Stereocontrol in organic synthesis using silicon-containing compounds. A synthesis of a (±)-carbacyclin analogue with the geometry of the exocyclic double bond controlled by the protodesilylation of an allylsilane

Article abstract of DOI:10.1039/a804274h

The known ketone, 7-tert-butyldimethylsilyloxybicyclo[3.3.0]oct-8-en-2-one 11, was converted in five steps into 3-(4′-methoxycarbonylbutylidene)-7-tert-butyldimethylsilyloxy-8- cyanobicyclo[3.3.0]octan-2-one 13. Reduction gave the diastereoisomeric pair of allylic alcohols 14 and 15, both of which were converted into the allylsilane, 3-(1′-dimethylphenylsilyl-4′-methoxycarbonyl)butyl-7-tert- butyldimethyl-silyloxy-8-cyanobicyclo[3.3.0]oct-2-ene 20. Protodesilylation of the allylsilane gave a high level of selectivity (>96:4) in favour of the carbacyclin analogue, 5-(4′-methoxycarbonyl)butylidene-3-tert-butyldimethylsilyloxy-2- cyanobicyclo[3.3.0]octane 22, having the exocyclic double bond with the E-configuration.

Full text of DOI:10.1039/a804274h

View Article Online / Journal Homepage / Table of Contents for this issue
Stereocontrol in organic synthesis using silicon-containing
compounds. A synthesis of a ( )-carbacyclin analogue with
the geometry of the exocyclic double bond controlled by the
protodesilylation of an allylsilane
Ian Fleming* and Dick Higgins
Department of Chemistry, Lensfield Road, Cambridge, UK CB2 1EW
The known ketone, 7-tert-butyldimethylsilyloxybicyclo[3.3.0]oct-8-en-2-one 11, was converted in five
steps into 3-(4Ј-methoxycarbonylbutylidene)-7-tert-butyldimethylsilyloxy-8-cyanobicyclo[3.3.0]octan-
2-one 13. Reduction gave the diastereoisomeric pair of allylic alcohols 14 and 15, both of which were
converted into the allylsilane, 3-(1Ј-dimethylphenylsilyl-4Ј-methoxycarbonyl)butyl-7-tert-butyldimethyl-
silyloxy-8-cyanobicyclo[3.3.0]oct-2-ene 20. Protodesilylation of the allylsilane gave a high level of
selectivity (>96:4) in favour of the carbacyclin analogue, 5-(4Ј-methoxycarbonyl)butylidene-3-tert-
butyldimethylsilyloxy-2-cyanobicyclo[3.3.0]octane 22, having the exocyclic double bond with the
E-configuration.
R
SiMe2Ph
R
SiMe2Ph
Introduction
We and several other groups established that the stereo-
chemistry of electrophilic attack on the double bond of an
allylsilane was stereospecifically anti, usually in the sense 1,1–4
giving an alkene 3 with the R group and the A group cis, but
with more or less of the reaction taking place from the altern-
ative conformation 2, creating an alkene 4 with the R group
and the A group trans (Scheme 1). Most attention has naturally
5
6
R
R
E+
+
Si
H
A
A
H
H
H
R
R
Si
E+
7
8
1
2
7:8
(from 5)
7:8
(from 6)
E
R = Pri 90:10
R = Me 48:52
12:88
57:43
H
A
A
R
H
H
E
R
H
Scheme 2 Reagent: i, CF3CO2H, CDCl3, 0 ЊC
3
4
McClure,4 and of Curran and Kim,7 establishing that electro-
philic attack on the double bond of an allylsilane often takes
place mainly from conformation 2 when the substituent on the
double bond cis to the stereogenic centre is only a hydrogen
atom, as here, and when, at the same time, the R group is
methyl. A methyl group suffers only a weak A1,3 repulsion with
the H atom on the double bond, but an isopropyl group suffers
a much larger repulsion. The problem of having a small group
like methyl on the stereogenic centre giving mixtures of geo-
metrical isomers in the double bond is not, of course, restricted
to allylsilane reactions—it turns up equally with allyl Grignard
reactions, where selectivities (E:Z) ranging from 5:1 to 0.2:1
have been observed, depending upon the electrophile.8
At first sight this limitation does not bode well for our plan
to use the protodesilylation of an allylsilane to control the
geometry of the exocyclic double bond in a synthesis of a carb-
acyclin. E-Carbacyclin 99 is a prostacyclin analogue much more
potent in inhibiting platelet aggregation than its Z-isomer 10.10,11
Controlling the geometry of the exocyclic double bond in this
molecule is therefore a significant challenge in synthesis, both
because the geometry matters, and because it is not obvious
how to set it up stereoselectively, given that it is remote from
Scheme 1
been paid to the stereochemistry at the new stereogenic
centre,1,2 which is opposite in sense in the two products 3 and
4, but it is sometimes possible to use the reaction to control
double bond geometry. For example, dihydroxylation using
osmium tetroxide, followed by convergent stereospecific elimin-
ation, can be made to give either the cis or the trans alkene.3,4
Alternatively, stereocontrol might be achieved by arranging
for a high proportion of the reaction to take place in conform-
ation 1.
To test this possibility, we investigated the protodesilylation
of allylsilanes, creating a double bond exocyclic to a five-
membered ring.5 In summary, we found that protodesilylation
of the allylsilanes 5 took place mainly in the sense 1 → 3 to
give the E-alkenes 7, provided that the group R was moderately
large. Similarly, protodesilylation of the allylsilanes 6 gave the
Z-alkenes 8 (Scheme 2). Thus when R was isopropyl, the select-
ivity for reaction taking place in the sense 1 over 2 was close to
90:10, both for 5 and for 6, but when R was a methyl group the
degree of selectivity was low and actually reversed in both
cases, consistent with our earlier work,2,6 and that of Vedejs and
J. Chem. Soc., Perkin Trans. 1, 1998
2673

View Article Online
CO2H
HO2C
CO2Me
O
O
O
H
i
ii, iii, iv, v
H
H
H
H
H
H
H
CN
CN
OH
OH
OH
OH
OTBDMS
OTBDMS
OTBDMS
9
10
11
12 71%
13 25%
vi
any resident steric influence. Although a few stereoselective syn-
theses have appeared,12 the original Wittig route gave, as one
might expect, both isomers in essentially equal amounts.11,13
Our own approach, based on the model reaction 5 → 7,
would seem to be doomed—the methylene chain is likely to be
more like a methyl group in its steric effect than an isopropyl,
and is therefore unlikely to impart useful levels of stereocontrol.
Nevertheless, we carried out the synthesis, for we were aware
that other factors in our favour would come into play, and
so they did. We reported this work in preliminary communi-
cations,14 and report it in full here.
