9,
179.1, 126.9, 78.1, 50.3, 32.3, 22.9, 20.7, 18.8, 17.7 ppm; IR (film): n˜ =
Acknowledgements
The Natural Sciences and Engineering Research Council of Canada, the
Killam Trusts and Dalhousie University are thanked for providing the fi-
nancial support for this work. We are grateful to Dr. T. Stanley Cameron
for the X-ray crystal structure of 23a, Dr. M. Lumsden for assistance
with nuclear magnetic resonance experiments and Mr. Xiao Feng for the
collection of mass spectrometry data.
[1] a) K. L. Habermas, S. E. Denmark, T. K. Jones, Organic Reactions,
Vol. 45 (Ed.: L. A. Paquette), Wiley, New York, 1994, pp. 1–158;
b) H. Pellissier, Tetrahedron 2005, 61, 6479–6517; c) A. J. Frontier,
C. Collison, Tetrahedron 2005, 61, 7577–7606; d) M. A. Tius, Eur. J.
Org. Chem. 2005, 2193–2206; e) W. Nakanishi, F. G. West, Curr.
Opin. Drug Discovery Dev. 2009, 12, 732–751.
[2] For a review on the interrupted Nazarov reaction, see: T. N. Grant,
C. J. Rieder, F. G. West, Chem. Commun. 2009, 5676–5688.
[3] a) W. He, X. Sun, A. J. Frontier, J. Am. Chem. Soc. 2003, 125,
14278–14279; b) W. He, I. R. Herrick, T. A. Atesin, P. A. Caruana,
C. A. Kellenberger, A. J. Frontier, J. Am. Chem. Soc. 2008, 130,
1003–1011.
[4] a) M. A. Tius, Acc. Chem. Res. 2003, 36, 284–290, and references
therein; for recent examples, see: b) N. Shimada, B. O. Ashburn,
A. K. Basak, W. F. Bow, D. A. Vicic, M. A. Tius, Chem. Commun.
2010, 46, 3774–3775; c) W. F. Bow, A. K. Basak, A. Jolit, D. A.
Vicic, M. A. Tius, Org. Lett. 2010, 12, 440–443; d) A. K. Basak, N.
Shimada, W. F. Bow, D. A. Vicic, M. A. Tius, J. Am. Chem. Soc.
2010, 132, 8266–8267.
[5] a) A. Stephen, K. Hashmi, J. W. Bats, J. H. Choi, L. Schwarz, Tetra-
hedron Lett. 1998, 39, 7491–7494; b) P. Cao, X.-L. Sun, B.-H. Zhu,
Q. Shen, Z. Xie, Y. Tang, Org. Lett. 2009, 11, 3048–3051.
[6] a) V. M. Marx, D. J. Burnell, Org. Lett. 2009, 11, 1229–1231;
b) V. M. Marx, T. S. Cameron, D. J. Burnell, Tetrahedron Lett. 2009,
50, 7213–7216; c) V. M. Marx, D. J. Burnell, J. Am. Chem. Soc.
2010, 132, 1685–1689; d) V. M. Marx, F. M. Le Fort, D. J. Burnell,
Adv. Synth. Catal. 2011, 353, 64–68.
[7] a) R. F. W. Bader, Atoms in Molecules: A Quantum Theory, Oxford
University Press, Oxford, 1990; b) C. F. Matta, R. J. Boyd in The
Quantum Theory of Atoms in Molecules (Eds.: C. F. Matta, R. J.
Boyd), Wiley-VCH, Weinheim, 2007, pp. 1–34; c) J. Hernꢂndez-Tru-
jillo, F. Cortes-Guzman, G. Cuevas in The Quantum Theory of
3437, 1695, 1610 cmÀ1
;
HRMS (ESI): m/z calcd for C10H14O2Na+:
189.0886; found: 189.0884.
