Diastereoselective nucleophilic substitution reactions of oxasilacyclopentane acetals: Application of the "inside attack" model for reactions of five-membered ring oxocarbenium ions

Article abstract of DOI:10.1021/jo010823m

The additions of nucleophiles to oxocarbenium ions derived from oxasilacyclopentane acetates proceeded with high diastereoselectivity in most cases. Sterically demanding nucleophiles such as the silyl enol ether of diethyl ketone add to the face opposite the C-2 substituent. These reactions establish the syn stereochemistry about the newly formed carbon-carbon bond. Small nucleophiles such as allyltrimethylsilane do not show this same stereochemical preference: they add from the same face as the substituent in C-2-substituted oxocarbenium ions. The stereoselectivities exhibited by both small and large nucleophiles can be understood by application of the "inside attack" model for five-membered ring oxocarbenium ions developed previously for tetrahydrofuran-derived cations. This stereoelectronic model requires attack of the nucleophile from the face of the cation that provides the products in their lower energy staggered conformations. Small nucleophiles add to the "inside" of the lower energy ground-state conformer of the oxocarbenium ion. In contrast sterically demanding nucleophiles add to the inside of the envelope conformer where approach is anti to the C-2 substituent of the oxocarbenium ion, regardless of the ground-state conformer population.

Full text of DOI:10.1021/jo010823m

2056
J . Org. Chem. 2002, 67, 2056-2064
Dia ster eoselective Nu cleop h ilic Su bstitu tion Rea ction s of
Oxa sila cyclop en ta n e Aceta ls: Ap p lica tion of th e “In sid e Atta ck ”
Mod el for Rea ction s of F ive-Mem ber ed Rin g Oxoca r ben iu m Ion s
Teresa J . Bear, J ared T. Shaw, and K. A. Woerpel*
Department of Chemistry, University of California, Irvine, California 92697-2025
kwoerpel@uci.edu
Received August 13, 2001
The additions of nucleophiles to oxocarbenium ions derived from oxasilacyclopentane acetates
proceeded with high diastereoselectivity in most cases. Sterically demanding nucleophiles such as
the silyl enol ether of diethyl ketone add to the face opposite the C-2 substituent. These reactions
establish the syn stereochemistry about the newly formed carbon-carbon bond. Small nucleophiles
such as allyltrimethylsilane do not show this same stereochemical preference: they add from the
same face as the substituent in C-2-substituted oxocarbenium ions. The stereoselectivities exhibited
by both small and large nucleophiles can be understood by application of the “inside attack” model
for five-membered ring oxocarbenium ions developed previously for tetrahydrofuran-derived cations.
This stereoelectronic model requires attack of the nucleophile from the face of the cation that
provides the products in their lower energy staggered conformations. Small nucleophiles add to
the “inside” of the lower energy ground-state conformer of the oxocarbenium ion. In contrast,
sterically demanding nucleophiles add to the inside of the envelope conformer where approach is
anti to the C-2 substituent of the oxocarbenium ion, regardless of the ground-state conformer
population.
In tr od u ction
In the course of research aimed at developing the
reactions of silacyclopropanes as methods for stereo-
selective organic synthesis,1 we developed procedures for
converting silacyclopropanes such as 1 to oxasilacyclo-
pentane acetals 2 in high yield and with a high degree
of stereochemical control (eq 1). We envisioned that these
acetals would be useful synthetic intermediates, because
Lewis acid-mediated nucleophilic substitution reactions
employing allylic silanes or silyl enol ethers should afford
oxasilacyclopentanes 3. The configuration at the newly
formed stereogenic center would be established by attack
on the intermediate oxocarbenium ion from the face
opposite the substituent at C-2. The resulting product
could be converted to diol 4 by oxidation of the carbon-
silicon bond (eq 2).2-6
In this paper, we report full details of the diastereo-
selective reactions of oxasilacyclopentane acetals such as
2 with carbon nucleophiles. We evaluated the reactions
of a range of oxasilacyclopentane acetals with various
substituents at both C-2 and C-3 with various nucleo-
philes to determine the generality and limitations of the
idea described in eq 2. Reactions of the trans-2,3-dimethyl
acetal 2 provided products where the nucleophile had
indeed approached the putative oxocarbenium ion inter-
mediate from the face opposite the methyl group at C-2
as anticipated.7 In several cases, however, the stereo-
selectivities were high, but the sense of selectivity was
opposite what might be predicted on the basis of a cursory
examination of the substrate.8 These seemingly contras-
teric results, as well as the less surprising selectivities,
can be accommodated by application of our recently
reported “inside attack” model to explain stereoselective
reactions of tetrahydrofuran-derived oxocarbenium ions.9
The results and analysis in this paper demonstrate that
the inside attack model9 can be applied with success to
systems that differ significantly from those for which the
model was originally derived.10 Furthermore, the experi-
ments described here suggest that oxasilacyclopentane
acetals are useful synthetic intermediates.
(1) Franz, A. K.; Woerpel, K. A. Acc. Chem. Res. 2000, 33, 813-
820.
(2) Tamao, K.; Ishida, N.; Tanaka, T.; Kumada, M. Organometallics
1983, 2, 1694-1696.
(7) Shaw, J . T.; Woerpel, K. A. Tetrahedron 1997, 53, 16597-16606.
(8) Shaw, J . T.; Woerpel, K. A. J . Org. Chem. 1997, 62, 6706-6707.
(9) Larsen, C. H.; Ridgway, B. H.; Shaw, J . T.; Woerpel, K. A. J .
Am. Chem. Soc. 1999, 121, 12208-12209.
(10) Iminium ions are believed to undergo nucleophilic attack by
transition structures similar to those for oxocarbenium ions: Bur, S.
K.; Martin, S. F. Org. Lett. 2000, 2, 3445-3447.
(3) Fleming, I. Chemtracts: Org. Chem. 1996, 9, 1-64.
(4) J ones, G. R.; Landais, Y. Tetrahedron 1996, 52, 7599-7662.
(5) Tamao, K. Advances in Silicon Chemistry; J AI: Greenwich, CT,
1996; Vol. 3; pp 1-62.
(6) Smitrovich, J . H.; Woerpel, K. A. J . Org. Chem. 1996, 61, 6044-
6046.
10.1021/jo010823m CCC: $22.00 © 2002 American Chemical Society
Published on Web 03/14/2002

Substitution Reactions of Oxasilacyclopentane Acetals
J . Org. Chem., Vol. 67, No. 7, 2002 2057
Nu cleop h ilic Su bstitu tion Rea ction s
all cases just as they had with the other nucleophiles (eq
3). The newly formed carbon-carbon bond (C-1-C-1′)
was formed with syn diastereoselectivity (up to 86%).
When the (E)-silyl26 and (Z)-silyl27 enol ethers of 3-penta-
none (10) were employed as the nucleophiles (eq 5), the
Oxasilacyclopentane acetals were prepared by stereo-
specific and regioselective insertion reactions of carbonyl
compounds into silacyclopropanes.7,11,12 In earlier publica-
tions of these reactions,7,11,12 we isolated and purified the
products of insertion. We have found that it is operation-
ally easier to carry the insertion products through
subsequent steps without purification and only purify the
desired oxasilacyclopentane acetal that would be used for
the nucleophilic substitution reaction. The details of the
syntheses of these substrates are included as Supporting
Information. In early experiments, the acetate derivatives
provided the highest yields of nucleophilic substitution
products with the fewest side products, so we focused on
these substrates.
2,3-Dim eth yl-Su bstitu ted Oxasilacyclopen tan e Ac-
eta ls. The acetate trans-513 underwent stereoselective
nucleophilic substitution reactions with both (2-phenyl-
allyl)trimethylsilane14,15 (6a ) and 1-(trimethylsilyloxy)-
styrene16 (6b; eq 3).17 These reactions proceeded with
high 1,2-trans diastereoselectivity,18 indicating that the
nature of the nucleophilic alkene does not determine the
stereoselectivity of its reactions.19 A variety of Lewis
acids, solvents, and temperatures were investigated for
these reactions to develop optimal conditions.7 The
optimized conditions employed SnBr4 as the Lewis acid
and CH2Cl2 as the solvent. Conducting the reactions at
low temperatures (-78 °C) minimized competing side
reactions such as overadditions in the case of silyl enol
ethers. These conditions were used for all nucleophilic
substitution reactions reported here.
