Influence of the double-bond geometry of the Michael acceptor on copper-catalyzed asymmetric conjugate addition

Article abstract of DOI:10.1002/ejoc.200700424

Focusing on mechanistic aspects, a study of the influence of the (E)/(Z) double-bond geometry of the Michael acceptor on the enantioselectivity of copper-catalyzed asymmetric conjugate addition reactions has been realized. In spite of numerous articles co

Full text of DOI:10.1002/ejoc.200700424

FULL PAPER
DOI: 10.1002/ejoc.200700424
Influence of the Double-Bond Geometry of the Michael Acceptor on Copper-
Catalyzed Asymmetric Conjugate Addition
Magali Vuagnoux-d’Augustin[a] and Alexandre Alexakis*[a]
Keywords: Michael addition / Asymmetric catalysis / Diethylzinc / Copper / Double-bond geometry
Focusing on mechanistic aspects, a study of the influence of
the (E)/(Z) double-bond geometry of the Michael acceptor on
the enantioselectivity of copper-catalyzed asymmetric conju-
gate addition reactions has been realized. In spite of numer-
ous articles concerning copper-catalyzed asymmetric conju-
gate addition reactions, the major factors of such a reaction
are quite difficult to elucidate. Although our experiments
have not allowed us to define strict rules, they have high-
lighted some factors not to be neglected when considering
the approach of the copper reagent to the double bond, such
as (E)/(Z) isomerization and steric aspects, which could
change the reactive conformation of the substrate. Electronic
effects could also modify the polarization of the double bond
and influence the nucleophilic attack.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
gents to α,β-unsaturated esters.[4] Changing the double-
bond geometry led to the formation of the opposite enantio-
mer with higher enantioselectivities. This was also the case in
a similar study on Cu-catalyzed ACA reactions of Grignard
reagents carried out by Feringa and co-workers.[5] An inter-
esting difference between enones and α,β-unsaturated esters
concerning the double-bond isomerization rate was re-
vealed. They reported that α,β-unsaturated esters slowly
isomerize in contrast to aromatic enones. However, there
are not enough instances in the literature to draw conclu-
sions about the influence of the double-bond geometry on
the copper-catalyzed asymmetric conjugate addition reac-
tions. Therefore, we decided to investigate this feature in
more detail in order to try to determine the general factors
that have an important effect on the enantioselectivity of
the Cu-catalyzed ACA of diethylzinc.
Introduction
The asymmetric conjugate addition (ACA) reaction has
received considerable interest over the past decade.[1]
Enantioselectivities reaching more than 99% can be ob-
tained with cyclic and acyclic substrates when dialkylzinc
species are used. Only a few examples concerning the influ-
ence of the double-bond geometry on the enantioselectivity
of copper-catalyzed ACA reactions have been reported in
the literature. Imamoto and Mukaiyama reported the first
instance, in 1980, of a Cu-catalyzed 1,4-addition of methyl
Grignard species to (Z)- and (E)-1,3-diphenylprop-2-en-1-
one with a prolinol derivative as the chiral ligand.[2] The
authors showed that the same major enantiomer is obtained
with the same level of enantioselectivity irrespective of the
geometry of the starting material. They also observed that
there is an isomerization of the double bond in favour of
the thermodynamically more stable (E) substrate since, in
reactions that do not go to completion, it is recovered at
the end of the reaction. Hird and Hoveyda reported the Cu-
catalyzed ACA of zinc species to (Z)- and (E)-α,β-unsatu-
rated N-acyloxazolidinone.[3] In this case, the opposite
major enantiomer is obtained if the double-bond geometry
is modified. This phenomenon means that the face selectiv-
ity of the nucleophilic attack remains the same irrespective
of the geometry of the double bond. The lower enantio-
selectivity could be explained by isomerization of the
double bond and addition to both isomers. Recently, Loh
and co-workers studied the influence of the double-bond
geometry during the Cu-catalyzed ACA of Grignard rea-
Results and Discussion
Several (E) and (Z) Michael acceptors with different ste-
ric or electronic properties were used (Figure 1).
[a] Département de Chimie Organique, Université de Genève,
30, quai Ernest Ansermet, 1211 Genève 4, Switzerland
Fax: +41-22-379-3215
E-mail: Alexandre.Alexakis@chiorg.unige.ch
Figure 1. Enones used in this study.
5852
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 5852–5860

Copper-Catalyzed Asymmetric Conjugate Addition
Synthesis of the Substrates
The non-commercially available substrates were prepared
according to reported procedures. (Z)-1[6] and (Z)-2[7] were
isolated as the pure (Z) compounds by flash chromatog-
raphy after photoisomerization of the (E) substrates using
a mercury-vapour lamp (Scheme 1).
Figure 2. Approaches of the nucleophile in the double-bond attack.
To verify this hypothesis, diethylzinc was added to the
(E) and (Z) substrates in the presence of copper and simple
phosphoramidite ligand L1 under experimental conditions
developed by our group (Scheme 3, Table 1).[10] Surpris-
ingly, the opposite major enantiomer was obtained only
Scheme 1. Isomerization of Michael acceptors.
The pure compounds (Z)-3 and (Z)-4 were prepared by a
sequence of carbocupration/carbonation[8a] followed by the
reaction of the resulting carboxylic acids with 2 equiv. of
methyllithium (Scheme 2). Thus, for (Z)-3, the reaction of
Bu2CuLi·LiI with acetylene in Et2O gave the bis(vinylic)
cuprate which was treated with carbon dioxide (dry ice) in
order to generate the pure (Z)-carboxylic acid after acidic
workup in quantitative yield.[8b] For (Z)-4, the carbo-
cupration reaction was performed in THF with a cyclohex-
ylmagnesium/copper reagent[8c] followed by carbonation
under Normant’s conditions[8d] in 45% yield. The reaction
of the acids with 2 equiv. of methyllithium in Et2O gener-
ated the desired pure (Z)-enones (Z)-3 and (Z)-4 in 80 and
71% yields, respectively. Finally, (E)-4 was prepared by an
aldolization/crotonization reaction.[9]
Scheme 3. General procedure used for Cu-catalyzed ACA reactions
of Et2Zn.
Table 1. Cu-catalyzed ACA reactions of Et2Zn.
Entry Substrate T [°C] Conv.[a] [%]
ee[b] [%] (abs. conf.)
1
2
3
4
5
6
7
8
(E)-1
(Z)-1
(E)-2
(Z)-2
(E)-3
(Z)-3
(E)-4
(Z)-4
–20
–20
–30
–30
–30
–30
–30
–30
Ͼ95
Ͼ95
Ͼ95
Ͼ95
Ͼ95
Ͼ95
Ͼ95
Ͼ95
64 (R)
44 (R)
76 (R)
92 (R)
42 (R)
62 (S)
82 (R)
76 (R)
Preliminary Results
Initially, we thought that the nucleophilic attack oc-
curred from the same face irrespective of the geometry of
the double bond. If this was generalized, the opposite major
enantiomer should be obtained when changing the geome-
try of the double bond and keeping the same catalytic sys-
tem (copper salt, ligand, solvent and temperature), as
shown in Figure 2.
