Stereoselective Synthesis of Carbobicyclics via Organoyttrium-Catalyzed Sequential Cyclization/Silylation Reactions

Article abstract of DOI:10.1021/jo9721352

The sequential cyclization/silylation of 1,5-dienes and 1,6-dienes was effected under mild reaction conditions using catalytic quantities of Cp₂*YMe-THF. The process provides carbobicyclics in high yields and with excellent selectivities. The active catalyst is postulated to be Cp₂*YH-THF, which is generated in situ. A variety of alkenyl-substituted cyclopentane and cyclohexane substrates were examined. The high diastereoselectivities apparently originate from a preference for a chairlike transition structure that minimizes unfavorable steric interactions between the bulky Cp* ligands of the catalyst and the preexisting ring of the substrate. Acyclic triene precursors, 4-ethenyl-substituted 1,5-heptadienes and 5-ethenyl-substituted 1,8-nonadienes were also examined. These triene substrates, when exposed to the cyclization/silylation protocol, provide the strained trans-bicyclo[3.3.0]octanes and trans-decalin systems in high yield with excellent diastereoselectivity. The high selectivity is again attributed to the preference for a chairlike transition structure. The cyclized organosilane products isolated from these reactions were easily converted to the more versatile alcohols utilizing known oxidation methods.

Full text of DOI:10.1021/jo9721352

2292
J . Org. Chem. 1998, 63, 2292-2306
Ster eoselective Syn th esis of Ca r bobicyclics via
Or ga n oyttr iu m -Ca ta lyzed Sequ en tia l Cycliza tion /Silyla tion
Rea ction s
Gary A. Molander,* Paul J . Nichols, and Bruce C. Noll†
Department of Chemistry and Biochemistry, University of Colorado, Boulder, Colorado 80309-0215
Received November 21, 1997
The sequential cyclization/silylation of 1,5-dienes and 1,6-dienes was effected under mild reaction
conditions using catalytic quantities of Cp2*YMe‚THF. The process provides carbobicyclics in high
yields and with excellent selectivities. The active catalyst is postulated to be Cp2*YH‚THF, which
is generated in situ. A variety of alkenyl-substituted cyclopentane and cyclohexane substrates
were examined. The high diastereoselectivities apparently originate from a preference for a chairlike
transition structure that minimizes unfavorable steric interactions between the bulky Cp* ligands
of the catalyst and the preexisting ring of the substrate. Acyclic triene precursors, 4-ethenyl-
substituted 1,5-heptadienes and 5-ethenyl-substituted 1,8-nonadienes were also examined. These
triene substrates, when exposed to the cyclization/silylation protocol, provide the strained trans-
bicyclo[3.3.0]octanes and trans-decalin systems in high yield with excellent diastereoselectivity.
The high selectivity is again attributed to the preference for a chairlike transition structure. The
cyclized organosilane products isolated from these reactions were easily converted to the more
versatile alcohols utilizing known oxidation methods.
In tr od u ction
reaction. The requisite diene, N-(2-propenyl)-2-ethe-
nylpiperidine, was easily synthesized in five steps from
commercially available (()-2-piperidinemethanol. When
reacted with methylphenylsilane in the presence of
Cp*2YCH3‚THF over 1 h, the expected phenylsilyl-
substituted quinolizidine was obtained in high yield.
Oxidation using the Woerpel oxidation conditions4 pro-
vided the natural product (()-epilupinine in yields rang-
ing from 51% to 62% over two steps (eq 3). Remarkably,
none of the trans isomer, naturally occurring (()-lupin-
ine, could be observed by NMR. The high diastereose-
lectivity associated with this process apparently results
from the avoidance of interactions between the catalyst
and the piperidine ring during the intramolecular inser-
tion of the more hindered vinyl group of the diene.
The highly selective (()-epilupinine synthesis sug-
gested that the organoyttrium-catalyzed cyclization/
silylation method could be used to prepare a variety of
interesting polycyclic ring systems stereoselectively.
Toward this end, a number of functionalized dienes and
acyclic trienes were examined to ascertain stereoselec-
tivity and functional group tolerances associated with
polycyclic ring formation using this cyclization protocol.
For the past several years we, along with others, have
been developing selective cyclization1 and silylation2
reactions employing group 3 and organolanthanide cata-
lysts. During the course of our research, the organoyt-
trium-catalyzed, stereoselective cyclization/silylation re-
action of substituted 1,5- and 1,6-dienes was developed.1f
A number of substituted dienes were cyclized in the
presence of the precatalyst Cp*2YCH3‚THF and phenyl-
silanes to provide a variety of phenylsilyl-substituted
carbocycles in high yield with moderate to excellent
stereoselectivity (eqs 1 and 2). The resulting silanes
could be easily oxidized to provide the more versatile
alcohols. The air sensitive, Lewis acidic catalyst was
compatible with ether, thioacetal, and tertiary amine
functionalities.
(1) (a) Parkin, G.; Bunel, E.; Burger, B. J .; Trimmer, M. S.; van
Asselt, A.; Bercaw, J . E. J . Mol. Catal. 1987, 41, 21. (b) Bunel, E.;
Burger, B. J .; Bercaw, J . E. J . Am. Chem. Soc. 1988, 110, 976. (c)
Heeres, H. J .; Heeres, A.; Teuben, J . H. Organometallics 1990, 9, 1508.
(d) Piers, W. E.; Shapiro, P. J .; Bunel, E. E.; Bercaw, J . E. Synlett 1990,
74. (e) Molander, G. A.; Hoberg, J . O. J . Am. Chem. Soc. 1992, 114,
3123. (f) Molander, G. A.; Nichols, P. J . J . Am. Chem. Soc. 1995, 117,
4415. (g) Molander, G. A.; Retsch, W. H. J . Am. Chem. Soc. 1997,
119, 8817.
(2) (a) Sakakura, T.; Lautenschlager, H.-J .; Tanaka, M. J . Chem.
Soc., Chem. Commun. 1991, 40. (b) Molander, G. A.; J ulius, M. J .
Org. Chem. 1992, 57, 6347. (c) Onozawa, S.-Y.; Sakakura, T.; Tanaka,
M. Tetrahedron Lett. 1994, 35, 8177. (d) Molander, G. A.; Retsch, W.
H. Organometallics 1995, 14, 4570. (e) Fu, P.-F.; Brard, L.; Li, Y.;
Marks, T. J . J . Am. Chem. Soc. 1995, 117, 7157. (f) Molander, G. A.;
Winterfeld, J . J . Organomet. Chem. 1996, 524, 275.
To demonstrate the synthetic viability of this protocol,
the catalytic cyclization/silylation method was utilized in
the key step of a natural product synthesis.3 The
structural features of (()-epilupinine, the simplest mem-
ber of the lupin alkaloids, made it an ideal candidate for
synthesis via an organoyttrium-catalyzed cyclization
† To whom all correspondence regarding X-ray crystal structures
should be addressed.
(3) Molander, G. A.; Nichols, P. J . J . Org. Chem. 1996, 61, 6040.
(4) Smitrovich, J . H.; Woerpel, K. H. J . Org. Chem. 1996, 61, 6044.
S0022-3263(97)02135-X CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/13/1998

Stereoselective Synthesis of Carbobicycles
J . Org. Chem., Vol. 63, No. 7, 1998 2293
pel oxidation conditions4 provided alcohol 5 as a single
diastereomer in 93% yield (eq 5). The structure of alcohol
5 was determined unambiguously by single-crystal X-ray
crystallography of the corresponding diol generated by
removal of the TBDMS protecting group.
Sch em e 1
Triene 6 was prepared from (2R*,3R*)-3-ethenyl-2-(2-
propenyl)cyclopentanone6 following the same protocol
outlined in Scheme 1. This substrate, in the presence of
1.1 equiv of PhMeSiH2 and 5 mol % of the precatalyst
Cp*2YCH3‚THF, provided the crude bicyclic organosilane
7 (eq 6). The crude silane mixture was then directly
subjected to Woerpel oxidation conditions to give diol 8
as a 6:1 mixture of inseparable diastereomers in 72%
yield (eq 7). Fortuitously, the Woerpel oxidation condi-
tions removed the TMS protecting group and oxidized the
silane, alleviating the need for a protecting group removal
step as was necessary for alcohol 5. The stereochemistry
shown for the major diastereomer of silane 7 was based
on the stereochemistry found previously for silane 4.
The results described herein provide a detailed account
of these annulation/silylation reactions mediated by an
organoyttrium catalyst.
Resu lts a n d Discu ssion
Cycliza tion of Mon ocyclic P r ecu r sor s. Initial
investigations of the catalytic cyclization/silylation reac-
tion involved the examination of simple, polyalkenyl-
substituted carbocycles. Triene 3 was prepared from the
known ketone 15 in two steps as outlined in Scheme 1.
In a typical catalytic reaction, 5 mol % of the precatalyst
Cp*2YCH3‚THF was dissolved in cyclohexane (0.5 M)
followed by the addition of 1.1 equiv of methylphenylsi-
lane and 1 equiv of the substrate at room temperature.
The sterically bulky methylphenylsilane was used as the
silylating agent because product dimerization had been
observed previously in six-membered ring formation
using phenylsilane.1f Thus the phenylsilane products
derived from 1,6-diene cyclization/silylation compete with
phenylsilane in the trapping of the cyclized organoyt-
trium intermediate resulting in unwanted product dimer-
ization. Exposure of triene 3 to the cyclization/silylation
protocol for 1 h at room temperature provided the bicyclic
organosilane 4 in 86% isolated yield as what appeared
to be a single diastereomer by gas chromatographic
analysis (eq 4). The diastereoselectivity of the cyclization
event in reactions that use methylphenylsilane as the
silylating agent is difficult to assess. The σ-bond me-
tathesis trapping of the cyclized organoyttrium interme-
diate with methylphenylsilane generates a product with
a stereocenter on silicon in addition to any stereocenters
formed in the annulation. Thus, when methylphenylsi-
lane was utilized, the diastereoselectivity was determined
after oxidation of the cyclized organosilane to the corre-
sponding alcohol. Oxidation of silane 4 using the Woer-
There are several interesting aspects of the cyclization/
silylation of substrates 3 and 6 that are worth noting.
First, initial reaction of the active yttrium hydride
catalyst occurs highly selectively at the least hindered
of the three available olefins. The allyl group reacts in
preference to the more hindered vinyl groups. This initial
olefin insertion is followed by a rapid intramolecular
cyclization resulting in a cyclic organoyttrium intermedi-
ate. This cyclization is presumed to be rapid because
uncyclized silane products, the result of a σ-bond me-
tathesis reaction between the uncyclized organoyttrium
intermediate and the methylphenylsilane present in the
(6) Posner, G. H.; Sterling, J . J .; Whitten, C. E.; Lentz, C. M.;
Brunelle, D. J . J . Am. Chem. Soc. 1975, 97, 107.
(5) Shimizu, I.; Ohashi, Y.; Tsuji, J . Tetrahedron Lett. 1983, 24, 3865.

2294 J . Org. Chem., Vol. 63, No. 7, 1998
Molander et al.
Sch em e 2
The next cyclic substrates that were investigated
possessed an allyl group at a quaternary center and a
vinyl group substituted at a tertiary carbon. The known
diol 13 was prepared following the addition of 2 equiv of
allylmagnesium bromide to 2-(hydroxymethyl)cyclohex-
anone.8 The two diastereomers were cleanly separated
at this point by flash chromatography on silica gel
(Scheme 2). Independent oxidation of the two diastere-
omers (one shown) followed by Wittig olefination provided
alcohols 14a and 14b in diastereomerically pure form.
Protection of the alcohols with either a TBDMS or TMS
group provided the required substrates.
When substrate 15 was subjected to the standard
cyclization/silylation conditions, a cyclized organosilane
16 was isolated in 86% yield as a 1.6:1 mixture of
diastereomers along with 5% of other, nonseparable
isomers (eq 10). As explained earlier, the diastereose-
lectivity of the annulation event could only be determined
after oxidation in cases where methylphenylsilane was
used as a silylating agent because of the silicon stereo-
center. Oxidation of silane 16 provided alcohol 17 as a
single diastereomer in 70% yield (eq 11). The stereo-
chemistry of alcohol 17 was determined unambiguously
from an X-ray crystal structure of the corresponding diol
following TBDMS removal with TBAF.
F igu r e 1. Proposed catalytic cycle for the organoyttrium-
catalyzed cyclization of triene 3.
reaction mixture, are not observed. The intramolecular
insertion reaction also occurs exclusively, this time at the
less hindered of the two available vinyl groups. Subse-
quent σ-bond metathesis trapping of the cyclized orga-
noyttrium intermediate with methylphenylsilane pro-
vides the bicyclic organosilane and regenerates the active
hydride catalyst (Figure 1). The high selectivity associ-
ated with the cyclization of 3 is attributed to the chairlike
transition structure shown in Figure 1. Unfavorable
steric interactions between the bulky Cp* ligands of the
catalyst and the preexisting cyclohexyl ring of the
substrate may help to explain the high selectivity in the
case of substrate 3. The lower selectivity seen for the
cyclization of substrate 6 may be attributed to diminished
steric interations between the more fluxional cyclopentyl
ring and the catalyst. The reaction of substrates 3 and
6 also demonstrates the ability of the catalyst to tolerate
silyl protected alcohols.
Other dienes were prepared to determine if the inter-
mediate organometallic could cyclize to a vinyl group at
a hindered quaternary center. Substrates 9 and 11 were
easily prepared from 2-(2-propenyl)cyclohexanone7 in two
steps via vinylmagnesium bromide addition followed by
the appropriate alcohol protection. When the substrates
were exposed to the standard reaction protocol, clean
formation of uncyclized organosilanes 10 and 12 resulted
(eqs 8 and 9). This result is not surprising given the
selectivity observed in eq 4. It is worth noting that the
THP ether protecting group in substrate 12 tolerated the
Lewis acidic catalyst.
Some interesting features were associated with the
cyclization/silylation of diene 15. The initial insertion
of the organoyttrium hydride catalyst again takes place
exclusively at the allyl group. The ensuing cyclization
event occurs with complete selectivity. Presumably this
selectivity is again associated with the desired chairlike
transition structure and unfavorable interactions be-
tween the preexisting ring and the catalyst in a boatlike
transition structure as depicted in Figure 2. The in-
(7) Vanderwerf, C. A.; Lemmerman, V. L. Org. Synth. 1955, Coll.
Vol. 3, 44.
(8) Kobayashi, S.; Hachiya, I. J . Org. Chem. 1994, 59, 3590.