CO2Me
CO2Me
OH
H
OH
H
+
H
H
CN
CN
OTBDMS
OTBDMS
14 80%
15 8%
vii
viii
vii
Results and discussion
We chose not to deal with the problem of the lower side chain,
which is, in any case, the main site of variation in the molecules
being tested for clinical usefulness, and assembled instead a
cyano analogue. We started from the known, racemic unsatur-
ated ketone 11,15 which was attacked exo on the bicyclic frame-
work by cyanide ion to give the nitrile 12. This addition took
place under milder conditions than usual for this conventional
procedure,16 as befits a bicyclooctenone with a bridgehead
double bond. More recent methods for the conjugate addition
of cyanide ion, based on trimethylsilyl cyanide in the presence
of Lewis acids,17 were too harsh, giving elimination products
lacking the silyloxy group or even more extensive decom-
position. A conventional aldol sequence then gave the ketone
13, with the usual E-geometry, but with an unimpressive overall
yield, although this was only to be expected from an earlier
report on aldol reactions on this type of bicyclooctanone.18 We
were even less successful with phenylthioalkylation19 of the silyl
enol ether derived from the ketone 12.
CO2Me
CO2Me
OAc
H
OAc
H
H
H
CN
CN
OTBDMS
OTBDMS
16 93%
17 95% from 15
70% from 14
Scheme 3 Reagents: i, KCN, NH4Cl, H2O, DMF; ii, LDA; iii,
MeO2C(CH2)3CHO; iv, MeSO2Cl, Et3N; v, DBU; vi, NaBH4, CeCl3; vii,
Ac2O, Et3N, DMAP; viii, AcOH, EtO2CN᎐NCO2Et, Ph3P
᎐
CO2Me
CO2Me
CO2Me
SiMe2Ph
Luche reduction20 of the ketone 13 gave a separable mixture
of the alcohols 14 and 15 in a ratio of 10:1, and acetylation
gave the acetates 16 and 17, respectively. The latter, the more
important isomer, could also be made by a Mitsunobu reaction
from the major alcohol 14 (Scheme 3). Acetic acid is not usually
the best carboxylic acid to use in Mitsunobu reactions,21 but it
worked well enough here.
OAc
H
SiMe2Ph
i
+
H
H
H
H
H
CN
CN
CN
OTBDMS
OTBDMS
OTBDMS
19 6%
16
18 58%
The acetate 16 reacted with our silylcuprate reagent, using
the solvent mixture that we had found best for secondary allylic
acetates,22 giving a mixture of the regioisomeric allylsilanes 18
and 19, with the latter, unfortunately, as the minor product,
just as we had found earlier in the model series.22 In welcome
contrast, the acetate 17 gave only the allylsilane 20, which was
discernibly different from the allylsilane 19 (Scheme 4). By
analogy with our earlier work,3 these reactions can be relied
upon to be stereospecifically anti. Presumably an allylsilane
analogous to 18, but with an endo silyl group, was not formed
from the acetate 17, because it would have involved endo attack
on the bicyclo[3.3.0]octane system. Because of the success of
this reaction, we were not obliged to use our other device for
ensuring high levels of regioselectivity—the use of a silyl-
cuprate reagent assembled on a carbamate group in place of the
acetate.22 This was fortunate, for in this case we found that the
N-phenyl carbamate derived from the alcohol 14 reacted with
the silylcuprate reagent, assembled on the carbamate as usual,
to give a low yield (12%) of the allylsilane 19 instead of 20.
CO2Me
CO2Me
SiMe2Ph
OAc
H
i
H
H
H
CN
CN
OTBDMS
OTBDMS
20 86%
17
Scheme 4 Reagent: i, (PhMe2Si)2CuLiؒLiCN, THF, Et2O, pentane
This appears to be an unprecendented anti SN2Ј reaction of a
cuprate, but it is probable that it is a result of a syn 1,3-shift of
the carbamate anion, away from the hindered endo position on
the bicyclic framework, followed by an SN2 reaction, with the
2674
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
usual inversion. We did not pursue this problem, since we
had by this time found the Mitsunobu alternative for achieving
convergence.
Protodesilylation of the allylsilane 18 merely moved the
double bond regioselectively into the ring 21, as expected,23 but
the protodesilylations of the two allylsilanes 19 and 20 were not
equally stereoselective with respect to the double bond geom-
etry exocyclic to the ring. Protodesilylation of the diastereo-
isomer 19 gave a mixture of the E- and Z-isomers 22 and 23
(Scheme 5) in a ratio of about 2:1, which looked, on the face of
H+
H+
NC
TBDMSO
NC
SiMe2Ph
H
TBDMSO
H
SiMe2Ph
H+
CO2Me
MeO2C
MeO2C
24
25
H+
H+
NC
TBDMSO
NC
SiMe2Ph
H
CO2Me
CO2Me
TBDMSO
SiMe2Ph
H
H+
CO2Me
SiMe2Ph
26
27
i
Scheme 6
H
H
CN
H
H
CN
present in as high a concentration, can be protonated in the
stereoelectronically favourable sense, that is exo on the bi-
cyclic system and anti to the silyl group. With the allylsilane
20 that does give good stereocontrol, everything is favourable
for the formation of the E-isomer 22: the more populated con-
formation 26 is protonated exo on the bicyclic system and anti
to the silyl group. The higher-energy conformation 27 of this
diastereoisomer, if it were to give any of the Z-isomer 23, would
have to be protonated either endo on the bicyclic system or syn
to the silyl group, and the combination of unfavourable factors
effectively suppresses this pathway.
We note finally that the convergent synthesis of the acetate
17, the stereospecific and highly regioselective synthesis of the
allylsilane 20, and the highly stereospecific protodesilylation
conspire to make the four-step sequence from the enone 13 to
the carbacyclin analogue 22 reasonably efficient (50%).
OTBDMS
OTBDMS
18
21 97%
CO2Me
CO2Me MeO2C
SiMe2Ph
i
+
H
H
CN
H
H
H
H
CN
OTBDMS
22 67:33 90%
CN
OTBDMS
OTBDMS
19
23
Experimental
CO2Me
CO2Me
SiMe2Ph
Ether refers to diethyl ether.
i
(5SR,7RS)-7-tert-Butyldimethylsilyloxybicyclo[3.3.0]oct-8-en-
2-one 11
H
H
CN
H
H
1,2-Epoxycyclooct-3-ene, prepared in 89% yield from cycloocta-
1,3-diene (160 mmol) was converted into bicyclo[3.3.0]oct-7-
en-2-ol (71%) using lithium diethylamide, and into the bicyclic
ketone using chromic acid.25 Treatment with N-bromosuccin-
imide gave the bromohydrin, which was protected as its tert-
butyldimethylsilyl ether. β-Elimination of the bromide using
DBU, and chromatography (SiO2, hexane–EtOAc, 4:1) gave
the ketone (29% overall based on the epoxide), which had 13C
NMR and IR in agreement with the literature;15 δH(CDCl3) 6.35
CN
OTBDMS
OTBDMS
20
22 >96:4 92%
Scheme 5 Reagent: i, CF3CO2H, CH2Cl2, 0 ЊC
it, very much what one might have expected from the model
series (Scheme 2) if we assume that a methylene chain is a little
more effective than a methyl group in causing the conformation
1 rather than 2 to be populated. However, the allylsilane 20 gave
very largely (>96:4, GC supported by 13C NMR) the E-isomer
22, exactly as we had hoped. The assignment of configuration
to the alkenes 22 and 23 was easy by comparison of the 13C
NMR spectra with those reported for the E- and Z-isomers of
the carbacyclin iloprost.24 Although our work with the model
compounds would suggest that a methylene chain will not be an
effective group in controlling the double bond geometry, we
have somehow achieved an extraordinarily high level of control
with one of the diastereoisomers but not the other.