[a] Conversion determined by GC/MS. [b] ee determined by chiral
GC.
Scheme 2. Carbocupration reaction to form (Z)-3 and (Z)-4.
Eur. J. Org. Chem. 2007, 5852–5860
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5853

M. Vuagnoux-d’Augustin, A. Alexakis
FULL PAPER
once when the linear substrate 3 was used (Table 1, En- Table 2, Entries 2 and 3, respectively). Moreover, when the
tries 5 and 6) and with a better enantioselectivity for the reaction was carried out with all the reagents, the reaction
adduct derived from the (Z) compound. Other substrates was complete in less than 15 min and the asymmetric induc-
gave the same major enantiomer, irrespective of the geome- tion was largely inferior (32%; Table 2, Entry 4) to that ob-
try of the double bond. The asymmetric induction of the tained with (E)-1 (64%; Table 2, Entry 1). It seemed that
adduct generated from the 4-phenylbut-3-en-2-one (2) was the double-bond isomerization was the major process in
higher when the (Z) substrate was used (Table 1, Entries 3 this reaction. As the same major enantiomer was obtained
and 4), in contrast to nitrostyrene (1) for which a lower irrespective of the double-bond geometry, the nucleophile
enantioselectivity was observed (Table 1, Entries 1 and 2). could have approached from the same face but at a slower
Finally, 4-cyclohexylbut-3-en-2-one (4) afforded the 1,4-ad- rate than the isomerization process. This could account for
duct with similar levels of enantioselectivity (Table 1, En- the decrease in the enantiomeric excess that was observed
tries 7 and 8). These first results are in contrast to our initial when (Z)-nitrostyrene [(Z)-1] was used.
hypothesis except for the oct-3-en-2-one (3) for which the
Similarly, the asymmetric induction remained constant
nucleophilic attack occurred at the same face of the Michael during the entire reaction when (E)-2 was used and no
acceptor.
isomerization of the double bond was observed (Table 3,
Entry 1). However, performing the ACA reaction using (Z)-
2 once without diethylzinc and once without the copper salt
revealed that no conjugate addition reaction occurred, but
that a small amount of isomerization had occurred after
Study of the Isomerization of the Double Bond During the
Cu-Catalyzed ACA of Et2Zn
The phenomenon of isomerization of the double bond only 15 min reaction time (7 and 11%, respectively; Table 3,
could explain and justify why the same major enantiomer Entries 3 and 4). If all components took part in the reac-
was obtained even though the geometry of the double bond tion, a small decrease in the asymmetric induction between
was inverted. This isomerization process was reported by the beginning (96%) and the end (92%) of the reaction was
Corey and Boaz in 1985 during the reaction between Gil- noticed (Table 3, Entries 2 and 5, respectively). When the
man reagents and α,β-unsaturated enones.[11a] The isomer- reaction was quenched before the complete consumption of
ization of the double bond is feasible because of the revers- the starting materials, a large quantity of isomerized sub-
ible d–π* complexation[11b–11d] which leads to the thermo- strate was formed (55% of the remaining substrate),
dynamically more stable (E) substrate. The isomerization whereas 56% of the 1,4-adduct was generated (Table 3, En-
process was studied by applying the general procedure de- try 5). We could not exclude the isomerization process dur-
scribed in Scheme 3 for nitrostyrene (1; Table 2) and 4- ing the Cu-catalyzed ACA of Et2Zn to 4-phenylbut-3-en-2-
phenylbut-3-en-2-one (2; Table 3).
one (2), but it did not seem to be the main one. In fact, the
The asymmetric induction was the same for the entire isomerization of the double bond could not explain why the
reaction when (E)-1 was used and no isomerization of the enantiomeric excess was largely superior when (Z)-2 was
double bond was observed (Table 2, Entry 1). Performing used. In theory, if the nucleophilic approach remained the
the ACA reaction using (Z)-1 once without diethylzinc and same and there was an isomerization of the double bond,
once without the copper salt revealed that no conjugate ad- the asymmetric induction should be, in the best case, equal
dition occurred, but an important isomerization phenome- to that obtained for the opposite isomer. This was not the
non did occur after only 15 min of reaction (around 70%; case and it can be assumed that other effects (steric or elec-
Table 2. Study of the isomerization of nitrostyrene (1) at –20 °C.
Entry
Substrate
Et2Zn
[equiv.]
Cu(OAc)2·H2O
Ligand
[mol-%]
Time
1H NMR proportion [%]
ee (%)[a]
(abs. conf.)
[mol-%]
(Z)-1
(E)-1
5
1
2
3
4
(E)-1
(Z)-1
(Z)-1
(Z)-1
1.2
–
1.2
1.2
2.0
2.0
–
4.0
4.0
4.0
4.0
48 h
0
31
30
0
0
69
70
0
100
0
0
64 (R)[b]
n.d.
n.d.
15 min
15 min
15 min
2.0
100
32 (R)
[a] ee was determined by chiral GC. [b] ee is constant (determined by chiral GC at different reaction times and conversions).
Table 3. Study of the isomerization of 4-phenylbut-3-en-2-one (2) at –20 °C.
Entry
Substrate
Et2Zn
[equiv.]
Cu(OAc)2·H2O
Ligand
[mol-%]
Time
GC/MS proportion [%]
ee [%][a]
(abs. conf.)
[mol-%]
(Z)-2
(E)-2
6
1
2
3
4
5
(E)-2
(Z)-2
(Z)-2
(Z)-2
(Z)-2
1.2
1.2
–
1.2
1.2
2.0
2.0
2.0
–
4.0
4.0
4.0
4.0
4.0
48 h
18 h
15 min
15 min
15 min
0
0
93
89
20
0
0
7
11
24
100
100
0
0
56
75 (R)[b]
92 (R)
n.d.
n.d.
96 (R)
2.0
[a] ee was determined by chiral GC. [b] ee is constant (determined by chiral GC at different reaction times and conversions).
5854 Eur. J. Org. Chem. 2007, 5852–5860
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Copper-Catalyzed Asymmetric Conjugate Addition
tronic, for instance) influenced the nucleophilic approach to
explain the formation of the same major enantiomers. In
both cases, the small amount of isomerization that occurred
with Et2Zn or the Cu salt alone could be explained by the
Lewis acidity of these metals.
In contrast to previous results, the experimental condi-
tions did not favour the double-bond isomerization of 4-
cyclohexylbut-3-en-2-one (4) even if the same major enanti-
omer was generated with similar enantioselectivities. It is
noteworthy that (Z)-4 reacted more slowly than the (E)
compound. When the reaction was quenched at 50% con-
version, the (Z)/(E) ratio of the unreacted starting material
was in excess of 95:5.
