On the stereochemical course of the addition of allylsilanes to aldehydes

Article abstract of DOI:10.1016/j.tet.2012.05.084

Model compounds 3 and 5 have been studied to determine the orientation of the reacting double bonds in the transition state of the allylmetal-aldehyde addition. These models were designed to remove any intrinsic steric bias for the formation of the bicyclic products that would obfuscate a stereoelectronic contribution to the transition states. Model system 3 revealed a modest preference for the synclinal transition state, albeit in very low yields. Model system 5 underwent selective and largely Lewis acid independent cyclization primarily via a synclinal transition state. The high proximal selectivity observed in these cyclizations likely reflects the selectivity of an unhindered allylmetal-aldehyde addition for the synclinal transition state and results from a stereoelectronic preference, not an intrinsic steric bias, for the synclinal arrangement of double bonds.

Full text of DOI:10.1016/j.tet.2012.05.084

Tetrahedron 68 (2012) 7701e7718
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
On the stereochemical course of the addition of allylsilanes to aldehydes
*
Scott E. Denmark , Eric J. Weber, Neil G. Almstead, Larry M. Wolf
Roger Adams Laboratory, Department of Chemistry, University of Illinois, Urbana, IL 61801, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Model compounds 3 and 5 have been studied to determine the orientation of the reacting double bonds
in the transition state of the allylmetalealdehyde addition. These models were designed to remove any
intrinsic steric bias for the formation of the bicyclic products that would obfuscate a stereoelectronic
contribution to the transition states. Model system 3 revealed a modest preference for the synclinal
transition state, albeit in very low yields. Model system 5 underwent selective and largely Lewis acid
independent cyclization primarily via a synclinal transition state. The high proximal selectivity observed
in these cyclizations likely reflects the selectivity of an unhindered allylmetalealdehyde addition for the
synclinal transition state and results from a stereoelectronic preference, not an intrinsic steric bias, for
the synclinal arrangement of double bonds.
Received 8 February 2012
Received in revised form 12 May 2012
Accepted 22 May 2012
Available online 29 May 2012
Keywords:
Allylsilane
Stereochemistry
Aldehyde
Lewis acid
Synclinal
Ó 2012 Elsevier Ltd. All rights reserved.
Antiperiplanar
1. Introduction
The controlled construction of stereocenters in open-chain
systems is of primary importance in the synthesis of acyclic natu-
ral and non-natural products. Many methods have been developed
recently to synthesize the long sequences of stereocenters present
in these molecules including the addition of allylmetal reagents to
aldehydes.1 The utility of the allylmetalealdehyde addition is partly
derived from the high yield, excellent site selectivity, and the mild
conditions under which it can be employed.
The reaction of a substituted allylmetal reagent with an alde-
hyde can result in the formation of diastereomeric homoallylic al-
cohols (Scheme 1).2 The allylmetalealdehyde addition has proven
to be successful with a wide variety of metals including boron,3 tin,4
silicon,5 chromium,6 and titanium.7 The diastereoselectivity ob-
served in the Lewis acid mediated allylmetalealdehyde additions is
dependent upon the allylmetal used. This dependence has been
classified into three groups that relate the stereochemical outcome
of the reaction to the geometry of the double bond.1f,8 Type 1 re-
actions wherein the syn/anti ratio reflects the Z/E ratio of the
starting allylmetal (B, Al, Sn); Type 2 reactions wherein the reaction
is syn-selective independent of the geometry of the allylmetal (Sn,
Si); and Type 3 reactions wherein the reaction is anti-selective in-
dependent of the geometry of the allylmetal (Cr, Ti, Zr).
Scheme 1.
Proposals for transition state geometry have been set forth for
all three types of reactions,1f,9 but those that fall in the Type 2
family are the focus of the studies described herein. Proposals for
Type 2 reactions invoke an open chain arrangement of the
reacting species.1f The two limiting hypotheses identify the tor-
sional angle between the double bonds (synclinal (60ꢀ) and
antiperiplanar (180ꢀ)) and minimization of nonbonded in-
teractions as key features for relative diastereoselection (Scheme
2). The internal induction process is governed primarily by the
relative disposition of the metal electrofuge and the aldehyde
(anti or syn SE0) in the transition structure (Scheme 3). Thus, the
orientation of the double bonds and the location of the metal in
the transition structure uniquely define the stereochemical out-
come of the reaction.
Previous investigations from these laboratories have described
the synthesis and cyclization of a model system that unambiguously
determined the stereochemical course of addition in an allylsila-
neealdehyde reaction.10 However, criticisms of this model focused
on a potential inherent bias given the diastereomeric relationship
of the products. We therefore undertook the investigation of two
new model systems that remove this bias.
* Corresponding author. Tel.: þ1 217 333 0066; fax: þ1 217 333 3984; e-mail
address: sdenmark@illinois.edu (S.E. Denmark).
0040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2012.05.084

7702
S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
The stereochemical course of the addition of electrophiles to
allylsilanes has been studied to establish the position of the silicon
electrofuge in the transition structure of these reactions.14 In an
early study, Fleming examined the additions of electrophiles to
stereochemically-defined allylsilanes, which were constrained in
either a five or six-membered ring.15,16 The addition of an electro-
phile to these substra0tes resulted in the formation of products from
both anti and syn SE pathways. Fleming concluded that the ste-
reochemical constraints of the ring systems were likely the domi-
nant influence in the observed stereoselectivity of these reactions.
Therefore, these models cannot be used to elucidate the intrinsic
0
preference of the SE reaction.
0
Fleming expanded the study of the SE reaction to include the
use of acyclic allylsilanes.17 In protiodesilylation experiments the
Scheme 2.
0
allylsilane cleanly gives products from an anti SE reaction whereas
the diastereomeric allylsilane affords a mixture of both syn and anti
0
SE products (Scheme 4). Results from a deuteration study indicate
that in addition to the anti selectivity of the allylsilane, the cyclo-
hexyl ring has a preference for axial protonation. When this axial
preference is in opposition to the anti selectivity of the allylsilane,
the molecule will find an alternative reaction where the stereo-
specificity is lost.
Scheme 3.
2. Background
The Lewis acid-mediated addition of electrophiles to allylsilanes
has been extensively studied.5 In most cases the addition of an
0
electrophile to an allylsilane is an anti SE process. In the ground
state, simple allylsilanes are known to prefer the conformation
where the small substituent H eclipses the double bond.11 The
electrophile can then approach the double bond from the same side
as the allylmetal (syn SE0) or from the side opposite the allylmetal
(anti SE0). The configuration of the newly formed stereogenic center
is therefore dependent upon the directionality of attack (Scheme 3).
After attack of the electrophile on the double bond only a slight
rotation of the CeC bond is necessary for the formation of the in-
termediate ii, which is stabilized by hyperconjugation with the
silicon atom (Scheme 3).12 The silyl group is then released, resulting
in the stereoselective formation of a trans-double bond.
The site selectivity and stereochemical course of electrophilic
additions to allylsilanes have been modeled computationally by
Hehre.13 In this study, the conformational profile of 2-silylbut-3-
ene was determined and three energy minima were observed
(Chart 1). In the two most stable conformers the CeSi bond is
perpendicular to the CeC double bond. The interaction of a point
Scheme 4.
Wetter18 and0Kitching19 have also examined the stereochemical
course of the SE reaction. In the studies performed by Wetter the
reaction of a disilylalkene proceeds through either the syn or anti
SE pathways depending upon the electrophile. Kitching examined
the SE reaction of some cyclohexenylsilanes, -germanes, and
-stannanes. For all three of the allylmetals, the results indicate that
attack by proton occurred with anti selectivity except when a trans-
4-tert-butylcyclohex-2-enyl derivative was used. In these reactions
the approach of the electrophile anti to the metal is impeded by the
presence of the tert-butyl substituent.
Kumada and Hayashi have carried out extensive studies to de-
fine all of the stereochemical features of the addition to aldehydes
with titanium tetrachloride (Scheme 5).20 The results from this
study can be summarized as follows:1 the enantiomeric excess of
the products was essentially the same as the starting materials;2
the E-allylsilanes reacted with high diastereoselectivity (syn/anti,
92:8e99:1);3 the Z-allylsilanes were less selective with the
resulting syn/anti ratio of products dependent upon the structure of
the aldehydes (syn/anti, 50:50e99:1). The configuration of the
products obtained for all of the reactions studied is interpreted in
terms of an anti SE0 reaction. To explain the observed selectivities an
acyclic transition structure was proposed in which the double-
bonds are arranged in an antiperiplanar relationship. The ob-
served diastereoselectivities are proposed to result from a minimi-
zation of steric interactions in the transition structures (Scheme 6).
0
0
charge (a proton) and the allylsilane was next studied in the three
low energy conformers. By using this ‘test’ electrophile an elec-
trostatic potential map was developed. The electrophilic attack
onto the two low energy conformers of 2-silylbut-3-ene, iii and iv,
was shown to occur anti to the silyl group. In the high-energy
conformer v, attack will occur anti to the methyl group.
Chart 1.

S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
7703
Scheme 5.
The forgoing stereochemical/mechanistic studies of the allyl-
metalealdehyde addition have primarily involved intermolecular
reactions. Although the position of the silicon electrofuge with
respect to the approaching electrophile has been defined, the exact
orientation of the double bonds in the transition state is, in most
cases, unknown. The investigation of an intramolecular allylme-
talealdehyde addition was undertaken in our laboratories to pro-
vide an unambiguous correlation between product stereochemistry
and transition state geometry.8,10 To achieve this objective, both the
position of the silicon electrofuge and the orientation of the
reacting double bonds in the transition structure was un-
ambiguously defined in an intramolecular allylmetalealdehyde re-
action. Although intramolecular cyclizations cannot exactly model
the corresponding intermolecular reactions, the results from these
studies provide useful insights into intrinsic preferences in the
transition structure of the allylmetalealdehyde additions.
The deuterium-labeled model 1 (Scheme 7) was designed to
differentiate between the syn and anti SE pathways.10d,f The posi-
0
tion of the deuterium atom in the products can be used to establish
0
if a syn or anti SE pathway has been followed. This model is able to
determine both the position of the silicon electrofuge and the rel-
ative disposition of the double bonds in the transition structure of
the allylmetalealdehyde addition simultaneously.
The selectivity observed in the cyclization of the model system 1
is dependent upon the Lewis acid used. The bulky Lewis acid SnCl4
led to a non-selective reaction while cyclization with triflic acid, the
Scheme 6.
Scheme 7.

7704
S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
sterically least demanding reagent, resulted in a very selective re-
action favoring the proximal diastereomer. If E-complexation ge-
ometry is assumed between the Lewis acid and the aldehyde10b,21
then the major steric contribution in the model system would
arise from the (phenyldimethylsilyl)methylene group. The forma-
tion of any of the proximal product with SnCl4, a Lewis acid known
to form 2:1 complexes with aldehydes,22 was interpreted as
a stereoelectronic advantage for the synclinal transition state. The
high distal selectivity observed with fluoride ion is thought to re-
sult from a change in mechanism. The fluoride ion is proposed to
the diastereomeric bias inherent in model 1. Intramolecluar cycli-
zation of 3 will lead to the formation of enantiomeric (ignoring the
13C label) and therefore isoenergetic products 4a and 4b.24 The
enantiomeric composition (as established by the 13C label) will
provide a direct measure of the energy difference between the
antiperiplanar and synclinal transition state geometries. Thus, this
model should be ideal for elucidating the existence of stereo-
electronic control alone. The 13C label, by synthetic design, will be
uniquely located in the methylidene group. The product distribu-
tion will be determined by integration of the signals for the pseudo-
enantiomers in the 13C NMR spectrum.
The second model, 5 seeks to minimize the inherent bias in
model 1 by removing the trialkylsilylmethylene group and locating
the silicon electrofuge in the ring, but trans to (and therefore away
from) the tethered aldehyde moiety. Cyclization of model system 5
leads to the formation of the diastereomeric alcohols 6a and 6b. The
alcohols 6a and 6b do not contain the exocyclic olefin present in the
cyclization products of model system 1. Therefore, the observed
selectivity should more accurately reflect the intrinsic preference
for the synclinal versus the antiperiplanar transition states.24
initiate
a nucleophilic attack on the aldehyde by an allyl-
fluorosiliconate or an allyl anion.23 This process is best accommo-
dated by an antiperiplanar transition state in which the developing
negative charges experience less repulsion than in a synclinal
transition state.
The products from both the synclinal and antiperiplanar tran-
0
sition structures were found to arise from an anti SE reaction. The
arrangement of double bonds in the transition structure does not
affect the relative disposition of the silicon electrofuge that must be
disposed away from the approaching electrophile. All of the cycli-
za0tions with Lewis acids were greater than 95% selective for the anti
SE reaction. The high selectivity observed demonstrates that in
0
4. Results and discussion
a sterically unbiased SE reaction an anti orientation of the elec-
trophile with respect to silicon is preferred.
4.1. Synthesis of model system 3
3. Model design
4.1.1. Construction of the pentadienylsilane unit. The construction of
the pentadienylsilyl moiety for 6 represented a major synthetic
challenge. Any synthetic approach had to take into account the
need to place the 13C label exclusively in the methylidene group. A
survey of the synthetic methods developed for the preparation of
pentadienylsilanes25 indicated that these methods would not be
applicable since they involve the intermediacy of pentadienyl an-
ions, which in model 3 would lead to scrambling of the 13C label.
The solution to this problem is outlined in Scheme 9.
Although model system 1 does not contain any bulky sub-
stituents at the methylene group attached to the silicon, an in-
herent bias might exist because the products are diastereoisomers
and thus the ground states are not isoenergetic. A difference in
ground state energies could manifest itself in the relative energies
of the transition states leading to formation of the products. Two
advanced models, 3 and 5 (Scheme 8) were formulated to address
Scheme 8.

S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
7705
Methylenation26 of 3-chloro-2-cyclohexenone 727 afforded 8 in 70%
yield. The chloro diene was very reactive and had to be quickly
subjected to the next set of reaction conditions. The nickel-
catalyzed coupling28 of 8 with trimethylsilylmethylmagnesium
chloride29 afforded diene 9 in 85% yield. Pentadienylsilane 9 was
stable to chromatography (alumina), GC, and distillation (bp 90 ꢀC/
15 mmHg). A pure sample could be stored at ꢁ20 ꢀC without de-
composition for extended periods of time.
again, the chloro diene could not be stored and was immediately
subjected to the nickel catalyzed cross-coupling with trimethylsi-
lylmethylmagnesium chloride29 to provide pentadienylsilane 15 in
89% yield. Treatment of 15 with 0.05 M aqueous sodium hydroxide
afforded alcohol 16 in 95% yield. Oxidation of 16 with 1,10-(azodi-
carbonyl)dipiperidine and tert-butoxymagnesium bromide33 affor-
ded aldehyde 10 in 60% yield. The basic character of the Mukaiyama
oxidation protocol was critical as all other oxidizing agents tested
Scheme 9.
Seyferth25 had previously shown that the Lewis acid catalyzed
reactions of (2,4-pentadienyl)trimethylsilane with various electro-
philes, such as aldehydes, acetals, and acid chlorides provided ad-
ducts in fair to high yields. Accordingly, the Lewis acid catalyzed
reactions of 9 with acetaldehyde, benzaldehyde, benzoyl chloride
and the dimethyl acetal of benzaldehyde were investigated. Only
the Et2AlCl-catalyzed addition of 9 to benzaldehyde afforded any
product (20% yield), which resulted from the attack of the elec-
trophile at a terminus of the pentadienylsilane. This result is con-
sistent with that reported by Seyferth, who observed adducts with
only this site selectivity. In all other reactions investigated, complex
mixtures with poor mass recovery were obtained.
led to decomposition of the pentadienylsilane. The model system
was stable to chromatography (Al2O3), GC, and could be stored at
ꢁ20 ꢀC for extended periods without decomposition.
4.1.3. Cyclization of model system 10. Treatmentof a 0.05 M solution
of 10 in CH2Cl2 at ꢁ70 ꢀC with 1.1 equiv of FeCl3 for 2.5 h led to the
formation of the expected bicyclic alcohol 17 in 20% yield
(Scheme 11). Bicyclic alcohol 17 was the only observable product by
GC analysis. The 13C NMR spectrum of 17 shows that the methylidene
carbon signals are clearly distinguishable at 108.99 and 110.99 ppm,
thus direct integration of these signals should be possible.
4.1.2. Preparation of model system 10 (unlabeled). Considering the
sensitivity of the intermediates and potential for self-addition of the
target, the initial goal was to synthesize model 3 without the 13C
label. The synthesis of the unlabeled model 10 is outlined in Scheme
10. Treatment of the known diketo acid 11,30 prepared in four steps
in 27% overall yield from diethyl 3-oxoglutarate, with excess oxalyl
chloride provided the unstable acid chloride 12. Reduction of 12
with lithium tris-(3-pentoxy)aluminum hydride31 and protection of
the resulting alcohol as its diphenylmethylsilyl ether32 afforded
chloro enone 13 in 53% overall yield from 11. The intermediate al-
cohol was very prone to polymerization upon evaporation of solvent
after chromatography, but polymerization could be minimized by
keeping the alcohol in solution at all times. Methylenation of 13 as
previously described26 afforded chloro diene 14 in 83% yield. Once
Scheme 11.
4.1.4. Preparation of 13C labeled model system 3. The synthesis of
model 3 is outlined in Scheme 12. Treatment of 13 with methyl-
enetriphenylphosphorane enriched with 40% 13C label afforded
chloro diene 18 in 84% yield. The 13C-enriched phosphonium salt
was prepared by mixing the appropriate amounts of unlabeled
Scheme 10.

