Copper-catalyzed asymmetric conjugate addition with chiral SimplePhos ligands

Article abstract of DOI:10.1002/chem.200901577

SimplePhos ligands represent a novel class of monodentate chiral ligands based on a chiral amine moiety and flexible diaryl groups on the phosphorous atom. They can be easily prepared by two different pathways and they can be highly functionalised. Herein we report the copper-catalysed asymmetric conjugate addition of diethyl zinc and trialkylaluminium reagents with SimplePhos ligands, which gives high enantioselectivity with cyclic enones, acyclic enones and nitro-olefins, with up to 98.6 % ee. Of particular interest is the reaction of trialkylaluminium reagents with a wide range of 3substituted enones, thus allowing the formation of stereogenic quaternary carbon centres.

Full text of DOI:10.1002/chem.200901577

FULL PAPER
DOI: 10.1002/chem.200901577
Copper-Catalyzed Asymmetric Conjugate Addition with Chiral
SimplePhos Ligands**
Laꢀtitia Palais and Alexandre Alexakis*[a]
Abstract: SimplePhos ligands represent
asymmetric conjugate addition of di-
ethyl zinc and trialkylaluminium re-
agents with SimplePhos ligands, which
gives high enantioselectivity with cyclic
a novel class of monodentate chiral li-
gands based on a chiral amine moiety
and flexible diaryl groups on the phos-
phorous atom. They can be easily pre-
pared by two different pathways and
they can be highly functionalised.
Herein we report the copper-catalysed
enones, acyclic enones and nitro-ole-
fins, with up to 98.6% ee. Of particular
interest is the reaction of trialkylalumi-
nium reagents with a wide range of 3-
substituted enones, thus allowing the
formation of stereogenic quaternary
carbon centres.
Keywords: asymmetric catalysis
·
·
conjugate addition
·
copper
enones · P ligands
Introduction
asymmetric copper-catalysed conjugate addition is nowadays
[2]
À
a well-developed methodology to create C C bonds. Many
Transition-metal-catalysed asymmetric transformations re-
quire an efficient chiral ligand to induce high levels of enan-
tioselectivity. The art of designing these ligands still remains,
in many respects, quite empirical. A way to simplify this
problem is to rely on modular structures that can easily
afford a huge diversity of ligands. Due to strong substrate
dependence in most cases, it is desirable that such ligands
remain tunable in terms of steric and/or electronic proper-
ties and can be easily synthesised. Subtle changes in confor-
mational, steric and/or electronic properties of the chiral
ligand can often lead to dramatic variation of the reactivity
and enantioselectivity.
efforts have been made in designing efficient systems and
identifying new ligands to improve enantioselectivities.
Phosphoramidite ligands appear to be among the most effi-
cient for the copper-catalysed asymmetric conjugate addi-
tion (ACA).[3] In 2001, we introduced tropos phosphorami-
dites, in which the rigid binaphthol unit was replaced by a
flexible biphenol unit.[4a] The success of these ligands was
based on the different turnover frequency (TOF) provided
by each conformer. A simple disconnection of the biphenol
and the chiral amine allows the synthesis of many variants,
both on the biphenol and the amine part.[4b]
Although bidentate ligands (C2 symmetric or not) were
almost exclusively used until the last decade, more and
more monodentate ligands are currently being investigated,
particularly in the field of conjugate addition.[1] Of particular
interest to our group were the copper-catalysed reactions in
which monodentate ligands are of prime importance. The
[a] L. Palais, Prof. Dr. A. Alexakis
Dꢀpartement de chimie organique
Universitꢀ de Genꢁve
30 quai Ernest Ansermet
1211 Geneve 4 (Switzerland)
Fax : (+41)223-793-215
In keeping the same amine part, we felt that it could be
possible to replace the biphenol part by another type of
chirality, that is, helicity. Here again, we are dealing with
two conformers that may not provide the same TOF in the
Cu-catalysed ACA.
E-mail: alexandre.alexakis@unige.ch
[**] SimplePhos ligands=phosphorus ligands based on the induced helici-
ty of a diaryl group and a chiral amine.
This new concept seemed quite successful, and we have
recently described the synthesis of some of these new effi-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200901577.
Chem. Eur. J. 2009, 15, 10473 – 10485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
10473

Ph2PCl (pathway 1). Some ligands are stable towards oxida-
tion in the solid state but slightly air-sensitive in solution.
Although traces of oxide do not impede catalysis, we have
decided to protect our ligands with borane to form phos-
phine–borane complexes,[7] which are air-stable and easily
purified by chromatography on alumina. It should be noted,
however, that hindered ligands such as L5, L12, L13, L17,
L19, L20, L22 and L23 are perfectly stable and do not easily
form the borane complexes.
For the other strategy (pathway 2), the same methodology
as for the synthesis of phosphoramidite ligands was used.
After formation of the aminophosphine dichloride, prepared
in situ from bis-(S)-(1-phenylethyl)amine or bis-(S)-(1-2-
naphthylethyl)amine (L22 and L23) and a solution of PCl3/
Et3N, three equivalents of aryl Grignard reagent[8] (prepared
in THF) were added to form the free ligand after one night
of heating at reflux. Some were not protected because it was
observed that these ligands crystallised spontaneously in
pentane without oxidation. Due to the presence of magnesi-
um salts, the ligands could not be protected in situ and
needed prior filtration on neutral alumina under an inert at-
mosphere.
cient chiral ligands called SimplePhos ligands.[5] The modu-
larity of these ligands allows for easy variations of the
amino unit[6] as well as the diaryl part on the phosphorus
atom. Moreover, SimplePhos li-
gands have shown their effi-
ciency not only for asymmetric
conjugate addition of Et2Zn to
cyclohexenone and cyclohepte-
none (with ee values up to
95%) but also for the addition
of Me3Al and Et3Al species to
methylcyclohexenone and phe-
nylcyclohexenone catalysed by
copper to form quaternary ste-
These ligands have a high potential modularity because
the amine part could be modified by the presence of elec-
tron-withdrawing or electron-donating substituents on the
Ar1 aryl moiety or by using more sterically hindered amine
reogenic carbon centres. These ligands seemed to be very ef-
ficient for ACA, since high enantioselectivity was obtained
in experiments in which phosphoramidite-type ligands gave
inferior results.
We herein report a full ac-
count of our results in the syn-
Table 1. Synthesis of SimplePhos–borane ligands by two different pathways.
thesis of several members of
the SimplePhos family of li-
gands.
Modifications
were
made to both the amine and
aryl scaffold with different elec-
tronic or steric requirements.
All these new ligands were
tested for the copper-catalysed
ACA on several substrates,
such as cyclic and acyclic
enones, nitro-olefins and substi-
tuted cyclohexenones for the
formation of quaternary carbon
centres.
Entry
Ligand
Ar1
Ar2
R
Isolated yield [%]
1
2
3
4
5
6
7
8
L1
Ph
Ph
Ph
o-OMe
2-naphthyl
1-naphthyl
o-tolyl
p-OMe
Ph, 2-naphthyl
Ph
Ph,
Ph, 1-pyridine
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
2-naphthyl
2-naphthyl
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
o-Me
p-Me
p-OMe
m-CF3A
m-OMe
xylyl
Me
Me
Et
Me
Me
Me
Me
Me
68
82
84
58
41
78
57
63
94
4
41
50
94
94
70
25
71
67
40
30
59
8
L1-BH3
L2-BH3
L3-BH3
L4-BH3
L5
L6-BH3
L7
L8-BH3
L9
(C6H4)
9
Me
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Me, iPr
Me, (CH2)2Ph
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
L10-BH3
L11-BH3
L12
Results and Discussion
G
Synthesis of SimplePhos li-
gands: Starting from a chiral
C2-symmetric amine (except for
L8–L11), SimplePhos ligands
have been prepared by two dif-
ferent pathways (Table 1). For
the first one, ligands L4 to L11
were obtained by deprotonation
of the chiral amine by nBuLi
followed by the addition of
L13
R
L14-BH3
L15-BH3
L16-BH3
L17
L18
L19
L20
L21
L22-BH3
L23
N
G
G
3,5-CF
3A
2-naphthyl
1-naphthyl
1-furyl
xylyl
m-CF
66
46
(C6H4)
10474
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10473 – 10485