Stereoselective Synthesis of Carbobicycles
J . Org. Chem., Vol. 63, No. 7, 1998 2295
F igu r e 3. Proposed orientation required for intramolecular
olefin insertion of dienes 22 and 25.
F igu r e 2. Proposed orientation required for intramolecular
olefin insertion of diene 16.
(eq 17). The stereochemistry of organosilanes 22 and 25
were subsequently proven unambiguously by obtaining
a single-crystal X-ray structure of diol 26.
tramolecular insertion is again rapid as evidenced by the
absence of uncyclized silylated products.
The analogous TMS protected diene 18 was also
exposed to the reaction protocol. Diene 18, in the
presence of 1.1 equiv of PhMeSiH2 and 5 mol % of the
precatalyst Cp*2YCH3‚THF, provided the crude bicyclic
organosilane 19 as a 2:1 diastereomeric mixture along
with a significant amount of an additional, undetermined
isomer (eq 12). The crude silane mixture was subjected
to Woerpel oxidation conditions to provide diol 20 as a
single diastereomer in 48% yield along with 14% of an
unidentified isomer (eq 13). Again, the Woerpel oxidation
conditions removed the TMS protecting group concurrent
with oxidation of the silane.
The dienes that originated from alcohol 14b were also
examined to determine what effect the trans arrange-
ment of the allyl and vinyl groups would have on the
selectivity of the ring closure. When substrate 21 was
exposed to the standard reaction conditions, the cyclized
silane 22 was isolated in 93% yield as a 1.5:1 mixture of
diastereomers (eq 14). Subsequent oxidation provided
alcohol 23 in 68% yield as a 17.8:1 mixture of diastere-
omers (eq 15). Unfortunately, all attempts to remove the
TBDMS protecting group in alcohol 23, in hopes of
obtaining a crystalline diol, were unsuccessful. Because
TMS groups are much easier to remove, substrate 24 was
prepared from alcohol 14b and subjected to the standard
cyclization/silylation conditions. Bicyclic organosilane 25
was isolated in 89% yield as a 1.5:1 mixture of diaster-
eomers (eq 16). Subsequent Woerpel oxidation provided
the desired diol 26 in 53% yield as a single diastereomer
Some features of the cyclizations of dienes 21 and 24
are worth highlighting. The annulation event most likely
proceeds exclusively though a chairlike transition struc-
ture and thus avoids unfavorable interactions between
the catalyst and the preexisting ring as shown in Figure
3. The initial insertion of the organoyttium hydride
catalyst is again completely chemoselective for the less
hindered olefin. Catalytic annulation/silylation reactions
of this type could be used to generate a wide variety of
either cis or trans fused decalin systems in a highly
selective manner.
The cyclopentane analogue of substrate 24 was the
next diene examined. (1R*,2S*)-2-ethenyl-1-(2-prope-
nyl)-1-(trimethylsiloxy)cyclopentane (27) was prepared
using the same reaction sequence outlined in Scheme 2
starting with allylmagnesium bromide addition to 2-(hy-

2296 J . Org. Chem., Vol. 63, No. 7, 1998
Molander et al.
droxymethyl)cyclopentanone to prepare the requisite diol.
When diene 27 was reacted with 1.1 equiv of PhMeSiH2
at room temperature in the presence of 5 mol % of the
precatalyst Cp*2YCH3‚THF for 45 min, bicyclic organosi-
lane 28 was isolated in 88% yield as a 1:1 mixture of
diastereomers (eq 18). The selectivity appeared similar
in this case to some of the aforementioned cyclization
reactions prior to oxidation of the silane stereocenter.
Oxidation of silane 28 provided a 51% yield of diol 29 as
a 1:1 mixture of diastereomers (eq 19). The lack of
selectivity in this cyclization was unexpected given the
results seen in the analogous cyclohexyl substrates. The
fluxionality of the cyclopentane ring may account for the
lack of selectivity in this reaction. There is apparently
no bias toward a chairlike transition structure in this
example. A single-crystal X-ray structure of one of the
diol diastereomers 29 confirmed the structure of sub-
strate 27 as well as the structure of the final product.
In a previous study the catalytic cyclization/silylation
reaction of 1,2-divinylbenzene was reported.3 From this
experiment it was known that the active yttrium hydride
catalyst was able to add to a vinyl substituent on an
aromatic ring. The diene 1-ethenyl-2-(2-propenyl)ben-
zene 33 was synthesized from the ozonolysis of indene
followed by Wittig olefination of the crude dialdehyde.
This substrate was prepared to examine the chemose-
lectivity for an allyl group versus a vinyl group ortho
substituted on a benzene ring. When aromatic diene 33
was reacted under the standard cyclization/silylation
reaction conditions two organosilane products (34 and 35)
were isolated in 99% crude yield as a 7:1 mixture (eq 22).
The major product 34 was the silane resulting from initial
organoyttrium hydride insertion at the less hindered allyl
substituent followed by cyclization and σ-bond metathesis
trapping by PhMeSiH2. The minor product was the cyclic
organosilane resulting from initial yttrium hydride inser-
tion at the more hindered vinyl group followed by
cyclization and silane trapping. This was the first
example, using the organoyttrium-catalyzed cyclization/
silylation method, in which a significant amount of
undesired material was produced via initial yttrium
hydride insertion to a more hindered olefin. This lapse
in chemoselectivity is perhaps due to the planarity of the
aromatic ring which makes the vinyl group more acces-
sible in this case, although electronic effects cannot be
ruled out.
Hoping that a larger alcohol protecting group might
lead to an increase in selectivity, the TMS substituent
in substrate 27 was removed and replaced with a bulkier
TBDMS protecting group providing diene 30. Reaction
under the standard cyclization/silylation protocol pro-
vided silane 31 in 92% yield as a 1.7:1 mixture of
diastereomers (eq 20). We had hoped that oxidation of
the silane stereocenter would reveal selective formation
of the corresponding alcohol. Unfortunately, oxidation
of silane 31 provided alcohol 32 in 86% yield as a 1:1
mixture of diastereomers (eq 21). Thus, no favorable
steric bias was gained in this cyclization through the use
of the larger TBDMS protecting group. Although the
cyclizations of these alkenyl-substituted cyclopentanes
lack selectivity, the annulation event was highly efficient
and could provide a variety of substituted 5,6-ring
systems from simple precursors.
The identities of silanes 34 and 35 were confirmed by
oxidation9 to the corresponding alcohols and comparison
of their NMR spectra to those previously reported in the
literature (eq 23).10 Curiously, the oxidation resulted in
isolation of alcohols 36 and 37 in 81% yield as a 14:1
mixture of isomers. The observed selectivity essentially
doubled from the cyclization/silylation event. The most
likely explanation for this peculiar result is partial
decomposition of the minor product during oxidation or
isolation.
The reaction of diene 33 is noteworthy for a few
reasons. First, preferential insertion of the yttrium
hydride catalyst still occurs at the allyl double bond.
(9) Bergens, S. H.; Nogeda, P.; Whelan J .; Bosnich, B. J . Am. Chem.
Soc. 1992, 114, 2121.
(10) (a) Taylor, S. K.; Davisson, M. E.; Hissom, B. R., J r.; Brown, S.
L.; Pristach, H. A.; Schramm, S. B.; Harvey, S. M. J . Org. Chem. 1987,
52, 425. (b) Bijoy, P.; Rao, G. S. R. S. Synth. Commun. 1993, 23, 2701.
(c) Beckwith, A. L. J .; Gerba, S. Aust. J . Chem. 1992, 45, 289.

Stereoselective Synthesis of Carbobicycles
J . Org. Chem., Vol. 63, No. 7, 1998 2297
Additionally, the organometallic that is initially formed
through insertion of the vinyl group has the metallocene
complex placed at the terminus of the alkene rather than
at the internal carbon. This observation is consistent
with our previous results using the yttrium-based cata-
lyst but is in contrast to selectivities reported elsewhere
for the silylation of styrene using sterically more open
samarium2e and neodymium2a catalysts. Steric concerns
are critically important in the reaction of a organolan-
thanide hydrides to styrene, and the less open Cp*2YH
catalyst has always been observed to undergo olefin
insertion in a manner that places the metallocene
complex at the terminal carbon of the original monosub-
stituted alkene unit.
Diene 38 was prepared in an identical fashion to diene
24 from 2-(hydroxymethyl)cyclohexanone using vinyl in
place of allyl to prepare the required diol (Scheme 2). The
results for cyclization of dienes 9 and 11 indicated that
diene 38 would be too sterically encumbered to cyclize.
Nonetheless, reaction of substrate 38 with PhSiH3 using
the standard reaction protocol provided the cyclized
organosilane 39, along with a small amount of other
isomers, in 48% yield as an 8:1 mixture of diastereomers
(eq 24). Some of the uncyclized silane 40 was also
isolated. This result served to illustrate that the 5-exo
cyclization onto a vinyl group at a quaternary center is
possible. The isolation of the uncyclized silane 40
indicates that σ-bond metathesis with the silane is
competing with a sterically inhibited cyclization event.
F igu r e 4. Proposed orientation required for intramolecular
olefin insertion of diene 39.
chemistry of silane 41 was determined unambiguously
by a single-crystal X-ray structure of diol 42.
Interestingly, the stereochemistry of silane 41 is dif-
ferent from that which would be predicted by a chairlike
transition structure. The cyclization of diene 38 proceeds
exclusively through a boatlike transition structure. This
mode of cyclization had not been previously observed in
these cyclization reactions. Unfavorable interactions
between the bulky Cp* rings of the catalyst and the TMS
ether apparently favor a boatlike transition structure as
shown in Figure 4. Additionally, unfavorable interac-
tions (A1,3-strain) between the OTMS group and the vinyl
group also contribute to the preference for the boatlike
transition structure. This result suggests that the or-
ganoyttrium-catalyzed cyclization/silylation protocol can
be used to selectively prepare substituted cyclopentane
rings from a variety of cyclic, divinyl substrates.
A bulkier silane was utilized in an attempt to slow the
σ-bond metathesis reaction enough to allow clean cy-
clization to occur. Thus, diene 38 was reacted with
PhMeSiH2 using the standard reaction protocol. The
reaction with PhMeSiH2 was slower, requiring 6 h at
ambient temperature. Isolation of the cyclic organosilane
41 was achieved in 80% yield as a 4.4:1 mixture of
diastereomers (eq 25). No uncyclized material was
isolated in this case which indicated that bulkier silanes
could be used effectively to enhance cyclization reactions
of this type. Determination of the stereoselectivity of the
cyclization again required oxidation of the silane to
remove the silicon stereocenter. In a subsequent NMR
experiment, diene 38 was reacted with PhMeSiH2 in the
presence of 5 mol % of the precatalyst Cp*2YCH3‚THF
in benzene-d6 over the course of 7 h at room temperature.
The crude cyclic organosilane 41 was isolated and im-
mediately oxidized to give diol 42 in 48% yield over two
steps as a single diastereomer (eq 26). Thus, not only
do bulky divinyl substrates of this type cyclize, they
apparently do so with complete selectivity. The stereo-
Cycliza tion of Acyclic P olyen e P r ecu r sor s
The selective conversion of simple acyclic polyene
precursors to complex polycyclic molecules is an impor-
tant transformation in organic synthesis. Cyclizations
of this type has been reported by Negishi and co-workers

