An achiral-acid-induced switch in the enantioselectivity of a chiral cis-diamine-based organocatalyst for asymmetric aldol and Mannich reactions

Article abstract of DOI:10.1002/anie.201107239

Pick and choose: The asymmetric synthesis of two different enantiomeric products has been achieved through the use of a single organocatalyst with or without achiral organic acid additives (see scheme). These additives may assist in altering the substrate

Full text of DOI:10.1002/anie.201107239

Angewandte
Chemie
DOI: 10.1002/anie.201107239
Organocatalysis
An Achiral-Acid-Induced Switch in the Enantioselectivity of a Chiral
cis-Diamine-Based Organocatalyst for Asymmetric Aldol and Mannich
Reactions**
Shin A. Moteki, Jianwei Han, Satoru Arimitsu, Matsujiro Akakura, Keiji Nakayama, and
Keiji Maruoka*
From a practical point of view, the development of a novel
approach for the asymmetric synthesis of both enantiomeric
products through catalytic asymmetric transformations with
the same chiral catalyst would be very useful.[1] Various types
of chiral metal complexes have already been introduced.[2]
However, strictly speaking, many of these examples have
employed a distinct three-dimensional association between a
chiral ligand and different metals, or vice versa. In marked
contrast, an initial effort for the synthetic application of chiral
organocatalysts has appeared very recently, but has not been
developed to a synthetically useful level.[3] In this context, we
explored the capability of an additive to induce an unex-
pected inversion in catalyst selectivity for certain asymmetric
transformations catalyzed by a single chiral organocatalyst.
Herein we present the first practical example of such a
system, employing achiral, organic acids as reliable additives
in asymmetric, direct aldol reactions catalyzed by a chiral, cis-
diamine-based, Tf-amido organocatalyst of type 1 (Tf = tri-
fluoromethanesulfonyl Scheme 1).[4–7]
Initally, the asymmetric direct aldol reaction of cyclo-
hexanone and a-keto ester 3a was carried out with organo-
catalyst 1a to establish the optimum reaction conditions. The
treatment of cyclohexanone 2a and a-keto ester 3a with
20 mol% of 1a in methanol at 08C gave the corresponding
aldol products 4a in moderate yields; the major isomer, syn-
4a, was obtained with high enantioselectivity (Table 1,
entry 1). We had previously observed a remarkable enhance-
ment in enantioselectivity for the asymmetric conjugate
addition of heterosubstituted aldehydes to vinyl sulfones by
using bulky benzoic acid additives under the influence of a
structurally rigid, trans-diamine-based, Tf-amide catalyst with
a dihydroanthracene framework.[8] This finding prompted us
to test a series of benzoic acid derivatives to probe their
potential for the reversal of enantioselectivity in these
products as well.
Table 1: Screening of reaction conditions for asymmetric aldol
reactions.[a]
Scheme 1. Chiral cis-diamine-based regioisomeric organocatalysts.
Tf =trifluoromethanesulfonyl.
Entry Additive[b]
Yield[c] [%]
d.r.[d]
aldol ee[e] [%]
(syn/anti)
1
2
3
4
5
6
7
8
none
C6F5OH
4-(NMe2)-C6H4CO2H
C6H5CO2H
4-FC6H4CO2H
4-NO2-C6H4CO2H
3,5-F2C6H3CO2H
2,6-F2C6H3CO2H
2,6-(OH)2C6H3CO2H
2,6-(NO2)2C6H3CO2H
C6F5CO2H
63
78
98
95
94
90
87
75
38
72
71
29
18
8.1:1
5.0:1
4.8:1
5.2:1
5.2:1
5.6:1
6.2:1
7.3:1
6.0:1
5.9:1
9.2:1
3.7:1
3.0:1
4a
4a
4a
4a
4a
4a
4a
4a
5a
4a
5a
5a
5a
94
58
88
81
79
72
72
20
45
59
42
0
[*] Dr. S. A. Moteki, J. Han, S. Arimitsu, Prof. Dr. K. Maruoka
Laboratory of Synthetic Organic Chemistry and Special Laboratory
of Organocatalytic Chemistry, Department of Chemistry
Graduate School of Science, Kyoto University
Sakyo, Kyoto 606-8502 (Japan)
E-mail: maruoka@kuchem.kyoto-u.ac.jp
9
Dr. M. Akakura
Department of Chemistry, Aichi University of Education
Igaya-cho, Kariya 448-8542 (Japan)
10
11
12
13
C6F5SO3H
CF3CO2H
5
Dr. K. Nakayama
Process Technology Research Laboratories, Daiichi Sankyo Co., LTD
Hiratsuka, Kanagawa 254-0014 (Japan)
[a] Unless otherwise specified, the asymmetric aldol reaction of cyclo-
hexanone and a-keto ester 3a was conducted in the presence of
20 mol% of catalyst 1a and 20 mol% of acid additives for 16 h under the
given conditions. [b] Order based on decreasing pKa values (approxi-
mate). [c] Yield of isolated product. [d] Diastereoselectivity was deter-
mined by 1H NMR spectroscopy. [e] Enantiopurity of aldols was
determined by HPLC analysis on a chiral stationary phase with hexane/2-
propanol as solvent (see the Supporting Information).