Scheme 4. Tandem Cu-catalyzed ACA silylation.
Table 4. Formation of silylenol ethers and the (E)/(Z) ratio.
Entry Substrate T [°C] (E)/(Z) ratio[a] Major conformer
Determination of the s-cis/s-trans Conformation of the
1
2
3
4
5
6
(E)-2
(Z)-2
(E)-3
(Z)-3
(E)-4
(Z)-4
–20
–20
–20
–20
–30
–30
81:19
71:29
86:14
79:21
58:42
30:70
s-trans
s-trans
s-trans
s-trans
s-trans
s-cis
Enones During the Cu-Catalyzed ACA of Et2Zn
Another explanation for the results obtained in Table 1
and the formation of the same major enantiomer by using
the opposite isomer is that the (E) and (Z) isomers of the
enones could react under different s-trans or s-cis confor-
mations. The reaction of each isomer in a different confor-
mation could explain the inversion of the nucleophilic ap-
proach by changing the steric constraints. To determine the
conformation of the enones studied during the copper-cata-
lyzed ACA of Et2Zn, the previous method developed in our
group by Knopff and Alexakis for the formation of enanti-
oenriched silylenol ethers by a tandem ACA/silylation reac-
tion was used.[12] It was demonstrated that the ratio of the
(E)- and (Z)-silylenol ethers synthesized was representative
of the ratio of the s-trans and s-cis forms of the enones used
for the Cu-catalyzed ACA of diethylzinc in the presence of
a phosphoramidite-type ligand. These conclusions were
made because there is no equilibration between the two zinc
enolate species generated by the Cu-catalyzed ACA reac-
tion. Moreover, the ratio between the s-trans and the s-cis
conformers of enones did not affect the enantiomeric ex-
cess. The ratio of each conformer of each substrate [(E) and
(Z)] during the Cu-catalyzed asymmetric conjugate ad-
dition of Et2Zn in the presence of a phosphoramidite ligand
(L1) was determined by this methodology (Scheme 4,
Table 4). The goal of these attempts was to find out if the
modification of the double-bond geometry could affect the
conformation of the enone and consequently explain a
modification of the facial approach of the nucleophile. The
1
[a] (E)/(Z) ratio was determined by H NMR spectroscopy.
the double-bond geometry led to an inversion of the major
conformer in favour of the s-cis compound for (Z)-4
(Table 4, Entries 5 vs. 6). Other enones preferred to react
mainly under an s-trans conformation, in accordance with
Knopff’s results. As (E)- and (Z)-4-phenylbut-3-en-2-ones
(2) reacted under the same s-trans conformation (Table 4,
Entries 1 and 2), a modification of the nucleophilic ap-
proach due to steric effects can be excluded. The formation
of the same major enantiomer cannot be explained by steric
hindrance and/or modification of the major conformer.
Oct-3-ene-2-one (3) reacted under an s-trans conformation
irrespective of the geometry of the double bond (Table 4,
Entries 3 and 4) although opposite major enantiomers were
obtained (Table 1, Entries 5 and 6). This allowed us to con-
firm that the face selectivity remained the same irrespective
of the double-bond geometry. Finally, (Z)-11 was obtained
from (Z)-4, whereas (E)-11 was generated from (E)-4. Based
on Knoppf’s results, we can assume that (Z)-4 reacted un-
der an s-cis conformation, whereas (E)-4 reacted under an
s-trans conformation. The same major enantiomer with the
same level of enantioselectivity was obtained (Table 1, En-
tries 7 and 8). By mainly reacting under an s-cis conforma-
tion, (Z)-4 could prevent the nucleophilic approach at one
face due to steric hindrance, and the nucleophilic species
has to invert its approach. This modification of conforma-
tion could be a key point in the explanation of the forma-
tion of the same major enantiomer starting from the sub-
strate with the opposite geometry of the double bond.[14]
1
(E)/(Z) ratio was determined by H NMR analysis of the
crude mixture by comparison with the results obtained by
House et al. for the preparation of trimethylsilylenol
ethers.[13] The position of the NMR peak due to the β-vinyl
proton of enol ethers allows the assignment of the stereo-
chemistry of silylenol ethers. In the isomer with the β-vinyl
proton and the oxygen function cis [(E) compound], the
position of the β-vinyl proton resonance is at a lower field
by around 0.2–0.3 ppm than that of the (Z) compound.
The conformation of the substrate did not seem influ-
Ligand Screening
Finally, it could also be assumed that there is no possible
enced by the double-bond geometry except for the sterically comparison or correlation between the (E) and (Z) sub-
hindered 4-cyclohexylbut-3-en-2-one (4) for which changing strates. They could react as different substrates with no
Eur. J. Org. Chem. 2007, 5852–5860
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5855

M. Vuagnoux-d’Augustin, A. Alexakis
FULL PAPER
similarities in their structure. That could explain why higher Table 6. Ligand screening as applied to oct-3-en-2-one (3).
enantioselectivities were obtained when the (Z) substrates
of 2 and 3 were used. To check this hypothesis ligand
screening was performed (Figure 3). If there were similari-
ties between the two isomers of a given substrate, the same
trend in enantioselectivity should be observed. For an inef-
Entry
Substrate
Ligand
Conv.[a] [%]
ee [%][b,c]
ficient ligand, the observed enantiomeric excesses for the
two adducts should decrease. In contrast, if there were no
similarities, an opposite trend should be noticed. The 4-
phenylbut-3-en-2-ones (E)-2 and (Z)-2 (Table 5) and the
oct-3-en-2-ones (E)-3 and (Z)-3 (Table 6) were submitted to
the Cu-catalyzed ACA of diethylzinc.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
(E)-3
(Z)-3
(E)-3
(Z)-3
(E)-3
(Z)-3
(E)-3
(Z)-3
(E)-3
(Z)-3
(E)-3
(Z)-3
(E)-3
(Z)-3
L1
L1
L2
L2
L3
L3
L4
L4
L5
L5
L6
L6
L8
L8
Ͼ98
Ͼ98
Ͼ98
95
Ͼ98
90
Ͼ98
95
Ͼ98
96
Ͼ98
69
Ͼ98
92
44 (R)
60 (S)
30 (R)
62 (S)
32 (R)
42 (S)
22 (S)
55 (R)
14 (R)
0
0
0
56 (S)
14 (R)
[a] Conversion was determined by GC/MS. [b] ee was determined
by chiral GC. [c] The absolute configuration is given in parentheses.
rivative was used, the enantiomeric excess was higher for
the 1,4-adduct derived from (E)-2 (L6; Table 5, Entries 7 vs.
