Direct asymmetric aldol reaction co-catalyzed by l-proline and group 12 elements Lewis acids in the presence of water

Article abstract of DOI:10.1016/j.tetlet.2010.11.014

An approach based on combinations of various water compatible Lewis acids and l-proline co-catalysts has been evaluated for the direct asymmetric aldol reaction. From this broad screening, chloride salts from group 12 elements (ZnCl₂, CdCl

Full text of DOI:10.1016/j.tetlet.2010.11.014

Tetrahedron Letters 52 (2011) 159–162
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Direct asymmetric aldol reaction co-catalyzed by L-proline and group 12
elements Lewis acids in the presence of water
⇑
Maël Penhoat , Didier Barbry, Christian Rolando
Miniaturization for Synthesis, Analysis and Proteomics, USR CNRS 3290 and Institut Eugène-Michel Chevreul, FR CNRS 2638, Université de Lille 1,
Sciences et Technologies, 59655 Villeneuve d’Ascq, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 June 2010
Revised 28 October 2010
Accepted 5 November 2010
Available online 11 November 2010
An approach based on combinations of various water compatible Lewis acids and L-proline co-catalysts
has been evaluated for the direct asymmetric aldol reaction. From this broad screening, chloride salts
from group 12 elements (ZnCl2, CdCl2, HgCl2) lead to the highest stereoselectivities. Optimized catalytic
conditions (catalytic system: L-proline: 20%/ZnCl2: 10%; solvent mixture: DMSO/H2O, 8:2) gave anti aldol
products with improved enantioselectivity (>99% ee) compared to a moderately stereoselective proce-
dure based on proline activation only.
Keywords:
Ó 2010 Elsevier Ltd. All rights reserved.
Aldol reaction
Dual activation
Lewis acids
Proline
The asymmetric aldol reaction is an important carbon–carbon
bond-forming reaction that results in the formation of either one
or two vicinal stereocenters.1 This process can be observed in many
different enzymatic transformations for the synthesis of complex
molecules. For many years the scientific community sought to ster-
eochemically control this reaction using small chiral organic mol-
ecules. The Mukaiyama aldol reaction of silyl enol ethers with
aldehydes catalyzed by Lewis acid transition metals and main
group elements chiral complexes is regarded as the first efficient
approach in this field.2 This strategy involves a preliminary trans-
formation of the ketone into a more active silyl enol ether, whereas
biocatalytic pathways based on aldol reactions between a ketone
and an aldehyde do not require this chemical activation. More re-
mation (3) side-reactions have been recently observed in the pres-
ence of aromatic aldehydes (Fig. 1).5a
Mono-Aldol Product 1 :
O
OH
O
OH
O
COOH
Ar
Ar
N
H
(20%)
R
R
R
R
R
R
R
R
(1'S,2S)-syn-1
OH
(1'R,2S)-anti-1
+
O
O
O
OH
cently, new metal-free organocatalysts derived from L-proline have
H
Ar
been reported to activate direct aldol reactions in some cases with
exceptional levels of stereoselectivity.3 Since the initial work of
Barbas III and coworkers on enamine processes,4 a wide variety
of small organic molecules have been successfully envisaged as
organocatalysts. However, the limitation of this approach lies in
the availability of such compounds, as their preparation is often
costly and requires many synthetic steps. In this regard, L-proline
may be considered as one of the most attractive of these
organocatalysts.
Ar
Ar
R
R
(1'R,2R)-syn-1
(1'S,2R)-anti-1
Side-Products :
H
O
H
O
N
O
N
Ar
R
R
Nevertheless numerous drawbacks still remain when using pro-
line as the catalyst, including moderate stereoselectivities and
dehydration by-products (2).5 Furthermore 1-oxapyrrolizidine for-
Ar
Ar
Ar
Ar
Dehydration Product 2
1-oxapyrrolyzidine 3
(syn and trans)
⇑
Corresponding author. Tel.: +33 3 20 43 70 75; fax: +33 3 20 33 61 36.