7706
S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
Scheme 12.
methyltriphenylphosphonium bromide with 99% 13C labeled
methyltriphenylphosphonium iodide, which in turn was prepared
by treating 99% 13C labeled methyl iodide with triphenylphosphine.
The high percentage (40%) of 13C label incorporated into 3 facili-
tated the stereochemical analysis of the reaction products 4a and
4b. Coupling of 18 with trimethylsilylmethylmagnesium chloride
afforded pentadienylsilane 19 in 80% yield. Deprotection with base
proceeded smoothly in 96% yield and oxidation33 of the resulting
alcohol 20 provided model system 3 in 51% yield. The incorporation
of the 13C label was clearly evident in the 1H NMR spectrum of 3.
The methylidene protons of the 13C labeled portion of 3 appeared as
a doublet of doublets centered at 4.61 ppm with a carbon-hydrogen
coupling constant of 156 Hz. The mass spectrum of 3 showed the
presence of two molecular ions at 222 and 223 m/z. The 13C NMR
spectrum of 3 showed an enhancement of only the exo-methyl-
idene carbon signal at 107.88 ppm, indicating that no scrambling of
the 13C label had occurred.
were determined by integration of the methylidene carbon signals
in the 13C NMR spectrum.34 Long delay times (>5 s) were used to
ensure complete relaxation of the 13C nuclei. Signal to noise ratios of
at least 20/1 were obtained for each 13C NMR spectrum to minimize
error in the integration of the methylidene carbon signals.
The results summarized in Table 1 for the cyclization of 3 show
a modest preference for the formation of the distal isomer, sug-
gesting a stereoelectronic advantage for synclinal orientation of
reactants under electrophilic conditions (Scheme 9).35 Since the
distribution of reaction products 4a and 4b is a direct measure of
the relative energies of the transition states for synclinal and
antiperiplanar geometry, simple energy calculations show that at
ꢁ70 ꢀC the synclinal geometry is preferred by only 0.40 kcal/mol.
The operation of an intrinsic electronic effect that favors synclinal
geometry, if present, is very small.
In contrast to model 1, the range of syn-selectivity for the Lewis
acid catalyzed cyclization of 3 is very narrow. The insensitivity to
the size of the Lewis acid indicates the lack of a steric component.
Taking into account the diminished reactivity of 3 compared to 1,10f
the lack of a steric component suggests that bond formation occurs
in a late transition state. In a late transition state the carbonecarbon
4.1.5. Cyclization of model system 3. The Lewis acid promoted re-
actions of 3 were performed by treating a 0.05 M solution of 3 in
CH2Cl2 at ꢁ70 ꢀC with 1.1 equiv of Lewis acid (Table 1). The fluoride
ion promoted cyclization of 3 was performed by treating a 0.05 M
solution of 3 in THF at 20 ꢀC with 1.0 equiv of a 0.32 M solution of
tetrabutylammonium fluoride in THF. The proximal/distal ratios
bond
a to the trimethylsilyl group acquires more double bond
character as bond formation takes place, thus minimizing the steric
interaction between the Lewis acidealdehyde complex and the
(trimethylsilyl)methylene protons as discussed earlier for the
model system 1. The lack of selectivity observed for the fluoride ion
induced cyclization of 3 suggests the formation of a delocalized
pentadienyl anion.
Table 1
Cyclization of model 3a
The low yields of reaction products in Table 1 are puzzling. An
alternative reaction pathway may be available to 3, such as the
intermolecular addition proceeding through the termini of the
pentadienylsilyl moiety. Alternatively, the pentadienylsilane may
be undergoing transmetalation with the Lewis acid prior to closure,
which would preclude any interpretation of the results.36
4.1.6. 13C NMR study of the transmetalation of 9 with Lewis acids.
The transmetalation of the pentadienylsilane 9 with selected Lewis
acids was studied by 13C NMR spectroscopy at low temperature.
Pentadienylsilane 9 was very reactive toward SnC14. The silane was
completely consumed with 1.0 equiv of SnC14 at ꢁ70 ꢀC, pre-
sumably to form the pentadienyltrichlorostannane derivative.
Though these results indicate that transmetalation with SnCl4 may
compete with formation of the active Lewis acidealdehyde com-
plex, this apparently is not the case. Transmetalation would pre-
sumably occur through the 13C labeled terminus of the
pentadienylsilyl moiety. Cyclization would then take place through
a Type I transition state with the result being a decrease in syn-
selectivity, yet the syn-selectivity of the SnCl4 promoted cyclization
of 3 is comparable to that of the other Lewis acids. Treatment of 9
Entry Reagent
Time, min Temp, ꢀC Yield, %b 4a/4b, %c DDGyd
e
1
2
3
4
5
BF3$OEt2
FeCl3
150
150
150
60
ꢁ70
ꢁ70
ꢁ70
ꢁ70
20
6
20
12 (16)
12
70/30
70/30
73/27
67/33
53/47
0.34
0.34
0.40
0.29
0.05
Et2AlCle
SnCl4
n-Bu4NþFꢁe,f
60
21
a
All cyclizations were performed with 1.05 equiv of Lewis acid in CH2Cl2
at ꢁ70 ꢀC except where noted.
b
Yields of isolated material, yield in parentheses based on recovered starting
material.
c
Ratios were determined by 13C NMR analysis.
Calculated at 203 K (in kcal/mol).
Reaction performed in THF at 20 ꢀC.
d
e
f
In addition to the bicyclic alcohols 4a and 4b, a 33% yield of aldehyde 21 was
obtained.

S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
7707
with BF3$OEt2 resulted in a complex reaction mixture whose 13C
NMR spectrum was not interpretable.
assigned by comparison of their 13C NMR shifts with published
values.41
4.2.3. Cyclization of model system 5. The cyclization of 5 was pro-
moted by various Lewis acids and the results are collected in Table 2.
All of the Lewis acids studied were selective for the proximal di-
astereomer. The results obtained with BF3$OEt2 and CF3SO3H were
almost identical to those obtained with model system 1.10f The cy-
clization with TiCl4 and SnCl4 were found to be highly selective for
the proximal diastereomer. The cyclization with SnCl4 (the sterically
most demanding Lewis acid) actually afforded a 90/10 ratio of di-
astereomers favoring the proximal isomer (entry 1). Foregoing
studies revealed that the size of the Lewis acidealdehyde complex
influences the selectivity of the cyclization. For model system 5 only
a small change in selectivity was observed between cyclization with
either SnCl4 or CF3SO3H (compare entries 1 and 4). These results
indicate that the steric bulk of the Lewis acid does not play a sig-
nificant role in determining the stereochemical outcome of the re-
action. In model system 5 no external methylene unit exists, which
could interact with the Lewis acidealdehyde complex. In fact, the
silane is fixed in an anti orientation with respect to the approaching
aldehyde (anti SE0). The cyclization of model system 5 with fluoride
afforded primarily the distal product resulting from an anti-
periplanar transition state. The results with fluoride indicate that
the antiperiplanar transition state is accessible, but is not favored in
reactions with the Lewis acids.
4.2. Synthesis and cyclization of model system 5
4.2.1. Synthesis of model system 5. The synthesis of model system 5
was achieved in four steps starting from the known bicyclic ke-
tone37 22 (Scheme 13). BaeyereVilliger oxidation of 22 with MCPBA
buffered with potassium carbonate afforded bicyclic lactone 23 in
87% yield. Treatment of the lactone with (PhMe2Si)2CuLi17 followed
by addition of diazomethane afforded one isomer of the desired
allylsilane 24 in 77% yield. The ester was reduced with lithium
aluminum hydride (72% yield) and the resulting alcohol 25 was
then oxidized with the Collins reagent39 to give the target aldehyde
5 in 84% yield.
Table 2
Cyclization of model 5a
Scheme 13.
4.2.2. Proof of relative configuration of model system 5. The con-
figuration of 5 was tentatively assigned as trans on the basis of an
analogous precedent, namely the addition of a silyl-cuprate reagent
to an allyl acetate, which proceeds via an anti SN20 process.17
Because the configurational assignment is crucial to the in-
terpretation of the results, the relative configuration of 5 was
independently established by conversion to a known compound
(Scheme 14). Hydrogenation of the allylsilane ester 24 afforded the
saturated silane ester 26 in 80% yield. The phenylsilane was treated
with fluoroboric acid39 in ether to afford fluorosilane 27 (95% yield),
which was converted to hydroxy ester 28 by treatment with
mCPBA39 in 81% yield. Closure of the hydroxy ester to the known
bicyclic lactone40 29 was effected by heating in refluxing benzene
with a catalytic amount of p-toluenesulfonic acid (74% yield).
Comparison of the 1H NMR data with published values indicates
that the two compounds are identical. Finally, the configurations of
the products of the allylmetalealdehyde addition of 5 (6a/6b) were
Entry
Reagent
Time, min
Mass recovery, %c
6a/6b, %b,c
DDGyd
1
2
3
4
5
6
SnCl4
TiCl4
20
3
40
1
90
720
80
88
80
80
89
71
90/10
94/6
80/20
95/5
78/22
16/84
0.85
1.07
0.54
1.14
0.49
BF3$OEt2
CF3SO3He
ZrCl4
n-Bu4NþFꢁf
ꢁ0.64
a
All cyclizations were performed with 1.05 equiv of Lewis acid in CH2Cl2
at ꢁ78 ꢀC except where noted.
b
Average of at least three runs within ꢂ3%.
c
Ratios and yields were calculated based on independently determined response
factors versus cyclododecane.
d
Calculated at 195 K (in kcal/mol).
Only 0.95 equiv of triflic acid used.
Reaction performed in THF at 66 ꢀC.
e
f
4.2.4. Stereochemical analysis. The two limiting transition states
leading to the formation of the proximal and distal diastereomers
are shown in Scheme 15. If the Lewis acid adopts an E-complexa-
tion geometry with the aldehyde10b,21 the steric influence from the
Lewis acid will be minimal. This is not true with the model system 1
where the (phenyldimethylsilyl)methylene unit can be influenced
by the bulky Lewis acid SnCl4. The results from the cyclization of
this model demonstrate that the allylmetalealdehyde addition will
proceed through a synclinal transition state when no external steric
bias is present.
Although several explanations for the origin of the syn selec-
tivity observed in the cyclization of model system 5 can be imag-
ined, the two most plausible will be considered here. The first
invokes a Coulombic attraction between the two charged centers in
an unsymmetrical transition state, favoring the synclinal over the
Scheme 14.

7708
S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
Scheme 15.
Scheme 16.
antiperiplanar transition state (Scheme 16). In the activated com-
plexes vi and viii, the charges are localized on proximal atoms,
whereas in the transition state the charges become separated; to
a greater extent in the distal transition state ix than in the proximal
transition state vii. In non-polar solvents charge separation is en-
ergetically disfavored, leading to the preferential formation of the
proximal diastereomer. With more polar solvents, the formation of
the distal diastereomer may be more prevalent. Huisgen has used
a similar explanation to explain the retention of configuration ob-
served in [2þ2] cycloadditions of electronically complementary
olefins.42 In these cycloadditions a large effect was observed be-
tween polar and non-polar solvents, influencing both the rate of
reaction and the selectivity of the cycloaddition.
been normalized so that water and tetramethylsilane are the two
possible extremes on the scale, at 1.0 and 0.0, respectively. Changing
solvent from nitropropane to hexanes had little effect on the selec-
tivity of the reaction. Unfortunately, solvents more polar than ni-
tromethane could not be used in this study because they generally
contain a Lewis basic atom which would compete with the aldehyde
for complexation with the Lewis acid. The lack of a solvent effect
does not necessarily rule out the possibility of Coulombic attraction
in the transition state of the cyclization of model system 5. In an
intramolecular reaction any solvent effect would only arise if the
intermediate was highly-polarized or long-lived. If cyclization of
model system 5 proceeds via an early transition state (with little
charge separation) the solvent effect is expected to be small.
To probe this explanation, a solvent study of the cyclization of 5
with BF3$OEt2 was performed. Boron trifluoride etherate was cho-
sen for this study because of the relatively low selectivity observed
for the proximal diastereomer with this reagent in CH2Cl2. The re-
sults obtained from this study are shown in Table 3. Apparently, the
effect of solvent polarity in the stereochemical outcome of these
cyclizations is marginal. The values shown in the table for ENT are
froma solvent polarityscale developed byReichardt.43 This scale has
A second explanation for the synclinal transition state preference
focused on secondary orbital interactions. Anh and Thanh have sug-
gested that the stereochemical outcome of an aldol reaction may be
controlled by the secondary overlap between the frontier molecular
orbitals.44a In this proposal two energetically favorable interactions
are identified: (1) the in-phase overlap between the carbonyl carbon
and the internal carbon of allylsilane moiety and (2) the in phase
overlap between the carbonyl oxygen and the silane bearing carbon
(Scheme 17). Although these interactions may be operative in model
system 1 (x), the location of the silyl group in model system 5 pre-
cludes this type of favorable overlap (xi). Given the similar synclinal
preferences for these two models, secondary orbital overlap does not
provide a convincingexplanation. Thus, to provide additionalinsights
into the origin of stereocontrol, computational investigation of the
transition states for these addition reactions were undertaken.
Table 3
Effect of solvent in the BF3$OEt2 mediated cyclization of 5a
Entry
Solvent
Time,
min
6a/6b, %b,c
Mass
ETN
DDGyd
recovery, %c
1
2
3
4
CH2Cl2
Hexane
Toluene
Nitropropane
40
80/20
82/18
82/18
90/10
80
76
80
72
0.309
0.009
0.099
0.373
0.54
0.59
0.59
0.85
24 h
180
24 h
4.3. Computational investigations
a
All cyclizations were performed with 1.05 equiv of BF3$OEt2 at ꢁ78 ꢀC.
Average of at least three runs within ꢂ3%.
b
c
Model system 5 and a simplified intermolecular system were
modeled using density functional theory for purposes of compari-
son and to judge whether the intramolecular case introduces
Ratios and yields were calculated based on independently determined response
factors versus cyclododecane.
d
Calculated at 195 K (in kcal/mol).9.

S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
7709
The transition states were located employing the M06-2X45
functional, which was chosen because of recent applications that
effectively treat reactions of main group elements. Additional
functionals were investigated but provided qualitatively similar
results and are not discussed further. The transition states were
located using the 6-31þG(d,p) basis set and solvation was included
using the CPCM46 model with dichloromethane as the solvent. The
electronic energies of the optimized transition state geometries
were further refined using the 6-311þþG(2d,2p) basis set. The
geometries were verified as transition states by analysis of the vi-
brational frequencies determined from numerically calculated
hessians (the method of central differences) applying the same
level of theory as that used for locating the geometries. All calcu-
lations were performed using the GAMESS47 program.
Scheme 17.
Initially the reaction of 5 was modeled wherein CF3SO3H was
substituted with H3Oþ (Fig.1). The preferred transition state takes up
a synclinal conformation (TS-a2,
agreement with the experimental results (
D
Gz¼2.1 kcal/mol) in qualitative
D
Gz¼1.14 kcal/mol). The
unwanted bias for the synclinal conformation (Scheme 18). It was
judged that if the calculated energy difference between the syn-
clinal and antiperiplanar transition states for the intramolecular
and intermolecular cases were similar to each other as well as to
the experimental values, model system 5 would indeed reflect the
intrinsic synclinal preference in the transition state.
electronic energy difference in the intermolecular analogue (TS-b1 e
TS-b2,
D
Ez¼1.1 kcal/mol) qualitatively compares well to the calcu-
lated electronic energy difference in the intramolecular case (TS-a2 e
TS-a1,
D
Ez¼1.9 kcal/mol). This agreement suggests that the intra-
molecular case reflects the electronic preference for the synclinal
Experimental
+O
Compute intramolecular
Compute intermolecular
CH3
H
+O
H
+O
H
H
H2O
H
H2O
H
F3CO2SO
SiPhMe2
SiMe3
SiMe3
SiMe3
antiperiplanar
CH3
O+
H
H
O+
H
O+
H
H2O
H
F3CO2SO
H2O
H
SiPhMe2
synclinal
SiMe3
Scheme 18.
Fig. 1. Relative free energies (kcal/mol) at 195 K and electronic energies in parentheses; energies for TS-a are relative to TS-a2; energies for TS-b and TS-c are relative to TS-c2;
ꢀ
selected distances are labeled in A; CPCM-M06-2X/6-311þþG(2d,2p)//M06-2X/6-31þG(d,p).