The allylsilane 19 that does not give good stereocontrol will
probably adopt most readily a conformation close to that
shown as 24, with the hydrogen atom eclipsing, or partly eclips-
ing, the double bond (Scheme 6). Protodesilylation in this con-
formation will lead to the E-isomer 22, but it will involve attack
by the proton either syn to the silyl group or endo on the bicyclic
system. Since neither of these pathways is likely to be favour-
able, it is not surprising that this diastereoisomer does not lead
cleanly to the E-carbacyclin 22. The alternative conformation
25, which will lead to the Z-isomer 23, although probably not
(1 H, dd, J 3 and 2, HC᎐C), 5.20 (1 H, ddt, J 8, 7 and 2, HCOSi),
᎐
3.06–2.98 (1 H, m, bridgehead H), 2.72 (1 H, dt, J 8 and 5, endo
CHAHBCO), 2.53–2.22 (3 H, m, exo CHAHBCO, endo Hs),
1.65–1.41 (2 H, m, exo Hs), 0.81 (9 H, s, Me3C) and 0.05 (6 H, s,
Me2Si).
(1SR,5SR,7RS,8SR)-7-tert-Butyldimethylsilyloxy-8-cyano-
bicyclo[3.3.0]octan-2-one 12
Potassium cyanide (26.0 mmol) was added to a solution of the
ketone 11 (20.0 mmol) and ammonium chloride (20.0 mmol) in
a mixture of DMF (90 cm3) and water (10 cm3) at 60 ЊC and
stirred for 4 h. The mixture was diluted with ether and washed
with aqueous ammonium chloride and brine. The organic layer
was dried (MgSO4) and evaporated under reduced pressure to
give a yellow oil. Chromatography (SiO2, CH2Cl2) gave the
ketone (71%); Rf(CH2Cl2) 0.25; νmax(film)/cmϪ1 2240 (C᎐N) and
᎐
᎐
1740 (C᎐O); δ (CDCl ) 4.41 (1 H, dt, J 5 and 2.5, HCOSi),
᎐
H
3
᎐
3.15–3.01 (2 H, m, HCC᎐N and endo CH H C᎐O), 2.9 (1 H,
᎐
᎐
B
A
dd, J 9 and 2, bridgehead CHC᎐O), 2.49–2.17 (4 H, m, exo
᎐
CH H C᎐O, endo Hs and bridgehead H), 1.89 (1 H, dddd, J 10,
᎐
B
A
8, 5 and 4, exo H β to C᎐O), 1.60 (1 H, ddd, J 15, 2.5 and 1.5,
᎐
J. Chem. Soc., Perkin Trans. 1, 1998
2675

View Article Online
exo H), 0.80 (9 H, s, Me3C) and 0.04 (6 H, s, Me2Si); δC(CDCl3)
mmol) and the ketone (1.5 mmol) were stirred in methanol (2.5
cm3) at 0 ЊC for 5 min. Aqueous ammonium chloride was added
and the mixture extracted with ether. The ether was washed
with brine, dried (MgSO4) and evaporated under reduced pres-
sure. The residue was chromatographed (SiO2, hexane–EtOAc,
2:1) to give the alcohol 14 (80%); Rf(hexane–EtOAc, 2:1) 0.33;
᎐
216.25 (C᎐O), 119.84 (C᎐N), 77.44 (COSi), 54.15, 41.18, 41.56,
᎐
᎐
37.10, 36.65, 26.57 (3 × CH2, 3 × CH), 25.44 (Me3C), 17.74
(Me3C), Ϫ5.21 (SiMeAMeB), Ϫ5.26 (SiMeAMeB); m/z 264 (17%,
M Ϫ Me), 222 (100%, M Ϫ But) (Found: Mϩ Ϫ 15, 264.1439.
C14H22NO2Si requires M, 264.1420). The expected exo configur-
ation of the cyano group was supported by the small coupling
constant (2.5) between the proton adjacent to the silyloxy group
at δ 4.41 and the proton adjacent to the cyano group.
νmax(CH2Cl2)/cmϪ1 3500 (OH), 2230 (C᎐N), 1740 (C᎐O) and
᎐
᎐
᎐
1650 (C᎐C); δ (CDCl ) 5.50 (1 H, t, J 7, HC᎐C), 4.48 (1 H, d,
᎐
᎐
H
3
J 6, HCOH), 4.27 (1 H, td, J 8 and 6, HCOSi), 3.66 (3 H, s,
Me), 2.80–2.72 (2 H, m, ring Hs), 2.58–2.48 (1 H, m, ring H),
2.36–2.11 (3 H, m, ring Hs), 2.30 (2 H, t, J 7, CH2COOMe),
Methyl 5-hydroxypentanoate
Sulfuric acid (conc., 0.1 cm3) was added to a stirred solution
of δ-valerolactone (120 mmol) in methanol (250 cm3) and the
mixture refluxed for 4 h. The mixture was cooled to 0 ЊC and
sodium hydrogen carbonate (1.50 g) added. After the mixture
had been stirred for 10 min, it was filtered through Celite and
the solvent removed under reduced pressure to give the ester
(94%) identical (1H NMR) to that reported.26
2.07 (2 H, q, J 7, CH CH C᎐C), 1.71 (2 H, quintet, J 7, CH -
᎐
2
2
2
CH C᎐C), 1.30–1.19 (1 H, m, ring H), 0.88 (9 H, s, Me C),
᎐
2
3
0.12 (3 H, s, SiMeAMeB) and 0.08 (3 H, s, SiMeAMeB); m/z 393
(4%, Mϩ), 336 (51, M Ϫ But) and 318 (100, M Ϫ But Ϫ H2O)
(Found: Mϩ, 393.2316. C21H35NO4Si requires M, 393.2335),
and the alcohol 15 (8%); Rf(hexane–EtOAc, 2:1) 0.20; νmax(CH2-
Cl2)/cmϪ1 3500 (OH), 2230 (C᎐N), 1740 (C᎐O) and 1650 (C᎐C);
᎐
᎐
᎐
᎐
δ (CDCl ) 5.59 (1 H, t, J 7, HC᎐C), 4.27 (1 H, s, HCOH), 4.22
(1 H, td, J 8 and 6, HCOSi), 2.81–2.48 (3 H, m, ring Hs), 2.40
(3 H, m, ring Hs), 2.31 (2 H, t, J 7, CH2COOMe), 2.25 (2 H, q,
᎐
H
3
Methyl 5-oxopentanoate
The ester (80 mmol) was added to a solution of pyridinium
chlorochromate (120 mmol) in dichloromethane (350 cm3) at
0 ЊC and allowed to warm to room temperature. The mixture
was stirred for 2 h and diluted with ether, filtered through Celite
and solvent removed under reduced pressure. The residue was
redissolved in ether and the filtration process repeated to
remove the remaining chromium residues. The product was
purified by column chromatography (hexane–EtOAc, 3:1) to
J 7, CH CH C᎐C), 1.72 (2 H, quintet, J 7, CH CH ), 1.27–1.16
᎐
2
2
2
2
(1 H, m, ring H), 0.87 (9 H, s, Me3C), 0.11 (3 H, s, SiMeAMeB)
and 0.08 (3 H, s, SiMeAMeB); m/z 336 (21%, M Ϫ But) and
75 (100) (Found: Mϩ Ϫ 57, 336.1637. C21H35NO4Si requires
M Ϫ 57, 336.1631).