SimplePhos Ligands
FULL PAPER
Table 2. Deprotection of SimplePhos–borane ligands by using morpholine.
with 2-naphthyl and 1-naphthyl
as Ar1 aryl groups. Coordina-
tion effects can also expected
by an ortho-OMe group.[9,10] We
have shown that these effects
have an important consequence
for the enantioselectivity in the
copper-catalysed ACA with
phosphoramidite-type ligands.
On the amine part, the C2 sym-
metry is not a requirement be-
cause we have also synthesised
non-symmetric chiral amines
(two different Ar1 aryl groups).
Again, electronic or coordina-
tion effects can be tuned by var-
iation of R (methyl, ethyl or
phenyl groups) and Ar1 (o-Me
and o-OMe).
Entry[
Ligand
Ar1
Ar2
R
Isolated yield [%]
1
L2
Ph
Ph
Et
98
93
97
100
81
88
87
80
83
77
97
2
3
4
5
6
7
8
9
L3
L4
L6
L8
L10
L11
L14
L15
L16
L22
o-OMe
2-naphthyl
o-tolyl
Ph, 2-naphthyl
Ph
Ph, 1-pyridine
Ph
Ph
Ph
2-naphthyl
(C6H4)
Ph
Ph
Ph
Ph
Ph
Ph
Me
Me
Me
Me
Me, (CH2)2Ph
Me
Me
Me
Me
Me
p-OMeACTHNGUETRN(NUG C6H4)
m-CF3ACHTUNGETRNNU(G C6H4)
m-OMeACTHNGUETRN(NUG C6H4)
xylyl
10
11
All the free ligands displayed a single set of signals in the
1H, 13C and 31P NMR spectra. This is a good indication of
the rapid interconversion of the two helical conformers.
Only the most hindered ligand, L22-BH3, gave rise to two
signals at low temperature (Figure 1). Thus, the free and
borane variants of ligand L22 were studied by variable-tem-
perature 31P NMR spectroscopy. From À908C (183K) to
room temperature a single set of sharp signals was observed
with the free ligand. Concerning borane ligand L22-BH3,
On the Ar2 aryl part on the phosphorus atom, similar var-
iations can be applied. For example, the xylyl group on L17
and L22 can decrease the free rotation around phosphorus
and modify the general helicity and consequently act on the
enantioselectivity. Such beneficial effects have been ob-
served with ortho-substituted biphenol phosphoramidite li-
gands.[4b] Moreover, and in contrast to phosphoramidite li-
gands, the substituents on Ar2 can have a direct electronic
effect on the coordination properties of the phosphorus
atom. This is why we have synthesised ligands L12–L16, L18
and L23 with CF3 or OMe groups in various positions. Com-
bination of these various effects were undertaken according
to the results of the conjugate additions we attempted.
Aromatic heterocycles may also be introduced as Ar2
groups. Ligand L21, with a 2-furyl group, is representative
of this class. A possible coordination of the heteroatom with
copper could be envisaged. This ligand was prepared accord-
ing the modified procedure 2 using 2-lithiofuran instead of
Grignard reagent. An excess of this lithium reagent had to
be added to the solution of phosphine amine dichloride and,
after one night of heating at reflux, ligand L21 was obtained
with very low yield.
In addition to these variations, it could be possible to
create a stereogenic phosphorus atom if the two Ar2 aromat-
ic substituents were different. Thus, by adding successively
two different aromatic Grignard reagents to the amine–PCl2
intermediate, we observed two signals by 31P NMR spectros-
copy on the resulting diastereomeric ligand. Unfortunately,
a 2:1 ratio indicated a low asymmetric induction by the
chiral amine.
SimplePhos ligands complexed by BH3 should be depro-
tected for further use because the borane[11] complex did not
provide good results in terms of enantioselectivity for the
addition of Et2Zn to cyclohexenone. For this reason, all the
borane complexes were deprotected with of morpholine
(10 equiv) in diethyl ether at room temperature.[7] A simple
filtration on neutral alumina under an inert atmosphere
gave the desired ligands with high yields (Table 2).
Figure 1. Variable-temperature 31P NMR spectra of ligand L22-BH3 in
[D8]toluene.
Chem. Eur. J. 2009, 15, 10473 – 10485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10475

L. Palais and A. Alexakis
the multiplicity did not change over the temperature range
of 258C (298 K) to À708C (203 K). However, by lowering
the temperature below À708C, the singlet broadened and
split into two signals with a coalescence temperature at
À808C (193 K). Further cooling to À908C resulted in the
generation of two resolved signals originating from two heli-
cal conformers in approximately 1:3.5 ratio. This slower in-
terconversion of the two conformers should be ascribed to
the increased rigidity of the borane complex.
Scheme 2. Michael acceptors used in this study.
Copper-catalysed asymmetric conjugate addition of R2Zn:
We first investigated the conjugate addition of diethyl zinc
reagent to cyclic enones using the same conditions as those
described previously[4b,5] with copper thiophene carboxylate
(CuTC; 5 mol%) and chiral ligand (10 mol%; Scheme 1).
Cyclohexenone 1 is the archetypical enone in the copper-
catalysed ACA. With this substrate, we have tested all the
synthesised ligands for the addition of Et2Zn (Table 3).
Among the variations on the amine part, ligands L4, L22
Table 3. Addition of Et2Zn to cyclohexenone catalysed by CuTC in the
presence of various ligands.
Entry
Ligand
Conv. [%][a]
ee [%][b]
1
2
3
4
5
6
7
8
L1
L2
L3
L4
L5
L6
L7
L8
97 (90)[c]
100
80 (83)[c]
81
63
95 (96)[c]
70
51
88
86
24
83
2
100
100 (100)[c]
100
100
100
99
100
100
60
100
100
100
100
100
100
100
100
100
100
9
L9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
L10
L11
L12
L13
L14
L15
L16
L17
L18
L19
L20
L21
L22
L23
4
67
42
90
75
73
82
56
23
75
91
95
100
100
[a] Determined by GC–MS. [b] Determined by chiral GC. [c] Reaction
was carried out at À108C.
Scheme 1. SimplePhos ligands tested for conjugate addition.
and L23 with Ar1 =2-naphthyl afforded the best results with
complete conversions and very high enantioselectivities up
to 95% ee (Table 3, entry 4). As we have observed previous-
ly for biphenol-based phosphoramidite ligands, an increase
of the steric bulk on the amine favours the enantioselectivi-
ty.[13] But with a too sterically hindered Ar1, such as the 1-
naphthyl group in L5 (entry 5), the enantiomeric excess de-
creased to 70%. Moreover, ortho substituents on this same
aryl group on the amine part is not favourable. Indeed, li-
gands L3 and L6, which respectively have a methoxy and
methyl group, show lower results (entries 3 and 6). Concern-
The cyclic and acyclic enones and nitroalkenes used for this
study with R2Zn are compiled in Scheme 2. All of them are
commercially available except for the nitroalkene 7, which
was prepared according to the literature.[12] In the following
study, the full screening of SimplePhos ligands is only de-
scribed for cyclohexenone 1. For the other Michael accept-
ors, we only summarised the interesting results. The full
screening of ligands with all the substrates are reported in
the Supporting Information.
10476
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10473 – 10485

