Simultaneous discrimination of diastereotopic groups and faces: The first example in intramolecular [3 + 2] and [2 + 2 + 1] cycloaddition reactions

Article abstract of DOI:10.1021/jo0017638

To explore a novel concept for controlling diastereoselectivity, systematic studies on the sense and degree of diastereotopic groups and face selections in intramolecular [3 + 2] (nitrile oxide and nitrone) and [2 + 2 + 1] (Pauson-Khand) cycloadditions have been conducted. Optically pure methyl (S)-3,4-O-isopropylidene-3,4-dihydroxybutanoate (5) and methyl (S)-2,3-O-isopropylidene-2,3-dihydroxypropanoate (6) were converted to substrate aldehydes (1-4) that bear geminal allyl groups and four types of controllers with the intention of imparting a stereochemical bias to the allylic groups and their faces. The controllers involve 1,2-bis(tert-butyldimethylsiloxy), 1,3-bis(tertbutyldimethylsiloxy), 1,2-acetonide, and 1,3-acetonide groups, which are referred to as 1,2-(TBDMSO)₂, 1,3-(TBDMSO)₂, 1,3-dioxolane, and 1,3-dioxane, respectively. Twelve runs of cycloaddition reactions as combinations between the three types of reactions and the four types of substrates were performed to provide bicyclo[4.3.0] or -[3.3.0] adducts of synthetic importance in which isoxazolidine, isoxazoline, or cyclopentenone segments were fused. For every case, high levels of diastereoselectivity have been achieved: >99% (in eight cases), 82%, and 76% for the discrimination of diastereotopic groups and 68→99% for the discrimination of diastereotopic faces. On the basis of the absolute structures of the cycloadducts, plausible stereochemical models are proposed.

Full text of DOI:10.1021/jo0017638

3834
J . Org. Chem. 2001, 66, 3834-3847
Sim u lta n eou s Discr im in a tion of Dia ster eotop ic Gr ou p s a n d F a ces:
Th e F ir st Exa m p le in In tr a m olecu la r [3 + 2] a n d [2 + 2 + 1]
Cycloa d d ition Rea ction s
Teruhiko Ishikawa,* Kazuo Shimizu, Hirokazu Ishii, Shushiro Ikeda, and Seiki Saito*
Department of Bioscience and Biotechnology, Faculty of Engineering, Okayama University,
Tsushima, Okayama, J apan
seisaito@biotech.okayama-u.ac.jp
Received December 18, 2000
To explore a novel concept for controlling diastereoselectivity, systematic studies on the sense and
degree of diastereotopic groups and face selections in intramolecular [3 + 2] (nitrile oxide and
nitrone) and [2 + 2 + 1] (Pauson-Khand) cycloadditions have been conducted. Optically pure methyl
(S)-3,4-O-isopropylidene-3,4-dihydroxybutanoate (5) and methyl (S)-2,3-O-isopropylidene-2,3-di-
hydroxypropanoate (6) were converted to substrate aldehydes (1-4) that bear geminal allyl groups
and four types of controllers with the intention of imparting a stereochemical bias to the allylic
groups and their faces. The controllers involve 1,2-bis(tert-butyldimethylsiloxy), 1,3-bis(tert-
butyldimethylsiloxy), 1,2-acetonide, and 1,3-acetonide groups, which are referred to as 1,2-
(TBDMSO)2, 1,3-(TBDMSO)2, 1,3-dioxolane, and 1,3-dioxane, respectively. Twelve runs of cyclo-
addition reactions as combinations between the three types of reactions and the four types of
substrates were performed to provide bicyclo[4.3.0] or -[3.3.0] adducts of synthetic importance in
which isoxazolidine, isoxazoline, or cyclopentenone segments were fused. For every case, high levels
of diastereoselectivity have been achieved: >99% (in eight cases), 82%, and 76% for the
discrimination of diastereotopic groups and 68f99% for the discrimination of diastereotopic faces.
On the basis of the absolute structures of the cycloadducts, plausible stereochemical models are
proposed.
In tr od u ction
with a high level of stereocontrol.1 With regard to
diastereotopic group selection, spiro lactonization or
acetalization,2 iodolactonization,3 radical cyclization,4
intramolecular Michael addition,5 and intramolecular bis-
silylation6 have been reported as notable examples. In
general, however, the stereochemical impact of preexist-
ing stereogenic centers on these diastereoselective pro-
cesses has been less understood and developed. Thus,
those reports and such situations strongly intrigued us
to apply the strategy of diastereotopic group selection to
other synthetic reactions capable of inducing multiple
stereogenic centers in one step relying on a preexisting
single stereogenic center. At the same time, we had a
strong desire to put forward a stereochemical model to
account for stereochemical results as well by which we
can introduce the novel concept of stereocontrol in this
field. It was expected that if not only diastereotopic allyl
groups but also their π-(CdC) faces are discriminated
then cycloaddition reactions such as [3 + 2] (nitrone or
nitrile oxide cycloaddition) or [2 + 2 + 1] (Pauson-Khand
Synthetic reactions that involve the selective trans-
formations of groups in mirror-image positions (enan-
tiotopic groups) or in stereochemically distinct positions
not related in mirror-image position (diastereotopic groups)
are a powerful tool to create new stereocenters. A number
of applications in this context have been demonstrated
(1) For a review of the group selective reactions, see: Waldmann,
H., Ed. Organic Synthesis Highlights II; VCH: Weinheim, 1995; p 203-
222. For recent applications, see: (a) Partridge, J . J .; Chadha, N. K.;
Uskokovic, M. R. J . Am. Chem. Soc. 1973, 95, 7171-7172. (b) Hoye,
T. R.; Peck, D. R.; Swanson, T. A. J . Am. Chem. Soc. 1984, 106, 2738-
2739. (c) Hatakeyama, S.; Sakurai, K.; Takano, S. J . Chem. Soc., Chem.
Commun. 1985, 1759-1761. (d) Fuji, K.; Node, M.; Terada, S.; Murata,
M.; Nagasawa, H. J . Am. Chem. Soc. 1985, 107. 6404-6406. (e) Hafele,
B.; Schroter, D.; J ager, V. Angew. Chem., Int. Ed. Engl. 1986, 25, 87-
89. (f) Schreiber, S. L.; Schreiber, T. S.; Smith, D, B. J . Am. Chem.
Soc. 1987, 109, 1525-1529. (g) Schreiber, S. L.; Goulet, M. T.; Schulte,
G. J . Am. Chem. Soc. 1987, 109, 4718-4720. (h) Aube, J .; Burgett, P.
M.; Wang, Y. Tetrahedron Lett. 1988, 29, 151-154. (i) Kobayashi, Y.;
Kato, N.; Shimazaki, T.; Sato, F. Tetrahedron Lett. 1988, 29, 6301-
6304. (j) Tamao, K.; Tohma, T.; Inui, N.; Nakayama, O.; Ito, Y.
Tetrahedron Lett. 1990, 31, 7337-7340. (k) Kagechika, K.; Shibasaki,
M. J . Org. Chem. 1991, 56, 4093-4094. (l) Wipf, P.; Kim, Y. Tetrahe-
dron Lett. 1992, 33, 5477-5480. (m) Wang, Z.; Deschenes, D. J . Am.
Chem. Soc. 1992, 114, 1090-1091. (n) Mikami, K.; Narisawa, S.;
Shimizu, M. Terada, M. J . Am. Chem. Soc. 1992, 114, 6566-6568. (o)
Kondo, K.; Sodeoka, M.; Mori, M.; Shibasaki, M. Tetrahedron Lett.
1993, 34, 4219-4222. (p) Inoue, T.; Kitagawa, O.; Saito, A.; Taguchi,
T. J . Org. Chem. 1997, 62, 7384-7389.
(3) Brown, E, G.; Kurth, M. J . J . Am. Chem. Soc. 1987, 109, 6844-
6845.
(4) Curran, D. P.; Geib, S. J .; Lin, C.-H. Tetrahedron: Asymmetry
1994, 5, 199-202
(5) Fujioka, H.; Kitagaki, S.; Ohno, N.; Kitagawa, H.; Kita, Y.
Tetrahedron: Asymmetry 1994, 5, 333-336.
(2) (a) Harada, T.; Hayashiya, T.; Wada, I.; Iwa-ake, N.; Oku, A. J .
Am. Chem. Soc. 1987, 109, 527-532. (b) Harada, T.; Wada, I.; Oku,
A. Tetrahedron Lett. 1987, 28, 4182-4184. (c) Harada, T.; Wada, I.;
Uchimura, J .; Inoue, A.; Tanaka, S.; Oku, A. Tetrahedron Lett. 1991,
32, 1219-1222. (d) Prasad, K.; Underwood, R. L.; Repic, O. J . Org.
Chem. 1996, 61, 384-383. (e) Schreiber, S. L.; Wang, Z. J . Am. Chem.
Soc. 1985, 107. 5303-5304. (f) Schreiber, S. L.; Schreiber, T. S.; Wang,
Z. Tetrahedron Lett. 1988, 29, 4085-4088.
(6) Suginome, M.; Yamamoto, Y.; Fujii, K.; Ito, Y. J . Am. Chem. Soc.
1995, 117, 9608-9609.
(7) (a) Khand, I. U.; Knox, G. R.; Pauson, P. L.; Watts, W. E. J .
Chem. Soc., Chem. Commun. 1971, 36. (b) Pauson, P. L. Tetrahedron
1985, 41, 5855-5860. (c) Mukai, C.; Kim, J . S.; Sonobe, H.; Hanaoka,
M. J . Org. Chem. 1999, 64, 6822-6831. (d) Mukai, C.; Uchiyama, M.;
Sakamoto, S.; Hanaoka, M. Tetrahedron Lett. 1995, 36, 5761-5764.
10.1021/jo0017638 CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/05/2001