8). However, comparison of absolute values of the enantio-
meric excesses of adducts derived from (E)-2 and (Z)-2
showed that the enantiomeric excesses did not follow the
same trend. For instance, comparison of the results ob-
tained with L1 and L2 or L3 showed an increase in the
enantiomeric excess for the adduct derived from the (E)
substrate [(E)-2] and a decrease for the (Z) one. In contrast,
comparison of the results obtained with L2 and L7 or L8
showed a decrease in the enantiomeric excess for the adduct
derived from the two substrates the magnitude of which dif-
fered depending on the double-bond geometry (Table 5, En-
tries 1, 2 and 9–12). Since they did not follow the same
trend, these two substrates have probably to be studied in-
dependently of each other.
The (Z) substrate (Z)-3 was less reactive than the (E)
isomer since the conversions were worse, except in the case
of L1. The TADDOL derivative L6 did not lead to any
asymmetric induction in either case (Table 6, Entries 11 and
12). In all cases, the opposite major enantiomer was ob-
tained, which seems to confirm that the nucleophilic ap-
proach remains the same. The asymmetric induction was
not always higher for the 1,4-adducts derived from the (Z)
substrate [(Z)-2; Table 6, Entries 9 vs. 10 and 13 vs. 14].
However, comparison of the absolute values of the enantio-
meric excesses of adducts derived from (E)-2 and (Z)-2
showed that the enantiomeric excesses did not follow the
Figure 3. Ligands used in this study.
Table 5. Ligand screening as applied to 4-phenylbut-3-en-2-one (2).
Entry
Substrate
Ligand
Conv.[a] [%]
ee [%][b,c]
1
2
3
4
5
6
7
8
9
(E)-2
(Z)-2
(E)-2
(Z)-2
(E)-2
(Z)-2
(E)-2
(Z)-2
(E)-2
(Z)-2
(E)-2
(Z)-2
L1
L1
L2
L2
L3
L3
L6
L6
L7
L7
L8
L8
Ͼ98
Ͼ98
Ͼ98
Ͼ98
92
96
74
86
94
99
75
76 (R)
96 (R)
88 (R)
94 (R)
82 (R)
88 (R)
36 (S)
20 (S)
16 (S)
88 (S)
34 (R)
90 (R)
10
11
12
98
[a] Conversion was determined by GC/MS. [b] ee was determined
by chiral GC. [c] The absolute configuration is given in parentheses.
The (Z) substrate (Z)-2 had a slightly higher reactivity same trend. For instance, comparison of the results ob-
than the (E) isomer since conversions were similar or supe- tained with L1 and L2 showed a decrease in the enantio-
rior. In all cases the same major enantiomer was obtained meric excess for the adduct derived from the (E) substrate
which seems to confirm that the nucleophilic approach is (E)-2 and an increase in the enantiomeric excess of the op-
changed by a modification of the double-bond geometry. posite one. Similarly, comparison of the results obtained
The asymmetric induction was always higher for the 1,4- with L1 and L8 showed an increase in the enantiomeric
adduct 6 derived from (Z)-2 in the presence of a phos- excess of the adduct derived from the (E) substrate and a
phoramidite-type ligand. In contrast, when a TADDOL de- decrease in the opposite one. In contrast, comparison of the
5856
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 5852–5860

Copper-Catalyzed Asymmetric Conjugate Addition
results obtained with L7 and L3 showed a decrease in the
enantiomeric excess for each adduct of around 30% of the
initial value. The same trend was observed between L1 and
L4 but the enantiomeric excess decreased to a different ex-
tent depending on the double-bond geometry. Lastly, the
usual matched and mismatched effects of L4 and L5 were
noticed here (Table 6, Entries 7–10). With regard to oct-3-
en-2-one (3), ligand screening allowed us to confirm that
the nucleophilic approach remained the same irrespective
of the geometry of the double bond of the starting material.
Since these two substrates did not follow the same tendency
on changing the ligand, we can assume that, probably, these
two substrates have to be studied independently of each
other.
Figure 4. α- and β-recognition patterns in the nucleophilic ap-
proach to the vinyl group.
If we consider the studied cases, the EWG is the same in
2 and 3 and the electronic influence on the α-carbon atom
in the carbonyl moiety is, thus, the same in both substrates.
The only difference is the replacement of a butyl group (3)
by a phenyl one (2) which leads to different electronic ef-
fects. The double bond in oct-3-en-2-one (3) is subject to
the high EWG effects of the ketone on the one hand and
to the weak inductive donor effects on the other. It is pos-
sible to assume that the copper complex preferentially “rec-
ognizes” the α-carbon atom with a high negative polariza-
tion during the d–π* interaction. In this case, modification
of the double-bond geometry does not influence the nucleo-
philic approach that occurs at the same face to give oppo-
site enantiomers. The double bond in 4-phenylbut-3-en-2-
one (2) is subject to the high EWG effects of the ketone on
the one hand and to the electronic effects of the phenyl on
the other. It is possible to assume that the influence of the
phenyl group is stronger than that of the ketone because
of π-electron delocalization and that the copper complex
preferentially “recognizes” the β-carbon atom just before
oxidative addition. Therefore, modification of the double-
bond geometry changes the nucleophilic approach which
occurs at the opposite face to give the same enantiomer.
Changes in the electronic properties of the aromatic group
by substitution could shed more light on this aspect. This
is, however, beyond the scope of this article.
Conclusions
Based on the results of the ligand screening and on the
different behaviour of the (E) and (Z) substrates, it is clear
that many reasons have to be considered to account for the
results obtained in the copper-catalyzed ACA reactions de-
scribed herein. In the case of nitrostyrene (1) the isomeriza-
tion of the double bond by a reversible d–π* interaction is
the main reason for the results obtained. The same enantio-
selectivity observed for 4 could be explained by a change in
the reactive conformer from s-trans to s-cis.[14] The results
with 3 with a linear aliphatic substituent are in line with
those of Hoveyda,[3] Loh[4] and Feringa[5] and their co-
workers who also observed a reversal of enantioselectivity.