SimplePhos Ligands
FULL PAPER
ing the modification of the R1 group, the replacement of the
methyl (L1) by an ethyl group (L2) did not have a consider-
able effect (entries 1 and 2). On the amine part, the C2 sym-
metry is not a requirement because ligand L8, on account of
having two different R or Ar1 groups (entry 8), was more ef-
ficient than the simple L1.
methyl-3-hexen-2-one (3) under the same experimental con-
ditions and the results are summarised in Table 5. The test
reaction for addition of Et2Zn was realised with the most
Table 5. Addition of Et2Zn to various acyclic enones catalysed by copper
in the presence of various SimplePhos ligands.
Concerning the diaryl scaffold, the first striking result
showed that the bulkier, electron-rich substituent in the
ortho position on the Ar2 part on the phosphorus atom (L12
and L20) was detrimental to enantioselectivity (Table 3, en-
tries 12 and 20). The enantioselectivity also dropped using li-
gands L13 and L14 with an electron-donating group in the
para position, such as methyl or methoxy moieties (en-
tries 13 and 14). In contrast, the presence of a substituent in
the meta position on Ar2 seemed to improve the enantiose-
lectivity. However, an electron-rich methoxy group (L16) af-
forded lower enantiomeric excess values than an electron-
withdrawing group such as m-CF3 (L15) (entries 15 and 16).
This surprising electronic effect was not observed with phos-
phoramidite ligands. As we were hoping to increase the
enantioselectivity by limiting the free rotation of aryl groups
around the phosphorus atom, we investigated ligands L17–
L19 with more sterically hindered substituents: for example,
with methyl and CF3 groups in 3,5-positions or with Ar2 =2-
naphthyl. Once again, electronic effects had a strong influ-
ence on the enantioselectivity. Ligand L18 with an electron-
withdrawing group (CF3) gave better results than ligand L17
with an electron-rich group (CH3) (entries 17 and 18). On
the other hand, when Ar2 =2-naphthyl, the system was too
sterically hindered to enhance the enantioselectivity
(entry 19). In some cases, we tried to perform the reaction
at higher temperature (À108C) but enantioselectivities were
almost the same (Table 3, entries 1 and 4).
Entry
1[a]
2
3
4
5
6
7
8
Substrate
CuX
Ligand
Conv. [%][a]
ee [%][b]
3
3
3
3
3
3
4
4
5
5
CuTC
L1
L1
L4
L17
L22
L23
L2
L5
L1
99
100
100
96
100
94
100
100
100
100
32 (R)
58 (R)
75 (R)
80 (R)
77 (R)
54 (R)
52 (R)
52 (R)
0
Cu
Cu
Cu
Cu
Cu
N
CuTC
CuTC
CuTC
CuTC
9
10
L23
68 (S)
[a] Determined by GC–MS. [b] Absolute configuration; determined by
chiral GC.
simple ligand, L1, in the presence of CuTC. In contrast to
cyclic enone, low enantioselectivities were observed under
these conditions (Table 5, entry 1). However, we chose to
screen various copper sources with ligand L1 and we man-
aged to improve the enantiomeric excess with copper tri-
fluoroacetylacetonate (CuACTHNUTRGEN(UNG F3acac)2) up to 58% ee (entry 2).
Then we applied this new condition to other ligands and the
best result was obtained with ligand L17 (Ar2 =xylyl), with
up to 80% ee (entry 4). Similar enantioselectivities of
around 76% ee were observed with ligands bearing a 2-
naphthyl group on the amine moiety, L4 and L22 (entries 3
and 5). Ligand L23, which previously gave high enantiomer-
ic excess, showed only 54% ee (entry 6). On the aryl part, it
seemed that steric hindrance overcame the beneficial elec-
tronic effects.
Next we studied the addition of the diethyl zinc reagent
to linear acyclic enone 4 catalysed by CuTC (Table 5, en-
tries 7 and 8). Only some of the ligands were tested in this
reaction but moderate enantioselectivity was obtained with
ligands L2 and L5, with up to 52% ee (entries 7 and 8). The
screening of a large range of copper salts to afford better re-
sults was unsuccessful.
The last acyclic a,b-unsaturated ketone used in this study
was benzalacetone (5). First, we tested some SimplePhos li-
gands under classical conditions. Ligand L22 could catalyse
the asymmetric conjugate addition of Et2Zn with complete
conversion and 68% ee (Table 5, entry 10). Many ligands
were unable to provide efficient asymmetric induction in
this reaction. Therefore, we tested an array of different
copper salts with ligand L1. A small improvement of enan-
The same observations were made with the cyclohepte-
none 2 (Table 4) but with slightly lower results in terms of
enantioselectivity, although up to 95% ee could be obtained
with L23 (Table 4, entry 5).
Table 4. Addition of E2Zn to cycloheptenone catalysed by CuTC in the
presence of various ligands.
Entry[a]
Ligand
Conv. [%][a]
ee [%][b]
1
2
3
4
5
L1
L4
L15
L22
L23
100
67
100 (100)[c]
100
92 (93)[c]
90
91
95
100
100
[a] Determined by GC–MS. [b] Determined by chiral GC. [c] Reaction
was carried out at À108C.
Then the challenge was to apply this methodology to
other Michael acceptors such as acyclic enones and nitroal-
kenes. We first screened some of these ligands with 5-
tioselectivity was observed with [CuACHTGNUETRN(NUG OTf)]2·C6H6 (Tf=tri-
fluoromethansulfonate). But after trying this new copper
source for the addition of Et2Zn to this substrate, we were
Chem. Eur. J. 2009, 15, 10473 – 10485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10477