2298 J . Org. Chem., Vol. 63, No. 7, 1998
Molander et al.
using catalytic titanium alkoxide.11 Additionally, a va-
riety of elegant palladium(II)-catalyzed “zipper” reactions
of enynes have been reported by Trost and co-workers.12
The only lanthanide-catalyzed polyene cyclization was
disclosed by Marks and co-workers, who used dienes and
enynes to prepare nitrogen-containing bicycles via a
catalytic hydroamination reaction.13 In an attempt to
expand the organoyttrium-catalyzed cyclization/silylation
method, a variety of triene precursors were prepared to
examine the propensity for multiple cyclization reactions
and study relevant stereochemistry issues.
The first triene substrate prepared was the sym-
metrical substrate 43. This substrate was prepared in
two steps by reacting methyl acrylate with 2 equiv of
allylmagnesium bromide followed by TBDMS protection
of the resulting alcohol. It was uncertain that this triene
precursor would even cyclize because it had already been
established that a vinyl substituent at a quaternary
center on a cyclohexane ring would not undergo cycliza-
tion with an adjacent allyl group. Nonetheless, when
triene 43 was treated with PhSiH3 in the presence of 5
mol % of the precatalyst Cp*2YCH3‚THF in cyclohexane
(0.5 M) the major product, bicyclic silane 44, was isolated
in 55% yield in a 3:1 diastereomeric ratio (eq 27).
Interestingly, a significant amount of the six-membered
ring byproduct 45 was also isolated. The major diaste-
reomer of silane 44 was easily isolated and was later
determined to be the trans 5,5-ring system shown.
Oxidation of the major diastereomer provided alcohol 46
in 44% yield (eq 28). A single-crystal X-ray structure of
alcohol 46 confirmed the stereochemistry shown for
silane 44. Because the second 5-exo-trig cyclization event
was presumed to proceed preferentially through a chair-
like transition structure, the minor diastereomer of silane
44 was assumed to be the cis fused bicyclic silane.
However, precise stereochemical identification of the
minor diastereomer of silane 44 was not possible.
uents. An ensuing intramolecular cyclization event then
takes place that favors formation of a five-membered ring
over a six-membered ring by a 5:1 ratio. Notably, a more
sterically inhibited 5-exo-trig cyclization is favored over
the 6-exo-trig cyclization. The organoyttrium intermedi-
ate from the first 5-exo-trig cyclization is apparently not
trapped by the PhSiH3 in the reaction mixture. Instead,
this intermediate rapidly undergoes a second 5-exo-trig
cyclization, and the resulting organometallic is then
converted to the corresponding silane 44 via a σ-bond
metathesis reaction with PhSiH3. Any organometallic
generated by a 6-exo process is transformed by a σ-bond
metathesis reaction with PhSiH3 because the second
cyclization event (which would generate a bridged bicyclic
sytem) is slow.
Thinking that the OTBDMS group was bulky enough
to slow the 5-exo-trig cyclization pathway and allow for
the formation of silane 45, a second triene was prepared
which replaced the OTBDMS group with the smaller
OTMS substituent. Substrate 47 was prepared in fash-
ion similar to triene 43. The reaction of the OTMS
containing precursor 47 under the standard organoyt-
trium-catalyzed cyclization/silylation protocol provided
the crude bicyclic silane 48 as a single diastereomer (eq
29). Attempts to purify silane 48 resulted in partial
decomposition of the product. Thus, direct oxidation of
the crude silane product provided diol 49 in 73% yield
over two steps as a single diastereomer (eq 30). No other
significant byproducts were isolated. Removal of the
TBDMS group in the known alcohol 46 and subsequent
1H NMR spectral comparison with diol 49 provided the
structural verification of silane 48. Remarkably, three
stereocenters are produced with complete selectivity from
the simple, achiral triene 47 via the organoyttrium-
catalyzed cyclization/silylation protocol.
There are again several interesting features associated
with this unique transformation. First, none of the silane
product from a 6-exo-trig cyclization was observed. Ap-
parently, the smaller OTMS group allows complete
preference for the 5-exo cyclization pathway. Addition-
ally, following the initial cyclization event the smaller
OTMS substituent must prefer the pseudoaxial orienta-
tion, allowing for the generation of a single trans-fused
(12) (a) For a review, see: Trost, B. M. Acc. Chem. Res. 1990, 23,
34. (b) Trost, B. M.; Lee, D. C.; Rise, F. Tetrahedron Lett. 1989, 30,
651. (c) Trost, B. M.; Pedregal, C. J . Am. Chem. Soc. 1992, 114, 7292.
(d) Trost, B. M.; Czeskis, A. Tetrahedron Lett. 1994, 35, 211. (e) Trost,
B. M.; Chung, J . Y. L. J . Am. Chem. Soc. 1985, 107, 4586. (f) Trost,
B. M.; Tanoury, G. J . J . Am. Chem. Soc. 1987, 109, 4753. (g) Trost, B.
M.; Lee, D. C. J . Org. Chem. 1989, 54, 2271. (h) Trost, B. M.; Shi, Y.
J . Am. Chem. Soc. 1991, 113, 701. (i) Trost, B. M.; Shi, Y. J . Am.
Chem. Soc. 1993, 115, 9421. (j) Trost, B. M.; Shi, Y. J . Am. Chem.
Soc. 1992, 114, 791.
Some distinct features of the multiple cyclization of
triene substrate 43 should be mentioned. From the
structures of the isolated products it is apparent that the
initial organoyttrium hydride insertion occurs exclusively
at one of the two (equivalent) less hindered allyl substit-
(11) Negishi, E.-I.; J ensen, M. D.; Kondakov, D. Y.; Wang, S. J . Am.
Chem. Soc. 1994, 116, 8404.
(13) Li, Y.; Marks, T. J . J . Am. Chem. Soc. 1996, 118, 707.

Stereoselective Synthesis of Carbobicycles
J . Org. Chem., Vol. 63, No. 7, 1998 2299
Sch em e 3
Sch em e 4
triene 55 as a 1:1 diastereomeric mixture. Following
extensive purification by flash chromatography, both
diastereomers could be isolated in pure form.
Triene substrate 55a was first reacted under the
organoyttrium-catalyzed cyclization/silylation conditions.
In the presence of
5 mol % of the precatalyst
Cp*2YCH3‚THF, the reaction with PhSiH3 was complete
in 1 h at room temperature providing bicyclic silane 56
in 72% yield as a 2.1:1 mixture of diastereomers, pre-
sumably at the nonbridgehead stereocenter (eq 31).
Unfortunately, the two diastereomers could not be sepa-
rated, and all attempts at oxidation of this silane mixture
resulted in decomposition of the products. Therefore, the
relative stereochemistry of the major product could not
be established with certainty.
diastereomer as shown in Scheme 3. The isolation of a
single diastereomer also suggests that the second 5-exo
cyclization is extremely selective as was assumed for the
reaction of triene 43. The outstanding selectivity can be
attributed to the preference for the two chairlike transi-
tion structures that minimize unfavorable steric interac-
tions between the bulky Cp* rings of the catalyst and
the substrate. Again, no uncyclized or partially cyclized
hydrosilylated products were observed, indicating that
the second cyclization event is faster than silane trap-
ping. The generation of 3-substituted trans-bicyclo-
[3.3.0]octanes via a tandem anionic cyclization reaction
has been disclosed by Bailey and co-workers.14 The
results herein demonstrate that the organoyttrium-
catalyzed cyclization/silylation reaction is also well suited
for the selective synthesis of the strained trans 5,5-ring
system. Finally, the isolation of the thermodynamically
less favored trans-bicyclo[3.3.0] system speaks to the
kinetic nature of the cyclization reactions. It has been
noted previously that olefin insertion reactions of orga-
nolanthanides are extremely exothermic and effectively
irreversible under reasonable reaction conditions.15
To demonstrate the versatility of this method further,
an additional triene precursor of the trans 5,5-ring
system was synthesized. The triene 3-(tert-butyldimeth-
ylsiloxy)-4-ethenyl-1,6-heptadiene (55) was efficiently
prepared over five steps as outlined in Scheme 4. The
addition of the ester enolate 5016 to acrolein provided
alcohol 51. Protection of alcohol 51 as the TBDMS ether
52 and subsequent DIBALH reduction of the ester
provided alcohol 53. Oxidation of alcohol 53 to aldehyde
54 followed by a Wittig olefination provided the desired
The second diastereomer of triene 55, substrate 55b,
was also subjected to the annulation/silylation reaction
conditions. Reaction with PhSiH3 over the course of 1 h
at ambient temperature provided the bicyclic organosi-
lane product 57 in 78% yield as a 7.2:1 diastereomeric
mixture, again presumably at the nonbridgehead stereo-
center (eq 32). As in the case of substrate 55a , separation
of the diastereomers was not possible. Additionally, all
attempts at oxidation and TBDMS removal resulted in
product decomposition. Thus, the relative stereochem-
istry of the major product could again not be established
with certainty. The major silane products 56 and 57
depict the relative stereochemistry that would result from
both of the cyclizations proceeding through chairlike
transition structures.
To gain insight regarding the stereochemistry of these
cyclizations the TMS protecting group was again used
because it could be easily removed during oxidation. The
TBDMS protecting group in substrate 55a was removed
and replaced with a TMS group providing triene 58.
Reaction of substrate 58 with PhSiH3 in the presence of
5 mol % of the precatalyst Cp*2YCH3‚THF provided the
crude silane 59 as a 4:1 mixture of diastereomers (eq 33).
Direct oxidation of the silane mixture 59 provided the
expected diol 60 in 37% yield over two steps as a 4:1
mixture of diastereomers at the nonbridgehead stereo-
center (eq 34). The major diastereomer was isolated and
the structure was determined unambiguously from a
single-crystal X-ray structure. From the cyclization
result of triene 47 it is likely that the annulation/
(14) (a) Bailey, W. F.; Khanolkar, A. D. Tetrahedron Lett. 1990, 31,
5993. (b) Bailey, W. F.; Khanolkar, A. D.; Gavaskar, K. V. J . Am.
Chem. Soc. 1992, 114, 8053.
(15) Haar, C. M.; Stern, C. L.; Marks, T. J . Organometallics 1996,
15, 1765.
(16) Procedure adapted from: Lotz, B. T.; Miller, M. J . J . Org. Chem.
1993, 58, 618.

2300 J . Org. Chem., Vol. 63, No. 7, 1998
Molander et al.
Sch em e 5
silylation reactions of substrates 55a and 55b, which both
contain vinyl groups at a tertiary center, provide trans-
fused 5,5-rings exclusively (Scheme 3). Therefore, the
diastereomeric mixtures observed in the above reactions
must originate from the second 5-exo-trig cyclization
event where the transition structure of the second ring
closure is apparently influenced by the sterically bulky
ether substituent. Nonetheless, these reactions serve to
demonstrate that highly functionalized trans-fused 5,5-
bicycles can be prepared using the organoyttrium-
catalyzed cyclization/silylation protocol.
oxidation and melting point comparison of the derived
acid with known literature values18 verified the stereo-
chemistry shown. A similar cyclization of triene 63 was
reported by Negishi and co-workers using catalytic Ti(i-
OPr)4 in the presence of Et2AlCl.11 This cascade carbo-
metalation reaction required 3 days for completion and
gave a 3.3:2.3:1 ratio of three diastereomeric alcohols 65.
Attention was next focused on the synthesis of trans-
decalin ring systems from triene precursors. The first
precursor examined was analogous to substrate 47.
Substrate 61 was prepared in two steps from 1,8-
nonadien-5-one17 by vinylmagnesium bromide addition
and subsequent TMS protection. Reaction of substrate
61 under the standard reaction conditions provided only
uncyclized material. When triene 61 was reacted with
2 equiv of PhMeSiH2 over the extended reaction time of
25 h at room temperature the disilylated product 62 was
cleanly isolated in 89% yield (eq 35). This result was
puzzling for a number of reasons. First, the analogous
substrate, triene 47, cyclized very efficiently to generate
the strained trans 5,5 ring system. Next, the reaction
time of 25 h was abnormally long considering most
cyclization/silylation reactions using the organoyttrium
precatalyst Cp*2YCH3‚THF were complete within 1 h.
These results and the results observed for cyclic sub-
strates 9 and 11 confirmed that cyclization onto a vinyl
group at a quaternary center to form a six-membered ring
is difficult using this protocol.
The origin of stereoselectivity in the cyclization of
substrate 63 is very similar to that found for the
cyclization of substrate 47. Initial insertion of the
organoyttrium hydride occurs exclusively at one of the
least hindered olefins (Scheme 5). The first cyclization
event occurs through a chairlike transition structure
providing a cyclized organoyttrium intermediate.
A
second, rapid cyclization then ensues, providing a bicyclic
organoyttrium species. This intermediate undergoes a
σ-bond metathesis reaction with PhMeSiH2, liberating
the product and regenerating the active catalyst. These
results suggest that the organoyttrium-catalyzed cycliza-
tion/silylation method may be used to prepare a variety
of functionalized trans-decalin ring systems in a highly
selective fashion starting with simple polyene precursors.
Con clu sion s
The reactions described herein serve to demonstrate
the usefulness and versatility of the organoyttrium-
catalyzed cyclization/silylation method. A wide variety
of polycyclic molecules were prepared from simple pre-
cursors in a highly selective manner using this reaction
protocol. The intrinsic steric bulk of the catalyst dictates
the mode of cyclization and accounts for the high dias-
tereoselectivies observed in these reactions. The cycliza-
tion and subsequent silylation reaction of acyclic precur-
sors best illustrates this selectivity as three new
stereocenters are created with complete selectivity. The
bicyclic organosilanes generated by this method were
The next trans-decalin precursor examined was the
known triene 63.11 Reaction of substrate 63 using the
standard reaction protocol provided the bicyclic silane 64
in 68% yield as a diastereomeric mixture (eq 36). As in
previous six-membered ring cyclizations, the use of
PhMeSiH2 was required to prevent product dimerization.
Thus, the diastereoselectivity of the ring closure could
only be determined after oxidation of the silane stereo-
center. Oxidation of silane 64 provided alcohol 65 in 85%
yield as a single diastereomer (eq 37). Subsequent PDC
(18) (a) Chapman, N. B.; Shorter, J .; Toyne, K. J . J . Chem. Soc. 1964,
1077. (b) Dauben, W. G.; Tweit, R. L.; Mannerskantz, C. J . Am. Chem.
Soc. 1954, 76, 4420. (c) DiBiase, S. A.; Wolak, R. P.; Dishong, D. M.;
Gokel, G. W. J . Org. Chem. 1980, 45, 3630.
(17) Nagai, M.; Lazor, J .; Wilcox, C. S. J . Org. Chem. 1990, 55, 3440.