Reactions with internal alkenes allowed for stereo-
chemical control at the exocyclic stereocenter as well as
the ring junction. Substitutions by various (E)- and (Z)-
crotylsilanes 820-25 were not highly stereoselective (eq 4),
but the 1,2-trans products 9a and 9b predominated in
nucleophilic substitution reactions were significantly
more stereoselective. The major diastereomer 11a , whose
structure was proven by X-ray crystallography,28 was
formed with high 1,2-trans selectivity, and the new
carbon-carbon bond was established with high syn
selectivity. The configurations of the minor products 9b,c
and 11b,c was assigned by chemical correlation.7,28
Although the structure of the major products did not
depend on the stereochemistry of the silyl enol ether, the
(E)-isomer proved to be more selective.
(11) Franz, A. K.; Woerpel, K. A. Angew. Chem., Int. Ed. 2000, 39,
4295-4299.
(12) Franz, A. K.; Woerpel, K. A. J . Am. Chem. Soc. 1999, 121, 949-
In sharp contrast to the results with the styrene
nucleophiles and the internal alkenes, 1,2-cis products
were formed preferentially with smaller, unhindered
957.
(13) X-ray crystallography was used to prove the stereochemistry
of this compound. The details are provided as Supporting Information.
(14) Narayanan, B. A.; Bunnelle, W. H. Tetrahedron Lett. 1987, 28,
6261-6264.
(15) For a review of the methods for the synthesis of allylsilanes,
see: Sarkar, T. K. Synthesis 1990, 969-983, 1101-1111.
(16) Walshe, N.; Goodwin, G.; Smith, G.; Woodward, F. Org. Synth.
1986, 65, 1-5.
(17) In all cases, stereoselectivities were determined by GC or GC-
MS analysis of unpurified reaction mixtures. These selectivities were
corroborated by 1H and 13C NMR spectroscopy. The yields are reported
for purified products.
(20) Wrighton, M. S.; Schroeder, M. A. J . Am. Chem. Soc. 1974, 96,
6235-6237.
(21) Hayashi, T.; Kabeta, K.; Hamachi, I.; Kumada, M. Tetrahedron
Lett. 1983, 24, 2865-2868.
(22) Matarasso-Tchiroukhine, E.; Cadiot, P. J . Organomet. Chem.
1976, 121, 155-168.
(23) Abel, E. W.; Rowley, R. J . J . Organomet. Chem. 1975, 84, 199-
229.
(18) The configuration of 7b was determined by protection of the
ketone, oxidation of the carbon-silicon bond to provide the diol, and
analysis of the 13C chemical shifts of the derived acetonides (Rych-
novsky, S. D.; Rogers, B. N.; Richardson, T. I. Acc. Chem. Res. 1998,
31, 9-17). The structure of 7a was proven by chemical correlation to
7b.
(24) Leboutet, L.; Courtois, G.; Miginiac, L. J . Organomet. Chem.
1991, 420, 155-161.
(25) Rajagopalan, S.; Zweiffel, G. Synthesis 1984, 113-115.
(26) Hall, P. L.; Gilchrist, J . H.; Collum, D. B. J . Am. Chem. Soc.
1991, 113, 9571-9574.
(27) Cazeau, P.; Duboudin, F.; Moulines, F.; Babot, O.; Dunogues,
J . Tetrahedron 1987, 43, 2075-2088.
(28) The details are provided as Supporting Information.
(19) Mayr, H.; Patz, M. Angew. Chem., Int. Ed. Engl. 1994, 33, 938-
957.

2058 J . Org. Chem., Vol. 67, No. 7, 2002
Bear et al.
terminal alkene nucleophiles. Allyltrimethylsilane under-
went substitution to give the 1,2-cis product 12 with 98%
diastereoselectivity (eq 6).29 Similarly, methallyltrimeth-
ylsilane (13a )30 and 2-((trimethylsilyl)oxy)propene (13b)16
underwent highly selective reactions, favoring the 1,2-
cis products 14a and 14b, respectively (eq 7).29
The control experiments shown in eq 8 cannot discount
such a mechanism.
The cis-disubstituted acetate cis-5, like trans-5, under-
went highly 1,2-trans-selective reactions with large nu-
cleophiles. Substitution of cis-5 with silyl enol ether 6b
provided the 1,2-trans product 16 (eq 9).36 Substitution
with the silyl enol ethers of diethyl ketone (10) gave the
expected trans product 17a with high diastereoselectivity
(eq 10). The configuration of 17a was proven by X-ray
A control experiment (eq 8) suggested that an oxocar-
benium ion was the reactive intermediate in the substi-
tution reaction with allyltrimethylsilane.31-33 The methyl
crystallography of its 2,4-dinitrophenylhydrazone deriva-
tive.28 As observed for trans-5, syn diastereoselectivity
about the C-1-C-1′ bond was favored, and the configu-
ration of the major product was independent of the alkene
isomer of the enol silane.
The reaction of cis-5 with the small nucleophile allyl-
trimethylsilane proceeded with low selectivity (eq 11).
This behavior of cis-5 is surprising when compared to the
reaction with trans-5 (eq 6). In defiance of intuition, when
acetals 15a ,b were prepared separately,34 and each
epimer was independently treated with allyltrimethyl-
silane and TiCl4 (eq 8).35 The selectivities of these
substitution reactions were within experimental error of
the selectivity observed in the reaction with the acetate
trans-5 (eq 6). Because the stereoselectivity was inde-
pendent of both the leaving group and the configuration
of the acetal, we propose that these reactions proceed via
a common oxocarbenium ion intermediate. If isomeriza-
tion of the acetal epimers occurred during the reaction,
however, a direct nucleophilic substitution reaction (SN2)
would also account for this stereochemical convergence.
the two methyl groups are on opposite sides of the ring
(as in trans-5), attack occurs overwhelmingly from one
face, but when they are on the same side (as in cis-5),
the nucleophile cannot differentiate the two faces of the
ring.
2- a n d 3-Alk yl-Su bstitu ted Oxa sila cyclop en ta n e
Aceta ls. To elucidate the role of each alkyl substituent
in the 2,3-dimethyl systems, substitution reactions on
monosubstituted oxasilacyclopentane acetals were per-
formed. Both small (methyl) and large (isopropyl) sub-
stituents were placed at both C-2 and C-3, and the
reactions with both small and large nucleophiles were
examined.
(29) The stereochemistry was proven by oxidation to the diol and
conversion to the derived acetonide. The details are provided as
Supporting Information.
(30) Ishihara, K.; Mouri, M.; Gao, Q.; Maruyama, T.; Furuta, K.;
Yamamoto, H. J . Am. Chem. Soc. 1993, 115, 11490-11495.
(31) There has been much discussion of the mechanism of Lewis
acid-promoted acetal substitution in the literature. See, for example:
(a) Mori, I.; Ishihara, K.; Flippin, L. A.; Nozaki, K.; Yamamoto, H.;
Bartlett, P. A.; Heathcock, C. H. J . Org. Chem. 1990, 55, 6107-6115.
(b) Denmark, S. E.; Almstead, N. G. J . Org. Chem. 1991, 56, 6458-
6467.
(32) Sammakia has demonstrated that the SnCl4-mediated reaction
of dimethyl acetals proceeds through an oxonium ion intermediate:
Sammakia, T.; Smith, R. S. J . Am. Chem. Soc. 1994, 116, 7915-7916.
(33) Acyclic alkoxyalkyl acetates undergo substitution reactions with
nucleophilic alkenes via oxonium ion intermediates: Matsutani, H.;
Ichikawa, S.; Yaruva, J .; Kusumoto, T.; Hiyama, T. J . Am. Chem. Soc.
1997, 119, 4541-4542.
Nucleophilic substitution reactions of the 2-alkyloxa-
silacyclopentane acetates with the large nucleophile (E)-
10 proceeded with high 1,2-trans selectivity. The reaction
(34) We do not know the stereochemistry of the two acetals, but that
knowledge is not important since we have both stereoisomers.
(35) TiCl4 was used as the Lewis acid since SnBr4 was not acidic
enough to promote the reaction with the methyl acetals.
(36) The configuration of 16 was proven by conversion to the 1,3-
diol acetonide. The details are provided as Supporting Information.