Discrimination of Diastereotopic Groups and Faces
J . Org. Chem., Vol. 66, No. 11, 2001 3835
Ch a r t 1
Sch em e 2a
a
For n ) 1 series: (a) allyl-MgBr/Et2O/-78 °C, 50 min; (b) (1)
2 N HCl/EtOH/0 °C to rt, 2 h, (2) PcCl/Py/CH2Cl2/-78 °C, 4 h; (c)
(1) TBSOTf/Et3N/CH2Cl2/0 °C, 2 h, (2) DIBAH/toluene/-78 °C, 1
h; (d) (1) (MeO)2CMe2/THF/p-TsOH/rt, 36 h, (2) c-2; (e) Swern
oxidation.
Sch em e 1
nitrones, nitrile oxides, and dienynes to be subjected to
above-mentioned intramolecular cycloaddition reactions.
One of the allyl groups can serve as a reaction partner
in these cycloaddition processes to furnish isoxazolidines,
isoxazolines, or cyclopentenones in a bicyclo[4.3.0] or
-[3.3.0] framework. In these transformations, the 1,3- or
1,2-bis(tert-butyldimethylsiloxy)9 [1,3-(TBDMSO)2 or 1,2-
(TBDMSO)2 controller] and 1,2- or 1,3-O-isopropylidene
groups [1,3-dioxolane or 1,3-dioxane controller] in the
substrates were expected to play a pivotal role of stereo-
control by biasing the diastereotopic allylic groups and
hopefully their diastereofaces as well.
Resu lts a n d Discu ssion
Syn th esis of P r ecu r sor s for Cycloa d d ition P r o-
cesses. In Scheme 2 are outlined the preparations of
aldehydes 1 and 3 bearing geminal allyl groups. Methyl
(3S)-3,4-O-isopropylidene-3,4-dihydroxybutanoate (5) pre-
pared from dimethyl (S)-malate8 was reacted with an
excess amount (2.5 equiv) of allylmagnesium bromide to
give a diallyl alcohol (7: 94%). The acetonide function of
this alcohol was deprotected under acidic conditions
followed by acylation of the thus-generated primary
hydroxy group with pivaloyl chloride in pyridine to
furnish a common intermediate (8: 78%) for the prepara-
tion of 1 and 3. The subsequent introduction of a tert-
butyldimethylsilyl group or an isopropylidene group to
both secondary and tertiary hydroxy groups followed by
DIBALH reduction led to 9 (78%) or 10 (67%), respec-
tively. The final Swern oxidation of the thus-generated
primary hydroxy groups uneventfully gave 1 (90%) with
reaction)7 process should afford functionalized carbocycles
with multiple stereogenic centers and functions.
Our plan is outlined in Chart 1 and Scheme 1, in which
the four chiral aldehydes bearing geminal allyl groups
(1-4), themselves accessible through a series of routine
reactions from two common chiral hydroxy esters 58 and
6, were converted to the corresponding four sets of
(9) (a) Saito, S.; Hirohara, Y.; Narahara, O.; Moriwake, T. J . Am.
Chem. Soc. 1989, 111, 4533-4535. (b) Saito, S.; Narahara, O.;
Ishikawa, T.; Asahara, M.; Moriwake, T.; Gawronski, J .; Kazmierczak,
F. J . Org. Chem. 1993, 58, 6292-6302. (c) Ishikawa, T.; Tajima, Y.;
Fukui, M.; Saito, S. Angew. Chem., Int. Ed. Engl. 1996, 35, 1863-
1864.
(8) Saito, S.; Ishikawa, T.; Kuroda, A.; Koga, K.; Moriwake, T.
Tetrahedron 1992, 48, 4067-4086. See also: Saito, S.; Hasegawa, T.;
Inaba, M.; Nishida, R.; Fujii, T.; Nomizu, S.; Moriwake, T. Chem. Lett.
1984, 1389-1392.

3836 J . Org. Chem., Vol. 66, No. 11, 2001
Ishikawa et al.
Ta ble 1. Resu lts of In tr a m olecu la r Cycloa d d ition Rea ction s of Nitr ile Oxid es Der ived fr om Ald eh yd es 1-4
a
Purified by silica gel column chromatography: crude oximes were prepared from 1-4 with hydroxylamine hydrochloride in the presence
b
of Et3N (0 °C to room temperature/12-20 h). Nitrile oxides were in situ generated by rigorously stirring a two-phase mixture of a
solution of the oximes in dichloromethane and an aqueous solution of sodium hypochlorite (5%). c Yields after purification by silica gel
column chromatography; dr ) diastereomeric ratios calculated using the weight of isolated products; dr (group) determined based on the
absolute configurations of the products (Si ) tert-butyldimethylsilyl).
Ch a r t 2
the 1,3-bis(tert-butyldimethylsilyloxy) [1,3-(TBDMSO)2]
controller or 3 (94%) with the 1,3-acetonide (1,3-dioxane)
controller.
When commercially available methyl (R)-2,2-dimethyl-
1,3-dioxolane-4-carboxylate (6) was subjected to the same
sequence of reactions as that for 5, the desired diallyl
aldehydes 2 and 4 with the vicinal TBDMSO [1,2-
(TBDMSO)2] controller and 1,2-acetonide (1,3-dioxolane)
controller, respectively, were furnished in good yield: the
yields of intermediates involved in these syntheses are
also indicated in Scheme 2 (n ) 0). Since the enantiomers
of 5 and 6 are available, the enantiomers of 1-4 may be
prepared by the same pathways as those in Scheme 2, if
so desired.
Discr im in a tion of Dia ster eotop ic Gr ou p s a n d
F a ces in In tr a m olecu la r Nitr ile Oxid e [3 + 2] Cy-
cloa d d ition Rea ction s. Oximes prepared from diallyl-
aldehydes 1-4 were treated, after brief purification by
Both the NOESY correlations and typical coupling
constants for 11 and 12 illustrated in Chart 2 unambigu-
a silica gel pad, with aqueous sodium hypochlorite in CH2-
ously led to their absolute configurations as indicated.10
Cl2 to generate the corresponding nitrile oxides, which
The similar structural analysis was conducted for cy-
immediately underwent intramolecular [3 + 2] cyclo-
addition reactions with perfect diastereotopic group
(10) The signal of H(5) appearing at 4.85 ppm as a ddd pattern with
selection, giving a mixture of two diastereomeric isox-
azolines 11 + 12 (13:1), 13 + 14 (10:1), and 15 + 16 (26:
1) from 1, 3, and 4, respectively, indicating the degree of
π-face discrimination [dr (face)], except for the case of
aldehyde 2. These cycloadducts could be separated by
silica gel column chromatography, and their absolute
configurations were carefully analyzed by means of NMR
spectroscopy involving J H-H values and NOESY correla-
tions. These results are summarized in Table 1, and some
selected NMR data are shown in Chart 2.
J values of 0.9 (long-range), 6.1, and 11.2 Hz indicated that the each
TBDMSO group at the preexisting stereogenic center [C(5)] and the
newly induced stereogenic center [C(3)] arrange equatorial and axial,
respectively. On the other hand, the C(5)-hydrogen of the minor product
12 appeared at 4.78 ppm as a dd pattern with J values of 2.1 and 3.9
Hz suggested its equatorial orientation, while the NOESY data for 12
illustrated in Chart 2 clearly indicated the axial TBDMSO-group at
C(5) and the axial allyl-group at C(3) in a chair conformation. The
existence of long-range spin-coupling (J ) 0.9 Hz) between H(5) and
H(1) indicates a 1,3-diaxial arrangement each other intervened by Cd
N π-system. The w-shaped long-range spin coupling between He(2) and
He(4) (J ) 2.4 Hz) was also observed to support the chair conformation
of 11.

Discrimination of Diastereotopic Groups and Faces
J . Org. Chem., Vol. 66, No. 11, 2001 3837
Ch a r t 3
nitrile oxide derived from 2 bearing the 1,2-(TBDMSO)2
controller was very slow (48 h) to unexpectedly result in
the formation of 17 as a mixture of isomers of unknown
stereochemistry (3:1): this cycloadduct probably stemmed
from normal cycloadducts followed by acid-promoted
tautomerization, electrophilic halogenation, and hydroly-
sis under the conditions.12 On the other hand, the
corresponding acetonide version (4) led to the formation
of 1513 and 1614 (26:1) in 59% yield. In any event, it
turned out that the diastereotopic group selection was
perfect again in this case, and the allyl group being
disposed syn to the 1,3-dipole in this 1,3-dioxolane ring
system was exclusively employed as a reaction partner.
Th e Role of Ga u ch e OCCC Ar r a n gem en t. The
absolute configurations of the cycloadducts determined
as above also gave insight into transition states leading
to these cycloadducts, which are summarized in Scheme
3.
The transition states TS11-(chair) and TS12-(boat) lead-
ing to 11 and 12 involve the axial-oriented TBDMSO
group at the prostereogenic center for which the smaller
15
A value (1.06 kcal mol-1
)
than that of an allyl group16
must be responsible partly at least: the difference
between these A values, however, is not sufficiently large
enough to specify one allyl group exclusively. Hence, an
another stabilizing factor contributing to the preference
of TS11-(chair) to TS11′ should be operating. We reasoned
that this factor might be a gauche OCCC arrangement
in which the electrostatic attractive interaction between
the silyl-protected oxygen and the vinyl hydrogen atoms
can operate as pointed out recently by Kishi17 and Houk.18
On the other hand, the degree (86% de) of diastereotopic
faces selection can reasonably be explained on the basis
of the energy difference between TS11-(chair) and less
stable near boatlike conformation [TS12 (boat)] leading
to the minor product (12). The initial cycloadduct pro-
duced through TS12 (boat) should have a structure such
as 12-(boat), which, however, is thermodynamically less
stable and flips to a more stable chair conformation,
giving 12-(chair) even though it contains the axially
oriented TBDMSO and allyl substituents.
Ch a r t 4
cloadducts 13 and 14, by which their structures were
elucidated as those shown in Chart 3 together with
selected J H-H values and NOESY correlations pertinent
to these analyses. A distinctive NOESY correlation
observed between Ha(4) and H(1) was the reason we
determined the conformation of 14 as a boat.11
These results mean that we had four possible diaster-
eomers in hand having the common preexisting C(5)
stereogenic center in an enantiomerically pure form. To
confirm this, the protecting groups of 11-14 were depro-
tected to give four diol derivatives 11(OH)2-14(OH)2 as
shown in Chart 4, the proton NMR spectra of which were
surely not identical to each other at all. In every case,
the exclusive level of diastereotopic groups selection was
achieved, and quite interestingly, their senses were
reversed depending whether a controller was the non-
cyclic 1,3-(TBDMSO)2 (1) or the cyclic 1,3-dioxane (3) as
we can see for 11 and 13: the former prefers the pro-R
allyl group as a reaction partner to the pro-S one, and
the latter prefers the pro-S one to the pro-R one.
The transition states TS13 and TS15 leading to 13 and
15, respectively, involve the 1,3-dioxane or 1,3-dioxolane
ring system in which the cycloaddition took place between
the nitrile oxide and the allyl group positioned cis to the
dipole to result in the exclusive level of diastereotopic
groups selection. The diastereotopic face selection was
controlled by a steric factor similarly to the case of 1 to
(12) As far as the careful NMR analysis of 17 is concerned, we were
able to assign its structure tentatively as shown.
(13) Three typical structure-supporting NOESY correlations were
observed between the proton at C(4) of known absolute configuration
and vinyl or one of C(8) protons, and between the C(1)-proton and the
remaining C(8)-proton. These results led us to unambiguously conclude
the absolute configurations of cycloadduct in this case as 15.
(14) The absolute configurations of 16 was estimated on the basis
of an idea that no chance of cycloaddition reaction was available for
the pro-R allyl group: hence remaining possibility of isomer should
be an epimer at C(1).
(15) Eliel, E. L.; Satici, H. J . Org. Chem. 1994, 59, 688-689.
(16) The A-value for CH2dCHCH2-group is not available from the
exhaustive listing in Eliel, E. L.; Wilen, S. H.; Mander L. N. Stereo-
chemistry of Organic Compounds; Wiley: New York, 1994; pp 695-
698. However, we can roughly estimate it to be near that of ethyl or
little more: the A value of CH3CH2-group is 1.79 kcal/mol (that of
propyl group, not available).
Although the final decision for the structure of 17 is
open to further studies, the cycloaddition reaction of
(17) Wu, T.-C.; Goekjian, P. G.; Kishi, Y. J . Org. Chem. 1987, 52,
4819-4823.
(18) Houk, K. N.; Eksterowicz, J . E.; Wu, Y.-D.; Fuglesang, C. D.;
Mitchell, D. B. J . Am. Chem. Soc. 1993, 115, 4107-4177.
(11) The long-range spin coupling observed between Ha(4) and lower
field equatorial-methyl proton (six-bond away from each other) in the
1,3-dioxane ring also supported this structure.