L. Palais and A. Alexakis
unable to increase the enantiomeric excess. Surprisingly,
ligand L22 catalysed the reaction with CuTC (68% ee) but
not with [CuACHTUNGTRENNUNG(OTf)]2·C6H6.
last few years, new methodologies for the construction of
quaternary carbon centres have appeared in the literature.[16]
Powerful catalytic methods using copper-catalysed asymmet-
ric conjugate reactions were described with reactive sub-
strates such as nitroalkenes[17] or doubly activated enones.[18]
To avoid the problem of a lack of reactivity of b-trisubstitut-
ed enones, both our group[19] and those of Hoveyda[20] and
Tomioka[21] have developed efficient catalyst systems that
tolerate this type of substrate.
Trisubstituted a,b-unsaturated ketones used for this study
are summarised in Scheme 3. Some of them were commer-
cially available; the others were prepared according to liter-
ature procedures.[22]
Finally, we envisaged another type of Michael acceptor
such as nitroalkenes. para-Methylnitrostyrene (6) was
chosen as the substrate because high enantioselectivity has
been reported for the asymmetric conjugate addition of the
diethyl zinc reagent with phosphoramidite-based li-
gands.[4b,14] Moreover, to see the influence of the substituent,
an acyclic nitroalkene bearing an alkyl group (7) and a
cyclic nitroalkene (8) were selected.
Classical conditions, with CuTC (5 mol%) and chiral
ligand (10 mol%), were used at À308C in Et2O. Several
SimplePhos ligands were screened, and the product formed
by addition of Et2Zn to substrate 6 gave the best enantio-
meric excess (61%) with L2 (Table 6, entry 1).
Table 6. Addition of R2Zn to nitroalkenes catalysed by CuTC Simple-
Phos ligand L2.
Scheme 3. Trisubstituted enones used in this study.
Entry
1
Substrate
R2
Et
Adduct
Conv. [%][a]
100
ee [%][b]
61 (S)
We first investigated the conjugate addition of Et3Al to 3-
methylcyclohexen-2-one (17). As we have recently reported,
the best experimental conditions are to perform the reaction
at À108C in diethyl ether with 5 mol% CuTC and 10 mol%
chiral ligand.[5] At a lower temperature (À308C), some prob-
lems of conversion were observed and more than two equiv-
alents of organometallic reagents were needed, yet this was
without any improvement of the enantioselectivity. The re-
sults with all the ligands are compiled in Table 7.
14
2
3
Et
15a
15b
100
100
50 (S)
89 (S)
Me
4
Et
16
100[c]
92 (cis)
As for conjugate addition of Et2Zn, bulky groups on the
amine part of the ligand such as the 2-naphthyl group (L4
and L22) led to an increased enantioselectivity (up to
95% ee; Table 7, entries 4 and 22) except with ligand L23
(entry 23), for which both conversion and enantioselectivity
dropped. In contrast to the previous results obtained for the
ACA of Et2Zn, electron-withdrawing substituents R3 on the
diaryl group were not favourable. However, the bulky R
part (L2) gave good enantiomeric excess (entry 2). As we
have previously shown, ligand L8 with a non-symmetric
chiral amine (entry 8) gave a comparable enantiomeric
excess (92% ee) to those obtained with L1 and L4 (90 and
93% ee; entries 1 and 4, respectively).
As previously observed, ortho substitution on the diaryl
groups of the ligand is highly detrimental. Only 1% conver-
sion was observed (Table 7, entry 8). On the other hand, the
presence of an electron-donating group (methyl and me-
thoxy) in the para position has a more positive effect and
gave good enantiomeric excess (entries 9 and 10). However,
a methoxy group in the meta position (L16) did not improve
the enantioselectivity and even showed a deleterious effect
(entry 16).
[a] Determined by GC–MS. [b] Absolute configuration; determined by
chiral GC or supercritical fluid chromatography (SFC). [c] cis/trans 85:15.
We then applied this methodology to nitroalkene 7 and,
as previously, ligand L2 catalysed the reaction with full con-
version but moderate enantiomeric excess (50% ee)
(Table 6, entry 2). In addition, we also tried the addition of
Me2Zn to 7. This was more challenging because it is known
that dimethyl zinc is much less reactive than diethyl zinc. By
combining this reagent with L2, the corresponding adduct
was obtained with a remarkable 89% ee (entry 3).
Finally, the addition of Et2Zn to the cyclic nitroalkene 8
gave the desired compound in full conversion and high
enantioselectivity: up 92% ee for the cis product (entry 4).
Unfortunately, Me2Zn was not reactive enough to give the
corresponding adduct of cyclic substrate 8.
Copper-catalysed asymmetric conjugate addition of R3Al re-
agents: Another more demanding copper-catalysed conju-
gate addition was tested: the addition of R3Al to trisubsti-
tuted cyclic enones to induce the formation of stereogenic
quaternary carbon centres. The synthesis of biologically
active natural products and drugs bearing quaternary stereo-
centres still remains a highly demanding task.[15] Over the
In contrast to dialkyl zinc species for conjugate addition,
an electron-withdrawing CF3 group in the meta position
(L15) was not as good and it was even worse with two
10478
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10473 – 10485

SimplePhos Ligands
FULL PAPER
Table 7. Addition of Et3Al to 3-methyl-2-cyclohexenone (17) catalysed
by CuTC in the presence of various SimplePhos ligands.
Scheme 4. ACA of R3Al to ethylcyclohexen-2-one (18) catalysed by
CuTC in the presence of ligand L22.
Entry
R
Ligand
Conv. [%][a] (yield[%])
ee [%][b]
1
2
3
4
5
6
7
8
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
Et
nPr
nBu
L1
L2
L3
L4
L5
L6
L7
L8
85
90
93
86
93
70
66
88
92
20
90
10
–
90
91
80
27
92
24
92
27
74
95
51
96
95
100 (71)
98 (73)
100
99
100
100
100
87
100
48
1
100 (60)
44
82
100
100 (81)
3
100
34
3-isobutylcyclohex-2-en-1-one (20) to determine the limits
of our system. In this case, the results obtained with Simple-
Phos ligands gave better yield and enantioselectivities than
with phosphoramidite counterparts.
We first envisaged the addition of a methyl group on 20
(Table 8). In the case of phosphoramidite ligands, Me3Al
gave better enantioselectivities than Et3Al.[19b] To check if
9
L9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24[c]
25[c]
L10
L11
L12
L13
L14
L15
L16
L17
L18
L19
L20
L21
L22
L23
L22
L22
Table 8. Addition of R3Al to 3-isobutyl-2-cyclohexenone (20) catalysed
by CuX in the presence of various SimplePhos ligands.
100
91 (71)
85
100 (52)
100 (66)
Entry
R
CuX
Ligand
Conv. [%][a]
(yield [%])
ee
N
[%][b]
1
2
3
4
5
6
7
8
Me
Me
Me
Me
Me
Me
Me
Me
Et
CuTC
L1
L1
L1
L1
100
100
100
85
85
83.5
85.5
95
96
94
97
94
95
96
98.6
97
96
[a] Determined by GC–MS. [b] Determined by chiral GC. [c] Reaction
was carried out at 08C with 3 equiv of R3Al.
Cu
A
[Cu
A
Cu
Cu
Cu
Cu
Cu
N
100 (81)
97 (79)
100 (90)
97 (78)
90
98 (77)
75 (60)
10 (85)
>99 (66)
79 (55)
L2
L17
L19
L22
L1
groups in this same position (L18; Table 7, entries 15 and
14). One of the best previous ligands, L23, showed lower en-
antiomeric excess for the addition of triethylaluminium to
this substrate (entry 23).
9
CuTC
CuTC
CuTC
CuTC
CuTC
CuTC
10
11
12
13[c]
14
Et
Et
Et
L4
L17
L22
L22
L22
Then, to enlarge the scope of organoaluminium reagents,
we decided to use tri-n-butyl- and tri-n-propylaluminium,
which are also commercially available at a low price. These
types of reagent had never been used for the formation of
quaternary carbon centres to substituted enones. After opti-
misation of the experimental conditions, we found that the
reaction should be performed at higher temperature (08C)
with three equivalents of nPr3Al or nBu3Al to get total con-
version. As previously, the best ligand for addition of such a
trialkylaluminium reagent is L22, with high enantioselectivi-
ties close to 96% ee for the addition of nPr3Al and 95% ee
with nBu3Al (Table 7, entries 24 and 25).
The same procedure was applied for the ACA of nPr3Al
and nBu3Al to 3-ethylcyclohexen-2-one (18; Scheme 4).
Once again, ligand L22 showed its efficiency with full con-
version and high enantioselectivities up to 97% ee.
The experimental conditions were further optimised by
copper-catalysed asymmetric conjugate addition of various
aluminium species to the bulkier cyclohexenone substrate,
nPr
nBu[c]
100 (71)
[a] Determined by GC–MS. [b] Determined by chiral GC. [c] Reaction
was carried out at 08C with 3 equiv of R3Al.
this was also the case with SimplePhos ligands, we tried to
add various groups such as methyl, ethyl, n-propyl and n-
butyl. Surprisingly, the reaction of Me3Al with ligand L1
gave a moderate 85% ee (Table 8, entry 1). Some optimisa-
tions of the experimental conditions were made by changing
the copper salts. Copper acetylacetonate appeared to be an
optimum one (entry 4). Then other ligands were tested
under these new optimised conditions. This test substrate 20
gave at best 98% ee with a biphenol-based phosphoramidite
ligand.[19c] Three SimplePhos ligands, L2 and, again, L17 and
L22, gave respectively close enantiomeric excess values
from 95 to 97% (entries 5, 6 and 8). Once again, the bulkier
ligand L22 proved its efficiency for the formation of quater-
Chem. Eur. J. 2009, 15, 10473 – 10485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10479