E-mail address: mael.penhoat@univ-lille1.fr (M. Penhoat).
Figure 1. Direct aldol reaction catalyzed by L-proline.
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.11.014

160
M. Penhoat et al. / Tetrahedron Letters 52 (2011) 159–162
Table 1
Lewis acid screening
In the past few years, different groups reported asymmetric al-
dol reactions catalyzed by proline in the presence of a large excess
of water.5a,6 In every cases, water addition leads to important
improvements of enantio- and diastereoselectivities. Under such
‘wet’ conditions the aldol reaction is slower but dehydration (2)
or 1-oxapyrrolizidine formation (3) side reactions are completely
suppressed.
O
OH
O
OH
O
COOH
N
H
NO2
NO2
(20%)
syn -1a
(1'R,2R) and (1'S,2S)
anti -1a
(1'R,2S) and (1'S,2R)
+
Considering a possible interaction between Lewis acids, L-pro-
O
Lewis Acid (y%)
DMSO/H20
(8 : 2)
line and the aldehyde generating a more activated catalytic system,
a suitable approach consists of the addition of various Lewis acids
to the reaction media. This hypothesis has recently raised an inter-
H
R.T., 24h
N
O
O2N
7
est among the scientific community and the most relevant work
reported in the literature has been described by Darbre and co-
workers with a Zn(prolinate)2 complex prepared under basic con-
ditions7a and by Mlynarski and co-workers with bis(prolina-
mides)/zinc(II) complexes.7e Aldol reactions catalyzed by
Zn(prolinate)2 were found to be moderately stereoselective (ee
up to 56% and de up to 54%). Bis(prolinamides)/zinc(II) complexes
proved to be one of the more stereoselective catalytic systems for
direct aldol reactions. In both cases, the use of water as a cosolvent
is required in order to obtain a complete solubilization of the com-
plex. However, with regards to these results, we were interested in
NO2
O2N
1-oxapyrrolyzidine 3a
(syn and trans)
Entrya
Lewis acid
(20%)
Conv.b
(%)
Ratiob
1a/3a
drb
anti/syn
ee
anti-1ac (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
—
>99
>99
>99
>99
<5
100:0
100:0
100:0
100:0
nd
100:0
100:0
100:0
100:0
100:0
nd
2.5:1
9:1
7:1
16:1
nd
7:1
3.5:1
1:1
2.5:1
2.5:1
nd
40
96
85
96
nd
91
80
24
9
61
nd
94
94
ZnCl2
FeCl3
HgCl2
CuCl2
FeSO4
InCl3
Sc(OTf)3
MgCl2
YbCl3
CuI
evaluating another strategy based on a combination of zinc(II)/
L-
82
30
proline and by extension of Lewis acid/ -proline combinations, by
L
simple mixing without pre-formation of the catalyst under basic
media. This alternative, realized under nearly neutral conditions,
was supposed to completely modify the structure of the active cat-
alytic complex previously studied by Darbre and to enable a rapid
screening of a wide range of Lewis acid co-catalysts. Interestingly,
both Lewis acid and organocatalytic activations are largely inspired
from enzymatic mechanisms encountered in nature.8 The enamine
>99
>99
>99
<5
>99
31
CdCl2
PdCl2
100:0
100:0
11:1
2:1
a
General procedure: to
a
solution of L-proline (20%), p-nitrobenzaldehyde
(1 mmol), and Lewis acid (20%) in a mixture of DMSO/H2O, 8:2 v/v (3 mL) is added
the cyclohexanone (5 mmol). The reaction is then stirred at room temperature for
24 h.
pathway in L-proline catalysis is analogous to the one observed in
class I aldolases and Lewis acid activation is similar to class II aldol-
ases mechanism involving a zinc(II) cofactor.
b
Determined by 1H NMR on the crude material.
c
Determined by chiral HPLC (Chiralpack IB) based on the major (10R,2S)-anti-1a
Lewis acid catalysis in the presence of water is not trivial, since
most of them are decomposed under aqueous conditions. Never-
theless, a few examples of water-tolerant Lewis acids have previ-
ously been reported by Kobayashi et al.9 Furthermore, Lewis
acids in aqueous media are known to coordinate to a molecule of
water to generate metallo-hydroxonium species leading to a nearly
neutral pKa value.10 In the presence of ligands, those metallo-
hydroxonium species are able to dissociate and equilibrate with
different ligand–metal complexes. Such dynamic behaviours are
enantiomer.
