Sequential Cyclization/Silylation of Dienynes Catalyzed by an Organoyttrium Complex

Article abstract of DOI:10.1021/jo980521t

The organoyttrium complex Cp*₂YCH₃·THF (Cp* = C₅Me₅) has been shown to be an effective precatalyst for the selective sequential cyclization/silylation of dienynes. The catalyst's ability to insert the alkyne in preference to the alkenes in a regioselective manner, combined with the high diastereoselectivity of the intramolecular insertion process, leads to bicyclo[3.3.0]octane products in high yield. The stereochemistry of the exocyclic olefin, the ring fusion, and the ring substituents are all controlled in the reaction. The cyclization of dienynes thus affords silylated carbobicyclics with high diastereoselectivities in excellent yields.

Full text of DOI:10.1021/jo980521t

J . Org. Chem. 1998, 63, 5507-5516
5507
Sequ en tia l Cycliza tion /Silyla tion of Dien yn es Ca ta lyzed by a n
Or ga n oyttr iu m Com p lex
Gary A. Molander* and William H. Retsch
Department of Chemistry and Biochemistry, University of Colorado, Boulder, Colorado 80309-0215
Received March 19, 1998
The organoyttrium complex Cp*2YCH3‚THF (Cp* ) C5Me5) has been shown to be an effective
precatalyst for the selective sequential cyclization/silylation of dienynes. The catalyst’s ability to
insert the alkyne in preference to the alkenes in a regioselective manner, combined with the high
diastereoselectivity of the intramolecular insertion process, leads to bicyclo[3.3.0]octane products
in high yield. The stereochemistry of the exocyclic olefin, the ring fusion, and the ring substituents
are all controlled in the reaction. The cyclization of dienynes thus affords silylated carbobicyclics
with high diastereoselectivities in excellent yields.
In tr od u ction
enynes. In addition to silylation, the cyclization reaction
can be terminated by hydrogenation. Use of a silylation
reaction to terminate the cyclization is of greater interest
because it leads to materials that can be further func-
tionalized by oxidation to the alcohol. The utility of this
catalytic cyclization/silylation method was demonstrated
through its use in the key step of a total synthesis of (()-
epilupinine.6c This method yielded the desired product
with extremely high diastereoselectivity.
The development of novel carbon-carbon bond-forming
reactions is an integral part of organic chemistry. Within
this context, the cyclization of dienes and enynes repre-
sents a powerful means to convert simple, readily avail-
able substrates to more complex organic molecules. In
fact, the selective conversion of readily available acyclic
polyene precursors into more highly elaborated polycyclic
molecules has been catalyzed by a host of metal com-
plexes. For example, a number of recently reported
protocols to effect Pauson-Khand cyclizations of enynes
employ a variety of metal complexes for the preparation
of bicyclic cyclopentenones.1 A polyene cyclization cata-
lyzed by a titanium alkoxide species2 has been reported,
as well as a variety of palladium(II)-catalyzed3 reactions
with polyenynes.
Carbobicyclics have also been synthesized from acyclic
precursors using this cyclization strategy (eq 1).7 For
Radical4 and organometallic cyclizations5 of unsatu-
rated systems terminated by a silylation reaction can also
be accomplished to form functionalized carbocycles. In
particular, significant progress has been made in the
development of organoyttrium-catalyzed cyclization reac-
tions.6 The precatalyst Cp*2YCH3‚THF (Cp* ) C5Me5)
utilized in these studies has demonstrated the ability to
generate five- and six-membered rings from dienes and
example, organoyttrium catalyzed sequential cyclization/
silylation reactions of trienes have been accomplished to
prepare polycyclic systems. The selectivity of the catalyst
for unsubstituted alkenes relative to allylically substi-
tuted alkenes permits the process to be carried out in a
highly selective manner.
The organoyttrium precatalyst has also demonstrated
a high degree of selectivity and stereocontrol in the
catalyzed cyclization/silylation of enynes (eq 2).6d The
(1) (a) For a review of transition metal catalyzed reactions, see:
Catellani, M.; Chiusoli, G. P.; Costa, M. J . Organomet. Chem. 1995,
500, 69. (b) Livinghouse, T.; Pagenkopf, B. L. J . Am. Chem. Soc. 1996,
118, 2285. (c) Hicks, F. A.; Kablaoui, N. M.; Buchwald, S. L. J . Am.
Chem. Soc. 1996, 118, 9450. (d) Hicks, F. A.; Buchwald, S. L. J . Am.
Chem. Soc. 1996, 118, 11688. (e) Hicks, F. A.; Berk, S. C.; Buchwald,
S. L. J . Org. Chem. 1996, 61, 2713. (f) Zhang, M.; Buchwald, S. L. J .
Org. Chem. 1996, 61, 4498.
(2) Negishi, E.-I.; J ensen, M. D.; Kondakov, D. Y.; Wang, S. J . Am.
Chem. Soc. 1994, 116, 8404.
(3) (a) For a review see: Trost, B. M. Acc. Chem. Res. 1990, 23, 34
and references therein. (b) Chatani, N.; Furukawa, N.; Sakurai, H.;
Murai, S. Organometallics 1996, 15, 901. (c) Trost, B. M.; Shi, Y. J .
Am. Chem. Soc. 1993, 115, 9421. (d) Shi, Y.; Trost, B. M. J . Am. Chem.
Soc. 1992, 114, 791.
(4) (a) Kulicke, K. J .; Giese, B. Synlett 1990, 91. (b) Kraus, G. A.;
Liras, S. Tetrahedron Lett. 1990, 31, 5265. (c) Kopping, B.; Chatgil-
ialoglu, C.; Zehnder, M.; Giese, B. J . Org. Chem. 1992, 57, 3994.
(5) Ojima, I.; Donovan, R. J .; Shay, W. R. J . Am. Chem. Soc. 1992,
114, 6580.
(6) (a) Molander, G. A.; Hoberg, J . O. J . Am. Chem. Soc. 1992, 114,
3123. (b) Molander, G. A.; Nichols, P. J . J . Am. Chem. Soc. 1995, 117,
4415. (c) Molander, G. A.; Nichols, P. J . J . Org. Chem. 1996, 61, 6040.
(d) Molander, G. A.; Retsch, W. H. J . Am. Chem. Soc. 1997, 119, 8817.
(e) Molander, G. A.; Dowdy, E. D. J . Org. Chem., in press.
catalytic cyclization of 1,6-enynes containing allylic sub-
stituents afforded trans-substituted (phenylsilyl)meth-
ylcyclopentanes in high yields. The diastereomeric ratios
generated by this protocol were in excess of 50:1 when
the substrate contained an allylic alkyl substituent. The
formation of the observed cyclized products was dictated
(7) Molander, G. A.; Nichols, P. J . J . Org. Chem., in press.
S0022-3263(98)00521-0 CCC: $15.00 © 1998 American Chemical Society
Published on Web 07/15/1998

5508 J . Org. Chem., Vol. 63, No. 16, 1998
Molander and Retsch
Sch em e 1
Sch em e 2
Sch em e 3a
by the ability of the organoyttrium complex to selectively
insert alkynes relative to allylically substituted alkenes
in the first step of the reaction. The Lewis acidic
organoyttrium complex tolerated ethers, acetals, and
tertiary amine functionalities in the reaction.