L. Palais and A. Alexakis
Table 9. Addition of R3Al to 3-phenyl-2-cyclohexenone (21) catalysed by
copper salt in the presence of various SimplePhos ligands.
nary carbon centres. This was not the case for all ligands be-
cause some of them showed a deleterious effect both on
conversions and enantioselectivities (for example, with
groups in meta positions for the diaryl part). To our surprise,
ligand L19, which generally did not give good results, cata-
lysed the reaction with full conversion and high enantiomer-
ic excess up to 94% (entry 7).
Next the scope of the aluminium reagents was studied.
After some optimisations, we found that 5 mol% CuTC and
10 mol% chiral ligand in diethyl ether offered the best con-
ditions (Table 8). With Et3Al, the first attempt with the
simple ligand L1 gave both high conversion and 94% enan-
tioselectivity (Table 8, entry 9). Moreover, ligands L4 (2-
naphthyl group on the amine part) and L17 (xylyl groups on
P) also gave good results (entries 10 and 11). However,
ligand L22, which is a mix of these two best ligands, im-
proved the results with 98.6% ee (entry 12). For all the
other ligands, the reaction gave good enantioselectivities but
low conversions. The addition of a propyl chain to this sub-
strate worked perfectly with ligand L22, with 55% isolated
yield and 97% ee (entry 13). In the case of enone 20, Sim-
plePhos ligands were very efficient for the addition of
Et3Al, nPr3Al, and nBu3Al species to this bulky trisubstitut-
ed cyclohexenone, clearly better than phosphoramidite-type
ligands, which gave no conversion with Et3Al.[19c]
Interestingly, the absolute configuration of the adducts
obtained by the addition of Me3Al, Et3Al, nPr3Al or nBu3Al
was the same as with a less bulky substrate (methylcyclohex-
enone 17). Thus the ACA of R3Al showed that the face se-
lectivity is the same whatever the R substituent (small or
bulky).
3-Arylcyclohex-2-en-1-ones 21–23 were quite problematic
substrates since both steric and electronic effects were com-
bined. In our preliminary study,[5] we reported that 3-phenyl-
cyclohex-2-en-1-one (21) gave the desired adduct with
Me3Al using CuTC as the copper salt with L17, and did so
in good conversion and with 86% ee. Having many new li-
gands in hand, we screened all our SimplePhos ligands to
find the best one (only a selection appear in Table 9). In ad-
dition, we also tested Et3Al, nPr3Al and nBu3Al.
Entry
R
CuX
Ligand
Conv. [%][b]
(yield [%])
ee
G
[%][c]
1
2
3
4
5
6
7
8
Me
Me
Me
Me
Me
Me
Me
Et
Et
Et
Et
Et
Et
Et
Et
nPr
nBu
CuTC
CuTC
CuTC
CuTC
CuTC
CuTC
CuTC
CuTC
CuTC
L1
L4
L8
L13
L16
L17
L22
L1
L1
L1
L1
L1
98
76
81
81
75
71
86
87
81
81
75
82
81
86
86
90
88
88
69 (63)
86 (63)
94
79
98 (71)
74 (61)
85
84
>99
9[a]
10
11
12
13
14
15
16[d]
17[d]
[Cu
A
Cu
Cu
Cu
Cu
Cu
N
94
98
L4
95 (70)
73 (64)
95 (69)
97 (63)
79 (56)
L17
L22
L17
L17
CuTC
CuTC
[a] Reaction was carried out at 08C. [b] Determined by GC–MS. [c] De-
termined by chiral GC. [d] Reaction was carried out at 08C with 3 equiv
of R3Al.
tions. Once again, ligand L22 with the amine 2-naphthyl
moiety gave the best results of up to 90% ee (entry 15). Li-
gands L4 and L17 also showed high conversions and good ee
values up to 86% (entries 13 and 14). But for the addition
of a longer alkyl chain (propyl or butyl), ligand L17 was the
best one with 88% ee in the both cases (Table 9, entries 16
and 17).
To check the influence of the electronic effects on the b
position of the enone, copper-catalysed asymmetric addi-
tions of trialkylaluminium reagents were carried out on two
other 3-arylcyclohexen-2-ones, 22 and 23, with electron-
withdrawing or electron-donating groups. For this study, we
tested just four different ligands (Table 10).
First, we investigated the addition of Me3Al to 22 bearing
an electron-donating OMe group in the para position. The
classical conditions with CuTC seemed to be the best and
the desired adduct could be obtained with 81% ee and good
yield with L17 (entry 2). This result was much better than
those obtained with phosphoramidite-based ligands (20%
conversion, 60% ee).[19c] Et3Al was tested with CuTC; how-
ever, low conversion was observed (Table 10, entry 4). We
have already seen that copper acetate monohydrate could
lead to increased reactivity (see Table 9). Therefore, the
three other ligands were tested with these new conditions
and good conversion and enantioselectivities were obtained
with L4, with up to 90% ee (Table 10, entry 6). Ligand L22
gave quite similar enantioselectivity but lower conversion
(89%; entry 8).
Ligand L22 gave similar enantioselectivity as L17 (up to
87% ee) (Table 9, entry 7), even if the conversion was not
full. Other ligands did not perform well, with either low con-
versions and enantiomeric excesses (not shown in the table),
except for ligands L1, L4, L8, L13 and L16, which gave
quite good conversions and ee values close to 75–81%.
We then investigated the addition of Et3Al on this same
substrate 21, first with optimised experimental conditions.
Thus, CuTC and ligand L1 catalysed the reaction with 85%
conversion and 81% ee (Table 9, entry 8). The same result
was obtained at a higher temperature (entry 9). Variation of
the copper salts allowed us to find that [Cu
gave complete conversion, but the enantioselectivity was
lower (entry 10). Cu(acac)2 gave a better enantiomeric
excess (entry 11) but using Cu(OAc)2·H2O improved the
ACHTNUGERTN(NUGN CH3CN)4]BF4
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
conversion with the same enantioselectivity (entry 12). Con-
sequently, all the ligands were tested under these new condi-
10480
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10473 – 10485

