Copper-catalyzed asymmetric conjugate addition of aryl aluminum reagents to trisubstituted enones: Construction of aryl-substituted quaternary centers

Article abstract of DOI:10.1002/anie.200803436

(Chemical Equation Presented) Al be back: Novel aryl and alkenyl aluminum reagents are generated through a halogen/Li exchange-Li/Al transmetalation sequence. These aryl alanes are used in the copper-catalyzed asymmetric conjugate addition reaction to a variety of cyclic enones giving chiral aryl-substituted quaternary centers (see scheme). Both, electron-donating and electron-withdrawing groups give full conversion and very good ee values.

Full text of DOI:10.1002/anie.200803436

Angewandte
Chemie
DOI: 10.1002/anie.200803436
Asymmetric Catalysis
Copper-Catalyzed Asymmetric Conjugate Addition of Aryl Aluminum
Reagents to Trisubstituted Enones: Construction of Aryl-Substituted
Quaternary Centers**
Christine Hawner, Kangying Li, Virginie Cirriez, and Alexandre Alexakis*
In the field of asymmetric conjugate addition (ACA) to
enones, much effort has been directed towards the develop-
ment of copper- and rhodium-catalyzed reactions of alkylme-
tal as well as aryl- and vinylboronic acid reagents.[1] While
most reports deal with disubstituted substrates, the construc-
tion of enantioenriched all-carbon quaternary stereocenters is
still a synthetic challenge.[2] Recently, a number of interesting
reports concerning the copper-catalyzed version provided
solutions to this problem. However, the use of the classical
alkylzinc reagents requires special conditions, such as reactive
substrates, for example, nitroalkenes[3] and doubly activated
enones.[4] Less-reactive trisubstituted cyclic enones are only
accessible with specialized N-heterocyclic carbene (NHC)
ligands[5] or more reactive nucleophiles, such as, trialkylalu-
minum reagents. In combination with stronger coordinating
solvents, their enhanced Lewis acidity has made trialkylalu-
minum reagents very successful in the ACA.[6] Grignard
reagents are also useful owing to their wide scope of reaction,
but the enantioselectivities are generally lower than with Al
and Zn reagents.[7]
Although quaternary centers are ubiquitous motifs in
natural and pharmaceutical products, only few reports deal
with their formation through the ACA. Most examples are
limited to the addition of phenyl and p-anisyl,[5,7] the aryl
group is already present in the substrate[6c,d] or activated
substrates[8] are required. To date, no general method has
been developed to introduce a wide range of aromatic groups
to trisubstituted enones. In our search for a solution to this
problem, we turned our attention to aluminum reagents, as
they had proven to be very successful in the introduction of
alkyl substituents. However, aryl aluminum reagents are not
commercially available. Similarly, the commercial availability
of aryl zinc reagents is restricted to diphenyl zinc and other
routes have been developed to obtain the desired compounds,
such as transmetalation of arylboronic acids with diethyl
zinc[9] or of aryl lithiums with zinc dichloride.[8b,10]
triethylaluminum was performed (Scheme 1). Three equiv-
alents of triethylaluminum were necessary for the ACA to
take place. We reasoned that in a first step, 2 equivalents of
Scheme 1. Transmetalation of phenylboronic acid to triethylaluminum.
CuTC=copper thiophencarboxylate.
Et3Al reacted with 1 equivalent of PhB(OH)2, which was
followed by the phenyl transfer from borane to the third e-
quivalent of Et3Al to form the phenyl alane. However, the
ethyl transfer to the substrate 3-methylcyclohex-2-enone (1)
lead to 36% of undesired byproduct.
Therefore, we were looking for another option to generate
aryl alanes and turned towards a sequence consisting of
halogen–metal exchange between an aryl halide and nBuLi
giving the corresponding aryl lithium compound which was
then transmetalated with a dialkylaluminum chloride reagent
to generate the desired dialkyl aryl aluminum species in situ
(Scheme 2).[11] The diethylphenyl alane thus generated was
used as nucleophile in the copper-catalyzed ACA of 3-
methylcyclohex-2-enone.
Scheme 2. I/Li-exchange–Li/Al-transmetalation sequence.
We therefore envisaged generating aryl alanes in similar
ways. First, the phenyl transfer from phenylboronic acid to
We optimized the experimental conditions using commer-
cial PhLi for the transmetalation step generating PhAlEt2. On
the basis of work by Hoveyda et al,[8b] we decided to start our
investigations with 9 equivalents of PhAlEt2 at À308C and to
use only the supernatant solution, as LiCl precipitated from
the reaction mixture (Scheme 3).
[*] C. Hawner, Dr. K. Li, V. Cirriez, Prof. Dr. A. Alexakis
Department of Organic Chemistry, University of Geneva
30 quai ernest Ansermet, 1211 Geneve 4 (Switzerland)
Fax: (+41)223-793-215
E-mail: alexandre.alexakis@chiorg.unige.ch
The reaction was very selective towards the aryl transfer
and went to full conversion giving 97% ee. Et2AlCl proved to
[**] We thank the European Community (Marie Curie Early Stage
Researcher Training, MEST-CT-2005-019780) for financial support.
be
a better transmetalation agent than Me2AlCl and
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200803436.
iBu2AlCl. From a large number of phosphoramidite ligands
Angew. Chem. Int. Ed. 2008, 47, 8211 –8214
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8211

