Synthesis of 3-methylenecyclohexan-1-ols by lewis acid catalyzed cyclization of (epoxy-allyl)silanes

Article abstract of DOI:10.1002/ejoc.200901263

A new route for the synthesis of (epoxy-allyl)silanes bearing the PhMe ₂Si group has been developed and their acid-catalyzed cyclization studied. The so-called normal products derived from 5-exo or 6-endo attack were never obtained. On the contrary, an interesting tandem rearrangement/cyclization process was observed, which selectively led to 3-methylenecyclohexan-1-ols. A mechanism is proposed to explain this tandem reaction. The stereoselectivity of the cyclization process depends on the nature of the catalyst.

Full text of DOI:10.1002/ejoc.200901263

FULL PAPER
DOI: 10.1002/ejoc.200901263
Synthesis of 3-Methylenecyclohexan-1-ols by Lewis Acid Catalyzed Cyclization
of (Epoxy–allyl)silanes
Francisco J. Pulido,*[a] Asunción Barbero,*[a] and Pilar Castreño[a]
Keywords: Epoxides / Cyclization / Lewis acids / Lewis acid catalysis / Silanes
A new route for the synthesis of (epoxy–allyl)silanes bearing
the PhMe2Si group has been developed and their acid-cata-
lyzed cyclization studied. The so-called normal products de-
rived from 5-exo or 6-endo attack were never obtained. On
the contrary, an interesting tandem rearrangement/cycliza-
tion process was observed, which selectively led to 3-methyl-
enecyclohexan-1-ols. A mechanism is proposed to explain
this tandem reaction. The stereoselectivity of the cyclization
process depends on the nature of the catalyst.
reaction involves nucleophilic ring-opening of the epoxide
group, activated by the Lewis acid, proceeding through the
most stable carbocation[6,7] (Scheme 1).
Introduction
Most biologically active molecules contain quite complex
ring skeletons. Among them, polycyclic natural products
containing cyclopentanoid, cyclohexanoid and cyclohep-
tanoid systems are abundant in nature, for example, hima-
chalane and perforenone sesquiterpenes. The structural di-
versity of these compounds has provided many challenges
for synthetic chemists and has inspired many creative pro-
cedures for the stereoselective synthesis of medium-sized
carbocyclic rings.[1] Cyclization strategies to give medium-
sized rings are often inhibited by entropic factors and trans-
annular interactions.[2] Moreover, a flexible route using sim-
ple reagents and mild conditions that allows the synthesis
of different-sized carbocycles is a valuable alternative.
During the past few decades, the importance of silicon-
containing compounds has increased tremendously due to
their use as versatile building blocks in a variety of synthetic
transformations.[3] Moreover, organosilanes have often been
used in the synthesis of natural products to control the
stereochemistry of reactions taking place in their neigh-
bourhood.[4] In particular, allylsilanes have proved to be
one of the most useful reagents in the construction of
carbo- and heterocyclic compounds. Our contribution to
this active field has focused on developing new procedures
for five-, six- or seven-membered ring annulations using al-
lylsilane chemistry.[5]
Scheme 1. Lewis acid catalyzed cyclization of (epoxy–allyl)silanes.
In a preliminary communication, we showed that (ep-
oxy–allyl)silanes bearing a dimethyl(phenyl)silyl group un-
dergo Lewis acid catalyzed cyclization reactions to give 3-
methylenecyclohexan-1-ols.[8] We now present in full the re-
sults of this methodology, together with a discussion of the
factors governing the stereocontrol of the process, and re-
port new examples reflecting the influence of the silyl group
(PhMe2Si vs. Me3Si) and the substitution of the epoxide on
the course of the reaction.
However, the corresponding acid-catalyzed cyclization of
allylsilanes containing an epoxide group has attracted less
attention. It has been observed, as a general trend, that the Results and Discussion
The synthesis of allylsilanes[9] and -stannanes[10] by the
[a] Departamento de Química Orgánica,
C/ Dr Mergelina s/n, 47011 Valladolid, Spain
Fax: +34-983-423013
metallocupration of allenes and their synthetic applications
have been the subject of intense study in our group for the
last decade. In particular, the reaction of allene with the
lower order cyano(silyl)cuprate PhMe2SiCu(CN)Li involves
addition of copper to the central carbon atom and a silicon
E-mail: pulido@qo.uva.es
barbero@qo.uva.es
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200901263.
Eur. J. Org. Chem. 2010, 1307–1313
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1307

F. J. Pulido, A. Barbero, P. Castreño
FULL PAPER
atom to the terminal carbon atom of the allenic system to Table 2. Lewis acid catalyzed cyclization of (epoxy–allyl)silanes 3.
form intermediates of type 1, which, in the presence of α,β-
unsaturated oxo compounds undergo Michael addition re-
actions to yield (oxo–allyl)silanes 2a–g (Table 1). All the re-
actions were carried out in the presence of BF3·OEt2 or
TMSCl, which considerably increased the yield (Table 1).
Subsequent reaction of compounds 2a–g with dimethylsul-
fonium methylide gave (epoxy–allyl)silanes 3a–g as mixtures
of diastereomers (Table 1).