As a primary observation, the addition of Lewis acids to the
reaction medium has no effect on the chemical selectivity, and
no 1-oxapyrrolizidines (3a) side products are observed. However,
Lewis acids have important effects on the stereoselectivity and
the rate of the reaction. It is interesting to note that with or with-
out Lewis acid the same enantiomer (10R,2S)-anti-1a is observed in
excess in every experiments. Furthermore, ZnCl2, FeCl3, CdCl2,
HgCl2, FeSO4, InCl3, PdCl2 and YbCl3 have a positive effect on the
enantioselectivity of the major anti-diastereomer (up to 96%, en-
tries 2 and 4). Depending on the nature of the Lewis acid, diastere-
oselectivity varies considerably, and important positive effects in
favor of the anti-diastereomer are observed in the case of ZnCl2,
FeCl3, HgCl2, CdCl2, and FeSO4. Other evaluated Lewis acids did
not improve significantly the diastereoselectivity of the reaction.
Interestingly, from these Lewis acids with a positive effect, the best
results in term of conversion and stereoselectivity were obtained
with chloride salts from group 12 elements.
observed in metallo-enzymes. In our case, L-proline could activate
the ketone via an enamine intermediate and at the same time
interact with various metal salts to activate the aldehyde partner.
Furthermore, the nature of the metal could have important effects
on both stereo- and chemical selectivities.
In order to evaluate this approach we first decided to screen
known water compatible Lewis acids (ZnCl2, FeCl3, HgCl2, CuCl2,
FeSO4, InCl3, Sc(OTf)3, MgCl2, YbCl3, CdCl2, PdCl2) in the model reac-
tion between cyclohexanone and p-nitrobenzaldehyde catalyzed by
L-proline in a 8:2 v/v, DMSO/water solvent mixture (Table 1). These
reaction conditions are suitable to evaluate the different effects of
Lewis acids on both stereo- and chemical selectivities and on the
rate of the reaction by comparison with more standard conditions
based on proline alone (entry 1). Without any Lewis acid, in the
For ecotoxicological considerations,11 we next turned our ef-
forts on the L-proline/zinc(II) couple that gave promising results
presence of 20 mol %
L
-proline a reaction time of 18 h has been
during the screening step. In order to optimize the conditions (Ta-
ble 2), we envisaged to modify the solvent mixture composition by
varying the amount of water (entries 1–3), the L-proline/zinc(II) ra-
tio (entries 1, 4, 6, and 7), and the nature of the zinc counterion
(entry 5). As expected, when DMSO is used alone as the solvent,
the absence of water affects dramatically both aldol chemical
selectivity (ratio 1a/3a = 59:41) and stereoselectivity. As previ-
ously described by Darbre and co-workers no reaction occurs in
the presence of the ZnCl2 only.7d When the reaction is done ‘in
found to be optimal for a complete conversion. Increasing the reac-
tion time to completion and decreasing the amount of catalyst re-
sulted in lower stereoselectivities compared to Hayashi’s best
result under
of 96% ee for Hayashi after 2 h using 30 mol % of
L
-proline organocatalysis,6e (40% ee in our case instead
L-proline), suggest-
ing a partial racemisation of the product over time. As expected, no
1-oxapyrrolizidines (3a) was observed even after 18 h, thus con-
firming a crucial effect of water on the chemical selectivity.