Communications
alane. Under these conditions, the reaction lead to full
conversion and the same ee values as obtained using the
commercial PhLi.
To investigate the scope of the reaction, we decided to use
a variety of different aryl groups and a vinyl alane in this
transformation (Table 1). As SimplePhos ligands give higher
yields and ee values than phosphoramidite ligands in the
addition of AlMe3 to 3-phenylcyclohex-2-enone,[6c] we tested
this new kind of ligand under the optimized reaction
conditions. This study showed that a wide range of electron-
donating and electron-withdrawing groups, which can be
substituted in para- or meta-position are successful in this
reaction giving full conversion and very high ee values. Only
the o-substituted aryl groups required special conditions.
Whereas ligand L12 led to only moderate conversion and low
ee values, the use of SimplePhos L13 improved the results.
Overnight reaction at 108C achieved full conversion and
84.6% ee (Table 1, entry 8). Aryl bromides could also be used
for the generation of the corresponding alane, as shown by 2-
bromonaphthalene (Table 1, entry 13). Although a catalyst
loading below 5 mol% was ineffective, 5 mol% of copper salt
and ligand were sufficient for the addition of a simple phenyl
Scheme 3. Conjugate addition with PhAlEt2 as the nucleophile.
Table 1: Screening of nucleophiles.
(see Supporting Information for
results) and eleven different
copper salts, L12 and CuTC were
chosen (Scheme 3). Interestingly,
the ligand had a significant influ-
ence on conversion and especially
ee value, whereas almost all the
copper salts gave the same results.
The use of less than 9 equivalents
of phenyl alane was detrimental for
the conversion and both conver-
sion and excess decreased when
THF was used as solvent.
Since the separation of the
supernatant from the resulting
salts is not convenient, we added
the whole suspension to the mix-
ture of CuTC and L12 in Et2O to
see if the salts influenced the
reaction. This was not the case
and exactly the same results were
obtained as in the salt-free reac-
tions.
Entry
R
CuTC [mol%]
L [mol%]
L*
Product
Yield [%]
ee [%]
1
Ph
5
5
11
11
5.5
11
10
11
11
11
11
5.5
10
11
11
11
11
10
L12
L12
L13
L12
L12
L12
L12
L13
L12
L12
L12
L12
L12
L12
L12
L12
L12
2
2
2
3
3
3
4
5
6
7
7
7
8
9
10
11
12
n.d.[c]
75
97
96.5
94
2
3
4
5
6
7
Ph
Ph
10
10
5
10
5
10
10
10
10
5
n.d.[c]
0
m-MeC6H4
m-MeC6H4
m-MeC6H4
p-MeC6H4
o-MeC6H4
m-OMeC6H4
p-OMeC6H4
p-OMeC6H4
p-OMeC6H4
2-naphthyl
m-CF3C6H4
p-CF3C6H4
m-BrC6H4
87
96.8
97.8
96.3
84.6
98
n.d.[c]
82
8[d]
9
82
66
81
0
10
11
12
13[a]
14
15
16
17[b]
95
5
n.d.[c]
59
97.2
98.6
94.9
98.4
97.7
82
10
10
10
10
10
73
74
76
93
=
Bu-CH CH
We were interested in adding
substituted aryl groups for which
the lithiated precursor is not com-
mercially available. Therefore, we
conducted further optimization
[a] 2-Bromonaphthalene served as aryl halide. [b] The reaction was stirred for 19 h at À308C, conversion
was 80%. [c] n.d.=not determined. [d] The reaction was stirred for 21 h at À108C.
with in situ prepared PhLi, which was generated through I/
Li-exchange from PhI. However, no conversion occurred. For
the tandem hydroalumination–ACA reaction of vinyl alanes,
30 mol% of CuX and 30 mol% of ligand are necessary for the
reaction to take place.[6d] Therefore, we increased the catalyst
loading to 10 mol% using equimolar amounts of copper and
ligand. At the same time, we added only 3 equivalents of
group (Table 1, entry 1), but double this amount proved to be
necessary for substituted aryl groups. In some cases, a
5:10 mol% ratio of CuTC:L12 was successful (Table 1,
entries 6 and 12). The absolute configuration of the phenyl
adduct was determined to be R, and by extension we ascribe
this configuration to the other aryl substituents as well. Our
methodology can be extended to vinyl alanes by starting from
8212
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 8211 –8214

