Significant asymmetric amplification in enantioselective Cu/DiPPAM-catalyzed 1,6- and 1,4-conjugate additions of diethylzinc to (Di)enones

Article abstract of DOI:10.1021/ol300866p

An unprecedented and strong positive nonlinear effect (NLE) was observed in Cu/DiPPAM-catalyzed asymmetric 1,6-conjugate addition of Et₂Zn on (E)-3-(prop-1-en-1-yl)cyclohex-2-enone (71% ee reached with a 40% ee ligand). Moreover, similar behaviors were observed in Cu/DiPPAM-catalyzed 1,4-asymmetric conjugate additions of Et₂Zn on both cyclohexenone and (E)-3-non-3-en-2-one (83% and 70% ee, respectively, reached with a 40% ee ligand).

Full text of DOI:10.1021/ol300866p

ORGANIC
LETTERS
2
012
Vol. 14, No. 14
576–3579
Significant Asymmetric Amplification in
Enantioselective Cu/DiPPAM-catalyzed
3
1,6- and 1,4-Conjugate Additions of
Diethylzinc to (Di)enones
†
,§
†,§
,‡
Magaly Magrez, Joanna Wencel-Delord, ,†,A§ lexandre Alexakis,*
,†,§
Christophe Cr ꢀe visy,* and Marc Mauduit*
Ecole Nationale Sup eꢀ rieure de Chimie de Rennes, CNRS, UMR 6226, Avenue du
G eꢀ n eꢀ ral Leclerc, CS 50837, 35708 Rennes Cedex 7, France, Department of Organic
Chemistry, University of Geneva, Quai Ernest Ansermet 30, 1211 Geneva, Switzerland,
and Universit eꢀ Europ eꢀ enne de Bretagne, 5 Bd La e€ nnec, 35000 Rennes, France
crevisy@ensc-rennes.fr; marc.mauduit@ensc-rennes.fr; Alexandre.Alexakis@unige.ch
Received April 5, 2012
ABSTRACT
An unprecedented and strong positive nonlinear effect (NLE) was observed in Cu/DiPPAM-catalyzed asymmetric 1,6-conjugate addition of Et Zn
2
on (E)-3-(prop-1-en-1-yl)cyclohex-2-enone (71% ee reached with a 40% ee ligand). Moreover, similar behaviors were observed in Cu/DiPPAM-
catalyzed 1,4-asymmetric conjugate additions of Et
with a 40% ee ligand).
2
Zn on both cyclohexenone and (E)-3-non-3-en-2-one (83% and 70% ee, respectively, reached
2
both cyclic and acyclic enones and also in 1,6-conjugate
,3
In order to extend the application of asymmetric cata-
lysis, particularly for industrial purposes, there is a need
for new chiral ligands that can be synthesized in very
short and efficient routes. In this context, we have recently
developed the DiPPAM family, which can be easily ob-
tained from diphenylphosphinobenzaldehyde and an ami-
no acid. These ligands, which are a simplified version of
4
addition of dialkylzinc on cyclic dienones. Although preli-
minary mechanistic studies had already been undertaken
dealing with the 1,4-conjugate addition reaction, they had
to be pursued and extended to the 1,6-conjugate addi-
tion. The observation of nonlinear effect (NLE) can help
1
(2) (a) Wencel, J.; Rix, D.; Crevisy, C.; Mauduit, M. PCT Int Appl
ꢀ
WO 2009050284, 2009. (b) Wencel, J.; Rix, D.; Jennequin, T.; Labat, S.;
Cr ꢀe visy, C.; Mauduit, M. Tetrahedron: Asymmetry 2008, 19, 1804–1809.
Hoveyda’s salen-peptide ligands, proved to be efficient in
Cu-catalyzed 1,4-conjugate additions of dialkylzinc to
(
3) Magrez, M.; Wencel-Delord, J.; Cr ꢀe visy, C.; Mauduit, M. Tetra-
hedron 2012, 68, 3507–3511.
(4) Wencel-Delord, J.; Alexakis, A.; Cr ꢀe visy, C.; Mauduit, M. Org.
Lett. 2010, 12, 4335–4337.
(5) For a recent review on NLE: Satyanarayana, T.; Abraham, S.;
Kagan, H. B. Angew. Chem., Int. Ed. 2009, 48, 456–494.
(6) For general reviews on Cu-catalyzed ACA, see: (a) Jerphagnon,
T.; Pizzuti, M. G.; Minnaard, A. J.; Feringa, B. L. Chem. Soc. Rev. 2009,
38, 1039–1075. (b) Alexakis, A.; B €a ckvall, J. E.; Krause, N.; P ꢁa mies, O.;
Di ꢀe guez, M. Chem. Rev. 2008, 108, 2796–2823.
†
Ecole Nationale Sup ꢀe rieure de Chimie de Rennes.
Universit ꢀe Europ ꢀe enne de Bretagne.
§
‡
University of Geneva.
1) (a) Degrado, S. J.; Mizutani, H.; Hoveyda, A. H. J. Am. Chem.
(
Soc. 2001, 123, 755–756. (b) Degrado, S. J.; Mizutani, H.; Hoveyda,
A. H. J. Am. Chem. Soc. 2002, 124, 13362–13363. (c) Brown, M. K.;
Degrado, S. J.; Hoveyda, A. H. Angew. Chem., Int. Ed. 2005, 44, 5306–
310. (d) Kacprzynski, M. A.; Kazane, S. A.; May, T. L.; Hoveyda,
A. H. Org. Lett. 2007, 9, 3187–3190.
5
1
0.1021/ol300866p r 2012 American Chemical Society
Published on Web 06/28/2012