In an effort to extend the organoyttrium-catalyzed
cyclization/silylation reaction of enynes to the preparation
bicyclic products, a variety of dienynes were prepared and
reacted under conditions similar to those used to cyclize
enynes. The process envisioned for the organoyttrium-
catalyzed sequential cyclization/silylation of dienynes
would be complementary to the previously reported
methods using other classes of catalysts. A viable
catalytic cycle, similar to that proposed for the cyclization/
silylation of enynes, is outlined in Scheme 1.6d The
catalytic cycle would be initiated by a σ-bond metathesis
between Cp*2YCH3‚THF and PhSiH3, producing a cata-
lytically active metal hydride species.6a-d,8 Next, the
catalyst would have to insert the alkyne selectively in
preference to the two alkenes in the substrate. Evidence
supports the proposition that the alkyne would be more
reactive and could be inserted regioselectively in order
to form a single alkenylyttrium intermediate.6d,8b This
intermediate could undergo cyclization with either of the
available alkenes. Evidence from enyne cyclizations
suggested that the 5-exo-trig cyclization would be pre-
ferred over the 6-exo-trig pathway.6d Cyclization would
provide a second intermediate alkylyttrium species. This
second intermediate, similar to those proposed in prior
cyclizations,6,7 could undergo a subsequent intramolecu-
lar alkene insertion with the remaining alkene as pre-
dicted by reactions with trienes.7 A σ-bond metathesis
with silane would generate the bicyclic product and
complete the catalytic cycle. All of the steps in this
catalytic cycle are well precedented, and the chemo-,
regio-, and diastereoselectivities demonstrated in preced-
ing studies suggested that this reaction would also be
accomplished with excellent selectivity.
a
Key: (a) (carbethoxymethylene)triphenylphosphorane, CH2Cl2;
(b) vinylmagnesium bromide, CuI, THF; (c) (1) LDA, THF, MeI,
HMPA, (2) LDA, THF, MeI, HMPA; (d) LAH, THF; (e) pyr‚SO3,
DMSO, NEt3, CH2Cl2; (f) Ph3PCH3Br, t-BuOK, Et2O.
to stereodefined, highly functionalized carbobicyclics from
relatively simple, readily accessible acyclic starting ma-
terials.
Resu lts a n d Discu ssion
A series of dienyne substrates designed to test this new
protocol was prepared according to Schemes 2-9. Scheme
2 outlines the synthesis of dienyne substrates 2 and 3.
Hydrazone 1 was prepared according to literature
procedures.6d Treatment of the hydrazone with LDA and
alkylation with the appropriate electrophile was followed
by removal of the hydrazone with wet Amberlyst. The
resultant aldehydes were subjected directly to a Wittig
reaction to yield the desired dienynes.9
Substrate 10 was synthesized according to the outline
in Scheme 3. The aldehyde 5-cyclohexyl-4-pentynal was
prepared according to a published procedure6d and was
treated with (carbethoxymethylene)triphenylphosphorane
to yield the unsaturated ester 5.10 Vinyl cuprate addition
followed by dialkylation with iodomethane provided the
substituted ester 7. Reduction of the ester to the alcohol
followed by oxidation using a modified Swern procedure
afforded aldehyde 9.11 The aldehyde was treated with a
Wittig reagent to yield the desired dienyne 10.
We now report that the organoyttrium hydride catalyst
derived from Cp*2YCH3‚THF is indeed an effective
catalyst for the sequential cyclization of dienynes in the
postulated manner. The reaction provides facile access
(9) Fitjer, L.; Quabek, U. Synth. Commun. 1985, 15, 855.
(10) Bunce, R. A.; Dowdy, E. D.; J ones, P. B.; Holt, E. M. J . Org.
Chem. 1993, 58, 7143.
(8) (a) Molander, G. A.; Hoberg, J . O. J . Org. Chem. 1992, 57, 3266.
(b) Molander, G. A.; Retsch, W. H. Organometallics 1995, 14, 4570.
(11) Wipf, P.; Fritch, P. C. J . Org. Chem. 1994, 59, 4875.

Sequential Cyclization/Silylation of Dienynes
J . Org. Chem., Vol. 63, No. 16, 1998 5509
Sch em e 4a
Sch em e 6a
a
Key: (a) (1) LDA, THF, (2) acrolein, (3) TBDMSCl; (b)
DIBALH, THF; (c) Ph3PCH3Br, t-BuOK, Et2O; (d) PPh3, imidazole,
I2, 3:1 Et2O/CH3CN; (e) alkynyllithium from 2,2-dimethyl-5-
ethynyl-1,3-dioxane, HMPA, THF; (f) TBAF, THF.
and 17, which were separable by flash chromatography.
The relative stereochemistry of the acetonides was as-
signed on the basis of the chemical shift and coupling
constants of the allylic proton. Deprotection of the
acetonide provided diastereomerically pure diols 18 and
19.
a
Key: (a) PDC, DMF; (b) DCC, DMAP, MeOH, CH2Cl2; (c) (1)
LDA, THF, (2) acrolein; (d) LAH, THF; (e) 2,2-dimethoxypropane,
cat. TsOH, acetone; (f) MeOH, cat. TsOH.
In Scheme 5, the conversion of diol 18 to the desired
substituted dienynes 22 and 25 is diagrammed.
A
Sch em e 5a
monoprotection of diol 18 yielded a separable 1:5 mixture
of silyl ethers 20a and 20b. The minor isomer 20a was
oxidized with TPAP13 to aldehyde 21 and treated with a
Wittig reagent to provide dienyne substrate 22. The
hydroxyl group of 20b was protected as a trityl ether,
followed by TBAF deprotection of the silyl ether to yield
24. Compound 24 was oxidized in the same manner as
21 and treated with a Wittig reagent to provide dienyne
substrate 25. An identical synthetic scheme was used
for the conversion of the other diastereomer, diol 19, to
dienyne substrates 56 and 58.
The synthesis of intermediate 30 is shown in Scheme
6. γ-Butyrolactone was treated with LDA, acrolein, and
tert-butyldimethylsilyl chloride to yield a single dia-
stereomer of compound 26. Reduction of the lactone to
the hemiacetal was followed by treatment with a Wittig
reagent to yield 27. The hydroxyl group of 27 was
converted to an iodide14 and displaced with the alkynyl-
lithium derived from 2,2-dimethyl-5-ethynyl-1,3-dioxane
to yield dienyne 29.15 The silyl ether of 29 was depro-
tected with TBAF to give intermediate 30. Protection of
the hydroxyl group of this intermediate provided sub-
strates 52 and 54.
Scheme 7 outlines the synthesis of substrate 35.
Lactone 31 was prepared according to a published
procedure and treated with LDA.16 Alkylation with
benzyl bromide provided diastereomerically pure 32. The
relative stereochemistry of 32 was assigned by NOE
experiments. The lactone 32 was reduced to the hemi-
acetal and treated directly with a Wittig reagent to
provide 33. The hydroxyl group of 33 was converted to
a
Key: (a) NaH, TBDMSCl, THF; (b) cat. TPAP, NMO, molec-
ular sieves, CH2Cl2; (c) Ph3PCH3Br, t-BuOK, Et2O; (d) TrCl, DBU,
CH2Cl2; (e) TBAF, THF.
The diastereomerically pure dienynes 22, 25, 56, and
58 were prepared according to Schemes 4 and 5. Alcohol
116d was oxidized with PDC and converted to a methyl
ester with DCC, DMAP, and MeOH.12 Ester 13 was
treated with LDA and acrolein in an aldol reaction to
provide â-hydroxy ester 14. Subsequent reduction with
LAH provided a 1:1 mixture of the desired 1,3-diols.
Treatment of 15 with 2,2-dimethoxypropane in acetone
and a catalytic amount of H2SO4 afforded acetonides 16
(13) Ley, S. V.; Norman, J .; Griffith, W. P.; Marsden, S. P. Synthesis
1994, 639.
(14) Corey, E. J .; Pyne, S. C. J . Org. Chem. 1986, 51, 4726.
(15) To establish the relative stereochemistry of 28, the iodide was
displaced with lithium acetylide, ethylendiamine complex in DMSO
to prepare a terminal alkyne. Installation of a cyclohexyl group on the
alkyne afforded material identical to compound 22 by GC and NMR.