7710
S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
orientation in the intermolecular case satisfactorily. The internal
nuclear coordinates of the relevant atoms are also similar. The dis-
tance of the forming bond is smaller by w0.2 A likely reflecting minor
yields were very low, but a modest preference for the synclinal
transition state persisted. In addition, the highly reactive pentadie-
nylsilane may have undergone transmetalation prior to cyclization.
Model system 5 eliminated the steric contribution of the (phenyl-
dimethylsilyl)methylene group and underwent selective cyclization
primarily via a synclinal transition state. The selectivities observed in
the cyclization of 5 were not highly dependent upon the nature of the
Lewis acid studied. The high proximal selectivity observed in these
cyclizations probably reflects the selectivity of an unhindered allyl-
metalealdehyde addition for the synclinal transition state. Model
ꢀ
intramolecular constraints rather than a later transition state.
However, upon the inclusion of thermal corrections, the agree-
ment is lost. In fact, little to no conformational preference is pre-
dicted to exist in the analogous intermolecular reaction (TS-b2 e TS-
b1, not shown,
D
Gz¼0.1 kcal/mol). This disagreement can be un-
derstood when considering additional steric interactions in the in-
termolecular case that are not present in the intramolecular case.
Inspection of Newman projections (Scheme 19) reveals a difference
in dihedral angle of w6o (CH3eCeCeH) and decreased distances
between the methyl group of the aldehyde and the allyl moiety of
the allylsilane in the synclinal conformation. These geometric dif-
ferences reflect a degree of rigidity present in TS-b2 not present in
TS-b1 causing greater unfavorable thermal contributions (mostly
entropic) to the total energy because of decreased levels of freedom.
system
5 removed the possible steric bias of the (phenyl-
dimethylsilyl)methylene unit present in model system 1. According
to computation analysis, the high selectivity observed for the proxi-
mal diastereomer in the cyclization of model system 5 most likely
results from a stereoelectronic preference, not an intrinsic steric bias,
for the synclinal arrangement of double bonds in the transition state.
Taken together, the results from model systems 1, 3 and 5 allow
a number of general conclusions to be drawn: (1) under activation
by Lewis and Brønsted acids, the addition of allylic silanes with
aldehydes experience a modest preference for the synclinal ar-
rangement of reacting double bonds, (2) the magnitude of this
preference is variable and dependent upon the size and co-
ordination geometry of the Lewis acid; smaller Lewis acids lead to
greater synclinal preferences, (3) the origin of the synclinal pref-
erence is most likely stereoelectronic.
6. Experimental section
6.1. General methods
Scheme 19.
1H NMR spectra were recorded at 200, 300, 400, or 500 MHz in
CDCl3 with CHCl3 as an internal reference (7.26 ppm). 13C NMR
spectra were recorded at 75.5, 100.6, or 125.8 MHz in CDCl3 solu-
tions with CHCl3 (77.0 ppm) as internal reference. Chemical shifts
A more relevant comparison can be made with TS-c in which the
position of the trimethylsilylmethyl group attached to the allylsilane
is switched (Fig. 1). The presence of differential steric interactions
would be small. Much better agreement is seen with the predicted
are reported in parts per million (d), coupling constants, J, are
reported in hertz. Infrared spectra were recorded either as thin
films, solutions (CCl4) or as KBr pellets, on an IBM FTIR-32 spec-
trometer. Peaks are reported in units of cmꢁ1 with the following
relative intensities: br (broad), s (strong 67e100%), m (medium
33e67%), or w (weak 0e33%). Mass spectra were recorded on
a Varian MAT CH-5 spectrometer with ionization voltages of 70 or
10 eV. Data are reported in the form m/z (intensity relative to
base¼100%). Analytical gas chromatography was performed on
a Hewlett Packard 5890 equipped with both split and on-column
injectors. The columns used were an HP 50 m OV-1 cross-linked
methyl silicone (column A), an HP 50 m HP-5 phenyl-methyl sili-
cone gum (column B), an HP 20 m HP-1 cross-linked methyl silicone
megabore (column C), HP 20 m 20 M Carbowax (column D) and an
HP 50 m OV-17 (column E) Retention times (tR) and integrated ratios
were obtained from either a HewlettePackard 3393A recorder or
a HewlettePackard 3396II recorder. Analytical high pressure liquid
chromatography (HPLC) was performed on a HewlettePackard HP
1090 liquid chromatograph with a PerkineElmer LC-75 Spectro-
photometric Detector. A Supelco LC-Si 5-m column was used. The
detector wavelength was set to 254 nm. Retention times (tR) and
integrated ratios were obtained from an HP 3390A recorder. Ana-
lytical thin-layer chromatography was performed on Merck silica
gel plates with F254 indicator, visualization was accomplished by UV
light, vanillin, and iodine. Solvents used in reactions were reagent
grade and were distilled from the indicated drying agents: hexane,
dichloromethane (CaH2); ether, THF (Na/benzophenone). Solvents
for extraction and chromatography were technical grade and dis-
tilled from the indicated drying agents: dichloromethane (CH2Cl2),
pentane, hexane, ethyl acetate: CaCl2; diethyl ether (Et2O), tert-
butyl methyl ether (TBME): CaSO4/FeSO4. Solvents for re-
crystallization were spectral grade. Column chromatography was
free energy difference in this case (TS-c1 e TS-c2,
D
Gz¼2.4 kcal/mol).
A counter argument for this being a relevant comparison is that this
arrangement now allows for secondary orbital overlap, which may
contribute to the synclinal preference (Scheme 17). However, as was
pointed out previously, the similarity of the synclinal preference
for model systems 1 (secondary orbital interactions possible,
D
D
Gz¼1.1 kcal/mol) and 5 (secondary orbital interactions not possible,
Gz¼1.1 kcal/mol) clearly suggests that secondary orbital overlap
may not be the major contributor to the selectivity.
Although the difference between the predicted and experi-
mental selectivity in 5 may be considered significant (0.96 kcal/
mol), the correctly predicted conformation as well as the excellent
agreement between the predicted selectivity in 5 and TS-c supports
a lack of bias in the probing interactions of 5. Additionally, the fact
that TS-c2 is predicted to be the lowest energy transition state by
1.5 kcal/mol provides further support for the relevance of com-
paring TS-c to 5. Overall this investigation suggests that the
selectivity in 5 reflects an intrinsic synclinal preference in the
transition state. The disparity between the experimental and cal-
culated preferences for the synclinal transition structures may be
ascribed to the difference in activator (triflic acid vs hydronium ion)
and/or the difference in electrofugal group (dimethylphenylsilyl vs
trimethylsilyl).
5. Conclusions
The synthesis and cyclization of two new models designed to
elucidate the stereoelectronic preferences in allylsilane aldehyde
addition reactions were described. Model system 3 required the re-
action to take place at C(3) of a pentadienylsilane and consequently

S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
7711
performed by the method of Still48 with 32e63 mm silica gel
(Merck). Bulb-to-bulb (Kugelrohr) distillations were performed on
Data for 9: bp 100 ꢀC (0.05 mmHg); 1H NMR (90 MHz, CCl4):
1.90e3.66 (m, 7H), 6.12 (s, 1H, HC(50)); IR: 1785 (s), 1695 (s), 1601
(s), 1445 (w), 1420 (w) 1385 (w), 1345 (w), 1310 (m), 1250 (w), 1160
(w), 1080 (w), 1060 (w), 1020 (m), 985 (m), 920 (w), 900 (w) cmꢁ1
d
€
a Buchi GKR-50 Kugelrohr; boiling points (bp) refer to air bath
temperatures and are uncorrected. Melting points (mp) were de-
termined on a Thomas-Hoover capillary melting point apparatus
and are uncorrected. Elemental analyses were performed by the
University of Illinois Microanalytical Service Laboratory. Organo-
lithium reagents were titrated according to the method of Gilman.49
.
6.2.1. Preparation of 5-[20-(diphenylmethylsilyloxyethyl)]-3-chloro-
2-cyclohexen-1-one (13). To a solution of 2.25 g (10.9 mmol) of 12 in
40 mL of THF at ꢁ78 ꢀC was added 32.6 mL (22.8 mmol) of a 0.70 M
solution of lithium tris-(3-ethyl-3-pentoxy)aluminum hydride in
THF over a period of 30 min. The resulting pale yellow solution was
stirred at ꢁ78 ꢀC for 4 h and then quenched with 50 mL of satd aq
NaHCO3 solution. After warming the mixture to rt, 200 mL of ether
was added and the resulting emulsion was filtered through a fritted
glass funnel. The residue on the funnel was washed with 50 mL of
ether and the aqueous layer of the filtrate was extracted with ether
(2ꢃ100 mL). The ether layers were individually washed in series
with one, 50 mL portion of satd NH4Cl solution, combined, dried
(MgSO4), and evaporated. The residue was immediately purified by
chromatography on silica gel with ethyl acetate/hexane, 2:1 as el-
uent. Prior to evaporation, 8 mL of DMF was added to the eluent.
After evaporation of the ethyl acetate/hexane, 1.90 mL (9.21 mmol)
of diphenylmethylsilyl chloride and 1.252 g (18.39 mmol) of imid-
azole were added to the DMF solution. The solution was stirred at rt
for 0.5 h, then was poured onto 25 mL of satd aq NaHCO3 solution
and extracted with ether (3ꢃ50 mL). The ether layers were in-
dividually washed in series with one, 25 mL portion of water and
25 mL of brine. The ether layers were combined, dried (K2CO3) and
evaporated. The residue was chromatographed on silica gel with
hexane/ethyl acetate, 5:1 as eluent to give 2.135 g (53% yield from
11) of 13 as a clear, colorless oil. Data for 13: Rf 0.28 (hexane/ethyl
6.1.1. Preparation of 1-(trimethylsilylmethyl)-3-methylidene (9).
Preparation of 1-chloro-3-methylidenecyclohex-l-ene (8). To a solu-
tion of 10.2 g (28.7 mmol) of methyltriphenylphosphonium bro-
mide in 50 mL benzene was added 47.8 mL (28.7 mmol) of 0.06 M
potassium tert-amylate in benzene. The suspension was heated to
reflux for 1 h. The resulting orange suspension was allowed to cool
to rt and 2.50 g (19.10 mmol) of 3-chlorocyclohex-2-en-l-one in
10 mL of benzene was added. The resulting dark reaction mixture
was stirred at rt for 1 h, then was poured onto 50 mL of satd aq
NaHCO3 solution, and then was extracted with ether (3ꢃ75 mL).
The ether extracts were individually washed in series with one,
25 mL portion of water and 25 mL of brine. The ether extracts were
combined, dried (K2CO3) and evaporated. The residue was chro-
matographed on silica gel with pentane as eluent. Distillation
afforded 1.45 g (60% yield) of 8 as a pale-yellow oil, which
decomposed readily. Data for 8: bp 50 ꢀC (7 mmHg); Rf 0.50 (pen-
tane); tR 5.62 min, column C (75 ꢀC (5 min), 10 ꢀC/min, 150 ꢀC); 1H
NMR (390 MHz, CCl4):
d 1.66e1.93 (m, 2H, H2C(5)), 2.16e2.53 (m,
4H, H2C(4) and H2C(6)), 4.68 (s, 2H, H2C(7)), 6.15 (s, 1H, HC(2)); IR:
3020 (m), 2970 (m), 2960 (w), 1530 (m), 1480 (w), 1163 (w), 1105
(w), 1060 (w), 1035 (w), 925 (m), 900 (s), 765 (w), 760 (w) cmꢁ1
.
acetate, 5:1); 1H NMR (200 MHz, CDCl3):
d 0.87 (s, 3H, SiPh2CH3),
6.1.2. Preparation of 1-(trimethylsilylmethyl)-3-methylenecyclohex-
l-ene (9). To a solution of 1.44 g (11.2 mmol) of 8 in 22 mL of ether
was added 22.1 mL (16.8 mmol) of a 0.76 M solution of trime-
thylsilylmethylmagnesium chloride in THF and 0.60 g (0.11 mmol)
of Ni(dppp)Cl2. The resulting pale-orange solution was heated at
reflux for 5 h, then was quenched with 35 mL of satd aq NaHCO3
solution and was extracted with pentane (3ꢃ50 mL). The pentane
extracts were individually washed in series with one, 25 mL portion
of water and 25 mL of brine. The pentane layers were combined,
dried (K2CO3) and evaporated. The residue was chromatographed
on alumina (neutral, activity I) with pentane as the eluent. Distil-
lation afforded 1.72 g (85%) of 9 as a clear. colorless liquid. Data for
9: bp 90 ꢀC (15 mmHg); tR 11.53 min, column C (75 ꢀC (5 min),
1.56e1.63 (m, 3H, HC(5) and 2 HC(10)),1.95e2.15 (m,1H), 2.40e2.43
(m, 3H), 2.55e2.73 (m, 1H), 3.73 (t, J¼6.0, 1H, HC(20)), 6.18 (d, J¼1.9,
1H, HC(2)), 7.33e7.58 (m, 10H, arom. H); 13C NMR (50 MHz, CDCl3):
d
ꢁ3.60 (Si(CH3)3), 31.20, 37.29, 39.82, 42.44, 60.04,128.15 (para-Cs),
128.35 (C(2)), 130.21 (ortho-C), 134.28 (meta-Cs), 135.77 (ipso-C),
158.31 (C(3)), 197.31 (C(1)); IR: 3070 (w), 3045 (w), 3005 (w), 2930
(w), 2910 (w), 2885 (w), 1673 (s), 1610 (m), 1428 (m), 1336 (w), 1289
(w),1255 (s),1234 (w),1210 (w),1119 (s),1111 (s),1100 (m),1081 (m),
995 (w), 951 (w), 883 (w), 806 (w), 706 (m) cmꢁ1; MS (70 eV): m/z
370 (Mþ, 1), 357 (16), 356 (11), 355 (42), 295 (45), 294 (25), 293
(100), 277 (10), 219 (12), 217 (34), 200 (16), 199 (90), 198 (12), 197
(60),195 (10),181 (20),171 (44),157 (14),155 (18),137 (25),121 (17),
115 (35), 105 (20), 91 (27), 77 (32), 67 (16), 65 (14), 45 (18), 39 (14).
Anal. Calcd for C21H23ClO2Si: C, 68.00; H, 6.25; Cl. 9.56. Found: C,
67.82; H, 6.38; Cl, 9.64.
10 ꢀC/min, 150 ꢀC); 1H NMR (200 MHz, CDCl3):
d 0.00 (s, 9H,
Si(CH3)3), 1.52 (s, 2H, CH2Si(CH3)3), 1.66 (q, J¼6.0, 2H, H2C(5)), 1.99
(t, J¼6.0, 2H, H2C(6)), 2.24 (t of t, J¼l.6 and 6.4, 2H, H2C(4)), 4.51 (d,
J¼4.8, 2H, ]CH2), 5.77 (s, 1H, HC(2)); 13C NMR (50 MHz, CDCl3):
6.2.2. Preparation of 2-(30-chloro-50-methylidenecyclohex-30-enyl)
ethan-1-yl diphenylmethylsilyl ether (14). To a solution of 3.09 g
(8.64 mmol) of methyltriphenylphosphonium bromide in 40 mL of
benzene was added 4.86 mL (8.21 mmol) of a 1.69 M solution of
potassium tert-amylate in benzene. The yellow suspension was
heated at reflux for 1 h. The resulting yellowish-orange suspension
was cooled to rt and 2.14 g (5.76 mmol) of 13 in 5 mL of benzene was
added. The dark reaction mixture was stirred at rt for 0.5 h, then was
poured onto 25 mL of satd aq NaHCO3 solution and was extracted
with ether (3ꢃ75 mL). The ether extracts were individually washed
with one, 30 mL portion of water and 30 mL of brine. The ether
extracts were combined, dried (K2CO3) and evaporated. The residue
was chromatographed on silica gel with pentane/ether, 30:1 as el-
uent to give 1.76 g (83% yield) of 14 as a cloudy, viscous oil, which
decomposed readily. Data for 14: Rf 0.30 (pentane/ether, 30:1); NMR
d
ꢁ0.46 (Si(CH3)3), 24.01, 29.46, 31.15, 32.22. 107.05 (eCH2), 123.64
(C(2)), 142.17 (C(l)), 145.01 (C(3)); IR: 3080 (w), 3010 (m), 2960 (s),
2830 (w), 1632 (m), 1600 (w), 1655 (w), 1440 (w), 1429 (w), 1417
(w), 1365 (w), 1250 (s), 1215 (m), 1210 (m), 1165 (m), 1133 (w), 1105
(w), 1054 (w), 890 (s), 855 (s) cmꢁ1; MS (70 eV): m/z 180 (Mþ, 9),
147 (14), 89 (50), 75 (43), 73 (100), 59 (35), 58 (10), 45 (13), 43 (35).
HRMS: Calcd for C11H20Si: m/z 180.13342. Found, 180.13365.
6.2. Synthesis and cyclization of 2-[50-methylidene-30-
(trimethylsilylmethyl)cyclohex-30enyl]ethanal (model system
10). 2-(30-keto-50-chlorocyclohex-40-enyl)ethanoyl chloride
(12)
A mixture of 2.042 g (10.85 mmol) of 11 in 9.51 mL (109 mmol)
of oxalyl chloride was heated to a gentle reflux for 0.5 h. The excess
oxalyl chloride was removed under reduced pressure. The resulting
yellow oil, which decomposed rapidly, was used without further
purification. Analytical data were obtained from a distilled sample.
(200 MHz, CDCl3):
d
0.62 (s, 3H, SiPh2CH3), 1.51e1.66 (q, J¼6.5, 2H, 2
HC(2)), 1.90e2.16 (m, 3H), 2.28e2.36 (m, 2H), 3.73 (t, J¼6.2, 2H, 2
HC(1)), 4.75 (d, J¼9.5, 2H, ]CH2), 6.24 (s, 1H, HC(40)), 7.33e7.59 (m,
10H, arom. H); IR: 3670 (m), 3610 (m), 3015 (s), 2970 (m), 2930 (m),