Stereoselective Synthesis of Carbobicycles
J . Org. Chem., Vol. 63, No. 7, 1998 2301
cyclization/silylation procedure given. The reaction was found
to be complete in 1 h by gas chromatography. Following
workup, the crude product was agitated in a Kugelrohr
distillation apparatus at 70 °C at reduced pressure to remove
any low molecular weight impurities. This provided 4 as a
clear oil (0.092 g, 0.22 mmol, 86% yield) as what appeared to
be a single diastereomer by gas chromatography: Rf 0.49
easily oxidized in most cases using known literature
procedures to provide the more versatile alcohols in good
to modest yields. The results outlined above demonstrate
the utility of the organoyttrium-catalyzed cyclization/
silylation method in the selective preparation of useful,
cyclic organic molecules.
1
(hexanes); H NMR (400 MHz, CDCl3) δ 7.56-7.50 (m, 2H),
7.37-7.31 (m, 3H), 5.04-4.92 (m, 1H), 4.46-4.36 (m, 2H),
2.08-1.96 (m, 1H), 1.79-1.58 (m, 5H), 1.57-1.44 (m, 2H),
1.32-1.06 (m, 5H), 1.04-0.82 (m, 11H), 0.78-0.58 (m, 3H),
0.32 (d, J ) 3.8 Hz, 3H), 0.1-0.02 (m, 6H); 13C NMR (100 MHz,
CDCl3) δ 146.7, 137.5, 134.3, 134.2, 129.0, 127.8, 111.6, 52.9,
52.8, 43.8, 40.1, 36.9, 35.4, 35.3, 30.6, 26.3, 26.1, 26.0, 21.2,
18.8, 18.7,-1.9, -2.3, -4.2, -4.8; IR (neat) 2928, 2117, 1250
cm-1; HRMS Calcd for C26H43OSi2+(M - H)+: 427.2853, found
427.2790; LRMS (EI) m/z 371 (31.2), 195 (47.3), 121 (100.0),
105 (16.1), 91 (12.9).
Exp er im en ta l Section
Ma ter ia ls a n d Meth od s. All catalytic organoyttrium
reactions were performed in a nitrogen-filled Vacuum Atmo-
spheres glovebox unless otherwise noted. All substrates were
extensively purified by flash chromatography and/or distilla-
tion. Prior to use, all substrates were freeze/pump/thaw-
degassed. Cyclohexane was purchased from Aldrich Chemical
Co., distilled from Na, and degassed before using. The
organoyttrium precatalyst Cp*2YCH3‚THF was synthesized
from the literature procedure19 using Schlenk techniques.
Phenylsilane (Aldrich, 99%) was degassed and stored in the
glovebox. Methylphenylsilane was purchased from United
Chemical Technologies and freeze/pump/thaw-degassed prior
to storage in a glovebox. Diethyl ether and THF were distilled
from sodium benzophenone ketyl under argon immediately
prior to use. CH2Cl2 was distilled from CaH2 immediately
prior to use. Triethylamine, DMSO, and DMF were all
distilled from CaH2 and stored over molecular sieves. Potas-
sium hydride was purchased as a 35% dispersion in mineral
oil from Aldrich. The white solid was isolated under argon
after several washings with dry hexane using Schlenk tech-
niques.20 Deuterated NMR solvents (CDCl3) were purchased
from Cambridge Isotope Laboratories and stored over 4 Å
molecular sieves. All other materials were commercially
available and were used without further purification unless
otherwise noted.
Gen er a l P r oced u r e for th e Or ga n oyttr iu m -Ca ta lyzed
Cycliza tion /Silyla tion Rea ction . In a nitrogen atmosphere
glovebox Cp*2YCH3‚THF (5 mol %) was dissolved in cyclohex-
ane (0.5 M) prior to the addition of either phenylsilane of
methylphenylsilane (at least 1.1 equiv) and the substrate (1
equiv). Reaction progress was monitored by gas chromatog-
raphy. Upon completion, the reaction mixture was removed
from the glovebox and filtered through a small plug of Florisil
to remove the catalyst prior to concentration by rotary
evaporation. The crude product was purified by flash chro-
matography. Concentration under reduced pressure (0.06
mmHg) or Kugelrohr distillation provided the isolated silane
products.
Gen er a l P r oced u r e for th e Woer p el Oxid a tion of
Sila n e P r od u cts.4 To a dry 100 mL round-bottom flask
equipped with a stopcock sidearm was added a 90% solution
of tert-butyl hydroperoxide (10 equiv) and DMF (2 mL per 0.5
mmol silane) under argon. Next, solid potassium hydride (10
equiv) was slowly added at room temperature using Schlenk
techniques under an atmosphere of argon. The addition was
exothermic and provided a slightly pink foamy suspension. To
this was added the silane (1 equiv) in DMF (0.2 M). The
mixture was then stirred for 12 h at 50 °C. Sodium thiosulfate
(1 g per 0.5 mmol silane) was then added, and the mixture
was stirred for 0.5 h. The reaction mixture was then exposed
to reduced pressure (0.06 mmHg) to remove the DMF. The
resulting solid residue was extracted several times with
chloroform and filtered through Celite. The filtrate was
concentrated by rotary evaporation and purified by flash
chromatography. Concentration under reduced pressure (0.06
mmHg) or Kugelrohr distillation provided the isolated alcohol
products.
(1R*,2R*,6S*,7S*)-2-(ter t-Bu tyld im eth ylsiloxy)-2-eth e-
n yl-7-(h yd r oxym eth yl)bicyclo[4.4.0]d eca n e (5). Silane 4
(73 mg, 0.17 mmol) was exposed to the Woerpel oxidation
procedure given. Following workup, the crude product was
purified by flash chromatography. Concentration under re-
duced pressure (0.06 mmHg) provided 5 (51.5 mg, 0.16 mmol,
93% yield) as a single stereoisomer which was >99% pure by
1
gas chromatography: Rf 0.30 (5:1 hexanes:EtOAc); H NMR
(400 MHz, CDCl3) δ 6.15 (dd, J ) 17.6, 11.2 Hz, 1H), 5.05-
4.92 (m, 2H), 3.70 (dd, J ) 10.7, 3.2 Hz, 1H), 3.48 (dd, J )
10.7, 6.2 Hz, 1H), 1.92-1.83 (m, 1H), 1.80-1.61 (m, 5H), 1.60-
1.46 (m, 2H), 1.45-1.32 (m, 1H), 1.31-1.01 (m, 5H), 0.99-
0.76 (m, 11H), 0.02 (d, J ) 4.0 Hz, 6H); 13C NMR (100 MHz,
CDCl3) δ 146.6, 111.8, 76.4, 65.9, 52.6, 45.5, 38.2, 36.6, 30.2,
29.8, 26.1, 25.9, 25.7, 20.9, 18.8, -1.9, -2.3; IR (neat) 3336,
2929, 2855, 1252 cm-1; HRMS Calcd for C19H36O2Si+: 324.2485,
found 324.2502; LRMS (EI) m/z 324 (0.5), 267 (75.3), 175 (33.3),
157 (18.2), 119 (22.0), 105 (22.1), 93 (39.0), 75 (100.0).
(1R *,2R *,6S *,7S *)-7-E t h e n y l-2-(h y d r o x y m e t h y l)-
bicyclo[4.4.0]n on a n -7-ol (8). (1R*,2R*,3R*)-1,3-Diethenyl-
2-(2-propenyl)-1-(trimethylsiloxy)cyclopentane (6) (62 mg, 0.25
mmol) was exposed to the general cyclization/silylation pro-
cedure given. The reaction mixture changed from faint yellow
to a darker, cloudy yellow over 1 h after which no starting
material was observed by gas chromatography. Following
workup, the crude silane 7 was directly subjected to oxidation.
Crude silane 7 in 2.5 mL was subjected to the Woerpel
oxidation procedure given. Following workup, the crude
product was purified by flash chromatography. Concentration
under reduced pressure (0.06 mmHg) provided 8 (35.2 mg, 0.18
mmol, 72% yield) as a 6:1 mixture of diastereomers which was
>99% pure by gas chromatography. Spectral data reported
1
for the major diastereomer: Rf 0.15 (1:1 hexanes:EtOAc); H
NMR (400 MHz, CDCl3) δ 5.85 (dd, J ) 17.1, 10.7 Hz, 1H),
5.22 (dd, J ) 17.1, 1.3 Hz, 1H), 5.04 (dd, J ) 10.7, 1.3 Hz,
1H), 3.65 (dd, J ) 10.7, 4.3 Hz, 1H), 3.52-3.38 (m, 1H), 2.02-
1.73 (m, 4H), 1.72-1.38 (m, 5H), 1.37-1.14 (m, 3H), 1.13-
0.76 (m, 3H); 13C NMR (100 MHz, CDCl3) δ 144.3, 111.6, 81.5,
67.1, 54.8, 46.5, 43.9, 39.3, 29.2, 27.6, 25.6, 23.7; IR (neat) 3374,
2925, 2857 cm-1; HRMS Calcd for C12H20O2+: 196.1463, found
196.1471; LRMS (EI) m/z 196 (5.6), 147 (17.4), 120 (18.6), 119
(18.6), 109 (18.0), 108 (21.9), 95 (26.2), 79 (2.4), 70 (44.6), 55
(100.0).
(1R*,2S*)-1-(ter t-Bu tyld im eth ylsiloxy)-1-eth en yl-2-[3-
(m eth ylp h en ylsilyl)p r op yl]cycloh exa n e (10). (1R*,2S*)-
1-tert-(Butyldimethylsiloxy)-1-ethenyl-2-(2-propenyl)cyclohex-
ane (9) (0.141 g, 0.50 mmol) was exposed to the general
cyclization/silylation procedure given. The reaction mixture
changed from clear to light yellow over 1 h and was found to
be complete. Following workup, the crude product was
agitated in a Kugelrohr distillation apparatus at 100 °C at
reduced pressure to remove any low molecular weight impuri-
ties. This provided 10 as a clear oil (0.2008 g, 0.498 mmol,
99% yield): 1H NMR (400 MHz, CDCl3) δ 7.58-7.49 (m, 2H),
7.41-7.31 (m, 3H), 6.13 (dd, J ) 17.8, 11.2 Hz, 1H), 5.03-
4.97 (m, 2H), 4.37-4.30 (m, 1H), 1.73-1.58 (m, 3H), 1.58-
1.39 (m, 4H), 1.39-1.09 (m, 6H), 0.90 (s, 9H), 0.84-0.72 (m,
(1R*,2R*,6S*,7S*)-2-(ter t-Bu tyld im eth ylsiloxy)-2-eth e-
n yl-7-[(m eth ylph en ylsilyl)m eth yl]bicyclo[4.4.0]decan e (4).
Diene 3 (0.076 g, 0.25 mmol) was exposed to the general
(19) den Haan, K. H.; de Boer, J . L.; Teuben, J . H.; Smeets, W. J .
J .; Spek, A. L. J . Organomet. Chem. 1987, 327, 31.
(20) Shriver, D. F.; Drezdzon, M. A. The Manipulation of Air-
Sensitive Compounds; Wiley-Interscience: New York, 1986.