Substitution Reactions of Oxasilacyclopentane Acetals
J . Org. Chem., Vol. 67, No. 7, 2002 2059
of the 2-methyl acetate 19 afforded predominately 20a ,
which possessed 1,2-trans stereochemistry as well as high
selectivity about the C-1-C-1′ bond (eq 12).37 A similar
major product (22a ) was observed with the more hindered
2-isopropyl acetate 21 (eq 13).38
trans selectivity with large nucleophiles (eqs 12 and 13)
as Reissig observed,41,42 in that study it was reported that
reactions with allyltrimethylsilane proceeded with 1,2-
trans selectivity as well (albeit of lower magnitude).41,42
We have previously demonstrated that a 2-methyltet-
rahydrofuran acetal with geminal substitution at C-4
(i.e., the carbon analogue of 19) reacts with 1,2-cis
selectivity comparable to the acetate 19.43 The attenuated
1,2-cis selectivity for the isopropyl-substituted acetal 21
suggests that developing steric interactions between the
nucleophile and the resident substituent can influence
the selectivity. Taken together, these experiments show
that the unusual selectivities observed for acetates 19
and 21 are not due to the presence of the silicon atom,
but are caused by the geminal substitution on the silicon.
Nucleophilic substitution reactions of the 3-alkyloxa-
silacyclopentane acetals 25 and 26 proceeded with high
1,3-trans diastereoselectivity even for small nucleophiles.
With both the 3-methyl and the 3-isopropyl groups, 1,3-
trans products were observed with allyltrimethylsilane
(eq 16).44 Similar selectivities were observed by Reissig
for the analogous tetrahydrofuran acetals.41,42
Nucleophilic substitution on the mono-alkyl-substitut-
ed acetates 19 and 21 with the alkene 6b occurred with
only modest diastereoselectivity (eq 14). In the case of
the 2-isopropyl substrate 21, the reaction was somewhat
more selective, favoring the 1,2-trans product 23b.39
Exp la n a tion for th e Obser ved Ster eoselectivities
The selectivities exhibited by oxasilacyclopentane ac-
etates cannot be rationalized by examining the disposi-
tions of the substituents on the five-membered ring. For
example, the allylations of some systems showed pref-
erential allylation cis to the substituent at C-2, even in
the absence of other stereocenters in the molecule (eq 15).
At the stage when we had examined a relatively small
collection of five-membered ring oxocarbenium ion sys-
tems, we proposed a tentative explanation.8 The selec-
tivities of additions to oxasilacyclopentane oxocarbenium
ions derived from cis-5, 19, and 21 were not consistent
with that model, however.
In contrast to the reactions with the larger nucleophiles
10 and 6b (eqs 12-14), substitution by allyltrimethylsi-
lane gave 1,2-cis products 24 (eq 15).40 These results are
Concurrent with the oxasilacyclopentane oxocarbenium
ion studies, we developed the inside attack model that
correctly predicts additions of allyltrimethylsilane to
oxocarbenium ions derived from tetrahydrofurans.5 The
inside attack model proposes that approach of a nucleo-
phile to a five-membered ring oxocarbenium ion occurs
to the “inside” face of the envelope conformer adopted
by the cation (transition structures 28a ,b, eq 17). Attack
to the inside of the envelope leads to an all-staggered,
lower energy conformer (29a ,b). Attack from the “outside”
is disfavored because it leads to a higher energy product
(31a ,b) that experiences eclipsing interactions between
substituents at C-1 and C-2 (eq 18). Extrapolation of this
stereoelectronic model can explain the selectivities ob-
served in the oxasilacyclopentane oxocarbenium ions
reported here. It should be noted that in the course of
his observations of five-membered ring oxocarbenium ion
inconsistent not only with intuition but also with studies
of simple 2-methyltetrahydrofuran acetals.41,42 Although
the 2-methyl-substituted acetate 19 showed high 1,2-
(37) The configuration of this product was assigned by an NMR
chemical shift correlation. The details are provided as Supporting
Information.
(38) The configuration of the ring junction was proven by NOE
measurements, and the stereochemistry at C-1′ was assigned by
analogy to the examples proven by X-ray crystallography. The details
are provided as Supporting Information.
(39) The configuration of 23a was not proven. The stereochemistry
of 23b was proven by an NMR shift correlation. The details are
provided as Supporting Information.
(40) The configuration of 24a was determined by conversion to a
known compound. The structure of 24b was assigned by conversion to
a cyclic acetal and examination of coupling constants. The details are
provided as Supporting Information.
(42) Schmitt, A.; Reissig, H.-U. Eur. J . Org. Chem. 2000, 3893-
3901.
(43) Shaw, J . T.; Woerpel, K. A. Tetrahedron 1999, 55, 8747-8756.
(44) The configuration of 27a was determined by conversion to a
known compound. The silane 27b was converted to the diol acetonide.
The details are provided as Supporting Information.
(41) Schmitt, A.; Reissig, H.-U. Synlett 1990, 40-42.

2060 J . Org. Chem., Vol. 67, No. 7, 2002
Bear et al.
on the lower energy conformers of the oxocarbenium ions.
This point is demonstrated by the analysis of the high
1,2-cis selectivity (98:2) observed for the addition of
allyltrimethylsilane to the oxocarbenium ion derived from
trans-5 (eq 6). Semiempirical (AM-1) calculations indicate
that 32b is the preferred conformer for the oxocarbenium
ion generated from trans-5 (eq 19). These calculations
reactivities, Reissig proposed a modified Felkin-Anh
model to explain the reactions of these intermediates.41,42
To understand the selectivities exhibited by substituted
oxasilacyclopentane acetates, the transition structures
for nucleophilic attack must be analyzed to determine
their relative energies. Because attack of the silyl enol
ether or allylic silane on the oxocarbenium ion is irre-
versible,45,46 this step determines the stereochemical
course of the reaction. Although interconversion between
the conformers of the oxocarbenium ions is likely to be
rapid relative to nucleophilic attack,47-49 the transition
structures for nucleophilic attack bear some resemblance
to the starting carbocations. Consequently, the relative
energies of the oxocarbenium ion conformers will be
reflected in the energies of the transition structures.
Developing steric interactions in the products will also
influence the energies of the transition structures.
The magnitude of the steric interactions that develop
upon nucleophilic attack (such as gauche- or syn-pentane
interactions) will depend on the size of the nucleophile.
Attack by a small nucleophile will engender less desta-
bilization in the transition state than attack by a large
nucleophile would. Consequently, when diastereomeric
transition states involving a small nucleophile are con-
sidered, developing destabilizing interactions will not
play as large a role in determining diastereoselectivity
as it would with a larger nucleophile.
indicate that the dihedral angle for the C-3 axial sub-
stituent and the pseudoequatorial tert-butyl is approxi-
mately 0°. Therefore, in addition to the pseudo-1,3-diaxial
interaction between the C-2 methyl and the pseudoaxial
tert-butyl group in 32a , the axial C-3 methyl group of
32a experiences an eclipsing interaction with the pseudo-
equatorial tert-butyl group. Consequently, the more
populated conformer is 32b, and allyltrimethylsilane
adds to the inside to generate the observed 1,2-cis adduct
12 with high diastereoselectivity.
This model can also be used to understand the lack of
selectivity observed for the cis analogue (cis-5, eq 11) with
allyltrimethylsilane. Considering the interactions de-
scribed for the trans-substituted case (vide supra), the
two ground-state conformers of the oxocarbenium ion
(33a ,b) do not differ much in energy, because each
conformer experiences an unfavorable interaction (eq 20).
Because the two conformers are similarly populated,
allyltrimethylsilane adds to the inside of both 33a and
33b to give a 57:43 mixture of diastereomeric products.
The idea that the stereoselective reactions involving
small nucleophiles should be more dependent upon
interactions in the reactant than in the product is
consistent with the experiments reported here. For small
nucleophiles such as allyltrimethylsilane, the relative
configuration of the product can be explained by nucleo-
philic attack to the inside of the lower energy envelope
conformer of the cation, because minimal steric inter-
actions develop in the transition state. The additions of
larger nucleophiles, on the other hand, are controlled by
developing steric interactions. For these nucleophiles,
addition occurs to the inside of the envelope conformer
that provides fewer steric interactions in the transition
state regardless of the relative energies of the oxocarbe-
nium ion conformers.
As the size of the nucleophiles increases, destabilizing
steric interactions between the nucleophile and the C-2
substituent develop in the transition structure. As a
result, knowledge of the ground-state conformer popula-
tions alone is insufficient to understand the selectivities
that are observed. The magnitudes of the interactions
between the nucleophile and the substituent depend on
the size of the nucleophile and the size of the substituent
with which it interacts. With nucleophiles of moderate
size, such as 2-((trimethylsilyl)oxy)styrene (6b), the
destabilizing interactions can be either modest or sig-
nificant depending on the size of the substituent (methyl
or isopropyl) at C-2 (eq 14). With the largest nucleophile
examined, silyl enol ether 10, however, the steric interac-
tions between the nucleophile and the C-2 substituent
in the transition state dominate the stereoselectivity,
thereby leading to 1,2-trans stereochemistry in all cases
(eqs 5, 10, 12, and 13). Nucleophilic addition still occurs
to the inside of the envelope conformation, but approach
occurs onto the higher energy conformer of the oxocar-
For small nucleophiles such as allyltrimethylsilane,
methallyltrimethylsilane, and 2-((trimethylsilyl)oxy)pro-
pene, the major products are obtained by inside attack
(45) Burfeindt, J .; Patz, M.; Mu¨ller, M.; Mayr, H. J . Am. Chem. Soc.