Table 1. Synthesis of (epoxy–allyl)silanes 3a–g by silylcupration of
allene.
Entry
R1
R2
R3
Yield [%]
2
3
a
b
c
d
e
f
Me
Me
Et
Et
iPr
H
Ph
Me
Me
H
H
H
H
Me
H
H
H
H
H
90[a]
87[a]
82[a]
80[b]
70[b]
91[b]
83[b]
72
78
65
76
68
75
67
g
H
Me
[a] BF3·Et2O was used as catalyst. [b] TMSCl was used as catalyst.
[a] Reactions with BF3·Et2O were performed in CH2Cl2 at 0 °C. [b]
Reactions with TiCl4 were performed in CH2Cl2 at –78 °C.
For the acid-mediated cyclization of (epoxy–allyl)silanes
3 we used 1 equiv. of the appropriate Lewis acid and dichlo-
romethane as solvent, which is the most commonly em-
ployed solvent for allylation reactions with allylsilanes. In
most cases the cyclization process was complete almost as
soon as the sample could be analysed. The use of alumin-
ium-based Lewis acids resulted in the formation of complex
mixtures and low yields of the cyclic products. The results
of this Lewis acid catalyzed cyclization reactions are shown
in Table 2.[8]
The first thing to note is that the Lewis acid mediated
cyclization of (epoxy–allyl)silanes 3 is mechanistically quite
different to the classic intramolecular SN2 opening of the
oxirane ring. In effect, compounds 4 and 5 thus obtained
are the reaction products of neither a 5-exo nor 6-endo cy-
clization. Instead, the cyclization of 3 unexpectedly afforded
3-methylenecyclohexan-1-ols in good yields (Scheme 2).
The relative stereochemistries of the products 4 and 5
were determined by extensive NMR experiments and in
particular by COSY and NOESY. For example, in com-
pound 4a the coupling pattern for 4-H (t, J = 5.0 Hz)
clearly indicates that it is equatorial (Ph axial), whereas the
coupling pattern for 1-H (td, J = 9.0 and 4.2 Hz) indicates
that 1-H and 2-H are axial (OH and Me, trans-diequato-
rial). The NOE correlations between 1-H and Me show
strong cross-peaks at the coordinates with no cross-peaks
being found for the correlation between 1-H and 2-H,
which supports the trans OH/Me relationship. At the same
time, 4-Ph shows a cross-peak with 2-H with no cross-peak
with Me observed, which confirms the anti relationship of
4-Ph and 2-Me.
Scheme 2. Possible pathways for the cyclization of (epoxy–allyl)-
silanes 3.
1308
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1307–1313

Methylenecyclohexanols by Cyclization of (Epoxy–allyl)silanes
Scheme 3. Plausible mechanism for the formation of 3-methylenecyclohexanols.
A plausible mechanism for the formation of 3-methylene- molecular acid-catalyzed cyclization of aldehydes contain-
cyclohexanols can be rationalized as shown in Scheme 3. ing in the alkyl side-chain an allylstannane or -silane when
Thus, the reaction pathway would involve two steps: an ini- a small Lewis acid was used. To rationalize the observed
tial Lewis acid catalyzed rearrangement of the epoxide stereoselectivity, Keck et al. invoked a transition-state con-
group to the corresponding aldehyde (via the most stable formation in which the aldehyde and allylstannane assume
carbocation to give intermediate I) followed by reaction a synclinal orientation that favors secondary orbital overlap
with the allylsilane unit to provide an intermediate carbo- (bonding and energy-lowering) between the π* LUMO of
cation (stabilized by strong hyperconjugative donation or the aldehyde and the HOMO of the allylstannane nucleo-
hyperconjugation[11] from the adjacent carbon–silicon phile. In the absence of steric and other interactions, this
bond), finally losing its silyl group to a nucleophile (e.g., a frontier orbital effect could account for the observed syn
halide ion) to complete the process. The degree of concert- stereoselectivity (Scheme 5).
edness between the two steps is still uncertain.[12]
The fact that the observed epoxide rearrangement takes
place prior to the cyclization is probably a consequence of
an unfavourable orbital alignment of the allylsilane/epoxide
pair in the reactive conformation, as we will discuss later.
It is also likely that both steric and electronic factors play
an important role in governing the stereochemical outcome
of the reaction.
Examination of the results for the Lewis acid mediated
cyclization of (epoxy–allyl)silanes 3, summarized in Table 2,
Scheme 5. Favorable secondary orbital interaction between the oxy-
gen atom of the carbonyl group and the silylmethylene carbon
atom.
indicates two different types of behaviour depending on the
Note, it is interesting that when TiCl4 is used for the cy-
clization process the stereoselectivity observed is noticeably
higher, but in the opposite sense. The main products are
methylenecyclohexanols 5, which have an anti relationship
between the hydroxy group and the C-2 substituent and
must result from an antiperiplanar transition state
(Scheme 6). The product ratio observed is the result of sim-
ple thermodynamic control via a Zimmerman–Traxler-type
transition state in which the bulky Lewis acid complexed to
the carbonyl group shows a steric preference to be equato-
rial and anti to the pseudoequatorial substituent on C-2.