3838 J . Org. Chem., Vol. 66, No. 11, 2001
Ishikawa et al.
Sch em e 3
Sch em e 4
of cycloaddition reactions between the 1-derived nitrile
oxide and A- or B-derived one: the contribution of such
electrostatic reinforcement to a transition state is typical
for the 1-derived nitrile oxide case [see TS11-(chair)] as
already mentioned.
Discr im in a tion of Dia ster eotop ic Gr ou p s a n d
F a ces in In tr a m olecu la r Nitr on e [3 + 2] Cycloa d -
d ition Rea ction s. Aldehydes 1-4 were converted to the
corresponding nitrones through a reaction with N-ben-
zylhydroxylamine in a usual manner and without any
purification, were subjected to the thermal cycloaddition
conditions. The nitrones derived from 1 and 2 furnished
a separable mixture of diastereomeric isoxazolidines 18
+ 19 + 20 (5:2.4:1) and 21 + 22 (11:1), respectively, and
the nitrone from 4 exclusively gave 23. The cycloaddition
reaction of the nitrone derived from 3 was not effected
when a solution of the nitrone in benzene was heated at
110 °C for 48 h. These results are summarized in Table
2.
NOESY correlations used for the determination of the
absolute configurations of 18-21 and 23 are summarized
in Chart 5. To avoid an any ambiguity in the analysis of
absolute configurations of 19 due to signal overlapping,
the TBDMS protecting group at the preexisting stereo-
genic center was deprotected (TBAF/THF) followed by
usual acetylation to give the corresponding acetyl deriva-
tive (19-(9-OAc)), in which NOESY correlations were
analyzed. The J H-H values including long-range cases
(19) (a) Ishikawa, T.; Ikeda, S.; Ibe, M.; Saito, S. Tetrahedron 1998,
54, 5869-5882. For other diastereoselective intramolecular nitrile
oxide cycloaddition reactions in which one to three stereogenic centers
exist, and dipolarophiles (CdC) directly link to one of such stereo
centers, see: (b) Kozikowski, A. P.; Chen, Y.-Y.; Wan, B. C.; Xu, Z.-B.
Tetrahedron 1984, 40, 2345-2358. (c) Nakata, M.; Akazawa, S.;
Kitamura, S.; Tatsuta, K. Tetrahedron Lett. 1991, 32, 5363-5366. (d)
Takahashi, T.; Shimayama, T.; Miyazawa, M.; Nakazawa, M.; Yamada,
H.; Takatori, K.; Kajiwara, M. Tetrahedron Lett. 1992, 33, 5973-5976.
(e) Kobayashi, Y.; Miyazaki, H.; Shiozaki, M. J . Am. Chem. Soc. 1992,
114, 10065-10066 and J . Org. Chem. 1994, 59, 813-822. (f) Taka-
hashi, T.; Nakazawa, M.; Sakamoto, Y.; Houk, K. N. Tetrahedron Lett.
1993, 34, 4075-4078. (g) Pearson, W. H.; Lovering, F. E. J . Am. Chem.
Soc. 1995, 117, 12336-12337. (h) Takahashi, T.; Iwamoto, H. Tetra-
hedron Lett. 1997, 38, 2483-2486. See also: (i) Irie, O.; Fujiwara, Y.;
Nemoto, H.; Shishido, K. Tetrahedron Lett. 1996, 37, 9229-9232.
result in the comparable level of diastereomeric excess
(10:1 or 26:1 for 3 or 4, respectively).
The degrees of diastereotopic faces selection observed
previously in the intramolecular [3 + 2] cycloaddition
reactions of nitrile oxides (A and B) were 3:1 and 2:1,
respectively, as shown in Scheme 4.19 It is of interest that
the cycloaddition reaction of the nitrile oxide derived from
1 resulted in a much higher level of diastereotopic faces
selection [dr (face) ) 13:1, Table 1). Again, the attractive
electrostatic interaction due to the gauche OCCC ar-
rangement17,18 should be responsible for such marked
difference in the degree of diastereotopic faces selection

Discrimination of Diastereotopic Groups and Faces
J . Org. Chem., Vol. 66, No. 11, 2001 3839
Ta ble 2. Resu lts of In tr a m olecu la r Cycloa d d ition Rea ction s of Nitr on es Der ived fr om Ald eh yd es 1-4
a
b
Conducted using benzene as a solvent containing 1.5 equiv of N-benzylhydroxylamine and aldehydes 1-4. Yields after purification
by silica gel column chromatography (Si ) tert-butyldimethylsilyl). c dr ) diastereomeric ratios calculated using the weight of isolated
products. >99:1 means that no isomer signals were detected in 500 MHz NMR analyses. e Diastereomers, inseparable by conventional
d
silica gel column chromatography; dr determined by NMR. f Other diastereomers not detected at all by both silica gel column
chromatography and NMR analysis.
Ch a r t 5
(0 °C, 0.7 h), the corresponding nitrone did not undergo
the expected cycloaddition reaction at all as mentioned
above. The structural difference between the nitrile oxide
(linear) and nitrone parts (flat but bent plus (Z)-geometry
in this case21) may be responsible for this reactivity
difference: the latter should become much more demand-
ing in terms of orbital overlapping for the [3 + 2]
cycloaddition process to be effected because rotation
about two carbon-carbon σ-bonds are prohibited due to
the incorporation of 1,3-dioxane ring system.22
The absolute configurations of cycloadducts 18 and 19
for the nitrone prepared from 1 indicated that a reversal
of diastereotopic groups selection occurred as compared
to the case of the nitrile oxide derived from 1 although
the same 1,3-(TBDMSO)2 controller was used. On the
basis of this fact, we can draw a possible transition state
such as TS18-C (pro-S) as far as a chair conformation is
concerned (Scheme 5). This structure, however, is not
only suffering from severe 1,3-diaxial interaction but also
is bearing the axial allyl group of larger A value16 than
the TBDMSO group15 and, therefore, should be ruled out.
Thus, we reached the conclusion that the transition state
leading to 18 might be boat in nature as TS18-B (pro-S)
bearing the axially oriented TBDMSO group in which
electrostatically stabilizing interaction through the gauche
OCCC arrangement17,18 is operating similarly to the case
(20) It should be mentioned that the NMR spectrum of an insepa-
rable mixture of 21 and 22 fortunately indicated that the J 1,2 values
of 21 and 22 are 8.5 and 8.9 Hz, respectively which were crucial for
the absolute configuration determination. In marked contrast to these
the J 1,2 value of 23 is nearly zero indicating 1,2-trans arrangement.
(21) Saito, S.; Ishikawa, T.; Kishimoto, N.; Kohara, T.; Moriwake,
T. Synlett 1994, 282-284.
(22) Molecular models provided us with a more insight into this
problem. We considered eight models of possible transition states,
among which only two cases can make orbital overlap possible, in
which, however, the olefin part is forced to be in too sterically congested
space for the reaction to be effected. One remaining chance might be
the isomerization of nitrone geometry from (Z) to (E), facilitating
effective orbital overlapping but apparently giving rise to unfavorable
torsional strain.
obtained for these cycloadducts were also used for the
analysis of absolute configurations in combination with
the NOESY data.20
In marked contrast to the case of nitrile oxide derived
from 3 in which [3 + 2] cycloaddition smoothly took place

3840 J . Org. Chem., Vol. 66, No. 11, 2001
Ishikawa et al.
Sch em e 5
Ch a r t 6
selection as the nitrile oxide case for the nitrone cyclo-
addition reaction with an exclusive level of diastereomeric
ratio, and at the same time, the π-facial discrimination
turned out to be exclusive (23). A model study gave more
insight into this outstanding stereocontrol, which sug-
gested that the orbital overlapping of the [3 + 2]
cycloaddition process of this class became possible only
when Si-Si face matching was achieved as in TS23: Re-
Re matching as in TS23′ seems impossible because of
extremely severe steric congestion and torsional strain.
Role of Con tr oller s in Bia sin g Ster eoch em ica l
Cou r ses of In tr a m olecu la r Nitr ile Oxid e or Nitr on e
Cycloa d d ition Rea ction s. Before moving to the studies
on the application of above-mentioned strategy to other
classes of cycloaddition reactions such as the Pauson-
Khand reaction, it seems important to summarize the
steric or electronic impact of the four types of controllers
on the transition-state models of the tandem stereo-
determining steps involving diastereotopic groups and
faces selections of this class (Chart 7).
Since the dipolar moieties are linking directly to the
preexisting stereogenic centers that are incorporated into
the controllers of 1,3-dioxolane or 1,3-dioxane ring sys-
tem, they can nicely bias the geminal allyl groups,
dictating those being disposed syn to the dipolar moieties
to become a reaction partner (>99%) as shown in
TS13,15,23. These heterocyclic controllers also provide the
facial bias to the allylic CdC bond imparted by steric and
electronic effects. The steric effects might be evaluated
positively or negatively as the result of delicate balance
among various types of steric interactions, whereas the
of TS11 as mentioned in Scheme 3. Hence, the stereo-
chemical outcome of 19 might also be controlled by a
similar boat transition state as TS19-B
.
With regard to the formation of 20, the pro-R allylic
group was selected as a reaction partner, and therefore,
the reaction should be intervened with a chair-type
transition state bearing an axially oriented TBDMSO
group as TS20(Z), in which, however, no positive orbital
overlapping seems possible even though it features the
gauche OCCC arrangement. Therefore, in order for the
formation of 20 to be effected, the geometry of the nitrone
must change from Z to E (TS20(E)) during the reaction.
The rate of such Z to E isomerization, however, is known
to be slow.23 This is the reason the thermodynamically
less stable adducts 18 and 19 than 20 became major
products in this case. To confirm that 18 was given under
kinetically controlled conditions, a solution of 18 in
toluene was heated at 110 °C for 24 h to result in the
recovery of 18 and also in the formation of neither 19
nor 20.
The 1,2-(TBDMSO)2 controller system, by which the
nitrile oxide cycloaddition did not proceed at all as shown
above in Table 1, provided an opportunity in favor of the
pro-R allyl group for the nitrone cycloaddition reaction.
The role of 1,2-(TBDMSO)2 controller has its origin in
their stable anti-periplanar arrangement (anti OCCO).9
In addition to this crucial backbone, the additional
favorable structural feature such as the gauche OCCC
arrangement (twice) is present in the transition state
TS21 responsible for the formation of cycloadduct 21,
which is lacking in the transition state TS22 leading to
the minor cycloadduct 22 (Chart 6).
(24) For asymmetric induction in intramolecular nitrone cyclo-
addition reactions, see: (a) Ihara, M.; Takahashi, M.; Fukumoto, K.;
Kametani, T. J . Chem. Soc. Chem. Commun. 1988, 9-10. (b) Roush,
W. R.; Watts, A. E. J . Am. Chem. Soc. 1984, 106, 721-723. (c)
Annunziata, R.; Cinquini, M.; Cozzi, F.; Raimondi, L. Tetrahedron
1987, 43, 4051-4056. (d) Ishikawa, T.; Kudo, T.; Shigemori, K.; Saito,
S. J . Am. Chem. Soc. 2000, 118, 7633-7637; For asymmetric induction
in intramolecular dipolar cycloaddition reactions other than nitrones,
see: (e) Righi, P. R.; Marotta, E.; Landuzzi, A.; Rosini, G. J . Am. Chem.
Soc. 1996, 118, 9446-9447. (f) Denmark, S. E.; Hurd, A. R.; Sacha, H.
J . J . Org. Chem. 1997, 621, 11668-1674. (g) Garner, P.; Cox, P. B.;
Anderson, J . A.; Protasiewicz, J .; Zaniewski, R. J . Org. Chem. 1997,
62, 493-498. (h) Young, D. G. J .; G.-Bengoa, E.; Hoveyda, A. H. J .
Org. Chem. 1999, 64, 692-693. (i) Marrugo, H.; Dogbe´avou, R.; Breau,
L. Tetrahedron Lett. 1999, 40, 8979-8983.
The 1,3-dioxolane controller system provided an op-
portunity of the same sense of diastereotopic group
(23) LeBel, N. A.; Banucci, E. G. J . Org. Chem. 1971, 36, 2440-
2448. See also ref 24d.