M. Penhoat et al. / Tetrahedron Letters 52 (2011) 159–162
161
Table 2
Optimisation of the L-proline/zinc(II) catalytic conditions
Entrya
Lewis acid (y%)
Ratio DMSO/H2O
Conv.b (%)
Ratiob 1a/3a
drb anti/syn
ee anti-1ac (%)
L
-Proline (x%)
1
2
3
4
5
6
7
20
20
20
—
20
20
10
ZnCl2 (20%)
ZnCl2 (20%)
ZnCl2 (20%)
ZnCl2 (20%)
Zn(OTf)2 (20%)
ZnCl2 (10%)
ZnCl2 (5%)
80:20
100:0
0:100
80:20
80:20
80:20
80:20
>99
>99
2
3
100:0
59:41
nd
9:1
1.5:1
nd
96
63
nd
nd
89
>99
74
nd
nd
>99
100:0
100:0
100:0
9:1
16:1
7:1
>99 (99d)
63
a
General procedure: to a solution of L-proline (x%), p-nitrobenzaldehyde (1 mmol), and Lewis acid (y%) in a mixture of DMSO/H2O, 8:2 v/v (3 mL) is added the cyclo-
hexanone (5 mmol). The reaction is then stirred at room temperature for 24 h.
b
Determined by 1H NMR on the crude material after extraction and evaporation.
c
Determined by chiral HPLC (Chiralpack IB) based on the major (10R,2S)-anti-1a enantiomer.
d
Isolated yield for the aldol product 1.
O
H
O
H
Cl Cl
Zn2-
procedure B). In order to evaluate the benefit of our approach this
methodology was compared to the standard procedure based on
L
-
O
O
N+
H
N+
H
proline (20%) catalysis alone in DMSO (Table 3, procedure A). In
every case, procedure B produces higher yields and stereoselectiv-
ities with an excellent 1 to 3 ratio. Our catalytic system thus offers
a good to excellent control toward the formation of the (10R,2S)-
anti-1 product (80% < ee < 99%).
Figure 2. Structure of the [(L
-Pro)2ZnCl2] complex.12
In summary, the study of the effect of various water compatible
Lewis acids on L-proline organocatalysis in the presence of water
water’, the conversion is extremely low, confirming the importance
of the organic solvent. Replacement of chloride anions by triflate
counterions which increase Lewis acidity resulted in lower stere-
oselectivities. In terms of conversion and stereoselectivities best
led us to new findings. From this broad screening, the chloride salts
of the group 12 elements (ZnCl2, CdCl2, HgCl2) led to the highest
stereoselectivities and conversions. The optimized catalytic condi-
tions (L-proline: 20%; ZnCl2: 10%) gave anti-products with im-
results were obtained with a 2:1 ratio of
tively 20/10 mol %) suggesting that the catalytic system may be
based on a [(
-Pro)2ZnCl2]12 complex (entry 6, Fig. 2) for which
L-proline/ZnCl2 (respec-
proved enantioselectivities ranging from 80–99% ee when
compared to the standard and moderately stereoselective proce-
dure without Lewis acid. At this stage, the exact nature of the ac-
tive catalytic complex is still an important question, as different
possible complexes in equilibrium with water can be envisaged.
This procedure appears to be an excellent and readily available bio-
mimetic model analogous to Class II aldolases, which opens new
interrogations concerning the exact nature of the chemical bonds
involved in such enzymes.
L
the structure has already been described by Lutz and Bakker after
crystallization under similar conditions. Attempts to decrease the
amount of both co-catalysts (L-proline, 10%; ZnCl2, 5%) lead to low-
er conversion, diastereoselectivity, and enantioselectivity.