7712
S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
2420 (m) (sh), 2390 (m),1530 (m),1475 (m),1425 (m),1200 (s),1110
(m),1040 (m), 925 (m), 910 (s), 750 (s) cmꢁ1; MS (70 eV): m/z no Mþ,
217 (10), 215 (18), 200 (49), 198 (19), 197 (96), 181 (11), 156 (29), 155
(22), 154 (84), 139 (15), 137 (20), 126 (22), 121 (21), 119 (100), 105
(12), 93 (29), 92 (16), 91 (81), 77 (22).
added 203 mg (0.906 mmol) of 16 in 3 mL of THF. After stirring the
solution at rt for 5 min, a solution of 275 mg (1.089 mmol) of 1,10-
(azodicarbonyl)dipiperidine in 2 mL of THF was added to the reaction
mixture. The resulting dark-red solution was stirred at rt for 12 h.
During thistimethecolorfaded and a pale-yellow precipitateformed.
The reaction mixture was poured onto 10 mL of satd aq NaHCO3 so-
lution, followed by extraction with ether (3ꢃ25 mL). The ether ex-
tracts were individually washed in series with a one,10 mL portion of
water and 10 mL of brine. The ether extracts were combined, dried
(K2CO3) and evaporated. The residue was chromatographed on alu-
mina with pentane/ether, 20:1 as eluent to give 115 mg (60% yield) of
10 as a clear, colorless oil. Data for 10: Rf 0.35, silica gel (pentane/ether,
20:1). tR 15.14 min, column E (125 ꢀC (5 min), 10 ꢀC/min, 250 ꢀC); 1H
6.2.3. Preparation of 2-[50methylidene-30-(trimethylsilylmethyl)cy-
clohex-30-enyl]ethan-1-yl diphenylmethylsilyl ether (15). To a mix-
ture of 1.75 g (4.74 mmol) of 14 and 25 mg (0.047 mmol) of Ni(dppp)
Cl2 in 10 mL of ether was added 9.48 mL (7.11 mmol) of a 0.75 M
solution of trimethylsilylmethylmagnesium chloride in THF. The
resulting pale-orange solution was heated to reflux for 12 h. The
dark-yellow reaction mixture containing a finely dispersed pre-
cipitate was poured onto 25 mL of satd aq NaHCO3 solution followed
by extraction with ether (3ꢃ25 mL). The ether extracts were in-
dividually washed with one, 25 mL portion of water and 25 mL of
brine. The ether extracts were combined, dried (K2CO3), and evap-
orated. The residue was chromatographed on Al2O3 (neutral, activity
I) with pentane/ether, 20:1 as eluent to give 1.767 g (89% yield) of 15
as a clear, colorless oil. Data for 15: Rf 0.45, silica gel (pentane/ether,
NMR (200 MHz, CDCl3): d
0.00 (s, 9H, Si(CH3)3), 1.51 (s, 2H, 2 HC(80)),
1.70e2.20 (m, 3H), 2.33e2.41 (m, 4H), 4.61 (d, J¼11.7, 2H, 2 HC(]
CH2)), 5.78 (s,1H, HC(40)), 9.77 (t, J¼1.7, 1H, HC(1)). 13C NMR (90 MHz,
CDCl3):
d
ꢁ1.16 (Si(CH3)3), 28.50, 28.91, 36.24, 27.47, 29.36 (C(2)),
107.88 (C(60)), 122.80 (C(40)), 139.40 (C(30)), 142.40 (C(60)), 201.95
(C(1)); IR: 1684 (w),1633 (m), 1600 (w),1425 (w),1410 (w),1373 (w),
1259(w)(sh),1249(s),1164(w), 887(m),852(s)cmꢁ1;MS(10 eV): m/
z 222 (Mþ,17), 207 (14),132 (13),117 (20),106 (15), 91 (20), 75 (14), 73
(100). HRMS: Calcd for C13H22OSi: m/z 222.14400. Found 222.14405.
20:1); 1H NMR (200 MHz, CDCl3):
d
ꢁ0.01 (s, 9H, Si(CH3)3), 0.63 (s,
3H, SiPh2CH3), 1.50e1.59 (m, 4H), 1.77e2.05 (m, 4H), 2.28e2.34 (m,
1H), 3.75 (t, J¼6.6, 2H HC(1)), 4.56 (br d, J¼13, 2H, ]CH2), 5.75 (s,1H,
HC(40)), 7.33e7.61 (m, 10H, arom. H); 13C NMR (50 MHz, CDCl3):
6.2.6. Cyclization of 2-[50-methylidene-30-(trimethylsilylmethyl)cy-
d
ꢁ3.06 (SiPh2CH3), ꢁ1.16 (Si(CH3)3), 28.62, 31.01 (C(10)), 36.60, 38.12,
clohex-30-enyl]-1-ethanal (10). To
a
suspension of 58 mg
38.46, 61.24 (C(1)), 106.80 (]CH), 122.75 (C(40)), 127.82 (para-Cs),
129.75 (ortho-C), 134.23 (meta-Cs), 136.10 (ipso-C), 140.49 (C(30)),
143.90 (C(50)); IR: 3065 (w), 3000 (m), 2900 (m) (sh), 2930 (m), 2870
(m), 1630 (w), 1589 (w), 1428 (m), 1255 (m), 1116 (s), 1100 (s) (sh),
1130 (s), 1021 (w), 996 (w), 960 (w), 891 (m), 695 (m) cmꢁ1; MS
(70 eV): m/z 420 (Mþ, 1), 410 (17), 397 (20), 396 (54), 395 (100), 333
(17), 332 (29), 331 (23), 319 (27), 318 (73), 317 (100), 257 (24), 256
(14), 255 (46), 209 (15), 206 (25),199 (13),198 (11),197 (46),196 (13),
195 (73), 193 (13), 190 (12), 181 (30), 180 (11), 165 (25), 163 (13), 152
(20), 151 (62), 120 (17), 119 (19), 105 (35), 91 (41), 89 (24), 79 (19), 77
(44), 75 (20), 74 (41), 73 (100), 59 (17), 45 (23). HRMS: Calcd for
C26H36OSi2: m/z 420.23046. Found, 420.23010.
(0.358 mmol) of FeCl3 in 5.0 mL of CH2Cl2 at ꢁ70 ꢀC was added
73 mg (0.328 mmol) of 10 in 1.5 mL of CH2Cl2. The resulting
reddish-orange solution was stirred at ꢁ70 ꢀC for 1.0 h and was
quenched with 5.0 mL of 1 M NaOH in MeOH. The orange solution
was allowed to warm to rt. The resulting orange suspension was
stirred at rt for 12 h, poured onto 20 mL of water, and extracted
with ether (3ꢃ25 mL). The ether extracts were individually washed
in series with one, 10 mL portion of water and one, 10 mL portion of
brine. The ether extracts were combined, dried (K2CO3) and evap-
orated. The residue was chromatographed on silica gel with hex-
ane/ethyl acetate, 5:1 as eluent to give 9.5 mg (20% yield) of 17 as
a white solid. Data for 17: mp 61e62 ꢀC; Rf 0.29 (hexane/ethyl ac-
etate, 5/1); tR 9.55 min, column E (125 ꢀC (5 min), 10 ꢀC/min to
6.2.4. Preparation of 2-[50-methylidene-30-(trimethylsilylmethyl)cy-
clohex-20-enyl]-l-ethanol (16). A solution of 1.745 g (4.15 mmol) of
15 and 4.15 mL (4.15 mmol) of a 1.0 M solution of sodium hydroxide
in MeOH in 84 mL of a THF/MeOH, 1:1 mixture was stirred at rt of
0.5 h. The reaction mixture was poured onto 25 mL of a satd aq
NaHCO3 solution and extracted with ether (3ꢃ50 mL). The ether
extracts were individually washed with one, 25 mL portion of water
and 25 mL of brine. The ether extracts were combined, dried
(K2CO3) and evaporated. The residue was chromatographed on
Al2O3 (neutral, activity 1) with hexane/ethyl acetate, 5:1 as eluent
to give 0.885 g (95% yield) of 16 as a clear, colorless oil. Data for 16:
Rf 0.28, silica gel (hexane/ethyl acetate, 5:1); 1H NMR (200 MHz,
250 ꢀC); 1H NMR (200 MHz, CDCl3):
d
1.30 (d, J¼13.7, 1H), 1.70 (d,
J¼8.3, 1H), 1.99e2.10 (m, 3H), 2.23e2.38 (m, 3H), 2.85 (d, J¼3.5, 1H,
eOH), 3.91e3.97 (m, 1H, HC(2)), 4.75 (d, J¼1.5, 1H, Htrans-C(10)),
4.90e4.92 (m, 3H, 2 HC(9) and Hcis-C(10)); 13C NMR (90 MHz,
CDCl3):
d 27.32, 33.85, 34.81, 37.82, 55.10, 68.80 (C(2)), 108.99
(C(10)), 110.99 (C(9)), 143.88 (C(6)), 145.99 (C(7)); IR: 3070 (w),
3015 (w), 3000 (w), 2975 (w), 2935 (m), 2850 (w), 2815 (w), 1645
(m), 1449 (w), 1431 (w), 1426 (w), 1391 (w), 1328 (w), 1285 (w),
1268 (w), 1250 (w), 1240 (w), 1140 (w), 1064 (s), 1032 (w), 1012 (m),
988 (w), 957 (w), 933 (w), 920 (w), 892 (s), 830 (w) cmꢁ1; MS
(70 eV): m/z 150 (Mþ, 9), 149 (100), 133 (8), 129 (7), 107 (20), 106
(84), 105 (18), 104 (8), 92 (15), 91 (85), 79 (17), 78 (10), 77 (18), 73
(24), 71 (12), 65 (12), 57 (28), 56 (11), 55 (17), 43 (17), 41 (33), 39
(19). HRMS: Calcd for C10H14O: m/z 150.10446. Found 150.10442.
CDCl3):
1H), 3.69 (t, J¼6.5, 2H, 2 HC(1)), 4.58 (d, J¼92, 2H, ]CH2), 5.76 (s,
1H, HC(40)); 13C NMR (50 MHz, CDCl3):
d
ꢁ0.01 (s, 9H, Si(CH3)3). 1.31e2.07 (m, 8H), 2.32e2.38 (m,
d
ꢁ1.21 (Si(CH3)3), 28.54,
31.04, 36.56, 38.08, 60.66 (C(1)), 106.95 (]CH2), 122.75 (C(40)),
140.29 (C(30)), 143.67 (C(50)), 175; IR: 3615 (w), 3070 (w), 3000 (w),
2950 (m), 2920 (m), 2905 (m), 2820 (w), 1632 m, 1425 (w), 1415
(w), 1372 (w), 1250 (s), 1165 (w), 1040 (w), 996 (w), 886 (w), 850 (s)
cmꢁ1; MS (10 eV): m/z 224 (Mþ, 50), 179 (12), 120 (12), 119 (63), 107
(22), 106 (62), 105 (14), 91 (15), 75 (20), 73 (100). HRMS: Calcd for
C13H24OSi: m/z 224.159648. Found, 224.159593.
6.3. Synthesis and cyclization of 2-[50-[70-13C]Methylidene-30-
(trimethylsilylmethyl)cyclohex-30-enyl]-1-ethanal (model
system 3). Preparation of 2-(30-chloro-50-[70-13C]
methylidenecyclohex-30-enyl)ethan-1-yl diphenylmethylsilyl
ether (18)
To 0.538 g (1.321 mmol) of methyltriphenylphosphonium iodide
(99%, 13C) and 1.708 g (1.982 mmol) of methyltriphenylphosphonium
bromide in 20 mL of benzene was added 1.95 mL (3.303 mmol) of
a 1.69 M solution of potassium tert-amylate in benzene. The yellow
suspension was heated at reflux for 1 h. The reaction mixture was
cooled to rt and a solution of 1.021 g (2.752 mmol) of 13 in 3 mL of
benzene was added. The resulting dark-yellow reaction mixture
6.2.5. Preparation of 2-[50-methylidene-30-(trimethylsilylmethyl)cy-
clohex-30-enyl]-1-ethanal (10). A solution of tert-butoxymagnesium
bromide was prepared by adding 0.103 mL (1.093 mmol) of tert-bu-
tanol to 0.376 mL (1.093 mmol) of a 2.91 M solution of methyl-
magnesium bromide in ether in 2.5 mL of THF. To this solution was