Discrimination of Diastereotopic Groups and Faces
J . Org. Chem., Vol. 66, No. 11, 2001 3841
Ch a r t 7. Su m m a r y of th e Sen se a n d Degr ee of Dia ster eotop ic Gr ou p s a n d F a ce Selection s in
In tr a m olecu la r Nitr ile Oxid e a n d Nitr on e Cycloa d d ition Rea ction sa
a
b
See also Tables 1 and 2: Si ) tert-butyldimethylslyl. See inset for “n” and “con tr oller ”. c DGS ) diastereotopic group selection;
DFS ) diastereotopic face selection; dr ) diastereomeric ratio.
electronic effect can operate positively through electro-
static interaction in the gauche Si-OCCCd arrange-
ments where the oxygen atom is linking to the pro-
stereogenic center: this kind of electrostatic effect be-
comes possible only in the transition states leading to
major diastereoisomers (TS13,15,23).24
reaction,26 we can illustrate the mechanism of this
process using 24-27 (Scheme 6) prepared from chiral
diallyl aldehydes 1-4 through Corey’s acetylenation
protocol27 followed by trimethylsilylation of the terminal
sp carbons (Scheme 7).28 The events of both diastereotopic
group and face selection of the Pauson-Khand reaction
might take place during N f O conversion. Furthermore,
since the first intermediate M, the alkyne-Co2(CO)6
complex, has two diastereotopic cobalt atoms, their
subsequent complexation to the terminal olefin should
affect the eventual stereochemical outcomes of Pauson-
Khand reaction of 24-27, in particular the degree of the
diastereotopic faces selection. Thus, to achieve the high
level of diastereoselectivity of this class, we expected that
the TMS group introduced to the acetylenic terminus
might serve as one of control elements with regard to
which of the cobalt atoms is able to be involved in the
complexation with the terminal olefin.
Discr im in a tion of Dia ster eotop ic Gr ou p s a n d
F a ces in P a u son -Kh a n d [2 + 2 + 1] Cycloa d d ition
Rea ction s. Treatment of 24-27 with Co2(CO)8 in dichlo-
romethane at 0 °C to room temperature for 3 h gave the
corresponding acetylenehexacarbonyldicobalt complexes
(M) as a rather stable dark red oil, which were roughly
purified by column chromatography over silica gel. A
solution of these complexes in acetonitrile was heated at
65-75 °C for 11-18 h, furnishing the [2 + 2 + 1]
On the other hand, we should discuss the role of the
acyclic controllers individually with regard to the 1,3-
(TBDMSO)2 and 1,2-(TBDMSO)2. The 1,3-(TBDMSO)2
controller involves the equatorial TBDMSO group at the
preexisting stereogenic center and the axial one at the
prostereogenic center due to the generalization that a
substituent of smaller A value arranges axial at the
bicyco[4.3.0]-type transition states including the six-
membered system where both TBDMSO groups are
involved (TS11,18-B). The most stable conformation at-
tained by the 1,2-(TBDMSO)2 controller should basically
involve anti-OCCO arrangement9 as shown in TS21, the
origin of which has been well approved so far.25 Similar
to the case of heterocyclic controllers, the gauche OCCC
arrangement is commonly involved in these most stable
transition states and might positively play an important
role in biasing the diastereotopic faces selection again.
Mech a n ism of P a u son -Kh a n d Rea ction a n d th e
P r ep a r a tion of Dien yn es. According to the current
level of mechanistic understanding of Pauson-Khand
(25) Gung, B, W.; Zhu, Z.; Fouch, R. A. J . Am. Chem. Soc. 1995,
122, 1783-1788 and references therein.
(26) (a) Schore, N. E. Org. React. 1991, 40, 1-90. (b) Verdaguer, X.;
Va´zquez, J .; Fuster, G.; B.-Ge´nisson, V.; Greene, A. E.; Moyano, A.;
Pericas, M. A.; Riera, A. J . Org. Chem. 1998, 63, 7073-7052 and
references therein.
(27) Corey, E. J .; Fuchs, P. L. Tetrahedron Lett. 1972, 36, 3769-
3772.
(28) Conversion of 2 to 26 through the Corey’s protocol resulted in
very low yield (7%). The enyne 26, therefore, was provided from 4
through a series of reactions shown bottom in Scheme 7.

3842 J . Org. Chem., Vol. 66, No. 11, 2001
Ishikawa et al.
Sch em e 6
Sch em e 7a
For 32 or 34, NOEs between the allyl group and both
C(2)-H (convex space) or C(4)-H (concave space) were
clearly observed (Scheme 8), which, in turn, left behind
ambiguity with regard to the sense of the diastereotopic
groups selection. To make this point clear, both 32 and
34 were converted to 38 (H2/Pd-C) and 40 [(Ph3P)CuH]6/
C6H6], respectively, in which we observed clear-cut NOEs
between C(2)-H and one of C(8)-H’s and no NOE between
this C(8)-H and C(5)-H, indicating that the both initial
cycloadducts 32 and 34 bear the bridgehead hydrogen
arranging in convex space. For 34, since it was quite clear
that the allyl group positioned cis to the trimethylsilyl-
ethynyl group became a reaction partner in this Pauson-
Khand reaction, the absolute configurations at C(1) and
C(3) should be those as shown in 34 in Scheme 8.
However, the absolute configuration at C(3) of 32 still
remained ambiguous. Since the absolute configurations
of 32 and 34 at the bridgehead centers are identical, the
most simple and reliable way of solving this ambiguity
is to convert 38 (+ 39) into 41 (+ 42), which can be
examined as to whether 34 is identical with the de-
acetonidation product of 40. Thus, 38 containing 39 was
treated with HCl in EtOH to give 41 (+ 42), whereas 34
was reduced (H2/Pd-C) and deprotected (HCl/EtOH) to
give a diol, the NMR of which was identical with that of
42 (Scheme 8). Thus, the absolute configuration of 32 was
revealed to be those shown in Table 3 or Scheme 8.
Or igin of th e Dia ster eoselection s. To determine
what is responsible for the sense of the diastereotopic
groups and faces selections, the results of Pauson-Khand
reactions are listed in Table 4. These data together with
those pertinent to this problem listed in Chart 7 clearly
indicate that the levels of control for the diastereotopic
a
Key: (a) (1) CBr4/PPh3/Zn/0 °C to rt, (2) BuLi/THF/-78 °C,
then TMS-Cl; (b) (1) (a) - 1, (2) (a) - 2 without TMS-Cl; (c) (1)
90% AcOH/MeOH/PTSA, (2) TBDMSOTf/Et3N; (d) BuLi/TMS-Cl.
cycloadducts 28 + 29 (9:1), 30 + 31 (10:1), 32 + 33 (10:
1), and 34 (exclusive) from M as summarized in Table 3.
The Pauson-Khand reaction of additional substrate 36,
which is missing the terminal trimethylsilyl group, was
carried out for comparison and resulted in a mixture of
four isomers in a ratio of 2.7:1.5:1.5:1. Therefore, it seems
that the terminal TMS groups tagged in 24-27 play an
important role in the stereochemistry-determining step
(N f O) of the Pauson-Khand reaction (Scheme 6) as
expected.
The absolute configurations of these cycloadducts were
determined not only by NOE experiments but also by
appropriate derivatizations, in particular for bicyclo-
[3.3.0]-type products such as 32 or 34. It should be
mentioned that, as quickly recognized from the absolute
configurations shown in Chart 8, adduct 28 apparently
less stable than 29 became a major product in the case
of 24. The exact reason for this result could not be given
an answer at present because we have no reliable idea
for the transition state structure during the N f O
conversion.29
(29) Based on a model inspection we tentatively have an idea about
the transition state structure leading, for instance, to 34 as shown
below.

Discrimination of Diastereotopic Groups and Faces
J . Org. Chem., Vol. 66, No. 11, 2001 3843
Ta ble 3. Resu lts of P a u son -Kh a n d Rea ction s
a
Heating a solution of cobalt-enyne complexes, prepared from 24-27 (120 mol % of Co2(CO)8/CH2Cl2/0 °C to room temperature, 1-5
h) in acetonitrile. Yields after purification by silica gel column chromatography (Si ) tert-butyldimethylsilyl). c Determined on the basis
b
of the absolute configurations of the products. Diastereomeric ratios calculated using the weight of isolated product. e Other diastereomers
d
not detected at all by both silica gel column chromatography and NMR analysis.
groups selection were uniformly high. Thus, the substrate-
dependent senses of the diastereotopic groups selection
were generally observed irrespective of the reaction type,
suggesting that the diastereotopic bias imposed on the
prostereogenic groups (gem-allyl groups) is the result of
stereocontrol dictated by the controllers. On the other
hand, the sense of the diastereotopic faces selection
exhibited no controller-dependent regular change, sug-
gesting that the diastereofacial bias was conferred on the
terminal olefins as the result of the delicate balance
between electrostatic and steric effects which should be
determined intrinsically depending on the nature of the
reaction. In any event, the degrees of the diastereotopic
faces selection were uniformly acceptable.
group discriminations observed in the combinations
between three types of reactions and four types of
substrates were uniformly excellent (dr ) >99:1 for seven
cases and 7.3:1-11:1 for the remaining three cases). We
have also achieved the high level of asymmetric induction
in the π-facial discrimination ranging from dr ) 9:1 to
dr ) >99:1 with only one exception (dr ) 2.1:1). In the
systems presented here, the multiple stereogenic centers
have been constructed from a single stereogenic center
in a single operation to realize the high levels of stereo-
control. The products themselves as well as these strate-
gies will be of high value toward the synthesis of complex
molecules.
Exp er im en ta l Section
1
Con clu sion
Gen er a l Meth od s. H NMR spectra were recorded at 300
or 500 MHz and 13C NMR at 75 MHz using CDCl3 as a solvent.
We have developed a novel method for realizing
simultaneous discrimination of diastereotopic groups and
their faces in intramolecular [3 + 2] and [2 + 2 + 1]
cycloaddition reactions. The processes afforded both
bicyclo[3.3.0] and bicyclo[4.3.0] adducts with three to four
stereogenic centers in which isoxazolidine, isoxazoline,
or cyclopentenone segments were fused. Special attention
was paid to the present stereochemical models to account
for how the 1,3-(TBDMSO)2, 1,2-(TBDMSO)2, 1,3-dioxane,
and 1,3-dioxolane controllers work. The degrees of the
The chemical shifts (δ) are given in parts per million relative
1
to internal CHCl3 (7.26 ppm for H) or CDCl3 (77 ppm for 13C).
The 1H NMR spectral data are indicated in the following
manner: δ value of signal (peak multiplicity, integrated
number of protons, and coupling constant (if any)). Splitting
patterns are abbreviated as follows: s, singlet; d, doublet; t,
triplet; m, multiplet; b, broad. All the chemical shift assign-
ments were consistent with COSY spectra, and conformational
analyses were carried out on the basis of a combination of J
values, NOE spectra, and NOESY spectra, and in every case
the absolute configurations were able to be unambiguously

