Enantioselective bimetallic catalysis of michael additions forming quaternary stereocenters

Article abstract of DOI:10.1002/anie.200803539

Robotlike: Low catalyst loadings of a planar-chiral ferrocenyl bispalladacycle are sufficient to catalyze the Michael addition of trisubstituted α-cyanoacetates to enones with excellent yields (TONs up to 2450) and high enantioselectivity. The reaction proceeds by a cooperative bimetallic mechanism and is superior to previous methods relying on soft Lewis acid catalysts. (Chemical Equation Presented)

Full text of DOI:10.1002/anie.200803539

Communications
DOI: 10.1002/anie.200803539
Bimetallic Catalysis
Enantioselective Bimetallic Catalysis of Michael Additions Forming
Quaternary Stereocenters**
Sascha Jautze and Renꢀ Peters*
Direct conjugate additions of a-carbonyl-stabilized nucleo-
philes to activated olefins are among the most attractive
also promote the same reaction utilizing iPr2NEt as cocata-
lyst,[6] but with low enantioselectivity. With a sterically
demanding PdII–pincer complex, Uozumi et al. later achieved
good enantioselectivity under similar reaction conditions yet
À
reactions for C C bond constructions owing to their ideal
atom economy and the versatility of the activating functional
groups involved. For catalytic asymmetric versions, a high
level of efficiency has been demonstrated with 1,3-dicarbonyl-
based nucleophiles.[1] In contrast, the realization of a general,
practical, highly active, and highly enantioselective catalyst
for the conjugate addition of a-cyanoacetates to enones
remains elusive. This might be explained by the fact that a-
cyanoacetates are incapable of two-point binding to a Lewis
acid. In this study we were particularly interested in the direct
Michael addition of trisubstituted a-cyanoacetates to enones,
found the same limitation with a-Me substituents.[7]
A
conceptually different approach was developed by Jacobsen
et al., who employed a dimeric O-bridged Al–salen com-
plex.[8] In contrast to the soft Lewis acid catalysts, this catalyst
tolerated an a-phenyl-substituted a-cyanoacetate. The appli-
cation of a variety of a-aryl- and a-amino-substituted a-
cyanoacetates was described for the addition to a,b-unsatu-
rated imides without the necessity of an additional base.[9] The
use of unsubstituted vinyl acceptors was not mentioned in this
study.
Herein we report the application of the bispalladacycle
complex FBIP-Cl which exploits the principal advantages of
soft Lewis acids like high catalytic activity as a consequence of
low oxophilicity, resulting in negligible product inhibition,[10]
and overcomes the narrow structural restrictions for the
previously reported late-transition-metal catalysts. The
rationale behind this development was that a soft bimetallic
complex capable of simultaneously activating both substrates
would not only lead to superior catalytic activity, but also to
an enhanced level of stereocontrol as a result of the highly
organized transition state: the a-cyanoacetate should be
activated by enolization promoted by coordination of the
nitrile moiety to one PdII center, while the enone should be
activated as an electrophile by coordination of the olefinic
double bond to the carbophilic Lewis acid. Cooperative
reactivity between two metal centers has been suggested for
enzymatic systems[11] and is emerging as an intriguing design
principle for artificial catalysts.[12]
À
in light of the demand for efficient catalytic asymmetric C C
bond-forming methods that create substituted quaternary
stereocenters[2] and thus provide access to broadly useful
multifunctional chiral building blocks.[3]
Enolate formation by deprotonation of trisubstituted a-
cyanoacetates with a Brønsted base such as a tertiary amine
can trigger the conjugate addition to enones,[4] but the basic
conditions might also induce various side reactions with base-