This hypothesis is reinforced by the fact that the greater the
bulkiness of the C-2 substituent, the higher the stereoselec-
tivity of the cyclization (Table 2, Entries 3, 6 and 8).
nature of the Lewis acid used. Thus, with a small Lewis
acid such as BF3·OEt2, the predominant methylenecy-
clohexanol in every case (except for 3a) is the syn adduct 4
(Table 2, Entries 2, 4, 5, 7 and 10). To rationalize this result
we propose that the reaction proceeds through a chair-like
transition state in which the R substituent at C-2 adopts a
pseudoequatorial orientation and the aldehyde function-
ality adopts a pseudoaxial orientation (Scheme 4).
Scheme 4. Synclinal transition state.
The predominance of the syn products 4 can be rational-
ized by using a frontier molecular orbital (FMO) argument.
In related work by Keck[13] and Cox[14] and their co-
workers, good stereoselectivities were obtained in the intra- Scheme 6. Antiperiplanar transition state.
Eur. J. Org. Chem. 2010, 1307–1313
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1309

F. J. Pulido, A. Barbero, P. Castreño
FULL PAPER
Unfortunately, cyclization of (epoxy–allyl)silanes 3a, 3c
and 3f with TiCl4 did not give the desired methylene-
cyclohexanols as a result of competitive chlorohydrin for-
mation or a protodesilylation process. Nevertheless, what-
ever the factors that influence the stereochemical outcome
of this cyclization, it is important to note that we are able
to access both 1,6-syn- and 1,6-anti-cyclohexanol diastereo-
isomers in a selective fashion through the appropriate
choice of Lewis acid. This attractive feature of the reported
strategy provides a route for the synthesis of both dia-
stereomers of menthol from methylenecyclohexanols 4e and
5e (i.e., neomenthol and isomenthol) according to the pro-
cedure described by Braddock and Brown.[15]
Scheme 7. Synthesis of (epoxy–allyl)trimethylsilanes 7a,b.
We then subjected (epoxy–allyl)silanes 7a and 7b to the
standard cyclization conditions with boron trifluoride–di-
ethyl ether at 0 °C. Surprisingly, the reaction again afforded
compounds 4 and 5 in similar ratios,[21] which means that
the nature of the silyl group is not the cause of this unusual
tandem process (Scheme 8).
Mechanistic Insights
The acid-catalyzed rearrangement of epoxides to alde-
hydes or ketones is a well-known synthetic transforma-
tion.[16,17] Moreover, the strong anionic nature of the oxy-
gen–Lewis acid bond in intermediate A (Scheme 3) would
favour the rearrangement.[18] However, this is the first time
that this sequential type of process (rearrangement/cycliza-
tion) has been observed[19] in the acid-catalyzed cyclization
of (epoxy–allyl)silanes. The former observation seems to
indicate that in the case of (epoxy–allyl)silanes 3, rearrange-
ment of the epoxide occurs much faster than nucleophilic
substitution on the epoxy group. In this sense, the fact that
cyclization of both diastereoisomers of 3a gives the same
result when treated with Lewis acids indicates that the reac-
tion proceeds through common intermediates, and it could
be considered as evidence that rearrangement takes place
prior to cyclization.
Scheme 8. Lewis acid catalyzed cyclization of (epoxy–allyl)trimeth-
ylsilanes 7a and 7b.
To give some support to this hypothesis, we performed a
simple semi-empirical analysis of a model of (epoxy–allyl)-
The question to be analysed is why substrates 3 show silane by using Gaussian 98,[22] and the preliminary results
preference to undergo acid-promoted epoxide rearrange- seem to indicate that the geometry, symmetry and orienta-
ment instead of the expected oxirane-cleavage/allylsilane- tion of the HOMO orbital of the allylsilane and the LUMO
terminated cyclization process leading to 5-exo or 6-endo of the epoxide are not favourable for a good overlap in (ep-
adducts.
oxy–allyl)silanes of type 3. However, a good electronic in-
The fact that the previously reported[6,7] cyclization of teraction could be observed in the regioisomeric (oxo–
(epoxy–allyl)trimethylsilanes is perceptibly different from allyl)silane I between the LUMO of the aldehyde and the
that observed for our dimethyl(phenyl)silyl derivatives HOMO of the allylsilane nucleophile. However, more theo-
(Scheme 1) led us to think that the nature of such a group retical work is probably needed before a definitive picture
may be the cause of the different behaviour shown. To study can be provided, and this is not the aim of this work.
such an influence we synthesized (epoxy–allyl)trimethylsil-
Another factor to be considered in these cyclization reac-
anes 7a and 7b, which are structurally analogous to the di- tions is the influence of the substitution of the epoxide on
methyl(phenyl)silyl substrates 3d and 3b (Scheme 7). The sil- the course of the reaction. For this purpose we prepared
anes 7a and 7b were synthesized by using the Trost meth- 1,1-disubstituted (epoxy–allyl)silanes 8a and 8b and tested
odology.[20]
them under the standard Lewis acid cyclization conditions.