3844 J . Org. Chem., Vol. 66, No. 11, 2001
Ishikawa et al.
Ch a r t 8
Sch em e 8a
determined by means of those experiments for the cycloadducts
or their derivatives. FR-IR spectra were recorded using NaCl
plates. Elemental analyses were carried out by Dr. Miyoko
Izawa of this laboratory. Optical rotations were measured on
a digital polarimeter using a 3.5 mm × 0.5 dm Pyrex cell.
Analytical thin-layer chromatography was performed on
Merck precoated silica gel 60 F-254 (0.25 mm thickness).
Column chromatography was performed on Merck silica gel
60 7734 using an appropriate ratio of ethyl acetate (EtOAc)-
hexane mixed solvent and abbreviated as CC. Although
diastereoisomeric cycloadducts were able to be separated
through flash column chromatography in many cases, it was
repeatedly required in order to achieve the complete separation
of the isomers.
a
Key: (a) (1) H2/Pd-C/EtOH, (2) TBAF/THF; (b) 2 N HCl/
EtOH.
Ta ble 4. Su m m a r y of th e Sen se a n d Degr ee of
Dia ster eotop ic Gr ou p s a n d F a ce Selection s in
In tr a m olecu la r P a u son -Kh a n d Rea ction sa
group selection
face selection
controller
sense
drb
sense
drb
All reactions, unless otherwise noted, were conducted under
a nitrogen or an argon atmosphere. Liquid reagents were
transferred via a dry hypodermic syringe and added through
a rubber septum wired onto a reaction flask from which a
steady stream of inert gas was flowing. Organic extracts were
concentrated by evaporation with a rotary evaporator evacu-
ated at around 60 mmHg. Unless otherwise noted, materials
were obtained from commercial suppliers, and reagent grade
materials were used without further purification. Toluene,
triethylamine (Et3N), dichloromethane (CH2Cl2), acetonitrile,
and benzene were freshly distilled from CaH2 prior to use.
Methanol (MeOH) and ethanol (EtOH) were distilled from
magnesium turnings under argon. Tetrahydrofuran purchased
from Kanto Chemical Co., Inc., is dehydrated and stabilizer-
free grade and was used as received.
c
1,2-(TBDMSO)2
1,3-(TBDMSO)2
1,3-dioxolane
1,3-dioxane
pro-R
pro-R
pro-S
pro-S
10:1
>99:1
>99:1
>99:1
Si face
Si face
Si face
Si face
>99:1
9:1
>99:1
10:1
See also Table 3. dr ) diastereomeric ratio. c dr of face
selection for the major diastereomer.
a
b
MHz) δ 1.35 (s, 3H), 1.40 (s, 3H), 1.61-1.78 (m, 2H), 2.22-
2.42 (m, 4H), 3.06 (s, 1H), 3.51 (dd, 1H, J ) 7.6, 7.9 Hz), 4.08
(dd, 1H, J ) 5.9, 7.9 Hz), 4.33-4.41 (m, 1H), 5.04-5.13 (m,
4H), 5.55-5.59 (m, 2H); 13C NMR (75 MHz) δ 25.7, 26.7, 41.4,
43.8, 44.2, 70.1, 72.7, 73.0, 109.0, 118.1, 118.3, 133.7, 133.8;
IR (film) 3492, 1640 cm-1. Anal. Calcd for C13H22O3: C, 68.99;
H, 9.80. Found: C, 68.77; H, 9.90.
(2S)-4-Allyl-1,2-O-(isop r op ylid en e)-6-h ep ten e-1,2,4-tr i-
ol (7). To a solution of ethyl (3S)-3,4-dihydroxybutanoate 5
(1.0 g, 5.4 mmol) in ether (10 mL) was added allylmagnesium
bromide (1.0 M, 2.5 equiv) at -78 °C. The mixture was stirred
at that temperature for 50 min. The reaction was quenched
by the addition of water and filtered through a Celite pad, the
filter cake being thoroughly rinsed with AcOEt. The combined
organic solutions were dried with Na2SO4 and concentrated
to give an oil, which, on careful CC (50:1 hexane/EtOAc), gave
(2S)-4-Allyl-1-O-p iva loyl-6-h ep ten e-1,2,4-tr iol (8). To a
solution of the diallyl acetonide 7 (1.21 g, 5.3 mmol) in EtOH
(6 mL) was added 2 N HCl (6 mL) at 0 °C. The mixture was
stirred at room temperature for 2 h, concentrated by a rotary
evaporator, and dried under high vacuum for overnight to give
an oil, which, on CC (2:1 hexane/EtOAc), gave (2S)-4-allyl-6-
heptene-1,2,4-triol (816 mg, 83%): [R]28D +15.7 (c 1.07, CHCl3);
1H NMR (300 MHz) δ 1.46-1.74 (m, 2H), 2.19-2.42 (m, 4H),
2.60-2.70 (b, 1H), 3.00 (b, 1H), 3.40-3.63 (m, 2H), 4.00 (b,
1H), 4.12 (m, 1H), 5.08-5.19 (m, 4H), 5.72-5.92 (m, 2H); 13C
7 (1.22 g, 94%): [R]21 +11.9 (c 1.16, CHCl3); 1H NMR (300
D