sensitive functionalities. To obtain synthetically useful enan-
tioselectivities and yields, low-temperature reaction techni-
ques, high catalyst loadings, and extended reaction times are
usually required. In their seminal study in 1992, Ito and co-
workers reported that a RhI complex bearing a trans-chelating
diphosphine ligand is able to catalyze the addition of a-
cyanopropionate to vinyl ketones with high enantioselectivity
in the absence of a base.[5] Unfortunately, a substituents
bulkier than Me impeded valuable enantioselectivities. Sub-
sequently, Richards et al. found that PdII–pincer complexes
Bispalladacycle FBIP-OTs, which was generated in situ
from FBIP-Cl by treatment with AgOTs,[13] was indeed able
to smoothly catalyze the addition of a-phenyl-substituted
cyanoacetate 1Aa (R = Me) to methyl vinyl ketone (MVK)
(precatalyst loading 0.5 mol%), albeit with poor enantiose-
lectivity (Table 1, entry 1).[14,15] The enantioselectivity was
considerably increased by use of bulky ester groups, though at
the expense the reaction rate (Table 1, entries 2 and 4; initial
[*] S. Jautze, Prof. Dr. R. Peters
Laboratory of Organic Chemistry, ETH Zꢀrich
Wolfgang-Pauli-Strasse 10, Hꢁnggerberg HCI E 111
8093 Zꢀrich (Switzerland)
Prof. Dr. R. Peters
New address: Institut fꢀr Organische Chemie, Universitꢂt Stuttgart
Pfaffenwaldring 55, 70569 Stuttgart (Germany)
Fax: (+49)711-685 64321
reaction rates at c = 0.20 molLÀ1: 1Ab: 40.6 mmolLÀ1 hÀ1
;
E-mail: rene.peters@oc.uni-stuttgart.de
[**] This work was supported financially by TH research grant TH-01/07-
1 and F. Hoffmann-La Roche. We thank Priv.-Doz. Dr. Martin Karpf
and Dr. Paul Spurr (both F. Hoffmann-La Roche, Synthesis and
Process Research) for carefully reading this manuscript. Moreover,
we are grateful to Paul Seiler (ETHZ) for X-ray crystal structure
analysis and Stefan Diethelm (ETHZ) for skillful experimental work
during a semester internship.
1Ad: 18.4 mmolLÀ1 hÀ1). To increase the reactivity of the tert-
butyl ester 1Ad, various solvents were screened. The reaction
medium was found to have a strong influence: the enantio-
selectivity decreased in all solvents tested relative to the
selectivity in CH2Cl2, while a significantly enhanced reaction
rate was noticed in cyclohexane, Et2O, diglyme, and EtOH
(Table 1, entries 7, 8, 11, and 13). Whereas in the protic
solvent EtOH, nearly racemic product was formed, the
reaction in in diglyme showed promising selectivity, which
Supporting information for this article including experimental
details is available on the WWW under http://dx.doi.org/10.1002/
anie.200803539.
9284
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 9284 –9288

Angewandte
Chemie
Table 1: Optimization of the model reaction.[a]
cal reasons tBu esters were chosen for the investigation of the
reaction scope.
Significantly, reactions of a range of a-aryl-a-cyanoace-
tate donors with various vinyl ketone acceptors proceeded
with excellent yield and high enantioselectivity (Table 2). The
Table 2: Scope and limitations of the reaction.[a]
Entry
1A
R
Solv.
Ar
Additive
(mol%)
Yield
[%][b]
ee
[%][c]
R1
R2
2
FBIP-Cl Yield ee
1
2
3
4
5
6
7
8
1Aa
1Ab
1Ac
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ad
1Ae
Me
Et
Bn
tBu
tBu
tBu
tBu
tBu
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CHCl3
toluene
c-C6H12
Et2O
p-Tol
p-Tol
p-Tol
p-Tol
p-Tol
p-Tol
p-Tol
p-Tol
p-Tol
p-Tol
p-Tol
p-Tol
p-Tol
Mes
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
95
93
>99
81
60
71
96
94
69
49
97
13
>99
99
>99
99
99
>99
94
93
>99
11
33
41
81
69
56
62
54
46
38
75
11
4
84
86
90
90
90
88
86
95
Entry
1
[mol%] [%][b] [%][c]
1
1Ad Ph
1Ad Ph
1Ad Ph
1Ad Ph
1Ad Ph
1Ad Ph
1Ad Ph
1Bd p-Br-C6H4
1Bd p-Br-C6H4
1Cd p-Cl-C6H4
Me
Me
Et
n-Pr
n-Pent
Ph
2Ada 0.2
2Ada 0.2
2Adb 0.75
2Adc 1.0
2Add 1.0
2Ade 1.0
>99 90
99 89
>99 88
2[d]
3
4
5
6
7
90
98
95
80
86
89
77
76
9
tBu THF
p-MeO-C6H4 2Adf 1.0
10
11
12