S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
7713
containing a heavy precipitate was stirred at rt for 0.5 h, then was
poured onto 20 mL of satd aq NaHCO3 solution and was extracted
with ether (3ꢃ50 mL). The ether extracts were individually washed
with one, 20 mL portion of water and one, 20 mL portion of brine. The
ether extracts were combined, dried (K2CO3) and evaporated. The
residue was chromatographed on silica gel with pentane/ether, 3:1 as
eluent to give 0.853 g (83% yield) of 18 as a cloudy, viscous oil, which
decomposed readily. Data for 18: Rf 0.30(pentane/ether, 3/1); 1H NMR
(w), 995 (w), 885 (m), 851 (s) cmꢁ1; MS (70 eV): m/z 225 (Mþ, 4),
224 (Mþ, 5), 120 (8), 119 (13), 107 (12), 106 (15), 105 (6), 92 (9), 91
(13), 75 (20), 74 (8), 73 (100), 59 (6), 45 (13). HRMS: Calcd for
C13H24OSi: m/z 224.15963. Found m/z 224.16003.
6.3.3. Preparation
of
2-[50-[70-13C]methylidene-30-(trimethylsi-
lylmethyl)cyclohex-30-enyl]-1-ethanal (3). This compound was pre-
pared in 51% yield by use of the same procedure as described above
in the preparation of 10. Data for 3: Rf 0.35 (pentane/ether, 20:1); tR
15.14 min, column E (125 ꢀC (5 min), 10 ꢀC/min, 250 ꢀC); 1H NMR
(200 MHz, CDCl3):
d
0.56 (s, 3H, SiPh2(CH3)3), 1.48 (q, J¼6.3, 2H, 2
HC(2)), 1.83e2.01 (m, 3H), 2.21e2.36 (m, 2H), 3.66 (t, J¼6.5, 2H, 2
HC(1)), 4.68 (d of d, J13CeH¼158 and JHeH¼8.2, 0.8H, HC (13C, 70)), 4.68
(d, J¼8.9, 1.2H, HC(70)), 6.17 (s, 1H, HC(40)), 7.28e7.52 (m, 10H, arom.
H); IR: 3670 (m), 3615 (m), 3600 (w), 3580 (w), 3020 (s), 2970 (m),
2960(w), 2930(m), 2420 (m), 2385(m), 2370 (w), 2350(w),1530 (m),
1480 (m), 1425 (m), 1190 (s), 1160 (w), 1100 (m), 1060 (w), 1040 (m),
925 (m), 900 (s), 750 (s) cmꢁ1; MS (70 eV): m/z no Mþ, 214 (12), 200
(18), 199 (100), 139 (11), 137 (13), 129 (17), 128 (30), 127 (39), 126 (2),
125 (10),120 (20),119 (24), 93 (19), 92 (59), 91 (76), 79 (15), 78 (51), 77
(35), 65 (16), 52 (12), 51 (18), 45 (11), 41 (13).
(200 MHz, CDCl3):
d
ꢁ0.05 (s, 9H, Si(CH3)3), 1.51 (s, 2H, 2 HC(80)),
1.73e2.16 (m, 3H), 2.30e2.40 (m, 4H), 4.61 (d of d, J13CeH¼156 and
JHeH¼11.9, 0.8H, ]13CH2), 4.61 (d, JHeH¼12.1, 1.2H, ]CH2), 5.77 (s,
1H, HC(40)), 9.76 (t, J¼1.8, 1H, HC(1)); 13C NMR (90 MHz, CDCl3):
d
ꢁ1.19 (Si(CH3)3), 28.46, 28.86, 36.20, 37.42, 49.34 (C(2)), 107.85 (C-
(
13C, 70)), 122.76 (C(40)), 139.36 (C(30)), 142.42 (d, J13CeC¼73, C (13C,
5)), 142.42 (C(50)), 201.90 (C(1)); IR: 3020 (w), 3000 (w), 2950 (m),
2820 (w), 2720 (w), 1722 (s), 1683 (w),1631 (m), 1600 (w),1580 (w),
1438 (w), 1425 (w), 1405 (w), 1372 (w), 1259 (w) (sh), 1250 (s), 1230
(w), 1210 (w), 1153 (w), 1133 (w), 1121 (w), 975 (w), 887 (m), 853 (s)
cmꢁ1; MS (70 eV): m/z 223 (Mþ, 6), 22 (Mþ, 6), 208 (5), 207 (7), 132
(6),118 (6),117 (11),106 (7),103 (6), 92 (10), 91 (17), 79 (5), 75 (12), 74
(9), 73 (100), 59 (9), 45 (18), 43 (5). HRMS: Calcd for C13H22OSi: m/z
222.14398. Found, 222.143831.
6.3.1. Preparation of 2-(50-[70-13C]methylidene-30-trimethylsilylme-
thylcyclohex-30-enyl)ethan-1-yl diphenylmethylsilyl ether (19). To
a mixture of 850 g (2.303 mmol) of 18 and 12 mg (0.023 mmol) of
Ni(dppp)Cl2 in 5.0 mL of THF was added 4.73 mL (3.46 mmol) of
a 0.73 M solution of trimethylsilylmethylmagnesium chloride in
THF. The resulting pale-orange solutionwas heated at reflux for 12 h.
The dark-yellow reaction mixture was poured onto 20 mL of satd aq
NaHCO3 solution and was extracted with ether (3ꢃ25 mL). The ether
extracts were washed with one, 20 mL portion of water and one,
20 mL portion of brine. The ether extracts were combined, dried
(K2CO3) and evaporated. The residue was chromatographed on
alumina (neutral, activity I) with pentane/ether, 20:1 and hexane/
ethyl acetate, 5:1 as eluent to give 770 mg (80% yield) of 19 as a clear,
colorless oil and 50 mg (10% yield) of 20 as a clear, colorless oil. Data
for 19: Rf 0.45, silica gel (pentane/ether, 20/1); 1H NMR (200 MHz,
6.3.4. Cyclization
of
2-[50-[70-13C]methylidene-30-(trimethylsi-
lylmethyl)cyclohex-30enyl]-1-ethanal (3). General procedures for
Et2AlCl, BF3$OEt2 and SnCl4. To a solution of 5 (1.1 equiv, 0.05 M) in
CH2Cl2 at ꢁ70 ꢀC was added 1 equiv of Lewis acid. The solution was
stirred at ꢁ70 ꢀC until complete reaction of 5 was observed and
quenched with excess 1 N NaOH in MeOH. For workup and iso-
lation of reaction products see the procedure as described above for
the cyclization of 10 with FeCl3.
FeCl3. See the procedure as described for the cyclization of 10
with FeCl3.
CDCl3):
d
ꢁ0.02 (s, 9H, Si(CH3)3), 0.62 (s, 3H, SiPh2CH3), 1.49e1.58
n-Bu4NþFꢁ. To a mixture of 5 (1.0 equiv, 0.05 M) and NaHCO3
(1.0 equiv) in CH2Cl2 at rt was added 1.0 equiv of 0.32 M solution of
n-Bu4NþFꢁ in THF. The reaction mixture was stirred at rt for 1.0 h
and was quenched with excess 1 N NaOH in MeOH. For workup and
isolation of reaction products see the procedure described above for
the cyclization of 10 with FeCl3.
(m, 4H), 1.76e2.02 (m, 4H), 2.27e2.32 (m, 1H), 3.74 (t, J¼6.5, 2H, 2
HC(1)), 4.56 (br d of d, J13CeH¼149 and JHeH¼13.8, 0.8H, ]13CH2),
4.56 (br d, J¼13.8, 1.2H, ]CH2), 5.74 (s, 1H, HC(40)), 7.29e7.59 (m,
10H, arom. H); 13C NMR (50 MHz, CDCl3):
d
ꢁ3.06 (SiPh2CH3), ꢁ1.15
(Si(CH3)3), 28.62, 31.02 (C(10)), 36.60, 38.13, 38.46, 61.24 (C(1)),
106.82 (]13CH2), 122.79 (C(40)), 127.81 (para-Cs), 129.74 (ortho-C),
134.82 (meta-Cs), 136.12 (ipso-C), 140.44 (C(30)), 143.88 (C(50)),
143.88 (d, J13CeC¼72, C(50)); IR: 3070 (w), 3050 (w), 3005 (w), 2955
(m), 2925 (m), 2910 (m), 2870 (m), 2820 (w), 1632 (w), 1590 (w),
1487 (w), 1429 (m), 1372 (w), 1250 (m), 1207 (w), 1165 (w), 1119 (s),
1110 (s),1091 (m),1177 (m),1021 (w), 996 (w), 950 (w), 887 (m), 855
(s), 850 (s) (sh), 800 (w), 701 (m) cmꢁ1; MS (70 eV): m/z no Mþ, 395
(1), 210 (7), 209 (33), 207 (7), 198 (8), 197 (42), 195 (6), 193 (11), 179
(6),154 (10),119 (5),105 (7), 91 (8), 85 (5), 75 (23), 74 (8), 73 (100), 71
(9), 59 (6), 57 (25), 56 (7), 55 (5), 45 (12), 43 (17), 41 (10). HRMS:
Calcd for C26H36OSi2: m/z 420.23047. Found: m/z 420.23053.
6.3.5. Preparation of rel-(1R,2S,4S)-6,7-[9-13C]-dimethylidenebicyclo
[2.2.2]octan-2-ol (4a) and rel-(1R,2R,4S)-6,7-[9-13C]-dimethylidene-
bicyclo[2.2.2]octan-2-ol (4b). Data for mixture isolated from FeCl3
promoted cyclization of 3. Data for 4a: white solid, mp 60e61 ꢀC; Rf
0.29 (hexane/ethyl acetate, 5:1); tR 9.55 min, column E (125 ꢀC
(5 min), 10 ꢀC/min to 250 ꢀC); 1H NMR (360 MHz, CDCl3):
d 1.30 (d,
J¼13.7, 1H), 1.69 (d, J¼8.1, 1H), 1.99e2.09 (m, 2H), 2.19 (s, 2H),
2.23e2.38 (m, 2H), 2.85 (d, J¼3.1, 1H, eOH), 3.92e3.96 (m, 1H,
HC(2)), 4.75 (d of br d, 1H, JHe13C¼155 and JHeH¼1.4, Htrans-C (13C,
9)), and 4.75 (d, J¼1.4, 1H, Htrans-C(10)), 4.91 (d of m, 1H, J¼136, 2
HC (13C, 9)), and Hcis-(13C, 9), 4.91 (m, 3H, 2 HC(10) and Htrans
-
6.3.2. Preparation
of
2-[50-[7-13C]methylidene-30-(trimethylsi-
C(10)); 13C NMR (90 MHz, CDCl3):
d 27.32, 33.85, 34.81, 37.92, 55.10,
lylmethyl)cyclohex-20-enyl]-1-ethanol (20). This compound was
prepared in 96% yield by use of the same procedure as described
above for the preparation of 16. Data for 20: Rf 0.28, silica gel
68.79 (C(2)), 109.00 (C (13C, 10)), 110.99 (C (13C, 9)), 143.88 (C(6)),
145.99 (C(7)); IR: 3590 (w), 3540 (w), 3065 (w), 3005 (m), 2985 (w),
2935 (s), 2850 (w), 2840 (w), 1645 (w), 1612 (w), 1480 (w), 1448
(w),1430 (w), 1425 (w),1391 (w), 1338 (w),1317 (w),1284 (w),1265
(w), 1235 (w), 1150 (w), 1140 (w), 1064 (s), 1030 (w), 1011 (m), 985
(w), 955 (w), 932 (w), 919 (w), 890 (s), 839 (w), 829 (w) cmꢁ1; MS
(70 eV): m/z 150 (Mþ, 9),150 (Mþ,1),108 (19),107 (88),106 (99),105
(14), 93 (11), 92 (51), 91(100), 79 (20), 78 (14), 73 (13), 41 (13), 39
(20). HRMS: Calcd for C10H14O: m/z 150.10446. Found, 150.10442.
(hexane/ethyl acetate, 5/1); 1H NMR (200 MHz, CDCl3):
d
ꢁ0.00 (s,
9H, Si(CH3)3), 1.37e2.07 (m, 8H), 2.32e2.38 (m, 1H), 3.69 (t, J¼6.8,
2H, H2C(1)), 4.58 (d of d, J13CeH¼156 and JHeH¼9.1 0.8H,
HC(]13CH2)), 4.58 (d, JHeH¼9.2, 1.2H, HC(]CH2)), 5.75 (s, 1H,
HC(40)); 13C NMR (50 MHz, CDCl3):
d
ꢁ1.12 (Si(CH3)3), 25.25, 31.10
(C(10)), 36.60, 38.10, 38.61, 60.79 (C(1)), 106.98 (]13CH2), 122.79
(C(40)), 140.32 (C(30)), 143.69 (d, J13CeC¼72, C(50)), 143.69 (C(50)); IR:
3610 (w), 3000 (w), 2950 (m), 2920 (m), 2900 (m), 2820 (w), 1630
(m), 1410 (w) (br), 1370 (w), 1257 (w) (sh), 1248 (s), 1164 (m), 1089
6.3.6. Preparation of 2-[20-(phenyldimethylsilyl)-30-cyclohexenyl]-1-
ethanal (model system 5). Preparation of 1-oxabicyclo[3.2.2]non-5-

