Copper(I) thiolate catalysts in asymmetric conjugate addition reactions

Article abstract of DOI:10.1021/ol049457u

Equation presented. Full conversion and enantioselectivities up to 83% have been obtained in the conjugate addition reactions of diethyl zinc to Michael acceptors catalyzed by well-defined (chiral) copper(I) aminoarenethiolates. Interesting differences between organozinc or Grignard reagents have been found: for cyclic enones R₂Zn reagents afford better results, whereas earlier work showed that RMgX reagents react more selectively with acyclic enones.

Full text of DOI:10.1021/ol049457u

ORGANIC
LETTERS
2004
Vol. 6, No. 12
1959-1962
Copper(I) Thiolate Catalysts in
Asymmetric Conjugate Addition
Reactions
Anne M. Arink,† Thijs W. Braam,† Roy Keeris,† Johann T. B. H. Jastrzebski,†
,‡
Cyril Benhaim,‡ Ste´phane Rosset,‡ Alexandre Alexakis,* and
,†
Gerard van Koten*
Debye Institute, Department of Metal-Mediated Synthesis, Utrecht UniVersity,
Padualaan 8, 3584 CH Utrecht, The Netherlands, and Department of Organic
Chemistry, UniVersity of GeneVa, 30, Quai Ernest Ansermet,
1211 GeneVa 4, Switzerland
g.Vankoten@chem.uu.nl; alexandre.alexakis@chiorg.unige.ch
Received March 23, 2004
ABSTRACT
Full conversion and enantioselectivities up to 83% have been obtained in the conjugate addition reactions of diethyl zinc to Michael acceptors
catalyzed by well-defined (chiral) copper(I) aminoarenethiolates. Interesting differences between organozinc or Grignard reagents have been
found: for cyclic enones R Zn reagents afford better results, whereas earlier work showed that RMgX reagents react more selectively with
2
acyclic enones.
Enantioselective copper-catalyzed 1,4-additions have been
extensively studied.1 Most of these studies involved the use
of chiral ligands in the presence of copper(II) salts2 or copper-
(I) complexes formed in situ, leaving salts (e.g., LiI) present
during catalysis.3 So far, only a few reports described the
use of salt-free, well-defined, and fully characterized chiral
copper(I) complexes as catalysts.4,5 Salt-free prepared copper-
(I) aminoarenethiolates4 1a and enantiopure 1b (Figure 1)
have been tested as catalysts in several types of C-C
coupling reactions, e.g., 1,4-additions,6 1,6-additions,7 cross-
couplings,8 and allylic substutions.9 In these initial studies
we found that, in the 1,4-addition of Grignard reagents to
R,â-unsaturated enones, enantioselectivities up to 76% were
(3) (a) Zhou, Q. L.; Pfaltz, A. Tetrahedron 1994, 50, 4467. (b) Spescha,
M.; Rihs, G. HelV. Chim. Acta 1993, 76, 1219. (c) Seebach, D.; Jaeschke,
G.; Pichota, A.; Audergon, L. HelV. Chim. Acta 1997, 80, 2515.
(4) Knotter, D. M.; Grove, D. M.; Smeets, W. J. J.; Spek, A. L.; van
Koten, G. J. Am. Chem. Soc. 1992, 114, 3400.
(5) Pichota, A.; Pregosin, P. S.; Valentini, M.; Worle, M.; Seebach, D.
Angew. Chem., Int. Ed. 2000, 39, 153.
(6) van Klaveren, M.; Lambert, F.; Eijkelkamp, J. F. M.; Grove, D. M.;
van Koten, G. Tetrahedron Lett. 1994, 35, 6135.
(7) Haubrich, A.; van Klaveren, M.; van Koten, G.; Handke, G.; Krause,
N. J. Org. Chem. 1993, 58, 5849.
† Utrecht University.
‡ University of Geneva.
(1) (a) Rossiter, B. E.; Swingle, N. M. Chem. ReV. 1992, 92, 771. (b)
Krause, N.; Hoffman-Ro¨der, A. Synthesis 2001, 171. (c) Alexakis, A.;
Benhaim, C. Eur. J. Org. Chem. 2002, 3221.
(2) (a) Alexakis, A.; Benhaim, C.; Rosset, S.; Humam, M. J. Am. Chem.
Soc. 2002, 124, 5262. (b) Duursma, A.; Minnaard, A. J.; Feringa, B. L.
Tetrahedron 2002, 58, 5773. (c) Degrado, S. J.; Mizutani, H.; Hoveyda, A.
H. J. Am. Chem. Soc. 2002, 124, 13362. (d) Hu, X.; Chen, H.; Zhang, X,
Angew. Chem., Int. Ed. 1999, 38, 3518. (e) Yan, M.; Yang, L.-W.; Wong,
K.-Y.; Chan, A. S. C. Chem. Commun. 1999, 11. (f) Escher, I.; Pfaltz, A.
Tetrahedron 2000, 56, 2879.
(8) van Klaveren, M., Ph.D. Dissertation, Utrecht University, Utrecht,
1996.
10.1021/ol049457u CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/13/2004

