Epimerization and kinetic resolution in copper-catalyzed enantioselective 1,4-additions of organozinc reagents to 6-substituted cyclohex-2-enones

Article abstract of DOI:10.1016/S0040-4039(02)01898-1

Enantioselective 1,4-addition reactions of diethyl-, dimethyl-, and di-n-butylzinc to 6-methylcyclohex-2-enone (1) and 6-t-butylcyclohex-2-enone (4), catalyzed by Cu(OTf)₂ and phosphoramidites L1-L4 were examined. The additions to enone 1 proce

Full text of DOI:10.1016/S0040-4039(02)01898-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 7887–7890
Epimerization and kinetic resolution in copper-catalyzed
enantioselective 1,4-additions of organozinc reagents to
6-substituted cyclohex-2-enones
Laura Mediavilla Urbaneja,a Alexandre Alexakisb and Norbert Krausea,*
aDortmund University, Organic Chemistry II, D-44221 Dortmund, Germany
bUniversite´ de Gene`ve, De´partement de Chimie Organique, 30, Quai Ernest Ansemet, CH-1211 Gene`ve 4, Switzerland
Received 28 August 2002; accepted 5 September 2002
Abstract—Enantioselective 1,4-addition reactions of diethyl-, dimethyl-, and di-n-butylzinc to 6-methylcyclohex-2-enone (1) and
6-t-butylcyclohex-2-enone (4), catalyzed by Cu(OTf)2 and phosphoramidites L1–L4 were examined. The additions to enone 1
proceeded with high enantioselectivity; by acid- or base-catalyzed epimerization, adduct (S,S)-2 can be obtained from racemic 1
in diastereo- and enantiomerically pure form. In contrast, Michael additions to substrate 4 were rather slow and could be used
for the kinetic resolution of the enone. © 2002 Elsevier Science Ltd. All rights reserved.
Copper-mediated and -catalyzed transformations
belong to the most important methods for the regio-
and stereoselective formation of CꢀC bonds.1 Among
these, enantioselective 1,4-additions of organozinc
reagents to simple enones, catalyzed by copper(II) tri-
flate and a chiral phosphoramidite, have received par-
ticular attention.2 Recently, these conditions have also
been used for the kinetic resolution of certain 4- and
5-substituted enones.3 Here, we present initial results of
a study devoted to the corresponding reaction of 6-sub-
stituted cyclohex-2-enones using the BINOL- and
biphenyl-derived phosphoramidites L1–L4 (Fig. 1).4–6
In contrast to the previous examples,3 these transforma-
tions cannot only be performed as a classical kinetic
resolution, but also with epimerization at C-6, which
might give access to a single stereoisomeric product
from the racemic enone.
chloric acid) gave mainly (S,S)-2 (corresponding to
82% ee). Clearly, the enantioselective addition is fol-
lowed by an acid-catalyzed epimerization of the cis to
the thermodynamically more stable trans isomer.9 This
isomerization can be avoided by using the less acidic
acetic acid for the protonation (entry 2), giving enan-
tiomeric excesses of 85% for the trans and >99% for the
cis product.10 As observed also in the corresponding
addition reactions to prochiral enones,4 a slightly
We first examined addition reactions to 6-methylcyclo-
hex-2-enone (1) which is readily available both in
racemic7 and enantiomerically pure8 form (Table 1).
The first reaction, performed under standard conditions
for enantioselective 1,4-addition reactions to enones,
catalyzed by Cu(OTf)2 and phosphoramidite (Sa,R,R)-
L14 (CH2Cl2, −30°C, 3 h, workup with dilute hydro-
Keywords: copper catalysis; enantioselectivity; epimerization; kinetic
resolution; Michael addition; phosphoramidites.
* Corresponding author. Tel.: +49-231-7553882; fax: +49-231-
7553884; e-mail: norbert.krause@uni-dortmund.de
Figure 1. Chiral phosporamidite ligands used in this work.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01898-1