With these optimum conditions in hand we then decided to
vary the electronic demand of the aromatic aldehydes (Table 3,
Table 3
Scope and limitation of the L-proline/ZnCl2 (2:1) dual activation in the presence of water
Entry
Ketone
Aldehyde
Procedurea
Conversionb (%)
Ratiob 1/3
drb anti/syn
ee anti-1c (%)
O
CHO
CHO
CHO
1
2
A
B
99 (60d)
65:35
100:0
1:1
16:1
39
>99
100 (99d)
O2N
O
O
O
O
3
4
A
B
15
7
75:25
100:0
1:1
8:1
nd
nd
MeO
5
6
A
B
99
65:35
100:0
1:1
3:1
14
87
100 (99d)
F C
3
CHO
CHO
7
8
A
B
90
85:15
100:0
1:1
5:1
8
85
100 (99d)
9
10
A
B
45
60
30:70
100:0
1:1
9:1
58
80
O
a
General procedure A: to a solution of
L-proline (20%) and aldehyde (1 mmol) in DMSO (3 mL) is added cyclohexanone (5 mmol). The reaction is then stirred at room
temperature for 24 h. General procedure B: to a solution of
L-proline (20%), aldehyde (1 mmol), and Lewis acid (10%) in a mixture of DMSO/H2O, 8:2 v/v (3 mL) is added the
cyclohexanone (5 mmol). The reaction is then stirred at room temperature for 24 h.
b
Determined by 1H NMR on the crude material.
c
Determined by average of two sample by chiral HPLC (Chiralpack IB) based on the major (10R,2S)-anti-1 enantiomer.
d
Isolated yield for the aldol product 1.

162
M. Penhoat et al. / Tetrahedron Letters 52 (2011) 159–162
5. (a) Zotova, N.; Franzke, A.; Armstrong, A.; Blackmond, D. G. J. Am. Chem. Soc.
Acknowledgments
2007, 129, 15100–15101; (b) Reaction between p-methoxybenzaldehyde and
cyclopentanone under proline catalysis in DMSO produce essentially the
dehydration product after 24 h. Reactions between substituted benzaldehydes
and cyclohexanone under proline catalysis in dry DMSO are not clean with
formation of 1-oxapyrrolizidines.
We acknowledge the European Community (FEDER), the Région
Nord-Pas de Calais (France) and the CNRS for their financial contri-
butions to this work.
6. (a) Hayashi, Y. Angew. Chem., Int. Ed. 2006, 45, 8103–8104; (b) Mase, N.; Nakai,
Y.; Ohara, N.; Yoda, H.; Takabe, K.; Tanaka, F.; Barbas, C. F., III J. Am. Chem. Soc.
2006, 128, 734–735; (c) Hayashi, Y.; Sumiya, T.; Takahashi, J.; Gotoh, H.;
Urushima, T.; Shoji, M. Angew. Chem., Int. Ed. 2006, 45, 958–996; (d) Aratake, S.;
Itoh, T.; Okano, T.; Usui, T.; Shoji, M.; Hayashi, Y. Chem. Commun. 2007, 2524–
2526; For important references on the role of water in aldol reaction see: (e)
Hayashi, Y.; Itoh, T.; Nagae, N.; Ohkubo, M.; Ishikawa, H. Synlett 2008, 1565–
1570.
7. (a) Darbre, T.; Machuqueiro, M. Chem. Commun. 2003, 1090–1091; (b) Kofoed,
J.; Machuqueiro, M.; Reymond, J.-L.; Darbre, T. Chem. Commun. 2004, 1540–
1541; (c) Kofoed, J.; Darbre, T.; Reymond, J.-L. Chem. Commun. 2006, 1482–
1484; (d) Fernandez-Lopez, R.; Kofoed, J.; Machuqueiro, M.; Darbre, T. Eur. J.
Org. Chem. 2005, 5268–5276; (e) Kofoed, J.; Reymond, J.-L.; Darbre, T. Org.
Biomol. Chem. 2005, 3, 1850–1855; (f) Paradowska, J.; Stodulski, M.; Mlynarski,
J. Adv. Synth. Catal. 2007, 349, 1041–1046; (g) Akagawa, K.; Sakamoto, S.; Kudo,
K. Tetrahedron Lett. 2005, 46, 8185–8187; (h) Nakagawa, M.; Nakao, H.;
Watanabe, K.-I. Chem. Lett. 1985, 391–394; For a review see: (i) Paradowska, J.;
Stodulski, M.; Mlynarski, J. Angew. Chem., Int. Ed. 2009, 48, 2–12.