However, when an aromatic substituent is present, such as
in 2, neither an isomerization process nor a change in the
reactive conformer can explain the higher enantioselectivity
obtained with the (Z) substrate. Therefore, other effects
should be considered. One possibility would be to consider
the carbon atoms in the α- and β-positions with respect to
the EWG and their properties due to the electronic effects
of their environment. Their influence could be very impor-
tant during d–π* interactions. Two possibilities could be en-
visaged (Figure 4). Duhamel and Plaquevent have already
reported these kinds of observations with regard the enan-
tioselective protonation of prochiral enolates.[15] The copper
cluster could preferentially “recognize” the carbon atom at
the α-position, so there is no influence of the geometry of
Experimental Section
General Methods: All reactions were carried out under argon with
oven-dried glassware. Solvents were dried by filtration through alu-
mina previously activated at 350 °C under nitrogen for 12 h before
use. All solvents were degassed by nitrogen-bubbling before use in
all experiments. Diethylzinc [15wt.-% in hexane (Acros)] and meth-
yllithium [1.6  in diethyl ether (Acros)] were used without any
further purification. Cu(OAc)2·H2O (Merck) was purchased and
used without any further purification. Evolution of the reaction
was monitored by GC/MS with a Hewlett-Packard (EI mode)
HP6890-5973 apparatus or by TLC (visualisation by UV and anis-
the double bond, and the opposite major enantiomers aldehyde, KMnO4 or PMA staining). Flash chromatography was
performed by using silica gel (32–63 µm, 60 Å). 1H (300, 400 or
should be obtained since the nucleophilic approach should
not differ. In this case, it is necessary to work with the pure
(Z) and (E) isomers since products with the opposite con-
figuration are obtained. Similarly, the copper cluster could
also “recognize” the carbon atom in the β-position. In this
case, modification of the double-bond geometry could
modify the facial approach of the nucleophile, leading to
500 MHz) and 13C (75, 100 or 125 MHz) NMR spectra were re-
corded in CDCl3 with Bruker AMX-300, -400 or -500 spectrome-
ters. Chemical shifts (δ) are given in ppm relative to residual deuter-
iated solvent. Coupling constants are reported in Hz. Enantiomeric
excesses were determined by chiral GC (capillary column, 10 psi
H2) and the temperature programs used are described as follows:
initial temperature [°C] – initial time [min] – temperature gradient
the same major enantiomer. It is not necessary to work with
the pure (Z) and (E) isomers since the two pure forms, as
well as their mixtures, afford the same enantiomer.
[°C/min] – final temperature [°C]. Retention times (Rt) are given in
min. Absolute configurations were assigned by comparison with
authentic samples.
Eur. J. Org. Chem. 2007, 5852–5860
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5857

M. Vuagnoux-d’Augustin, A. Alexakis
FULL PAPER
(Z)-Nitrostyrene [(Z)-1]:[6] A solution of (E)-nitrostyrene (2.02 g,
13.5 mmol) in cyclohexane (125 mL) was irradiated with a mer-
cury-vapour arc lamp (254 nm) for 16 h. The solvent was removed
in vacuo, and the crude mixture was purified by flash chromatog-
raphy (pentane/Et2O, 4:1) to afford the desired product as a yellow
0.36 mmol), while being kept at 30 °C and then stirred at 30 °C for
3 h. After the reaction was complete, the solution was acidified (pH
= 4) with concentrated hydrochloric acid and heated to 100 °C to
distil the acetone. The remaining residue was cooled to room tem-
perature and extracted twice with dichloromethane. The combined
organic layers were concentrated in vacuo. The crude oily product
oil in 35% yield [97% of the (Z) isomer]. 1H NMR (300 MHz,
1
CDCl3): δ = 7.58–7.26 (m, 5 H, Ar), 6.94 (d, JH,H = 9.6 Hz, 1 H, was distilled under reduced pressure (10–3 mbar) at 65 °C to give
CH), 6.75 (d, 1JH,H = 9.6 Hz, 1 H, CH) ppm. 13C NMR (75 MHz,
CDCl3): δ = 135.5, 133.9, 130.6, 130.4, 129.8, 128.5 ppm.
the desired product in 37% yield. H NMR (300 MHz, CDCl3): δ
1
1
= 3.61 (s, 1 H, CH), 3.47 (s, 1 H, CH), 2.88 (d, JH,H = 6.0 Hz, 2
H, CH2), 1.99 (s, 3 H, Me), 1.67–0.79 (m, 11 H, c-Hex) ppm. 13C
NMR (70 MHz, CDCl3): δ = 210.1, 71.6, 47.3, 43.1, 28.8, 28.1,
26.4, 26.0 ppm.
(Z)-4-Phenylbut-3-en-2-one [(Z)-2]:[7] A solution of (E)-4-phen-
ylbut-3-en-2-one (E)-2 (10.34 g, 70.73 mmol) in diethyl ether
(125 mL) was irradiated with a mercury-vapour arc lamp (254 nm)
for 20 h. The solvent was removed in vacuo, and the crude mixture
was purified by flash chromatography (pentane/Et2O, 4:1) to afford
the desired product in 41% yield [97% of the (Z) isomer] as a yel-
(E)-4-Cyclohexylbut-3-en-2-one [(E)-4]:[9] 4-Cyclohexyl-4-hydroxy-
butan-2-one (28.0 mmol) and an aqueous hydrochloric acid solu-
tion (5 wt.-%, 10 mL) in toluene (25 mL) were heated to reflux for
4 h. After the reaction was complete, the mixture was cooled, and
the aqueous layer was separated. The toluene layer was concen-
trated, and the residue was distilled at 50 °C under reduced pres-
sure (10–2 mbar) to give the colourless desired product in 63%
1
low oil. H NMR (300 MHz, CDCl3): δ = 7.46–7.29 (m, 5 H, Ar),
1
1
6.85 (d, JH,H = 12.6 Hz, 1 H, CH), 6.14 (d, JH,H = 12.6 Hz, 1 H,
CH), 2.11 (s, 3 H, Me) ppm. 13C NMR (75 MHz, CDCl3): δ =
200.7, 139.9, 135.2, 129.3, 129.2, 129.1, 128.4, 30.7 ppm.
1
yield. 1H NMR (400 MHz, CDCl3): δ = 6.70 (dd, JH,H = 16.0,
(Z)-Hept-2-enoic Acid:[8b] In a reactor equipped with a mechanical
stirrer and under nitrogen, copper iodide (10.55 g, 100.8 mmol) was
suspended in dry diethyl ether (250 mL) at room temperature. The
solution became grey. At –50 °C, n-butyllithium (63.0 mL of a 1.6 
solution in hexane) was added dropwise. The mixture turned black.
Acetylene (2.25 L) was bubbled into the mixture. After 30 min of
stirring, dry ice (CO2) was bubbled from a flask into the green
solution during 2 h. The crude mixture was quenched with a solu-
tion of NH4Cl/HCl (3:1, v/v) for 30 min. After filtration through
Celite and separation, the aqueous layer was extracted three times
with diethyl ether. The organic layers were washed with NaOH
(1.5 ), and concentrated HCl was added to the aqueous layer until
pH = 1. A last extraction of the aqueous layer with diethyl ether
was performed (three times). The combined organic layers were
dried with Na2SO4 and concentrated in vacuo to give 6.47 g of the
title compound as a yellow-brown oil in quantitative yield. 1H
1
2
2JH,H = 6.6 Hz, 1 H, CH), 5.99 (dd, JH,H = 16.0, JH,H = 1.3 Hz,
1 H, CH), 2.22 (s, 3 H, Me), 2.17–2.06 (m, 1 H, CH), 1.75–1.63 (m,
5 H, c-Hex), 1.33–1.07 (m, 5 H, c-Hex) ppm. 13C NMR (100 MHz,
CDCl3): δ = 199.2, 153.4, 128.8, 40.5, 31.7, 26.8, 25.8, 25.6 ppm.