(16) Molander, G. A.; Harris, C. H. J . Am. Chem. Soc. 1995, 117,
3705.
(12) Neises, B.; Steglich, W. Organic Syntheses; Wiley: New York,
1990; Vol. VIII, p 93.

5510 J . Org. Chem., Vol. 63, No. 16, 1998
Molander and Retsch
Sch em e 7a
Sch em e 9a
a
Key: (a) (1) LDA, THF, (2) benzyl bromide; (b) DIBALH, THF;
(c) Ph3PCH3Br, t-BuOK, Et2O; (d) PPh3, imidazole, I2, 3:1 Et2O/
CH3CN; (e) alkynyllithium from 2,2-dimethyl-5-ethynyl-1,3-diox-
ane, HMPA, THF.
Sch em e 8a
a
Key: (a) (1) n-BuLi, (C6H11)3B, I2, THF, (2) NaOH, H2O2; (b)
TBAF, THF; (c) PPh3, I2, imidazole, 3:1 Et2O/CH3CN; (d) (1) LDA,
THF, (2) 42, HMPA; (e) (1) DIBALH, THF, (2) Ph3PCH3Br,
t-BuOK, Et2O; (f) TBDPSCl, imidazole, DMF.
Ta ble 1. Sequ en tia l Cycliza tion /Silyla tion of Dien yn es
Ca ta lyzed by Cp *2YCH3‚THF a
a
Key: (a) pyr‚SO3, DMSO, NEt3, CH2Cl2; (b) PPh3, CBr4, Zn,
CH2Cl2; (c) 2 equiv of n-BuLi, THF; (d) (1) n-BuLi, (C6H11)3B, I2,
THF, (2) NaOH, H2O2.
the iodide and displaced with the appropriate alkynyl-
lithium to yield dienyne substrate 35 as a 30:1 mixture
of diastereomers.
Scheme 8 depicts the synthesis of substrate 38. Alco-
hol 27, prepared in a previous synthesis, was oxidized to
provide 36. Aldehyde 36 was converted to a terminal
alkyne via a Corey-Fuchs sequence.17 A cyclohexyl
group was placed on the terminal alkyne according to
published procedures to yield substrate 38.18
Finally, Scheme 9 provides an outline for the synthesis
of compound 46. Commercially available 3-butyn-1-ol
was protected as the silyl ether and converted to iodide
42 in high yield. Lactone 43 was prepared according to
the published procedure19 and alkylated with iodide 42
to give a 4:1 mixture of diastereomers. The lactone was
reduced to the hemiacetal and treated with the Wittig
reagent. Flash column chromatography provided the
major isomer 45 as a single diastereomer. Protection of
45 as the silyl ether provided the final dienyne substrate.
With the substrates in hand, investigations were
undertaken to determine the optimum reaction condi-
tions for the sequential cyclization/silylation reaction and
to outline the scope and limitations of the process.
Compound 2 was treated with 1.1 equiv of phenylsilane
and 5 mol % of the precatalyst in cyclohexane for 2.5 h
(entry 1, Table 1). A mixture of compounds was gener-
a
Reactions were run in cyclohexane with 5 mol % precatalyst
b
and 1.1-1.3 equiv PhSiH3 at rt. Diastereomeric ratio was
determined by fused silica gas chromatographic analysis.
ated in the reaction. Only compound 47 was isolated
cleanly from the reaction mixture. The expected stereo-
chemistry of the bicyclic product with respect to the ring
fusion was predicted to be trans by comparison to the
products isolated from previous enyne and polyene
cyclizations.6b-d,7,8a The stereochemistry of the meth-
ylphenylsilyl substituent was not determined. The lack
of selectivity of the catalyst for initial insertion of the
alkyne over the unsubstituted alkenes was assumed to
be responsible for the formation of undesirable byprod-
ucts in this reaction. To overcome this difficulty and
(17) Corey, E. J .; Fuchs, P. L. Tetrahedron Lett. 1972, 36, 3769.
(18) Brown, H. C.; Sinclair, J . A.; Midland, M. M. J . Am. Chem. Soc.
1973, 95, 3080.
(19) Annby, U.; Stenkula, M. Tetrahedron Lett. 1993, 34, 8545.

Sequential Cyclization/Silylation of Dienynes
J . Org. Chem., Vol. 63, No. 16, 1998 5511
Ta ble 2. Sequ en tia l Cycliza tion /Silyla tion of
Dia ster eom er ica lly P u r e Dien yn es Ca ta lyzed by
Cp *2YCH3‚THF a
excess of phenylsilane at 25 °C for 2 h (Table 2, entry 1).
The reaction was run in benzene-d6 and was monitored
1
by H NMR. The creation of two new stereocenters in
the cyclization reaction results in the formation of a
mixture of two diastereomeric products in 93% isolated
yield. The ratio of products as determined by fused silica
gas chromatographic analysis of the crude reaction
mixture was 7:1.
Results observed in the cyclization of enynes suggested
that improvement of the diastereoselectivity of the cy-
clized products could be achieved if larger alkoxy groups
were incorporated into the starting materials.6d The
addition of a (triphenylmethyl)oxy substituent in the
allylic position of substrate 25 demonstrated that higher
selectivities could be achieved (Table 2, entry 2). Sub-
strate 25 was cyclized according to the general experi-
mental procedure to provide 86% of cyclized product 51.
The diastereoselectivity was found to be >30:1 by 1H
NMR of the crude reaction mixture. The enhancement
of the diastereomeric ratio can be rationalized by the
increased steric size of the (triphenylmethyl)oxy sub-
stituent relative to the (tert-butyldimethylsilyl)oxy sub-
stituent of substrate 22. NOE data for 51 supports the
relative stereochemical assignments of the products in
entries 1 and 2 (Table 2).
Figure 1 suggests conformations for the expected
intermediates in the sequential cyclization reaction of
substrates 22 and 25. The presence of the cyclohexyl
group as a substituent on the alkyne of entries 1 and 2
(Table 2) provides branching in the propargylic position.
This branching serves as a regiocontrol element in the
alkyne insertion leading to the formation of proposed
intermediate I. Intermediate I is drawn in a conforma-
tion that suggests a chairlike transition structure pro-
posed for the first intramolecular alkene insertion. In
this conformation, the alkyl substituent occupies a pseudo-
equatorial position on the ring being formed. This
controls the formation of the first new stereocenter of the
bicyclic ring system. A transition structure of this type
was first proposed for the cyclization of enynes and is
also supported by products observed in the organo-
yttrium-catalyzed cyclization of trienes.6d,7 The second
alkylyttrium intermediate can adopt a number of con-
formations. The proposed conformations designated as
IIa and IIb are drawn to suggest those that lead to the
transition structures for the second intramolecular inser-
tion. Cyclization through a second chairlike transition
structure, similar to the conformation of IIa , is proposed
for the formation of the major diastereomer. Cyclization
through a boatlike transition structure, as suggested by
IIb, is expected to give rise to the minor diastereomer of
the reaction.