7714
S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
ene-2-one (23). A flame-dried, 300 mL flask equipped with a mag-
netic stir bar and nitrogen inlet tube was charged with the ketone 22
(3.57 g, 29.26 mmol), Na2CO3 (12.40 g, 117.0 mmol, 4 equiv) and
CH2Cl2 (180 mL). The mixture was allowed to stir and a solution of
m-chloroperbenzoic acid (5.05 g, 29.26 mmol) in CH2Cl2 (20 mL)
was added by syringe. The reaction mixture was stirred at room
temperature for 12 h. The reaction mixture was poured into H2O
(150 mL) and was extracted with CH2Cl2 (150 mL). The organic ex-
tracts were washed with satd aq NaHCO3 solution (2ꢃ100 mL) and
brine. The organic layers were dried (MgSO4) and concentrated to
leave a yellow oil. Purification was accomplished by silica gel col-
umn chromatography (hexane/ethyl acetate, 3:2) to afford 3.52 g
(87%) of 23 as a colorless oil. An analytical sample was obtained by
Kugelrohr distillation. Data for 23: bp 145 ꢀC (3 mmHg, air bath); 1H
in one portion. The reaction was quenched after 2 h by the addition
of 1 mL of water followed by 1 mL of 15% aq NaOH and 3 mL of
water. A gray precipitate formed, which was filtered off through
Celite with the aid of diethyl ether. The ethereal filtrate was evap-
orated to leave a yellow oil. Purification was accomplished by silica
gel column chromatography (hexane/ethyl acetate, 92:8) to afford
0.32 g (72%) of 25 as a colorless oil. An analytical sample was
obtained by Kugelrohr distillation. Data for 25: bp 200 ꢀC
(0.2 mmHg, air bath); 1H NMR: (300 MHz, CDCl3)
d 7.55 (m, 2H, Ph),
7.37 (m, 3H, Ph), 5.62 (m, 2H, HC(30), HC(40)), 3.58 (t, J¼7, 2H,
H2C(2)), 1.20e2.00 (m, 9H), 0.35 (s, 6H, (SiPh(CH3)2)); 13C NMR:
(75.5 MHz, CDCl3)
d 138.40 (Ph), 133.77 (Ph), 128.91 (Ph), 127.66
(Ph), 126.42 (C(30)), 124.27 (C(40)), 60.86 (C(1)), 38.05 (C(2)), 31.64
(C(10)), 28.75 (C(20)), 25.77 (C(50)), 21.46 (C(60)), ꢁ3.85 (SiPhCH3),
ꢁ4.14 (SiPhCH3); IR: (CCl4) 3638 (m), 3071 (w), 3021 (m), 2930 (s),
2841 (m), 1428 (m), 1250 (s), 1111 (s), 1061 (m) cmꢁ1; MS: (70 eV)
m/z 260 (Mþ, 1), 246 (1), 245 (3), 182 (5), 137 (22), 136 (14), 135
(100),108 (19),107 (11), 80 (65), 79 (24), 43 (12), 28 (15). Anal. Calcd
for C16H24OSi: C, 73.78; H, 9.29. Found: C, 73.74; H: 9.21.
NMR: (300 MHz, CDCl3):
HC(5)), 4.84 (m,1H, HC(7)), 2.80 (m, 2H, H2C(3)), 2.61 (m,1H, HC(4)),
1.7e2.4 (m, 4H); 13C NMR: (75.5 MHz, CDCl3)
171.85 (C(1)), 139.38
d
6.59 (t, J¼8, 1H, HC(6)), 6.36 (t, J¼8, 1H,
d
(C(4)), 129.33 (C(5)), 70.30 (C(6)), 40.84 (C(2)), 27.61 (C(3)), 26.56
(C(7)), 23.25 (C(8)); IR: (CCl4) 3051 (w), 2938 (m), 2880 (w), 1784
(w),1728 (s), 1387 (m),1371 (m), 1210 (m), 1181 (s),1022 (s), 963 (m)
cmꢁ1: MS: (70 eV) m/z 139 (Mþ, 1.5), 138 (7), 110 (10), 109 (11), 94
(14), 91 (11), 82 (22), 81 (16), 79 (100), 77 (33), 67 (38), 53 (27). Anal.
Calcd for C8H10O: C, 69.54; H, 7.30. Found: C, 69.32; H, 7.40.
6.3.9. 2-[20-(Phenyldimethylsilyl)-30-cyclohexenyl]-1-ethanal (5). A
flame-dried, 100-mL flask equipped with a magnetic stir bar and
nitrogen inlet tube was charged with pyridine (2.99 mL, 36.9 mmol,
12 equiv) and methylene chloride (20 mL). The mixture was cooled
to 0 ꢀC and chromium (III) oxide (1.84 g, 18.43 mmol, 6 equiv) was
added in portions to give a reddish-brown solution. After 15 min
a solution of alcohol 25 (0.80 g, 3.07 mmol) in CH2Cl2 (15 mL) was
added by syringe and the black mixture was stirred for 2 h. The
reaction mixture was filtered through Celite with the aid of ether.
The ethereal extracts were washed with 5% aq NaOH (150 mL) and
brine. The organic layers were dried (MgSO4) and the solvent
evaporated to leave a yellow oil. Purification was accomplished by
silica gel column chromatography (hexane/ethyl acetate, 93:7) to
afford 664 mg (84%) of 5 as a colorless oil. An analytical sample was
obtained by Kugelrohr distillation. Data for 5: bp 180 ꢀC (0.1 mmHg,
6.3.7. 2-[20-(Phenyldimethylsilyl)-30-cyclohexenyl]acetic acid methyl
ester (24). Lithium (dispersion in mineral oil (30%), 2.66 g,
115.2 mmol, 20 equiv) was washed with hexane (2ꢃ20 mL) and
suspended in THF (80 mL). A solution of phenyldimethylsilyl chlo-
ride (19.68 g,115.2 mmol, 20 equiv) inTHF (20 mL) was added to the
reaction mixture via syringe at 0 ꢀC and allowed to stir for 15 min.
Gilman titration gave a 0.71 M solution. A flame-dried, 250 mL flask
equipped with a magnetic stir bar and nitrogen inlet tube was
charged with CuI (3.00 g, 15.93 mmol, 1.1 equiv) and THF (75 mL).
The mixture was cooled to 0 ꢀC and a solution of PhMe2SiLi in THF
(45.0 mL, 0.71 M, 28.9 mmol, 2.2 equiv) was added by syringe. A
solution of the lactone, 23 (2.00 g, 14.5 mmol) in THF (20 mL) was
added by syringe to the red reaction mixture at 0 ꢀC. After 2 h at 0 ꢀC
the reaction mixture was quenched by the addition of water
(30 mL). The mixture was acidified by addition of 100 mL of a 10% aq
oxalic acid solution. The mixture was extracted with ethyl acetate
(3ꢃ100 mL) and the organic extracts were dried (MgSO4) and con-
centrated to a yellow oil. Initial purification of the acid was ac-
complished by silica gel column chromatography (hexane/ethyl
acetate, 19:1/1:1). Esterification of the acid with diazomethane
(40 mmol) afforded the methyl ester. Purification of the ester by
silica gel column chromatography (hexane/ethyl acetate, 93:7) gave
3.19 g (77%) of 24 as a colorless oil. An analytical sample was
obtained by Kugelrohrdistillation. Data for 24: bp 150 ꢀC (0.1 mmHg,
air bath); 1H NMR: (300 MHz, CDCl3)
(m, 2H, Ph), 7.36 (m, 3H, Ph), 5.61 (m, 2H, CH]CH), 2.37 (m, 3H),
d
9.64 (t, J¼1.8, 1H, CHO), 7.53
1.94 (m, 2H), 1.58 (m, 2H), 1.36 (m, 1H), 0.36 (s, 6H, (SiPh(CH3)2));
13C NMR: (75.5 MHz, CDCl3)
d 202.87 (C(1)), 137.83 (Ph), 133.80
(Ph), 129.08 (Ph), 127.75 (Ph), 125.97 (C(30)), 124.21 (C(40)), 49.37
(C(2)), 31.78 (C(10)), 27.24 (C(20)), 25.83 (C(50)), 21.44 (C(60)), ꢁ3.84
(SiPhCH3), ꢁ4.19 (SiPhCH3); IR: (CCl4) 3023 (m), 2926 (m), 2840
(m), 1727 (s), 1428 (m), 1250 (m), 1113 (m), 833 (m) cmꢁ1; MS:
(70 eV) m/z 181 (Mþ, 0.46),180 (1.4),179 (7), 163 (7),137 (6),136 (9),
135 (64), 81 (7), 80 (100), 79 (13), 75 (11), 43 (13). Anal. Calcd for
C16H22OSi: C, 74.36; H, 8.58. Found: C, 74.29; H, 8.63.
6.3.10. Preparation of authentic samples of 5-hydroxybicyclo[2.2.2]
oct-2-ene (6a and 6b). A flame-dried, 25-mL flask equipped with
a magnetic stir bar and nitrogen inlet was charged with a solution
of ketone 22 (0.5 g, 4.09 mmol) in EtOH (10 mL). The mixture was
cooled to 0 ꢀC, whereupon NaBH4 (155 mg, 4.09 mmol, 4 equiv) was
added and the mixture was stirred for 2 h at 0 ꢀC. The mixture was
poured into 1 M HCl (50 mL) and was extracted with diethyl ether
(3ꢃ50 mL). The combined ethereal extracts were washed with satd
aq NaHCO3 and brine was dried (MgSO4). The solvent was evapo-
rated to leave a white solid. Separation of the diastereomers was
accomplished by radial chromatography (4 mm, ethyl acetate/
hexane as eluent) to afford 80 mg of the minor diastereomer 6b and
170 mg of the major diastereomer 6a. Data for 6a (proximal): bp
air bath); 1H NMR: (300 MHz, CDCl3)
d 7.53 (m, 2H, Ph), 7.36 (m, 3H,
Ph), 5.59 (m, 2H, H-C(30,40)), 3.62 (s, 3H, (CO2CH3)), 2.78 (m, 2H),1.92
(m, 2H), 1.57 (m, 2H), 1.36 (m, 2H), 0.35 (s, 6H, (SiPh(CH3)2); 13C
NMR: (75.5 MHz, CDCl3)
d 173.60 (C(1)), 138.02 (Ph), 133.88 (Ph),
128.97 (Ph), 127.67 (Ph), 126.02C(30)), 123.99 (C(40)), 51.29 (C(3)),
39.71 (C(2)), 31.59 (C(10)), 29.60, 25.29, 21.25, ꢁ3.96 (Si(CH3)2); IR:
(CCl4) 3071(w), 3022 (w), 2953 (m), 2842 (w), 1738 (s), 1435 (m),
1428 (m),1250 (m),1167 (m),1113 (m), 831 (m) cmꢁ1; MS: (70 eV) m/
z 290 (Mþ,1), 289 (3), 288 (14), 209 (35), 208 (78), 207 (50),194 (10),
193 (64),177 (39),151 (63),137 (20),136 (59),135 (100),134 (34),121
(27),1119 (29), 118 (76), 117 (55), 89 (55). Anal. Calcd for C17H24O2Si:
C, 70.78; H, 8.39. Found: C, 70.82; H: 8.40.
160 ꢀC (100 mmHg, air bath): 1H NMR: (300 MHz, CDCl3)
d 6.43 (t,
6.3.8. 2-[20-(Phenyldimethylsilyl)-30-cyclohexenyl]-1-ethanol (25). A
flame-dried, 100-mL flask equipped with a magnetic stir bar and
nitrogen inlet tube was charged with the ester 24 (0.50 g,
1.75 mmol) and diethyl ether (45 mL). The mixture was cooled to
0 ꢀC and lithium aluminum hydride (66 mg, 1.75 mmol) was added
J¼7.3, 1H, HC(2)), 6.10 (t, J¼7.3, 1H, HC(3)), 3.91 (s, 1H, OH), 2.71 (m,
1H), 2.55 (m, 1H), 1.95 (dd, 1H), 1.06e1.49 (m, 6H); 13C NMR:
(75.5 MHz, CDCl3)
37.65C(1), 30.00C(4), 23.92C(8), 21.75C(7). Data for 6b (distal): bp
160 ꢀC (100 mmHg, air bath); 1H NMR: (300 MHz, CDCl3)
6.24 (t,
d 136.70C(3), 129.62C(2), 70.41C(6), 39.11C(5),
d

S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
7715
J¼6.8, 1H, HC(2)), 6.16 (t, J¼6.9, 1H, HC(3)), 3.81 (s, 1H, OH), 2.50 (t,
J¼3, 2H), 1.07e2.03 (m, 7H); 13C NMR: (75.5 MHz, CDCl3)
analytical sample was obtained by Kugelrohr distillation. Data
from 29: bp 110 ꢀC (5 mmHg, air bath); 1H NMR: (300 MHz,
d
135.33C(3), 131.91C(2), 69.27C(6), 37.65C(1), 35.56C(5), 29.93C(4),
CDCl3)
2.18 (m, 2H, H2C(3)), 1.2e1.9 (m, 8H); 13C NMR: (75.5 MHz, CDCl3)
176.41 (C(2)), 84.91 (C(8)), 44.54 (C(3)), 35.63 (C(4)), 29.97,
28.07, 25.08, 23.83; IR: (CCl4) 2944 (s), 2863 (m), 1788 (s), 1447
d
3.73 (td, J¼4, 11, 1H, HC(8)), 2.44 (dd, J¼6, 16, 1H, HC(4)),
25.91C(8), 17.16C(7). These data matched the literature values.42
d
6.3.11. Proof of configuration of 2-[20-(phenyldimethylsilyl)-30-cyclo-
hexenyl]-1-ethanal (5). Preparation of 2-[20-(phenyldimethylsilyl)cy-
clohexyl]acetic acid methyl ester (26). An oven-dried, 100 mL, 3-
necked flask equipped with a magnetic stir bar was charged with
24 (1.25 g, 4.33 mmol), EtOH (50 mL), and 5% Pd/C (50 mg). The
mixture was stirred under 1 atm of hydrogen for 12 h. The solvent
was removed by evaporation. Purification was accomplished by
silica gel column chromatography (hexane/ethyl acetate, 92:8) to
afford 0.90 g (80%) of 26 as a colorless oil. Data for 26: 1H NMR:
(m), 1210 (s), 1186 (s), 1075 (s), 1036 (s), 930 (m), 818 (m) cmꢁ1
;
MS: (70 eV) m/z 140 (Mþ, 2), 139 (2), 121 (2), 97 (s), 96 (18), 81
(20), 79 (12), 69 (6), 68 (39), 67 (100), 66 (15). These data
matched the literature values.41
6.4. General procedures for the cyclization of model system 5
(300 MHz, CDCl3)
(CO2CH3)), 2.4e1.0 (m, 12H), 0.34 (s, 3H, (SiPhCH3)), 0.26 (s, 3H,
(SiPhCH3)); 13C NMR: (75.5 MHz, CDCl3)
173.51 (C(1)),139.67 (Ph),
133.62 (Ph), 128.76 (Ph), 127.78 (Ph), 51.28 (C(2)), 41.64, 36.10,
29.85, 27.92, 27.30, 25.86, ꢁ2.65 (SiPhCH3), ꢁ3.94 (SiPhCH3).
d
7.49 (m, 2H, Ph), 7.30 (m, 3H, Ph), 3.58 (s, 3H,
6.4.1. SnCl4, BF3$OEt2, SiCl4, Et2AlCl, CF3CO2H, CF3SO3H. A magneti-
cally stirred solution of 5 (0.05 M), cyclododecane (1.0 equiv, in-
ternal standard), and dichloromethane were cooled to ꢁ78 ꢀC. At
ꢁ78 ꢀC, 1.05 equiv of the Lewis acid was added by syringe. The
reaction was stirred at ꢁ78 ꢀC until the complete reaction was
observed. The reaction mixture was then quenched at ꢁ78 ꢀC by
the addition of 5 equiv of a 1 N NaOH/MeOH solution and sub-
sequent stirring at room temperature for 2 h. The mixture was
extracted with ethyl acetate and washed with water (5 mL). The
aqueous layer was back extracted with ethyl acetate (2ꢃ2 mL) and
the combined organic extracts were passed through a short Florisil
column (5ꢃ75 mm). The organic layer was analyzed by ‘on-column’
capillary GC (column B). Program: 60 ꢀC (2 min), 10 ꢀC/min, 90 ꢀC
(5 min), 20 ꢀC/min, 250 ꢀC (5 min). Final ratios and yields were
calculated based on independently determined response factors
d
6.3.12. 2-[20-(Fluorodimethylsilyl)cyclohexyl]acetic acid methyl ester
(27). A flame-dried, 50-mL, three-necked flask equipped with
a magnetic stir bar and nitrogen inlet tube was charged with a so-
lution of the silane 26 (850 mg, 2.96 mmol) and CH2Cl2 (15 mL). The
mixture was cooled to 0 ꢀC and HBF4$2Et2O (312 mg, 3.55 mmol,
1.2 equiv) was added by syringe. The mixture was warmed to room
temperature and stirred for 15 h. The reaction mixture was diluted
with diethyl ether (50 mL) and was poured into satd aq NaHCO3
solution. The ethereal extract was washed with brine, dried
(MgSO4) and evaporated to leave 650 mg (95%) of 27 as a light
relative to the cyclododecane as internal standard. GC:
5
yellow oil. Data for 27: 1H NMR: (300 MHz, CDCl3)
d
3.65 (s, 3H,
tR¼15.44 min, response factor¼1.18; 6a tR¼8.18 min, response
(CO2CH3)), 2.57 (dd, J¼4, 15, 1H), 2.10 (m, 1H), 1.9e0.9 (m, 10H), 0.22
factor¼0.616; 6b tR¼8.62 min, response factor¼0.641.
(2d, J¼6, 6H, SiPh(CH3)2); 13C NMR: (75.5 MHz, CDCl3)
d 173.27
(C(1)), 51.37 (C(2)), 41.12, 34.77, 33.59, 31.42, 26.75, 26.08, 25.77,
6.4.2. ZrCl4. In a flame-dried, 5-mL flask, equipped with a magnetic
stir bar and a nitrogen inlet tube, was placed the Lewis acid
(1.05 equiv) and cyclododecane (1.0 equiv, internal standard) in
dichloromethane. The suspension was cooled to ꢁ78 ꢀC and a so-
lution of 5 was added. The reaction mixture was stirred at ꢁ78 ꢀC
until complete, quenched at ꢁ78 ꢀC by the addition of 5 equiv of
1 M NaOH/MeOH and subsequent stirring at room temperature for
2 h. The mixture was extracted with ethyl acetate (3 mL) and
washed with water (5 mL). The aqueous layer was back-extracted
with ethyl acetate (2ꢃ2 mL) and the combined organic extracts
were passed through a short Florisil column (5ꢃ75 mm). The or-
ganic layer was analyzed by ‘on-column’ capillary GC.
ꢁ2.03 (JCeF¼15), ꢁ2.87 (JCeF¼15).
6.3.13. 2-(20-Hydroxycyclohexyl)acetic acid methyl ester (28). An
oven-dried, 50-mL, three-necked flask equipped with a magnetic
stir bar and nitrogen inlet tube was charged with the fluorosilane
27 (464 mg, 2.00 mmol), DMF (25 mL), KF (232 mg, 4.0 mmol,
2 equiv) and m-chloroperbenzoic acid (1.035 g, 6.0 mmol, 3 equiv).
After stirring the mixture for 10 min at room temperature the
mixture became warm and was left to stir for 4 h at room tem-
perature. The reaction mixture was poured into water (50 mL) and
extracted with ether (3ꢃ25 mL). The ethereal extracts were washed
with satd aq NaHSO4 (50 mL) and satd aq NaHCO3 (3ꢃ50 mL). The
organic layer was dried over MgSO4 and the solvent was evaporated
to leave a light yellow oil. Purification was accomplished by silica
gel column chromatography using successively 4:1 and then 1:1
hexane/ethyl acetate as the eluent to give 280 mg (81%) of 28 as
6.4.3. n-Bu4NþFꢁ. A 0.05 M solution of 6 in THF was added to
3.0 equiv of n-Bu4NþFꢁ buffered with 3.0 equiv of NaHCO3. The
mixture was heated to reflux for 6 h. After cooling to room tem-
perature a 0.5 mL aliquot was removed and washed with 1 mL of
satd aq NaHCO3 solution and water (1 mL). The organic layer was
extracted with ethyl acetate (3ꢃ2 mL). The combined organic ex-
tracts were passed through a short Florisil column (5ꢃ75 mm) and
analyzed by ‘on-column’ capillary GC.
a colorless oil. Data for 28: 1H NMR: (300 MHz, CDCl3)
d 3.62 (s, 3H
(CO2CH3)), 3.15 (m, 1H, HC(20)), 2.66 (dd, J¼5.15, 1H, HC(2)), 2.35 (s,
1H, OH), 2.13 (dd, J¼7, 14, 1H, HC(2)), 1.94 (m, 1H, HC(10)), 1.64 (m,
4H), 1.19 (m, 4H); 13C NMR: (75.5 MHz, CDCl3)
d 174.53 (C(1)), 74.56
(C(20)), 57.58 (C(2)), 42.46, 38.32, 35.78, 31.37, 25.38, 24.89.
6.4.4. Procedure for calculation of relative response factors. Stock
solutions of known molarity of cyclododecane, proximal di-
astereomer (6a) and distal diastereomer (6b) were prepared by
dissolving known amounts of each in separate volumetric flasks
and diluting to the mark with CH2Cl2. A portion of each of the so-
lutions containing the epimeric alcohols were then separately
mixed with an equimolar amount of the cyclododecane stock so-
lution. The resulting solutions were then injected onto GC column B
and the area of the alcohol peak was divided by the area of the
cyclododecane peak. The injections were repeated in triplicate and
averaged to give a response factor relative to cyclododecane. The
6.3.14. trans-Hexahydro-2(3H)-benzofuranone (29). A flame-dried,
25-mL flask equipped with a magnetic stir bar and reflux con-
denser was charged with hydroxy ester, 28 (150 mg, 0.87 mmol),
benzene (10 mL), and p-toluenesulfonic acid (25 mg). The mix-
ture was heated at reflux for 3 h, during, which time methanol
was removed by azeotropic distillation. The reaction mixture was
washed with satd aq NaHCO3 and the solvent was removed by
evaporation. Purification of the resulting oil was accomplished by
silica gel column chromatography on silica gel (hexane/ethyl
acetate, 4:1) to afford 90 mg (74%) of 29 as a colorless oil. An