7888
L. Media6illa Urbaneja et al. / Tetrahedron Letters 43 (2002) 7887–7890
Table 1. Copper-catalyzed enantioselective 1,4-addition of diethylzinc to enone rac-1a
Entry
Ligand
Solvent
Protonation
S,S (%)
2S,5R (%)
2R,5S (%)
RR (%)
1
2
3
4
5
6
7
(Sa,R,R)-L1
(Sa,R,R)-L1
(Sa,R,R)-L1
(Sa,S,S)-L1
(R,R)-L2
CH2Cl2
CH2Cl2
Toluene
Toluene
Toluene
Toluene
Toluene
HCl
89
51
52
51
52
48
53.5
0
0
0
0
1
0
0
2
45
48
41
41
49
45
9
4
0
8
6
3
1.5
AcOH
AcOH
AcOH
AcOH
AcOH
AcOH
(R,R)-L3
(R,R)-L4
a Product ratio determined by gas chromatography on LIPODEX E.
improved enantioselectivity resulted with toluene as
solvent (entry 3), whereas the diastereomeric ligand
(Sa,S,S)-L1 was inferior (entry 4). The biphenyl-derived
ligands L2–L4 provided slightly lower enantiomeric
excesses, compared to (Sa,R,R)-L1; here, the best values
(95%/>99% ee) were observed with (R,R)-L4. Interest-
ingly, in all cases examined here, the (S)-enantiomer of
enone 1 is reacting with higher enantioselectivity than
the (R)-enantiomer; the absolute configuration of the
new stereogenic center is the same as in the correspond-
ing additions to cyclohex-2-enone.4,5 Still, both enan-
tiomers of the enone reacted with virtually the same
rate, so that a kinetic resolution was not possible. This
was also true for the corresponding addition reactions
of dimethylzinc to rac-1 which, however, gave a 1:1-
mixture of the cis and trans adduct in enantiomerically
pure form when (Sa,R,R)-L1 or (R,R)-L4 were used as
chiral ligand (Fig. 2).
Figure 2. Copper-catalyzed enantioselective 1,4-addition of
diethylzinc to enone rac-1.
The possibility to conduct the catalytic enantioselective
Michael addition to enone 1 under reagent control and
with or without subsequent epimerization enables the
deliberate preparation of any stereoisomer of 2. Thus,
(S,S)-2 was obtained with >99% ds and ee by treating
rac-1 with diethylzinc, as well as catalytic amounts of
Cu(OTf)2 and (Sa,R,R)-L1 under epimerization condi-
tions (treatment with sodium methanolate proved to be
more efficient on larger scale than with dilute HCl),
whereas (2R,5S)-2 was formed with 96% ds and >99%
ee in the corresponding reaction of enantiomerically
pure enone (R)-(+)-18 under non-epimerization condi-
tions (workup with acetic acid; Fig. 3). Consequently,
the enantiomeric ketones (R,R)-2 and (2S,5R)-2 are
accessible by using the enantiomeric ligand with rac-1
or (S)-1.
Figure 3. Selective formation of (S,S)-2 and (2R,5S)-2 by
copper-catalyzed enantioselective Michael addition.
lective reaction (Table 2, entry 4). In all other cases,
(S,S)-5 was the major addition product, although the
reaction rates differ considerably; whereas with ligand
(R,R)-L2 the addition was virtually complete within 12
h at −20°C (entry 3), the corresponding reactions with
(Sa,R,R)-L1 and (R,R)-L4 were sufficiently slow to
allow a kinetic resolution of the enone. Thus, with
(Sa,R,R)-L1 and 0.8 equiv. of Et2Zn, (R)-(−)-414 was
recovered with 81% ee after 42% consumption (entry 2),
whereas (R,R)-L4 gave the enone with 67% ee/65%
consumption (entry 5).
Not surprisingly, the corresponding copper-catalyzed
enantioselective 1,4-additions to the t-butyl-substituted
enone rac-411 gave distinctly different results (Table 2).
In contrast to substrate 1, the (S)-enantiomer of enone
412 is reacting with higher rate and enantioselectivity
with the catalysts used here than (R)-4. An exception is
the ligand (R,R)-L3 which gave a very slow and unse-