(E)(1SR,2RS,5SR,7RS,8SR)-3-(4Ј-Methoxycarbonylbutyl-
idene)-7-tert-butyldimethylsilyloxy-8-cyanobicyclo[3.3.0]octan-
2-yl acetate 16
1
give the aldehyde (57%), identical (IR and H NMR) to the
known compound.26
The alcohol 14 (0.6 mmol), acetic anhydride (0.7 mmol), tri-
ethylamine (0.7 mmol) and N,N-dimethylaminopyridine (0.7
mmol) were kept in dichloromethane (1 cm3) for 2 h at room
temperature. Aqueous ammonium chloride was added and the
mixture extracted with ether. The ether was washed with brine,
dried (MgSO4) and evaporated under reduced pressure. The
residue was chromatographed (SiO2, hexane–EtOAc, 3:1) to
give the acetate (93%); Rf(2:1 hexane–EtOAc) 0.47; νmax(CH2-
(E)(1SR,5SR,7RS,8SR)-3-(4Ј-Methoxycarbonylbutylidene)-7-
tert-butyldimethylsilyloxy-8-cyanobicyclo[3.3.0]octan-2-one 13
LDA (1.1 mol dmϪ3 in THF, 6.0 cm3) was added dropwise to
a solution of the ketone 12 (6.0 mmol) in THF (5 cm3) under
argon at Ϫ78 ЊC and stirred for 20 min. Methyl 5-oxopentano-
ate (9.0 mmol) was added and the solution stirred for a further
3 min, after which acetic acid (6.6 mmol) was added. The
mixture was warmed to room temperature and then diluted
with ether, washed with ammonium chloride, dried (MgSO4)
and evaporated under reduced pressure. The residue was dis-
solved in dichloromethane, triethylamine (12.0 mmol) and
methanesulfonyl chloride (12.0 mmol) were added at 0 ЊC, and
the mixture was stirred for 1 h. DBU (12 mmol) was then added
and the mixture stirred for a further 1 h, allowing it to warm to
room temperature. After dilution with ether, the mixture was
washed with aqueous ammonium chloride and brine, dried
(MgSO4) and evaporated under reduced pressure. The residue
was chromatographed (SiO2, EtOAc–hexane, 1:2) to give the
Cl2)/cmϪ1 2230 (C᎐N), 1740 (C᎐O × 2) and 1650 (C᎐C);
᎐
᎐
᎐
᎐
δH(CDCl3) 5.50 (1 H, dd, J 8 and 1, HCOAc), 5.41 (1 H, tq, J 7
and 2, HC᎐C), 4.18 (1 H, td, J 9 and 7, HCOSi), 3.66 (3 H, s,
᎐
MeO), 2.86 (1 H, dt, J 10 and 8, CHCHOAc), 2.49–2.41 (2 H,
m, ring Hs), 2.32–2.03 (3 H, m, ring Hs), 2.30 (2 H, t, J 7,
CH COOMe), 2.17 (3 H, s, MeC᎐O), 2.08 (2 H, q, J 7,
᎐
2
CH CH C᎐C), 1.71 (2 H, quintet, J 7, CH CH C᎐C), 1.25–1.17
᎐
᎐
2
2
2
2
(1 H, m, ring H), 0.87 (9 H, s, Me3C), 0.11 (3 H, s, SiMeAMeB)
and 0.07 (3 H, s, SiMeAMeB); m/z 420 (7%, M Ϫ Me) and 378
(100, M Ϫ But) (Found: Mϩ Ϫ 15, 420.2218. C23H37NO5Si
requires M Ϫ 15, 420.2206).
ketone (25%); Rf(EtOAc–hexane, 1:2) 0.41; νmax(film)/cmϪ1
᎐
2230 (C᎐N), 1740 (COOMe), 1720 (C᎐O) and 1650 (C᎐C);
᎐
᎐
᎐
(E)(1SR,2SR,5SR,7RS,8SR)-3-(4Ј-Methoxycarbonylbutyl-
idene)-7-tert-butyldimethylsilyloxy-8-cyanobicyclo[3.3.0]octan-
2-yl acetate 17
δ (CDCl ) 6.53 (1 H, tt, J 8 and 3, HC᎐C), 4.43 (1 H, q, J 4,
᎐
H
3
HCOSi), 3.67 (3 H, s, MeO), 3.14–2.81 (4 H, m, ring Hs) and
2.51–2.29 (2 H, m, ring Hs), 2.33 (2 H, t, J 7, CH2COOMe),
Method A. The alcohol 15 was acetylated similarly to give the
2.15 (2 H, q, J 7, CH CH C᎐C), 1.80 (2 H, quintet, J 7, CH -
᎐
2
2
2
acetate (95%); Rf(hexane–EtOAc, 3:1) 0.25; νmax(CH2Cl2)/cmϪ1
CH C᎐C), 1.66–1.50 (1 H, m, remaining ring H), 0.77 (9 H, s,
᎐
2
᎐
2230 (C᎐N), 1740 (C᎐O × 2) and 1650 (C᎐C); δ (CDCl ) 5.67
᎐
᎐
᎐
H
3
Me3C), 0.02 (3 H, s, SiMeAMeB) and 0.01 (3 H, s, SiMeAMeB);
(1 H, t, J 7, HC᎐C), 5.27 (1 H, s, HCOAc), 4.23 (1 H, dt, J 8 and
᎐
δ (CDCl ) 203.69 (C᎐O), 173.45 (COOMe), 137.38 (C᎐C),
᎐
᎐
C
3
7, HCOSi), 3.65 (3 H, s, MeO), 2.77–2.03 (7 H, m, 2 bridgehead
᎐
136.97 (C᎐C), 120.18 (C᎐N), 77.31 (COSi), 51.58 (MeO), 55.36,
᎐
᎐
Hs, 3 ring Hs and CH CH C᎐C), 2.29 (2 H, t, J 7, CH -
᎐
2
2
2
42.80, 42.06, 33.65, 33.39, 32.36, 29.02, 23.40 (5 × CH2,
3 × CH), 25.37 (Me3C), 17.70 (Me3C), Ϫ5.22 (SiMeAMeB) and
Ϫ5.27 (SiMeAMeB); m/z 376 (13%, M Ϫ Me) and 334 (100,
M Ϫ But) (Found: Mϩ Ϫ 15, 376.1968. C21H33NO4Si requires
M Ϫ 15, 376.1944).