[**] This work was partially supported by a Grant-in-Aid for Scientific
Research from MEXT (Japan). We thank the Research Center for
Computational Science, Okazaki (Japan) for the use of their facility
in our computational studies.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201107239.
Angew. Chem. Int. Ed. 2012, 51, 1187 –1190
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1187

.
Angewandte
Communications
Initially, several para-substituted benzoic acids were
tested to identify the correlation between acidity and the
reaction outcomes. As shown in Table 1, the addition of
benzoic acid derivatives lowered the overall enantioselectiv-
ities obtained in the reaction (entries 3–6 vs. entry 1). The acid
strength seems to influence the reaction rate as well as the
stereoselectivity. Although weaker acids resulted in higher
yields, as acid strength increased, diastereoselectivity
improved and a further reduction in enantioselectivity was
observed (entries 2–6).
Inversion of catalyst selectivity (that is, generation of syn-
5a) was observed with both 2,6-dihydroxybenzoic acid and
pentafluorobenzoic acid (Table 1, entries 9 and 11). Changing
the nature of the acids induced a drastic effect on the reaction
outcome; a further increase in acidity resulted in a negative
effect on the reaction rate (entries 12 and 13). As shown in
Table 2, the acid stoichiometry also has a significant effect on
tions of various ketones and a-keto esters in the presence of
catalyst 1a (Table 3). The use of catalyst 1a alone always gave
syn-aldol products (syn-4) with one (2R,1’R for syn-4a)
Table 3: Practical synthesis of both enantiomeric aldols in asymmetric
direct aldol reactions catalyzed by 1a with or without pentafluorobenzoic
acid as additive.[a]
Entry Product
Additive & sol-
Yield[d] [%]
(syn/anti)
ee[e] [%]
vent[b,c]
1
2
MeOH
C6F5CO2H/H2O/
MeCN
93 (8:1)
>99 (8:1)
94
À88
3[f]
MeOH
96 (7.3:1)
À72
4
5
MeOH
C6F5CO2H/H2O/
MeCN
83 (2.3:1)
98 (8:1)
89
À87
6
7
MeOH
C6F5CO2H/H2O/
MeCN
87 (8:1)
91 (7:1)
90
À93
Table 2: Effect of additives in asymmetric aldol reactions.[a]
8
9
MeOH
C6F5CO2H/brine
62 (20:1)
70 (20:1)
95
À94
10
11
neat
C6F5CO2H/brine
44 (15:1)
57 (16:1)
87
À93
Entry Solvent
Equivalents Yield[b] [%] syn/anti[c] ee[d] [%]
of additive
1
2
3
4
5
6
7
8
9
MeOH
MeOH
MeOH
MeOH
0
1
3
10
0
1
0
1
1
63
71
37
32
69
65
47
68
74
8.1:1
94
12
13
MeOH
C6F5CO2H/H2O/
MeCN
88 (1:2.7)
98 (1:6.7)
93
À87
9.2:1 À42
13:1 À58
13:1 À60
MeCN
MeCN
H2O
9:1
92
14
15
MeOH
C6F5CO2H/H2O/
MeCN
57 (20:1)
41 (18:1)
95
À96
9:1 À43
6:1 À58
8:1 À90
8:1 À88
H2O/MeOH[e]
H2O/MeCN[e]
[a] Unless otherwise specified, asymmetric aldol reaction of cyclic ketone
2 and a-keto ester 3 was conducted in the presence of 20 mol% of
catalyst 1a for 40 h under the given conditions. [b] Volume ratio of H2O/
MeCN=1:1. [c] Certain substrates do not dissolve in aqueous MeOH,
and thus the use of aqueous MeCN is recommended instead. [d] Yield of
isolated product. [e] Enantiopurity of major aldol isomers was deter-
mined by HPLC analysis on a chiral stationary phase with hexane/2-
propanol as solvent (see the Supporting Information). [f] With catalyst
1b.