L. Media6illa Urbaneja et al. / Tetrahedron Letters 43 (2002) 7887–7890
7889
Table 2. Copper-catalyzed enantioselective 1,4-addition of diethylzinc to enone rac-4a
Entry
Ligand
Time (h)
(2R,5S)-5 (%)
(R,R)-5 (%)
(S,S)-5 (%)
(2S,5R)-5 (%)
(R)-4 (%)
(S)-4 (%)
1
2
3
4
5
(Sa,R,R)-L1
(Sa,R,R)-L1b
(R,R)-L2
(R,R)-L3
(R,R)-L4
36
36
12
72
72
27.7
3.2
5.9
5.2
6.5
16.9
2.6
42.2
5.4
51.3
35.5
50.5
6.5
0.4
0.4
0.6
3.0
1.9
3.7
52.7
0.8
42.3
29.0
0
5.6
0
37.6
5.8
6.6
50.2
a Product ratio determined by gas chromatography on heptakis-(2,6-di-O-methyl-3-O-pentyl)-g-cyclodextrin.10,12,13
b Reaction with 0.8 equiv. of Et2Zn, 1 mol% of Cu(OTf)2, and 2 mol% of (Sa,R,R)-L1.
trans isomer. The latter strategy allows the deliberate
preparation of any stereoisomer of a 2,5-disubstituted
ketone. Further work with differently substituted
enones is continuing in our laboratories.
Acknowledgements
This work was supported by the Deutsche Forschungs-
gemeinschaft, the European Community (COST D12/
0022/99 and D24/0003/01), the Fonds der Chemischen
Industrie, and the Swiss National Fund.
References
1. Modern Organocopper Chemistry; Krause, N., Ed.;
Wiley-VCH: Weinheim, 2002.
2. (a) Alexakis, A.; Benhaim, C. Eur. J. Org. Chem., in
press; (b) Krause, N.; Hoffmann-Ro¨der, A. Synthesis
2001, 171–196; (c) Feringa, B. L.; Naasz, R.; Imbos, R.;
Arnold, L. A. In Modern Organocopper Chemistry;
Krause, N., Ed.; Wiley-VCH: Weinheim, 2002; pp. 224–
258.
3. Naasz, R.; Arnold, L. A.; Minnaard, A. J.; Feringa, B. L.
Angew. Chem. 2001, 113, 953–956; Angew. Chem., Int.
Ed. 2001, 40, 927–930.
4. Feringa, B. L.; Pineschi, M.; Arnold, L. A.; Imbos, R.;
De Vries, A. H. M. Angew. Chem. 1997, 109, 2733–2736;
Angew. Chem., Int. Ed. Engl. 1997, 36, 2620–2623.
5. Alexakis, A.; Rosset, S.; Allamand, J.; March, S.; Guil-
len, F.; Benhaim, C. Synlett 2001, 1375–1378.
6. Bispidine-derived ligands (Huttenloch, O.; Spieler, J.;
Waldmann, H. Chem. Eur. J. 2000, 6, 671–675) were also
examined but gave inferior enantioselectivities.
7. Cory, R. M.; Chan, D. M. T.; Naguib, Y. M. A.; Rastall,
M. H.; Renneboog, R. M. J. Org. Chem. 1980, 45,
1852–1863.
Figure 4. Kinetic resolution of enone 4 by copper-catalyzed
1,4-addition of di-n-butylzinc (product ratio determined by
gas chromatography on heptakis-(2,6-di-O-methyl-3-O-pent-
yl)-g-cyclodextrin; (R,R)- and (S,S)-6 could not be sepa-
rated).
A similar efficiency was observed in the corresponding
Michael addition of di-n-butylzinc to rac-4, catalyzed
by Cu(OTf)2 and (Sa,R,R)-L1 (Fig. 4) which was even
slower than the addition of diethylzinc. After 48 h at
−20°C, 55% of enone (R)-4 was recovered with 84% ee,
whereas enantiomerically pure enone resulted after 96 h
(78% consumption).
8. (a) Solladie´, G.; Hutt, J. J. Org. Chem. 1987, 52, 3560–
3566; (b) Laval, G.; Audran, G.; Galano, J.-M.; Monti,
H. J. Org. Chem. 2000, 65, 3551–3554.
9. Equilibrium ratio obtained by base-catalyzed epimeriza-
tion with NaOMe/MeOH: trans-2:cis-2=91:9.
10. Slight deviations from the expected 1:1 ratio of (2R)- and
(2S)-enantiomers are probably due to a slow epimeriza-
tion even under the less acidic workup conditions.
To summarize the results of this work, we have found
that copper-catalyzed enantioselective Michael addi-
tions of organozinc reagents to 6-substituted enones
can be conducted ‘substrate-oriented’, i.e. as a kinetic
resolution, or ‘product-oriented’ with or without subse-
quent enolization to the thermodynamically more stable

7890
L. Media6illa Urbaneja et al. / Tetrahedron Letters 43 (2002) 7887–7890
that (R)-1 and (S)-4 have the same absolute configuration.
14. Optical rotation for (R)-4 (72% ee): [h]2D0 −22 (c 3.0,
CHCl3), corresponding to a value of [h]D:−30 for the
enantiomerically pure enone (cf.: [h]2D1 +70 (c 3.0, CHCl3)
for (R)-(+)-18a).
11. Marino, J. P.; Jae´n, J. C. J. Am. Chem. Soc. 1982, 104,
3165–3172.
12. Equilibrium ratio obtained by base-catalyzed epimeriza-
tion with NaOMe/MeOH: trans-5:cis-5=90:10.
13. The Cahn–Ingold–Prelog rules cause a priority change, so