COOMe), 2.04 (3 H, s, MeC᎐O), 1.70 (2 H, quintet, J 7,
᎐
CH CH C᎐C), 1.62–1.23 (2 H, m, ring Hs), 0.87 (9 H, s, Me C),
᎐
2
2
3
0.12 (3 H, s, SiMeAMeB) and 0.08 (3 H, s, SiMeAMeB); m/z 420
(4%, M Ϫ Me), 378 (52, M Ϫ But), 117 (100) (Found: Mϩ Ϫ 15,
420.2220. C23H37NO5Si requires M Ϫ 15, 420.2206).
(E)(1SR,2RS,5SR,7RS,8SR)-3-(4Ј-Methoxycarbonylbutyl-
idene)-7-tert-butyldimethylsilyloxy-8-cyanobicyclo[3.3.0]octan-
2-ol 14 and (E)(1SR,2SR,5SR,7RS,8SR)-3-(4Ј-methoxy-
carbonylbutylidene)-7-tert-butyldimethylsilyloxy-8-cyano-
bicyclo[3.3.0]octan-2-ol 15
Method B. Diethyl azodicarboxylate (4.8 mmol) was added
to a solution of the alcohol 14 (1.6 mmol), triphenylphosphine
(4.8 mmol) and acetic acid (4.8 mmol) in ether (2 cm3) at room
temperature and stirred for 8 h. The mixture was washed with
aqueous sodium hydrogen carbonate and brine, dried (Mg-
SO4) and evaporated under reduced pressure. The residue was
Sodium borohydride (1.67 mmol), cerium() chloride (1.7
2676
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
chromatographed (SiO2, hexane–EtOAc, 3:1) to give the acet-
ate (70%), identical (TLC, IR, 1H NMR) with the other sample.
The allylsilane 19 (12%) was the only identifiable product
from the reaction of (E)(1SR,2SR,5SR,7RS,8SR)-3-(4Ј-meth-
oxycarbonylbutylidene)-7-tert-butyldimethylsilyloxy-8-cyano-
bicyclo[3.3.0]octan-2-yl N-phenyl carbamate with, successively,
butyllithium, copper(I) iodide, triphenylphosphine and the silyl-
lithium reagent for 2 h at 0 ЊC.22
(E)(1SR,2RS,5SR,7RS,8SR)-3-(4Ј-Methoxycarbonylbutyl-
idene)-7-tert-butyldimethylsilyloxy-8-cyanobicyclo[3.3.0]octan-
2-yl N-phenyl carbamate
n-Butyllithium (1.6 mol dmϪ3 in hexane, 0.1 cm3) was added to
a solution of the alcohol 14 (0.155 mmol) in THF (1.0 cm3)
under argon at Ϫ78 ЊC and stirred for 10 min. Phenyl isocyan-
ate (0.18 mmol) was added and the mixture warmed to room
temperature over 1 h. The mixture was diluted with ether,
washed with aqueous ammonium chloride and brine, dried
(MgSO4) and solvent evaporated under reduced pressure. The
residue was chromatographed (SiO2, hexane–EtOAc, 3:1) to
give the carbamate (74%); Rf(hexane–EtOAc, 3:1) 0.27;
νmax(CH2Cl2)/cmϪ1 3300 (NH), 1740 (C᎐O), 1690 (C᎐O), 1620
(E)(1ЈSR,1RS,5SR,7RS,8SR)-3-(1Ј-Dimethylphenylsilyl-4Ј-
methoxycarbonyl)butyl-7-tert-butyldimethylsilyloxy-8-cyano-
bicyclo[3.3.0]oct-2-ene 20
The allylic acetate 17 (0.23 mmol) was similarly treated with the
silylcuprate reagent to give the allylsilane 20 (86%); Rf(CH2Cl2)
0.43; νmax(CH2Cl2)/cmϪ1 2230 (C᎐N), 1740 (C᎐O) and 1630
᎐
᎐
᎐
(C᎐C); δ (CDCl ) 5.12 (1 H, d, J 2, HC᎐C), 4.07 (1 H, td, J 9
᎐
᎐
H
3
and 6, HCOSi), 3.63 (3 H, s, MeO), 3.13 (1 H, td, J 9 and 2,
allylic bridgehead H), 2.51–1.13 (9 H, m, ring Hs and CH2-
CHSi), 2.37 (2 H, t, J 7, CH2COOMe), 1.45 (2 H, quintet,
CH2CH2COOMe), 0.90 (9 H, s, Me3C), 0.29 (3 H, s, SiMeA-
᎐
᎐
(C᎐C), 1600, 1580 and 1500 (Ph); δ (CDCl ) 7.42–7.07 (5 H, m,
᎐
H
3
Ph), 6.76 (1 H, s, NH), 5.58 (1 H, d, J 9, HCO2CNHPh), 5.50
(1 H, t, J 7, HC᎐C), 4.25 (1 H, td, J 9 and 7, HCOSi), 3.66 (3 H,
MeB), 0.27 (3 H, s, SiMeAMeB
D), 0.12 (3 H, s, Si-
᎐
or
C
or
s, MeO), 2.93 (1 H, dt, J 10 and 8, CHCHCN), 2.62–2.07 (7 H,
MeAMeB
D) and 0.08 (3 H, s, SiMeAMeB
D);
or
or
C
or
or
C
m, ring Hs and side-chain CH C᎐C), 2.30 (2 H, t, J 7, CH -
δ (CDCl ) 173.21 (C᎐O), 144.87, 133.77, 127.71 and 129.17
᎐
C 3
᎐
2
2
᎐
COOMe), 1.71 (2 H, quintet, J 7, CH2CH2COOMe), 1.33–1.19
(1 H, m, ring H), 0.89 (9 H, s, Me3C), 0.12 (3 H, s, SiMeAMeB)
and 0.09 (3 H, s, SiMeAMeB); m/z 512 (3%, Mϩ), 455 (30,
M Ϫ But) and 119 (100) (Found: Mϩ, 512.2743. C28H40N2O5Si
requires M, 512.2707).