1998, 120, 3629-3634.
(46) Hagen, G.; Mayr, H. J . Am. Chem. Soc. 1991, 113, 4954-4961.
(47) Seeman, J . I. J . Chem. Educ. 1986, 63, 42-48.
(48) Seeman, J . I. Chem. Rev. 1983, 83, 83-184.
(49) Chandrasekhar, S. Res. Chem. Intermed. 1997, 23, 55-62.

Substitution Reactions of Oxasilacyclopentane Acetals
J . Org. Chem., Vol. 67, No. 7, 2002 2061
benium ion to avoid development of steric interactions,
in accord with the Curtin-Hammett principle.47-49
The stereoselective reaction of 2-methyl acetate 19 with
(E)-10 (eq 12) illustrates the role of destabilizing inter-
actions on stereoselectivity. The preferred conformer of
the derived oxocarbenium ion is 34a with the methyl
group in the equatorial position (eq 21), but inside attack
would not greatly influence the preferences for these two
reaction pathways. As the prochiral sp2-hybridized nu-
cleophile approaches the oxocarbenium ion along the
Bu¨rgi-Dunitz trajectory, the position antiperiplanar to
the oxygen is more sterically crowded than the synclinal
position.54,55 We believe that transition structures resem-
bling A are favored because the smaller sp2-hybridized
carbon (Rmedium) adopts the more sterically demanding
antiperiplanar position (A′) as compared to the methyl
group (Rlarge) in this position (transition structure B′).
of the sterically demanding nucleophile 10 on this
conformer would develop a severe steric interaction
between the methyl group and the approaching nucleo-
phile. Because the two conformers 34a and 34b are in
rapid equilibrium, the lower energy transition state can
be accessed by the addition of the large nucleophile to
the higher energy conformer 34b (eq 21). Addition to the
inside of the less populated conformer 34b affords the
observed 1,2-trans product 20a with high (g92%) dia-
stereoselectivity.
The reactions of internal alkenes such as crotylsilanes
8 and silyl enol ethers 10 not only control the stereo-
chemical relationships on the five-membered ring, but
also establish the syn configuration about the C-1-C-1′
bond. This stereochemical result is independent of the
nucleophile configuration (E or Z) or the nature of the
nucleophile (allylic silane or silyl enol ether). These
observations contrast with other work in the literature.50
For example, Scolastico has shown that nucleophilic
additions to a cyclic oxocarbenium ion proceed with either
syn or anti stereochemistry depending upon whether silyl
enol ethers51 or crotylstannanes52 were employed.
In the case of the oxocarbenium ions derived from
oxasilacyclopentanes, the configuration of the exocyclic
stereogenic center is most likely established by an anti
transition structure. A priori, two possible staggered
transition structures can be considered for the addition:
A, in which the orientation of the alkene nucleophile to
the oxygen atom of the electrophile is antiperiplanar (eq
22), or B, in which it is synclinal (eq 23).53 In both
structures, the hydrogen atom is directed toward the
inside of the ring, which is the most sterically demanding
position. The configuration of the silyl enol ether 10
Con clu sion
The additions of nucleophiles to oxocarbenium ions
derived from oxasilacyclopentane acetates occurred with
high diastereoselectivity in most cases. These stereose-
lectivities can be understood by application of our inside
attack model9 for five-membered ring oxocarbenium ions.
These results demonstrate that small nucleophiles add
to the inside of the lower energy ground-state conformer
of the oxocarbenium ion. In contrast, sterically demand-
ing nucleophiles add to the inside of the envelope
conformer where approach is anti to the C-2 substituent
of the oxocarbenium ion, regardless of the ground-state
conformer population. The products generated from these
substitution reactions provide access to 1,3-diols with
high diastereoselectivity. Future studies will investigate
the use of this methodology for the synthesis of polyoxy-
genated structures such as those embedded in natural
products.
(50) Other systems show varying degrees of dependence upon these
two variables. For example, see: (a) Danishefsky, S. J .; DeNinno, S.;
Lartey, P. J . Am. Chem. Soc. 1987, 109, 2082-2089. (b) Murata, S.;
Suzuki, M.; Noyori, R. Tetrahedron 1988, 44, 4259-4270. (c) Panek,
J . S.; Schaus, J . V. Tetrahedron 1997, 53, 10971-10982. (d) Rychnovsky,
S. D.; Sinz, C. J . Tetrahedron Lett. 1998, 39, 6811-6814. (e) Comins,
D. L.; Kuethe, J . T.; Hong, H.; Lakner, F. J .; Concolino, T. E.;
Rheingold, A. L. J . Am. Chem. Soc. 1999, 121, 2651-2652. (f)
Matsumura, Y.; Kanda, Y.; Shirai, K.; Onomura, O.; Maki, T. Org. Lett.
1999, 1, 175-178. (g) Pilli, R. A.; Riatto, V. B.; Vencato, I. Org. Lett.
2000, 2, 53-56.
Exp er im en ta l Section
Gen er a l P r oced u r es. General experimental details are
provided as Supporting Information. Microanalyses were
performed by Atlantic Microlab, Atlanta, GA. All reactions
were carried out under a stream of nitrogen in glassware that
had been flame-dried. Solvents were dried and distilled prior
to use.
1-Oxa -3,4-d im et h yl-5-(3-(2-p h en yl-1-p r op en yl))-2-d i-
(ter t-bu tyl)sila cyclop en ta n e (7a ). To a stirred solution of
acetate trans-5 (0.070 g, 0.25 mmol) in 3 mL of CH2Cl2 at -78
(51) Bernardi, A.; Cardani, S.; Carugo, O.; Columbo, L.; Scolastico,
C.; Villa, R. Tetrahedron Lett. 1990, 31, 2779-2782.
(52) Pasquarello, A.; Poli, G.; Scolastico, C. Tetrahedron: Asymmetry
1990, 1, 429-432.
(53) Intramolecular additions of silyl enol ethers to aldehydes in the
presence of Lewis acids exhibit only a modest preference for anti-
periplanar transition states: Denmark, S. E.; Lee, W. J . Org. Chem.
1994, 59, 707-709.
(54) Seebach, D.; Golinski, J . Helv. Chim. Acta 1981, 64, 1413-1423.
(55) Bassindale, A. R.; Ellis, R. J .; Lau, J . C.-Y.; Taylor, P. G. J .
Chem. Soc., Chem. Commun. 1986, 98-100.

2062 J . Org. Chem., Vol. 67, No. 7, 2002
Bear et al.
°C was added SnBr4 (0.246 mL, 1 M, CH2Cl2) followed by
2-phenyl-3-(trimethylsilyl)-1-propene14 (0.104 mL, 0.492 mmol,
density of 0.9 assumed). The reaction was stirred for 0.5 h and
then quenched by the addition of MeOH/triethylamine (0.5 mL,
2:1). The mixture was diluted in 10 mL of CH2Cl2 and poured
into 40 mL of saturated aqueous NaHCO3. The organic layer
was recovered, and the aqueous layer was extracted with 2 ×
10 mL of CH2Cl2. The combined organic layers were filtered
through a cotton plug and reduced in vacuo. Purification by
flash chromatography (10:90 to 30:70 CH2Cl2/hexanes) yielded
the product as a clear oil (0.079 g, 93%). Analysis by GC
indicated an 89:11 mixture of diastereomers: 1H NMR (CDCl3,
500 MHz) δ 7.40 (m, 2H), 7.32 (m, 2H), 7.25 (m, 1H), 5.28 (d,
J ) 0.8, 1H), 5.22 (d, J ) 1.5, 1H), 4.28 (m, 1H), 2.66 (dd, J )
4.3, 15.1, 1H), 2.43 (dd, J ) 9.0, 15.2, 1H), 2.15 (m, 1H), 1.17
(d, J ) 7.3, 3H), 1.05 (s, 9H), 1.01 (s, 9H), 0.97 (d, J ) 6.8, 3H,
overlapped with m, 1H); 13C NMR (CDCl3, 125 MHz) δ 146.7,
142.4, 128.1, 126.5, 114.3, 78.24, 78.17, 43.3, 38.4, 28.8, 28.4,
22.3, 21.9, 20.1, 15.3, 12.8; IR (neat) 3081, 1731, 1630, 1473,
tion by flash chromatography (15:85 to 20:80 CH2Cl2/hexanes)
yielded the product as a clear oil (0.279 g, 93%). Analysis by
GC indicated a 99:1 mixture of diastereomers: 1H NMR
(CDCl3, 500 MHz) δ 5.83 (m, 1H), 5.04 (m, 2H), 4.16 (m, 1H),
2.36 (m, 1H), 2.14 (m, 1H), 1.87 (m, 1H), 1.75 (m, 1H), 1.42
(m, 1H), 1.24 (d, J ) 7.6, 3H), 1.06 (s, 9H), 1.04 (s, 9H
overlapped with m, 1H); 13C NMR (CDCl3, 125 MHz) δ 135.8,
116.5, 76.5, 42.5, 39.5, 28.4, 28.2, 22.0, 20.0, 15.5, 15.0; IR
(neat) 3076, 1642, 1474, 1388, 1023 cm-1; HRMS (CI/isobu-
tane) m/ z calcd for C12H25OSi (M - C3H5)+ 213.1675, found
213.1674. Anal. Calcd for C15H30OSi: C, 71.57; H, 12.01.
Found: C, 71.49; H, 12.02.