[a] Unless otherwise specified, the asymmetric aldol reaction of cyclo-
hexanone and a-keto ester 3a was conducted in the presence of
20 mol% of catalyst 1a for 16 h under the given conditions. [b] Yield of
isolated product. [c] Diastereoselectivity was determined by 1H NMR
spectroscopy. [d] Enantiopurity of aldols was determined by HPLC
analysis on a chiral stationary phase with hexane/2-propanol as solvent
(see the Supporting Information). [e] Volume ratio=1:1.
the reaction outcome. As acid stoichiometry increases,
regardless of their acid strengths, improved diastereoselctiv-
ities while diminishing enantioselectivity values (Table 2,
entries 1–4; see also the Supporting Information). The use
of acetonitrile did not alter reactivity or selectivity in this
aldol reaction and can be employed as a solvent when the
substrate shows poor solubility in methanol (entries 5 and 6).
Interestingly, switching the solvent to water also resulted in a
reversal in enantioselectivity (entry 7). Similarly, a large
reversal effect was observed when aqueous methanol (or
acetonitrile) was used along with one equivalent of penta-
fluorobenzoic acid (entries 8 and 9). However, the enantio-
meric excess remained unchanged when 2,6-dihydroxyben-
zoic acid was employed in the aqueous methanol system.[9]
With the optimal reaction conditions in hand, we further
investigated the generality of asymmetric, direct aldol reac-
configuration, while use of pseudoenantiomeric catalyst 1b
afforded products (syn-5) with the opposite (2S,1’S for syn-
5a) configuration (entry 1 vs. 3). The absolute structure of the
aldol product was unambiguously confirmed by X-ray crys-
tallographic analysis (see the Supporting Information). For
reactions with an acid additive, an aqueous acetonitrile
solvent system was employed instead of aqueous methanol
owing to solubility issues with some of the a-keto esters in
these experiments. Under these conditions, the reverse in
enantioselectivity was higher than that of the reaction
catalyzed by 1b (entry 2 vs. 3). Both cyclic and acyclic
ketones were tested for various a-keto esters, including
alkenyl- and Phenyl-substituted a-keto esters (entries 1–15).
In all cases, an efficient inversion of product configuration
was observed. To our knowledge, such a high magnitude of
1188
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1187 –1190

Angewandte
Chemie
enantiomeric inversion using achiral additives has never been
previously reported.
A similar reversal in catalyst selectivity was observable for
the asymmetric, anti-selective Mannich reaction of a cyclic
imino ester[10] (Table 4, entries 1–7). Optimum results were
Table 4: Practical synthesis of both enantiomeric Mannich products in
asymmetric Mannich reactions catalyzed by 1a with or without an acid
additive.[a]
Entry Product
Additive/solvent
none/DMF
Yield[b] [%] ee[c]
(anti/syn) [%]
Figure 1. Possible transition-state structures optimized by B3LYP/6-
90
(>20:1)
93
(>20:1)
86 (18:1)
31G*.[15,16]
1
2
>99
2,6-(NO2)2C6H3CO2H/
DMF
C6F5CO2H/DMF
À88
3
90
ketone carbonyl group while the acidic enamine NH hydro-
gen atom is involved in the recruitment of the acid additive,
coordinating onto the carbonyl oxygen of the acid through a
hydrogen bond. From several scenarios that we tested, it
appears that the involvement of one molecule of water is
required for the observed inversion in catalyst enantioselec-
tivity, as shown in Figure 1.[13] The water molecule, which
might be generated by formation of imine from ketones, is
involved in bridging the added acid to the a-keto ester,
leading to the favorable formation of transition state (B) and
thus furnishing the enantiomer syn-5.
In the absence of acid additives, a switch in solvent from
methanol (or acetonitrile) to water also resulted in inversion
of the enantiomeric configuration of the product (Table 2,
entries 1 vs. 7). However, in direct aldol reactions between
various ketones and benzaldehyde derivatives, we did not
observe any major differences in enantioselectivity or diaste-
reoselectivity when the solvent was changed from methanol
to water. Therefore, it is less likely that the catalyst ring
conformation changes, unlike those cases previously reported
with peptide based catalysts.[14] More detailed mechanistic
studies using DFT calculations are currently underway.