Supplementary data
Supplementary data (general procedures, spectroscopic data,
and chiral chromatography conditions) associated with this article
can be found, in the online version, at doi:10.1016/j.tetlet.
2010.11.014.
References and notes
1. For a review: Machajewski, T. D.; Wong, C.-H. Angew. Chem., Int. Ed. 2000, 39,
1352–1374.
2. (a) Mukaiyama, T.; Banno, K.; Narasaka, K. J. Am. Chem. Soc. 1974, 96, 7503–
7506; (b) Kobayashi, S.; Fujishita, Y.; Mukaiyama, T. Chem. Lett. 1990, 1455–
1458.
8. For reviews on aldolases and L-proline mechanism: (a) Gijsen, H. J. M.; Qiao, L.;
3. For reviews concerning proline catalyzed reactions see: (a) Groger, H.; Wilken,
J. Angew. Chem., Int. Ed. 2001, 40, 529–532; (b) Notz, W.; Tanaka, F.; Barbas, C.
F., III Acc. Chem. Res. 2004, 37, 580–591; (c) Saito, S.; Yamamoto, H. Acc. Chem.
Res. 2004, 37, 570–579; (d) Allemann, C.; Gordillo, R.; Clemente, F. R.; Cheong,
P. H.-C.; Houk, K. N. Acc. Chem. Res. 2004, 37, 558–569; (e) Barbas, C. F., III
Angew. Chem., Int. Ed. 2007, 46, 2–8; (f) Mukherjee, S.; Yang, J. W.; Hoffmann, S.;
List, B. Chem. Rev. 2007, 107, 5471–5569.
4. (a) Wagner, J.; Lerner, R. A.; Barbas, C. F., III Science 1995, 270, 1797–1800; (b)
Barbas, C. F., III; Heine, A.; Zhong, G.; Hoffmann, T.; Gramatikova, S.;
Bjornestedt, R.; List, B.; Anderson, J.; Stura, E. A.; Wilson, I. A.; Lerner, R. A.
Science 1997, 278, 2085–2092; (c) Zhong, G.; Hoffmann, T.; Lerner, R. A.;
Danishefsky, S.; Barbas, C. F., III J. Am. Chem. Soc. 1997, 119, 8131–8132; (d) List,
B.; Lerner, R. A.; Barbas, C. F., III Org. Lett. 1999, 1, 59–61.
Fitz, W.; Wong, C.-H. Chem. Rev. 1996, 96, 443–473; (b) Gröger, H.; Wilken, J.
Angew. Chem., Int. Ed. 2001, 40, 529–532.
9. (a) Kobayashi, S.; Manabe, K. Pure App. Chem. 2000, 1373–1380; (b) Kobayashi,
S. Eur. J. Org. Chem. 1999, 15227–15232; (c) Kobayashi, S.; Nagayama, S.;
Busujima, T. J. Am. Chem. Soc. 1998, 120, 8287–8288; (d) Otto, S.; Bertoncin, F.;
Engberts, J. B. F. N. J. Am. Chem. Soc. 1996, 118, 7702–7707; (e) Engberts, J. B. F.
N.; Feringa, B. L.; Keller, E.; Otto, S. Rec. Trav. Chim. Pays-Bas 1996, 115, 457–
464; (f) Kobayashi, S.; Ogawa, C. Chem. Eur. J. 2006, 12, 5954–5960.
10. For a review on the roles of zinc in zinc enzymes: Kimura, E.; Kikuta, E. J. Biol.
Inorg. Chem. 2000, 5, 139–155.
11. Shaw, B. P.; Rout, N. P. Biol. Plant. 2002, 45, 267–271.
12. Lutz, M.; Bakker, R. Acta Crystallogr., Sect. C 2003, 59, 18–20.