(Z)-3-Cyclohexylacrylic Acid:[8b,8d] In a reactor equipped with a
mechanical stirrer and under nitrogen, copper iodide (13.84 g,
73 mmol) was suspended in dry THF (220 mL) at room tempera-
ture. At –60 °C, cyclohexylmagnesium chloride (63.0 mL of a
1.45  solution in THF) was added dropwise and the mixture
stirred for 45 min. Acetylene (2.0 L) was bubbled into the mixture.
After 30 min, the temperature of the mixture was allowed to rise
to –20 °C, trimethyl phosphite (ca. 0.6 mL) was added, and dry ice
(CO2) was bubbled from a flask into the solution for 4 h. The crude
mixture was stirred at room temperature for 1 h before being
quenched with a solution of NH4Cl/HCl (3:1, v/v) for 30 min. After
filtration through Celite and separation, the aqueous layer was ex-
tracted three times with diethyl ether. The organic layers were
washed with NaOH (1.5 ), and concentrated HCl was added to
the aqueous layer until pH = 1. A last extraction of the aqueous
layer with diethyl ether was performed (3 times). The combined
organic layers were dried with MgSO4, and the solvent was evapo-
1
NMR (300 MHz, CDCl3): δ = 11.73 (s, 1 H, OH), 6.32 (dt, JH,H
2
1
2
= 11.5, JH,H = 7.5 Hz, 1 H, CH), 5.75 (dt, JH,H = 11.5, JH,H
=
=
=
1
2
3
1.4 Hz, 1 H, CH), 2.64 (ddt, JH,H = 7.5, JH,H = 7.0, JH,H
1
1.5 Hz, 2 H, CH2), 1.46–1.27 (m, 4 H, CH2), 0.88 (t, JH,H
7.0 Hz, 3 H, Me) ppm. 13C NMR (75 MHz, CDCl3): δ = 172.4,
153.5, 119.1, 31.0, 28.9, 22.3, 13.8 ppm.
1
(Z)-Oct-3-en-2-one [(Z)-3]:[16] In a reactor equipped with a mechan-
ical stirrer and under nitrogen, (Z)-hept-2-enoic acid (3.32 g,
25.9 mmol) was dissolved in dry diethyl ether (250 mL) at room
temperature. The mixture was cooled to –30 °C, and methyllithium
(2.0 equiv.) was quickly added. The reaction was exothermic, and
a white precipitate was formed. The temperature was increased to
0 °C during 1 h. An aqueous saturated solution of NH4Cl (150 mL)
was added to quench the reaction. The mixture was stirred for
30 min before the aqueous layer was extracted twice with diethyl
ether. The combined organic layers were washed with an aqueous
saturated solution of NaHCO3 and dried with MgSO4. The solvent
was distilled using a Rashig column (4 mm) to give 2.60 g of the
title compound as a pale yellow oil in 80% yield. 1H NMR
rated in vacuo to give 5.0 g (45% yield) of the desired product. H
NMR (400 MHz, CDCl3): δ = 10.5 (s, 1 H), 6.15 (dd, 1JH,H = 11.5,
2JH,H = 10.1 Hz, 1 H, CH), 5.68 (dd, 1JH,H = 11.5, 2JH,H = 0.8 Hz,
1 H, CH), 3.34–3.28 (m, 1 H), 1.73–1.03 (m, 10 H, c-Hex) ppm.
13C NMR (125 MHz, CDCl3): δ = 171.5, 158.3, 116.9, 37.4, 32.2,
25.9, 25.4 ppm. HRMS (EI): calcd. for C9H14O2 154.09938; found
154.09838.
(Z)-4-Cyclohexylbut-3-en-2-one [(Z)-4]:[16] In a reactor equipped
with a mechanical stirrer and under nitrogen, (Z)-3-cyclohex-
ylacrylic acid (3.08 g, 20.0 mmol) was dissolved in dry diethyl ether
(250 mL) at room temperature. The mixture was cooled to –30 °C,
and methyllithium (25 mL, 2.0 equiv.) was added quickly. The reac-
tion was exothermic, and the mixture turned yellow. The tempera-
ture was increased to 0 °C over 1 h before saturated NH4Cl
(150 mL) was added. The two layers were stirred for 30 min. The
aqueous layer was extracted twice with diethyl ether. The combined
organic layers were washed with a saturated solution of NaHCO3,
dried with MgSO4 and concentrated in vacuo. The residue was dis-
tilled at room temperature under reduced pressure (10–2 mbar) to
give 2.2 g of a colourless oil in 71% yield. 1H NMR (400 MHz,
1
(500 MHz, CDCl3): δ = 6.02 (d, JH,H = 11.5 Hz, 1 H, CH), 5.96
(dt, 1JH,H = 11.5, 2JH,H = 7.1 Hz, 1 H, CH), 2.49 (dt, 1JH,H = 14.3,
2JH,H = 7.1 Hz, 2 H, CH2), 2.09 (s, 3 H, Me), 1.33–1.20 (m, 4 H,
CH2), 0.79 (t, 1JH,H = 7.3 Hz, 3 H, Me) ppm. 13C NMR (125 MHz,
CDCl3): δ = 199.1, 148.5, 126.8, 31.2, 31.1, 28.9, 22.2, 13.7 ppm.
4-Cyclohexyl-4-hydroxybutan-2-one:[9] Cyclohexanecarbaldehyde
(12.0 mL, 100 mmol) was added dropwise over 1 h to a solution of
acetone (40 mL), water (30 mL) and pyrrolidine (0.3 mL,
1
1
CDCl3): δ = 5.98 (d, JH,H = 11.6 Hz, 1 H, CH), 5.82 (dd, JH,H
=
5858
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 5852–5860

Copper-Catalyzed Asymmetric Conjugate Addition
2
11.6, JH,H = 9.6 Hz, 1 H, CH), 3.22–3.12 (m, 1 H), 1.75–0.97 (2 (E)- and (Z)-(4-Phenylhex-2-en-2-yloxy)trimethylsilane (9). (E)-9:[12]
m, 10 H) ppm. 13C NMR (125 MHz, CDCl3): δ = 199.0, 153.4,
125.1, 37.3, 32.2, 31.5, 25.8, 25.3 ppm. HRMS (EI): calcd. for
C10H16O 152.12012; found 152.12031.