5
explain reaction mechanisms. NLEs have been already
first prepared by reacting diphenylphosphinobenzaldehyde
with various mixtures of L- and D-tert-leucine (L/D ratio of
100/0, 90/10, 80/20, 70/30, 60/40, 55/45, 50/50; method A).
The ligands of optical purities between 10 and 80% were
tested two or three times to secure the reproducibility
(see Supporting Information). The results are presented in
Figure 1.
6
observed in Cu-catalyzed 1,4-conjugate additions to enones.
7
Early observations were reported in 1994 by Pfaltz and van
8
Koten with Grignard reagents. In the case of addition of
organozinc nucleophiles, both negative and positive NLEs were
reported. Moderate depletions were observed by Feringa in the
Cu/phosphoramidite-catalyzed addition on cyclohexenone
9
and chalcone, Fu in the Cu/phosphaferroceneÀoxazoline
1
0
catalyzed addition on 4-chlorochalcone, and Ito in the
1
1
Cu/phosphino-phenol-catalyzed addition to chalcone. An
amplification (58% ee was reached with a 40% ee ligand) in
agreement with the Kagan ML model was observed by Hu in
2
the Cu/bidentate phosphiteÀpyridine-catalyzed addition on
1
2
chalcone. In 2004, Alexakis and van Koten observed a posi-
tive NLE in the Cu/aminoarenthiolate-catalyzed addition on
1
3
cyclohexenone (50% ee was obtained with a 36% ee ligand).
Nevertheless, it must be underlined that similar observa-
tion has never been reported to date on the Cu-catalyzed
1
,6-conjugate addition reaction.
Scheme 1. Cu(II)/DiPPAM-Catalyzed 1,6-ACA
Figure 1. NLE in Cu(II)/DiPPAM-catalyzed 1,6-ACA.
a
b
ee was determined on chiral GC using a Betadex column. Average
values from multiple experiments (see supporting information for
details).
A strong positive NLE was observed as, for instance, the
product was obtained in 71% ee when a ligand having a ee
of 40% was used. It was checked first that the preparation
mode of the ligand did not have any influence on the result.
To that aim, (L)-L1 and (D)-L1 ligands were first prepared
separately then mixed in a 60/40 ratio before being used in
the catalysis (method B). In this case, the product was
isolated in a similar ee (45%) to this which was obtained
using method A. To check that the removal of the hexane
We decided to study the relationship between optical
purities of the tert-butyl DiPPAM ligand L1 and the
product, and we report our results here.
The1,6-conjugateadditionwasstudied usingdiethylzinc
and (E)-3-(prop-1-en-yl)cyclohex-2-enone 1 under the ex-
4
perimental conditions reported previously: 5 mol % of
Cu(OTf) and 10 mol % of the ligand were used, and the
2
reactions were run in Me-THF at 0 °C. After acidic quench-
ing, the product being partially reconjugated, the reconjuga-
tion step was completed in the presence of DBU (Scheme 1).
DiPPAM ligands L1 having different optical purities were
from the Et Zn solution before the addition of solvent and
2
reagents did not interfere with the result, an experiment
was run without the removal of hexanes using a 20% ee
ligand. In this case, the reaction was slower, but a similar ee
value was reached (43%). Moreover the evolution of the ee
value in the course of time was monitored in the reaction
using a ligand having an optical purity of 20%: seven
reactions were run simultaneously in the same conditions
but quenched at different times (from 1 h 30 to 24 h), and it
was observed that in all cases the ee’s were somewhat
constant (ranging from 37% to 50%). Besides, it can be
noted that the isolated yields dropped when low ee ligands
were used (10À40%). First, it was observed that the con-
version of the conjugate addition step was lower when a
(7) Zhou, Q.-L.; Pfaltz, A. Tetrahedron 1994, 50, 4467–4478.
(8) van Koten, G. Pure Appl. Chem. 1994, 66, 1455–1462.
(9) Arnold, L. A.; Imbos, R.; Mandoli, A.; de Vries, A. H. M.; Naasz,
R.; Feringa, B. L. Tetrahedron 2000, 56, 2865–2878.
(10) Shintani, R; Fu, G. C. Org. Lett. 2002, 4, 3699–3702.
(11) Ito, K.; Eno, S.; Saito, B.; Katsuki, T. Tetrahedron Lett. 2005, 46,
3
981–3985.
(
12) Hu, Y.; Liang, X.; Zheng, Z.; Hu, X. Tetrahedron: Asymmetry
2003, 14, 2771–2774.
(
13) Arink, A. M.; Braam, T. W.; Keeris, R.; Jastrzebski, J. T. B. H.;
Benhaim, C.; Rosset, S.; Alexakis, A.; van Koten, G. Org. Lett. 2004, 6,
959–1962.
1
Org. Lett., Vol. 14, No. 14, 2012
3577