Discrimination of Diastereotopic Groups and Faces
J . Org. Chem., Vol. 66, No. 11, 2001 3845
NMR (75 MHz) δ 39.6, 42.7, 45.2, 67.0, 69.1, 74.2, 119.0, 119.2,
p-TsOH (2.5 mg, 0.07 equiv). The mixture was stirred at room
temperature for 36 h and neutralized by the addition of Et3N
(0.002 mL, 0.07 equiv). The mixture was concentrated by a
rotary evaporator to give an oil, which, on CC (30:1 hexane/
EtOAc), afforded (2S)-4-allyl-2,4-O-(isopropylidene)-1-O-
pivaloyl-6-heptene-1,2,4-triol (42 mg, 68%): 1H NMR (300
MHz) δ 1.21 (s, 9H), 1.39 (s, 3H), 1.44 (s, 3H), 1.40-1.56 (m,
2H), 2.18-2.60 (m, 4H), 4.07 (m, 2H), 4.10-4.18 (m, 1H), 5.00-
5.14 (m, 4H), 5.75-5.90 (m, 2H).
133.0, 133.2; IR (film) 3368 cm-1
.
To a stirred solution of the 1,2,4-triol (816 mg, 4.3 mmol) in
CH2Cl2 (8 mL) were added pyridine (0.45 mL, 1.3 equiv) and
pivaloyl chloride (0.64 mL, 1.2 equiv) at -78 °C. The mixture
was stirred at -78 °C for 4 h, quenched by the addition of
water, and extracted with CH2Cl2 (3 times). The combined
extracts were dried with Na2SO4 and concentrated by a rotary
evaporator to give an oil, which, on CC (3:1 hexane/EtOAc),
1
gave 8 (1.09 g, 94%): [R]29 +8.29 (c 1.64, CHCl3); H NMR
To a solution of the triol (583 mg, 1.88 mmol) in toluene
(6.0 mL) was added dropwise DIBALH (1.01 M in toluene, 5.6
mL, 5.63 mmol) at -78 °C, and the mixture was stirred at
-78 °C for 40 min. The reaction was quenched by the addition
of water, and the solid was removed by filtration through a
Celite pad. The combined organic solutions were dried with
Na2SO4 and concentrated to give an oil, which, on CC (2:1
D
(300 MHz) δ 1.22 (s, 9H), 1.54-1.72 (m, 2H), 2.18-2.46 (m,
4H), 2.85 (b, 1H), 3.48 (b, 1H), 4.03 (d, 2H, J ) 5.22 Hz), 4.21-
4.31 (m, 1H), 5.09-5.20 (m, 4H), 5.73-5.94 (m, 2H); 13C NMR
(75 MHz) δ 27.1, 38.8, 40.1, 42.8, 45.1, 67.2, 68.4, 74.0, 119.0,
119.1, 133.0, 133.3, 178.6; IR (film) 3419, 1732 cm-1. Anal.
Calcd for C15H26O4: C, 66.64; H, 9.69. Found: C, 66.39; H,
9.83.
hexane/EtOAc), afforded 10 (400 mg, 94%): [R]24 +73.7 (c
D
(2S)-4-Allyl-2,4-b is-(O-ter t-b u t yld im et h ylsilyl)-6-h ep -
ten e-1,2,4-tr iol (9). To a solution of 8 (424 mg, 1.57 mmol) in
CH2Cl2 (5 mL) was added TBDMSOTf (1.13 mL, 3.2 equiv)
followed by the addition of Et3N (0.88 mL, 3.9 equiv) at 0 °C.
The reaction was continued at 0 °C for 2 h, quenched by the
addition of water, and extracted with AcOEt. The combined
extracts were dried with Na2SO4 and concentrated to give a
clear colorless oil, which, on CC (10:1 hexane/EtOAc), gave
(2S)-4-allyl-1-O-pivaloyl-2,4-bis-(O-tert-butyldimethylsilyl)-6-
heptene-1,2,4-triol (504 mg, 83%): [R]21D -9.98 (c 1.24, CHCl3);
1H NMR (300 MHz) δ 0.09 (s, 3H), 0.10 (s, 3H), 0.11 (s, 3H),
0.12 (s, 3H), 0.87 (s, 18H), 1.12 (s, 9H), 1.62-1.81 (m, 2H),
2.20-2.52 (m, 2H), 3.96-4.02 (m, 2H), 4.12-4.21 (m, 1H),
4.96-5.12 (m, 4H), 5.78-5.95 (m, 2H); 13C NMR (75 MHz) δ
-4.2, -4.0, -1.6, -1.4, 17.9, 18.5, 25.8, 26.0, 27.2, 38.7, 43.9,
44.6, 46.0, 67.4, 69.0, 76.7, 117.5, 117.6, 134.4, 134.9, 178.3;
0.867, CHCl3); 1H NMR δ 1.34 (dd, 1H, J ) 3.0, 13.4 Hz), 1.40
(s, 3H), 1.49 (s, 3H), 1.53 (dd, 1H, J ) 11.5, 13.4 Hz), 2.01-
2.07 (b, 1H), 2.19-2.33 (m, 3H), 2.55-2.64 (m, 1H), 3.50-3.68
(m, 2H), 4.02-4.11 (m, 1H), 5.00-5.15 (m, 4H), 5.73-5.91 (m,
2H); 13C NMR (75 MHz) δ 25.7, 31.4, 32.6, 42.6, 45.9, 66.1,
66.4, 74.0, 98.7, 118.1, 118.2, 133.6, 133.8; IR (film) 3447 cm-1
.
Anal. Calcd for C13H22O3: C, 68.99; H, 9.80. Found: C, 68.75;
H, 9.90.
(1R,3R,5S)-3-Allyl-3,5-bis-(ter t-bu tyld im eth ylsilyloxy)-
7-a za -8-oxa bicyclo[4.3.0]n on -6-en (11). To a solution of the
aldehyde 1 (385 mg, 0.94 mmol) in EtOH (5 mL) were added
NH2OH-HCl (457 mg, 6.60 mmol) and Et3N (0.92 mL, 6.60
mmol) at 0 °C, and the mixture was stirred at room temper-
ature for 14 h. The reaction was diluted with water and
extracted with AcOEt. The combined extracts were dried with
Na2SO4 and concentrated to afford an oil, which, on CC (50:1
hexane/EtOAc), gave (2R)-4-allyl-2,4-bis-(O-tert-butyldimethyl-
silyl)-2,4-dihydroxy-6-heptenal oxime (381 mg, 95%). To a
stirred solution of this aldoxime (381 mg, 0.90 mmol) in CH2-
Cl2 (8 mL) was added an aqueous NaOCl solution (1.5 mL),
and the mixture was stirred at room temperature for 1.5 h.
The reaction was diluted with water and extracted with
AcOEt. The combined extracts were dried with Na2SO4 and
concentrated to give an oil, which, on CC (125:1 hexane/
EtOAc), gave 11 (more polar, 267 mg, 72%) and 12 (less polar,
20 mg, 5%). Data for the major isomer 11: [R]31D -67.4 (c 1.00,
IR (film) 1734 cm-1
.
To a solution of (2S)-4-allyl-1-O-pivaloyl-2,4-bis-O-(tert-
butyldimethylsilyl)-6-heptene-1,2,4-triol (550 mg, 1.10 mmol)
in toluene (6 mL) was added dropwise DIBALH (1.01 M in
toluene, 2.0 equiv) at -78 °C. The mixture was stirred at -78
°C for 1 h and quenched by the addition of EtOH-water
mixture. The precipitates were removed by filtration through
a Celite pad, which was rinsed with several portions of ethyl
acetate. The combined organic solutions were dried with Na2-
SO4 and concentrated to give an oil, which, on CC (50:1 hexane/
EtOAc), afforded 9 (430 mg, 94%): [R]24D +6.36 (c 1.44, CHCl3);
1H NMR (300 MHz) δ 0.09 (s, 6H), 0.88 (s, 9H), 0.89 (s, 9H),
0.12 (s, 6H), 1.71-1.78 (m, 2H), 1.97-2.07 (m, 1H), 2.19-2.43
(m, 4H), 3.42-3.52 (m, 1H), 3.64-3.74 (m, 1H), 3.98-4.09 (m,
1H), 4.98-5.12 (m, 4H), 5.75-5.94 (m, 2H); 13C NMR (75 MHz)
δ -4.3, -4.2, -1.5, -1.4, 18.0, 18.5, 25.8, 26.1, 43.9, 44.6, 46.0,
1
CHCl3); H NMR (500 MHz) δ 0.09 (s, 6H), 0.13 (s, 3H), 0.14
(s, 3H), 0.16 (s, 3H), 0.88 (s, 9H), 0.91 (s, 9H), 1.49 (dd, 1H, J
) 12.2, 13.1 Hz), 1.79 (dd, 1H, J ) 11.2, 12.8 Hz), 1.91 (ddd,
1H, J ) 2.4, 6.4, 13.1 Hz), 2.05 (ddd, 1H, J ) 2.4, 6.1, 12.8
Hz), 2.31-2.42 (m, 2H), 3.55-3.63 (m, 1H), 3.83 (dd, 1H, J )
7.6, 10.3 Hz), 4.53 (dd, 1H, J ) 7.9, 10.3 Hz), 4.85 (ddd, 1H, J
) 0.9, 6.1, 11.2 Hz), 5.05-5.11 (m, 2H), 5.67-5.77 (m, 1H);
13C NMR (75 MHz) δ -9.7, -9.2, -6.33, -6.31, 13.9, 14.0, 21.3,
21.4, 37.6, 40.7, 42.6, 43.2, 61.3, 69.1, 72.7, 114.5, 128.4, 157.2;
IR (film) 2857, 1471 cm-1. Anal. Calcd for C22H43NO3Si2: C,
62.06; H, 10.18; N, 3.29. Found: C, 61.87; H, 10.39; N, 3.10.
67.5, 70.0, 76.5, 117.8, 117.9, 134.2, 134.4; IR (film) 3467 cm-1
.
Anal. Calcd for C22H46O3Si2: C, 63.71; H, 11.18. Found: C,
63.56; H, 11.08.
(2S)-4-Allyl-2,4-b is-O-(ter t-b u t yld im et h ylsilyl)-2,4-d i-
h yd r oxy-6-h ep ten a l (1). A solution of 1-O-deacylated 9 (2.77
g, 6.67 mmol) in CH2Cl2 (30 mL) was mixed with Swern
reagent prepared from (COCl)2 (1.17 mL, 13.3 mmol) and
DMSO (1.89 mL, 26.7 mmol) in CH2Cl2 (5 mL) at -78 °C. After
15 min of stirring, Et3N (7.45 mL, 0.053 mol) was added to
the mixture at -78 °C, and stirring was continued for
additional 30 min at 0 °C. The reaction was quenched by the
addition of water and extracted with AcOEt. The combined
organic solutions were dried with Na2SO4 and concentrated
to give an oil, which, on CC (50:1 hexane/EtOAc), afforded 1
(2.47 g, 90%): [R]27D -8.66 (c 1.12, CHCl3); 1H NMR (300 MHz)
δ 0.06 (s, 3H), 0.08 (s, 3H), 0.11 (s, 3H), 0.13 (s, 3H), 0.88 (s,
9H), 0.91 (s, 9H), 1.74 (dd, 1H, J ) 6.6, 14.6 Hz), 1.90 (dd, 1H,
J ) 4.9, 14.6 Hz), 2.28-2.44 (m, 2H), 4.27-4.32 (m, 1H), 5.00-
Data for the minor (1S,3R,5S)-isomer (12): [R]31 +2.4 (c
D
2.21, CHCl3); 1H NMR (500 MHz) δ 0.05 (s, 6H), 0.06 (s, 3H),
0.11 (s, 3H), 0.84 (s, 9H), 0.88 (s, 9H), 1.47 (dd, 1H, J ) 12.2,
12.8 Hz), 1.78 (dd, 1H, J ) 3.9, 14.0 Hz), 2.14 (dt, 1H, J ) 2.1,
14.0 Hz), 2.20 (ddd, 1H , J ) 2.1, 5.8, 12.2 Hz), 2.46-2.51 (m,
1H), 2.76-2.82 (m, 1H), 3.41-3.49 (m, 1H), 3.92 (dd, 1H, J )
7.9, 8.2 Hz), 4.44 (dd, 1H, J ) 8.2, 10.6 Hz), 4.78 (dd, 1H, J )
2.1, 3.9 Hz), 5.04-5.10 (m, 2H), 5.91-5.99 (m, 1H); 13C NMR
(75 MHz) δ -5.1, -4.8, -1.8, -1.7, 17.9, 18.1, 25.6, 25.8, 43.2,
43.7, 44.7, 46.8, 63.4, 73.9, 74.5, 117.3, 134.8, 159.7; IR (film)
2857, 1472 cm-1. Anal. Calcd for C22H43NO3Si2: C, 62.06; H,
10.18; N, 3.29. Found: C, 61.90; H, 10.22; N, 3.27.
(1R ,3S ,5S ,6S )-3-Allyl-7-b e n zyl-3,5-b is-(t er t -b u t yld i-
m eth ylsilyloxy)-7-a za -8-oxa bicyclo[4.3.0]n on a n e (18). To
a solution of the aldehyde 1 (970 mg, 2.35 mmol) in benzene
(20 mL) was added N-benzylhydroxylamine (435 mg, 3.53
mmol) at 0 °C. The mixture was stirred at 50 °C for 18 h. The
reaction was diluted with water and extracted with AcOEt.
The combined extracts were dried with Na2SO4 and concen-
trated to give an oil, which, on CC (50:1 hexane/EtOAc),
afforded 19 (253 mg, 21%), 18 (537 mg, 44%), and 20 (107 mg,
5.12 (m, 2H), 5.76-5.93 (m, 1H), 9.52 (d, 1H, J ) 2.5 Hz); 13
C
NMR (75 MHz) δ -4.7, -4.2, -1.7, -1.6, 18.1, 18.5, 25.8, 25.9,
26.1, 26.2, 41.9, 44.8, 45.8, 75.1, 76.8, 118.0, 118.1, 134.0, 134.3,
202.4; IR (film) 1737, 1472 cm-1. Anal. Calcd for C22H44O3Si2:
C, 64.02; H, 10.74. Found: C, 63.80; H, 10.99.
(2S)-4-Allyl-2,4-O-(isop r op ylid en e)-6-h ep ten e-1,2,4-tr i-
ol (10). To a solution of 8 (54 mg, 0.20 mmol) in THF (2 mL)
were added 2,2-dimethoxypropane (0.05 mL, 2 equiv) and