13
14
15
16
17
18[d]
19[e]
20[e]
21
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
tBu
R[f]
DME
8
Me
Me
Me
Me
Me
Et
Me
Me
Me
Me
Me
2Bda 0.1
2Bda 0.05
2Cda 0.1
2Cda 0.05
2Cda 0.02
2Cdb 0.5
2Dda 0.5
2Dda 0.2
2Eda 0.25
2Fda 0.5
2Gda 0.5
2Hda 0.5
2Ida 1.0
2Jda 0.5
>99 94
>99 89
>99 95
diglyme
MeCN
EtOH
9[e]
10
11[f] 1Cd p-Cl-C6H4
99
98
91
85
diglyme
12[e,f] 1Cd p-Cl-C6H4
diglyme Tipp
diglyme Tipp
diglyme Tipp
diglyme Tipp
diglyme Tipp
diglyme Tipp
diglyme Tipp
13
14
15
16
17
18
19
20
21
1Cd p-Cl-C6H4
1Dd p-F-C6H4
1Dd p-F-C6H4
1Ed p-Me-C6H4
1Fd m-Br-C6H4
1Gd m-Cl-C6H4
1Hd m-CF3-C6H4 Me
1Id m-MeO-C6H4 Me
1Jd m-Me-C6H4
>99 94
>99 95
>99 93
>99 91
>99 91
>99 91
>99 91
>99 80
>99 90
HOAc (10)
HOAc (1)
HOAc (20)
HOAc (20)
HOAc (10)
HOAc (20)
[a] 25 mg 1A, 2.0 equiv MVK, 0.5 mL solvent if not mentioned otherwise.
[b] Determined by 1H NMR spectroscopy. [c] Determined by HPLC
methods. [d] 0.2 mol% FBIP-Cl, 1.2 mol% Ag salt, 110 mL diglyme.
[e] 0.1% FBIP-Cl, 0.6% Ag salt, 110 mL diglyme. [f] R=CH(iPr)2, 250 mg
1Ae,1.4 mL diglyme. Mes=2,4,6-Me3-C6H2; Tipp=2,4,6-iPr3-C6H2.
Me
[a] 100–250 mg 1, 2.0 equiv MVK, 0.2–1.4 mL diglyme. [b] Yield of
isolated product. [c] Determined by HPLC. [d] 1.2 equiv MVK. [e] Reac-
tion time 48 h. [f] 408C.
could be further enhanced by increasing the steric demand of
the sulfonate counteranion without diminishing the catalytic
activity (Table 1, entries 14 and 15).
class of targeted products was recently shown to be highly
useful for the synthesis of enantioenriched esters of b2,2-amino
acids.[4b] As shown in entry 2 of Table 2, a large excess of
MVK was not required. The reaction rate slowed with
increasing size of the alkyl substituent R2 on the vinyl
ketone, and precatalyst loadings of up to 1.0 mol% were
required to attain quantitative yields (Table 2, entries 2–4),
though the enantioselectivity was not considerably influ-
enced. In contrast, the reactions of aryl vinyl ketones were
less enantioselective (76–77% ee, Table 2, entries 5 and 6).
On the a-aryl substituent R1, both electron-withdrawing and
-donating substituents were well tolerated (Table 2, entries 8–
21). Acceptor substituents in para position (Table 2, entries 8–
15) slightly increased the reactivity and asymmetric induction
and made it possible to further decrease the precatalyst
loading to 0.02% (Table 2, entry 12, turnover number: 2450
(monomeric catalyst)), while alkyl donors decelerated the
process only a little. Substituents in meta position generally
retarded the addition, necessitating higher precatalyst load-
ings, but they had no negative impact on the enantiomer ratio
The results from entries 1–15 in Table 1 were obtained
with technical-grade MVK. Surprisingly, purification of MVK
by distillation led to a sharp decrease in enantioselectivity
(74% ee, 0.2% mol% FBIP-Cl, 1.2% mol% Ag-O3S-Tipp,
2 equiv MVK, diglyme, RT). Since commercial MVK is
stabilized by acetic acid and hydroquinone, the impact of
these two additives was studied. While hydroquinone slightly
retarded the reaction, the presence of catalytic amounts of
acetic acid proved beneficial in terms of enantioselectivity.
Almost identical results were obtained with 1, 10, and
20 mol% HOAc (Table 1, entries 16–20). Reducing the
precatalyst loading to 0.2 mol% led to identical results
(Table 1, entry 18), and even with as little as 0.1 mol%
precatalyst, the reaction was still relatively efficient after a
reaction time of 20 h (Table 1, entries 19 and 20). Although
substrate 1Ae bearing the especially bulky (diisopropyl)-
methyl ester unit led to even higher selectivity under the
optimized reaction conditions (Table 1, entry 21), for practi-
Angew. Chem. Int. Ed. 2008, 47, 9284 –9288
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9285

Communications
irrespective of their electronic nature.[16] The developed
coordination proceeds selectively by MeCN substitution.