The magnesium derivative of 2-bromo-3-(trimethylsilyl)- As is shown in Scheme 9, we obtained the normal 6-endo
propene was generated by metal/halogen exchange with products 9a and 9b, which corresponds to a direct opening
tBuLi followed by the addition of anhydrous magnesium of the epoxide. This result reinforces our hypothesis because
bromide. Transmetallation with CuI followed by Michael in substrates 8a and 8b the poorer intrinsic migratory apti-
addition to the corresponding enone gave the desired allyl- tude of the methyl group (compared with hydrogen) pre-
trimethylsilanes 6a and 6b in satisfactory yields. The (ep- vents the migration from being much faster than further
oxy–allyl)silane derivatives were obtained by standard reac- attack of the nucleophilic allylsilane on the stable tertiary
tion with dimethylsulfonium methylide (Scheme 7).
carbocation intermediate.
1310
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1307–1313

Methylenecyclohexanols by Cyclization of (Epoxy–allyl)silanes
J = 4.8 Hz, 1 H, CHHO), 4.56 (s, 1 H, =CHH), 4.64 (s, 1 H,
=CHH) ppm. 13C NMR (75 MHz, CDCl3, recognizable signals): δ
= 8.5, 19.2, 24.9, 26.3, 40.1, 53.1, 59.0, 106.4 ppm. IR (neat): ν =
˜
1632, 1248, 1100, 836 cm–1. MS (EI): m/z = 289 [M + 1]+, 273 [M –
Me]+, 203, 135.
2-Ethyl-5-{[dimethyl(phenyl)silyl]methyl}-1,2-epoxy-5-hexene (3d):
1
Colourless liquid (167 mg, 76%). H NMR (300 MHz, CDCl3): δ
= 0.35 (s, 6 H, SiMe2), 0.90 (t, J = 7.5 Hz, 3 H, Me), 1.78–1.50 (m,
4 H), 1.79 (s, 2 H, CH2Si), 1.89 (t, J = 8.1 Hz, 2 H, CH2C=), 2.48
(d, J = 4.7 Hz, 1 H, CHHO), 2.55 (d, J = 4.7 Hz, 1 H, CHHO),
4.57 (s, 1 H, =CHH), 4.64 (s, 1 H, =CHH), 7.57–7.37 (m, 5 H, Ph)
ppm. 13C NMR (75 MHz, CDCl3): δ = –3.0, 8.8, 26.2, 26.8, 32.1,
33.0, 52.0, 59.8, 107.7, 127.7, 129.0, 133.5, 138.9, 146.3 ppm. IR
Scheme 9. Cyclization of substituted (epoxy–allyl)silanes 8a,b.
(neat): ν = 1650, 1225, 1100, 910 cm–1. MS (EI): m/z = 275 [M +
Conclusions
˜
1]+, 259 [M – Me]+, 197, 135. C17H26OSi (274.18): calcd. C 74.39,
A new synthesis of (γ,δ-epoxy–allyl)silanes by silyl-
cupration of allene and capture of the intermediate cuprate
with enones has been developed. Surprisingly, the Lewis
acid catalyzed cyclization of these substrates leads to 3-
methylenecyclohexanols through an interesting tandem pro-
cess of an epoxide/aldehyde rearrangement and an allyl-
silane-terminated cyclization.
The factors governing this tandem reaction have been
studied, and a mechanism is proposed. Moreover, a study
of the influence of the Lewis acid used on the stereoselecti-
vity of the cyclization shows two different types of behav-
iour. Thus, the major diastereomeric syn-cyclohexanol ob-
H 9.55; found C 74.78, H 9.84.
2-Isopropyl-5-{[dimethyl(phenyl)silyl]methyl}-1,2-epoxy-5-hexene
(3e): Colourless liquid (157 mg, 68%). 1H NMR (300 MHz,
CDCl3): δ = 0.34 (s, 6 H, SiMe2), 0.90 (d, J = 6.8 Hz, 3 H, Me),
0.99–0.87 (m, 1 H, CHMe2), 0.93 (d, J = 6.8 Hz, 3 H, Me), 1.79
(s, 2 H, CH2Si), 1.83–1.62 (m, 4 H, CH2CH2), 2.43 (d, J = 4.6 Hz,
1 H, CHHO), 2.53 (d, J = 4.6 Hz, 1 H, CHHO), 4.56 (s, 1 H,
=CHH), 4.62 (s, 1 H, =CHH), 7.56–7.35 (m, 5 H, Ph) ppm. 13C
NMR (75 MHz, CDCl3): δ = –3.0, 17.8, 18.1, 26.3, 29.2, 32.0, 32.4,
50.3, 62.1, 107.5, 127.7, 128.9, 133.5, 138.9, 146.6 ppm. C18H28OSi
(288.19): calcd. C 74.94, H 9.78; found C 75.26, H 10.07.