low ee ligand was used. Moreover, the difficulties within
this reaction may arise mainly from the DBU-mediated
reconjugation step. Indeed, although the amount of non-
conjugated product was not dependent on the optical purity
of the ligand used, the reconjugation step was more difficult
when reaction was run with low ee ligand, requiring the
addition of more (2Â) DBU. Thus, the lower yields ob-
served in these cases may arise from the excess of DBU
at rt. In addition, it was found that similar shapes were
obtained whatever the configuration (L or D) of the ligand
used in excess was (Figure 2). Moreover, the evolution of
the ee in the course of time was monitored in the reaction
using ligand having an optical purity of 20%: four reac-
tions were run simultaneously in the same conditions but
quenched at different times (from 1 to 4 h). The determina-
14
which probably causes partial decomposition of products.
We then tried to evidence NLE in Cu-DiPPAM-cata-
lyzed 1,4-ACA. Experiments were first done on cyclohex-
enone 3 and diethylzinc. Although the reaction was
2
originally developed using Cu(I) triflate, efficiency of
Cu(I) and Cu(II) sources were first compared (Table 1).
Table 1. Cu/DiPPAM-Catalyzed 1,4-ACA on Cyclohexenone 3
d
a
copper
Cu/
L
Cu
temp
time
(h)
ee
(%)
c
cosolvent
source
(mol %)
(°C)
AcOEt
AcOEt
THF
(CuOTf)
(CuOTf)
Cu(OTf)
Cu(OTf)
Cu(OTf)
Cu(OTf)
Cu(OTf)
2
2
2
2
2
2
2
2/1
2/1
1/1
1/1
1/1
1/1
2/1
2
2
2
2
1
1
2
rt
À18
rt
2
89
6
93À98
2
94
b
THF
À18
rt
6
96
AcOEt
AcOEt
THF
3.5
5
92
91
94
À18
rt
2
Figure 2. NLE in Cu(II)/DiPPAM-catalyzed 1,4-ACA on 3.
a
b
ee was determined by chiral GC on a GTA column. 62% isolated
c
yield. (CuOTf)
d
was used as Cu(I) triflate source. Full conver-
a
2
3
C
sion was obtained in all experiments.
7
H
8
Average values from multiple experiments (see supporting informa-
tion for details).
Slightly better results were reached at room temperature
1
tion of the ee values showed that it was constant with the
progress of reaction (69, 67, 67, and 65%, respectively,
were obtained).
The study was then extended to acyclic enones using
non-3-en-2-one 5 as the substrate. Although the reaction
5
when a 1/1 Cu(II)/ligand ratio was used in THF/hexanes.
Thus, the following experiments were performed in the
1
6
latter conditions, using method A. (Figure 2). A signifi-
cant positive NLE was observed as, for instance, the
product was obtained in 83% ee when a ligand having a
ee of 40% was used. It was observed that the reaction was
slower with ligands having a low ee; nevertheless, conver-
sions >90% were obtained in all cases after 2À4 h reaction
17
Table 2. Cu/DiPPAM-Catalyzed 1,4-ACA on Acyclic Enone 5
(
14) The re-conjugation step using DBU is required to convert the
nonconjugate product and consequently to isolate a pure conjugated
product after chromatography. Nevertheless, NMR studies using inter-
nal standard showed that the yield in the conjugate product is not
increased but slightlydecreased after the DBU treatment. According to a
recent report, DBU mediated re-conjugation in 1,6-ACA products
appears problematic: Nishimura, T.; Noishiki, A.; Hayashi, T. Chem.
Commun. 2012, 48, 973–975.
e
d
a
cosolvent copper source Cu/L Cu (mol %) time (h) ee (%)
(
15) Hexanes from the Et
elimination has deleterious consequences on the conversions and ee’s.
16) When method B was used with a 20% ee ligand, the product was
2
Zn solution was not removed as its
AcOEt
THF
(CuOTf)
Cu(OTf)
Cu(OTf)
2
2
2
2/1
1/1
1/1
2
2
2
4
96
77
97
b
c
20
(
AcOEt
1.5
isolated with an ee (71%) similar to that which was obtained using
method A.
(
catalyzed 1,4-addition of thioglycolates to chalcone: Hui, Y.; Jiang, J.;
Wang, W.; Chen, W.; Cai, Y.; Lin, L.; Liu, X.; Feng, X Angew. Chem.,
Int. Ed. 2010, 49, 4290–4293.
a
b
Determined on chiral GC on a GTA column. Average value from
17) An outstanding NLE has recently been achieved in the La-
c
two experiments. 64% isolated yield. Full conversions were reached in
d
e
all experiments. (CuOTf)
2
3 C7
H
8
was used as Cu(I) triflate source.
3
578
Org. Lett., Vol. 14, No. 14, 2012