Activation of FBIP-Cl with Ag-O3S-Tipp provided a similar
monomeric species as indicated by 1H NMR analysis. For
FBIP-O3S-Tipp, NOESY/ROESY experiments revealed that
the active catalyst sites are, as expected, trans to the imidazo-
line N donor (see the Supporting Information). Treatment of
the activated catalyst with a-cyanoacetate 1Ad (2.5 equiv/Pd)
in CDCl3 effected rapid partial MeCN–substrate exchange as
revealed by 1H NMR spectroscopy (see the Supporting
Information). Based on the hard–soft acid–base (HSAB)
concept it is expected that the softer N rather than the harder
O atoms would bind to the soft Lewis acid to promote
enolization. This is further supported by the fact that no
Michael addition product was formed when 1Ad was
replaced with diethyl a-phenyl malonate.[18]
process is operationally simple and does not require the use
of inert gas; the described experiments were conducted in air.
In most cases chromatographic purification was not necessary
to obtain analytically pure products as no side products were
formed. Excess enone was removed by evaporation, and the
catalyst was removed by filtration over a short pad of silica
gel.
In the development of the described methodology, we
assumed that both nucleophile and electrophile are activated
by a cooperative bimetallic mechanism (Scheme 1). Coordi-
The initial reaction rate using the corresponding Sp-
configured ferrocenyl monoimidazoline monopalladacycle
FIP-Cl[19] activated by Ag-O3S-Tipp was only 2.1 times
slower than that with the bispalladacycle under identical
reaction conditions (0.5 mol% precatalyst, 25 mg 1Ad,
2 equiv MVK, 0.2 equiv HOAc, 0.5 mL diglyme, RT), but
the product was formed with low and inversed enantioface-
selectivity (À45% ee, preferential attack of MVK on the Re
face of 1).[20]
The initial rate of the model reaction catalyzed by FBIP-
O3S-Tipp follows a first-order dependence for the activated
catalyst, the Michael donor and the Michael acceptor (see the
Supporting Information, Plots S1–S6). However, the rate-
À
determining step is not the formation of the C C bond (i.e.
the enantioselectivity-determining step) but the decomplex-
ation of the bidentate product which is able to form a chelate
complex with the bimetallic system (chelate effect) thus
stabilizing complex III against ligand exchange. This was
evidenced by the relationship between the initial conversion
monitored by HPLC methods (sample quenching by hexane/
diglyme/iPrOH leads to the release of product 2 from III) and
the reaction time (Figure 1). Extrapolation of the straight line
to t0 = 0 h provides a positive y intercept. In other words:
Scheme 1. Proposed cooperative intramolecular bimetallic mechanism
and X-ray crystal structure of 2Cda.
nation of the nitrile group to the PdII center would facilitate
enolization, while the enone would be activated by coordi-
II
=
nation of the C C bond to the carbophilic Pd . The
configurational outcome would depend upon the face selec-
tivity of the enol approaching the Michael acceptor. To
differentiate between the enantiotopic faces, the catalyst must
À
upon addition of the reagents, C C bond formation occurs
À
thus control the conformation with regard to the C CN s
bond and also direct the enone. Control of the reactive
conformation is achieved by the use of a bulky ester moiety
and an especially large sulfonate counteranion, which should
point away from each other to minimize unfavorable steric
interactions. The direction of the enone is accomplished by
the cooperative mechanism, which is in accordance with the
absolute configuration (Si-face attack on 1) determined for
2Cda by X-ray crystal structure analysis (Scheme 1).[17]
Kinetic, spectroscopic, and enantioselectivity data pro-
vide strong evidence for a mechanism involving bimetallic
catalysis. In previous studies on the aza-Claisen rearrange-
ment, we showed that the Cl-bridged dimer FBIP-Cl forms a
monomeric catalyst species by activation with silver tosylate,
in which one MeCN ligand reversibly and diastereoselectively
adopts one of two possible coordination sites on the PdII
center, while the other coordination site remains blocked by
the sulfonate anion.[13b] The latter ligand is not replaced by an
excess of MeCN. It was therefore likely that substrate
Figure 1. Relationship between the initial conversion (determined by
HPLC) and reaction time (25 mg 1Ad, 2.0 equiv MVK, cat. Ag-O3S-
Tipp, 0.5 mL diglyme) for different loadings of the FBIP-Cl precatalyst;
~
&
*
: 1.0 mol% (y=13.3x + 2.06), : 0.8 mol% (y=10.4x + 1.61),
:
0.6 mol% (y=7.78x + 1.42), +: 0.4 mol% (y=5.80x + 0.92), –:
0.25 mol% (3.10x + 0.57); R2 ꢀ0.995 in all cases.