5-{[Dimethyl(phenyl)silyl]methyl}-1,2-epoxy-5-hexene (3f): Colour-
1
tained when a small Lewis acid (e.g., BF3) is used indicates less liquid (148 mg, 75%). H NMR (300 MHz, CDCl3): δ = 0.35
(s, 6 H, SiMe2), 1.66–1.59 (m, 2 H), 1.81 (s, 2 H, CH2Si), 2.11–1.94
(m, 2 H), 2.42 (dd, J = 5.0, 2.7 Hz, 1 H, CHHO), 2.72 (dd, J =
5.0, 4.1 Hz, 1 H, CHHO), 2.88–2.82 (m, 1 H, CHO), 4.60 (s, 1 H,
=CHH), 4.67 (s, 1 H, =CHH), 7.60–7.34 (m, 5 H, Ph) ppm. 13C
NMR (75 MHz, CDCl3): δ = –3.0, –2.9, 26.0, 30.6, 34.1, 47.0, 51.9,
a synclinal transition state. However, when the size of the
Lewis acid used is larger, for example, TiCl4, excellent dia-
stereoselectivities are observed in favor of the anti-cyclo-
hexanol, which is the result of simple thermodynamic con-
trol via an antiperiplanar transition state.
108.1, 127.7, 129.0, 133.5, 138.8, 145.8 ppm. IR (neat): ν = 1634,
˜
1249, 1100, 837 cm–1. MS (EI): m/z = 246 [M]+, 231 [M –
Me]+, 135. HRMS (ESI): calcd. for C15H22OSi 246.1440 [M]+;
found 246.1473.
Experimental Section
Synthesis of (Epoxy–allyl)silanes: BuLi (0.6 mL, 1 mmol, 1.6  in
hexanes) was added dropwise to a solution of trimethylsulfonium
iodide (220 mg, 1 mmol) in dry THF (5 mL) and the mixture
stirred at 0 °C for 5 min. Then a solution of the (oxo–allyl)silane
(0.8 mmol) in THF (1 mL) was added. After stirring at 0 °C for an
additional 30 min and at room temp. for 1 h, brine (10 mL) was
added and the mixture extracted with diethyl ether, dried and con-
centrated to dryness. The residue was purified by chromatography
to give (epoxy–allyl)silanes 3a–g and 7a,b. Compounds 3a,[8] 3b[8]
and 3g[8] have been described previously.
2-Ethyl-5-[(trimethylsilyl)methyl]-1,2-epoxy-5-hexene (7a): Colour-
less liquid (115 mg, 68%). H NMR (300 MHz, CDCl3): δ = 0.03
(s, 9 H, SiMe3), 1.22 (t, J = 7.5 Hz, 3 H, Me), 1.53 (s, 2 H, CH2Si),
1.78–1.55 (m, 4 H, CH2CH2), 1.89 (t, J = 8.3 Hz, 2 H, CH2Me),
2.61–2.58 (m, 2 H, CH2O), 4.53 (s, 1 H, =CHH), 4.60 (s, 1 H,
=CHH) ppm. 13C NMR (75 MHz, CDCl3): δ = –1.4, 8.8, 26.9,
1
27.0, 32.2, 33.1, 52.1, 59.9, 106.9, 146.9 ppm. IR (neat): ν = 1630,
˜
1463, 1248, 854 cm–1. C12H24OSi (212.16): calcd. C 67.86, H 11.39;
found C 68.19, H 11.62.
2,4,4-Trimethyl-5-[(trimethylsilyl)methyl]-1,2-epoxy-5-hexene (7b):
2-Ethyl-5-(dimethylphenylsilylmethyl)-4-methyl-1,2-epoxy-5-hexene
(3c): Colourless liquid of a 1:1 mixture of diastereoisomers A + B
(150 mg, 65%).
1
Colourless liquid (127 mg, 70%). H NMR (300 MHz, CDCl3): δ
= 0.04 (s, 9 H, SiMe3), 1.19 (s, 3 H, Me), 1.20 (s, 3 H, Me), 1.30
(s, 3 H, Me), 1.39 (d, J = 14.3 Hz, 1 H, CHHSi), 1.57 (s, 2 H,
CH2CO), 1.98 (d, J = 14.3 Hz, 1 H, CHHSi), 2.54 (d, J = 4.9 Hz,
1 H, CHHO), 2.63 (d, J = 4.9 Hz, 1 H, CHHO), 4.65 (s, 1 H,
=CHH), 4.82 (s, 1 H, =CHH) ppm. 13C NMR (75 MHz, CDCl3):
δ = –0.50, 21.1, 22.1, 27.6, 28.5, 39.4, 47.7, 54.7, 55.9, 108.2,
153.4 ppm. C13H26OSi (226.18): calcd. C 68.96, H 11.57; found C
69.27, H 11.80.
1
A: H NMR (300 MHz, CDCl3): δ = 0.33 (s, 6 H, SiMe2), 0.87 (t,
J = 7.5 Hz, 3 H, CH3CH2), 1.02 (d, J = 6.8 Hz, 3 H, Me), 1.30
(dd, J = 14.5, 8.6 Hz, 1 H, CHHCO), 1.41 (dd, J = 14.5, 7.5 Hz, 1
H, CHHCO), 1.67–1.53 (m, 1 H), 1.78 (s, 2 H, CH2Si), 1.85–1.74
(m, 2 H), 2.44 (d, J = 4.6 Hz, 1 H, CHHO), 2.52 (d, J = 4.6 Hz, 1
H, CHHO), 4.56 (s, 1 H, =CHH), 4.66 (s, 1 H, =CHH), 7.55–7.35
(m, 5 H, Ph) ppm. 13C NMR (75 MHz, CDCl3): δ = –2.8, 9.1, 19.8,
24.7, 26.4, 37.1, 40.3, 51.6, 59.2, 106.7, 127.7, 128.9, 133.5, 139.0,
151.7 ppm.