SimplePhos Ligands
FULL PAPER
Table 10. Addition of R3Al to 3-aryl-2-cyclohexenones 22 and 23 cata-
Table 11. Addition of Et3Al to 3-methyl-2-cyclopentenone (24) catalysed
lysed by copper in the presence of various SimplePhos ligands.
by CuTC in the presence of various SimplePhos ligands.
Entry
R
Ligand
Conv. [%][a] (yield [%])
ee [%][b]
1
2
3
4
5
6
Et
Et
Et
Et
nBu
nBu
L1
L4
L8
L19
L4
97
100
100
93
27
100 (65)
58
77
71
57
75
74
Entry Substrate
R
CuX
Ligand Conv. [%][a]
(yield [%])
ee
[%][b]
1
2
3
4
5
6
7
8
22
22
22
22
22
22
22
22
23
23
Me CuTC
Me CuTC
Me CuTC
Et CuTC
Et Cu
Et Cu
Et Cu
Et Cu
Me Cu
L1
91
73
81
79
85
82
90
89
90
91
93
L17
L22
L1
100 (72)
68
48
L22
[a] Determined by GC–MS. [b] Determined by chiral GC.
A
80
100 (71)
71
89 (60)
100 (79)
100 (73)
N
ACHTUNGTRENNUNG
The second best, L8, was also the ligand with a non-C2-sym-
metric amine (entry 3). But bulkier substituents on the
diaryl part seemed to be very detrimental; a decrease in
enantioselectivity was observed with ligand L19 (entry 4).
Moreover, electronic effects on the diaryl part did not have
a strong influence on the enantioselectivity. The amine part
was more important for the control of the catalytic system.
Another aluminium reagent, nBu3Al, was also tested. Using
the best conditions with ligand L4, the adduct 43 was ob-
tained in low conversion but with 75% ee (entry 5). After
screening some ligands, we found that ligand L22, which was
inefficient for the addition of Et3Al (not shown in Table 11),
gave the desired compound with complete conversion and
the same enantiomeric excess value (74%; entry 6). This
result shows that the length of the alkyl chain does not have
any influence on the enantioselectivity.
U
9
10
ACHTUNGTRENNUNG
Et CuTC
[a] Determined by GC–MS. [b] Determined by chiral GC.
Then we tested the addition of Me3Al to enone 23 bearing
an electron-withdrawing group (CF3) in the para position.
Copper acetate monohydrate was the best copper salt and
both good conversion and enantioselectivity were obtained
with ligand L22, with up to 91% ee (Table 10, entry 9). Bi-
phenol phosphoramidite-based ligand gave only 93% con-
version and a moderate enantiomeric excess of 66%.[19c] The
addition of Et3Al showed also full conversion and high ee
values of up to 93% with the same ligand (entry 10).
As expected, substrate 23 bearing an electron-withdraw-
ing group on the aromatic moiety had a higher reactivity
than 22 bearing an electron-donating group in the same
para position. Phenylcyclohexenone 21 presented an inter-
mediate reactivity; this is because electron-withdrawing
groups tend to decrease the electronic density of the b posi-
tion and make it more electrophilic and reactive. This
remark is true for the addition of Me3Al. In the case of the
addition of Et3Al, we only observed an improvement of the
conversion.
It was reported that Me3Al was a better reagent than
Me2Zn for the ACA to nitroalkene with phosphoramidite li-
gands.[14] To test if this was also the case with SimplePhos li-
gands, the addition of the Me3Al reagent to nitroalkene 8
was attempted (Scheme 5). Using the simplest ligand, L1,
Finally, an even more challenging substrate was 3-methyl-
cyclopenten-2-one (24), representative of five-membered
ring systems.[19b] For the first attempt, we tested the simple
ligand L1 under the same experimental conditions described
above. Moderate enantioselectivity was observed (up to
58% ee), although with good conversion, whereas the six-
membered ring system gave better results both in terms of
conversion and enantioselectivity with the same ligand
(Table 7, entry 1 versus Table 10, entry 1). We extensively
tried to optimise these conditions but whatever the copper
salt, the solvent, the catalyst loading or the temperature, we
did not manage to improve these results. Consequently, we
have screened some ligands to find a better one (Table 11).
Once again, the ligand with a 2-naphthyl group on the
amine scaffold, L4, gave the best result, even if we did not
manage to obtain more than 77% ee (Table 11, entry 2).
Scheme 5. Addition of Me3Al to nitroalkene 8 catalysed by CuTC and
SimplePhos ligand L1.
we could obtain the adduct with the same enantiomeric
excess value as with Me2Zn (90 versus 89% ee with Me2Zn).
To enlarge the scope of reaction, we examined a tandem
hydroalumination–ACA reaction (Scheme 6).[19c] As with
phosphoramidite ligands, we expected the preferential trans-
fer of the vinylic group. After some experimental optimisa-
tions, the desired product 44 was obtained in full conversion
but with only 50% ee with ligand L8. No transfer of the iBu
group was observed.
To ascertain if the face selectivity of the transfer was the
same as usual, we hydrogenated 44 to 44H (H2, Pd/C,
Chem. Eur. J. 2009, 15, 10473 – 10485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10481

L. Palais and A. Alexakis
Scheme 8. Retrosynthesis.
Scheme 6. Tandem hydroalumination–ACA hydrogenation reaction
(DIBAL-H=diisobutylaluminium hydride).
The synthesis of product 45 was prepared according to the
usual procedure, by addition of the corresponding Grignard
reagent to ethoxycyclohexenone with 86% yield (see the
Supporting Information). Finally, the ACA of the trimethyl-
aluminium reagent to enone 45 catalysed by copper thio-
phene carboxylate in the presence of L22 allowed the for-
mation of the chiral adduct 46 in 65% yield and 94% ee
(Scheme 9).
MeOH). In turn, ent-44H could be obtained by the conju-
gate addition of Me3Al to 18 bearing a long alkyl chain in
the b position (Scheme 7).
Scheme 7. Addition of Me3Al to 3-hexyl-2-cyclohexenone (19) catalysed
by CuTC and SimplePhos ligand L17.
Scheme 9. Synthesis of 46.
The adduct ent-44H was obtained with high yield and
enantioselectivity (up to 93% ee) with ligand L17 and the
absolute configuration of S was assigned. This was deter-
mined by analogy with the general trend observed for the
adducts of R3Al to 3-methylcyclohexen-2-one (17; see
Table 7). Thus, the alkenylalane attacked with the same
facial selectivity as Me3Al and Et3Al when using substrates
17, 18, 20 and 24.
Another interesting target for natural product synthesis
has recently been described. Combining ACA with R3Al
and a iodination of trisubstituted adducts, we have finally
envisaged the formation of a chiral key intermediate for the
synthesis of axane derivatives, which were isolated from the
marine sponge Axinella cannabia. The synthesis of such nat-
ural products has been reported in a racemic version.[24]
The proposed sequential retrosynthesis involves the bi-
cyclic compound 51, which is easily prepared by radical cyc-
lisation of the intermediate 50.[24] This could be obtained by
tandem ACA of Me3Al with a SimplePhos ligand to the tri-
substituted cyclic enone 49 and then an a iodination of the
corresponding enol acetate (Scheme 10).
Synthetic applications: The asymmetric conjugate addition
is an essential method for the specific introduction of a hy-
drocarbon unit to the b position of a carbonyl function. Fur-
thermore, the eminent nucleophilicity of the metal enolate
intermediate allows for reaction with various electrophiles,
thereby affording the a,b-vicinal structural modification of
enones and providing a power-
ful tool for the synthesis of
complex molecules.
To illustrate these two as-
pects, we have applied our
methodology to create stereo-
genic quaternary carbon cen-
tres for the preparation of key
intermediates in the synthesis
Scheme 10. Retrosynthesis.
of natural products. Firstly, we
were interested in the forma-
tion of product 46 in an enantio-enriched version
(Scheme 8), which is an intermediate for the synthesis of
several natural products such as the Vibsane family,[23a–e] iso-
lated from Viburnum awabuki, the axane family[23f] or the
ambra derivatives.[23g] All these previous syntheses were de-
scribed in a racemic version.
We have previously described the formation and the use
of aluminium enolates generated from the ACA of R3Al
with a phosphoramidite ligand to form the corresponding
enol silyl ethers and acetates.[19a] As it was not possible to
functionalise directly the a position of the aluminium eno-
late, probably due to the strong bond between the alumini-
10482
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10473 – 10485