A new set of modified complexes was tested with ligand
variations at the benzylic position (1c) and the amine
functionality; dimethylamine was replaced for either a
pyrrolidinyl (3) or a piperidyl ring (4). Furthermore, trim-
ethylsilyl groups were introduced meta to the thiocopper
moiety (2a, 2b, 6, and 8). Another modification was the use
of a naphthalene instead of a benzene backbone (5-8).11
First, chiral catalysts 1c, 2b, 3, and 4 were tested in the
Michael addition of MeMgI to trans-4-phenylbuten-2-one
(S1) (Table 1).6 Complex 1b proved to be the best catalyst
under these conditions. Catalysts 3 and 4 gave slightly lower
regioselectivity and showed partial 1,2-product formation.
For the diethylzinc addition to Michael acceptors, fre-
quently applied conditions, i.e., 2 mol % Cu complex for 4
h at -20 °C, were used.1,2 For 2-cyclohexen-1-one (S2) full
conversion and an enantiomeric excess of 83% were achieved
with catalyst 1b (Table 2).
Figure 1. Modified copper(I) aminoarenethiolates 1-8.
achieved by using the enantiopure copper(I) aminoarene-
thiolate 1b in catalytic amounts (Table 1).4,6 These results
Table 2. Addition of Et2Zn to 2-Cyclohexen-1-one (S2) Using
Catalyst (R)-1b
Table 1. Addition of MeMgI to trans-4-Phenylbuten-2-one
(S1) Using Chiral Copper Aminoarenethiolates
conversion
(%)
ee %,
regioselectivityb
entry
solvent
additive
LiI
config
entry
catalysta
S1-a :S1-b
ee %, config
1
2
3
4
5
Et2O
Et2O
Et2Ob
toluene
THF
>99
85
3
90
23
83, R
15, R
13, R
70, R
10, R
1
2
3
4
5
1b
1c
2b
3
>99:<1
>99:<1
>99:<1
94:6
76, S
63, S
72, S
70, S
50, S
4
95:5
a Selective to 1,4-product. b Me2Zn instead of Et2Zn
a All catalysts have (R)-configuration. b Conversion was quantitative.
In contrast to the formation of the (S)-enantiomer of the
1,4-product when reacting MeMgI with S1 catalyzed by (R)-
1b, addition of MeMgI or Et2Zn to enone S2 resulted in the
preferential formation of the (R)-enantiomer. An additive
such as LiI (1 equiv) resulted in slower reaction rates and a
drastic drop in ee (entry 2). The addition of Me2Zn to S2
instead of Et2Zn was extremely slow.12 Variations in solvent
indicated that diethyl ether was the best solvent under these
conditions; the reaction rate in THF was much slower, while
the ee was lower too. However, for entries 2-5 the reaction
is still selective to the 1,4-product.
The influence of the ee of the catalyst on the ee of the
1,4-product was studied for the 1b-catalyzed addition of Et2-
Zn to cyclohexenone (S2) in diethyl ether (Figure 2).
A positive correlation ((+)-NLE) was observed between
the ee of catalyst 1b and the ee of the 1,4-product (Figure
2). Both positive and negative NLE () nonlinear effect)
relationships have been observed for Cu-catalyzed additions
were obtained for acyclic enones, whereas for cyclic enones
poor ee’s and low selectivities were found. For example, for
the addition of n-BuMgI to cyclohexenone, an ee of 23%
was obtained with a selectivity (1,4- vs 1,2-addition) of 64%.8
Interestingly, when RLi reagents were tested instead of
RMgX, no ee was found for both acyclic and cyclic enones.10
As recently high selectivities in Michael additions with R2-
Zn reagents have been achieved,2 we decided to test our
chiral catalysts also in zinc-mediated 1,4-additions.
In the present study we report on the addition of dieth-
ylzinc to a variety of Michael acceptors in the presence of
catalytic amounts of well-defined, salt-free copper(I) ami-
noarenethiolates 1-8 (Figure 1).
(9) (a) Meuzelaar, G. J.; Karlstro¨m, A. S. E.; van Klaveren, M.; Persson,
E. S. M.; del Villar, A.; van Koten, G.; Ba¨ckvall, J.-E. Tetrahedron 2000,
56, 2895. (b) van Klaveren, M.; Persson, E. S. M.; del Villar, A.; Grove,
D. M.; Ba¨ckvall, J.-E.; van Koten, G. Tetrahedron Lett. 1995, 36, 3059.
(c) van Klaveren, M.; Persson, E. S. M.; Grove, D. M.; Ba¨ckvall, J.-E.;
van Koten, G. Tetrahedron Lett. 1994, 35, 5931, (d) Karlstro¨m, A. S. E.;
Huerta, F. F..; Meuzelaar, G. J.; Ba¨ckvall, J.-E. Synlett 2001, 923.
(10) Lambert, F.; Knotter, D. M.; Janssen, D. M.; van Klaveren, M.;
Boersma, J.; van Koten, G. Tetrahedron: Asymmetry 1991, 2, 1097.
(11) Arink, A. M.; Braam, T. W.; Keeris, R.; Jastrzebski, J. T. B. H.;
Kooijman, H.; Ellis, D.; Spek, A. L.; van Koten, G. Unpublished results.
(12) Compare with: Kitamura, M.; Miki, T.; Nakano, K.; Noyori, R.
Bull. Chem. Soc. Jpn. 2000, 73, 999.
1960
Org. Lett., Vol. 6, No. 12, 2004