Synthesis of (1,1-Dimethylepoxy–allyl)silanes: tert-Butyllithium
(0.6 mL, 1.7  in pentane, 1 mmol) was added dropwise to a solu-
tion of isopropyldiphenylsulfonium fluoroborate (1 mmol) in dry
THF at –40 °C, and the resulting mixture was stirred at this tem-
1
B: H NMR (300 MHz, CDCl3, recognizable signals): δ = 0.97 (d,
J = 6.8 Hz, 3 H, Me), 2.54 (d, J = 4.8 Hz, 1 H, CHHO), 2.60 (d, perature for 1 h. A solution of the (oxo–allyl)silane (0.85 mmol) in
Eur. J. Org. Chem. 2010, 1307–1313
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1311

F. J. Pulido, A. Barbero, P. Castreño
FULL PAPER
dry THF (4 mL) was added; after 1 h at –40 °C, the mixture was
warmed to 0 °C and quenched with brine (10 mL). After standard
workup, the residue was purified by chromatography to give (ep-
oxy–allyl)silanes 8a,b.
Me), 0.98 (s, 3 H, Me), 1.61–1.46 (m, 2 H), 1.84–1.76 (m, 1 H),
1.88 (d, J = 13.4 Hz, 1 H, CHHC=), 2.06 (d, J = 13.4 Hz, 1 H,
CHHC=), 2.12–2.02 (m, 1 H), 2.31 (dt, J = 13.6, 5.1 Hz, 1 H), 3.46
(dd, J = 9.1, 3.9 Hz, 1 H, CHOH), 4.60 (s, 1 H, =CHH), 4.70 (s, 1
H, =CHH) ppm. 13C NMR (75 MHz, CDCl3): δ = 20.1, 27.4, 31.0,
31.9, 36.7, 46.1, 76.4, 109.2, 145.9 ppm. C9H16O (140.12): calcd. C
77.09, H 11.50; found C 77.43, H 11.76.
5-{[Dimethyl(phenyl)silyl]methyl}-1,1-dimethyl-1,2-epoxy-5-hexene
(8a): Colourless liquid (121 mg, 52 %). 1H NMR (300 MHz,
CDCl3): δ = 0.34 (s, 6 H, SiMe2), 1.24 (s, 3 H, Me), 1.30 (s, 3 H,
Me), 1.71–1.57 (m, 2 H), 1.80 (s, 2 H, CH2Si), 2.04–1.93 (m, 2 H),
2.65 (t, J = 6.2 Hz, 1 H, CHO), 4.59 (s, 1 H, =CHH), 4.69 (s, 1 H,
=CHH), 7.56–7.31 (m, 5 H, Ph) ppm. 13C NMR (75 MHz, CDCl3):
δ = –3.0, 18.7, 24.8, 26.0, 27.1, 34.7, 58.2, 64.0, 108.1, 127.7, 129.0,
133.5, 138.8, 145.9 ppm. C17H26OSi (274.18): calcd. C 74.39, H
9.55; found C 74.72, H 9.89.
1,2,2-Trimethyl-4-methylenecyclohexanol (9b): Colourless liquid
(109 mg, 71%). 1H NMR (300 MHz, CDCl3): δ = 0.89 (s, 3 H,
Me), 0.97 (s, 3 H, Me), 1.19 (s, 3 H, Me), 1.74–1.55 (m, 4 H,
CH2CH2), 2.08 (d, J = 13.2 Hz, 1 H, CHHC=), 2.18–2.03 (m, 1
H), 2.32 (d, J = 13.2 Hz, 1 H, CHHC=), 4.59 (s, 1 H, =CHH), 4.68
(s, 1 H, =CHH) ppm. 13C NMR (75 MHz, CDCl3): δ = 23.4, 24.2,
24.3, 31.1, 37.1, 39.1, 45.5, 73.7, 108.4, 146.6 ppm. C10H18O
(154.14): calcd. C 77.87, H 11.76; found C 78.11, H 12.02.
5-{[Dimethyl(phenyl)silyl]methyl}-1,1,2-trimethyl-1,2-epoxy-5-hex-
ene (8b): Colourless liquid (135 mg, 55%). 1H NMR (300 MHz,
CDCl3): δ = 0.33 (s, 6 H, SiMe2), 1.18 (s, 3 H, Me), 1.26 (s, 3 H,
Me), 1.28 (s, 3 H, Me), 1.54–1.50 (m, 1 H), 1.77–1.69 (m, 1 H),
1.78 (s, 2 H, CH2Si), 1.94–1.79 (m, 2 H), 4.55 (s, 1 H, =CHH), 4.62
(s, 1 H, =CHH), 7.56–7.31 (m, 5 H, Ph) ppm. 13C NMR (75 MHz,
CDCl3): δ = –3.0, 19.5, 20.8, 21.3, 26.2, 33.7, 33.8, 62.1, 64.4, 107.5,
127.7, 129.0, 133.5, 138.8, 146.5 ppm. C18H28OSi (288.19): calcd.
C 74.94, H 9.78; found C 75.31, H 10.11.