1H NMR (400 MHz, CDCl3): δ = 7.30–7.24 (m, 2 H, Ar), 7.21–
7.12 (m, 3 H, Ar), 4.83 (d, JH,H = 9.6 Hz, 1 H, CH), 3.22–3.15
1
(m, 1 H, CH), 1.76 (s, 3 H, Me), 1.67–1.54 (m, 2 H, CH2), 0.89 (t,
1JH,H = 7.6 Hz, 3 H, Me), 0.18 (s, 9 H, SiMe3) ppm. 13C NMR
(100 MHz, CDCl3): δ = 148.0, 146.4, 128.2, 127.2, 125.6, 112.7,
General Procedure for Cu-Catalyzed ACA of Et2Zn:[10] A solution
of Cu(OAc)2·H2O (0.02 mmol) and a phosphoramidite-type ligand
(0.04 mmol) in dry diethyl ether (2.5 mL) was stirred at room tem-
perature for 30 min and then cooled to –30 °C. Et2Zn (1.2 equiv.,
1.2 mL of a 1.0  solution in hexane) was added dropwise in such
a way that the temperature did not rise above –30 °C. The solution
was stirred for 5 min, and the Michael acceptor (1.0 mmol) in dry
diethyl ether (0.5 mL) was added dropwise. The reaction mixture
was stirred at –30 °C overnight before the reaction was quenched
with a 2  aqueous solution of HCl.
1
45.5, 30.7, 18.2, 12.2 ppm. (Z)-9: H NMR (400 MHz, CDCl3): δ
1
= 7.30–7.24 (m, 2 H, Ar), 7.21–7.12 (m, 3 H), 4.63 (d, JH,H
=
9.9 Hz, 1 H, CH), 3.54–3.47 (m, 1 H, CH), 1.79 (s, 3 H, Me), 1.79–
1
1.67 (m, 2 H, CH2), 0.85 (t, JH,H = 7.3 Hz, 3 H, Me), 0.14 (s, 9
H, SiMe3) ppm. 13C NMR (100 MHz, CDCl3): δ = 146.6, 146.5,
128.1, 127.5, 125.5, 112.8, 43.4, 30.2, 22.7, 12.3, 0.7 ppm.
(E)- and (Z)-(4-Ethyloct-2-en-2-yloxy)trimethylsilane (10): IR
(neat): ν = 2958, 2922, 2856, 1670, 1251, 842, 632 cm–1. HRMS
˜
(EI): calcd. for C13H28OSi 228.190944; found 228.190840. The en-
antiomeric excess was measured by chiral GC on deprotected
1-Nitro-2-phenylbutane (5):[10] 1H NMR (500 MHz, CDCl3): δ =
7.47–7.26 (m, 5 H, Ar), 4.72–4.64 (m, 2 H, CH2), 3.51–3.45 (m, 1
1
1
ketone. (E)-10: H NMR (500 MHz, CDCl3): δ = 4.36 (d, JH,H
=
1
H, CH), 1.92–1.77 (m, 2 H, CH2), 0.96 (t, JH,H = 7.3 Hz, 3 H,
10.1 Hz, 1 H, CH), 1.94–1.85 (m, 1 H, CH), 1.73 (s, 3 H, Me),
1.44–1.11 (m, 6 H), 0.89–0.83 (t, 3 H, Me), 0.19 (s, 9 H, SiMe3)
ppm. 13C NMR (125 MHz, CDCl3): δ = 147.4, 114.1, 39.7, 36.1,
29.7, 22.9, 18.2, 14.1, 11.9, 0.32 ppm. (Z)-10: 1H NMR (500 MHz,
Me) ppm. 13C NMR (125 MHz, CDCl3): δ = 139.2, 128.8, 127.5,
127.4, 80.7, 45.9, 26.1, 11.4 ppm. The enantiomeric excess was mea-
sured by chiral GC [100–0–1–170; Rt1 = 15.3 min, (S); Rt2
15.8 min, (R); lipodex E].
=
1
CDCl3): δ = 4.15 (d, JH,H = 9.5 Hz, 1 H, CH), 2.33–2.23 (m, 1 H,
4-Phenylhexan-2-one (6):[10] 1H NMR (300 MHz, CDCl3): δ = 7.35–
7.20 (m, 5 H, Ar), 3.12–3.00 (m, 1 H, CH), 2.76 (d, 1JH,H = 7.3 Hz,
2 H, CH2), 2.05 (s, 3 H, Me), 1.79–1.53 (m, 2 H, CH2), 0.82 (t,
1JH,H = 7.3 Hz, 3 H, Me) ppm. 13C NMR (75 MHz, CDCl3): δ =
207.9, 144.2, 50.5, 42.9, 30.5, 29.3, 11.9 ppm. The enantiomeric ex-
cess was measured by chiral GC [lipodex E; isotherm 75 °C; Rt1 =
35.05 min, (S); Rt2 = 37.66 min, (R)].
CH), 1.79 (s, 3 H, Me), 1.44–1.11 (m, 6 H), 0.89–0.83 (t, 3 H, Me),
0.19 (s, 9 H, SiMe3) ppm. 13C NMR (125 MHz, CDCl3): δ = 146.1,
113.9, 36.7, 35.2, 29.7, 28.5, 23.0, 22.7, 14.1, 11.8, 0.8 ppm.
(Z)- and (E)-(4-Cyclohexylhex-2-en-2-yloxy)trimethylsilane (11): IR
(neat): ν = 2956, 2925, 2853, 1738, 1366, 1217 cm–1. HRMS (EI):
˜
calcd. for C15H30OSi 254.206594; found 254.206540. The enantio-
meric excess was measured by chiral GC on deprotected ketone.
(E)-11: 1H NMR (500 MHz, CDCl3): δ = 4.48 (d, 1JH,H = 10.8 Hz,
1 H, CH), 2.22–2.17 (m, 1 H, CH), 1.77 (s, 3 H, Me), 1.74–1.67
(m, 5 H), 1.57–1.43 (m, 1 H), 1.29–1.14 (m, 5 H), 1.10–0.94 (m, 2
4-Ethyloctan-2-one (7):[10] 1H NMR (300 MHz, CDCl3): δ = 2.20
1
(d, JH,H = 6.8 Hz, 2 H, CH2), 1.98 (s, 3 H, Me), 1.77–1.67 (m, 1
H, CH), 1.24–1.03 (m, 8 H), 0.74 (t, 1JH,H = 6.8 Hz, 3 H, Me), 0.71
1
H), 0.88 (t, JH,H = 7.3 Hz, 3 H, Me), 0.25 (s, 9 H, SiMe3) ppm.
1
(t, JH,H = 7.5 Hz, 3 H, Me) ppm. 13C NMR (75 MHz, CDCl3):
13C NMR (125 MHz, CDCl3): δ = 146.4, 111.9, 43.0, 42.2, 31.5,
29.4, 26.9, 26.7, 26.1, 22.8, 12.1, 0.4 ppm. (Z)-11: 1H NMR
δ = 208.7, 48.1, 35.0, 33.0, 28.6, 26.1, 22.7, 13.8, 10.5 ppm. The
enantiomeric excess was measured by chiral GC [lipodex E; 60–15–
20–170; Rt1 = 15.8 min, (+)-(R); Rt2 = 16.1 min, (–)-(S)].