1
8
equiv of Et Zn and using both copper sources (Figure 3).
2
It can be seen that NLE was little more pronounced when
CuOTf) was used and less important to the effect ob-
(
2
served when cyclohexenone was employed asthe substrate.
It must be noted that, contrary to what was observed in the
1,6-ACA, the optical purity of the ligand has no influence on
the isolated yield. This difference of behavior may be explained
by the absence of reconjugation step in the 1,4-addition
reaction. Finally, the evolution of the ee versus the catalyst
loading was studied using a 20% ee ligand, and no significant
differences could be noticed, both with Cu(I) and Cu(II)
triflates. For instance, ee’s of 33%, 38%, and 31% were
obtained using, respectively, 1, 2, and 4 mol % of Cu(OTf)2.
To conclude, a strong positive NLE was evidenced for
the first time in the Cu-catalyzed 1,6-conjugate addition of
dialkylzinc on dienones. A similar effect was observed in
the 1,4-conjugate addition on both cyclic and acyclic
enones. This effect probably arises from the presence of
di(oligo)meric species in Cu/DiPPAM-catalyzed conju-
gate additions. All attempts (mass spectrometry, liquid
or solid NMR, X-ray) to determine the structure of these
oligomeric species have not been a success until now, but
efforts will continue to reach this goal.
Figure 3. NLE in Cu(I)- and Cu(II)/DiPPAM-catalyzed 1,4-
ACA on 5.
Acknowledgment. This work was supported by the
Minist ꢁe re de la Recherche et de la Technologie and by
the Universit ꢀe Europ ꢀe enne de Bretagne (grant to J.W.-D
and M.M.). The authors thank Pr. Henri Kagan for
fruitful exchange of information.
a
Average values from multiple experiments (see supporting informa-
tion for details).
2
was originally developed using Cu(I) triflate, efficiencies
of Cu(I) triflate and Cu(II) triflate were first compared
(
Supporting Information Available. Experimental pro-
cedures and spectral and chromatographic data for pro-
ducts. This material is available free of charge via the
Internet at http://pubs.acs.org.
Table 2).
The ee values obtained using a 2/1 Cu(I)/ligand ratio
were similar to those obtained with a 1/1 Cu(OTf) /ligand
2
ratio, but the reaction was faster in the latter case. NLE
experiments were run in AcOEt/hexanes at 30 °C with 3
Note Added after ASAP Publication. This manuscript
was published ASAP on June 28, 2012. Reference 18 was
erroneously removed during production. The corrected
version was reposted on July 3, 2012.
(
18) When method B was employed with a 20% ee ligand and the
Cu(II) source, the product was isolated in a similar ee (33%) to this
which was obtained using method A, but the ee was lower (32%) when
Cu(I) source was used. This is probably connected to the poorer
reproducibility of the reaction with Cu(I) triflate.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 14, 2012
3579