1-Oxa -3,4-d im et h yl-5-(3-(2-m et h yl-1-p r op en yl))-2-d i-
(ter t-bu tyl)sila cyclop en ta n e (14a ). To a stirred solution of
acetate trans-5 (0.302 g, 1.06 mmol) in 3 mL of CH2Cl2 at -78
°C was added SnBr4 (1.06 mL, 1 M, CH2Cl2) followed by
methallyltrimethylsilane30 (0.380 mL, 2.12 mmol). The reaction
was stirred for 0.5 h and then quenched by the addition of
MeOH/triethylamine (0.5 mL, 2:1). Then the mixture was
diluted in 10 mL of CH2Cl2 and poured into 40 mL of saturated
aqueous NaHCO3. The organic layer was recovered, and the
aqueous layer was extracted with 2 × 10 mL of CH2Cl2. The
combined organic layers were filtered through a cotton plug
and reduced in vacuo. Purification by flash chromatography
(15:85 to 20:80 CH2Cl2/hexanes) yielded the product as a clear
oil (0.279 g, 93%). Analysis by GC indicated a 94:6 mixture of
diastereomers: 1H NMR (CDCl3, 500 MHz) δ 4.79 (s, 1H), 4.75
(s, 1H), 4.33 (m, 1H), 2.18 (m, 1H), 2.10 (m, 1H), 2.01 (m, 1H),
1.80 (s, 3H), 1.17 (d, J ) 7.2, 3H), 1.04 (s, 18H), 0.93 (d, J )
6.8, 3H, overlapped with m, 1H); 13C NMR (CDCl3, 125 MHz)
δ 144.1, 111.8, 78.5, 43.1, 40.4, 28.7, 28.4, 22.6, 22.1, 21.9, 20.0,
1364, 1075 cm-1; HRMS (CI/isobutane) m/ z calcd for C18H27
-
OSi (M - t-Bu)+ 287.1831, found 287.1831. Anal. Calcd for
C
22H36OSi: C, 76.68; H, 10.53. Found: C, 76.76; H, 10.60.
1-Oxa -3,4-d im eth yl-5-(3-(1-bu ten yl))-2-d i(ter t-bu tyl)si-
la cyclop en ta n e (9). To a stirred solution of acetate trans-5
(1.654 g, 5.773 mmol) in 60 mL of CH2Cl2 at -78 °C was added
SnBr4 (5.8 mL, 1 M, CH2Cl2) followed by (Z)-crotyltrimethyl-
silane20 (1.50 mL, 11.6 mmol, >99:1 Z/E). The reaction mixture
was stirred at -78 °C for 0.5 h and then warmed to room
temperature over 0.5 h. The reaction mixture was poured into
40 mL of saturated aqueous NaHCO3 and diluted in 10 mL of
CH2Cl2. The organic layer was saved and the aqueous layer
extracted with 2 × 20 mL of CH2Cl2. The combined organic
layers were washed once with brine, filtered through a pad of
Na2SO4, and reduced in vacuo. Purification by flash chroma-
tography (2:98 to 4:96 to 10:90 to 20:80 to 30:70 CH2Cl2/
hexanes) yielded the product as a clear oil (1.49 g, 91%) in a
75:11:14 mixture of 9a /9b/9c by GC. The isomers were
sufficiently resolved by flash chromatography to allow indi-
vidual characterization.
15.1, 12.9; IR (neat) 3075, 1651, 1474, 1363, 1078, 1030 cm-1
;
HRMS (CI/isobutane) m/ z calcd for C13H27OSi (M - t-Bu)+
227.1831, found 227.1834. Anal. Calcd for C17H34OSi: C, 72.27;
H, 12.13. Found: C, 72.37; H, 12.14.
1-Oxa -3,4-d im eth yl-5-(3-(2-oxop r op yl))-2-d i(ter t-bu tyl)-
sila cyclop en t a n e (14b ). To a stirred solution of acetate
trans-5 (0.069 g, 0.24 mmol) in 3 mL of CH2Cl2 at -78 °C was
added SnCl4 (0.240 mL, 1 M, CH2Cl2) followed by 2-(trimeth-
ylsiloxy)propene16 (0.052 mL, 0.32 mmol). The reaction was
stirred for 0.5 h and then quenched by the addition of MeOH/
triethylamine (0.5 mL, 2:1). The mixture was diluted in 10
mL of CH2Cl2 and poured into 40 mL of saturated aqueous
NaHCO3. The organic layer was recovered, and the aqueous
layer was extracted with 2 × 10 mL of CH2Cl2. The combined
organic layers were filtered through a cotton plug and reduced
in vacuo. Purification by flash chromatography (3:97 to 5:95
EtOAc/hexanes) yielded the product as a clear oil (0.031 g,
60%). Analysis by GC indicated a 94:6 mixture of diastereo-
mers: 1H NMR (CDCl3, 500 MHz) δ 4.70 (td, J ) 5.7, 8.1, 1H),
2.43 (d, J ) 4.8, 1H), 2.41 (d, J ) 2.0, 1H), 2.22 (s, 3H), 1.17
(d, J ) 7.5, 3H), 1.038 (s, 9H), 1.035 (s, 9H), 0.89 (d, J ) 6.8,
3H, overlapped with m, 1H); 13C NMR (CDCl3, 125 MHz) δ
208.2, 77.3 (overlapped with CDCl3), 47.5, 42.6, 30.0, 28.6, 28.3,
22.1, 21.9, 20.0, 14.8, 12.7; 13C NMR (CD2Cl2, 125 MHz) 208.2,
77.7, 47.4, 42.9, 30.3, 28.8, 28.4, 22.4, 22.1, 20.2, 14.9, 12.8;
IR (neat) 2934, 2892, 1716, 1474, 1364, 1079 cm-1; HRMS (CI/
isobutane) m/ z calcd for C12H23O2Si (M - t-Bu)+ 227.1467,
found 227.1464. Anal. Calcd for C16H32O2Si: C, 67.55; H, 11.25.
Found: C, 67.45; H, 11.30.
1-Oxa -3,4-d im eth yl-5-(2-p h en yl-2-oxoeth yl)-2,2-d i(ter t-
bu tyl)sila cyclop en ta n e (16). To a cooled (-78 °C) solution
of cis-5 (1.00 g, 3.49 mmol) in 35 mL of CH2Cl2 was added
SnBr4 (3.80 mL, 0.93 M, CH2Cl2) followed by 1-phenyl-1-
((trimethylsilyl)oxy)ethylene10 (0.671 g, 3.49 mmol). After 10
min at -78 °C, the reaction mixture was poured into 200 mL
of saturated aqueous NaHCO3 and diluted in 200 mL of CH2-
Cl2. The aqueous layer was extracted with 2 × 50 mL of CH2-
Cl2. The combined organic layers were washed with brine,
dried over Na2SO4, and concentrated in vacuo. Analysis of the
unpurified product by GC-MS indicated a pair of diastereo-
mers in a ratio of 90:10. Purification by flash chromatography
(1:99 to 2:98 EtOAc/hexanes) yielded the product (as an
inseparable mixture of diastereomers) as a colorless oil (0.94
g, 79%): 1H NMR (500 MHz, CDCl3) δ 8.00 (d, J ) 7.4, 2H),
Da ta for 9a : 1H NMR (CDCl3, 500 MHz) δ 6.02 (m, 1H)
5.04 (m, 1H), 4.94 (m, 1H), 3.29 (dd, J ) 7.8, 10.0, 1H), 2.33
(m, 1H), 1.51 (m, 1H), 1.19 (d, J ) 7.5, 3H), 1.04 (s, 9H), 1.01
(d, J ) 7.0, 3H), 1.00 (s, 9H), 0.93 (d, J ) 6.1, 3H), 0.85 (m,
1H); 13C NMR (CDCl3, 125 MHz) δ 143.9, 112.6, 86.6, 43.2,
41.0, 27.9, 25.8, 21.3, 15.9, 13.8, 12.8; IR (neat) 3074, 1640,
1473, 1386, 1097, 1000 cm-1; HRMS (CI/isobutane) m/ z calcd
for C17H34OSi (M + H)+ 283.2457, found 283.2457. Anal. Calcd
for C17H34OOSi: C, 72.27; H, 12.13. Found: C, 72.35; H, 12.10.