3846 J . Org. Chem., Vol. 66, No. 11, 2001
Ishikawa et al.
8.8%), which were eluted in this order. Physical data for 18:
allyl-3,4-isopropylidenedioxy-1,6-heptadiene: [R]23 -31.2 (c
D
[R]28 -21.2 (c 1.56, CHCl3); 1H NMR (500 MHz) δ 0.09 (s,
1.00, CHCl3); 1H NMR (300 MHz) δ 1.38 (s, 3H), 1.48 (s, 3H),
2.16-2.18 (m, 1H), 2.22-2.33 (m, 1H), 2.39-2.56 (m, 2H), 4.59
(d, 1H, J ) 8.8 Hz), 5.06-5.22 (m, 2H), 5.78-5.94 (d, 1H, J )
8.8 Hz); 13C NMR (75 MHz) δ 26.8, 28.3, 39.5, 39.8, 80.8, 84.3,
94.0, 108.6, 118.6, 118.8, 132.5, 133.0, 133.5; IR (film) 1738
D
3H), 0.10 (s, 6H), 0.12 (s, 3H), 0.88 (s, 9H), 0.89 (s, 9H), 1.49-
1.57 (m, 2H), 1.61 (dd, 1H, J ) 12.5, 13.4 Hz), 1.99 (dd, 1H, J
) 4.6, 13.4 Hz), 2.30-2.32 (m, 2H), 2.88 (dd, 1H, J ) 8.2, 9.5
Hz), 2.97-3.05 (m, 1H), 3.44 (dd, 1H, J ) 5.8, 8.5 Hz), 3.70-
3.76 (m, 1H), 3.88 & 4.14 (ABq, 2H, J ) 14.3 Hz), 4.12 (1H,
dd, J ) 7.6, 8.5 Hz), 5.08-5.13 (m, 2H), 5.79-5.88 (m, 1H),
7.22-7.42 (m, 5H); 13C NMR (75 MHz) δ -4.3, -4.1, -2.0,
-1.9, 18.0, 18.2, 25.9, 25.9, 37.4, 39.3, 45.6, 48.4, 61.9, 70.4,
70.5, 72.2, 75.5, 118.4, 126.9, 128.2, 128.7, 133.9, 138.3; IR
(film) 2858, 1472 cm-1. Anal. Calcd for C29H51NO3Si2: C, 67.26;
H, 9.93; N, 2.70. Found: C, 66.95; H, 10.13; N, 3.00.
cm-1
.
This geminal dibromo olefin (5.00 g, 13.6 mmol) was also
treated in the similar way as that described for 1 f 24
employing water for quenching of the reaction in place of the
trimethylchlorosilane to give 35 (1.60 g, 57%). 35: [R]25D -31.6
(c 1.50, CHCl3); 1H NMR (300 MHz) δ 1.36 (s, 3H), 1.51 (s,
3H), 2.30-2.50 (m, 1H), 2.55-2.64 (m, 1H), 2.56 (d, 1H, J )
2.2 Hz), 4.61 (d, 1H, J ) 2.2 Hz), 5.09-5.20 (m, 2H), 5.75-
6.00 (m, 1H); 13C NMR (75 MHz) δ 26.6, 27.8, 39.6, 39.7, 71.4,
76.3, 78.3, 83.4, 108.9, 118.1, 118.4, 132.5, 133.1; IR (film) 3313
(shoulder), 3251, 2127, 1640 cm-1. Anal. Calcd for C13H18O2:
C, 75.69; H, 8.80. Found: C, 75.51; H, 9.01.
Physical data for (1S,3S,5S,6R)-isomer (19): [R]28 -61.4
D
(c 1.08, CHCl3); 1H NMR (500 MHz) δ 0.10 (s, 3H), 0.12 (s,
9H), 0.88 (s, 9H), 0.89 (s, 9H), 1.58 (dd, 1H, J ) 9.1, 13.7 Hz),
1.71-1.81 (m, 2H), 1.89 (dd, 1H, J ) 3.9, 13.7 Hz), 2.31-2.35
(m, 1H), 2.41-2.46 (m, 1H), 2.74-2.83 (m, 2H), 3.72 (dd, 1H,
J ) 5.2, 7.3 Hz), 3.93 & 4.02 (ABq, 2H, J ) 14.0 Hz), 4.04-
4.08 (m, 2H), 5.03-5.12 (m, 2H), 5.84-5.93 (m, 1H), 7.23-
7.42 (m, 5H); 13C NMR (75 MHz) δ -4.5, -4.4, -1.8, -1.7,
18.0, 18.3, 25.9, 25.9, 26.0, 36.4, 39.8, 43.7, 47.6, 61.9, 68.6,
70.7, 71.4, 75.7, 117.6, 127.0, 128.2, 128.2, 128.6, 128.7, 134.6,
(5R)-4-Allyl-4,5-bis-(ter t-bu tyld im eth ylsilyloxy)-1-h ep -
ten -6-yn e (36). To a solution of 35 (200 mg, 0.969 mmol) in
90% AcOH/MeOH (1:1, 4 mL) was added p-toluenesulfonic acid
hydrate (18 mg, 0.0969 mmol), and the mixture was stirred
at 80 °C for 14 h. After the reaction was quenched by the
addition of saturated aqueous NaHCO3 solution, the mixture
was extracted with several portions of EtOAc. The combined
EtOAc solutions were dried (MgSO4) and concentrated to give
an oil, which, on CC (hexane), afforded deprotected diol (145
138.1; IR (film) 2857, 1472 cm-1
.
Physical data for (1S,3R,5S,6S)-isomer (20): [R]28 -50.2
D
(c 1.33, CHCl3); 1H NMR (500 MHz) δ 0.10 (s, 3H), 0.11 (s,
3H), 0.12 (s, 3H), 0.14 (s, 3H), 0.88 (s, 9H), 0.89 (s, 9H), 1.39
(dd, 1H, J ) 12.8, 15.8 Hz), 1.45 (dd, 1H, J ) 10.6, 13.4 Hz),
1.66 (dt, 1H, J ) 2.1, 12.8 Hz), 1.90 (ddd, 1H, J ) 2.1, 4.3,
13.4 Hz), 2.27-2.35 (m, 2H), 2.39-2.44 (m, 1H), 2.69-2.79 (m,
1H), 3.53 (dd, 1H, J ) 6.4, 10.9 Hz), 3.78 & 4.48 (ABq, 2H, J
) 14.2 Hz), 3.90 (t, 1H, J ) 6.4 Hz), 4.16-4.22 (m, 1H), 5.05-
5.13 (m, 2H), 5.76-5.84 (m, 1H), 7.22-7.44 (m, 5H); 13C NMR
(75 MHz) δ -4.3, -4.2, -1.8, -1.7, 18.0, 18.4, 25.9, 26.0, 37.5,
42.5, 46.7, 47.8, 64.7, 68.9, 71.3, 75.5, 77.7, 118.5, 126.8, 128.2,
mg, 90%). (5R)-4-Allyl-4,5-dihydroxy-1-hepten-6-yne: [R]25
D
1
+4.90 (c 0.960, CHCl3); H NMR (300 MHz) δ 2.33-2.43 (m,
3H), 2.50-2.59 (m, 1H), 2.54 (d, 1H, J ) 2.2 Hz), 4.28 (d, 1H,
J ) 2.2 Hz), 5.12-5.23 (m, 2H), 5.81-6.00 (m, 1H); 13C NMR
(75 MHz) δ 39.6, 39.7, 75.3, 75.5, 81.7, 119.2, 119.3, 132.8,
132.9; IR (film) 3421, 3304, 2116 (very weak), 1726 cm-1
.
To a solution of the diol (320 mg, 1.92 mmol) in CH2Cl2 (5
mL) was added TBDMSOTf (1.77 mL, 4.0 equiv) followed by
the addition of Et3N (1.34 mL, 5 equiv) at 0 °C. The reaction
was continued at 0 °C to room temperature for 16 h, quenched
by the addition of saturated aqueous NaHCO3 solution, and
extracted with CH2Cl2 and successively with several portions
of AcOEt. The combined organic extracts were dried with
MgSO4 and concentrated to give a colorless oil, which, on CC
128.9, 133.7, 138.4; IR (film) 2857, 1472 cm-1
.
(6S )-4-Allyl-4,6-b is-(t er t -b u t yld im e t h ylsiloxy)-8-t r i-
m eth ylsilyl-1-octen -7-yn e (24). To a solution of aldehyde 1
(800 mg, 1.93 mmol) in CH2Cl2 were successively added CBr4
(1.92 g, 5.79 mmol), PPh3 (1.52 g, 5.79 mmol), and Zn (368
mg, 5.79 mmol) at 0 °C. The mixture was stirred at 0 °C to
room temperature for 13 h. The reaction was quenched by the
addition of water and filtered through a Celite pad, and the
filter cake was thoroughly rinsed with AcOEt. The combined
organic solutions were dried with Na2SO4 and concentrated
to give an oil, which, on CC (hexane), afforded (3S)-5-allyl-
(hexane), afforded 36 (720 mg, 95%). 36: [R]25 -5.91 (c 1.10,
D
1
CHCl3); H NMR (300 MHz) δ 0.21 (s, 3H), 0.23 (s, 3H), 0.25
(s, 3H), 0.27 (s, 3H), 0.97 (s, 9H), 1.00 (s, 9H), 2.37-2.68 (m,
4H), 2.48 (d, 1H, J ) 2.2 Hz), 4.30 (d, 1H, J ) 2.2 Hz), 5.10-
5.22 (m, 2H), 5.90-6.08 (m, 1H); 13C NMR (75 MHz) δ -5.0,
-4.1, -2.9, -1.8, 18.2, 18.7, 25.7, 26.0, 40.8, 41.3, 68.8, 74.6,
79.4, 83.8, 117.4, 117.9, 133.9, 134.4; IR (film) 3310, 2154, 1640
cm-1. Anal. Calcd for C22H42O2Si2: C, 72.07; H, 11.55. Found:
C, 71.86; H, 11.63.
1,1-dibromo-3,5-bis-(tert-butyldimethylsiloxy)-1,7-octadiene (940
1
mg, 86%): [R]24 -12.3 (c 1.70, CHCl3); H NMR (300 MHz) δ
D
0.10 (s, 3H), 0.14 (s, 3H), 0.15 (s, 3H), 0.16 (s, 3H), 0.87 (s,
9H), 0.89 (s, 9H), 1.65 (dd, 1H, J ) 4.9, 14.3 Hz), 1.77 (dd, 1H,
J ) 7.4, 14.3 Hz), 2.23-2.38 (m, 3H), 2.42-2.50 (m, 1H), 4.58
(ddd, 1H, J ) 4.9, 7.4, 8.5 Hz), 5.00-5.11 (m 2H), 5.81-5.96
(m, 1H), 6.38 (d, 1H, J ) 8.5 Hz); 13C NMR (75 MHz) δ -4.5,
-3.9, -1.5, -1.4, 17.9, 18.5, 25.8, 26.1, 45.0, 45.7, 45.9, 70.5,
76.6, 88.5, 117.6, 117.7, 134.5, 134.8, 142.5; IR (film) 1639
(5R)-4-Allyl-4,5-b is(ter t-b u t yld im et h ylsilyloxy)-7-t r i-
m eth ylsilyl-1-h ep ten -6-yn e (26). To a solution of 36 (450
mg, 1.13 mmol) in THF (10 mL) was added a solution of
butyllithium (1.50 M in hexane, 1.13 mL, 1.71 mmol) at 0 °C,
and the mixture was stirred at 0 °C for 15 min. To this solution
was added TMSCl (0.29 mL, 2.26 mmol) at 0 °C, and the
mixture was stirred at 0 °C for 30 min followed by the addition
of AcOEt and saturated aqueous NaHCO3 solution. The
mixture was extracted with several portions of EtOAc, and
the combined AcOEt solutions were dried (MgSO4) and con-
centrated to give an oil, which, on CC (hexane), afforded 26
cm-1
.
To a solution of this 1,1-dibromo olefin (940 mg, 1.65 mmol)
in THF (20 mL) was added n-BuLi (1.5 M in hexane, 2.75 mL,
4.13 mmol) at -78 °C. The reaction mixture was stirred for 1
h before TMSCl (0.53 mL, 4.2 mmol) was added to the mixture.
The mixture was stirred for 30 min at that temperature,
quenched by the addition of water, and extracted with AcOEt.
The combined extracts were dried with Na2SO4 and concen-
(403 mg, 89%): [R]26 -11.7 (c 0.93, CHCl3); 1H NMR (300
D
MHz) δ 0.11 (s, 3H), 0.13 (s, 3H), 0.15 (s, 9H), 0.16 (s, 3H),
0.17 (s, 3H), 0.89 (s, 9H), 0.91 (s, 9H), 2.24-2.58 (m, 4H), 4.18
(s, 1H), 5.01-5.10 (m, 2H), 5.80-5.97 (m, 1H); 13C NMR (75
MHz) δ -4.9, -4.0, -1.8, -1.6, -0.3, 18.2, 18.7, 25.9, 26.0,
40.9, 41.4, 69.2, 79.4, 91.0, 106.1, 117.2, 117.7, 134.1, 134.6;
IR (film) 2175 cm-1. Anal. Calcd for C25H50O2Si3: C, 64.31; H,
10.79. Found: C, 64.17; H, 10.99.
trated to give an oil, which, on CC (hexane), afforded 24 (540
1
mg, 69%): [R]25 -27.1 (c 0.80, CHCl3); H NMR (300 MHz) δ
D
0.11 (s, 3H), 0.12 (s, 3H), 0.14 (s, 9H), 0.15 (s, 3H), 0.16 (s,
3H), 0.88 (s, 18H), 1.92 (d, 2H, J ) 6.0 Hz), 2.29-2.46 (m,
4H), 4.64 (t, 1H, J ) 6.0 Hz), 4.98-5.09 (m, 2H), 5.79-5.95
(m, 1H); 13C NMR (75 MHz) δ -4.7, -4.0, -1.7, -0.3, -0.2,
18.1, 18.5, 25.8, 26.0, 44.8, 45.0, 47.8, 60.0, 89.0, 108.7, 117.4,
(1R,3S,5S)-3-Allyl-3,5-bis-(ter t-bu tyld im eth ylsilyloxy)-
7-(tr im eth ylsilyl)bicyclo[4.3.0]n on -6-en -8-on e (28). To a
solution of 24 (190 mg, 0.36 mmol) in CH2Cl2 (6 mL) was added
Co2(CO)8 (149 mg, 0.43 mmol) at 0 °C, and the mixture was
stirred at 0 °C to room temperature for 5 h. The mixture was
concentrated by a rotary evaporator to give an oil, which on
117.6, 134.6, 134.8; IR (film) 2172 cm-1
.
(5R)-4-Allyl-4,5-isopr opyliden edioxy-1-h epten -6-yn e (35).
The aldehyde 4 (3.55 g, 16.7 mmol) was subjected to olefination
reaction to give a geminal dibromo olefin (5.50 g, 88%) in the
same way as that described for 1 f 24. (3R)-1,1-Dibromo-4-