9286 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 9284 –9288

Angewandte
Chemie
almost instantaneously. The amount of product correlates
within the experimental error to twice the amount of
precatalyst loading since the dimeric precatalyst forms two
active monomeric catalyst species. In contrast, for the
corresponding monopalladacycle catalyst FIP-Cl, for which
we also found a first-order dependence for the activated
catalyst excluding a bimetallic intermolecular double activa-
tion mechanism, the plots start at the zero point (see the
with success for soft Lewis acid complexes as a result of the
difficulties encountered by remote enantiofacial discrimina-
tion of the cyano-substituted enol. To overcome the previous
limitations, a catalytic system has been designed capable of
À
a) controlling the conformation with regard to the C CN s
bond of the Michael donor and b) directing the enone to
differentiate the enantiotopic faces by a bimetallic coopera-
tive mode of action. The proposed mechanism is supported by
enantioselectivity data and by spectroscopic and kinetic
investigations.[23] Remarkably the same precatalyst, which is
readily prepared in diastereomerically pure form in four steps
from ferrocene, has been found to be exceptionally efficient
for aza-Claisen rearrangements and direct Michael additions
despite the fundamental differences with regard to the
involved transition-state geometries. The reaction has various
operational advantageous as it proceeds at room temperature
with low catalyst loadings and high concentrations, does not
require inert gas, and typically provides excellent yields.
Current efforts are directed towards an acceleration of the
product decomplexation to further accelerate the catalytic
turnover.
À
Supporting Information, Plot S9) showing that the C C bond
formation does not proceed instantaneously in this case. Re-
face attack of the enone to complex I (monometallic
mechanism) is therefore significantly slower than the intra-
À
molecular C C bond formation. This explains why the
monometallic activation pathway does not destroy the
enantioselectivity for FBIP-Cl despite the similar overall
activity of monometallic FIP-Cl. The fact that there is no
large difference in the overall rate is consequently just
because of the slow decomplexation in III, whereas for the
monopalladacycle there is no chelate effect which explains in
that case decomplexation is considerably faster.
The decomplexation of 2 is a reversible step for the
bispalladacycle since a slight product inhibition was deter-
mined (see the Supporting Information, Plots S7–S8). Larger
R2 residues retard the decomplexation and turnover, since
initial attack of substrate 1 on III is likely to occur by an
associative mechanism at the Pd center, which coordinates to
the more labile ketone donor.
This mechanism also explains why phenyl vinyl ketones
(PVKs) reacted slower in this study than the less bulky MVK
although they are intrinsically more electrophilic.[5] It is also
in line with the reduced ee values obtained for PVKs (Table 2,
entries 5 and 6), as the higher electrophilicity of the acceptor
leads to increased product formation by the background
reaction following the monometallic pathway (attack from
the sterically better accessible Re face). Acceleration of the
monometallic pathway is also observed with increasing
amounts of MVK (see the Supporting Information,
Plot S10). On the other hand, electron-withdrawing substitu-
ents on the a-aryl group of 1 decrease the inherent
nucleophilicity of the generated enols and consequently
Received: July 21, 2008
Revised: August 22, 2008
Published online: October 27, 2008
À
Keywords: bimetallic catalysis · C C coupling · imidazolines ·
Michael addition · palladacycles
.
[1] Recent reviews about conjugate additions: a) M. Kanai, M.
Shibasaki in Catalytic Asymmetric Synthesis, 2nd ed. (Ed.: I.
Ojima), Wiley, New York, 2000, pp. 569 – 592; b) N. Krause, A.
Hoffmann-Rꢁder, Synthesis 2001, 171; c) J. Christoffers, A.
Baro, Angew. Chem. 2003, 115, 1726; Angew. Chem. Int. Ed.
2003, 42, 1688; d) J. Christoffers, G. Koripelly, A. Rosiak, M.