SimplePhos Ligands
FULL PAPER
um and the oxygen atom, we have envisaged the a halogen-
ation of the adduct starting from the corresponding enol
acetate 47. To avoid any problems with the alkyne, a test re-
action was done by ACA of the Et3Al reagent to 3-methyl-
cyclohexenone (17) followed by trapping of the correspond-
ing aluminium enolate with Ac2O.
Many methods have been tested for the a iodination, and
we chose a modified procedure developed by Cort[25] using
CuII salt and iodine as reagents in acetonitrile at room tem-
perature.
The corresponding adduct 48 was obtained in a good
overall yield (70%) and 95% ee (Scheme 11). It is impor-
tant to note that no loss of enantioselectivity was observed
in comparison with the corresponding hydrolysed adduct
(see Table 7, entry 22).
Scheme 12. Synthesis of 50.
Conclusion
In summary, we have described a series of versatile new
phosphorus ligands based on the induced helicity of a diaryl
part and C2-symmetric chiral amine. In many cases, some li-
gands could provide high enantioselectivities, except for the
copper-catalysed addition of the Et2Zn reagent to acyclic
enones and nitroalkenes, for which only moderate ee values
were obtained.
Concerning the formation of stereogenic quaternary
carbon centres by addition of trialkylaluminium species,
some ligands, and particularly L22, afforded much better re-
sults than the parent phosphoramidite-based ligand and
sometimes the best reported in the literature. Ligand L22
showed generally high enantioselectivity on a large scope of
substituted enones and seemed to be an optimal ligand for
this study. Theoretical studies are underway to gain a deeper
insight into the importance of the helicity of the diaryl part
on the phosphorus atom.
Scheme 11. a-Iodination of the adduct 48.
The synthesis of product 49 was prepared according to the
classical procedure by addition of the corresponding
Grignard reagent to ethoxycyclohexenone, which gave 71%
yield (see the Supporting Information).
Finally, SimplePhos ligand L22 showed its efficiency for
the formation of chiral intermediates in the synthesis of nat-
ural products.
Then product 49 could undergo the asymmetric conjugate
addition of Me3Al catalysed by CuTC and the most efficient
ligand, L22. Keeping in mind that no racemisation of the
chiral centre occurred during the formation of the enol ace-
tate nor during the a iodination, the enantiomeric excess
was measured just before quenching the aluminium enolate
with Ac2O. For reasons of chiral separation, the de-silylated
corresponding adduct gave 96% ee (Scheme 12). The abso-
lute configuration was determined on the corresponding sa-
turated product, obtained after hydrogenation, and the
methyl group came, as usual, from the back face of the sub-
strate. Then, after quenching the aluminium enolate with
acetic anhydride, the enol acetate 52 could be reacted with
iodine in presence of a stoichiometric amount of copper ace-
tate to allow the formation of the iodo adduct 50 with 73%
overall isolated yield. Finally, as described in the litera-
ture,[24] we could envisage a radical cyclisation to afford the
bi-cyclic product 51. It is interesting to note that the iodina-
tion procedure did not interfere with the triple bond, nor
with the trimethylsilyl group.
Experimental Section
Full experimental details are provided in the Supporting Information
(111 pages).
Typical procedure for ligand synthesis
Pathway 1: nBuLi (solution in hexane, 1 equiv) was added dropwise to a
stirred solution of amine (1 equiv) in THF at À788C. The reaction mix-
ture was stirred at À788C for 10 min, then allowed to warm up to room
temperature. After stirring for 2 h, the mixture was cooled to À788C and
a solution of Ph2PCl (1 equiv) in THF was slowly added dropwise. The
reaction mixture was allowed to warm to room temperature and was
stirred overnight. A solution of dimethylsulfide complex (2 equiv) was
added dropwise at room temperature. The reaction was stirred overnight
at this temperature. Then the reaction mixture was concentrated under
reduced pressure. The desired product was purified by flash chromatogra-
phy on neutral alumina with toluene as eluent.
Pathway 2: A solution of amine (1 equiv) in THF was added to a stirred
mixture of Et3N (5 equiv) and PCl3 (1 equiv) at 08C in THF under a ni-
trogen atmosphere, and the reaction mixture was stirred at room temper-
Chem. Eur. J. 2009, 15, 10473 – 10485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10483

L. Palais and A. Alexakis
ature for 4 h. Then, the solution was cooled to À308C and a solution of
Grignard reagent (3 equiv) was added slowly. The reaction mixture was
allowed to warm at room temperature and then was heated to reflux
overnight. Then, the solution was directly evaporated under vacuum. The
product was purified by filtration on neutral alumina under an inert at-
mosphere.
Finally, the corresponding adduct was hydrogenated with palladium on
charcoal in MeOH at room temperature over 2 d to afford the adduct
44H after filtration on Celite.
Then, under an inert atmosphere, free ligand was dissolved in dry THF
and a solution of dimethylsulfide complex (2 equiv) was added dropwise
at room temperature. The reaction was stirred overnight at this tempera-
ture. Then the reaction mixture was concentrated under reduced pres-
sure. The desired product was purified by flash chromatography on neu-
tral alumina with toluene as eluent.
Acknowledgements
The authors thank the Swiss National Research Foundation (grant no.
200020-113332) and COST action D40 (SER contact no. C07.0097) for fi-
nancial support, as well as BASF for the generous gift of chiral amines,
Stefan Kerhli for a generous gift of trisubstituted enones and Stephane
Rosset for his help with chiral separations.
Typical procedure for ligand deprotection: Phosphine borane ligand
(1 equiv) was dissolved in diethyl ether under a nitrogen atmosphere and
a solution of morpholine (10 equiv) was added at room temperature. The
solution was stirred at this temperature for 2 d.
[1] a) B. E. Rossiter; N. M. Swingle, Chem. Rev. 1992, 92, 771–806;
N. M. Swingle, Chem. Rev. 1992, 92, 771–806; b) A. Alexakis in
Transition Metal Catalysed Reactions (Eds.: S. I. Murahashi, S. G.
Davies), IUPAC Blackwell Science, Oxford (U.K.), 1999, p. 303;
c) K. Tomioka, Y. Nagaoka in Comprehensive Asymmetric Catalysis
(Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer, New
York, 2000, p. 1105; d) M. P. Sibi, S. Manyem, Tetrahedron 2000, 56,
8033–8061; e) N. Krause, A. Hoffmann-Rçder, Synthesis 2001, 171–
196; f) T. Hayashi, Acc. Chem. Res. 2000, 33, 354–362; g) J. Chris-
toffers, G. Koripelly, A. Rosiak, M. Rçssle, Synthesis 2007, 1279–
1300.
[2] For reviews, see: a) A. Alexakis, C. Benhaim, Eur. J. Org. Chem.
2002, 3221–3223; b) A. Alexakis, J. E. Backvall, N. Krause, O.
Pamies, M. Dieguez, Chem. Rev. 2008, 108, 2796–2823; c) S. R. Har-
utyunyan, T. den Hartog, K. Geurts, A. J. Minnard, B. L. Feringa,
Chem. Rev. 2008, 108, 2824–2852.
After complete deprotection, the solution was filtered on neutral alumina
under a nitrogen atmosphere and with toluene as eluent.
Typical procedure for 3-substituted enone synthesis: A flame-dried flask
was charged with Grignard reagent (2.0 equiv) and cooled to 08C. Then
3-ethoxycyclohex-2-en-1-one (50 mmol) in THF (40 mL) was added drop-
wise. Once the addition was complete, the reaction mixture was left at
room temperature until complete disappearance of the starting material.
The reaction was hydrolysed by the addition of aqueous sulfuric acid
(5% w/w). Et2O (50 mL) was added and the aqueous phase was separat-
ed and extracted further with Et2O (3ꢃ20 mL). The combined organic
fractions were washed with NaHCO3, brine and water, dried over
Na2SO4, filtered and concentrated in vacuo. The oily residue was purified
by flash chromatography on silica gel (pentane/Et2O 5:1).
Typical procedure for enantioselective copper-catalysed conjugate addi-
tion with diethyl zinc: A solution of CuTC (0.025 mmol) and chiral
ligand (0.05 mmol) in dry diethyl ether (2 mL) was stirred for 30 min at
room temperature and then cooled to À308C. Diethyl zinc (1 mL, 1m in
hexane) was slowly added to the reaction mixture and the solution was
stirred for 15 min at the same temperature. Michael acceptor (0.5 mmol)
was then added dropwise in 1 min. The reaction mixture was stirred at
À308C overnight before being quenched by MeOH/saturated aqueous
NH4Cl solution. Enantiomeric excess was determined by chiral GC.
[3] B. L. Feringa, Acc. Chem. Res. 2000, 33, 346–353.
[4] a) A. Alexakis, D. Polet, S. Rosset, S. March, Synlett 2001, 1375–
1378; b) A. Alexakis, D. Polet, S. Rosset, S. March, J. Org. Chem.
2004, 69, 5660–5667.
[5] L. Palais, I. S. Mikhel, C. Bournaud, L. Micouin, C. A. Falciola, M.
Vuagnoux-d’Augustin, S. Rosset, G. Bernardinelli, A. Alexakis,
Angew. Chem. 2007, 119, 7606–7609; Angew. Chem. Int. Ed. 2007,
46, 7462–7465.
Typical procedure for enantioselective copper-catalysed conjugate addi-
tion with trialkylaluminium reagent: A flame-dried Schlenk tube was
charged with copper (5 mol%) and the ligand (10 mol%). Diethyl ether
(1 mL) was added and the mixture was stirred at room temperature for
30 min. Then the Michael acceptor (0.5 mmol) in ether (1 mL) was added
at room temperature and the reaction mixture was stirred for a further
5 min before being cooled to À108C. Then trialkylaluminium (2 equiv)
was introduced dropwise over 1 min. Once the addition was completed,
the reaction was left at À108C overnight. The reaction was hydrolysed by
the addition of MeOH followed by saturated aqueous NH4Cl solution.
Enantiomeric excess was determined by chiral GC.
[6] A. Alexakis, S. Gille, F. Prian, S. Rosset, K. Ditrich, Tetrahedron
Lett. 2004, 45, 1449–1451.
[7] T. Imamoto, Pure Appl. Chem. 1993, 65, 655–660.
[8] a) J. L. Leaser, R. Cvetovich, Org. Synth. 2005, 82, 115; b) A. K.
Beck, P. Gysi, L. La Vecchia, D. Seebach, Org. Synth. 1999, 76, 12.
[9] K. Tissot-Croset, D. Polet, A. Alexakis, Angew. Chem. 2004, 116,
2480–2482; Angew. Chem. Int. Ed. 2004, 43, 2426–2428.
[10] D. Polet, A. Alexakis, Org. Lett. 2005, 7, 1621–1624.
[11] a) S. Jugꢀ, N. Riegel, C. Darcel, O. Stephan, J. Organomet. Chem.
1998, 567, 219–223; b) S. Jugꢀ, C. Darcel, E. B. Kaloum, R. Merdꢁs,
D. Moulin, N. Riegel, S. Thorimbert, J. P. GenÞt, J. Organomet.
Chem. 2001, 624, 333–343.
[12] O. Andrey, A. Alexakis, A. Tomassini, G. Bernardinelli, Adv. Synth.
Catal. 2004, 346, 1147–1168.
[13] A. Alexakis, D. Polet, C. Benhaim, S. Rosset, Tetrahedron: Asymme-
try 2004, 15, 2199–2203.
[14] D. Polet, A. Alexakis, Tetrahedron Lett. 2005, 46, 1529–1532.
[15] Quaternary Stereocenters: Challenges and Solutions for Organic Syn-
thesis (Eds.: J. Christoffers, A. Baro), Wiley-VCH, Weinheim (Ger-
many), 2006.
[16] For reviews, see: a) P. G. Cozzi, R. Hilgraf, N. Zimmermann, Eur. J.
Org. Chem. 2007, 36, 5969–5994; b) B. M. Trost, C. Jiang, Synthesis
2006, 3, 369–396; c) J. Christoffers, A. Baro, Adv. Synth. Catal.
2005, 347, 1473–1482.
Tandem hydroalumination—ACA procedure: A 1n solution of diisobuty-
laluminium hydride in n-heptane (1 mmol, 1 mL) was added to 1-hexyne
(1 mmol, 88 mL) while maintaining the temperature below 408C. When
the initial exothermic reaction had subsided, the reaction mixture was
heated for 2 h at 508C. The vinylalane formed was used in solution in
heptane. A flame-dried Schlenk tube was charged with CuTC (28.8 mg,
30 mol%) and the chiral ligand L8 (68.9 mg, 30.0 mol%). Dry Et2O
(1 mL) was added and the mixture was stirred at room temperature for
20 min. Then, the reaction mixture was cooled to À208C and the freshly
prepared vinylalane in solution in heptane was introduced dropwise over
1 min. After 15 min at this temperature, the 3-methylcyclohexenone
(48 mL, 0.5 mmol) in Et2O (1 mL) was added dropwise and, once the ad-
dition was complete, the reaction mixture was left at À208C overnight.
The reaction was hydrolysed by the addition of MeOH at À308C, fol-
lowed by 2n HCl (3 mL) at room temperature. Diethyl ether (10 mL)
was added and the aqueous layer was separated and extracted further
with diethyl ether (3ꢃ3 mL). The combined organic fractions were
washed with brine (5 mL), dried over anhydrous sodium sulfate, filtered
and concentrated in vacuo.
[17] J. Wu, D. M. Mampreian, A. H. Hoveyda, J. Am. Chem. Soc. 2005,
127, 4584–4585.
[18] a) A. W. Hird, A. H. Hoveyda, J. Am. Chem. Soc. 2005, 127, 14988–
14989; b) E. Fillion, A. Wilsily, J. Am. Chem. Soc. 2006, 128, 2774–
2775; c) A. Wilsily, E. Fillion, Org. Lett. 2008, 10, 20801–20804.
10484
www.chemeurj.org
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 10473 – 10485