Discrimination of Diastereotopic Groups and Faces
J . Org. Chem., Vol. 66, No. 11, 2001 3847
CC (hexane), afforded cobalt-enyne complex (244 mg, 83%).
A solution of this complex (93 mg, 0.12 mmol) in acetonitrile
(10 mL) was heated at 75 °C for 18 h. Purple precipitates were
removed through a short silica gel pad, and the filtrate was
concentrated to give an oil, which, on CC (30:1 hexane/EtOAc),
afforded 28 (less polar, 37 mg, 59%) and 29 (more polar, 4 mg,
7%). Physical data for the major isomer 28: [R]25D +28.4 (c 1.70,
3H), 1.46 (s, 3H), 1.86 (dd, 1H, J ) 11.6, 19.5 Hz), 2.14 (ddd,
1H, J ) 0.9, 7.6, 14.6 Hz), 2.18 (d, 1H, J ) 20 Hz), 2.34-2.46
(m, 3H), 2.61-2.65 (m, 1H), 2.74 (td, 1H, J ) 7.0, 10.3 Hz),
3.06-3.14 (m, 1H), 4.29 (s, 1H), 5.10-5.16 (m, 2H), 5.80-5.88
(m, 1H); 13C NMR (75 MHz) δ 27.5, 28.1, 38.1, 38.8, 42.8, 43.5,
44.7, 47.3, 89.1, 93.6, 110.3, 118.7, 133.3, 218.6.
(1S,2R,3S,5R)-2,3-Dih yd r oxy-3-p r op ylb icyclo[3.3.0]-
octa n -7-on e (42). To a solution of 34 (78 mg, 0.25 mmol) in
EtOH (2 mL), placed in a hydrogenation vessel, was added a
spatula tip of 10% Pd-C. The reaction vessel was flushed with
hydrogen gas, and the mixture was vigorously stirred under
a positive pressure of hydrogen for 2 days. The mixture was
filtered through a Celite pad, and the filtrate was dried over
Na2SO4 and concentrated to give an oil, which, on CC (30:1
hexane/EtOAc), afforded hydrogenated product (9.3 mg, 16%).
To a solution of the product (9.3 mg, 0.04 mmol) in MeOH (2
mL) was added 2 N HCl (1 mL) at room temperature. The
reaction was continued at room temperature for 14 h, quenched
by the addition of NaHCO3, and extracted with AcOEt. The
combined extracts were dried with Na2SO4 and concentrated
to give a clear colorless oil, which, on CC (EtOAc), gave 42
(3.0 mg, 28%): 1H NMR (500 MHz) δ 0.96 (t, 3H), 1.35-1.66
(m, 5H), 1.76 (br, 2H), 2.16-2.26 (m, 2H), 2.34-2.43 (m, 1H),
2.49-2.62 (m, 2H), 2.67-2.77 (m, 1H), 2.90-3.01 (m, 1H), 3.75
(m, 1H); 13C NMR (75 MHz) δ 14.7, 16.9, 35.2, 37.1, 43.2, 43.8,
46.4, 47.0, 84.7, 85.7, 220.6.
(1S,2R,3R,5R)-2,3-Dih yd r oxy-3-p r op ylb icyclo[3.3.0]-
octa n -7-on e (41). To a solution of the mixture of 38 and 39
(4.1 mg, 0.0096 mmol) in THF (2 mL) was added TBAF (1 mL,
1.0 M in THF) at room temperature. The reaction was
continued at room temperature for 12 h, quenched by the
addition of water, and extracted with AcOEt. The combined
extracts were dried with Na2SO4 and concentrated to give a
clear colorless oil, which, on CC (EtOAc), gave the product (1.5
mg, 79%) as a mixture of diastereomeric diols 41 and 42. The
following list of physical data for the major isomer 41 was read
from those recorded for the mixture: 1H NMR (500 MHz) δ
0.95 (t, 3H), 1.30-1.50 (m, 5H), 1.90-1.96 (m, 1H), 2.01 (br,
1H), 2.13 (br, 1H), 2.21-2.26 (dd, J ) 8.5, 14.4 Hz, 1H), 2.27-
2.33 (m, 1H), 2.54-2.60 (dd, J ) 10.1, 19.5 Hz, 2H), 2.71-
2.78 (m, 1H), 2.88-2.97 (m, 1H), 3.51 (d, J ) 7.3 Hz, 1H).
1
CHCl3); H NMR (500 MHz) δ 0.00 (s, 3H), 0.03 (s, 3H), 0.06
(s, 3H), 0.11 (s, 3H), 0.24 (s, 9H), 0.83 (s, 9H), 0.88 (s, 9H),
1.14 (dd, 1H, J ) 11.9, 12.5 Hz), 1.68 (dd, 1H, J ) 3.0, 13.7
Hz), 1.91 (dd, 1H, J ) 1.8, 18.9 Hz), 2.15 (dt, 1H, J ) 2.7, 13.7
Hz), 2.32 (ddd, 1H, J ) 2.7, 4.8, 12.5 Hz), 2.48-2.53 (m, 1H),
2.53 (dd, 1H, J ) 6.7, 18.9 Hz), 2.93 (dd, 1H, J ) 9.1, 14.6
Hz), 3.05-3.11 (m, 1H), 4.98 (t, 1H, J ) 3.0 Hz), 5.04-5.11
(m,2H), 5.91-6.01 (m, 1H); 13C NMR (75 MHz) δ -4.3, -1.7,
-0.4, 1.0, 18.0, 18.1, 25.7, 25.8, 36.1, 42.3, 45.6, 46.9, 48.1,
66.9, 74.3, 117.1, 135.3, 136.1, 188.7, 213.6; IR (film) 1699
cm-1. Anal. Calcd for C27H52O3Si3: C, 63.72; H, 10.30. Found:
C, 63.51; H, 10.19.
Physical data for the minor (1S,3S,5S)-isomer (29): [R]20
D
+8.46 (c 0.78, CHCl3); 1H NMR (500 MHz) δ 0.14 (s, 6H), 0.16
(s, 6H), 0.24 (s, 9H), 0.92 (s, 9H), 0.94 (s, 9H), 1.18 (dd, 1H, J
) 12.8, 13.4 Hz), 1.55 (dd, 1H, J ) 11.9, 12.2 Hz), 1.88 (dd,
1H, J ) 1.8, 18.3 Hz), 2.02 (ddd, 1H, J ) 2.7, 4.6, 13.4 Hz),
2.15 (ddd, 1H, J ) 2.7, 4.9, 12.2 Hz), 2.21-2.27 (m, 1H), 2.39-
2.44 (m, 1H), 2.51 (dd, 1H, J ) 7.0, 18.3 Hz), 2.97-3.04 (m,
1H), 4.97 (dd, 1H, J ) 4.9, 11.9 Hz), 5.03-5.12 (m, 2H), 5.70-
5.79 (m, 1H); 13C NMR (75 MHz) δ -4.0, -3.0, -1.7, -1.6,
1.4, 1.5, 18.5, 18.8, 25.7, 26.7, 38.4, 41.5, 47.1, 47.5, 49.3, 72.0,
76.8, 118, 133, 192, 212; IR (film) 1699 cm-1
.
(1S,2R,3R,5R)-2,3-Bis(ter t-b u t yld im et h ylsilyloxy)-3-
p r op ylbicyclo[3.3.0]octa n -7-on e (38). To a solution of 32
(5 mg, 0.01 mmol: containing 33 in a 10:1 ratio) in EtOH (1
mL), placed in a hydrogenation vessel, was added a spatula
tip of 10% Pd-C. The reaction vessel was flushed with
hydrogen gas, and the mixture was stirred vigorously under
a positive pressure of hydrogen for 16 h. The mixture was
filtered through a Celite pad, and the filtrate was dried over
Na2SO4 and concentrated to give an oil, which, on CC (30:1
hexane/EtOAc), afforded 38 + 39 (4.2 mg, 97%). The following
list of physical data for the major isomer 38 was read from
those recorded for the mixture: 1H NMR (500 MHz) δ 0.04 (s,
3H), 0.06 (s, 6H), 0.08 (s, 3H), 0.85 (s, 9H), 0.88 (s, 9H), 0.90-
0.92 (m, 3H), 1.20-1.62 (m, 5H), 2.12 (ddd, 1H, J ) 1.2, 4.9,
19 Hz), 2.21 (dd, 1H, J ) 8.5, 13.1 Hz), 2.29 (dd, 1H, J ) 1.2,
16 Hz), 2.48-2.56 (m, 3H), 2.65-2.73 (m, 1H), 3.75 (d, 1H, J
) 5.5 Hz).
(1S,2R,3S,5R)-3-Allyl-2,3-(isopr opyliden edioxy)bicyclo-
[3.3.0]octa n -7-on e (40). To a mixture of [(Ph3P)CuH]6 (191
mg, 0.097 mmol), weighted under nitrogen atmosphere, and
34 (30 mg, 0.09 mmol) was added benzene containing water,
and the resulting red solution was stirred at room temperature
for 14 h. The mixture was opened to air, and stirring was
continued for additional 2 h. The mixture was filtered through
a Celite pad and the filtrate was concentrated to give an oil,
which, on CC (20:1 hexane/EtOAc), afforded 40 (21 mg, 91%):
1H NMR (300 MHz) δ 1.31 (dd, 1H, J ) 11.6, 14.6 Hz), 1.39 (s,
Ack n ow led gm en t. This work was supported by a
Grant-in-Aid for Scientific Research from the Ministry
of Education, Science, Sports, and Culture, J apan. We
are also grateful to the “SC-NMR Laboratory of Okaya-
ma University” for high-field NMR experiments.
Su p p or tin g In for m a tion Ava ila ble: Copies of 1H and/
or 13C NMR spectra for compounds 1-4, 11-16, 18-23, 28-
32, 34, and 37-42. Synthetic procedures and physical data
for compounds 2-4, 13-16, 21-23, 25, 27, and 30-34. This
material is available free of charge via the Internet at
http://pubs.acs.org.
J O0017638