(aromatic C’s), 137.49 (HC᎐C), 123.56 (HC᎐C), 122.20 (C᎐N),
᎐ ᎐ ᎐
76.43 (COSi), 52.01 (MeO), 51.48, 41.43, 36.09 and 27.86
(remaining CHs), 43.41, 42.39 (2 CH2s), 33.66 (CH2COOMe),
31.40 and 24.78 (remaining CH2s), 25.69 (MeC), 17.93 (Me3C),
Ϫ4.05, Ϫ4.33, Ϫ4.82 and Ϫ4.89 (4 MeSis); m/z 511 (40%, Mϩ)
and 135 (100%, PhMe2Siϩ) (Found: Mϩ, 511.2913. C29H45-
NO3Si2 requires M, 511.2938).
(E)(1SR,2SR,5SR,7RS,8SR)-2-Dimethylphenylsilyl-3-(4Ј-
methoxycarbonylbutylidene)-7-tert-butyldimethylsilyl-8-cyano-
bicyclo[3.3.0]octane 18 and (E)(1ЈRS,1RS,5SR,7RS,8SR)-3-
(1Ј-dimethylphenylsilyl-4Ј-methoxycarbonyl)butyl-7-tert-butyl-
dimethylsilyloxy-8-cyanobicyclo[3.3.0]oct-2-ene 19
(E)(1RS,2SR,3RS,5SR)-7-(4Ј-Methoxycarbonyl)butyl-3-tert-
butyldimethylsilyloxy-2-cyanobicyclo[3.3.0]oct-7-ene 21
Trifluoroacetic acid (0.10 mmol) and the allylsilane 18 (0.01
mmol) were kept in CH2Cl2 (1 cm3) at 0 ЊC for 2 h. The mixture
was diluted with ether, washed with aqueous sodium hydrogen
carbonate, dried (MgSO4) and evaporated under reduced pres-
sure. The residue was chromatographed (SiO2, CH2Cl2) to give
the endocyclic alkene (97%); Rf(CH2Cl2) 0.38; νmax(CH2Cl2)/
The allylic acetate 16 (0.55 mmol) and the silylcuprate reagent
(1 mol dmϪ3 in THF, 1 cm3) were kept under nitrogen in a
mixture of ether (1.5 cm3) and pentane (1.5 cm3) at 0 ЊC for 2 h.
The mixture was diluted with ether, washed with aqueous
ammonium chloride and brine, dried (MgSO4) and solvent
evaporated under reduced pressure. The residue was chromato-
graphed to give the allylsilane 18 (58%); Rf(CH2Cl2) 0.44;
cmϪ1 2230 (C᎐N), 1740 (C᎐O) and 1650 (C᎐C); δ (CDCl ) 5.33
᎐
᎐
᎐
᎐
H
3
(1 H, d, J 2, HC᎐C), 4.20 (1 H, ddd, J 10, 9 and 6, HCOSi), 3.66
᎐
νmax(film)/cmϪ1 2230 (C᎐N), 1740 (C᎐O), 1660 (C᎐C) and 1600
(3 H, s, MeO), 3.23 (1 H, td, J 8 and 2, allylic bridgehead H),
2.67 (1 H, dtd, J 16, 9 and 3, remaining bridgehead H), 2.51
(1 H, dd, J 16 and 9, one allylic ring H), 2.36–2.29 (1 H, m,
HCCN), 2.30 (2 H, t, J 7, CH2COOMe), 2.17 (1 H, ddd, J 13, 9
᎐
᎐
᎐
᎐
(Ph); δH(CDCl3) 7.48–7.32 (5 H, m, Ph), 5.01 (1 H, td, J 7 and 1,
HC᎐C), 4.12 (1 H, dt, J 8 and 6, HCOSi), 3.66 (3 H, s, MeO),
᎐
2.53 (1 H, ddd, J 9, 8 and 3, CHCHSi), 2.34–1.94 (8 H, m, ring
Hs and CH CH C᎐C), 2.26 (2 H, t, J 7, CH CH COOMe), 1.62
and 6, endo CH H CHOSi), 2.06–1.98 (3 H, m, CH CH C᎐C
᎐
A B 2 2
᎐
2
2
2
2
(2 H, quintet, J 7, CH2CH2COOMe), 1.20 (1 H, dt, J 8 and 6,
exo CHAHBCHOSi), 0.86 (9 H, s, Me3C), 0.31 (6 H, s, SiMeA-
MeB or C or D), 0.09 (3 H, s, SiMeAMeB or C or D) and 0.05 (3 H, s,
and allylic ring H), 1.60 (2 H, quintet, J 7, CH2CH2COOMe),
1.43 (2 H, quintet, J 7, CH CH C᎐C), 1.32 (1 H, dt, J 13 and
10, exo CHAHBCHOSi), 0.87 (9 H, s, Me3C), 0.11 (3 H, s,
᎐
2
2
SiMeAMeB
); δ (CDCl ) 174.05 (C᎐O), 142.03, 133.68,
SiMeAMeB) and 0.08 (3 H, s, SiMeAMeB); δC(CDCl3) 173.94
᎐
or
C
or
D
C
3
᎐
129.31 and 127.79 (aromatic Cs), 136.69 (HC᎐C), 121.39
(C᎐O), 144.47 and 124.29 (HC᎐C), 122.04 (C᎐N), 76.99
᎐
᎐
᎐
᎐
᎐
(C᎐N), 120.47 (HC᎐C), 76.30 (COSi), 51.44 (MeO), 47.12,
(COSi), 52.83 (MeO), 51.42, 43.28 and 36.91 (remaining CHs),
42.50, 41.78, 33.81, 30.44 and 27.04 (remaining CH2s), 25.64
(Me3C), 17.98 (Me3C), Ϫ4.83 (SiMeAMeB) and Ϫ4.88 (SiMeA-
MeB); m/z 362 (3%, M Ϫ Me) and 320 (100, M Ϫ But) (Found:
Mϩ Ϫ 15, 362.2171. C21H35NO3Si requires M Ϫ 15, 362.2152).