7716
S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
entire procedure was repeated twice more and the numbers
obtained averaged to give the final relative response factors.
H 1.0 ꢁ4.1829122004 ꢁ1.7467826564 ꢁ0.1374283299
H 1.0 ꢁ5.0514757129 ꢁ0.2088679943 ꢁ0.2726719473
H 1.0 ꢁ4.7202673407 ꢁ0.8924293225 1.3244917997
C 6.0 ꢁ2.9753285714 1.8562633087 1.1850423305
H 1.0 ꢁ2.1178725323 2.5302872901 1.1084569633
H 1.0 ꢁ3.2268944375 1.7293741987 2.2434082164
H 1.0 ꢁ3.8252825810 2.3473478734 0.6988770319
C 6.0 ꢁ1.9747547897 0.3770154501 ꢁ1.3699585853
H 1.0 ꢁ1.0361002447 0.9391401034 ꢁ1.3661434390
H 1.0 ꢁ2.6870117503 0.9168085437 ꢁ2.0026619691
H 1.0 ꢁ1.7924535748 ꢁ0.6017593169 ꢁ1.8246387640
C 6.0 0.4448012967 ꢁ1.5126420254 3.4752696344
H 1.0 1.0818164073 ꢁ0.8891793510 4.1073870754
O 8.0 0.0538906898 ꢁ2.6466923119 4.0429014854
H 1.0 0.8705218510 1.2332028304 3.6246450903
H 1.0 0.0949944414 ꢁ2.5551282076 5.0340731482
H 1.0 ꢁ0.5529237896 ꢁ2.1584880328 7.1464569604
O 8.0 0.2153714893 ꢁ2.0354998448 6.5733816904
H 1.0 0.9807052562 ꢁ2.3294926073 7.0847851933
6.4.5. Electronic energies, Gibbs free energies, and Cartesiancoordinates
for calculated transition structures
TS-a1
E¼ꢁ872.714264
G(195 K)¼ꢁ872.409913
C 6.0 0.4368091839 ꢁ0.6264807398 ꢁ0.6893720498
C 6.0 ꢁ0.5724557795 ꢁ2.6587123862 0.4040993718
C 6.0 1.0241093939 ꢁ0.1749654226 0.6565121235
C 6.0 0.3144514037 ꢁ2.1534382257 ꢁ0.7551977615
H 1.0 1.3062016937 ꢁ2.6111993687 ꢁ0.7146126083
H 1.0 ꢁ0.1164008265 ꢁ3.5237439681 0.8971881195
H 1.0 1.1832506358 0.9110119238 0.6948495705
H 1.0 ꢁ0.1203295873 ꢁ2.4414827061 ꢁ1.7161052227
C 6.0 ꢁ0.9819665849 ꢁ0.0182761670 ꢁ0.8031516183
H 1.0 ꢁ1.4977590189 ꢁ0.4335846664 ꢁ1.6733167094
H 1.0 ꢁ0.9058097909 1.0648785635 ꢁ0.9331133896
SI 14.0 2.8229847500 ꢁ0.8458955211 1.0713560702
C 6.0 3.5082836161 0.2993732670 2.3887379086
H 1.0 3.5747930016 1.3308215529 2.0303703994
H 1.0 4.5142758137 ꢁ0.0276504041 2.6706937215
H 1.0 2.8875273225 0.2851622105 3.2903619174
C 6.0 2.7636976762 ꢁ2.6025500540 1.7398498976
H 1.0 2.4607713658 ꢁ3.3386818350 0.9908677021
H 1.0 2.0866299515 ꢁ2.6818963821 2.5968663014
H 1.0 3.7670624124 ꢁ2.8748048725 2.0844721366
C 6.0 3.7876665937 ꢁ0.7367557278 ꢁ0.5319277511
H 1.0 3.3939422717 ꢁ1.4235313279 ꢁ1.2875801499
H 1.0 4.8324838374 ꢁ1.0044151880 ꢁ0.3441717129
H 1.0 3.7643656677 0.2788281119 ꢁ0.9386406622
H 1.0 ꢁ1.5518410655 ꢁ2.9884937470 0.0452718805
C 6.0 ꢁ1.8605503205 ꢁ0.2833517426 0.4042123882
C 6.0 ꢁ0.7934070923 ꢁ1.6055858363 1.4681164037
H 1.0 ꢁ1.4719462680 ꢁ1.8738528790 2.2753772907
H 1.0 1.0501817520 ꢁ0.2576747378 ꢁ1.5154281730
H 1.0 ꢁ2.0095114852 0.5336607258 1.1101341971
O 8.0 ꢁ2.9556965213 ꢁ0.9924650626 0.1576565253
C 6.0 0.1143132096 ꢁ0.5925105119 1.7068738499
H 1.0 0.0545500683 ꢁ0.0309002520 2.6392084600
H 1.0 ꢁ3.6328940523 ꢁ0.8576505736 0.8792792424
H 1.0 ꢁ5.5497327733 ꢁ0.2388048931 1.8069629335
O 8.0 ꢁ4.6310365337 ꢁ0.4022444878 2.0579524887
H 1.0 ꢁ4.6612572367 ꢁ0.8690160754 2.9028181250
TS-a2
TS-b1
E¼ꢁ757.189334
G(195 K)¼ꢁ756.935374
C 6.0 ꢁ2.9427756569 ꢁ2.0622451355 ꢁ0.5188670145
H 1.0 ꢁ3.2101322331 ꢁ2.5064662551 0.4425594369
H 1.0 ꢁ3.7833355550 ꢁ2.2118355145 ꢁ1.2078439704
C 6.0 ꢁ2.7653340415 ꢁ0.5938918279 ꢁ0.3974987901
H 1.0 ꢁ2.2278209804 ꢁ0.0498053576 ꢁ1.1740511636
O 8.0 ꢁ3.6937931788 0.0471660140 0.2339185418
H 1.0 ꢁ3.6118021957 1.0509679598 0.0736209432
H 1.0 ꢁ2.0607372877 ꢁ2.5453443223 ꢁ0.9352650431
H 1.0 ꢁ3.4390684994 3.1635810236 0.3667648544
O 8.0 ꢁ3.2194034792 2.4742657991 ꢁ0.2760376538
H 1.0 ꢁ3.4727992987 2.8280220729 ꢁ1.1400769737
C 6.0 ꢁ0.1737354416 ꢁ1.0700919756 0.4024754169
C 6.0 0.1435239897 ꢁ2.4677016912 0.6753177744
H 1.0 0.4108090189 ꢁ3.0222905167 ꢁ0.2306249941
H 1.0 ꢁ0.6536129074 ꢁ2.9846514075 1.2211816940
SI 14.0 1.7123811382 ꢁ2.5570324547 1.8197890625
C 6.0 3.1114186816 ꢁ1.6282733520 0.9793241547
H 1.0 2.8874602567 ꢁ0.5599193280 0.8984305213
H 1.0 3.3006595974 ꢁ2.0192548577 ꢁ0.0254260399
H 1.0 4.0325910921 ꢁ1.7345791799 1.5610229049
C 6.0 2.1149234259 ꢁ4.3793451255 2.0140818998
H 1.0 1.2808771254 ꢁ4.9216483486 2.4695410271
H 1.0 2.9918681307 ꢁ4.5000429140 2.6581936356
H 1.0 2.3387611638 ꢁ4.8376844262 1.0459421079
C 6.0 1.2612519834 ꢁ1.7628619286 3.4592787289
H 1.0 0.4348576030 ꢁ2.2942296959 3.9417899619
H 1.0 0.9654789644 ꢁ0.7180964154 3.3206346198
H 1.0 2.1201940503 ꢁ1.7835395668 4.1373275922
C 6.0 ꢁ1.1369441151 ꢁ0.3369575705 1.0331151504
H 1.0 ꢁ1.1549689351 0.7431263516 0.9209533925
H 1.0 ꢁ1.6932182253 ꢁ0.7617570083 1.8662222134
H 1.0 0.4196732606 ꢁ0.5738924354 ꢁ0.3672183402
E¼ꢁ872.717268
G(195 K)¼ꢁ872.413289
H 1.0 0.3162195211 ꢁ0.7892937744 0.0768247203
C 6.0 0.0512508130 ꢁ0.6144496108 1.1225350136
C 6.0 ꢁ0.0112707011 1.0930280477 2.9910626805
C 6.0 ꢁ1.6830483595 ꢁ0.7968925071 2.7838653887
C 6.0 ꢁ0.8649576603 ꢁ0.0146550657 3.5746324426
C 6.0 ꢁ1.4383952233 ꢁ0.9148933005 1.3569635978
C 6.0 0.3971945383 0.8222368969 1.5235619534
H 1.0 ꢁ1.0846288031 0.0484687717 4.6384919806
H 1.0 ꢁ0.1093518103 1.5213088513 0.8524367619
H 1.0 ꢁ0.5860789556 2.0216645289 3.0708592976
H 1.0 ꢁ2.4314618118 ꢁ1.4269915283 3.2618031166
H 1.0 ꢁ1.7046234820 ꢁ1.9241102437 1.0145080105
H 1.0 1.4696448201 0.9896802681 1.3906822150
C 6.0 0.8211685427 ꢁ1.6146060473 2.0162534340
H 1.0 1.8989132386 ꢁ1.4374021431 1.9325291232
H 1.0 0.6133493036 ꢁ2.6381924524 1.6926815531
SI 14.0 ꢁ2.6977765853 0.1950709508 0.3502729144
C 6.0 ꢁ4.3121446765 ꢁ0.7600409074 0.3173418192
TS-b2
E¼ꢁ757.191168
G(195 K)¼ꢁ756.935409
C 6.0 3.2978863346 1.0867330920 0.9782481766
H 1.0 4.1656082702 1.7359011044 0.8071661550
H 1.0 2.5224514448 1.6683666383 1.4790084403
C 6.0 2.8478241645 0.6265449785 ꢁ0.3604268155
H 1.0 3.4961601925 ꢁ0.0523886867 ꢁ0.9171070661
O 8.0 2.1005374069 1.4291320084 ꢁ1.0434007719
H 1.0 2.0205580522 1.1195110184 ꢁ2.0101485968
H 1.0 3.6172003668 0.2538994806 1.6040028867
H 1.0 1.1645266625 0.4697540809 ꢁ3.8642997234

S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
7717
O 8.0 1.9809481309 0.3951694911 ꢁ3.3509564162
H 1.0 ꢁ0.2575058709 0.8989186921 ꢁ0.6134671012
H 1.0 ꢁ0.3550054211 1.4066741342 1.1061489986
SI 14.0 ꢁ2.1105468936 ꢁ0.2522562115 0.5444938620
C 6.0 ꢁ2.1991299251 ꢁ1.6338967137 ꢁ0.7241220704
H 1.0 ꢁ1.4559177905 ꢁ2.4102632751 ꢁ0.5172897134
H 1.0 ꢁ2.0285119601 ꢁ1.2544018827 ꢁ1.7367138190
H 1.0 ꢁ3.1895745108 ꢁ2.0994588437 ꢁ0.7002650791
C 6.0 ꢁ3.3453910242 1.1090560219 0.1604390122
H 1.0 ꢁ3.2685278757 1.9227442352 0.8881404746
H 1.0 ꢁ4.3653933358 0.7135251299 0.1989765061
H 1.0 ꢁ3.1779602666 1.5224212692 ꢁ0.8388731905
C 6.0 ꢁ2.3540874043 ꢁ0.9116773349 2.2860190857
H 1.0 ꢁ2.2389747933 ꢁ0.1145724566 3.0272785667
H 1.0 ꢁ1.6368293413 ꢁ1.7053619101 2.5181486110
H 1.0 ꢁ3.3609837861 ꢁ1.3281573903 2.3908245741
C 6.0 1.3862099517 ꢁ1.1794369696 ꢁ0.1091461625
H 1.0 1.1639294419 ꢁ1.1597085872 ꢁ1.1743258381
C 6.0 2.8124605304 0.5566338435 ꢁ0.3033823934
H 1.0 3.4043246537 ꢁ0.2334886299 ꢁ0.7667509131
O 8.0 2.2081484660 1.3835802262 ꢁ1.0816138730
H 1.0 2.1836403237 1.0304706064 ꢁ2.0409331535
H 1.0 2.0492330823 ꢁ1.9646915640 0.2417505784
H 1.0 0.8820182954 ꢁ0.5683928283 1.8338948232
H 1.0 3.4382273132 0.2178986978 1.7150710786
H 1.0 1.4346109795 0.3280412024 ꢁ3.9122658776
O 8.0 2.2047498291 0.2484516445 ꢁ3.3328812474
H 1.0 2.9833671990 0.3185557351 ꢁ3.9028410343
H 1.0 2.7080267275 0.5343984385 ꢁ3.9734140384
C 6.0 0.9392291627 ꢁ0.6558446804 0.8511103409
C 6.0 1.0978985936 ꢁ0.8759789685 2.2876860063
H 1.0 0.9630363567 0.0422481937 2.8707139659
H 1.0 2.0535863570 ꢁ1.3504395250 2.5361426854
SI 14.0 ꢁ0.2749272859 ꢁ2.1016205653 2.9002488008
C 6.0 ꢁ1.9549670655 ꢁ1.3370335586 2.5535380665
H 1.0 ꢁ2.1409379004 ꢁ1.2617627277 1.4773591351
H 1.0 ꢁ2.0357522067 ꢁ0.3362226988 2.9892478515
H 1.0 ꢁ2.7454662099 ꢁ1.9585216166 2.9862031431
C 6.0 0.0139760765 ꢁ2.3340572687 4.7392737769
H 1.0 1.0073301678 ꢁ2.7505986415 4.9318107620
H 1.0 ꢁ0.7304644984 ꢁ3.0213747235 5.1537052374
H 1.0 ꢁ0.0692319062 ꢁ1.3803562514 5.2691052388
C 6.0 ꢁ0.0674552391 ꢁ3.7064424503 1.9477335694
H 1.0 0.9454527429 ꢁ4.1069445613 2.0583896467
H 1.0 ꢁ0.2628102206 ꢁ3.5553036526 0.8812369959
H 1.0 ꢁ0.7705032828 ꢁ4.4599938047 2.3165497087
C 6.0 1.6877610758 ꢁ1.2423323503 ꢁ0.1237172670
H 1.0 1.3827623757 ꢁ1.2107674376 ꢁ1.1652041664
H 1.0 2.4672169160 ꢁ1.9553450536 0.1391471791
H 1.0 0.1444365725 0.0250914288 0.5407853064
TS-c1
E¼ꢁ757.189449
G(195 K)¼ꢁ756.934111
C 6.0 ꢁ0.1676418667 ꢁ0.8325579049 0.3755505841
C 6.0 ꢁ2.8281001452 ꢁ2.1201898483 ꢁ0.3766226030
H 1.0 ꢁ3.0555647520 ꢁ2.5132124337 0.6161899471
H 1.0 ꢁ3.6704564115 ꢁ2.3564751407 ꢁ1.0381976960
C 6.0 0.6277535167 ꢁ0.2288455296 ꢁ0.6926807789
H 1.0 0.2179782565 0.7315128339 ꢁ1.0266913740
H 1.0 0.7541046144 ꢁ0.9045537211 ꢁ1.5464779635
SI 14.0 2.4246650507 0.1390327908 ꢁ0.0578828840
C 6.0 2.2808902710 1.2672993660 1.4351087455
H 1.0 1.7373382071 0.7781723758 2.2494468034
H 1.0 1.7563985301 2.1928464209 1.1770145374
H 1.0 3.2763603969 1.5345567937 1.8033767622
C 6.0 3.3119014653 0.9769993316 ꢁ1.4833492519
H 1.0 3.3084067837 0.3419720974 ꢁ2.3744064057
H 1.0 4.3529654857 1.1816320895 ꢁ1.2144012385
H 1.0 2.8319983612 1.9279600581 ꢁ1.7339182045
C 6.0 3.2271468823 ꢁ1.4976414111 0.3944322903
H 1.0 3.2190048028 ꢁ2.1875346985 ꢁ0.4552734765
H 1.0 2.7126241114 ꢁ1.9758826482 1.2338730895
H 1.0 4.2690941487 ꢁ1.3364572245 0.6887805736
C 6.0 ꢁ1.1197700228 ꢁ0.1871542919 1.1041567644
H 1.0 ꢁ1.2527527380 0.8876245853 0.9951237036
C 6.0 ꢁ2.7195937249 ꢁ0.6382853661 ꢁ0.3488861841
H 1.0 ꢁ2.1708795200 ꢁ0.1172126634 ꢁ1.1363810150
O 8.0 ꢁ3.6924609120 ꢁ0.0034860326 0.2102360637
H 1.0 ꢁ3.6333529338 0.9967473938 0.0081435614
H 1.0 0.0078034081 ꢁ1.8881598757 0.5883744161
H 1.0 ꢁ1.9262756530 ꢁ2.5836014357 ꢁ0.7741776848
H 1.0 ꢁ1.5826709055 ꢁ0.6643241426 1.9621198469
H 1.0 ꢁ3.4487951845 3.1029633554 0.2811171091
O 8.0 ꢁ3.2303215846 2.4074103598 -0.3552617183
H 1.0 ꢁ3.4728329231 2.7584139222 ꢁ1.2236882093
Acknowledgements
We gratefully acknowledge the financial support from the Na-
tional Science Foundation (NSF CHE 8818147 and 9121631) for this
research.
References and notes
1. Forpertinentreviewsofallylmetaladditionssee:(a)Hoffmann,R.W. Angew. Chem.,
Int. Ed. Engl. 1987, 26, 489e503; (b) Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93,
2207e2293; (c) ; Helmchen, G., Hoffmann, R., Mulzer, J., Schaumann, E., Eds. ;
Thieme: Stuttgart, 1996; Vol. 3, pp 1357e1602; (d) Chemler, S. R.; Roush, W. R. In
ModernCarbonylChemistry; Otera, J., Ed.;Wiley-VCH:Weinheim,2000, Chapter11;
(e) Yanagisawa, A. In; Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds. Comprehensive
Asymmetric Catalysis; Springer: Berlin, 1999; vol. 2, Chapter 27; (f) Denmark, S. E.;
Almstead, N. G. InModern Carbonyl Chemistry; Otera, J., Ed.;Wiley-VCH:Weinheim,
2000, Chapter 10; (g) Denmark, S. E.; Fu, J. Chem. Rev. 2003, 103, 2763e2793; (h)
Yus, M.; Gonzalez-Gomez, J. C.; Foubelo, F. Chem. Rev. 2011, 111, 7774e7854.
2. To effectively describe the products observed in these reactions the stereo-
chemical descriptors suggested by Masamune will be employed: Masamune, S.;
Kaiko, T.; Garvey, D. S. J. Am. Chem. Soc. 1982, 104, 5521e5523.
3. Reviews: (a) Hoffmann, R. W. In Stereocontrolled Organic Synthesis; Trost, B. M.,
Ed.; Blackwell Scientific Publications: Cambridge, 1994, pp 259e274; (b) Roush,
W. R. In; Helmchen, G., Hoffmann, R. W., Mulzer, J., Schaumann, E., Eds. Ster-
eoselective Synthesis, Methods of Organic Chemistry (Houben-Weyl); E21; Thieme
Stuttgart: New York, 1996; Vol. 3, pp 1410e1486; (c) Hall, D. G.; Lachance, H.
Org. React. 2008, 73, 1e573.
4. Reviews: (a) Gung, B. W. Org. React. 2004, 64, 1e413; (b) Yamamoto, Y. Acc.
Chem. Res. 1987, 20, 243e249.
5. Reviews: (a) Sakurai, H. Pure Appl. Chem. 1982, 54, 1e22; (b) Majetich, G. In
Organic Synthesis; Hudlicky, T., Ed. Theory and Applications; JAI: Greenwich,
1989; Vol. 1, pp 173e240; (c) Hosomi, A. Acc. Chem. Res. 1988, 21, 200e206; (d)
ꢁ
Fleming, I.; Dunogues, J.; Smithers, R. Org. React. 1989, 37, 57e575; (e) Fleming,
I.; Barbero, A.; Walter, D. Chem. Rev. 1997, 97, 2063e2192; (f) Langkopf, E.;
Schinzer, D. Chem. Rev. 1995, 95, 1375e1408; (g) Masse, C. E.; Panek, J. S. Chem.
Rev. 1995, 95, 1293e1316; (h) The intramolecular reactions of allylsilanes have
also been reviewed Schinzer, D. Synthesis 1988, 263e273.
6. Reviews: (a) Furstner, A. Chem. Rev. 1999, 99, 991e1045; (b) Cintas, P. Synthesis
1992, 248e257; (c) Avalos, M.; Babiano, R.; Cintas, P.; Jimenez, J. L.; Palacios, J. C.
Chem. Soc. Rev. 1999, 28, 169e177; (d) Hargaden, G. C.; Guiry, P. J. Adv. Synth.
Catal. 2007, 349, 2407e2424.
7. Reviews: (a) Duthaler, R. O.; Hafner, A. Chem. Rev. 1992, 92, 807e832; (b)
Duthaler, R. O.; Hafner, A.; Alsters, P. L.; Rothe-Streit, P.; Rihs, G. Pure Appl. Chem.
1992, 64, 1897e1910; (c) Hoppe, D. In Stereoselective Synthesis; Helmchen, G.,
Hoffmann, R. W., Mulzer, J., Schaumann, E., Eds. Thieme: Stuttgart: New York,
1995; Vol. E 21b, pp 1551e1583.
TS-c2
E¼ꢁ757.192915
G(195 K)¼ꢁ756.937875
C 6.0 0.6730404514 ꢁ0.4293965984 0.7725663210
C 6.0 3.1848839704 1.0431993882 1.0508041278
H 1.0 4.0785687317 1.6656749175 0.9179593344
H 1.0 2.3966181408 1.6659044561 1.4762069578
C 6.0 ꢁ0.3551281882 0.5507306448 0.4217065961
8. Denmark, S. E.; Weber, E. J. Helv. Chim. Acta 1983, 66, 1655e1660.