Entries 5 and 6 of Table 2 demonstrate that the syn
diastereomers of substrates 22 and 25 also cyclized in
high yield to form bicyclic products. The (triphenyl-
methyl)oxy substituent of 58 gave rise to excellent
diastereoselectivity in the formation of product 59. The
sequential cyclization/silylation reaction of 56 generated
only a 2:1 ratio of diastereomers. The major products
57a and 59 were demonstrated to have the same relative
stereochemistry. This was accomplished by treatment
of compound 59 with formic acid in ether21 to liberate
the alcohol, which was reprotected as the tert-butyldi-
a
Reactions were run in an NMR tube according to the general
procedure. b Diastereomeric ratios were determined by fused silica
gas chromatographic analysis or by 1H NMR integration of the
crude reaction mixture.
increase the catalyst’s preference for initial insertion of
the alkyne, substrates with substituents on the alkene
and in the allylic position were prepared. The presence
of a methyl substituent on the alkene of dienyne 3 was
expected to favor the initial insertion of the alkyne by
lowering the reactivity of the alkene.6a,20 Unfortunately,
the intramolecular insertion of this 1,1-disubstituted
alkene was also hindered. Only the initial cyclization
occurred before the intermediate organometallic was
trapped with the phenylsilane to afford compound 48
(Table 1, entry 2). Substrate 10, prepared with a gem-
dimethyl group allylic to the alkene, was effectively
cyclized under similar reaction conditions (Table 1, entry
3). The bicyclic product 49 was isolated in 76% yield as
a 3.9:1 mixture of diastereomers. This example indicated
that allylic substitution at both alkenes would allow the
sequential cyclization reaction to proceed cleanly. It also
demonstrated that the alkenylmetallic species pro-
duced by the initial insertion undergoes a highly selective
5-exo-trig cyclization in preference to a 6-exo-trig cycliza-
tion.
The addition of an alkoxy substituent in the allylic
position of alkenes has been shown to permit favorable
insertion of the alkyne in preference to the alkene in
enyne substrates.6d The allylic alkoxy substituent lowers
the reactivity of the alkene by both steric and electronic
effects. Entries 1-6 of Table 2 demonstrated that higher
yields of bicyclic products could be obtained by the
incorporation of an allylic alkoxy substituent in the
substrates. The reaction of substrate 22 with an allylic
(tert-butyldimethylsilyl)oxy substituent was carried out
with 10 mol % of the precatalyst in the presence of an
(21) Bessodes, M.; Komiotis, D.; Antonakis, K. Tetrahedron Lett.
1986, 27, 579.
(20) Molander, G. A.; J ulius, M. J . Org. Chem. 1992, 57, 6347.

5512 J . Org. Chem., Vol. 63, No. 16, 1998
Molander and Retsch
F igu r e 1. Intermediates in the formation of bicyclic products from anti-dienynes.
F igu r e 2. Intermediates in the formation of bicyclic products from syn-dienynes.
methylsilyl ether. The 1H NMR of this compound matched
that of 57, the major product observed in entry 5 (Table
2). NOE Data for 59 supports the relative stereochem-
istry assigned to the major products of entries 5 and 6
(Table 2).
Substrates in entries 1, 2, 5, and 6 (Table 2) all possess
a cyclohexyl group as a substituent on the alkyne to
provide branching in the propargyl position. This branch-
ing serves as a regiocontrol element and is required to
attain complete regioselectivity in the initial alkyne
insertion. Entries 3, 4, and 7 (Table 2) provide examples
of functionalized alkyne substituents that can also be
utilized to obtain regiocontrol without hindering reactiv-
ity. The oxygens of these alkyne substrates did not
appreciably affect the reactivity of the highly electron-
deficient catalyst. The reactions proceeded cleanly at
room temperature, complete selectivity for the alkyne
insertion was observed, and the cyclized products were
obtained in high yield. The triphenylmethoxy substitu-
ent of entry 3 (Table 2) again demonstrated that excellent
diastereoselectivity, >40:1, could be achieved with this
bulky protecting group. NOE data supports the relative
stereochemistry assigned for 53.
The expected intermediates formed in the cyclization
reactions of 56 and 58 are shown in Figure 2. The second
alkylyttrium intermediate is drawn in two conformations,
IVa and IVb, to suggest possible transition-state struc-
tures. The reactions of 56 and 58 are expected to react
through boatlike transition structures, as suggested by
IVb, to give rise to the major product. This is predicted
on the basis of the pseudoaxial orientation of the allylic
substituent R′ in the formation of the second ring. In a
chairlike conformation, allylic A1,3-strain between R′ and
vinyl protons of the alkene and steric interactions
between the bulky metal ligands and the axial substitu-
ent R′ are postulated to be responsible for forcing the
reaction to proceed through a boatlike transition struc-
ture. The improvement of diastereoselectivity observed
when an allylic triphenylmethoxy substituent is used
supports this suggestion.
A tert-butyldiphenylsilyl protecting group was incor-
porated in substrate 54 and the cyclization reaction of
this substrate provided a 14:1 ratio of diastereomers in
the product (Table 2, entry 4). Although not as selective

Sequential Cyclization/Silylation of Dienynes
J . Org. Chem., Vol. 63, No. 16, 1998 5513
Ta ble 3. Sequ en tia l Cycliza tion /Silyla tion Rea ction s
conformation, which places the most substituents in
pseudoequatorial orientation, is preferred and prohibits
the second cyclization. The relative stereochemistry of
compound 61 was assigned by an NOE experiment. This
cyclization again demonstrated the preference of the
catalyst for the insertion of an alkyne over allylically
substituted alkenes. It also demonstrated that the
alkenylmetallic species produced by the initial insertion
undergoes a 5-exo-trig cyclization in preference to a 4-exo-
trig cyclization.
Ca ta lyzed by Cp *2YCH3‚THF a
Substrate 46 was prepared with the expectation that
it might cyclize to provide a bicyclo[4.3.0]nonane. Under
conditions identical to those that allowed the cyclization
of other dienynes to occur, 46 was reacted with the
precatalyst and 1.1 equiv of phenylsilane (Table 3, entry
2). The cyclization reaction readily occurred to produce
a single diastereomer of the desired carbobicyclic 62 in
excellent yield. An NOE experiment confirmed the
relative stereochemistry as drawn.
a
Reactions were run in an NMR tube using benzene-d6 as the
solvent according to the general procedure.
Con clu sion s
The organoyttrium complex Cp*2YCH3‚THF has been
used to catalyze the sequential cyclization/silylation of
suitable dienynes with phenylsilane to form functional-
ized bicyclic structures in a single step. In the sequential
process, two carbon-carbon bonds and a carbon-silicon
bond were formed selectively. The catalyst accomplishes
this reaction through its ability to react selectively in
each step of the reaction. In the first step, the alkyne
inserts regioselectively in preference to an alkene. The
second step requires that the catalyst preferentially
cyclize through a 5-exo-trig pathway over a 6-exo-trig
cyclization. This cyclization via intramolecular olefin
insertion engenders high diastereoselectivity in the first
cyclization and sets up a second intramolecular insertion
that is also accomplished with a high degree of dia-
stereoselectivity. The diastereoselectivity of the reaction
can be enhanced by increasing the size of the allylic
substituent in the second cyclization. In addition to the
catalyst’s ability to react selectively with the substrates,
these reactions proceed at very reasonable rates under
extremely mild conditions, and a variety of ether func-
tionalities can be tolerated. This method complements
and extends other catalyzed cyclization reactions of
enynes and polyenes and does so in a manner that is
atom economical, with no byproducts formed in the
process.22
F igu r e 3. Conformations proposed for the intermediate in the
reaction of 38.
as the triphenylmethoxy substituent of entries 2 and 3
(Table 2), the (tert-butyldiphenylsilyl)oxy group of entry
4 (Table 2) achieved selectivity greater than that ob-
served with the (tert-butyldimethylsilyl)oxy group utilized
in substrate 22 (Table 2, entry 1).
The diastereoselectivity of enyne cyclizations was
exceedingly high when allylic substituents with A values
greater than that of a protected alcohol were incorporated
into the enyne substrates.6d For example, substrate 35
was prepared with a benzyl group in the allylic position,
and formation of the minor diastereomer was not ob-
served (Table 2, entry 7). The 30:1 diastereomeric ratio
of the product was the result of the mixture of diaster-
eomers in the starting material (also 30:1). The cycliza-
tion of 35 demonstrated that allylic alkyl substituents
are effective in achieving excellent diastereoselectivities
in the cyclization of dienynes.