1
(500 MHz, CDCl3): δ = 4.25 (d, JH,H = 9.8 Hz, 1 H, CH), 1.85 (s,
3 H, Me), 1.83–1.80 (m, 1 H, CH), 1.74–1.67 (m, 5 H), 1.57–1.43
1
4-Cyclohexylhexan-2-one (8): H NMR (400 MHz, CDCl3):[10] δ =
1
(m, 1 H), 1.29–1.14 (m, 5 H), 1.10–0.94 (m, 2 H), 0.88 (t, JH,H
=
7.3 Hz, 3 H, Me), 0.24 (s, 9 H, SiMe3) ppm. 13C NMR (125 MHz,
CDCl3): δ = 147.7, 111.7, 45.4, 42.2, 31.2, 29.6, 26.9, 26.8, 26.7,
25.2, 18.3, 12.1, 0.9 ppm.
2.40–2.18 (AB system, 2 H, CH2), 2.09 (s, 3 H, Me), 1.74–0.85 (2
1
m, 14 H, c-Hex, CH2), 0.80 (t, JH,H = 3.0 Hz, 3 H, Me) ppm. 13C
NMR (100 MHz, CDCl3): δ = 209.5, 45.6, 40.6, 40.0, 30.1, 30.1,
29.1, 29.1, 26.6, 23.9, 11.7 ppm. The enantiomeric excess was mea-
sured by chiral GC [lipodex E; isotherm 70 °C; Rt1 = 34.3 min, (S);
Rt2 = 36.3 min, (R)].
Acknowledgments
General Procedure for the Tandem Cu-Catalyzed ACA of Et2Zn/
Silylation Sequence:[12] A solution of Cu(OAc)2·H2O (0.02 mmol)
and a phosphoramidite-type ligand (0.04 mmol) in dry diethyl ether
(2.5 mL) was stirred at room temperature for 30 min and then co-
oled to –30 °C. Et2Zn (1.2 equiv., 1.2 mL of a 1.0  solution in
hexane) was added dropwise so that the temperature did not rise
above –30 °C. The solution was stirred for 5 min, and the Michael
acceptor (1.0 mmol) in dry diethyl ether (0.5 mL) was added drop-
wise. The reaction mixture containing the zinc enolate species was
stirred at –30 °C overnight. A solution of Et2Zn (0.1 mL) was
added to TMSOTf (0.218 mL, 1.2 mmol) in order to eliminate
traces of water. The mixture was added to the solution containing
the zinc enolate at –30 °C and stirred overnight. The reaction mix-
ture was diluted with dry diethyl ether (2 mL) and filtered through
SiO2 (2 g) previously neutralized with Et3N (0.25 mL in 7 mL
Et2O). The solvents were removed in vacuo. The crude mixture
was purified by fast flash chromatography on SiO2 (8 g) previously
neutralized with Et3N (0.25 mL) and with pentane as eluent to give
pure silylenol ether as a mixture of (E) and (Z) compounds.
The authors thank the Swiss National Research Foundation (No.
20-068095.02 and COST action D24/0003/01 (OFES contract No.
c02.0027) for financial support.
[1] For reviews, see: a) N. Krause in Modern Organocopper Chem-
istry, Wiley-VCH, Weinheim, 2001; b) M. P. Sibi, S. Manyem,
Tetrahedron 2000, 56, 8033–8061; c) N. Krause, A. Hoffmann-
Röder, Synthesis 2001, 7, 171–196; d) A. Alexakis, C. Benhaim,
Eur. J. Org. Chem. 2002, 3221–3236; e) T. Hayashi, Acc. Chem.
Res. 2000, 33, 354 –362.
[2] T. Imamoto, T. Mukaiyama, Chem. Lett. 1980, 45–46.
[3] A. W. Hird, A. H. Hoveyda, Angew. Chem. Int. Ed. 2003, 42,
1276–1279.
[4] S.-Y. Wang, S.-J. Ji, T.-P. Loh, J. Am. Chem. Soc. 2007, 129,
276–277.
[5] S. R. Harutyunyan, F. Lopez, W. R. Browne, A. Correa, D.
Pena, R. Badorrey, A. Meetsma, A. J. Minnaard, B. L. Feringa,
J. Am. Chem. Soc. 2006, 128, 9103–9118.
[6] T. Ohwada, T. Ohta, K. Shudo, J. Am. Chem. Soc. 1986, 108,
3029–3032.
Eur. J. Org. Chem. 2007, 5852–5860
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5859

M. Vuagnoux-d’Augustin, A. Alexakis
FULL PAPER
[7] H. O. House, J. Org. Chem. 1959, 24, 1374–1375.
[8] a) J.-F. Normant, A. Alexakis, Synthesis 1981, 841–870; b) A.
Alexakis, G. Cahiez, J.-F. Normant, Tetrahedron 1980, 36,
1961–1969; c) H. Westmijze, J. Meijer, H. J. T. Bos, P. Vermeer,
Recl. Trav. Chim. Pays-Bas 1976, 95, 299–303; d) J. F. Norm-
ant, G. Cahiez, C. Chuit, J. Villlieras, J. Organomet. Chem.
1974, 77, 281–287.
11026; c) S. Woodward, Chem. Soc. Rev. 2000, 29, 393–401; d)
E.-I. Nakamura, S. Morokuma, Angew. Chem. Int. Ed. 2000,
39, 3751–3771.
[12] O. Knopff, A. Alexakis, Org. Lett. 2002, 4, 3835–3837.
[13] H. O. House, L. J. Czuba, M. Gall, H. D. Olmstead, J. Org.
Chem. 1969, 34, 2324–2336.
[14] P. K. Fraser, S. Woodward, Chem. Eur. J. 2003, 9, 776–783.
[9] M. Nakajima, T. Kyotani, M. Sawaki, K. Tsukashima, WO [15] For reviews, see: a) L. Duhamel, P. Duhamel, J.-C. Plaquevent,
09107368, 1991; Chem. Abstr. 1991, 115, 114022.
Tetrahedron: Asymmetry 2004, 15, 3653–3691; b) C. Fehr, An-
gew. Chem. Int. Ed. Engl. 1996, 35, 2566–2587.
[16] N. Jabri, A. Alexakis, J. F. Normant, Tetrahedron 1986, 24,
1369–1380.
[10] a) A. Alexakis, C. Benhaim, S. Rosset, M. Human, J. Am.
Chem. Soc. 2002, 124, 5262–5263; b) C. Benhaim, Ph. D. Dis-
sertation No. 3368, University of Geneva, 2002.
[11] a) E. J. Corey, N. W. Boaz, Tetrahedron Lett. 1985, 26, 6015–
6018; b) J. P. Snyder, J. Am. Chem. Soc. 1995, 117, 11025–
Received: May 12, 2007
Published Online: October 9, 2007
5860
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 5852–5860