7718
S.E. Denmark et al. / Tetrahedron 68 (2012) 7701e7718
9. For an early, general discussion of the topological features of reactions of
double bonds see: Seebach, D.; Golinski, J. Helv. Chim. Acta 1981, 64, 1413e1423.
10. (a) Denmark, S. E.; Weber, E. J. J. Am. Chem. Soc. 1984, 106, 7970e7971; (b)
Denmark, S. E.; Henke, B. R.; Weber, E. J. J. Am. Chem. Soc. 1987, 109, 2512e2514;
(c) Denmark, S. E.; Weber, E. J.; Wilson, T. M.; Willson, T. M. Tetrahedron 1989,
45, 1053e1065; (d) Denmark, S. E.; Almstead, N. G. J. Org. Chem. 1994, 59,
5130e5132; (e) Denmark, S. E.; Hosoi, S. J. Org. Chem. 1994, 59, 5133e5135; (f)
Denmark, S. E.; Almstead, N. G. J. Mex. Chem. Soc. 2009, 53, 174e192.
11. (a) Imachi, M.; Nakagawa, J.; Hayashi, M. J. Mol. Struct. 1983, 102, 403e412; (b)
Hayashi, M.; Imachi, M.; Saito, M. Chem. Lett. 1977, 221e222; (c) Beagley, B.;
Foord, A.; Moutran, R.; Roszondai, B. J. Mol. Struct. 1977, 117e120; (d) Ohno, K.;
Toga, K.; Murata, M. Bull. Chem. Soc. Jpn. 1977, 50, 2870e2876.
products (proximal, distal) and the transition structures that lead to their
formation (synclinal, antiperiplanar). The structural descriptors allow un-
ambiguous designation of the location of the hydroxyl group with respect to
a defined locant in the molecule; the 13C label in 4 and the double bond in 6.
The transition structures leading to the two isomeric products are deduced by
inspection of the models.
25. (a) Seyferth, D.; Pornet, J. J. Org. Chem. 1980, 45, 1721e1722; (b) Seyferth, D.;
Pornet, J.; Weinstein, R. M. Organometallics 1982, 1, 1651e1658; (c) Hosomi, A.;
Saito, M.; Sakurai, H. Tetrahedron Lett. 1980, 21, 3783e3786.
26. Schmidt, J. P.; Piranx, M.; Phillete, J. F. J. Org. Chem. 1975, 40, 1586e1588.
27. Gruber, L.; Tomoskozi, I.; Radies, L. Synthesis 1975, 708.
28. Tamao, K.; Sumitami, K.; Kumada, M. J. Am. Chem. Soc. 1972, 94, 4374e4376.
29. Pederson, D. J. Org. Chem. 1968, 33, 780e784.
30. Mokotoff, M.; Cavestri, R. C. J. Org. Chem. 1974, 39, 409e411 and references cited
therein.
12. White, J. M.; Clark, C. I. In Topics in Stereochemistry; Denmark, S. E., Ed.; Wiley:
New York, 1999, pp 137e200.
13. Kahn, S. D.; Pau, C. F.; Chamberlin, A. R.; Hehre, W. J. J. Am. Chem. Soc. 1987, 109,
650e663.
31. Krishnamurthy, S. J. Org. Chem. 1981, 46, 4628e4829.
14. For a thorough disquisition on the state of the art in SE0 reactions see: (a)
32. The tert-butyldimethylsilyl ether was originally employed as the protecting
group in the synthesis of 10, but this protecting group could not be removed
without affecting the pentadienylsilyl moiety. The diphenylmethylsilyl ether
was expressly developed for this purpose Denmark, S. E.; Hammer, R. P.; Weber,
E. J.; Habermas, K. L. J. Org. Chem. 1987, 52, 165e168.
€
Matassa, V. G.; Jenkins, P. R.; Kumin, A.; Damm, L.; Schreiber, J.; Felix, D.; Zass,
E.; Eschenmoser, A. Isr. J. Chem. 1989, 29, 321e343; (b) Denmark, S. E.; Werner,
N. S. J. Am. Chem. Soc. 2010, 132, 3612e3620.
15. (a) Fleming, I.; Terrett, N. K. Pure Appl. Chem. 1983, 55, 1707e1713; (b) Fleming,
I. Pure Appl. Chem. 1988, 60, 71e78.
33. Narasaka, K.; Morikawa, A.; Saigo, K.; Mukaiyama, T. Bull. Chem. Soc. Jpn. 1977,
16. (a) Carter, M. J.; Fleming, I. J. Chem. Soc., Chem. Commun. 1976, 679e680; (b) Au-
Yeung, B.-W.; Fleming, I. J. Chem. Soc., Chem. Commun.1977, 79e80; (c) Au-Yeung,
B.-W.; Fleming, I. J. Chem. Soc., Chem. Commun.1977, 81; (d) Fleming, I.; Williams,
R. V. J. Chem. Soc., Perkin Trans. 1 1981, 684e688; (e) Carter, M. J.; Fleming, I.;
Percival, A. J. Chem. Soc., PerkinTrans.11981, 2415e2434; (f) Chan, T. H.; Fleming, I.
Synthesis 1979, 761e786; (g) Fleming, I.; Au-Yeung, B.-W. Tetrahedron 1981, 37,
13e24; (h) Fleming, I.; Terrett, N. K. Tetrahedron Lett. 1984, 25, 5103e5104.
17. (a) Fleming, I.; Terrett, N. K. J. Organomet. Chem. 1984, 264, 99e118; (b) Fleming,
I.; Terrett, N. K. Tetrahedron Lett. 1983, 24, 4153e4156; (c) Chow, H.-F.; Fleming,
I. Tetrahedron Lett. 1985, 26, 397e400; (d) Fleming, I.; Sarkar, A. K.; Thomas, A. P.
J. Chem. Soc., Chem. Commun. 1987, 157e159.
50, 2773e2776.
34. The assignment of configuration was made by the relative slope of the lan-
thanide induced shift (using tris(6,6,7,7,8,8,8-heptafluoro-2,2-dimethyl-3,5-
octanedionato)europium) for the 13C-enriched methylidene carbon signals. The
‘proximal’ isomer 4b was expected to have the larger LIS because the hydroxyl
group in this isomer is closer to the 13C-enriched methylidene carbon. The
slope of the LIS was 5.27 ppm/equiv. for the ‘distal’ isomer 4a and 12.77 ppm/
equiv. for the ‘proximal’ isomer 4b (correlation coefficients of 0.996 and 0.993,
respectively).
35. Treatment of 3 with trifluoroacetic acid or SiCl4 gave only the desilylated
product 21.
18. (a) Wetter, H.; Scherer, P.; Schweizer, W. B. Helv. Chim. Acta 1979, 62,
1985e1989; (b) Wetter, H.; Scherer, P. Helv. Chim. Acta 1983, 66, 118e122.
19. (a) Young, D.;Kitching, W. Tetrahedron Lett.1983, 24, 5793e5796; (b) Wickham, G.;
Kitching, W. J. Org. Chem. 1983, 48, 612e614; (c) Wickham, G.; Kitching, W. Or-
ganometallics 1983, 2, 541e547; (d) Young, D.; Kitching, W.; Wickham, G. Tetra-
hedron Lett.1983, 24, 5789e5792; (e) Young, D.;Kitching, W. Aust. J. Chem.1985, 38,
1767e1777; (f) Young, D.; Jones, M.; Kitching, W. Aust. J. Chem.1986, 39, 563e573;
(g) Wickham, G.; Young, D.; Kitching, W. Organometallics 1988, 7, 1187e1195.
20. (a) Hayashi, T.; Konishi, M.; Kumada, M. J. Am. Chem. Soc. 1982, 104, 4963e4965;
(b) Hayashi, T.; Konishi, M.; Kumada, M. J. Org. Chem. 1983, 48, 281e282; (c)
Hayashi, T.; Ito, H.; Kumada, N. Tetrahedron Lett. 1982, 23, 4605e4606; (d)
Hayashi, T.; Okamoto, Y.; Kabeta, K.; Hagihara, T.; Kumada, M. J. Org. Chem.
1984, 49, 4224e4226.
36. To establish that the reaction conditions for cyclization of 3 were not cata-
lyzing the migration of silicon before cyclization, the 13C NMR spectrum of
unreacted 3 recovered from the reaction with Et2AlCl was obtained. The
ꢀ
distance between the termini in the pentadienylsilane system is about 5 A
ruling out a [1,5] silicon migration. However, two successive [1,3] migrations
could not be ruled out but could be easily detected by the 13C label. The 13C
NMR spectrum of the recovered aldehyde 3 indicated no scrambling of the
13C label.
37. Kozikowski, A. P.; Schmiesing, R. J. J. Org. Chem. 1983, 48, 1000e1007.
39. (a) Tamao, K.; Kakui, T.; Akita, M.; Iwahara, T.; Kanatani, R.; Yoshida, J.; Kumada,
M. Tetrahedron 1983, 39, 983e990; (b) Oppolzer, W.; Mills, R. J.; Pachinger, W.;
Stevenson, T. Helv. Chim. Acta 1986, 69, 1542e1545.
40. (a) Fristad, W. E.; Peterson, J. R. J. Org. Chem. 1985, 50, 10e18; (b) Pirkle, W. H.;
21. (a) Shambayati, S.; Crowe, W. E.; Schreiber, S. L. Angew. Chem., Int. Ed. Engl. 1990,
29, 256e272; (b) Ooi, T.; Maruoka, K. In Modern Carbonyl Chemistry; Otera, J.,
Ed.; Wiley-VCH: Weinheim, 2000, Chapter 1; (c) Saito, S.; Yamamoto, H. In
Modern Carbonyl Chemistry; Otera, J., Ed.; Wiley-VCH: Weinheim, 2000, Chapter
Adams, P. E. J. Org. Chem. 1980, 45, 4111e4117.
41. Stothers, J. B.; Tan, C. T. Can. J. Chem. 1976, 54, 917e925.
42. Huisgen, R. Acc. Chem. Res. 1977, 10, 117e124.
43. Reichardt, C. In Solvent and Solvent Effects in Organic Chemistry; VCH: Weinheim,
1988, pp 363e376.
€
2; (d) Reetz, M. T.; Hullmann, M.; Massa, W.; Berger, S.; Rademacher, P.; Hey-
manns, P. J. Am. Chem. Soc. 1986, 108, 2405e2408; (e) Denmark, S. E.; Almstead,
N. G. J. Am. Chem. Soc. 1993, 115, 3133e3139.
22. (a) Keck, G. E.; Castellino, S. J. Am. Chem. Soc. 1986, 108, 3847e3849; (b) Keck, G.
E.; Castellino, S.; Wiley, M. R. J. Org. Chem. 1986, 51, 5478e5480; (c) Filippini, F.;
Susz, B.-P. Helv. Chim. Acta 1971, 54, 835e845; (d) Beattie, I. R. Quart. Rev. 1963,
17, 382e405.
23. Biddle, M. M.; Reich, H. J. J. Org. Chem. 2006, 71, 4031e4039.
24. A brief comment about nomenclature used in this paper is warranted. We take
care to differentiate the descriptors for the structures of the final cyclization
44. (a) Anh, N. T.; Thanh, B. T. Nouv. J. Chim. 1986, 10, 681e683; (b) Mulzer, J.;
Buntrup, G.; Finke, J.; Zippel, M. J. Am. Chem. Soc. 1979, 101, 7723e7725.
45. Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157e167.
46. Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995e2001.
47. Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon, M. S.; Jensen,
J. H.; Koseki, S.; Matsunaga, N.; Nguyen, L. A.; Su, S.; Windus, T. L.; Dupuis, M.;
Montgomery, J. A. J. Comput. Chem. 1993, 14, 1347e1363.
48. Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923e2925.
49. Gilman, H.; Schulze, F. J. Am. Chem. Soc. 1925, 47, 2002e2005.