SimplePhos Ligands
FULL PAPER
[19] a) M. d’Augustin, L. Palais, A. Alexakis, Angew. Chem. 2005, 117,
1400–1402; Angew. Chem. Int. Ed. 2005, 44, 1376–1378; b) M.
Vuagnoux-d’Augustin, A. Alexakis, Synlett 2007, 2057–2060; c) M.
Vuagnoux-d’Augustin, A. Alexakis, Chem. Eur. J. 2007, 13, 9647–
9662; d) D. Martin, S. Kehrli, M. d’Augustin, H. Clavier, M. Maud-
uit, A. Alexakis, J. Am. Chem. Soc. 2006, 128, 8416–8417; e) C.
Hawner, K. Li, V. Cirriez, A. Alexakis, Angew. Chem. 2008, 120,
8334–8337; Angew. Chem. Int. Ed. 2008, 47, 8211–8214; f) H.
Hꢀnon, M. Mauduit, A. Alexakis, Angew. Chem. 2008, 120, 9262–
9264; Angew. Chem. Int. Ed. 2008, 47, 9122–9124.
[20] a) K. S. Lee, M. K. Brown, A. W. Hird, A. H. Hoveyda, J. Am.
Chem. Soc. 2006, 128, 7182–7184; b) M. K. Brown, T. L. May, C. A.
Baxter, A. H. Hoveyda, Angew. Chem. 2007, 119, 1115–1118;
Angew. Chem. Int. Ed. 2007, 46, 1097–1100; c) T. L. May, M. K.
Brown, A. H. Hoveyda, Angew. Chem. 2008, 120, 7468–7472;
Angew. Chem. Int. Ed. 2008, 47, 7358–7362.
[22] a) W. Giersch, G. Ohloff, patent no. DE2938979, p. 9; b) R. S. Bho-
sale, S. V. Bhosale, T. Wang, P. K. Zubaidha, Tetrahedron Lett. 2004,
45, 7187–7188; c) M. Curini, F. Epifano, S. Genovese, Tetrahedron
Lett. 2006, 47, 4697–4700.
[23] a) M. Gallen, C. M. Williams, Eur. J. Org. Chem. 2008, 4697–4705;
b) M. Gallen, C. M. Williams, Org. Lett. 2008, 10, 713–715; c) B. D.
Schwartz, D. P. Tilly, R. Heim, S. Wiedemann, C. M. Williams, P. V.
Bernhardt, Eur. J. Org. Chem. 2006, 3181–3192; d) D. P. Tilly, C. M.
Williams, P. V. Bernhardt, Org. Lett. 2005, 7, 5155–5157; e) R.
Heim, S. Wiedemann, C. M. Williams, P. V. Bernhardt, Org. Lett.
2005, 7, 1327–1329; f) A.-C. Guevel, D. J. Hart, J. Org. Chem. 2008,
73, 473–479; g) G. Ohloff, F. Naef, R. Decorzant, W. Thomen, E.
Sundt, Helv. Chim. Acta 1973, 56, 1414–1448.
[24] Y. L. Kuo, M. Dhanasekaran, C. K. Sha, J. Org. Chem. 2009, 74,
2033–2038.
[25] A. D. Cort, J. Org. Chem. 1991, 56, 6708–6709.
[21] Y. Matsumoto, K. Yamada, K. Tomioka, J. Org. Chem. 2008, 73,
4578–4581.
Received: June 10, 2009
Published online: August 28, 2009
Chem. Eur. J. 2009, 15, 10473 – 10485
ꢂ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10485