It was expected that the sequential cyclization/silyl-
ation reaction of dienynes could be readily extended to
include the formation of other bicyclic ring structures.
Substrate 38 was prepared with the expectation that it
might cyclize to provide a bicyclo[3.2.0]heptane or a
bicyclo[2.2.1]heptane. Under conditions identical to those
that allowed the cyclization of other dienynes to occur,
38 was reacted with the precatalyst and 1.1 equiv of
phenylsilane (Table 3, entry 1). No bicyclic product was
identified, but a mixture of 61 and starting material was
observed. By addition of excess silane the dienyne was
cyclized and hydrosilylated to form the highly function-
alized cyclopentane 61 cleanly and in high yield. The
absence of a bicyclic product in the reaction of 38 can be
explained by the intermediates depicted in Figure 3. To
form a second ring, the conformation of the intermediate
would be such that two of the three substituents are in
pseudoaxial orientations. Presumably, the flip-chair
Exp er im en ta l Section
All operations involving the organoyttrium complex were
performed with rigorous exclusion of oxygen and moisture in
flamed Schlenk-type glassware on an argon line connected to
a vacuum system (<0.04 mmHg) or in a nitrogen-filled,
Vacuum Atmospheres glovebox. The cyclohexane and benzene-
d6 used as solvents for the reactions were distilled from Na/
benzophenone under argon and then stored in the glovebox.
In preparation for use with the catalyst, substrates were dried
as a solution with MgSO4, concentrated in vacuo, and freeze/
pump/thaw degassed. The phenylsilane was purchased from
Aldrich and was dried with activated 4 Å molecular sieves and
degassed before use. Anhydrous YCl3 was purchased from
(22) (a) Trost, B. M. Science 1991, 254, 1471. (b) Trost, B. M. Angew.
Chem., Int. Ed. Engl. 1995, 34, 259.

5514 J . Org. Chem., Vol. 63, No. 16, 1998
Molander and Retsch
Cerac. The complex Cp*2YCH3‚THF was prepared according
to published procedures.23
Analysis of the crude reaction mixture by gas chromatography
indicated that a 3.9:1 diastereomeric ratio of products was
generated. Purification by flash chromatography (Rf 0.34 in
hexanes followed by Kugelrohr distillation (ot 85-95 °C/0.02
mmHg) afforded 76% (76 mg, 0.22 mmol) of 49 as a 2.9:1
mixture of diastereomers. Major diastereomer: 1H NMR (400
MHz, CDCl3) δ 0.73 (s, 3H), 0.76-0.94 (m, 2H with overlapping
singlet at 0.92, 3H), 0.95-1.07 (m, 3H), 1.10-1.31 (m, 4H),
1.34-1.44 (m, 1H), 1.48-1.54 (m, 1H), 1.57-1.71 (m, 6H),
1.92-2.12 (m, 2H), 2.39-2.55 (m, 3H), 4.25-4.35 (m, 2H),
Gen er a l Exp er im en ta l P r oced u r e for NMR Sca le Se-
qu en tia l Cycliza tion /Silyla tion Rea ction s. The catalytic
sequential cyclization/silylation of 22 is representative. In the
nitrogen atmosphere glovebox a solution of 6 mg (0.013 mmol)
of the precatalyst Cp*2YCH3‚THF was weighed in an NMR
tube equipped with a J . Young valve. Benzene-d6 (0.6 mL)
was added. To this solution was added 52 mg (0.015 mmol)
of 22 followed by 21 mg (0.019 mmol) of phenylsilane. The
NMR tube was sealed and removed from the glovebox. The
4.81-4.83 (m, 1H), 7.33-7.41 (m, 3H), 7.56-7.58 (m, 2H); 13
C
1
NMR (100 MHz, CDCl3) δ 12.06, 16.17, 21.64, 26.12, 26.18,
27.23, 30.32, 31.71, 33.24, 33.28, 37.74, 38.97, 51.40, 51.96,
61.99, 123.45, 127.95, 129.47, 132.82, 135.18, 140.36; IR (neat)
2134 cm-1; HRMS calcd for C24H36Si 352.2586, found 352.2579;
LRMS (EI) m/z (relative intensity) 352 (1.2), 107 (65), 55 (62),
41 (100). Anal. Calcd for C24H36Si2: C, 81.75; H, 10.29.
Found: C, 81.78; H, 10.22.
progress of the reaction was monitored by H NMR. After 2
h, the reaction mixture was diluted with 2 mL of Et2O and
filtered through a 0.5 g column of Florisil. The Florisil was
rinsed two times with 2 mL portions of Et2O. Analysis of the
crude reaction mixture by gas chromatography indicated that
a 7:1 ratio of diastereomers was generated. The organics were
combined and concentrated by rotary evaporation. The crude
material was purified by flash chromatography with hexanes
followed by solvent removal at 50 °C/0.04 mmHg to afford 93%
(62 mg, 0.014 mmol) of 50 as a 7:1 mixture of diastereomers.
(1E,2R*,6R*)-1-(Cycloh exylm eth ylen e)-4-[(ph en ylsilyl)-
m eth yl]bicyclo[3.3.0]octa n e (47). Gen er a l Exp er im en ta l
P r oced u r e for Ca ta lytic Cycliza tion /Silyla tion . In the
nitrogen atmosphere glovebox was added a solution of 6 mg
(0.013 mmol) of the precatalyst Cp*2YCH3‚THF in 1 mL of
cyclohexane to a reaction flask equipped with a magnetic stir
bar. To this solution was added 51 mg (0.24 mmol) of 2
followed by 31 mg (0.29 mmol) of phenylsilane. The reaction
flask was sealed and stirred in the glovebox for 2.5 h at
ambient temperature. After 2.5 h, the reaction mixture was
removed from the glovebox, diluted with 1 mL of Et2O, and
filtered through a 0.5 g column of Florisil. The Florisil was
rinsed two times with 2 mL portions of Et2O. The organics
were combined and concentrated by rotary evaporation. The
crude material was purified by flash chromatography (Rf 0.51
in hexanes) followed by Kugelrohr distillation (ot 80-85 °C/
0.1 mmHg) to afford 33% (26 mg, 0.08 mmol) of 47 (GC purity
>97%): 1H NMR (400 MHz, CDCl3) δ 0.84 (dt, J ) 11.6, 8.6
Hz, 1H), 0.96-1.04 (m, 2H), 1.10-1.29 (m, 6H), 1.31-1.38 (m,
1H), 1.43-1.72 (m, 8H), 1.90-2.00 (m, 2H), 2.19-2.28 (m, 1H),
(5E,1R*,2R*,4S*,8R*)-1-[(ter t-Bu tyld im eth ylsilyl)oxy]-
5-(cycloh exylm eth ylen e)-2-[(p h en ylsilyl)m eth yl]bicyclo-
[3.3.0]octa n e (50) was prepared from 22 according to the
general experimental procedure. Analysis of the crude reac-
tion mixture by gas chromatography indicated that a 7:1
diastereomeric ratio of products was generated. Flash chro-
matography (Rf 0.46 in hexanes) followed by removal of
solvents at 50 °C/0.01 mmHg afforded 93% yield of 50 as a
mixture of diastereomers: 1H NMR (400 MHz, CDCl3) δ 0.01
(s, 3H), 0.02 (s, 3H), 0.85 (s, 9H), 0.93-1.05 (m, 3H), 1.07-
1.28 (m, 4H), 1.32-1.42 (m, 2H), 1.55-1.82 (m, 8H), 1.89-
2.01 (m, 1H), 2.26-2.36 (m, 2H), 2.49-2.54 (m, 2H), 3.46 (dd,
J ) 8.9, 6.0 Hz, 1H), 4.24-4.32 (m, 2H), 4.77-4.80 (m, 1H),
7.31-7.39 (m, 3H), 7.53-7.55 (m, 2H); 13C NMR (100 MHz,
CDCl3) δ -4.49, -4.07, 16.74, 17.96, 25.81, 25.88, 25.96, 26.08,
26.15, 30.29, 32.48, 33.15, 33.19, 37.79, 49.32, 51.69, 57.83,
83.49, 124.00, 127.96, 129.49, 132.48, 135.21, 138.99; IR (neat)
2136 cm-1; HRMS calcd for C28H46OSi2 454.3087, found
454.3038; LRMS (EI) m/z (relative intensity) 454 (7), 371 (100).