᎐
᎐
45.38, 41.65 and 38.35 (CHs), 42.15, 36.10, 33.39, 29.03 and
25.10 (CH2s), 25.65 (Me3C), 17.94 (Me3C), Ϫ4.49, Ϫ4.58,
Ϫ4.83 and Ϫ4.92 (4 × MeSi); m/z 511 (29%, Mϩ) and 135 (100,
PhMe2Siϩ) (Found: Mϩ, 511.2942. C29H45NO3Si2 requires M,
511.2938), and the allylsilane 19 (6%); Rf(CH2Cl2) 0.40;
νmax(film)/cmϪ1 2220 (C᎐N), 1740 (C᎐O) and 1660 (C᎐C);
(E and Z)(1SR,2SR,3RS,5SR)-5-(4Ј-Methoxycarbonyl)butyl-
idene-3-tert-butyldimethylsilyloxy-2-cyanobicyclo[3.3.0]octane
22 and 23
᎐
᎐
᎐
᎐
δ (CDCl ) 7.52–7.27 (5 H, m, Ph), 5.15 (1 H, d, J 2, HC᎐C),
᎐
H
3
4.10 (1 H, ddd, J 10, 9 and 6, HCOSi), 3.62 (3 H, s, MeO),
3.18 (1 H, td, J 8 and 2, allylic bridgehead H), 2.58–2.37 (1 H,
m, bridgehead H), 2.30–1.12 (12 H, m, CH2s and CHs), 0.90
(9 H, s, Me3C), 0.29 (3 H, s, SiMeAMeB), 0.27 (3 H, s,
SiMeAMeB or C or D), 0.12 (3 H, s, SiMeAMeB or C or D) and 0.08
Trifluoroacetic acid (0.10 mmol) and the allylsilane 19 (0.01
mmol) were kept in CH2Cl2 (1 cm3) at 0 ЊC for 2 h, and a similar
work-up gave a mixture of the E- and Z-alkenes (90%, E:Z
67:33); Rf(CH2Cl2) 0.37; νmax(CH2Cl2)/cmϪ1 2230 (C᎐N), 1740
᎐
᎐
(3 H, s, SiMeAMeB
); δ (CDCl ) 173.65 (C᎐O), 145.38,
(C᎐O) and 1650 (C᎐C); E-isomer 22: δ (CDCl ) 5.28 (1 H, t,
᎐ ᎐
H 3
᎐
or
C
or
D
C
3
133.87, 129.19 and 127.78 (aromatic C), 137.44 (HC᎐C), 122.78
J 7, HC᎐C), 4.18 (1 H, td, J 9 and 7, HCOSi), 3.65 (3 H, s,
᎐
᎐
᎐
(HC᎐C), 121.90 (C᎐N), 76.68 (COSi), 52.43 (MeO), 51.44,
MeO), 2.63–2.44 (3 H, m, bridgehead Hs and allylic ring H),
2.35–2.14 (3 H, m, HCCN, endo CHAHBCHOSi, and allylic
ring H), 2.29 (2 H, t, J 7, CH2COOMe), 2.10–1.93 (4 H, m,
᎐
᎐
43.73, 36.00 and 30.91 (CHs), 42.78, 42.15, 33.81, 28.56 and
24.75 (CH2s) and 25.70 (Me3C), 17.97 (Me3C), Ϫ4.17, Ϫ4.43,
Ϫ4.83 and Ϫ4.88 (4 × MeSi); m/z 511 (37%, Mϩ) and 135 (100,
PhMe2Siϩ) (Found: Mϩ, 511.2907. C29H45NO3Si2 requires M,
511.2938).
CH CH C᎐C and allylic ring Hs), 1.65 (2 H, quintet, J 7,
᎐
2
2
CH2CH2COOMe), 1.21 (1 H, dt, J 13 and 8, exo CHAHB-
CHOSi), 0.87 (9 H, s, Me3C), 0.11 (3 H, s, SiMeAMeB) and 0.08
J. Chem. Soc., Perkin Trans. 1, 1998
2677

View Article Online
Organometallics, 1988, 7, 1187; W. Kitching, K. G. Penman,
(3 H, s, SiMe Me ); δ (CDCl ) 174.03 (C᎐O), 140.51 (HC᎐C)
᎐
᎐
A
B
C
3
B. Laycock and I. Maynard, Tetrahedron, 1988, 44, 3819.
2 M. J. C. Buckle and I. Fleming, Tetrahedron Lett., 1993, 34, 2383.
3 I. Fleming and N. K. Terrett, J. Organomet. Chem., 1984, 264, 99.
4 E. Vedejs and C. K. McClure, J. Am. Chem. Soc., 1986, 108, 1094.
5 I. Fleming and D. Higgins, J. Chem. Soc., Perkin Trans. 1, 1992,
3327.
᎐
122.27 (HC᎐C), 121.74 (C᎐N), 76.74 (COSi), 51.44 (MeO),
᎐
᎐
44.15, 43.45 and 38.41 (bridgehead Cs and CHCN), 42.00
(CH2CHOSi), 38.73 (CH2 trans to side chain), 34.75 (CH2 cis
to side chain), 33.39 (CH COOMe), 28.71 (CH CH C᎐C),
᎐
2
2
2
25.66 (Me3C), 24.89 (CH2CH2COOMe), 17.95 (Me3C), Ϫ4.82
(SiMeAMeB) and Ϫ4.89 (SiMeAMeB); Z-isomer 23: δH(CDCl3)
6 I. Fleming, N. J. Lawrence, A. K. Sarkar and A. P. Thomas, J. Chem.
Soc., Perkin Trans. 1, 1992, 3303.
7 D. P. Curran and B. H. Kim, Synthesis, 1986, 312.
8 H. Felkin, C. Frajerman and G. Roussi, Ann. Chim. (Paris), 1971, 6,
17; M. El Idrissi and M. Santelli, J. Org. Chem., 1988, 53, 1010.
9 R. C. Nickolson, M. H. Town and H. Vorbrüggen, Med. Chem. Rev.,
1985, 5, 1.
10 B. J. R. Whittle and S. Moncada, Prog. Med. Chem., 1984, 21, 237.
11 M. Mori, reported in M. Shibasaki, J. Ueda and S. Ikegami,
Tetrahedron Lett., 1979, 433; D. R. Morton, Jr. and F. C. Brokaw,
J. Org. Chem., 1981, 46, 4808.
12 H. Rehwinkel, J. Skupsch and H. Vorbrüggen, Tetrahedron Lett.,
1988, 29, 1775; H.-J. Gais, G. Schmiedle, W. A. Ball, J. Bund,
G. Hellman and I. Erdelmeier, Tetrahedron Lett., 1988, 29, 1773;
D. K. Hutchinson and P. L. Fuchs, J. Am. Chem. Soc., 1987, 109,
4755; M. Shibasaki, M. Sodeoka and Y. Ogawa, J. Org. Chem., 1984,
49, 4096; M. Sodeoka, S. Satoh and M. Shibasaki, J. Am. Chem.
Soc., 1988, 110, 4823; I. Edelmeier and H.-J. Gais, J. Am. Chem.
Soc., 1989, 111, 1125.