Angewandte
Chemie
Table 2: Screening of substrates.[a]
the corresponding iodide (Table 1, entry 17). The desired
adduct, which also had the R-configuration, could be obtained
with higher ee value and lower catalyst loading than previ-
ously reported for the tandem hydroalumination–A.C.A.[6d]
To understand why a higher catalyst loading was necessary
when the in situ prepared PhLi was used instead of the
commercial one, we studied the influence of nBuI and LiI in
the ACA. The commercial PhLi is free of these two reagents.
However, 1 equivalent of nBuI is formed during the I/Li-
exchange step, and LiI, which is soluble in Et2O, can be
generated through the attack of PhLi on nBuI, even though
this reaction is very slow at low temperature (Scheme 4).[12]
Entry
Product
Ligand Conversion [%][b] Yield [%]
ee [%]
1[c]
L12
L13
94
91
77
58
95
2
98.6
96.5
77.5
3[d]
L13
L13
99
62
67
4[e]
100
Scheme 4. Generation of LiI during the I/Li-exchange.
[a] Reaction conditions: 10 mol% CuTC, 11 mol% ligand, Et2O, over-
night, À108C. [b] Determined by GC-MS. [c] 13% ethyl transfer deter-
mined by GC-MS. [d] 3% ethyl transfer determined by GC-MS. [e] 7%
ethyl transfer determined by GC-MS.
Although no traces of nBuPh were detectable by GC-MS
below À158C, we wondered if even small, undetectable
amounts of LiI could disrupt the reaction. Therefore, we
added nBuI and LiI to the commercial PhLi to study the
influence of these two reagents on the reaction (Scheme 5).
reaction times were needed to give satisfactory degrees of
conversion. Unfortunately, these conditions lead to an
increase of ethyl transfer with the exception of the iBu-
substituted cyclohexenone, where no ethyl adduct was
detected. However, in all cases the product could be isolated
in good yields. Furthermore, the more reactive SimplePhos
ligand L13 gave better conversion and slightly higher
ee values for products 14–16 than the equivalent phosphor-
amidite ligand.
In summary, we have developed the first copper-catalyzed
ACA, which allows the addition of a wide range of aryl alanes
to 3-methylcyclohex-2-enone with excellent conversion and
ee values leading to the formation of chiral aryl-substituted
quaternary centers. Furthermore, we have shown that this
reaction is general to a variety of substrates carrying different
b-substituents or having different ring sizes. The desired alane
can easily be generated from the corresponding aryl iodide or
bromide through an I/Li-exchange–Li/Al-transmetalation
sequence and no separation of the resulting insoluble LiCl
is necessary, which makes this method highly convenient.
Even vinyl alanes can be generated in this way and used in the
ACA successfully.
Scheme 5. Study of the role of LiI and nBuI in the ACA.
While nBuI did not influence the reaction at all, the
presence of LiI prevented any conversion from taking place.
Consequently, careful temperature control during the I/Li
exchange is necessary to restrain the formation of LiI.
We were also interested in understanding if LiI interacted
with the ligand or any other reagent of the transformation.
However, when LiI was added to a solution of L12 in Et2O
and a small amount of C6D6, no change in the shift of the
31P NMR signal could be detected. We therefore suppose that
LiI interacts at a different point of the reaction.
Experimental Section
Typical procedure: The aryl iodide (1.05 mmol) was dissolved in Et2O
(0,5 mL) under inert atmosphere and cooled to À558C before the
addition of nBuLi (3.15 mmol, 656 mL, 1.6m in hexanes). The mixture
was stirred for 30 min at À558C. Then, Et2AlCl (1.05 mmol, 1.05 mL,
1m in hexanes) was added dropwise and stirred for 30 min between
À50 and À208C. In a separate flask, the copper salt (10 mol%), the
ligand (11 mol%), and Et2O (1.5 mL) were stirred thoroughly at
room temperature for 30 min. The flask was cooled to À308C and the
alane suspension including the salts was added to the copper/ligand
mixture. After 5 min, the substrate (0.35 mmol) was introduced. After
5 h at À308C the reaction was quenched with 10% HCl (20 mL).
Et2O was added (20 mL) and the layers were separated. The aqueous
Finally we tested differently substituted cyclohexenones
as well as enones with different ring sizes (Table 2). After
optimization of the reaction conditions, the phenyl alane
could also add to other substrates giving excellent ee values in
all cases except with 3-methylcyclopent-2-enone (Table 2,
entry 4), where only 77.5% ee was obtained. This result is,
however, still the highest excess ever reported for the Michael
addition to this substrate. However, the reaction conditions
and the ligand had to be adapted to the more challenging
substrates. We found that higher temperature and longer
Angew. Chem. Int. Ed. 2008, 47, 8211 –8214
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8213