Anal. Calcd for C28H46OSi2: C, 73.94; H, 10.19. Found: C,
73.79; H, 10.37.
(5E,1R*,2R*,4S*,8R*)-5-(Cycloh exylm eth ylen e)-2-[(ph en -
ylsilyl)m et h yl]-1-[(t r ip h en ylm et h yl)oxy]b icyclo[3.3.0]-
octa n e (51) was prepared from 25 (10 mg, 0.021 mmol)
according to the general experimental procedure. Analysis of
1
2.52-2.61 (m, 3H), 4.28 (t, J ) 3.7 Hz, J 29Si,H ) 196.5 Hz,
2H), 4.82 (dd, J ) 9.0, 2.3 Hz, 1H), 7.32-7.38 (m, 3H), 7.55-
7.56 (m, 2H); 13C NMR (100 MHz, CDCl3) δ 20.24, 26.12, 26.19,
26.82, 32.98, 33.09, 33.24, 37.85, 38.09, 40.57, 52.91, 57.09,
123.98, 127.94, 129.45, 132.79, 135.23, 139.56; IR (neat) 2136
cm-1; HRMS calcd for C22H32Si 324.2273, found 324.2271;
LRMS (EI) m/z (rel intensity) 324 (4), 55 (100), 41 (97).
(1E,2R*,3S*)-3-[3′-(2-Meth yl-1-pr open yl)]-1-(cycloh exyl-
m eth ylen e)-2-[(ph en ylsilyl)m eth yl]cyclopen ta n e (48) was
prepared from 3 (99 mg, 0.43 mmol) according to the general
experimental procedure outlined for the preparation of 47.
Purification by flash chromatography (Rf 0.41 in hexanes)
followed by Kugelrohr distillation afforded 84% (121 mg, 0.36
mmol) of 48 (GC purity >98%): 1H NMR (400 MHz, CDCl3) δ
0.97-1.07 (m, 2H), 1.10-1.31 (m, 6H), 1.58-1.88 (m, 8H, with
overlapping singlet at 1.65, 3H), 1.89-2.09 (m, 1H), 2.12-2.24
(m, 3H), 2.28-2.36 (m, 1H), 4.29 (t, J ) 4.0 Hz, 1J 29Si,H ) 197.8
Hz, 2H), 4.61 (s, 1H), 4.67 (s, 1H), 5.01-5.04 (m, 1H), 7.31-
7.39 (m, 3H), 7.54-7.56 (m, 2H); 13C NMR (100 MHz, CDCl3)
δ 14.16, 22.31, 26.10, 26.19, 27.16, 29.84, 33.01, 33.11, 38.47,
42.86, 44.31, 46.65, 110.99, 127.34, 127.92, 129.36, 133.38,
135.21, 144.57, 144.96; IR (neat) 2134 cm-1; HRMS calcd for
1
the crude reaction mixture by H NMR indicated that >30:1
diastereomeric ratio of products was generated. Purification
by flash chromatography (Rf 0.21 using 3% Et2O in hexanes)
followed by removal of solvents at 45 °C/0.01 mmHg afforded
86% (10 mg, 0.018 mmol) of 51: 1H NMR (500 MHz, CDCl3) δ
0.58-0.63 (m, 1H), 0.65-0.74 (m, 1H), 0.79-1.00 (m, 4H),
1.06-1.22 (m, 4H), 1.47-1.68 (m, 6H), 1.75-1.92 (m, 2H),
2.04-2.10 (m, 1H), 2.24-2.30 (m, 2H), 2.63-2.69 (m, 1H), 3.08
(dd, J ) 9.0, 5.6 Hz, 1H), 4.15-4.18 (m, 2H), 4.73-4.76 (m,
1H), 7.16-7.26 (m, 9H), 7.33-7.50 (m, 11H); 13C NMR (125
MHz, CDCl3) δ 16.47, 25.87, 26.08, 26.15, 30.31, 32.29, 33.16,
37.73, 48.35, 52.67, 57.69, 84.04, 86.30, 124.00, 126.82, 127.51,
127.91, 129.19, 129.46, 132.59, 135.25, 135.85, 145.36; IR
(neat) 2130 cm-1; HRMS calcd for C41H46OSi 582.3318, found
582.3318; LRMS (EI) m/z (relative intensity) 350 (31), 243
(100), 165 (72).
(5E,1R*,2R*,4S*,8R*)-5-[5′-(2′,2′-Dim eth yl-1′,3′-dioxan yl)-
m eth ylen e]-2-[(ph en ylsilyl)m eth yl]-1-[(tr iph en ylm eth yl)-
oxy]bicyclo[3.3.0]octa n e (53) was prepared from 52 (24 mg,
0.046 mmol) according to the general experimental procedure.
1
Analysis of the crude reaction mixture by H NMR indicated
C
23H34Si 338.2430, found 338.2425; LRMS (EI) m/z (relative
that a >40:1 diastereomeric ratio of products was generated.
Purification by flash chromatography using 20% Et2O in
hexanes (Rf 0.16 using 10% Et2O in hexanes) followed by
removal of solvents at 50 °C/0.01 mmHg afforded 98% (28 mg,
intensity) 338 (3), 282 (73), 199 (64), 107 (59), 81 (71), 55 (100),
41 (90).
(1E,2R*,4R/S,6R*)-1-(Cycloh exylm eth ylen e)-5,5-dim eth -
yl-4-[(p h en ylsilyl)m eth yl]bicyclo[3.3.0]octa n e (49) was
prepared from 10 (70 mg, 0.29 mmol) according to the general
experimental procedure outlined for the preparation of 47.
1
0.046 mmol) of 53 as a white solid: mp 54-60 °C; H NMR
(500 MHz, C6D6) δ 0.57-0.64 (m, 1H), 0.66-0.75 (m, 1H),
0.81-0.90 (m, 2H), 1.28-1.32 (m, 1H), 1.37 (s, 3H), 1.42 (s,
3H), 1.61-1.67 (m, 1H), 1.78-1.85 (m, 1H), 2.05-2.11 (m, 1H),
2.26-2.36 (m, 2H), 2.51-2.59 (m, 1H), 2.64-2.70 (m, 1H), 3.09
(dd, J ) 9.1, 5.7 Hz, 1H), 3.52-3.71 (m, 4H), 4.12-4.15 (m,
2H), 4.57-4.60 (m, 1H), 7.17-7.25 (m, 9H), 7.33-7.49 (m,
(23) (a) Threlkel, R. S.; Bercaw, J . E. J . Organomet. Chem. 1977,
136, 1. (b) Den Haan, K. H.; Teuben, J . H. J . Organomet. Chem. 1987,
322, 321. (c) Den Haan, K. H.; Wielstra, Y.; Eshuis, J . J . W.; Teuben,
J . H. J . Organomet. Chem. 1987, 323, 181.