Rꢁssle, Synthesis 2007, 1279.
[2] a) Quaternary Stereocenters: Challenges and Solutions for
Organic Synthesis (Eds.: J. Christoffers, A. Baro), Wiley-VCH,
Weinheim, 2005; b) J. Christoffers, A. Baro, Adv. Synth. Catal.
2005, 347, 1473; c) E. J. Corey, A. Guzman-Perez, Angew. Chem.
1998, 110, 402; Angew. Chem. Int. Ed. 1998, 37, 388.
minimize
a monometallic background reaction, which
[3] R. Peters, D. F. Fischer, S. Jautze, F. M. Koch, T. Kull, P. S. Tiseni,
Z.-q. Xin, M. Zajac, Chimia 2008, 62, 497.
explains the higher enantioselectivity.
The bimetallic mechanism involving coordination of the
[4] a) H. Wynberg, R. Helder, Tetrahedron Lett. 1975, 16, 4057;
b) T.-Y. Liu, R. Li, Q. Chai, J. Long, B.-J. Li, Y. Wu, L.-S. Ding,
Y.-C. Chen, Chem. Eur. J. 2007, 13, 319; acrolein as acceptor:
c) F. Wu, R. Hong, J. Khan, X. Liu, L. Deng, Angew. Chem. 2006,
118, 4407; Angew. Chem. Int. Ed. 2006, 45, 4301; d) M. Bell, K.
Frisch, K. A. Jørgensen, J. Org. Chem. 2006, 71, 5407; nitro
olefins as acceptors: e) H. Li, Y. Wang, L. Tang, F. Wu, X. Liu, C.
Guo, B. M. Foxman, L. Deng, Angew. Chem. 2005, 117, 107;
Angew. Chem. Int. Ed. 2005, 44, 105; vinyl sulfones as acceptors:
f) H. Li, J. Song, X. Liu, L. Deng, J. Am. Chem. Soc. 2005, 127,
8948; g) T.-Y. Liu, J. Long, B.-J. Li, L. Jiang, R. Li, Y. Wu, L.-S.
Ding, Y.-C. Chen, Org. Biomol. Chem. 2006, 4, 2097; a-Cl-
acrylonitrile as acceptor: h) Y. Wang, X. Liu, L. Deng, J. Am.
Chem. Soc. 2006, 128, 3928; i) Y. Wang, X. Liu, L. Deng, J. Am.
Chem. Soc. 2007, 129, 768; phase-transfer catalysis with an
enone/ynoate: j) X. Wang, M. Kitamura, K. Maruoka, J. Am.
Chem. Soc. 2007, 129, 1038.
=
acceptorꢀs C C bond also points to the necessity for catalytic
amounts of HOAc to obtain high enantioselectivity although
it has no influence on the reaction rate. The conjugate
addition should initially form complex II in which PdII is
connected to the a-C atom. To form the targeted product, a
proton transfer must take place from the protonated ester
group to the nucleophilic enolate C atom. This event might
compete with a b-hydride elimination to form a Pd0 species if
the proton transfer is not sufficiently fast.[21] Addition of
HOAc could thus speed up the required proton transfer, while
the generated acetate anion would liberate the neutral ester
group.
In conclusion, we have developed a soft Lewis acid
catalyst that is capable of promoting a highly enantioselective
Michael addition of a-aryl-substituted a-cyanoacetates to
vinyl ketones.[22] This challenge has previously not been met
[5] a) M. Sawamura, H. Hamashima, Y. Ito, J. Am. Chem. Soc. 1992,
114, 8295; b) M. Sawamura, H. Hamashima, Y. Ito, Tetrahedron
Angew. Chem. Int. Ed. 2008, 47, 9284 –9288
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9287

Communications
1994, 50, 4439; alternative Rh catalysts: c) K. Inagaki, K. Nozaki,
[15] For recent applications of imidazolines in asymmetric catalysis,
see for example, a) F. Menges, M. Neuburger, A. Pfaltz, Org.
Lett. 2002, 4, 4713; b) C. A. Busacca, D. Grossbach, R. C. So,
E. M. OꢀBrien, E. M. Spinelli, Org. Lett. 2003, 5, 595; c) S. Bhor,
G. Anilkumar, M. K. Tse, M. Klawonn, C. Dobler, B. Bitterlich,
A. Grotevendt, M. Beller, Org. Lett. 2005, 7, 3393; d) K. Ma, J.
You, Chem. Eur. J. 2007, 13, 1863; e) T. Arai, T. Mizukami, A.