Communications
phase was extracted one more time with Et2O (20 mL). The combined
[6] a) M. dꢁAugustin, L. Palais, A. Alexakis, Angew. Chem. 2005,
117, 1400 – 1402; Angew. Chem. Int. Ed. 2005, 44, 1376 – 1378;
b) N. Fuchs, M. dꢁAugustin, M. Humam, A. Alexakis, R. Taras, S.
Gladiali, Tetrahedron: Asymmetry 2005, 16, 3143 – 3146; c) L.
Palais, I. S. Mikhel, C. Bournaud, L. Micouin, C. A. Falciola, M.
Vuagnoux-dꢁAugustin, S. Rosset, G. Bernardinelli, A. Alexakis,
Angew. Chem. 2007, 119, 7606 – 7609; Angew. Chem. Int. Ed.
2007, 46, 7462 – 7465; d) M. Vuagnoux-dꢁAugustin, A. Alexakis,
Chem. Eur. J. 2007, 13, 9647 – 9662; e) M. Vuagnoux-dꢁAugustin,
S. Kehrli, A. Alexakis, Synlett 2007, 2057 – 2060.
organic phases were extracted once with NaOH (1m, 20 mL) and once
with H2O (20 mL). The organic layers were combined and dried over
MgSO4. The solution was then collected by filtration over silica, and
concentrated in vacuo. The crude was purified by flash chromatog-
raphy on a silica column with a mixture of pentane/Et2O (9:1) as
eluent.
Received: July 16, 2008
Published online: September 22, 2008
[7] a) D. Martin, S. Kehrli, M. dꢁAugustin, H. Clavier, M. Mauduit,
A. Alexakis, J. Am. Chem. Soc. 2006, 128, 8416 – 8417; b) Y.
Matsumoto, K. Yamada, K. Tomioka, J. Org. Chem. 2008, 73,
4578 – 4581.
[8] a) A. J. Wilsily, E. Fillion, J. Am. Chem. Soc. 2006, 128, 2774 –
2775; 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) A. Wilsily, E. Fillion, Org. Lett. 2008, 10,
2801 – 2804.
[9] J. Rudolph, C. Bolm, J. Am. Chem. Soc. 2002, 124, 14850 –
14851.
[10] a) D. Peꢂa, F. Lꢃpez, S. R. Harutyunyan, A. J. Minnaard, B. L.
Feringa, Chem. Commun. 2004, 1836 – 1837; b) J. G. Kim, P. J.
Walsh, Angew. Chem. 2006, 118, 4281 – 4284; Angew. Chem. Int.
Ed. 2006, 45, 4175 – 4178.
Keywords: asymmetric catalysis · copper · enones ·
Michael addition · quaternary stereocenters
.
[1] For reviews on asymmetric conjugate additions, see: a) A.
Alexakis, C. Benhaim, Eur. J. Org. Chem. 2002, 3221 – 3226;
b) N. Krause, A. Hoffmann-Rꢀder, Synthesis 2001, 171 – 196;
c) T. Hayashi, Acc. Chem. Res. 2000, 33, 354 – 362.
[2] a) C. Jiang, B. M. Trost, Synthesis 2006, 369 – 396; b) A. Baro, J.
Christoffers, Adv. Synth. Catal. 2005, 347, 1473 – 1482.
[3] J. Wu, D. M. Mampreian, A. H. Hoveyda, J. Am. Chem. Soc.
2005, 127, 4584 – 4585.
[4] A. W. Hird, A. H. Hoveyda, J. Am. Chem. Soc. 2005, 127, 14988 –
14989.
[5] K. Lee, K. Brown, A. W. Hird, A. H. Hoveyda, J. Am. Chem.
Soc. 2006, 128, 7182 – 7184.
[11] M. Oishi, in Science of Synthesis, Vol. 7 (Ed.: H. Yamamoto),
Thieme, Stuttgart, 2004, p. 261.
[12] G. Cahiez, D. Bernard, J. F. Normant, Synthesis 1976, 245 – 248.
8214
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 8211 –8214