Figure 3. Substrates S1-S4 used in this work.
always very high. For S2 the selectivity was 100%. For
substrates S1, S3, and S4 the selectivity is larger than 95%.
S1 and S3 afford 1-5% of side products that were identified
as the Michael adduct of the resulting enolate to the starting
enone. For S4 the formation of two side products (1-5%)
is observed. One of the side products could be identified as
arising from an ipso-type substitution (vinylic substitution).15
Since all reactions were quenched after the same period
(4 h) the reaction rates can be compared. The reaction rates
for these conditions increase for the substrates in the order
S1 < S3 < S4 ≈ S2. The best results are obtained for
catalysts 1a,b and 2a,b. The worse performance of catalysts
3, 4, and 5 can be explained by low solubility of these
complexes and the intermediates in diethyl ether. Complexes
1b, 2a,b, and 6-8 are very soluble in diethyl ether, whereas
1a is moderately soluble.
Differences in reaction rate and enantioselectivity can also
be explained by the differences in the CH(R)NZ2 substituent.
For instance, the conversion and enantioselectivity of all
substrates with 1b and 2b are higher than for 1c, where R
) Et instead of Me. Catalysts 3 and 4, where Z2 ) (CH2)4
and (CH2)5, respectively, have more steric bulk around the
catalytic center than 1b. Especially for the acyclic substrates
S1 and S3 this has a negative effect on both the reaction
rate and enantioselectivity.
Figure 2. Nonlinear relation between the ee of catalyst 1b and
the ee of the 1,4-product in the Et2Zn addition to cyclohexenone.
of organozinc or Grignard reagents to enones.3b,4,13,14 Espe-
cially the 1b-catalyzed addition of MeMgI to S1 was very
complex.13a
Subsequently, complexes 2-8 were tested using the same
protocol as applied for 1b in diethyl ether (cf. Table 3).
Table 3. Reaction of Et2Zn with Substrates S1-S4, with 2
mol % of Cu(SAr) Catalyst (1-8)
catalyst )
1,4-addition product
(yield (%),a ee (%),b configuration
en-
(R)-enan-
try achiral
tiomer
S1a
S2a
S3a
S4a
1
2
3
4
5
6
7
8
1b
1c
2b
3
48, 59, S >99, 83, R 86, 35, (+) 99, 22, S
39, 3, S >99, 40, R 71, 16, (-) 96, 10, S
67, 62, S >99, 81, R 87, 37, (+) 97, 15, S
10, 28, S 52, 6, R
17, 15, S 98, 63, R
48, 8, (-) 93, 5, S
35, 12, (-) 79, 0, -
4
1a
2a
5
6
7
60
84
9
99
>99
72
89
96
50
84
93
88
95
96
90
96
96
98
9
10
11
44
41
39
99
93
93
For substrates S1, S2, and S4 the respective enantiomers
formed in excess have the same configuration. However, the
configuration of the product resulting from S3 is dependent
on the catalyst.
8
a The yield of 1,4-product was determined by GC-MS. b The ee was
determined by chiral GC on a LipodexE (25 m × 0.25 mm) capillary
column.
Whereas 1b and 2b afford the (+)-enantiomer of substrate
S3 in excess, 1c, 3, and 4, each having more bulky amino
substituents than 1b and 2b, afford the (-)-enantiomer of
substrate S3. Of the achiral catalysts, 2a gives the best yields
for all substrates.
Besides acyclic enone S1 and cyclic enone S2, trans-3-
nonen-2-one (S3), and trans-â-nitrostyrene (S4) were also
tested (Figure 3).
The observation of a positive NLE indicates that aggregate
formation influences the rate-determining step in the 1,4-
addition reaction. Several preliminary experiments were
carried out to get information about the nature of possible
intermediates. Stoichiometric mixing of 1b with diethylzinc
in diethyl ether resulted in the formation of a soluble
complex, which at room temperature decomposed, leaving
a Cu0 mirror and a black precipitate. Analysis of the
supernatant revealed that [Zn(SAr)Et]2 had been formed
The results of the both chiral and achiral complexes in
catalysis are listed in Table 3 (entries 1-11). For 2-cyclo-
hexen-1-one (S2) and trans-â-nitrostyrene (S4) (almost) full
conversion was obtained. In nearly all cases, the catalysts
showed a slower reaction rate for trans-4-phenylbuten-2-
one (S1) and trans-3-nonen-2-one (S3). Irrespective of this
slow reaction rate, the selectivity toward the 1,4-product is
(13) (a) van Koten, G. Pure Appl. Chem. 1994, 66, 1455. (b) Arnold, L.
A.; Imbos, R.; Mandoli, A.; de Vries, A. H. M.; Naasz, R.; Feringa, B. L.
Tetrahedron 2000, 56, 2865.
(15) (a) Namboothiri, I. N. N.; Hassner, A. J. Organomet. Chem.1996,
518, 69. (b) Rimkus, A., Sewald, N. Org. Lett. 2002, 4, 3289.
(14) Girard, C.; Kagan, H. B. Angew. Chem., Int. Ed. 1998, 37, 2922.
Org. Lett., Vol. 6, No. 12, 2004
1961