Yanagisawa, Org. Lett. 2007, 9, 1145; f) S. Enthaler, B. Hage-
mann, S. Bhor, G. Anilkumar, M. K. Tse, B. Bitterlich, K. Junge,
G. Erre, M. Beller, Adv. Synth. Catal. 2007, 349, 853; g) T. Arai,
N. Yokoyama, A. Yanagisawa, Chem. Eur. J. 2008, 14, 2052;
h) Z.-q. Xin, D. F. Fischer, R. Peters, Synlett 2008, 1495; i) S.
Nakamura, K. Hyodo, Y. Nakamura, N. Shibata, T. Toru, Adv.
Synth. Catal. 2008, 350, 1443; j) T. Arai, N. Yokoyama, Angew.
Chem. 2008, 120, 5067; Angew. Chem. Int. Ed. 2008, 47, 4989.
[16] ortho-Substituted phenyl groups were not well tolerated and
moderate enantioselectivity was obtained, e.g.: o-FC6H4:
57% ee, 1-naphthyl: 50% ee.
H. Takaya, Synlett 1997, 119; d) Y. Motoyama, Y. Koga, K.
Kobayashi, K. Aoki, H. Nishiyama, Chem. Eur. J. 2002, 8, 2968;
for previous non-enantioselective studies with Ru and Rh
catalysts, see: e) T. Naota, H. Taki, M. Mizuno, S.-I. Murahashi,
J. Am. Chem. Soc. 1989, 111, 5954; f) S. Paganelli, A. Schionato,
C. Botteghi, Tetrahedron Lett. 1991, 32, 2807; Ir-catalyst: g) D.
Carmona, J. Ferrer, M. Lorenzo, F. J. Lahoz, I. T. Dobrinovitch,
L. A. Oro, Eur. J. Inorg. Chem. 2005, 1657.
[6] a) M. A. Stark, C. J. Richards, Tetrahedron Lett. 1997, 38, 5881;
b) M. A. Stark, G. Jones, C. J. Richards, Organometallics 2000,
19, 1282.
[7] a) K. Takenaka, Y. Uozumi, Org. Lett. 2004, 6, 1833; b) K.
Takenaka, M. Minakawa, Y. Uozumi, J. Am. Chem. Soc. 2005,
127, 12273; alternative PdII catalysts: c) C. Mazet, L. H. Gade,
Chem. Eur. J. 2003, 9, 1759; PtII catalyst: d) A. J. Blacker, M. L.
Clarke, M. S. Loft, M. F. Mahon, J. M. J. Williams, Organo-
metallics 1999, 18, 2867.
[8] M. S. Taylor, D. N. Zalatan, A. M. Lerchner, E. N. Jacobsen, J.
Am. Chem. Soc. 2005, 127, 1313.
[17] CCDC 695750 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif. HPLC analysis of the crystal used on
a chiral stationary phase subsequently confirmed that the major
enantiomer of 2Cda was investigated.
[9] M. S. Taylor, E. N. Jacobsen, J. Am. Chem. Soc. 2003, 125, 11204.
[10] Recent reviews about palladacycles: a) W. A. Herrmann,
V. P. W. Bohm, C. P. Reisinger, J. Organomet. Chem. 1999, 576,
23; b) M. Albrecht, G. van Koten, Angew. Chem. 2001, 113,
3866; Angew. Chem. Int. Ed. 2001, 40, 3750; c) J. Dupont, M.
Pfeffer, J. Spencer, Eur. J. Inorg. Chem. 2001, 1917; d) R. B.
Bedford, Chem. Commun. 2003, 1787; e) M. E. van der Boom,
D. Milstein, Chem. Rev. 2003, 103, 1759; f) J. T. Singleton,
Tetrahedron 2003, 59, 1837; g) F. Bellina, A. Carpita, R. Rossi,
Synthesis 2004, 2419; h) R. B. Bedford, C. S. J. Cazin, D. Holder,
Coord. Chem. Rev. 2004, 248, 2283; i) I. Omae, Coord. Chem.
Rev. 2004, 248, 995; j) I. P. Beletskaya, A. V. Cheprakov, J.
Organomet. Chem. 2004, 689, 4055; k) J. Dupont, C. S. Consorti,
J. Spencer, Chem. Rev. 2005, 105, 2527. For pioneering work
using planar-chiral palladacycles for aza-Claisen rearrange-
ments, see: l) T. K. Hollis, L. E. Overman, J. Organomet.