Sequential Cyclization/Silylation of Dienynes
J . Org. Chem., Vol. 63, No. 16, 1998 5515
11H); 13C NMR (125 MHz, C6D6) δ 16.75, 19.68, 26.15, 28.98,
30.47, 32.67, 35.91, 48.73, 53.15, 57.71, 64.13, 64.16, 84.15,
86.84, 97.39, 114.00, 127.18, 127.87, 128.34, 129.56, 129.90,
132.60, 135.61, 145.26, 145.82; IR (neat) 2130 cm-1; HRMS
calcd for C40H43O3Si (M - CH3)+ 599.2981, found 599.2958;
LRMS (EI) m/z (relative intensity) 350 (14), 243 (100), 165 (76).
(5E,1R*,2R*,4S*,8R*)-1-[(ter t-Bu tyld ip h en ylsilyl)oxy]-
5-[5′-(2′,2′-d im et h yl-1′,3′-d ioxa n yl)m et h ylen e]-2-[(p h en -
ylsilyl)m eth yl]bicyclo[3.3.0]octa n e (55) was prepared from
54 (29 mg, 0.058 mmol) according to the general experimental
procedure. Analysis of the crude reaction mixture by 1H NMR
indicated that a 14:1 diastereomeric ratio of products was
generated. Purification by flash chromatography using 0-10%
Et2O in hexanes (Rf 0.43 with 20% Et2O in hexanes) followed
by removal of solvents at 50 °C/0.01 mmHg afforded 90% (32
mg, 0.052 mmol) of 55 as a 19:1 mixture of diastereomers: 1H
NMR (500 MHz, CDCl3) δ 0.93-0.83 (m, 2H), 1.00 (s, 9H),
1.10-1.15 (m, 1H), 1.29-1.33 (m, 1H), 1.38 (s, 3H), 1.39-1.45
(m, 1H with overlapping singlet at 1.43, 3H), 1.65-1.72 (m,
1H), 1.77-1.85 (m, 1H), 2.12-2.18 (m, 1H), 2.39-2.42 (m, 2H),
2.46-2.52 (m, 1H), 2.54-2.60 (m, 1H), 3.49 (dd, J ) 9.1, 3.1
Hz, 1H), 3.54-3.72 (m, 4H), 4.18-4.20 (m, 2H), 4.58-4.61 (m,
1H), 7.31-7.41 (m, 9H), 7.46-7.47 (m, 2H), 7.62-7.65 (m, 4H);
13C NMR (125 MHz, CDCl3) δ 16.77, 19.26, 19.30, 25.36, 26.94,
27.08, 28.65, 30.19, 32.82, 35.39, 49.41, 51.98, 57.84, 64.03,
64.06, 84.10, 97.29, 113.01, 127.43, 127.51, 127.92, 129.51,
129.55, 129.58, 132.34, 134.18, 134.26, 135.22, 135.88, 135.95,
145.81; IR (neat) 2130 cm-1; HRMS calcd for C38H50O3Si2
610.3299, found 610.3279; LRMS (EI) m/z (relative intensity)
610 (3), 305 (84), 227 (59), 199 (100), 135 (98), 91 (50). Anal.
Calcd for C38H50O3Si2: C, 74.70; H, 8.25. Found: C, 74.86; H,
8.55.
(5E,1R*,2R*,4R*,8S*)-1-[(ter t-Bu tyld im eth ylsilyl)oxy]-
5-(cycloh exylm eth ylen e)-2-[(p h en ylsilyl)m eth yl]bicyclo-
[3.3.0]octa n e (57a ) was prepared from 56 according to the
general experimental procedure. Analysis of the crude reac-
tion mixture by gas chromatography indicated that a 2:1
diastereomeric ratio of products was generated. Purification
by flash chromatography (Rf 0.44 in hexanes) followed by
removal of solvents at 50 °C/0.01 mmHg afforded 56% of the
major diastereomer 57a : 1H NMR (500 MHz, CDCl3) δ -0.04
(s, 3H), -0.03 (s, 3H), 0.77-0.90 (m, 1H with overlapping
singlet at 0.83, 9H), 0.94-1.04 (m, 3H), 1.09-1.29 (m, 4H),
1.45-1.62 (m, 6H), 1.64-1.70 (m, 2H), 1.92-2.05 (m, 2H),
2.30-2.36 (m, 1H), 2.49-2.59 (m, 2H), 2.64-2.71 (m, 1H), 3.64
(d, J ) 3.8 Hz, 1H), 4.26-4.32 (m, 2H), 4.82-4.85 (m, 1H),
7.32-7.39 (m, 3H), 7.54-7.56 (m, 2H); 13C NMR (125 MHz,
CDCl3) δ -4.76, -4.54, 17.94, 18.17, 21.14, 25.89, 26.13, 26.21,
32.49, 33.24, 33.26, 33.38, 37.82, 52.64, 52.75, 55.72, 78.11,
124.32, 127.97, 129.55, 132.38, 135.19, 139.68; IR (neat) 2136
cm-1; HRMS calcd for C28H46OSi2 454.3087, found 454.3058;
LRMS (EI) m/z (relative intensity) 454 (4.3), 305 (63), 171 (59),
73 (100).
(5E,1R*,2S*,4R*,8S*)-1-[(ter t-Bu tyld im eth ylsilyl)oxy]-
5-(cycloh exylm eth ylen e)-2-[(p h en ylsilyl)m eth yl]bicyclo-
[3.3.0]octa n e (57b) was prepared from 56 according to the
general experimental procedure. Purification by flash chro-
matography (Rf 0.54 in hexanes) followed by removal of
solvents at 50 °C/0.01 mmHg afforded 10 mg (0.022 mmol) of
the minor diastereomer 57b in 30% yield: 1H NMR (500 MHz,
CDCl3) δ -0.01 (s, 3H), 0.02 (s, 3H), 0.89 (s, 9H), 0.92-1.03
(m, 3H), 1.09-1.31 (m, 5H), 1.39-1.60 (m, 7H), 1.64-1.69 (m,
2H), 1.91-1.98 (m, 1H), 2.42-2.55 (m, 3H), 2.75-2.82 (m, 1H),
3.80 (dd, J ) 4.1, 2.9 Hz, 1H), 4.26-4.32 (m, 2H), 4.81-4.84
(m, 1H), 7.32-7.39 (m, 3H), 7.54-7.56 (m, 2H); 13C NMR (125
MHz, CDCl3) δ -4.59, -4.36, 12.49, 18.32, 22.18, 26.04, 26.13,
26.19, 29.92, 32.16, 33.27, 37.80, 46.66, 51.56, 57.07, 72.33,
123.88, 127.94, 129.42, 132.92, 135.19, 139.89; IR (neat) 2130
cm-1; HRMS calcd for C28H46OSi2 454.3087, found 454.3073;
LRMS (EI) m/z (relative intensity) 454 (3.2), 171 (100), 73 (97).
(5E,1R*,2R*,4R*,8S*)-5-(Cycloh exylm eth ylen e)-2-[(ph en -
ylsilyl)m et h yl]-1-[(t r ip h en ylm et h yl)oxy]b icyclo[3.3.0]-
octa n e (59) was prepared from 58 (15 mg, 0.032 mmol)
according to the general experimental procedure. Analysis of
the crude reaction mixture by 1H NMR indicated that a >60:1
diastereomeric ratio of products was generated. Purification
by flash chromatography (Rf 0.24 using 5% CH2Cl2 in hexanes)
followed by removal of solvents at 45 °C/0.01 mmHg afforded
97% (18 mg, 0.031 mmol) of 59: 1H NMR (500 MHz, CDCl3) δ
0.51-0.61 (m, 2H), 0.60-0.69 (m, 1H), 0.94-1.05 (m, 2H),
1.09-1.28 (m, 4H), 1.47-1.71 (m, 7H), 1.91-1.97 (m, 2H),
2.08-2.14 (m, 1H), 2.41-2.48 (m, 1H), 2.56-2.61 (m, 1H),
2.91-2.97 (m, 1H), 3.69 (d, J ) 4.4 Hz, 1H), 3.95-3.98 (m,
1H), 4.05-4.08 (m, 1H), 4.84-4.87 (m, 1H), 7.16-7.23 (m, 9H),
7.29-7.45 (m, 11H); 13C NMR (125 MHz, CDCl3) δ 17.91, 22.27,
26.11, 26.21, 32.59, 33.06, 33.19, 37.85, 50.74, 54.09, 55.05,
80.08, 86.74, 124.80, 126.88, 127.57, 127.87, 129.24, 129.41,
132.71, 135.17, 139.48, 145.54; IR (neat) 2130 cm-1; HRMS
calcd for C41H46OSi: 582.3318, found 582.3282; LRMS (EI) m/z
(relative intensity) 350 (23), 243 (100), 165 (87), 107 (68), 81
(51).