quantitatively. These observations can be explained by a
transmetalation reaction, which results in the formation of
known [Zn(SAr)Et]2 and EtCu (Figure 4). The latter is
such cannot catalyze this reaction and indicates that a mixed
copper-zinc thiolate complex is the active intermediate.
Interestingly, the use of a combination of (R)-[Zn(SAr)Et]2
(2 mol %) with (R)-1b (2 mol %) led to full conversion to
the 1,4-product, but with a somewhat lower ee (65% ee (R)
vs 83% ee (R)). The same experiment, but now using (R)-
[Zn(SAr)Et]2 and (S)-1b (both 2 mol %) as a mixed catalyst,
again gave full conversion to the 1,4-product, now with 15%
ee (S). The addition of (R)-[Zn(SAr)Et]2 to the 1b-catalyzed
reaction seemed to slow the reaction and made it less
selective.
It was shown that there are significant differences between
the reactivity of Grignard and diorganozinc reagents to
enones catalyzed by copper(I) aminoarenethiolates. For a
cyclic enone, such as S2, the use of diethylzinc in the
presence of copper(I) aminoarenethiolates is preferred, in
contrast to acyclic enones (S1) where Grignard reagents give
the best result. For the Grignard addition to acyclic enones
a multishaped NLE has been reported, whereas the present
study shows a positive NLE for the diethylzinc addition to
cyclic enones. The latter observation could be explained by
a mechanism involving dimeric resting states of the type [Zn-
(SAr)Et]2.
Figure 4. Stoichiometric mixing of 1b with Et2Zn.
thermally unstable and decomposes to yield metallic copper.
The NLE behavior of the addition of diethylzinc to
benzaldehyde in the presence of zinc bis(aminoarenethiolate)
is comparable to that of the 1b-catalyzed addition of
diethylzinc to S2.16 A mechanism involving dimeric resting
states of the type [Zn(SAr)Et]2 could explain the positive
NLE in the latter reaction.
Complex (R)-[Zn(SAr)Et]2 was prepared16 and tested as
catalyst (2 mol %) in the addition reaction of Et2Zn to S2.
In the standard reaction time (4 h) used in this study, a
conversion of only 8% was achieved to a racemic 1,4-
addition product. This result shows that the zinc thiolate as
Acknowledgment. Financial support from NWO/CW and
COST D12 is gratefully acknowledged.
Supporting Information Available: Experimental details
and spectral data. This material is available free of charge
via the Internet at http://pubs.acs.org.
(16) Rijnberg, E.; Hovestad, N. J.; Kleij, A. W.; Jastrzebski, J. T. B. H.;
Boersma, J.; Janssen, M. D.; Spek, A. L.; van Koten, G. Organometallics
1997, 16, 2847.
OL049457U
1962
Org. Lett., Vol. 6, No. 12, 2004