In summary, we have succeeded in obtaining both
enantiomeric aldol and Mannich products by using a single
chiral organocatalyst 1a in the presence or absence of achiral
acids as additives. In principle, this strategy is applicable to
other keto and imino esters, as well as other catalytic systems.
Further efforts towards this end, as well as more detailed
mechanistic studies, are currently underway in our laboratory.
98
(>20:1)
93
4
5
none/DMF
>99
À83
2,6-(NO2)2C6H3CO2H/
DMF
(>20:1)
70
(16:1)
74
6
7
none/DMF
98
2,6-(NO2)2C6H3CO2H/
DMF
À87
(19:1)
[a] Unless otherwise specified, asymmetric Mannich reaction of various
ketones 2 and cyclic imine 6a was conducted in the presence of 10 mol%
of catalyst 1a for 60 h under the given conditions. [b] Yield of isolated
product. [c] Enantiopurity of major aldol isomers was determined by
HPLC analysis on a chiral stationary phase.
obtained when the reactions were conducted in DMF using
10 mol% of catalyst 1a in the absence of an acid additive
(entries 1 and 4). In contrast to the aldol reaction described
above, the use of pentafluorobenzoic acid did not reverse the
enantioselectivity (entry 3). However, the use of 10 mol%
2,6-dinitrobenzoic acid did provide for an efficient inversion
in product configuration (entries 2, 5, and 7). Unlike the aldol
reaction with an a-keto ester, the addition of water did not
improve this enantioselectivity reversing effect.[11]
Based on the above observations, we propose two possible
transition state models (Figure 1a and b), which are opti-
mized at the B3LYP/6-31G* level of approximation, to
account for the observed absolute configuration of both
aldol products. In the absence of acid additives, a-ketoesters
are chelated and activated through the carbonyl group of the
ketone by both the acidic Tf-amide hydrogen atom and the
acidic enamine NH hydrogen atom. The enamine moiety
would then attack from the back side, as shown in (Figure 1a)
to furnish aldol product syn-4.
Received: October 13, 2011
Published online: December 21, 2011
Keywords: additives · aldol reaction · asymmetric synthesis ·
.
Mannich reaction · organocatalysis
On the other hand, in the presence of acid additives and
water, the two point chelation ability of a-keto esters plays an
important role in the reversal of enantioselectivities.[12] The
acidic Tf-amide hydrogen atom in catalyst 1a activates the
[1] Recent reviews: a) M. P. Sibi, M. Liu, Curr. Org. Chem. 2001, 5,
719; b) G. Zanoni, F. Castronovo, M. Franzini, G. Vidari, E.
Angew. Chem. Int. Ed. 2012, 51, 1187 –1190
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1189

.
Angewandte
Communications
Giannini, Chem. Soc. Rev. 2003, 32, 115; c) T. Tanaka, M.
Hayashi, Synthesis 2008, 3361; d) M. Bartꢀk, Chem. Rev. 2010,
110, 1663.
Commun. 2011, 47, 6716; e) X. Tian, C. Cassani, Y. Liu, A.
Moran, A. Urakawa, P. Galzerano, E. Arceo, P. Melchiorre, J.
Am. Chem. Soc. 2011, 133, 17934.
[2] Selected examples: a) S. Kobayashi, H. Ishitani, J. Am. Chem.
Soc. 1994, 116, 4083; b) M. P. Sibi, J. Chen, J. Am. Chem. Soc.
2001, 123, 9472; c) K. Yabu, S. Masamoto, S. Yamasaki, Y.
Hamashima, M. Kanai, W. Du, D. P. Curran, M. Shibasaki, J.
Am. Chem. Soc. 2001, 123, 9908; d) D. M. Du, S. F. Lu, T. Fang, J.
Xu, J. Org. Chem. 2005, 70, 3712; e) F. Lutz, T. Igarashi, T.
Kawasaki, K. Soai, J. Am. Chem. Soc. 2005, 127, 12206; f) W.
Zeng, G. Y. Chen, Y. G. Zhou, Y. Li, J. Am. Chem. Soc. 2007,
129, 750; g) H. Y. Kim, H. Shih, W. E. Knabe, K. Oh, Angew.