Chem. 1999, 576, 290; m) C. E. Anderson, L. E. Overman, J.
Am. Chem. Soc. 2003, 125, 12412; n) L. E. Overman, C. E.
Owen, M. M. Pavan, C. J. Richards, Org. Lett. 2003, 5, 1809.
[11] a) N. Strꢂter, W. N. Lipscomb, T. Klabunde, B. Krebs, Angew.
Chem. 1996, 108, 2158; Angew. Chem. Int. Ed. Engl. 1996, 35,
2024; b) D. E. Wilcox, Chem. Rev. 1996, 96, 2435; c) H.
Steinhagen, G. Helmchen, Angew. Chem. 1996, 108, 2489;
Angew. Chem. Int. Ed. Engl. 1996, 35, 2339.
[12] Reviews: a) E. K. van den Beuken, B. L. Feringa, Tetrahedron
1998, 54, 12985; b) M. Shibasaki, S. Matsunaga, Chem. Soc. Rev.
2006, 35, 269; for selected recent examples, see: c) C. Mazet,
E. N. Jacobsen, Angew. Chem. 2008, 120, 1786; Angew. Chem.
Int. Ed. 2008, 47, 1762; d) S. Handa, K. Nagawa, Y. Sohtome, S.
Matsunaga, M. Shibasaki, Angew. Chem. 2008, 120, 3274;
Angew. Chem. Int. Ed. 2008, 47, 3230; e) S. Takizawa, T.
Katayama, H. Sasai, Chem. Commun. 2008, 35, 4113.
[13] a) S. Jautze, P. Seiler, R. Peters, Angew. Chem. 2007, 119, 1282;
Angew. Chem. Int. Ed. 2007, 46, 1260; b) S. Jautze, P. Seiler, R.
Peters, Chem. Eur. J. 2008, 14, 1430.
[14] For the first preparation of chiral ferrocenyl imidazolines, see:
a) R. Peters, D. F. Fischer, Org. Lett. 2005, 7, 4137; b) R. Peters,
Z.-q. Xin, D. F. Fischer, W. B. Schweizer, Organometallics 2006,
25, 2917.
1
[18] Complexation of MVK was not detected by H NMR spectros-
copy, indicating that such a complex is arguably too short-lived
on the NMR timescale.
[19] a) M. E. Weiss, D. F. Fischer, Z.-q. Xin, S. Jautze, W. B.
Schweizer, R. Peters, Angew. Chem. 2006, 118, 5823; Angew.
Chem. Int. Ed. 2006, 45, 5694; Angew. Chem. 2006, 118, 5823; for
the catalytic asymmetric formation of quaternary stereocenters
by aza-Claisen rearrangements, see: b) D. F. Fischer, Z.-q. Xin,
R. Peters, Angew. Chem. 2007, 119, 7848; Angew. Chem. Int. Ed.
2007, 46, 7704.
[20] The corresponding ferrocenyl imidazoline monopalladacycle
complex with a C5Ph5 spectator ligand resulted in ca. À50% ee
under identical reaction conditions.
[21] A b-hydride elimination would result in the formation of Pd0
species which are usually not stable as palladacycles and form
“naked” Pd0 (see Ref. [10]). The monopalladacycle formed by
decomplexation of one Pd0 would lead to a competing pathway
providing preferentially the other enantiomer.
[22] The optimized conditions for a-aryl-substituted a-cyano acetates
are not useful for a-alkyl-substituted substrates and result in
poor enantioselectivity. Current studies are therefore directed
toward the extension to this substrate class.
[23] One referee proposed a reversible Michael addition step as
mechanistic alternative to explain the enantioselectivity. How-
ever, our data can almost rule out such a scenario. If the step I to
II in Scheme 1 is reversible, lower ee values should be obtained
with an increasing amount of acetic acid, since in that case the
lifetime of II would be reduced. Before the equilibrium would be
reached (in that scenario necessary for high ee), the intermediate
would be trapped. For that reason we can also rule out that
protonation would be the enantioselectivity-determining step.
Cross experiments in which 2Ada and 2Cdb were treated with
2 mol% bispalladacycle catalyst at room temperature revealed
overall irreversibility as the formation of 2Adb and 2Cda could
not be detected by HPLC or NMR spectroscopy.
9288 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 9284 –9288