2302 J . Org. Chem., Vol. 63, No. 7, 1998
Molander et al.
2H), 0.31 (d, J ) 3.8 Hz, 3H), 0.04 (s, 6H); 13C NMR (100 MHz,
CDCl3) δ 146.6, 136.8, 134.3, 134.2, 129.3, 129.1, 127.9, 127.8,
111.9, 47.3, 37.1, 33.4, 33.3, 27.1, 26.2, 26.1, 25.9, 22.5, 22.4,
21.7, 18.8, 13.7, 13.6, -1.9, -2.2, -5.5, -5.7; IR (neat) 2114
cm-1; HRMS Calcd for C24H42OSi2+: 402.2774, found 402.2780;
LRMS (EI) m/z 402 (4.7), 197 (11.0), 196 (19.4), 195 (100.0),
121 (37.2), 75 (43.7). Anal. Calcd for C24H42OSi2: C, 71.57%;
H, 10.51%. Found: C, 71.74%; H, 10.81%.
darker, cloudy yellow over 1 h after which no starting material
was observed by gas chromatography. Following workup, the
crude product was purified by flash chromatography. Con-
centration under reduced pressure (0.06 mmHg) provided
(1R *,5S *,6R *)-1-(t r im et h yls iloxy)-5-[(m et h ylp h en yl-
silyl)methyl]bicyclo[4.4.0]decane (19) (157 mg, 0.44 mmol, 88%
yield) as a 2:1 mixture of diastereomers which also contained
another unidentified, cyclized isomer (approximately 18% of
the isolated material by GC). In an attempt to determine the
identity of this other isomer, an oxidation was performed on
the silane mixture.
(1R*,2S*)-1-Eth en yl-2-[3-(m eth ylp h en ylsilyl)p r op yl]-1-
(tetr a h yd r op yr a n yloxy)cycloh exa n e (12). (1R*,2S*)-1-
Ethenyl-2-(2-propenyl)-1-(trimethylsiloxy)cyclohexane (11) (0.130
g, 0.52 mmol) was subjected to the general cyclization/silylation
procedure given. The reaction mixture changed from faint
yellow to dark yellow over 1 h and was found to be complete.
Following workup, the crude product was purified by flash
chromatography. The product was then agitated in a Kugel-
rohr distillation apparatus at 100 °C at reduced pressure to
remove any low molecular weight impurities. This provided
12 as a clear oil (0.1914 g, 0.51 mmol, 98% yield): Rf 0.38 (10:1
hexanes: EtOAc); 1H NMR (400 MHz, CDCl3) δ 7.57-7.47 (m,
2H), 7.39-7.28 (m, 3H), 6.37 (dd, J ) 17.8, 11.3 Hz, 0.5H),
5.86 (dd, J ) 17.8, 11.3 Hz, 0.5H), 5.16-4.99 (m, 2H), 4.65-
4.59 (m, 1H), 4.32 (q, J ) 3.5 Hz, 1H), 4.03-3.93 (m, 1H), 3.48-
3.39 (m, 1H), 1.93-1.10 (m, 19H), 0.92-0.67 (m, 2H), 0.31 (dd,
J ) 3.8, 1.1 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 144.8,
143.8, 136.8, 134.3, 129.1, 127.8, 114.6, 113.2, 94.4, 93.4, 80.2,
79.6, 63.1, 62.8, 47.5, 46.5, 33.2, 33.1, 32.6, 32.4, 32.3, 26.9,
26.8, 25.8, 25.7, 25.3, 22.6, 22.5, 21.7, 20.1, 20.0, 13.6,
13.5, -5.6, -5.7; IR (neat) 2112 cm-1; HRMS Calcd for
The mixture of 19 and the unknown cyclized silane isomer
(157 mg, 0.44 mmol) was subjected to the Woerpel oxidation
procedure given. Following workup, the major and minor
products were separated by flash chromatography. Concen-
tration of the major product under reduced pressure (0.06
mmHg) provided 20 (39.1 mg, 0.21 mmol, 48% yield) which
was >99% pure by gas chromatography. The minor product
was also isolated (11.1 mg, 0.06 mmol, 13.6% yield), but its
structure was ambiguous. Characterization data is given for
the major product: mp 110 °C; Rf 0.19 (3:1 EtOAc:hexanes);
1H NMR (400 MHz, CDCl3) δ 3.72 (dd, J ) 10.7, 4.3 Hz, 1H),
3.65-3.57 (m, 1H), 1.81-1.53 (m, 9H), 1.51-1.21 (m, 9H); 13
C
NMR (100 MHz, CDCl3) δ 71.7, 66.5, 44.3, 39.8, 22.7; IR (neat)
3347, 2927, 2861 cm-1; HRMS Calcd for C11H20O2+: 184.1463,
found 184.1471; LRMS (EI) m/z 184 (8.3), 153 (100.0), 111
(28.0), 97 (23.3). Anal. Calcd for C11H20O2: C, 71.69; H, 10.94.
Found: C, 71.77; H, 10.69.
( 1 R *,5 R *,6 S *) -1 -( t e r t -B u t y l d i m e t h y l s i l o x y ) -5 -
[(m eth ylph en ylsilyl)m eth yl]bicyclo[4.4.0]decan e (22). The
diene (1R*,2S*)-1-(tert-butyldimethylsiloxy)-2-ethenyl-1-(2-pro-
penyl)cyclohexane (21) (143 mg, 0.51 mmol) was exposed to
the general cyclization/silylation procedure. The reaction
mixture changed from faint yellow to a darker, cloudy yellow
over 1 h after which no starting material was observed by gas
chromatography. Following workup, the crude product was
purified by flash chromatography. Concentration under re-
duced pressure (0.06 mmHg) provided 22 as a 1.5:1 mixture
of diastereomers (191 mg, 0.473 mmol, 93% yield) which was
>98% pure by gas chromatography: Rf 0.53 (hexanes); 1H
NMR (400 MHz, CDCl3) δ 7.55-7.48 (m, 2H), 7.36-7.28 (m,
3H), 4.43-4.37 (m, 1H), 1.86-1.74 (m, 1H), 1.73-1.51 (m, 7H),
1.43-1.31 (m, 2H), 1.27-1.10 (m, 4H), 1.10-0.96 (m, 2H),
0.93-0.84 (m, 9H), 0.83-0.74 (m, 1H), 0.54-0.39 (m, 1H),
0.32-0.28 (m, 3H), 0.09-0.03 (m, 6H); 13C NMR (100 MHz,
CDCl3) δ 137.4, 137.2, 134.3, 129.0, 128.9, 127.8, 74.6, 74.5,
53.5, 53.4, 40.1, 40.0, 35.7, 35.6, 33.3, 26.8, 26.7, 26.3, 26.2,
24.4, 21.7, 21.6, 19.1, 19.0, 17.9, 17.6, -1.79, -1.83, -1.87,
-4.44, -5.12; IR (neat) 2927, 2854, 2116 cm-1; HRMS Calcd
for C24H42OSi2+: 402.2774, found 402.2754; LRMS (EI) m/z 402
(0.2), 346 (25.1), 345 (56.1), 270 (44.8), 269 (80.0), 201 (31.7),
196 (51.8), 195 (96.4), 191 (65.1), 122 (38.0), 121 (100.0), 107
(20.2), 105 (41.3), 91 (33.5).
C
23H36O2Si+: 372.2485, found 372.2507; LRMS (EI) m/z 372
(0.1), 371 (0.1), 243 (12.1), 193 (13.8), 165 (15.0), 137 (34.6),
122 (13.5), 121 (75.0), 105 (12.9), 86 (17.3), 85 (100.0), 84 (33.7).
Anal. Calcd for C23H36O2Si: C, 74.13%; H, 9.74%. Found: C,
73.84%; H, 9.74%.
( 1 R *,5 S *,6 R *) -1 -( t e r t -B u t y l d i m e t h y l s i l o x y ) -5 -
[(m et h ylp h en ylsilyl)m et h yl]b icyclo[4.4.0]d eca n e (16).
(1R*,2R*)-1-(tert-Butyldimethylsiloxy)-2-ethenyl-1-(2-propenyl-
)cyclohexane (15) (147 mg, 0.52 mmol) was exposed to the
general cyclization/silylation procedure given. The reaction
mixture changed from faint yellow to a darker, cloudy yellow
over 1 h after which no starting material was observed by gas
chromatography. Following workup and purification, Kugel-
rohr distillation provided 16 as a 1.6:1 mixture of diastereo-
mers (194 mg, 0.48 mmol, 92% yield) which also contained 6%
of inseparable isomeric impurities by gas chromatography: at
160 °C/0.06 mmHg; Rf 0.48 (hexanes); 1H NMR (400 MHz,
CDCl3) δ 7.53-7.49 (m, 2H), 7.36-7.30 (m, 3H), 4.43-4.36 (m,
1H), 1.91-0.90 (m, 17H), 0.87 (s, 9H), 0.68-0.49 (m, 1H),
0.35-0.28 (m, 3H), 0.06-0.01 (m, 6H); 13C NMR (100 MHz,
CDCl3) δ 137.2, 134.2, 129.1, 127.8, 75.5, 51.0, 32.8, 26.1, 21.9,
18.6, -1.5, -1.8, -4.9; IR (neat) 2932, 2118 cm-1; HRMS Calcd
+
for C24H41OSi2 (M - H)+: 401.2696, found 401.2674; LRMS
(EI) m/z 196 (20.2), 195 (100.0), 121 (43.4), 105 (15.0).
( 1 R *,5 S *,6 R *) -1 -( t e r t -B u t y l d i m e t h y l s i l o x y ) -5 -
(h yd r oxym eth yl)bicyclo[4.4.0]d eca n e (17). Silane 16 (161
mg, 0.40 mmol) in 2 mL of DMF was subjected Woerpel
oxidation conditions given previously. Following workup, the
crude alcohol was purified by flash chromatography. After
prolonged exposure to reduced pressure (0.06 mmHg), 17 (84
mg, 0.28 mmol, 70% yield) was isolated as a clear oil which
was >96% pure by gas chromatography: ot 152 °C/0.06 mmHg;
Rf 0.29 (2:1 hexane: Et2O); 1H NMR (400 MHz, CDCl3) δ 3.67
(dd, J ) 10.7, 3.7 Hz, 1H), 3.54-3.46 (m, 1H), 1.90-1.63 (m,
6H), 1.53-1.12 (m, 11H), 0.87 (s, 9H), 0.06 (s, 6H); 13C NMR
(100 MHz, CDCl3) δ 75.5, 65.8, 44.8, 26.1, 22.1, 18.5, -1.5,
-1.8; IR (neat) 3336, 2933, 2858 cm-1; HRMS Calcd for
( 1 R *,5 R *,6 S *) -1 -( t e r t -B u t y l d i m e t h y l s i l o x y ) -5 -
(h yd r oxym eth yl)bicyclo[4.4.0]d eca n e (23). Silane 22 (172
mg, 0.43 mmol) in 2 mL of DMF was subjected to the Woerpel
oxidation procedure given. Following workup and purification,
Kugelrohr distillation provided 23 as a 17.8:1 mixture of
isomers (86.5 mg, 0.29 mmol, 68% yield) which was >98% pure
by gas chromatography: ot 152 °C/0.06 mmHg; Rf 0.30 (3:1
1
hexanes:EtOAc); H NMR (400 MHz, CDCl3) δ 3.59 (dd, J )
10.7, 3.2 Hz, 1H), 3.52 (dd, J ) 10.7, 5.7 Hz, 1H), 1.77-0.96
(m, 17H), 0.91 (s, 9H), 0.08 (d, J ) 1.9 Hz, 6H); 13C NMR (100
MHz, CDCl3) δ 74.3, 65.6, 47.6, 40.0, 39.8, 38.9, 29.9, 26.5,
26.4, 24.2, 21.4, 21.0, 19.1, -1.80, -1.83; IR (neat) 3331, 2928,
2855 cm-1; HRMS Calcd for C17H34O2Si+: 298.2328, found
298.2316; LRMS (EI) m/z 298 (0.1), 267 (12.4), 150 (24.9), 149
(100.0), 107 (15.4).
C
17H34O2Si+: 298.2328, found 298.2328; LRMS (EI) m/z 267
(20.3), 159 (12.3), 150 (26.7), 149 (100.0), 145 (12.9), 107 (33.0).
Anal. Calcd for C17H34O2Si: C, 68.39; H, 11.48. Found: C,
69.38; H, 11.30.
(1R *,5S *,6R *)-5-(H y d r o x y m e t h y l)b i c y c lo [4.4.0]-
d eca n ol (20). The diene (1R*,2R*)-1-(trimethylsiloxy)-2-
ethenyl-1-(2-propenyl)cyclohexane (18) (119 mg, 0.50 mmol)
was reacted under the general cyclization/silylation conditions
given. The reaction mixture changed from faint yellow to a
(1R *,5R *,6S *)-5-[(M e t h y lp h e n y ls i ly l)m e t h y l]-1-
(t r im et h ylsiloxy)b icyclo[4.4.0]d eca n e (25). The diene
(1R*,2S*)-2-ethenyl-1-(2-propenyl)-1-(trimethylsiloxy)cyclohex-
ane (24) (121 mg, 0.51 mmol) was subjected to the general
cyclization/silylation procedure given. The reaction mixture
changed from faint yellow to a darker, cloudy yellow over 1 h
after which no starting material was observed by gas chro-