Chem. 2009, 121, 7556; Angew. Chem. Int. Ed. 2009, 48, 7420;
h) S. M. Smith, J. M. Takacs, Org. Lett. 2010, 12, 4612; i) T.
Inagaki, A. Ito, J. Ito, H. Nishiyama, Angew. Chem. 2010, 122,
9574; Angew. Chem. Int. Ed. 2010, 49, 9384.
[7] K. Nakayama, K. Maruoka, J. Am. Chem. Soc. 2008, 130, 17666.
[8] S. A. Moteki, S. Xu, S. Arimitsu, K. Maruoka, J. Am. Chem. Soc.
2010, 132, 17074.
[9] The use of 2,6-(OH)2C6H3CO2H in H2O/MeOH gave: 47%
yield, d.r. (syn/anti) = 5.8:1, À44% ee.
[10] B. T. Hahn, R. Froehlich, K. Harms, F. Glorius, Angew. Chem.
2008, 120, 10134; Angew. Chem. Int. Ed. 2008, 47, 9985.
[11] In the case of 50% water in DMF without an acid additive, the ee
value of syn-4a is identical (94% ee) to that in entry 1 of Table 4.
In contrast, a racemic product was obtained when 20 mol% of
2,6-dinitrobenzoic acid was used along with a solvent system of
50% water in DMF.
[12] No enantiomeric inversion was observed in the direct aldol
reaction between cyclohexanone and p-nitrobenzaldehyde in the
presence of various organic acids and/or water.
[13] In aldol reactions without a water molecule present in the
transition state, B3LYP/6-31G* calculations indicated the favor-
able formation of syn-4a. Furthermore, the calculation involving
two molecules of water in the transition state by did not match
with our observations.
[14] a) W.-P. Li, L. Yi, Y.-N. Jia, X. Ma, J. Org. Chem. 2009, 74, 4812;
b) M. Messerer, H. Wennemers, Synlett 2011, 499; c) Y. Soh-
tome, S. Tanaka, K. Takada, T. Yamaguchi, K. Nagasawa, Angew.
Chem. 2010, 122, 9440; Angew. Chem. Int. Ed. 2010, 49, 9254.
[15] M. J. Frisch, et al. Gaussian03, revisionE.01; Gaussian Inc.:
Wallingford, CT, 2004(see the Supporting Information).
[16] Some of the substrate and the catalyst functionalities have been
omitted to simplify the calculations. The role of the omitted ethyl
ester moiety on the catalyst is to lock the Tf-amide group in the
axial position and the enamine nitrogen in the equatorial
position.
[3] a) P. Mazꢀn, R. Chinchilla, C. Nꢁjera, G. Guillena, R. Kreiter,
R. J. M. Gebbink, G. van Koten, Tetrahedron: Asymmetry 2002,
13, 2181; b) S.-H. Chen, B.-C. Hong, C.-F. Su, S. Sarshar,
ˇ
ˇ ´
Tetrahedron Lett. 2005, 46, 8899; c) Z. Hamersak, T. Ivsic,
Tetrahedron: Asymmetry 2009, 20, 1095; d) N. Abermil, G.
Masson, J. Zhu, Org. Lett. 2009, 11, 4648; e) M.-Q. Hua, H.-F.
Cui, J. Nie, J.-A. Ma, Angew. Chem. 2010, 122, 2832; Angew.
Chem. Int. Ed. 2010, 49, 2772; f) J. Wang, B. L. Feringa, Science
2011, 331, 1429.
[4] Review: D. Enders, A. A. Narine, J. Org. Chem. 2008, 73, 7857.
[5] For supermolecular assembly in asymmetric catalysis, see: D.
Uraguchi, Y. Ueki, T. Ooi, Science 2009, 326, 120.
[6] For reversal of enantioselectivity or diastereoselectivity by
additives in asymmetric reactions, see: a) L. Zhong, J. Xiao, C.
Li, J. Catal. 2006, 243, 442; b) X. J. Li, G. W. Zhang, L. Wang,
M. Q. Hua, J. A. Ma, Synlett 2008, 1255; c) D. G. Blackmond, A.
Moran, M. Hughes, A. Armstrong, J. Am. Chem. Soc. 2010, 132,
7598; d) J.-S. Gao, S.-Y. Bai, Q. Gao, Y. Liu, Q.-H. Yang, Chem.
1190
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1187 –1190