Da ta for 9b: 1H NMR (CDCl3, 500 MHz) δ 5.88 (m, 1H)
4.99 (m, 1H), 4.96 (m, 1H), 3.19 (dd, J ) 2.0, 10.0, 1H), 2.34
(m, 1H); 1.47 (m, 1H), 1.16 (d, J ) 6.8, 3H), 1.15 (d, J ) 6.7,
3H), 1.02 (s, 9H), 1.00 (s, 9H), 0.89 (d, J ) 6.1, 3H), 0.85 (m,
1H); 13C NMR (CDCl3, 125 MHz) δ 140.6, 114.4, 87.4, 43.2,
41.2, 27.9, 25.5, 21.3, 18.7, 15.3, 12.6; IR (neat) 3072,1642,
1473, 1387, 1096, cm-1; HRMS (EI) m/ z calcd for C17H34OSi
(M)+ 282.2379, found 282.2386.
Da ta for 9c: 1H NMR (CDCl3, 500 MHz) δ 5.79 (m, 1H)
4.96 (m, 1H), 4.90 (m, 1H), 3.91 (dd, J ) 7.4, 8.9, 1H), 2.26
(m, 1H), 2.14 (m, 1H), 1.16 (d, J ) 7.6, 3H), 1.09 (d, J ) 6.4,
3H), 1.06 (s, 18H), 0.97 (d, J ) 6.8, 3H overlapped with m,
1H); 13C NMR (CDCl3, 125 MHz) δ 143.9, 113.0, 83.5, 44.2,
41.2, 28.8, 28.7, 23.0, 22.2, 20.3, 19.7, 16.6, 13.6; IR (neat) 3076,
1639, 1473, 1386, 1085, 907 cm-1. Anal. Calcd for C17H34OSi:
C, 72.27; H, 12.13. Found: C, 72.31; H, 12.10.
1-Oxa -3-m et h yl-5-(3-p r op en yl)-2-d i(ter t-b u t yl)sila cy-
clop en ta n e (12). To a stirred solution of acetate trans-5 (0.260
g, 0.904 mmol) in 3 mL of CH2Cl2 at -78 °C was added SnBr4
(0.913 mL, 1 M, CH2Cl2) followed by methallyltrimethylsilane30
(0.327 mL, 1.83 mmol). The reaction was stirred for 0.5 h and
then quenched by the addition of MeOH/triethylamine (0.5 mL,
2:1). Then the mixture was diluted in 10 mL of CH2Cl2 and
poured into 40 mL of saturated aqueous NaHCO3. The organic
layer was recovered, and the aqueous layer was extracted with
2 × 10 mL of CH2Cl2. The combined organic layers were
filtered through a cotton plug and reduced in vacuo. Purifica-

Substitution Reactions of Oxasilacyclopentane Acetals
J . Org. Chem., Vol. 67, No. 7, 2002 2063
7.51 (t, J ) 7.9, 1H), 7.42 (t, J ) 7.7, 2H), 4.06 (ddd, J ) 10.3,
7.4, 2.9, 1H), 3.15 (dd, J ) 14.5, 7.9, 1H), 3.00 (dd, J ) 14.5,
3.4, 1H), 1.99 (m, 1H), 1.41 (quintet, J ) 8.2, 1H), 1.19 (d, J )
7.9, 3H), 1.02 (s, 9H), 0.96 (d, J ) 7.2, 3H), 0.95 (s, 9H); 13C
NMR (125 MHz, CDCl3) δ 200.3, 138.5, 133.0, 129.0, 128.7,
79.3, 45.5, 42.9, 29.0, 28.5, 22.4, 20.7, 20.3, 13.2, 11.4; IR (thin
film) 2932, 2858, 1716, 1473, 1364 cm-1; HRMS (CI/isobutane)
m/z calcd for C17H25O2Si (M - C4H9)+ 289.1624, found 289.1621.
Anal. Calcd for C21H34O2Si: C, 72.78; H, 9.89. Found: C, 72.55;
H, 9.87.
δ 213.8, 88.1, 49.4, 36.0, 33.2, 27.6, 27.5, 20.7, 19.9, 19.4, 16.0,
8.9, 7.9; IR (thin film) 2932, 2858, 1716, 1473, 1364 cm-1
;
HRMS (CI/isobutane) m/z calcd for C17H35O2Si (M + H)+
299.2406, found 299.2405. Anal. Calcd for C17H34O2Si: C,
68.39; H, 11.48. Found: C, 68.11; H, 11.48.
1-Oxa-4-(2-pr opyl)-5-(2-(3-oxopen tyl))-2,2-di(ter t-bu tyl)-
sila cyclop en ta n e (22). The procedure used for the prepara-
tion of 17 was employed. An experiment starting with acetate
21 (0.100 g, 0.333 mmol) provided an impure oil that by GC
analysis was a 93:3:2:2 mixture of diastereomers. Purification
by flash chromatography (1:99 to 2:98 EtOAc/hexanes) yielded
the product as a colorless oil (0.079 g, 73%): 1H NMR (500
MHz, CDCl3) δ 4.00 (dd, J ) 10.6, 2.6, 1H), 2.54 (m, 3H), 1.77
(m, 2H), 1.09 (d, J ) 6.9, 3H), 1.06 (t, J ) 7.3, 3H), 0.99 (s,
9H), 0.98 (s, 9H), 0.96 (d, J ) 6.7, 3H), 0.84 (d, J ) 6.7, 3H),
0.69 (dd, J ) 14.6, 7.4, 1H), 0.58 (dd, J ) 14.6, 12.0, 1H); 13C
NMR (125 MHz, CDCl3) δ 213.8, 79.7, 49.2, 46.3, 33.0, 27.7,
27.5, 27.2, 22.6, 21.0, 20.1, 15.3, 8.4, 7.9, 4.6; IR (thin film)
2958, 2857, 1716, 1472, 1387, 1365 cm-1; HRMS (CI/isobutane)
m/z calcd for C19H39O2Si (M + H)+ 327.2719, found 327.2712.
Anal. Calcd for C19H38O2Si: C, 69.88; H, 11.73. Found: C,
69.95; H, 11.61.
1-Oxa-3,4-dim eth yl-5-(2-(3-oxopen tyl))-2,2-di(ter t-bu tyl)-
sila cyclop en ta n e (17). To a cooled (-78 °C) solution of
acetate cis-5 (0.150 g, 0.524 mmol) in 5.7 mL of CH2Cl2 was
added SnBr4 (0.56 mL, 0.93 M, CH2Cl2) followed by (E)-3-
((trimethylsilyl)oxy)-2-pentene26 (0.493 g, 3.11 mmol). After 10
min at -78 °C, the reaction mixture was poured into 50 mL
of saturated aqueous NaHCO3 and diluted in 25 mL of CH2-
Cl2. The aqueous layer was extracted with 2 × 10 mL of CH2-
Cl2. The combined organic layers were washed with brine,
dried over Na2SO4, and concentrated in vacuo. Analysis of the
unpurified reaction mixture by GC showed a 94:5:1 mixture
of diastereomers. Purification by flash chromatography (1:99
to 2:98 EtOAc/hexanes) yielded the product as a colorless oil
(0.129 g, 79%) as a 93:5:2 mixture of diastereomers: 1H NMR
(500 MHz, CDCl3) δ 3.86 (dd, J ) 10.5, 2.8, 1H), 2.56 (q, J )
7.2, 2H), 2.48 (dq, J ) 7.0, 2.8, 1H), 1.98 (m, 1H), 1.41 (q, J )
8.1, 1H), 1.08 (d, J ) 6.9, 3H), 1.07 (d, J ) 8.2, 3H), 1.04 (t, J
1-Oxa -4-m eth yl-5-(2-p h en yl-2-oxoeth yl)-2,2-d i(ter t-bu -
tyl)sila cyclop en ta n e (23a ). The procedure used for the
preparation of 16 was employed. An experiment starting with
acetate 19 (0.116 g, 0.426 mmol) provided an impure oil that
by GC-MS analysis was a pair of diastereomers in a 59:41
ratio. Purification by flash chromatography (2:98 to 4:96
EtOAc/hexanes) yielded the major diastereomer and an en-
riched sample of the minor diastereomer, both as colorless oils
(0.072 g, 52% combined). HRMS and IR data were obtained
for the mixture of diastereomers: IR (thin film) 3061, 2963,
2859, 1747, 1598, 1471 cm-1; HRMS (CI/isobutane) m/z calcd
for C20H33O2Si (M + H)+ 333.2250, found 333.2255.