Supporting Information (see footnote on the first page of this arti-
cle): General experimental methods, details of the synthesis of
oxoallylsilanes, characterization of compounds 2c–f and 6a,b and
1H and 13C NMR spectra for all new compounds.
Acknowledgments
Cyclization of (Epoxy–allyl)silanes: BF3·OEt2 (0.15 mL, 1.2 mmol)
or TiCl4 (0.13, 1.2 mmol) was slowly added to a solution of the
(epoxy–allyl)silane (1 mmol) in DCM (10 mL) under nitrogen. Af-
ter stirring at the appropriate temperature (0 or –78 °C) for 30 min,
MeOH (2 mL) was added and the mixture warmed to room tem-
perature. The organic layer was washed with brine and dried with
MgSO4. The solvent was evaporated and the residue purified by
chromatography to give methylenecyclohexanols 4a–g, 5a–g and
9a,b. Compounds 4a,b,[8] 4d,e,[23] 5a,b,[8] 5d,e,[23] 4g[8] and 5g[8] have
been described previously. The relative stereochemistries of the cy-
clic products were determined by extensive NMR experiments and
in particular by COSY and NOESY data.
We thank the Ministry of Science and Technology of Spain
(CTQ2006-02436) and the Junta de Castilla y León (VA074A08)
for financial support.
[1] L. Yet, Chem. Rev. 2000, 100, 2963–3007.
[2] G. Illuminati, L. Mandolini, Acc. Chem. Res. 1981, 14, 95–102.
[3] E. Langkopf, D. Schinzer, Chem. Rev. 1995, 95, 1375–1408.
[4] a) I. Fleming, A. Barbero, D. Walter, Chem. Rev. 1997, 97,
2063–2192; b) S. E. Denmark, J. P. Fu, Chem. Commun. 2003,
167–170; c) M. G. Organ, D. D. Winkle, J. Huffmann, J. Org.
Chem. 1997, 62, 5254–5266; d) M. Suginome, T. Iwanami, A.
Yamamoto, Y. Ito, Synlett 2001, 1042–1045.
(1R*,2S*,4S*)-2-Ethyl-4-methyl-5-methylenecyclohexanol
(4c):
1
[5] a) A. Barbero, C. García, F. J. Pulido, Tetrahedron 2000, 56,
2739–2751; b) A. Barbero, Y. Blanco, F. J. Pulido, Chem. Com-
mun. 2001, 1606–1607; c) A. Barbero, Y. Blanco, F. J. Pulido,
J. Org. Chem. 2005, 70, 6876–6883.
[6] a) L. E. Overman, P. A. Renhowe, J. Org. Chem. 1994, 59,
4138–4142; b) R. J. Armstrong, L. Weiler, Can. J. Chem. 1986,
64, 584–596; c) T. S. Tan, A. N. Mather, G. Procter, A. H. Da-
vidson, J. Chem. Soc., Chem. Commun. 1984, 585–586.
[7] a) G. Proctor, A. T. Russell, P. J. Murphy, T. S. Tan, A. N.
Mather, Tetrahedron 1988, 44, 3953–3973; b) G. A. Molander,
S. W. Andrews, J. Org. Chem. 1989, 54, 3114–3120.
[8] A. Barbero, P. Castreño, F. J. Pulido, Org. Lett. 2003, 5, 4045–
4048.
[9] For the silylcupration of allenes, see: a) A. Barbero, F. J. Pulido,
Acc. Chem. Res. 2004, 37, 817–825; b) A. Barbero, F. J. Pulido,
Synthesis 2004, 779–785.
[10] For the stannylcupration of allenes, see: a) A. Barbero, F. J.
Pulido, Chem. Soc. Rev. 2005, 34, 913–920; b) A. Barbero, P.
Cuadrado, A. M. González, F. J. Pulido, R. Rubio, I. Fleming,
Tetrahedron Lett. 1992, 33, 5841–5842; c) A. Barbero, P. Cuad-
rado, I. Fleming, A. M. González, F. J. Pulido, J. Chem. Soc.,
Chem. Commun. 1990, 1030–1031.
[11] a) S. Zhang, F. G. Borwell, J. Org. Chem. 1996, 61, 51–54; b)
M. Sugawara, J. I. Yoshida, J. Org. Chem. 2000, 65, 3135–3142;
c) J. B. Lambert, Tetrahedron 1990, 46, 2677–2689; d) J. B.
Lambert, Y. Zhao, R. W. Emblide, L. A. Salvador, X. Liu, J.-
H. So, E. C. Chelius, Acc. Chem. Res. 1999, 32, 183–190.
[12] We were not able to isolate or detect aldehydes resulting from
isomerization of the starting (epoxy–allyl)silane.