Stereoselective Synthesis of Carbobicycles
J . Org. Chem., Vol. 63, No. 7, 1998 2303
matography. Following workup and purification by flash
chromatography, concentration under reduced pressure (0.06
mmHg) at 80 °C provided 25 as an approximate 1.5:1 mixture
of diastereomers (162 mg, 0.45 mmol, 89% yield) which was
>98% pure by gas chromatography: Rf 0.41 (hexanes); 1H
NMR (400 MHz, CDCl3) δ 7.56-7.48 (m, 2H), 7.38-7.29 (m,
3H), 4.43-4.36 (m, 1H), 1.81-1.30 (m, 10H), 1.25-0.85 (m,
6H), 0.80-0.68 (m, 1H), 0.57-0.41 (m, 1H), 0.35-0.27 (m, 3H),
0.13-0.01 (m, 9H); 13C NMR (100 MHz, CDCl3) δ 137.6, 134.3,
129.0, 127.7, 74.8, 74.7, 52.9, 40.0, 39.9, 35.7, 35.5, 33.4, 26.7,
26.6, 24.5, 21.8, 21.7, 17.7, 17.6, 2.6, 2.5, -4.3, -4.9; IR (neat)
2930, 2850, 2116 cm-1; HRMS Calcd for C21H36OSi2+: 360.2305,
found 360.2323; LRMS (EI) m/z 360 (4.7), 303 (22.6), 225 (81.8),
209 (78.1), 192 (35.9), 121 (100.0), 105 (14.9). Anal. Calcd
for C21H36Si2O: C, 69.93; H, 10.06. Found: C, 70.25; H, 10.36.
(1R *,5R *,6S *)-5-(H y d r o x y m e t h y l)b i c y c lo [4.4.0]-
d eca n ol (26). The 1.5:1 diastereomeric mixture of silanes 25
(140 mg, 0.39 mmol) in 2.0 mL of DMF was subjected to the
Woerpel oxidation procedure given. After workup and puri-
fication, concentration under reduced pressure (0.06 mmHg)
provided 26 (38 mg, 0.21 mmol, 53% yield) as a single
stereoisomer which was >99% pure by gas chromatography.
Recrystallization from chloroform and petroleum ether pro-
vided crystals suitable for X-ray crystallography. An X-ray
crystal structure provided unambiguous structure identifica-
tion of the major isomer: mp 111 °C; Rf 0.08 (3:1 hexanes:
HRMS Calcd for C10H18O2+: 170.1307, found 170.1303; LRMS
(EI) m/z 170 (7.9), 139 (100.0), 121 (12.1), 109 (11.0), 97 (68.1).
Anal. Calcd for C10H18O2: C, 70.55; H, 10.66. Found: C,
71.10; H, 10.72.
(1R*,5S*,6S*)-5-(Hydroxymethyl)bicyclo[4.3.0]nonanol: mp
88 °C; Rf 0.15 (1:1 hexanes:EtOAc); 1H NMR (400 MHz, CDCl3)
δ 3.4 (d, J ) 5.9 Hz, 2H), 2.02-1.85 (m, 2H), 1.84-1.72 (m,
2H), 1.71-1.57 (m, 4H), 1.56-1.25 (m, 5H), 1.14 (br s, 1H),
1.08-0.88 (m, 2H); 13C NMR (100 MHz, CDCl3) δ 78.8, 68.4,
47.7, 39.5, 38.8, 36.1, 29.2, 27.3, 24.6, 20.3; IR (neat) 3369,
2924, 2864, 1046 cm-1; HRMS Calcd for C10H18O2+: 170.1307,
found 170.1323; LRMS (EI) m/z 170 (21.0), 141 (52.4), 139
(67.4), 128 (61.8), 123 (47.8), 121 (14.6), 109 (28.4), 97 (100.0),
95 (46.6), 84 (66.0).
(1R *,5R */S *,6S *)-1-(t er t -B u t y ld i m e t h y ls i lo x y )-5-
[(m et h ylp h en ylsilyl)m et h yl]b icyclo[4.3.0]n on a n e (31).
(1R*,2S*)-1-(tert-Butyldimethylsiloxy)-2-ethenyl-1-(2-propenyl-
)cyclopentane (30) (67 mg, 0.25 mmol) was exposed to the
general cyclization/silylation procedure given. The reaction
mixture was initially yellow and faded to a light, cloudy yellow
over 1 h after which no starting material was observed by gas
chromatography. Following workup and purification, Kugel-
rohr distillation provided 31 (90 mg, 0.23 mmol, 92% yield) as
a 1.7:1 mixture of diastereomers which was >99% pure by gas
chromatography: ot 152 °C/0.06 mmHg; Rf 0.60 (hexanes); 1H
NMR (400 MHz, CDCl3) δ 7.57-7.49 (m, 2H), 7.38-7.29 (m,
3H), 4.45-4.38 (m, 1H), 2.06-1.86 (m, 1H), 1.85-1.28 (m,
11H), 1.24-1.09 (m, 1H), 1.08-0.90 (m, 2H), 0.88-0.56 (m,
10H), 0.36-0.29 (m, 3H), 0.09- -0.08 (m, 6H); 13C NMR (100
MHz, CDCl3) δ 137.3, 134.3, 129.1, 129.0, 127.8, 127.7, 82.1,
82.0, 57.8, 49.7, 47.4, 47.1, 38.7, 38.0, 36.7, 36.2, 35.9, 34.9,
34.8, 32.9, 29.7, 29.6, 27.7, 26.5, 26.1, 25.1, 22.3, 22.1, 22.0,
20.8, 20.0, 19.1, 18.7, 18.6, -2.4, -2.5, -2.6, -2.7, -4.7, -5.2;
IR (neat) 2115, 1046 cm-1; LRMS (EI) m/z 388 (0.2), 332 (18.1),
333 (44.9), 256 (30.5), 255 (59.8), 253 (51.9), 196 (48.4), 195
(82.4), 178 (40.8), 177 (51.3), 122 (44.8), 121 (100.0), 107 (15.2),
105 (35.4), 75 (90.8). Anal. Calcd for C23H40OSi2: C, 71.06;
H, 10.37. Found: C, 71.30; H, 10.42.
(1R *,5R */S *,6S *)-1-(t er t -B u t y ld i m e t h y ls i lo x y )-5-
(h yd r oxym eth yl)bicyclo[4.3.0]n on a n e (32). Silane 31 (72
mg, 0.185 mmol) was subjected to the Woerpel oxidation
conditions. Following workup, the products appeared to be a
1:1 mixture of diastereomeric alcohols by gas chromatography.
The diastereomers were purified, but could not be separated
by flash chromatography. Kugelrohr distillation provided 32
(41 mg, 0.16 mmol, 86% yield) as a 1:1 mixture of diastereo-
mers which was >99% pure by gas chromatography: ot 132°/
0.06 mmHg; 1H NMR (400 MHz, CDCl3) δ 3.65-3.36 (m, 2H),
2.04-1.89 (m, 1H), 1.84-1.11 (m, 12H), 1.10-0.84 (m, 11H),
0.09-0.04 (m, 6H); 13C NMR (100 MHz, CDCl3) δ 82.0, 81.5,
68.6, 67.2, 51.5, 50.1, 40.0, 39.2, 38.2, 38.0, 36.7, 36.0, 29.4,
29.0, 27.6, 26.1, 26.07, 26.0, 24.4, 21.4, 20.7, 20.5, 18.7, -2.3,
-2.4, -2.6; IR (neat) 3326, 2928, 2856, 1046 cm-1; LRMS (EI)
m/z 254 (3.1), 253 (13.7), 171 (40.8), 151 (19.5), 136 (46.2), 135
(100.0), 107 (38.4), 105 (16.1), 93 (73.7), 79 (60.5), 75 (86.1).
Anal. Calcd for C16H32O2Si: C, 67.54; H, 11.34. Found: C,
67.83; H, 11.55.
1,2,3,4-Tetr a h yd r o-1-n a p h th a len em eth a n ol (36). The
aromatic diene 1-ethenyl-2-(2-propenyl)benzene (33) (79 mg,
0.55 mmol) was exposed to the general cyclization/silylation
procedure given. The reaction mixture changed from faint
yellow to dark yellow over 1 h after which no starting material
was observed by gas chromatography. Following workup, the
crude product was purified by flash chromatography in hex-
anes and then concentrated to provide a 7:1 mixture of
cyclized/silylated products (34 and 35) (145 mg, 0.542 mmol,
99% yield) as a clear oil which was >97% pure by GC analysis.
The silane mixture was then diluted with chloroform (5 mL)
and cooled to 0 °C prior to the addition of HBF4‚OEt2 (169 mg,
10.4 mmol). This mixture was stirred for 1 h at 0 °C. The
mixture was then stirred for 15 min at 0 °C under reduced
pressure (0.04 mmHg) prior to warming to room temperature
for 5 min at reduced pressure. A light purple residue remained
to which 5 mL of both THF and MeOH were added followed
by KF (146 mg, 2.5 mmol), KHCO3 (303 mg, 3.0 mmol), and a
1
EtOAc); H NMR (400 MHz, CDCl3) δ 4.05 (dd, J ) 10.7, 3.1
Hz, 1H), 3.95 (dd, J ) 10.7, 5.9 Hz, 1H), 2.30 (br s, 1H), 2.26-
2.06 (m, 4H), 2.05-1.89 (m, 6H), 1.83 (br s, 1H), 1.79-1.66
(m, 3H), 1.65-1.55 (m, 3H); 13C NMR (100 MHz, CDCl3) δ 70.2,
65.1, 44.9, 40.0, 39.9, 39.2, 29.7, 26.1, 24.3, 21.2, 20.8; IR (neat)
3382, 2927, 2855 cm-1; HRMS Calcd for C11H20O2+: 184.1463,
found 184.1467; LRMS (EI) m/z 184 (8.9), 154 (10.8), 153
(100.0), 111 (18.7), 97 (22.4). Anal. Calcd for C11H20O2: C,
71.69; H, 10.94. Found: C, 71.96; H, 11.37.
(1R *,5R */S *,6S *)-5-[(Me t h ylp h e n ylsilyl)m e t h yl]-1-
(tr im eth ylsiloxy)bicyclo[4.3.0]n on a n e (28). (1R*,2S*)-2-
Ethenyl-1-(2-propenyl)-1-(trimethylsiloxy)cyclopentane (27) (88
mg, 0.39 mmol) was exposed to the general cyclization/
silylation procedure. The reaction mixture was initially bright
yellow and faded to light yellow over 10 min. After 45 min no
starting material was observed by gas chromatography. Fol-
lowing workup and purification, Kugelrohr distillation pro-
vided 28 (115 mg, 0.34 mmol, 88% yield) as a mixture of
diastereomers which was >98% pure by gas chromatography:
1
ot 136 °C/0.05 mmHg; Rf 0.44 (hexanes); H NMR (400 MHz,
CDCl3) δ 7.55-7.49 (m, 2H), 7.36-7.29 (m, 3H), 4.45-4.36 (m,
1H), 2.51-1.80 (m, 1H), 1.79-1.22 (m, 8H), 1.19-1.01 (m, 2H),
0.95-0.54 (m, 5H), 0.32-0.30 (m, 3H), 0.78-0.00 (m, 9H); 13
C
NMR (100 MHz, CDCl3) δ 134.3, 129.1, 129.0, 127.8, 127.7,
82.1, 81.9, 57.5, 49.3, 46.7, 38.7, 37.9, 36.7, 36.5, 34.7, 33.1,
29.8, 29.7, 27.6, 26.4, 25.1, 22.2, 22.0, 20.8, 20.0, 19.2, 2.1, 2.0,
-4.8, -4.9; IR (neat) 2116, 1052 cm-1; HRMS Calcd for
C
20H34OSi2+: 346.2148, found 346.2138; LRMS (EI) m/z 346
(6.6), 211 (38.9), 209 (73.3), 178 (41.2), 121 (100.0), 75 (38.7),
73 (72.9).
( 1 R *, 5 R */S *, 6 S *) -5 -( H y d r o x y m e t h y l ) b i c y c l o -
[4.3.0]n on a n ol (29). The silane 28 (94 mg, 0.28 mmol) in 1.5
mL of DMF was subjected to the Woerpel oxidation procedure
given. After workup, the products appeared to be a 1:1
mixture of diastereomeric diols by gas chromatography. The
diastereomers could be easily separated by flash chromatog-
raphy. Concentration under reduced pressure (0.06 mmHg)
provided a crystalline solid for the separate diastereomers of
29 (25 mg, 0.15 mmol, 51% yield). Recrystallization of one of
the diastereomers from chloroform provided crystals suitable
for X-ray crystallography. An X-ray crystal structure provided
unambiguous structure identification of this diastereomer.
(1R*,5R*,6S*)-5-(Hydroxymethyl)bicyclo[4.3.0]nonanol: mp 127
°C; Rf 0.21 (1:1 hexanes:EtOAc); 1H NMR (400 MHz, CDCl3) δ
3.64 (dd, J ) 10.7, 4.0 Hz, 1H), 3.44 (dd, J ) 10.4, 6.7 Hz,
1H), 1.94-1.72 (m, 4H), 1.71-1.55 (m, 5H), 1.54-1.37 (m, 2H),
1.36-1.21 (m, 2H), 1.20-1.09 (m, 1H), 1.08-0.92 (m, 2H); 13
NMR (100 MHz, CDCl3) δ 79.2, 67.0, 49.2, 39.7, 39.0, 36.3,
28.9, 25.8, 21.3, 20.1; IR (neat) 3331, 2936, 2849, 1042 cm-1
C
;