) 7.2, 3H), 1.03 (s, 9H), 0.97 (s, 9H), 0.87 (d, J ) 6.9, 3H); 13
C
NMR (125 MHz, CDCl3) δ 214.0, 81.0, 49.2, 38.5, 33.1, 28.5,
28.0, 22.1, 20.5, 19.8, 12.3, 10.8, 8.3, 7.8; IR (thin film) 2936,
2850, 1716, 1472, 1388, 980 cm-1; HRMS (CI/isobutane) m/z
calcd for C18H37O2Si (M + H)+ 313.2563, found 313.2562. Anal.
Calcd for C18H36O2Si: C, 69.17; H, 11.61. Found C, 69.26; H,
11.54.
1
1-Oxa -3,4-d im eth yl-5-(2-p r op en yl)-2-d i(ter t-bu tyl)sila -
cyclop en ta n e (18). The procedure used for the preparation
of 14a was employed. An experiment starting with acetate
cis-5 (0.200 g, 0.696 mmol) provided an impure oil that GC
analysis showed was a pair of diastereomers in a 57:43 ratio
as well as 5% of the trans isomer (12), which presumably arose
from small quantities of acetate trans-5 in acetate cis-5.
Purification by flash chromatography (0:100 to 1:99 EtOAc/
hexanes) yielded the product 18 as a colorless oil (0.161 g, 87%
combined) as an inseparable mixture of diastereomers. The
resonances for each isomer were identifiable in the NMR
spectra. Combustion analysis, HRMS, and IR data were
obtained for the mixture of diastereomers: IR (thin film):
3075, 2857, 1642, 1473, 1388, 1364 cm-1; HRMS (CI/isobutane)
m/z calcd for C16H33OSi (M + H)+ 269.2300, found 269.2295.
Anal. Calcd for C16H32OSi: C, 71.57; H, 12.01. Found: C, 71.28;
H, 12.03.
NMR Da ta for th e Ma jor Isom er of 23a : H NMR (500
MHz, CDCl3) δ 8.00-8.02 (m, 2H), 7.52-7.56 (m, 1H), 7.41-
7.46 (m, 2H), 3.93 (ddd, J ) 10.5, 8.0, 3.2, 1H), 3.18 (dd, J )
14.5, 8.0, 1H), 3.00 (dd, J ) 14.5, 3.2, 1H), 1.83 (m, 1H), 1.09
(d, J ) 6.5, 3H), 1.05 (dd, J ) 15.0, 7.6, 1H), 0.95 (s, 18H),
0.50 (dd, J ) 15.0, 11.8, 1H); 13C NMR (125 MHz, CDCl3) δ
199.9, 138.0, 132.7, 128.6, 128.3, 81.7, 45.1, 39.8, 29.7, 27.6,
27.5, 20.5, 19.7, 15.9.
1
NMR Da ta for th e Min or Isom er of 23a : H NMR (500
MHz, CDCl3) δ 7.94-7.96 (m, 2H), 7.53-7.56 (m 1H), 7.44-
7.47 (m, 2H), 4.80 (q, J ) 6.7, 1H), 3.14 (dd, J ) 16.2, 6.8,
1H), 3.04 (dd, J ) 16.2, 6.4, 1H), 2.56 (m, 1H), 1.04 (s, 9H and
m, 1H), 1.00 (s, 9H), 0.98 (d, J ) 7.0, 3H), 0.54 (dd, J ) 15.1,
9.0, 1H); 13C NMR (125 MHz, CDCl3) δ 199.0, 137.5, 132.8,
128.5, 128.1, 77.5, 42.4, 35.0, 29.7, 28.5, 28.0, 20.9, 18.5, 13.8.
1-Oxa -4-(2-p r op yl)-5-(2-p h en yl-2-oxoeth yl)-2,2-d i(ter t-
bu tyl)sila cyclop en ta n e (23b). The procedure used for the
preparation of 16 was employed. An experiment starting with
acetate 21 (0.200 g, 0.666 mmol) provided an impure oil that
by GC analysis was the product as a pair of diastereomers in
a ratio of 85:15. Purification by flash chromatography (1:99
to 2:98 EtOAc/hexanes) yielded the major diastereomer and a
sample enriched with the minor diastereomer, both as colorless
oils (0.177 g, 70% combined). Combustion analysis, HRMS, and
IR data were obtained for the mixture of diastereomers: IR
(thin film): 3060, 2958, 2857, 1690, 1598, 1471 cm-1; HRMS
(CI/isobutane) m/z calcd for C22H37O2Si (M + H)+ 361.2562,
found 361.2555. Anal. Calcd for C22H36O2Si: C, 73.28; H, 10.06.
Found: C, 73.11; H, 10.22.
NMR Da ta for th e Ma jor Isom er of 18: 1H NMR (500
MHz, CDCl3) δ 6.01 (m, 1H), 5.08 (m, 2H), 3.56 (ddd, J ) 10.0,
5.9, 3.7, 1H), 2.49 (m, 1H), 2.15 (m, 1H), 1.90 (m, 1H), 1.38
(quintet, J ) 8.2, 1H), 1.09 (s, 9H), 1.07 (d, J ) 8.1, 3H), 0.99
(s, 9H), 0.87 (d, J ) 7.0, 3H); 13C NMR (CDCl3, 125 MHz) δ
135.6, 116.1, 81.0, 43.0, 39.1, 28.7, 28.1, 22.1, 20.5, 20.0, 12.7,
11.0.
NMR Da ta for th e Min or Isom er of 18: 1H NMR (500
MHz, CDCl3, not all signals were sufficiently resolved) δ 5.89
(m, 1H), 3.91 (dt, J ) 7.1, 4.8, 1H), 2.38 (m, 1H), 1.63 (m, 1H),
1.08 (s, 9H), 1.03 (s, 9H); 13C NMR (125 MHz, CDCl3) δ 136.2,
115.9, 80.1, 40.3, 38.1, 29.0, 28.5, 21.9, 20.5, 20.4, 12.3, 9.6.
1-Oxa -4-m eth yl-5-(2-(3-oxop en tyl))-2,2-d i(ter t-bu tyl)si-
la cyclop en ta n e (20). The procedure used for the preparation
of 17 was employed. An experiment starting with acetate 19
(0.214 g, 0.786 mmol) provided an impure oil that GC-MS
analysis showed was a 92:5:3 mixture of diastereomers.
Purification by flash chromatography (1:99 to 2:98 EtOAc/
hexanes) yielded the product as a colorless oil (0.23 g, 74%)
as an inseparable mixture of diastereomers in a 92:5:3 ratio:
1H NMR (500 MHz, CDCl3) δ 3.71 (dd, J ) 10.2, 2.7, 1H), 2.55
(m, 3H), 1.86 (m, 1H), 1.10 (d, J ) 7.0, 3H), 1.04 (t, J ) 7.3,
3H), 1.01 (d, J ) 6.5, 3H), 0.98 (s, 9H), 0.97 (s, 9H and m,
1H), 0.47 (dd, J ) 14.8, 11.6, 1H); 13C NMR (125 MHz, CDCl3)
1
NMR Da ta for th e Ma jor Isom er of 23b: H NMR (500
MHz, CDCl3) δ 8.00 (d, J ) 7.4, 2H), 7.53 (t, J ) 7.4, 1H), 7.43
(t, J ) 7.7, 2H), 4.19 (ddd, J ) 10.2, 8.2, 3.0, 1H), 3.18 (dd, J
) 8.0, 14.6, 1H), 3.01 (dd, J ) 14.6, 3.0, 1H), 1.83 (m, 1H),
1.74 (m, 1H), 0.97 (d, J ) 8.7, 3H), 0.96 (s, 9H), 0.95 (s, 9H),
0.85 (d, J ) 6.7, 3H), 0.70 (dd, J ) 14.8, 8.0, 1H), 0.62 (dd, J
) 14.8, 12.1, 1H); 13C NMR (125 MHz, CDCl3) δ 199.9, 138.0,
132.6, 128.5, 128.2, 77.8, 50.7, 45.7, 28.0, 27.7, 27.4, 22.6, 20.8,
19.7, 15.9, 5.0.