(1E,2R*,4S*,5S*,6R*)-1-[5′-(2′,2′-Dim eth yl-1′,3′-dioxan yl)-
m eth ylen e]-5-[p h en yl(m eth yl)]-4-[(p h en ylsilyl)m eth yl]-
bicyclo[3.3.0]octa n e (60) was prepared from 35 (22 mg, 0.065
mmol, 30:1 ratio of diastereomers) according to the general
experimental procedure. Analysis of the crude reaction mix-
ture by gas chromatography indicated that a 30:1 diastereo-
meric ratio of products was generated. Purification by flash
chromatography (Rf 0.39 using 20% Et2O in hexanes) followed
by removal of solvents at 50 °C/0.01 mmHg afforded 89% (26
mg, 0.058 mmol) of 60: 1H NMR (400 MHz, CDCl3) δ 0.83-
0.98 (m, 3H), 1.25-1.41 (m, 1H with overlapping singlet at
1.39, 3H), 1.45 (s, 3H), 1.47-1.53 (m, 1H), 1.84-2.00 (m, 3H),
2.16-2.24 (m, 1H), 2.32-2.43 (m, 2H), 2.46-2.68 (m, 3H), 2.74
(dd, J ) 13.9, 6.7 Hz, 1H), 3.59-3.75 (m, 4H), 4.10-4.16 (m,
2H), 4.69-4.73 (m, 1H), 7.11-7.38 (m, 10H); 13C NMR (100
MHz, CDCl3) δ 19.28, 23.21, 28.71, 32.72, 34.69, 35.48, 37.54,
45.92, 47.54, 53.91, 54.45, 64.10, 97.31, 113.73, 125.65, 127.89,
128.16, 129.03, 129.41, 132.30, 135.14, 141.09, 146.00; IR
(neat) 2130 cm-1; HRMS calcd for C29H38O2Si 446.2641, found
446.2644; LRMS (EI) m/z (relative intensity) 446 (0.8), 267
(58), 107 (100), 91 (99).
(1E,2R*,3S*,4R*)-3-[(ter t-Bu tyld im eth ylsilyl)oxy]-1-(cy-
cloh exylm eth ylen e)-4-[2′-(p h en ylsilyl)eth yl]-2-[(p h en yl-
silyl)m eth yl]cyclop en ta n e (61) was prepared from 38 (20
mg, 0.060 mmol) according to the general experimental
procedure. Purification by flash chromatography (Rf 0.29 in
hexanes) followed by removal of solvents at 50 °C/0.01 mmHg
afforded 70% (23 mg, 0.042 mmol) of 61: GC purity >99%; 1H
NMR (400 MHz, CDCl3) δ -0.06 (s, 3H), -0.05 (s, 3H), 0.78-
1.03 (m, 6H with overlapping singlet at 0.79, 9H), 1.08-1.27
(m, 3H), 1.36-1.45 (m, 1H), 1.53-1.68 (m, 6H), 1.94-2.02 (m,
3H), 2.29-2.37 (m, 1H), 2.41-2.44 (m, 1H), 3.74-3.76 (m, 1H),
1
4.27 (t, J ) 3.7 Hz, J 29Si,H ) 192.0 Hz, 2H), 4.31 (t, J ) 4.0
1
Hz, J 29Si,H ) 194.1 Hz, 2H), 5.02 (dd, J ) 8.9, 1.1 Hz, 1H),
7.31-7.38 (m, 6H), 7.53-7.56 (m, 4H); 13C NMR (100 MHz,
CDCl3) δ -4.71, -4.38, 8.68, 15.15, 18.02, 24.57, 25.77, 26.09,
26.22, 31.32, 32.82, 33.10, 38.50, 44.71, 49.46, 79.82, 127.94,
127.97, 128.74, 129.47, 132.74, 132.77, 135.21, 142.88; IR
(neat) 2133 cm-1; HRMS calcd for C33H51OSi3 (M-H)+ 547.3248,
found 547.3236; LRMS (EI) m/z (relative intensity) 491 (30),
355 (52), 257 (70), 181 (100), 135 (66), 107 (86), 73 (77). Anal.
Calcd for C33H52OSi3: C, 72.19; H, 9.55. Found: C, 72.47; H,
9.91.
(5E,1R*,2R*,4S*,8S*)-1-[(ter t-Bu tyld im eth ylsilyl)oxy]-
5-(cycloh exylm eth ylen e)-2-[(p h en ylsilyl)m eth yl]bicyclo-
[4.430]n on a n e (62) was prepared from 46 according to the
general experimental procedure. Purification by flash chro-
matography using 10% CH2Cl2 in hexanes followed by removal
of solvents at 50 °C/0.01 mmHg afforded 62 in 90% yield: Rf
1
0.34 (10% CH2Cl2 in hexanes); H NMR (500 MHz, CDCl3) δ
0.49-0.55 (m, 1H), 0.66 (q, J ) 11.9 Hz, 1H), 0.81-1.25 (m,
7H with overlapping singlet at 1.02, 9H), 1.46-1.72 (m, 8H),
1.77-1.82 (m, 1H), 1.88-1.96 (m, 2H), 2.05-2.13 (m, 2H),
2.20-2.25 (m, 1H), 3.30 (dt, J ) 10.3, 4.4 Hz, 1H), 4.30 (t, J )
4.4 Hz, 1J 29Si,H ) 191.7 Hz, 2H), 4.71-4.75 (m, 1H), 7.31-7.41
(m, 8H), 7.51-7.53 (m, 2H), 7.65-7.69 (m, 4H); 13C NMR (125
MHz, CDCl3) δ 14.72, 20.21, 26.83, 26.90, 27.89, 28.43, 30.25,
33.84, 34.84, 38.57, 41.52, 42.92, 45.33, 50.06, 79.97, 124.92,
128.04, 128.23, 128.63, 130.08, 130.11, 130.20, 133.64, 134.90,

5516 J . Org. Chem., Vol. 63, No. 16, 1998
Molander and Retsch
135.64, 136.00, 136.69, 136.76, 142.79; IR (neat) 2133.4 cm-1
;
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails and characterization for intermediates 2, 3, 5-10, 12-
HRMS calcd for C35H43OSi2 (M - C4H9)+ 535.2852, found
538.2835; LRMS (EI) m/z (relative intensity) 535 (24), 305
(100), 275 (68), 201 (55).
1
30, 32-42, 44-46, 52, 54, 56, and 58; H and 13C spectra for
all compounds for which elemental analysis is not reported;
NOE and other spectral structural studies for compounds 16,
17, 32, 51, 53, 59, 61, and 62 (175 pages). This material is
contained in libraries on microfiche, immediately 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.
Ack n ow led gm en t. We gratefully acknowledge the
National Institutes of Health (GM48580), the National
Science Foundation (INT-9422011), and NATO (950674)
for their generous support of this research. In addition,
we would also like to thank Eric Dowdy for his help with
the NOE experiments for compound 62.
J O980521T