(3aR*,7aS*)-3a,4,5,6,7,7a-Hexahydro-3-methyl-1-oxo -1H-inden-7a-yl-
3,5-dinitrobenzoate (23a): Triethylamine (1 mL, 7.2 mmol) and then 3,5-
dinitrobenzoylchloride (0.48, 2.1 mmol) were added to a solution of 23
(0.17 g, 1.0 mmol) and 4-(dimethylamino)pyridine (0.13 g, 1.1 mmol) in
CH2Cl2 (25 mL) at RT. The solution was stirred for 18 h, diluted with
ethyl acetate and washed successively with aqueous solutions of HCl
(1m), NaOH (10%), saturated NaHCO3 and brine. The solution was
dried over Na2SO4 and concentrated under vacuum. Flash column chro-
matography (10–25% EtOAc in hexanes) of the residue provided 23a
(0.24 g, 67%) as
a peach-coloured solid. M.p. 176–1798C (CH2Cl2);
1H NMR: d=9.24 (1H, m), 9.12 (2H, m), 6.14 (1H, s), 3.39 (1H, m),
2.18 (3H, s), 2.14 (1H, m), 1.87 (2H, m), 1.70 (4H, m), 1.42 ppm (1H,
m); 13C NMR: d=203.3, 176.1, 161.4, 148.8, 133.8, 129.8 (2C), 128.3,
122.8, 85.6, 47.0, 30.0, 22.4, 20.5, 18.8, 17.7 ppm; IR (film): n˜ =1720,
1619 cmÀ1; HRMS (ESI): m/z calcd for C17H16N2O7Na+: 383.0850; found:
383.0847.
(E)-1-(3,4-Dimethyl-6-methylenecyclohex-3-enyl)-3-phenylprop-2-en-1-
one (27): BF3·OEt2 (1.1 equiv) was added to a solution of AVK 10
(0.070 g, 0.41 mmol) and 2,3-dimethylbutadiene (0.25 mL, 2.2 mol) in
CH2Cl2 (40 mL) at À788C. The solution was stirred for 1 h and poured
into a separatory funnel containing a saturated solution of sodium bicar-
bonate. The aqueous layer was extracted thoroughly with CH2Cl2. The
combined organic solutions were dried over Na2SO4 and concentrated
under vacuum. Flash column chromatography of the residue (2.5%
EtOAc in hexanes) yielded 27 (0.051 g, 51%) as an off-white solid. M.p.
71–748C (CH2Cl2/pentane); 1H NMR: d=7.63 (1H, d, J=16 Hz), 7.55
(2H, m), 7.38 (3H, m), 6.97 (1H, d, J=16 Hz), 4.98 (1H, m), 4.86 (1H,
Chem. Eur. J. 2011, 17, 8098 – 8104
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8103

J. Burnell et al.
Atoms in Molecules (Eds.: C. F. Matta, R. J. Boyd), Wiley-VCH,
Weinheim, 2007, pp. 375–397.
[13] For a recent review, see: M. Harmata, Chem. Commun. 2010, 46,
8886–8903.
[8] C. F. Matta, J. Hernꢂndez-Trujillo, J. Phys. Chem. A 2003, 107,
7496–7504.
[14] The stereochemical assignment is in accord with NOE measure-
ments.
[9] a) M. Feyereisen, G. Fitzgerald, A. Komornicki, Chem. Phys. Lett.
1993, 208, 359–363; b) F. Weigend, M. Haser, Theor. Chem. Acc.
1997, 97, 331–340; c) R. A. Distasio, Jr., P. Steele, Y. M. Rhee, Y.
Shao, M. Head-Gordon, J. Comput. Chem. 2007, 28, 839–856.
[10] B. Alcaide, P. Almendros, C. Aragoncillo, Chem. Eur. J. 2002, 8,
1719–1729.
[15] The regiochemical outcome was confirmed through detailed analysis
of 2D NMR spectra (COSY, HSQC and HMBC) of the products.
[16] X. Ma, J.-X. Wang, S. Li, K.-H. Wang, D. Huang, Tetrahedron 2009,
65, 8683–8689.
[17] M. L. Maddess, M. Lautens, Org. Lett. 2005, 7, 3557–3560.
[18] R. Aumann, H. J. Weidenhaupt, Chem. Ber. 1987, 120, 23–27.
[11] J. A. Marshall, G. M. Schaaf, J. Org. Chem. 2001, 66, 7825–7831.
[12] CCDC-803794 contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
Received: February 16, 2011
Published online: June 7, 2011
8104
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 8098 – 8104