1
NMR Da ta for th e Min or Isom er of 23b: H NMR (500
MHz, CDCl3, not all signals were sufficiently resolved) δ 7.93
(m, 2H), 7.53-7.56 (m, 1H), 7.44-7.48 (m, 2H), 4.90 (ddd, J

2064 J . Org. Chem., Vol. 67, No. 7, 2002
Bear et al.
) 10.0, 7.1, 3.0, 1H), 3.08 (dd, J ) 15.6, 10.0, 1H), 2.90 (dd, J
) 15.6, 3.0, 1H), 2.01 (m, 1H), 1.43 (m, 1H), 1.01 (d, J ) 6.5,
1H), 0.99 (s, 9H), 0.98 (s, 9H), 0.94 (d, J ) 6.4, 3H), 0.48 (t, J
) 14.2, 1H); 13C NMR (125 MHz, CDCl3) δ 198.9, 137.7, 132.7,
128.5, 128.1, 76.2, 49.8, 41.0, 31.4, 28.5, 28.2, 22.6, 21.9, 20.7,
19.5, 9.6.
1-Oxa -4-m et h yl-5-(3-p r op en yl)-2-d i(ter t-b u t yl)sila cy-
clop en ta n e (24a ). The procedure used for the preparation of
14a was employed. An experiment starting with acetate 19
(0.349 g, 1.28 mmol) yielded 0.256 g (79%) of the product as a
colorless oil in a 94:6 ratio by GC analysis: 1H NMR (CDCl3,
500 MHz) δ 5.95 (m, 1H), 5.06 (m, 2H), 4.09 (q, J ) 7.2, 1H),
2.42 (m, 1H), 2.12 (m, 2H), 1.04 (s, 9H), 1.02 (s, 9H), 1.01 (d,
J ) 6.0, 3H), 0.93 (dd, J ) 7.4, 14.9, 1H), 0.50 (dd, J ) 11.3,
14.9, 1H); 13C NMR (CDCl3, 125 MHz) δ 136.8, 116.0, 81.1,
36.9, 35.9, 28.4, 28.1, 20.7, 19.4, 18.5, 13.5; IR (neat) 2858,
1474, 1364, 1039, 824, 628 cm-1; HRMS (CI/isobutane) m/ z
calcd for C12H25OSi (M - C3H5)+ 213.1675, found, 213.1674.
Anal. Calcd for C15H30OSi: C, 70.80; H, 11.88. Found: C, 70.50;
H, 11.78.
IR (neat) 3076, 2858, 1642, 1474, 1388, 1023, 822 cm-1; HRMS
(CI/isobutane) m/ z calcd for C12H25OSi (M - C3H5)+ 213.1675,
found 213.1674. Anal. Calcd for C15H30OSi: C, 71.57; H, 12.01.
Found: C, 71.49; H, 12.02.
1-Oxa -3-(2-p r op yl)-5-(2-p r op en yl)-2-d i(ter t-bu tyl)sila -
cyclop en ta n e (27b). The procedure used for the preparation
of 14a was employed. A reaction employing acetate 26 (0.516
g, 1.72 mmol) yielded an impure oil that GC analysis indicated
was a single diastereomer. Purification by flash chromatog-
raphy (1:99 to 2:98 EtOAc/hexanes) yielded the product as a
colorless oil (0.461 g, 95%): 1H NMR (500 MHz, CDCl3) δ 5.82
(m, 1H), 5.04 (m, 2H), 4.19 (q, J ) 7.3, 1H), 2.31 (ddd, J )
13.8, 7.2, 6.5, 1H), 2.08 (ddd, J ) 14.0, 7.2, 7.0, 1H), 1.93 (dd,
J ) 12.9, 8.4, 1H), 1.83 (m, 2H), 1.14 (m, 1H), 1.06 (s, 9H),
1.04 (s, 9H and m, 3H), 0.94 (d, J ) 6.4, 3H); 13C NMR (125
MHz, CDCl3) δ 136.0, 116.6, 75.8, 43.0, 36.3, 31.7, 29.1, 28.7,
28.3, 26.7, 23.9, 22.7, 19.7; IR (neat) 3076, 2858, 1642, 1474,
1386, 1034, 821 cm-1; HRMS (CI/isobutane) m/z calcd for
C
17H35OSi (M + H)+ 283.2457, found 283.2464. Anal. Calcd
for C17H34OSi: C, 72.27; H, 12.13. Found: C, 71.95; H, 12.00.
1-Oxa -4-(2-p r op yl)-5-(3-p r op en yl)-2-d i(ter t-bu tyl)sila -
cyclop en ta n e (24b). The procedure used for the preparation
of 14a was employed. An experiment starting with acetate 21
(0.391 g, 1.30 mmol) yielded 0.358 g (97%) of the product as a
colorless oil in an 80:20 mixture by GC analysis: 1H NMR
(CDCl3, 500 MHz) δ 6.01 (m, 1H), 5.09 (s, 1H), 5.07 (d, J )
10.2, 1H), 4.22 (m, 1H), 2.14 (m, 1H), 2.05 (m, 1H), 1.88 (m,
1H), 1.42 (m, 1H), 1.04 (s, 9H), 1.02 (s, 9H), 0.97 (d, J ) 6.4,
3H), 0.91 (d, J ) 6.4, 3H overlapped with m, 1H), 0.42 (t, J )
14.2, 1H); 13C NMR (CDCl3, 125 MHz) δ 137.1, 116.0, 80.1,
50.1, 35.7, 31.0, 28.5, 28.3, 22.6, 21.7, 20.7, 19.6, 9.3; IR (neat)
3075, 2859, 1472, 1365, 1021 cm-1; HRMS (CI/isobutane) m/ z
calcd for C14H29OSi (M - C3H5)+ 241.1988, found 241.1988.
Anal. Calcd for C17H34OSi: C, 72.27; H, 12.13. Found: C, 72.16;
H, 12.17.
Ack n ow led gm en t. This research was supported by
the National Institutes of Health (General Medical
Sciences, GM 54909). K.A.W. thanks AstraZeneca, the
Camille and Henry Dreyfus Foundation, Glaxo-Wellcome,
Merck Research Laboratories, J ohnson & J ohnson, and
the Sloan Foundation for awards to support research.
K.A.W. is a Cottrell Scholar of the Research Corpora-
tion. We thank Dr. J ohn Greaves and Dr. J ohn Mudd
for mass spectrometric data, Dr. J oe Ziller for X-ray
crystallographic analysis, and Mr. Zhi-hui Peng and Dr.
Claudia Roberson for their contributions to computa-
tional chemistry.
1-Oxa -3-m et h yl-5-(3-p r op en yl)-2-d i(ter t-b u t yl)sila cy-
clop en ta n e (27a ). The procedure used for the preparation of
14a was employed. An experiment starting with acetate 25
(0.260 g, 0.955 mmol) yielded 0.228 g (94%) of the product as
a 99:1 mixture of diastereomers as indicated by GC analysis:
1H NMR (CDCl3, 500 MHz) δ 5.83 (m, 1H), 5.04 (m, 2H), 4.16
(m, 1H), 2.36 (m, 1H), 2.14 (m, 1H), 1.87 (m, 1H), 1.75 (m,
1H), 1.42 (m, 1H), 1.24 (d, J ) 7.6, 3H), 1.06 (s, 9H), 1.04 (s,
9H overlapped with m, 1H); 13C NMR (CDCl3, 125 MHz) δ
135.8, 116.5, 76.5, 42.5, 39.5, 28.4, 28.2, 22.0, 20.0, 15.5, 15.0;
Su p p or tin g In for m a tion Ava ila ble: General experimen-
tal details, description of the preparation of oxasilacyclopen-
tane acetates trans-5, cis-5, 19, 21, 25, and 26, stereochemical
proofs for compounds 7a ,b, 9a -c, 11a -c, 12, 14a ,b, 16, 17a ,
20a , 22a , 23b, 24a ,b, and 27a ,b, X-ray crystal structure data
for trans-5, S8, and S16, and NMR spectra and GC traces for
compounds 16, 17a , 20a , 22a , and 27b. This material is
available free of charge via the Internet at http://pubs.acs.org.
J O010823M