5.26 (1 H, t, J 7, HC᎐C), 4.18 (1 H, td, J 9 and 7, HCOSi), 3.66
᎐
(3 H, s, MeO), 2.63–1.93 (10 H, m, ring Hs and CH CH C᎐C),
᎐
2
2
2.30 (2 H, t, J 7, CH2COOMe), 1.68 (2 H, quintet, J 7,
CH CH C᎐C), 1.22 (1 H, dt, J 13 and 8, exo CH H CHOSi),
᎐
2
2
A
B
0.87 (9 H, s, Me3C), 0.11 (3 H, s, SiMeAMeB) and 0.08 (3 H, s,
SiMe Me ); δ (CDCl ) 170.04 (C᎐O), 140.51 (HC᎐C) 122.30
᎐
᎐
A
B
C
3
᎐
(HC᎐C), 121.81 (C᎐N), 77.00 (COSi), 51.49 (MeO), 44.85,
᎐
᎐
44.11 and 37.59 (2 bridgehead CHs and CHCN), 41.45 (CH2-
CHOSi), 40.07 (CH2 trans to side chain), 33.41 (CH2COOMe),
33.40 or 29.69 (CH cis to side chain), 28.74 (CH CH C᎐C),
᎐
2
2
2
25.67 (Me C), 24.88 (CH CH C᎐C), 17.98 (Me C), Ϫ4.80
᎐
3
2
2
3
(SiMeAMeB) and Ϫ4.86 (SiMeAMeB); m/z 362 (12%, M Ϫ Me)
and 320 (100, M Ϫ But).
The assignment of configuration rests mainly on the compar-
isons of the chemical shifts of the ring allylic methylene
carbons of our compounds E-22 (34.75 and 38.73) and Z-23
(40.07 and 33.41 or 29.69) with those reported24 for E-iloprost
(35.88 and 38.14 or 38.11) and Z-iloprost (41.23 and 32.59 or
32.50), with the upfield signals from the methylene carbon syn
to the methoxycarbonylpropyl sidechain.
13 K. C. Nicolaou, W. J. Sipio, R. L. Magolda, S. Seitz and W. E.
Barnette, J. Chem. Soc., Chem. Commun., 1978, 1067; K. Kojima
and K. Sakai, Tetrahedron Lett., 1978, 29, 3743; C. Gandolfi,
Symposium on the Chemistry & Biochemistry of Prostanoids,
Salford, July 1978; W. Skuballa, E. Schillinger, C.-St. Stürzebecher
and H. Vorbrüggen, J. Med. Chem., 1986, 29, 313.
14 I. Fleming and D. Higgins, Tetrahedron Lett., 1989, 30, 5777;
I. Fleming, D. Higgins and A. K. Sakar, in Selectivities in Lewis
Acid Promoted Reactions, ed. D. Schinzer, Kluwer, Dordrecht, 1989,
pp. 265–280.
15 M. F. Haslanger and S. Ahmed, J. Org. Chem., 1981, 46, 4808.
16 W. Nagata, S. Hirai, H. Itazaki and K. Takeda, J. Org. Chem., 1961,
26, 2413.
17 T. Utimoto, Y. Wakabayashi, T. Horiie, M. Inoue, Y. Shishiyama,
M. Obayashi and H. Nozaki, Tetrahedron, 1983, 39, 967.
18 M. F. Haslanger, US P 4,192,891, 1980; Chem. Abstr., 1981, 94,
15282h.
(E)(1SR,2SR,3RS,5SR)-5-(4Ј-Methoxycarbonylbutylidene)-3-
tert-butyldimethylsilyloxy-2-cyanobicyclo[3.3.0]octane 22
Trifluoroacetic acid (0.10 mmol) and the allylsilane 20 (0.01
mmol) were kept in CH2Cl2 (1 cm3) at 0 ЊC for 2 h, and a similar
work-up gave the E-alkene (92%, E:Z > 96:4) identical (1H
NMR, 13C NMR) with the sample above (Found: Mϩ Ϫ 15,
362.2130. C21H35NO3Si requires M Ϫ 15, 362.2152).
Acknowledgements
19 I. Paterson, Tetrahedron, 1988, 44, 4207.
20 A. L. Gemal and J.-L. Luche, J. Am. Chem. Soc., 1981, 103, 5454.
21 J. A. Dodge, J. I. Trujillo and M. Presnell, J. Org. Chem., 1994, 59,
234; P. J. Harvey, M. von Itzstein and I. D. Jenkins, Tetrahedron,
1997, 53, 3933.
22 I. Fleming, D. Higgins, N. J. Lawrence and A. P. Thomas, J. Chem.
Soc., Perkin Trans. 1, 1992, 3331.
23 M. Suzuki, H. Koyano and R. Noyori, J. Org. Chem., 1987, 52, 5583;
M. Shibasaki, H. Fukasawa and S. Ikegami, Tetrahedron Lett., 1983,
24, 3497; A. Takahashi and M. Shibasaki, Tetrahedron Lett., 1987,
28, 1893.
24 K. V. Schenker, W. von Philipsborn, C. A. Evans, W. Skuballa and
G.-A. Hoyer, Helv. Chim. Acta, 1986, 69, 1718.
We thank the SERC (now the EPSRC) for a studentship (D. H.).
References
1 T. Hayashi, M. Konishi, H. Ito and M. Kumuda, J. Am. Chem. Soc.,
1982, 104, 4962; T. Hayashi, M. Konishi and M. Kumada, J. Am.
Chem. Soc., 1982, 104, 4963; T. Hayashi, H. Ito and M. Kumada,
Tetrahedron Lett., 1982, 23, 4605; T. Hayashi, M. Konishi and
M. Kumada, J. Chem. Soc., Chem. Commun., 1983, 736; T. Hayashi,
K. Kabeta, T. Yamamoto, K. Tamao and M. Kumuda, Tetrahedron
Lett., 1983, 24, 5661; T. Hayashi, Y. Okamoto, K. Kabeta,
T. Hagihara and M. Kumada, J. Org. Chem., 1984, 49, 4224;
T. Hayashi, M. Konishi, Y. Okamoto, K. Kabeta and M. Kumada,
J. Org. Chem., 1986, 51, 3772; T. Hayashi, Y. Matsumoto and Y. Ito,
Organometallics, 1987, 6, 884; H. Wetter and P. Scherer, Helv. Chim.
Acta, 1983, 66, 118; V. G. Matassa, P. R. Jenkins, A. Kümin,
L. Damm, J. Schreiber, D. Felix, E. Zass and A. Eschenmoser, Isr.
J. Chem., 1989, 29, 321; G. Wickham and W. Kitching, Organo-
metallics, 1983, 2, 541; G. Wickham, D. Young and W. Kitching,
25 J. K. Crandall, D. B. Banks, R. A. Colyer, R. J. Watkins and J. P.
Arrington, J. Org. Chem., 1968, 33, 423.
26 M. Huckstep and R. J. K. Taylor, Synthesis, 1982, 881.
Paper 8/04274H
Received 5th June 1998
Accepted 26th June 1998
2678
J. Chem. Soc., Perkin Trans. 1, 1998