2304 J . Org. Chem., Vol. 63, No. 7, 1998
Molander et al.
30% H2O2 solution (1.2 g, 10.5 mmol). This mixture heated
at reflux for 18 h. The mixture was then cooled to room
temperature, and approximately 1 mL of water was added
followed by enough K2CO3 to make a paste. The organics were
separated, and the paste was extracted several times with
EtOAc. The combined organics were concentrated by rotary
evaporation, and the crude products were purified by flash
chromatography. Kugelrohr distillation provided a 14:1 mix-
ture of 36 and 1,2,3,4-tetrahydro-2-naphthalenemethanol (37)
(71.7 mg, 0.442 mmol, 81% yield) as a clear oil which was
>99% pure by GC analysis. The major product 36 could be
partially isolated by flash chromatography to allow for com-
plete characterization. The 1H NMR spectra were in agree-
ment with known literature spectra for 36:10 ot 104 °C/0.06
mmHg; Rf 0.13 (10:1 hexanes:EtOAc); 1H NMR (400 MHz,
CDCl3) δ 7.24-7.19 (m, 1H), 7.16-7.07 (m, 3H), 3.78 (d, J )
6.4 Hz, 2H), 2.76 (pent, J ) 5.6 Hz, 1H), 2.81-2.74 (m, 2H),
1.98-1.81 (m, 3H), 1.81-1.68 (m, 1H), 1.55 (br s, 1H); 13C NMR
(100 MHz, CDCl3) δ 138.2, 136.6, 129.3, 128.7, 126.1, 125.7,
67.1, 40.2, 29.7, 25.1, 19.7; IR (neat) 3334 br, 3060, 3016, 2930,
1035, 758 cm-1; HRMS Calcd for C11H14O+: 162.1045, found
162.1045; LRMS (EI) m/z 162 (10.7), 132 (12.0), 131 (100.0),
128 (10.2), 116 (10.4), 115 (12.9), 91 (25.8).
mmol, 80% yield) as a 4.4:1 mixture of diastereomers which
also contained 7% of other, unidentified isomers as indicated
by gas chromatography: ot 122 °C/0.06 mmHg; Rf 0.45
1
(hexanes); H NMR (400 MHz, CDCl3) δ 7.55-7.49 (m, 2H),
7.37-7.31 (m, 3H), 4.47-4.34 (m, 1H), 2.09-1.83 (m, 2H),
1.74-1.23 (m, 10H), 1.22-0.88 (m, 4H), 0.61-0.48 (m, 1H),
0.38-0.31 (m, 3H), 0.16-0.07 (m, 1H), 0.05-0.00 (m, 6H),
-0.07 to -0.10 (m, 1H); 13C NMR (100 MHz, CDCl3) δ 136.6,
134.4, 134.3, 129.2, 127.9, 85.2, 45.6, 44.5, 33.6, 30.2, 28.2, 25.9,
25.7, 21.8, 18.2, 2.1, 2.0, -5.6; IR (neat) 2118, 1036 cm-1
;
LRMS (EI) m/z 346 (16.4), 373 (13.0), 209 (73.5), 183 (100.0),
178 (33.9), 170 (31.7), 121 (93.0), 105 (14.4), 91 (19.9), 75 (35.3).
Anal. Calcd for C21H34OSi2: C, 70.32; H, 9.56. Found: C,
69.98; H, 9.59.
(1R*,2S*,5R*)-2-(Hyd r oxym eth yl)bicyclo[4.3.0]n on a n -
1-ol (42). In a nitrogen atmosphere glovebox, Cp*2YCH3‚THF
(6 mg, 0.013 mmol) was dissolved in benzene-d6 (0.5 mL) prior
to the addition of methylphenylsilane (35 mg, 0.28 mmol) and
38 (56 mg, 0.25 mmol). The reaction mixture was transferred
to a J . Young NMR tube and removed from the glovebox. The
mixture remained light yellow over a 7 h period after which
1
no starting material was observed by H NMR. The reaction
mixture was filtered through a small plug of Florisil to remove
the catalyst prior to concentration by rotary evaporation. After
further concentration under reduced pressure (0.06 mmHg),
an oxidation was performed on the crude silane 41.
A small amount of the minor product 37 was also isolated
1
allowing for NMR analysis. Both the H NMR and 13C NMR
spectra were in good agreement with known literature spectra
for 37:10 1H NMR (400 MHz, CDCl3) δ 7.10 (s, 4H), 3.62 (dd, J
) 6.4, 2.6 Hz, 2H), 2.94-2.79 (m, 3H), 2.50 (dd, J ) 16.5, 10.6
Hz, 1H), 2.07-1.91 (m, 2H), 1.58 (br s, 1H), 1.49-1.38 (m, 1H);
13C NMR (100 MHz, CDCl3) δ 136.7, 135.9, 129.2, 128.8, 125.6,
112.1, 67.7, 37.0, 32.4, 28.7, 25.9; IR (neat) 3342 br, 2920, 1065,
1024 cm-1; HRMS Calcd for C11H14O+: 162.1045, found
162.1073; LRMS (EI) m/z 162 (34.9), 144 (29.6), 129 (100.0),
116 (24.5).
The crude silane 41 (assume 0.25 mmol) in 1.5 mL of DMF
was subjected to the Woerpel oxidation conditions given.
Following workup and purification, concentration under re-
duced pressure (0.06 mmHg) provided the crystalline solid 42
as a single diastereomer (20 mg, 0.12 mmol, 48% yield) that
was >99% pure by gas chromatography. Recrystallization
from chloroform provided crystals suitable for X-ray crystal-
lography. An X-ray crystal structure provided unambiguous
structure identification of the diastereomers: mp 132 °C; Rf
0.47 (EtOAc); 1H NMR (400 MHz, CDCl3) δ 3.65-3.53 (m, 1H),
3.39-3.26 (m, 1H), 2.09-1.90 (m, 2H), 1.82 (d, J ) 12.6 Hz,
1H), 1.75-1.49 (m, 5H), 1.48-1.06 (m, 8H); 13C NMR (100
MHz, CDCl3) δ 79.7, 64.9, 51.7, 45.6, 33.7, 27.6, 25.5, 25.0,
21.1; IR (neat) 3322, 2939, 1061 cm-1; LRMS (EI) m/z 170 (0.4),
152 (15.3), 98 (100.0).
(1R *,2S *,5S *)-2-[(P h e n ylsilyl)m e t h yl]-1-(t r im e t h yl-
siloxy)bicyclo[4.3.0]n on a n e (39). The diene (1R*,2S*)-1,2-
diethenyl-1-(trimethylsiloxy)cyclohexane (38) (55 mg, 0.245
mmol) was subjected to the general cyclization/silylation
procedure. The reaction mixture was light yellow and became
cloudy yellow over 1 h after which no starting material was
observed by gas chromatography. Following workup, the crude
products were separated and purified by flash chromatogra-
phy. After exposure to reduced pressure (0.06 mmHg), the
major product 39 (37 mg, 0.12 mmol, 48% yield) was isolated
as an 8:1 mixture of diastereomers which also contained 6%
of other, unidentified isomers by gas chromatography: Rf 0.46
( 1 R *,3 S *,5 S *) -1 -( t e r t -B u t y l d i m e t h y l s i l o x y ) -3 -
[(p h en ylsilyl)m eth yl]bicyclo[3.3.0]octa n e (44). The triene
4-tert-(butyldimethylsiloxy)-4-ethenyl-1,6-heptadiene (43) (0.127
g, 0.50 mmol) was exposed to the general cyclization/silylation
procedure. The reaction mixture changed from faint yellow
to dark yellow over 1 h and was found to be complete.
Following workup, the crude products were purified by flash
chromatography using hexanes. The diastereoselectivity was
3:1 by gas chromatography. The major diastereomer could be
completely separated from the minor diastereomer and iso-
meric impurities. Kugelrohr distillation provided the clear oil
44 (0.0718 g, 0.2 mmol, 40% yield). The minor diastereomer
was separated along with other isomers. Kugelrohr distillation
provided the minor diastereomer [(1R*,3R*,5R*) isomer] as a
clear oil (0.0328 g, 0.09 mmol, 18% yield) which also contained
18% impurities by gas chromatography. The data are given
for the isolated major diastereomer: ot 110°/0.05 mmHg; Rf
1
(hexanes); H NMR (400 MHz, CDCl3) δ 7.58-7.51 (m, 2H),
7.40-7.31 (m, 3H), 4.40-4.24 (m, 2H), 2.04-1.80 (m, 2H),
1.78-0.70 (m, 13H), 0.65-0.53 (m, 1H), 0.16- -0.06 (m, 9H);
13C NMR (100 MHz, CDCl3) δ 135.2, 134.4, 132.5, 129.6, 128.0,
85.0, 45.6, 44.9, 33.6, 30.1, 28.1, 25.9, 25.6, 21.7, 15.2, 3.0, 2.6,
+
2.1; IR (neat) 2135, 1046 cm-1; HRMS Calcd for C19H32OSi2
:
332.1992, found 332.1967; LRMS (EI) m/z 332 (2.7), 259 (11.0),
195 (89.5), 183 (100.0), 121 (23.9), 107 (36.2), 105 (19.4).
A minor, uncyclized product, (1R*,2S*)-1-ethenyl-2-[(2-
phenylsilyl)ethyl]-1-(trimethylsiloxy)cyclohexane (40) (11.5 mg,
0.04 mmol, 15% yield) was also isolated: Rf 0.28 (hexanes);
1H NMR (400 MHz, CDCl3) δ 7.59-7.49 (m, 2H), 7.41-7.29
(m, 3H), 6.0 (dd, J ) 17.0, 11.0 Hz, 1H), 5.15-4.94 (m, 2H),
4.2 (t, J ) 3.6 Hz, 2H), 1.77-1.38 (m, 6H), 1.32-1.07 (m, 4H),
1.06-0.94 (m, 1H), 0.93-0.77 (m, 1H), 0.76-0.63 (m, 1H), 0.06
(s, 9H); 13C NMR (100 MHz, CDCl3) δ 146.2, 135.2, 132.9,
1
0.44 (hexanes); H NMR (400 MHz, CDCl3) δ 7.62-7.47 (m,
2H), 7.41-7.24 (m, 3H), 4.28-4.24 (t, J ) 3.7 Hz, 2H), 2.86-
2.73 (m, 1H), 2.17-2.03 (m, 1H), 2.02-1.88 (m, 3H), 1.81-
1.68 (m, 1H), 1.64-1.54 (m, 1H), 1.47-1.31 (m, 3H), 1.22-
1.09 (m, 3H), 1.04 (dd, J ) 12.6, 8.3 Hz, 1H), 0.82 (s, 9H), 0.02
(d, J ) 16.1 Hz, 6H); 13C NMR (100 MHz, CDCl3) δ 135.2,
132.8, 129.4, 127.9, 94.2, 56.0, 43.8, 40.3, 33.0, 32.5, 28.3, 25.8,
22.9, 20.0, 18.3, -3.1; IR (neat) 2134 cm-1; HRMS Calcd for
129.4, 127.9, 112.2, 77.7, 49.2, 37.1, 26.1, 25.5, 24.6, 21.8, 7.8,
+
2.5; IR (neat) 2131, 1046 cm-1; HRMS Calcd for C19H32OSi2
:
332.1992, found 332.1958; LRMS (EI) m/z 332 (2.3), 259 (14.7),
195 (63.0), 181 (25.0), 155 (33.5), 107 (63.8), 105 (32.0), 78
(45.5), 73 (100.0).
C
21H35OSi2+ (M - H)+: 359.2226, found 359.2219; LRMS (EI)
(1R *,2S *,5R *)-2-[(M e t h y lp h e n y ls i ly l)m e t h y l]-1-
(tr im eth ylsiloxy)bicyclo[4.3.0]n on a n e (41). (1R*,2S*)-1,2-
Diethenyl-1-(trimethylsiloxy)cyclohexane (38) (111 mg, 0.495
mmol) was exposed to the cyclization/silylation procedure
given. The reaction mixture was light yellow and became
cloudy yellow over 6 h after which no starting material was
observed by gas chromatography. Following workup and
purification, Kugelrohr distillation provided 41 (142 mg, 0.396
m/z 359 (0.1), 303 (15.1), 228 (10.2), 227 (44.9), 199 (11.5), 183
(16.0), 182 (27.3), 181 (100.0), 151 (18.5), 121 (11.5), 107 (35.3),
105 (19.8). Anal. Calcd for C21H36OSi2: C, 69.93%; H, 10.06%.
Found: C, 70.32%; H, 10.38%.
( 1 R *,3 S *,5 S *) -1 -( t e r t -B u t y l d i m e t h y l s i l o x y ) -3 -
(h yd r oxym eth yl)bicyclo[3.3.0]octa n e (46). Silane 44 (125
mg, 0.35 mmol) was subjected to the Woerpel oxidation
procedure given. The mixture was heated to 50 °C and stirred