Colourless liquid (79 mg, 51%). H NMR (300 MHz, CDCl3): δ =
0.92 (t, J = 7.2 Hz, 3 H, MeCH2), 1.06 (d, J = 6.5 Hz, 3 H, Me),
1.37–1.21 (m, 3 H), 1.51–1.39 (m, 2 H), 1.64 (dt, J = 13.0, 3.7 Hz,
1 H), 2.10–2.01 (m, 1 H), 2.31 (dd, J = 13.5, 2.9 Hz, 1 H, CHHC=),
2.44 (dd, J = 13.5, 3.4 Hz, 1 H, CHHC=), 3.92 (br. s, 1 H, CHOH),
4.80 (s, 2 H, =CH2) ppm. 13C NMR (75 MHz, CDCl3): δ = 11.7,
18.0, 25.1, 36.8, 37.1, 43.5, 43.9, 69.2, 108.7, 149.0 ppm. IR (neat):
ν = 3450, 3083, 1648, 1013, 890 cm–1. C H O (154.14): calcd. C
˜
10 18
77.87, H 11.76; found C 78.19, H 11.95.
(1R*,2S*,4R*)-2-Ethyl-4-methyl-5-methylenecyclohexanol
(5c):
1
Colourless liquid (26 mg, 17%). H NMR (300 MHz, CDCl3): δ =
0.94 (t, J = 7.4 Hz, 3 H, MeCH2), 1.10 (d, J = 7.0 Hz, 3 H, Me),
1.63–1.22 (m, 6 H), 2.24 (dd, J = 13.7, 5.0 Hz, 1 H), 2.54–2.48 (m,
2 H), 3.89 (br. s, 1 H, CHOH), 4.71 (s, 1 H, =CHH), 4.83 (s, 1 H,
=CHH) ppm. 13C NMR (75 MHz, CDCl3): δ = 11.8, 18.9, 23.6,
34.3, 35.9, 38.1, 38.8, 70.1, 110.1, 149.2 ppm. IR (neat): ν = 3580,
˜
3072, 1648, 1038, 803 cm–1.
1
3-Methylenecyclohexanol (4f): Colourless liquid (90 mg, 80%). H
NMR (300 MHz, CDCl3): δ = 1.52–1.39 (m, 2 H), 1.66 (br. s, 1 H,
OH), 1.97–1.89 (m, 2 H), 2.12–2.02 (m, 2 H), 2.38–2.30 (m, 2 H),
3.82 (tt, J = 9.0, 3.8 Hz, 1 H, CHOH), 4.65 (s, 2 H, =CH2) ppm.
13C NMR (75 MHz, CDCl3):
δ = 31.6, 35.9, 69.2, 107.8,
147.6 ppm. IR (neat): ν = 3350, 1650, 1061, 894 cm–1. MS (EI): m/z
˜
= 112 [M]+, 94, 79. C7H12O (112.09): calcd. C 74.95, H 10.78;
found C 75.29, H 11.06.
2,2-Dimethyl-4-methylenecyclohexanol (9a): Colourless liquid
(106 mg, 76%). 1H NMR (300 MHz, CDCl3): δ = 0.86 (s, 3 H,
[13] G. E. Keck, S. M. Dougherty, K. A. Savin, J. Am. Chem. Soc.
1995, 117, 6210–6223.
1312
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 1307–1313

Methylenecyclohexanols by Cyclization of (Epoxy–allyl)silanes
[14] J. Beignet, P. J. Jervis, L. R. Cox, J. Org. Chem. 2008, 73, 5462–
5475.
[22] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Mont-
gomery, R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Mil-
lam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J.
Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pom-
elli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P. Y.
Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, B. B.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P. M. W. Gill, B. G. Johnson, W. Chen, M. W. Wong, J. L.
Andres, M. Head-Gordon, E. S. Replogle, J. A. Pople,
Gaussian 98 (Revision A.7), Gaussian, Inc., Pittsburgh, PA,
1998. Theoretical level: MP2(fc)/3-21G*//HF/3-21G*.
[23] T. Ooi, K. Maruoka, H. Yamamoto, Tetrahedron 1994, 50,
6505–6522.
[15] D. C. Braddock, J. M. Brown, Tetrahedron: Asymmetry 2000,
11, 3591–3607.
[16] a) R. E. Parker, N. S. Isaacs, Chem. Rev. 1959, 59, 737–799; b)
B. Rickborn in Comprehensive Organic Synthesis Carbon-Car-
bon α-Bond Formation (Ed.: G. Pattenden), Pergamon Press,
Oxford, 1991, vol. 3, p. 733.
[17] A. Barbero, Y. Blanco, C. García, F. J. Pulido, Synthesis 2000,
1223–1228.
[18] Y. Kita, A. Furukawa, J. Futamura, K. Ueda, Y. Sawama, H.
Hamamoto, F. Fujioka, J. Org. Chem. 2001, 66, 8779–8786.
[19] Products of epoxide rearrangement have been obtained in a
similar cyclization reaction: X.-Y. Xiao, S.-K. Park, G. D.
Prestwich, J. Org. Chem. 1988, 53, 4869–4872.
[20] a) B. M. Trost, D. M. T. Chan, J. Am. Chem. Soc. 1982, 104,
3733–3735; b) B. M. Trost, B. P. Coppola, J. Am. Chem. Soc.
1982, 104, 6879–6881; c) B. M. Trost, M. Shimizu, J. Am.
Chem. Soc. 1982, 104, 4299–4301.
Received: November 4, 2009
Published Online: January 12, 2010
[21] None of the direct 5-exo or 6-endo epoxy-cyclization products
(see Scheme 4) were observed.
Eur. J. Org. Chem. 2010, 1307–1313
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1313