Stereoselective Synthesis of Carbobicycles
J . Org. Chem., Vol. 63, No. 7, 1998 2305
for 5 h over which time it became a faint yellow color and
contained a white precipitate. Following workup, the products
were purified by flash chromatography. Concentration under
reduced pressure (0.06 mmHg) provided 46 (39.6 mg, 0.15
mmol, 44% yield) as a single stereoisomer which was >99%
pure by gas chromatography. Recrystallization from hexanes
provided suitable crystals for X-ray crystallography. An X-ray
crystal structure provided unambiguous structure identifica-
tion of the major isomer: mp 95-96 °C; Rf 0.13 (5:1 hexanes:
EtOAc); 1H NMR (400 MHz, CDCl3) δ 3.61-3.49 (m, 2H),
2.80-2.74 (m, 1H), 2.19-2.06 (m, 1H), 2.05-1.91 (m, 1H),
1.90-1.77 (m, 2H), 1.69-1.53 (m, 2H), 1.49-1.32 (m, 4H),
1.30-1.19 (m, 1H), 1.10 (dd, J ) 12.6, 8.0 Hz, 1H), 0.85 (s,
9H), 0.074 (s, 3H), 0.064 (s, 3H); 13C NMR (100 MHz, CDCl3)
δ 93.6, 67.8, 56.2, 46.6, 37.2, 32.8, 28.5, 26.2, 25.8, 22.7, 18.3,
-3.0, -3.1; IR (neat) 3218, 2951, 2928, 2856 cm-1; HRMS
Calcd for C15H30O2Si+: 270.2019, found 270.1987; LRMS (EI)
m/z 270 (0.4), 213 (10.9), 157 (17.6), 121 (79.2), 119 (12.1), 93
(37.5), 79 (42.8), 77 (11.7), 75 (100.0), 73 (31.4). Anal. Calcd
for C15H30O2Si: C, 66.60; H, 11.18. Found: C, 66.90; H, 11.50.
(1R *,3S *,5S *)-3-(H y d r o x y m e t h y l)b i c y c lo [3.3.0]-
octa n ol (49). The triene 4-ethenyl-4-(trimethylsiloxy)-1,6-
heptadiene (47) (107 mg, 0.51 mmol) was exposed to the
general cyclization/silylation procedure. The reaction mixture
changed from faint yellow to bright yellow over 1 h after which
no starting material was observed by gas chromatography.
After workup and concentration under reduced pressure (0.06
mmHg), a 1H NMR spectrum of the crude mixture showed no
olefinic protons were present. The crude silane 48 appeared
to be one isomer by gas chromatography. Because of difficul-
ties in purification, an oxidation was performed on the crude
silane.
The crude silane 48 (assume 0.51 mmol) in 2.5 mL of DMF
was subjected to the Woerpel oxidation protocol given. Fol-
lowing workup, the products were purified by flash chroma-
tography. Concentration under reduced pressure (0.06 mmHg)
provided the crystalline solid 4921 as a single diastereomer (58
mg, 0.37 mmol, 73% yield) which was >99% pure by gas
chromatography: mp 95-96 °C; Rf 0.15 (2:1 EtOAc:hexanes);
1H NMR (400 MHz, CDCl3) δ 3.61-4.49 (m, 2H), 2.93-2.79
(m, 1H), 2.28-2.12 (m, 1H), 2.11-1.99 (m, 1H), 1.99-1.88 (m,
1H), 1.78 (dd, J ) 12.9, 8.0 Hz, 1H), 1.64-1.28 (m, 8H), 1.18
(dd, J ) 12.9, 8.0 Hz, 1H); 13C NMR (100 MHz, CDCl3) δ 91.9,
67.4, 54.1, 46.2, 36.9, 32.8, 28.0, 25.8, 22.4; IR (neat) 3363,
2953, 2868 cm-1; HRMS Calcd for C9H16O2+: 156.1150, found
156.1158; LRMS (EI) m/z 156 (1.5), 138 (14.5), 127 (22.8), 125
(37.3), 123 (16.8), 114 (29.2), 109 (53.3), 97 (70.3), 83 (100.0),
81 (42.5). Anal. Calcd for C9H16O2: C, 69.19; H, 10.32.
Found: C, 69.36; H, 10.67.
m/z 360 (1.1), 225 (16.5), 211 (52.9), 195 (10.6), 181 (100.0),
107 (15.2). Anal. Calcd for C21H36Si2O: C, 69.93; H, 10.06.
Found: C, 70.17; H, 10.24.
(1R*,2S*,3R*/3S*,4S*)-2-(ter t-Bu t yld im et h ylsiloxy)-3-
[(p h en ylsilyl)m eth yl]bicyclo[3.3.0]octa n e (57). The triene
(3R*,4R*)-3-(tert-butyldimethylsiloxy)-4-ethenyl-1,6-heptadi-
ene (55b) (63 mg, 0.26 mmol) was subjected to the general
cyclization/silylation procedure. The reaction mixture changed
from faint yellow to dark yellow over 1 h after which no
starting material was observed by gas chromatography. Fol-
lowing workup and purification, Kugelrohr distillation pro-
vided 57 as a 7.2:1 mixture of diastereomers (72 mg, 0.20
mmol, 78% yield) which was >97% pure by gas chromatogra-
phy. The major product could be partially separated from the
minor product by HPLC. The NMR data given is for the major
product. All other characterization data was obtained on the
diastereomeric mixture: ot 130 °C/0.07 mmHg; Rf 0.19 (hex-
anes); 1H NMR (400 MHz, CDCl3) δ 7.57-7.53 (m, 2H), 7.39-
7.31 (m, 3H), 4.31-4.24 (m, 2H), 3.44-3.40 (m, 1H), 2.39-
2.31 (m, 1H), 1.98-1.89 (m, 2H), 1.82-1.52 (m, 5H), 1.37-
1.29 (m, 2H), 1.12-1.06 (m, 3H), 1.85 (s, 9H), 0.20 (s, 3H), 0.10
(s, 3H); 13C NMR (100 MHz, CDCl3) δ 135.2, 132.6, 129.5,
128.0, 83.8, 59.6, 50.3, 47.8, 32.9, 28.6, 26.8, 25.9, 25.5, 18.0,
16.9, -4.0, -4.5; IR (neat) 2955, 2866, 2133 cm-1; HRMS Calcd
for C21H36OSi2+: 360.2305, found 360.2306; LRMS (EI) m/z 360
(0.07), 225 (15.5), 211 (52.4), 182 (29.8), 181 (100.0), 121 (11.6),
107 (28.7). Anal. Calcd for C21H36OSi2: C, 69.93; H, 10.06.
Found: C, 69.79; H, 10.33.
(1R*,2R*,3R*,5S*)-3-(H yd r oxym et h yl)b icyclo[3.3.0]-
octa n -2-ol (60). The triene (3R*,4S*)-4-ethenyl-3-(trimeth-
ylsiloxy)-1,6-heptadiene (58) (57 mg, 0.27 mmol) was subjected
to the general cyclization/silylation protocol. The reaction
mixture changed from faint yellow to clear over 1 h after which
no starting material was observed by gas chromatography.
After workup and concentration under reduced pressure (0.06
mmHg), a 1H NMR spectrum of the crude mixture showed no
olefinic protons were present. The crude silane 59 appeared
to be a 4:1 diastereomeric mixture by gas chromatography.
Because of difficulties in purification, an oxidation was per-
formed on the crude silane.
The crude silane 59 (assume 0.27 mmol) in 1.5 mL of DMF
was exposed to the Woerpel oxidation procedure given. Fol-
lowing workup and purification, concentration under reduced
pressure (0.06 mmHg) provided the crystalline solid 60 (15.4
mg, 0.1 mmol, 37% yield over two steps) as a 4:1 mixture of
diastereomers. The major diastereomer could be cleanly
separated by flash chromatography. The structure of 60 was
determined unambiguously from
a single-crystal X-ray
1
structure: mp 84 °C; Rf 0.26 (1:1 hexanes:EtOAc); H NMR
(400 MHz, CDCl3) δ 4.24-4.18 (m, 1H), 3.8 (dd, J ) 11.0, 4.6
Hz, 1H), 3.74-3.66 (m, 1H), 2.79-2.57 (m, 2H), 2.20-1.95 (m,
4H), 1.71-1.57 (m, 2H), 1.49-1.20 (m, 4H), 1.10-0.91 (m, 1H);
13C NMR (100 MHz, CDCl3) δ 71.0, 63.8, 59.2, 51.5, 47.6, 28.8,
28.1, 26.6, 19.7; IR (neat) 3362, 2935, 2865 cm-1; HRMS Calcd
for C9H14O+ (M - H2O)+: 138.1045, found 138.1021; LRMS
(EI) m/z 156 (0.5), 111 (11.2), 95 (32.3), 79 (36.6), 67 (99.5), 57
(56.9), 41 (100.0). Anal. Calcd for C9H16O2: C, 69.16; H, 10.32.
Found: C, 68.94; H, 10.46.
(1R*,2R*,3R*/3S*,5S*)-2-(ter t-Bu t yld im et h ylsiloxy)-3-
[(p h en ylsilyl)m eth yl]bicyclo[3.3.0]octa n e (56). The triene
(3R*,4S*)-3-(tert-butyldimethylsiloxy)-4-ethenyl-1,6-heptadi-
ene (55a ) (127 mg, 0.50 mmol) was exposed to the general
cyclization/silylation procedure given. The reaction mixture
changed from faint yellow to dark yellow over 1 h after which
no starting material was observed by gas chromatography.
Following workup and purification, Kugelrohr distillation
provided 56 as a 2:1 mixture of diastereomers (139 mg, 0.385
mmol, 77% yield) which also contained 7% of inseparable
isomeric impurities by gas chromatography. The major prod-
uct could be partially separated from the minor product by
flash chromatography allowing for complete characteriza-
5-E t h en yl-1,9-b is(m et h ylp h en ylsilyl)-5-(t r im et h ylsi-
loxy)n on a n e (62). Following the general cyclization/silyla-
tion procedure given, Cp*2YCH3‚THF (12 mg, 0.025 mmol) was
dissolved in cyclohexane (1 mL) prior to the addition of
methylphenylsilane (141 mg, 1.15 mmol) and 5-ethenyl-5-
(trimethylsiloxy)-1,8-nonadiene (61) (120 mg, 0.50 mmol). The
reaction mixture was initially bright yellow and remained this
color for 25 h after which no starting material was observed
by gas chromatography. A small amount of monomeric silane,
less than 4%, could still be observed by gas chromatography.
Following workup, the crude product was purified by flash
chromatography. Kugelrohr distillation provided 62 (215 mg,
0.45 mmol, 89% yield) as a clear oil which was >98% pure by
gas chromatography: ot 180 °C/0.06 mmHg; Rf 0.29 (20:1
hexanes:EtOAc); 1H NMR (400 MHz, CDCl3) δ 7.55-7.49 (m,
4H), 7.39-7.31 (m, 6H), 5.78-5.69 (m, 1H), 5.12-4.97 (m, 2H),
4.34 (q, J ) 3.5 Hz, 2H), 1.52-1.42 (m, 4H), 1.40-1.20 (m,
1
tion: ot 130 °C/0.07 mmHg; Rf 0.34 (hexanes); H NMR (400
MHz, CDCl3) δ 7.57-7.53 (m, 2H), 7.41-7.32 (m, 3H), 4.33-
4.24 (m, 2H), 3.6 (d, J ) 4.3 Hz, 1H), 2.39-2.29 (m, 1H), 2.15-
2.04 (m, 1H), 2.04-1.93 (m, 3H), 1.65-1.51 (m, 2H), 1.38-
1.23 (m, 3H), 1.05-0.93 (m, 2H), 0.84 (s, 9H), 0.71-0.60 (m,
1H), -0.02 (s, 6H); 13C NMR (100 MHz, CDCl3) δ 135.2, 132.5,
129.5, 128.0, 78.1, 57.5, 53.7, 48.6, 36.0, 29.0, 26.8, 25.9, 20.5,
18.2, 17.9, -4.6, -4.8; IR (neat) 2955, 2856, 2136 cm-1; HRMS
Calcd for C21H36Si2O+: 360.2305, found 360.2314; LRMS (EI)
(21) The stereochemistry of 54 was confirmed after comparing NMR
spectra with the diol obtained from alcohol 51 after TBDMS removal
with TBAF.

2306 J . Org. Chem., Vol. 63, No. 7, 1998
Molander et al.
8H), 0.92-0.76 (m, 4H), 0.32 (dd, J ) 3.8, 1.1 Hz, 6H), 0.08
(d, J ) 1.0 Hz, 9H); 13C NMR (100 MHz, CDCl3) δ 144.3, 136.7,
134.3, 129.1, 127.8, 112.5, 78.4, 40.0, 27.3, 24.8, 13.4, 2.6, -5.7;
IR (neat) 2938, 2113 cm-1; LRMS (EI) m/z 305 (35.5), 215
(46.9), 211 (36.5), 210 (71.1), 209 (100.0), 135 (38.3), 133 (38.8),
121 (78.2), 105 (12.2), 73 (74.4). Anal. Calcd for C28H46OSi3:
C, 69.64; H, 9.60. Found: C, 70.05; H, 9.81.
DMF was subjected to the Woerpel oxidation procedure given.
Following workup and purification, Kugelrohr distillation
provided the clear oil 65 (47.1 mg, 0.28 mmol, 85% yield) as a
single diastereomer which was >97% pure by gas
chromatography: ot 106 °C/0.07 mmHg; Rf 0.27 (5:1 hexanes:
EtOAc); 1H NMR (400 MHz, CDCl3) δ 3.44-3.38 (m, 2H),
1.78-1.45 (m, 8H), 1.39-1.31 (m, 1H), 1.28-1.14 (m, 2H),
1.04-0.76 (m, 6H), 0.72-0.58 (m, 1H); 13C NMR (100 MHz,
CDCl3) δ 68.8, 43.3, 42.5, 40.6, 36.9, 34.1, 33.8, 33.3, 29.4, 26.7,
26.6; IR (neat) 3358, 2915, 2849 cm-1; HRMS Calcd for
(1 R *,3 R *,6 R *)-3 -[(M e t h y l p h e n y l s i l y l )m e t h y l ]-
bicyclo[4.4.0]d eca n e (64). The triene 5-ethenyl-1,8-nona-
diene (63) (35 mg, 0.23 mmol) was subjected to the general
cyclization/silylation procedure given. The reaction mixture
was initially bright yellow and faded to light yellow over 1 h
after which no starting material was observed by gas chro-
matography. Following workup and purification, Kugelrohr
distillation provided 64 (43 mg, 0.16 mmol, 68% yield) as a
12.8:1 diastereomeric mixture which was >98% pure by gas
chromatography: ot 118 °C/0.07 mmHg; Rf 0.71 (hexanes); 1H
NMR (400 MHz, CDCl3) δ 7.55-7.48 (m, 2H), 7.38-7.30 (m,
3H), 4.4 (q, J ) 3.5 Hz, 1H), 1.78-1.36 (m, 7H), 1.30-1.10
C
11H20O+: 168.1514, found 168.1524; LRMS (EI) m/z 168 (2.4),
150 (38.7), 137 (30.1), 121 (22.6), 108 (15.1), 95 (77.4), 81 (84.9),
67 (62.4), 41 (94.6), 31 (100.0); Analysis Calcd for C11H20O: C,
78.51; H, 11.98; found: C, 78.54; H, 12.11.
Ack n ow led gm en t. We thank the National Insti-
tutes of Health (GM48580) for their generous support
of this research.
(m, 3H), 1.07-0.62 (m, 9H), 0.31 (dd, J ) 3.8, 1.1 Hz, 3H); 13
C
Su p p or tin g In for m a tion Ava ila ble: Complete experi-
mental details and 1H NMR and 13C NMR spectral data for
the preparation of cyclization substrates and X-ray crystal
structure data of the products described herein (136 pages).
This material is contained in libraries on microfiche, im-
mediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
NMR (100 MHz, CDCl3) δ 137.2, 134.3, 129.0, 127.8, 44.1, 44.0,
43.1, 43.0, 42.9, 36.5, 36.4, 34.6, 34.5, 34.1, 34.0, 33.8, 26.7,
26.6, 22.4, -4.8, -4.9; IR (neat) 2914, 2848, 2115 cm-1; HRMS
Calcd for C18H29Si+ (M + H)+: 273.2038, found 273.2005;
LRMS (EI) m/z 272 (0.4), 195 (57.6), 135 (20.2), 121 (100.0),
107 (10.9), 105 (14.2). Anal. Calcd for C18H28Si: C, 79.33; H,
10.36. Found: C, 79.34; H, 10.43.
(1R *,3R *,6R *)-3-(H y d r o x y m e t h y l)b i c y c lo [4.4.0]-
d eca n e (65). Silane 64 (89.5 mg, 0.33 mmol) in 1.5 mL of
J O9721352