Protonated N′-benzyl-N′-prolyl proline hydrazide as highly enantioselective catalyst for direct asymmetric aldol reaction

Article abstract of DOI:10.1039/b511992h

Protonated N′-benzyl-N′-l-prolyl-l-proline hydrazide has been developed as a highly enantioselective catalyst for the asymmetric direct aldol reaction of aromatic aldehydes with ketones. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b511992h

View Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Protonated N9-benzyl-N9-prolyl proline hydrazide as highly
enantioselective catalyst for direct asymmetric aldol reaction{
Chuanling Cheng,ade Jian Sun,*a Chao Wang,ae Yu Zhang,ae Siyu Wei,a Fan Jiangb and Yundong Wu*bc
Received (in Cambridge, UK) 24th August 2005, Accepted 18th October 2005
First published as an Advance Article on the web 18th November 2005
DOI: 10.1039/b511992h
Table 1 Screening L-proline hydrazides as the catalysts in the direct
aldol reaction of p-nitrobenzaldehyde with acetonea
Protonated N9-benzyl-N9-L-prolyl-L-proline hydrazide has been
developed as a highly enantioselective catalyst for the asym-
metric direct aldol reaction of aromatic aldehydes with ketones.
Proline catalysis in the asymmetric direct intermolecular aldol
reactions1,2 has recently received extraordinary attention due to its
fundamental importance to the evolution of so-called organoca-
talysis.3 The acidic proton of the carboxylic acid group of proline
was shown to be critical for both the reactivity and stereoselecti-
vity.1b However, such an acidic proton has been proven to be not
indispensable by the finding that chiral aminoalcohol derived
proline amides catalyze the same aldol reactions with high
efficiency and excellent enantioselectivity.4 A new window has
thus been opened for developing new proline-based organocata-
lysts through modifying its carboxylic acid function. For such a
kind of proline amide catalysts,5 the hydrogen-bonding in the
transition state involving the amide NH and the terminal hydroxyl
group was demonstrated to be the key to their catalytic
effectiveness in the aldol reactions. Along the same principle, we
envisioned that properly substituted proline hydrazides could
potentially also be developed into effective organocatalysts for the
same aldol reactions. The additional nitrogen atom in the
hydrazides should provide the catalysts with an excellent open
site to anchor hydrogen-bonding elements. Here we describe the
first example of chiral proline hydrazides as highly enantioselective
catalysts for the direct intermolecular aldol reactions.
Entry
Catalyst
Time (h)
Yield (%)b
ee (%)c
1
2
1
2
3
3
4
5
6
7
7
24
24
24
24
24
24
24
24
24
24
24
24
24
24
8
16
,5
,10
,10
29
28
32
,10
41
54
n.d.d
54
3
4e
5
20
48
42
59
17
77
n.d.
n.d.
n.d.
n.d.
51
83
70
90
58
6
7
8
9e
10
11
12
13e
14
15e
16
17e
18
a
8
9
,10
,10
,10
,10
67f
10
11
12
12
13
13
14
65f
10
30
24
24
24
,10
Unless specified otherwise, the concentration of aldehyde is 0.1 M
b
c
in neat acetone. Isolated yield based on the aldehyde. The ee
values were determined by HPLC and the configuration was
assigned as R by comparison of retention time. Not detected.
d
e
f
20 mol% TFA was added. Conversion is more than 95%.
We first screened a series of easily prepared N9-substituted
L-proline hydazides (1–14,Fig. 1) as the catalysts (20 mol%) in the
model reaction of p-nitrobenzaldehyde with neat acetone. As
shown in Table 1, the reactions with most of these catalysts were
found to be sluggish, affording the product in low to moderate ee
values. Interestingly, the catalyst with an N9-L-prolyl group that
bears a free cyclic secondary amine (12 in Fig. 1) stood out with
much higher efficiency than the others (entry 14, Table 1). The
conversion reached over 95% after 24 hours at room temperature.
The reaction time was further shortened to 8 hours if one
equivalent (based on the catalyst) of trifluoroacetic acid (TFA) was
added (entry 15). Due to some side reactions,6 only around 65% of
isolated yield was obtained in both cases. On the other hand,
catalyst 12 exhibited only moderate enantioselectivity (51%) in the
absence of Brønsted acid, which was, however, significantly lifted
to 83% in the presence of TFA (entry 15 vs 14). It should be noted
that capping the secondary amine of the N9-prolyl group (see
catalysts 13 and 14 in Fig. 1) was found to be detrimental to the
reactivity of the catalyst (entries 16–18) even though the
enantioselectivity was slightly enhanced (entries 16 and 17 vs 14
aChengdu Institute of Biology, Chinese Academy of Sciences,
Chengdu 610041, China. E-mail: sunjian@cib.ac.cn;
Fax: +86-28-85222753; Tel: +86-28-85211220
bCollege of Chemistry, Peking University, Beijing 100871, China
cDepartment of chemistry, The Hong Kong University of Science and
Technology, Kowloon, Hong Kong, China
dChengdu Institute of Organic Chemistry, Chinese Academy of Sciences,
Chengdu, 610041
eGraduate School of Chinese Academy of Sciences, Beijing, China
{ Electronic supplementary information (ESI) available: experimental
data. See DOI: 10.1039/b511992h
Fig. 1 L-Proline hydrazides evaluated in this study.
Chem. Commun., 2006, 215–217 | 215
This journal is ß The Royal Society of Chemistry 2006

View Online
+
…
Table 2 Optimizing the reaction conditions with 15a
and 15, respectively). Apparently, the existence of the [HN H]
group of the protonated N9-prolyl group of catalyst 12 is crucial to
both the reactivity and enantioselectivity. To further optimize the
structure of the catalyst, we thus fixed this protonated N9-prolyl
substituent and turned our attention to the CO–NH–NH–CO unit
of the hydrazide. Gratifyingly, installation of a benzyl group on the
N9-position of 12 led to an excellent catalyst (15 in Fig. 2), of which
the catalysis in the model aldol reaction afforded excellent
reactivity, enantioselectivity and isolated yield (entry 13 in Table 2).
Catalyst15wasfirstexaminedforthemodelreactioninacetone
at room temperature in the presence of an equal molar amount of
TFA. The reaction was found to complete in 3 hours, affording
92% ee (entry 2). Again, due to the side reactions, only 54%
isolated yield was obtained. As expected, both the reactivity and
enantioselectivity of the catalyst sharply dropped in the absence
of TFA (entry 1). Thus pre-mixed equal molar amounts of 15 and
TFA (15.TFA) were used for further optimization of the reaction
conditions.7
Entry
Solvent
Temp. (uC) Time (h) Yield (%)b ee (%)c
1d
2
3
4
5
6
7
8
Acetone
Acetone
DMF
DMSO
DCM
DCE
CH3CN
Et2O
THF
THF
THF
20
20
20
20
20
20
20
20
20
0
220
20
0
220
0
12
3
24
24
11
3
11
3
3
5
72
3
7
40
70
54
54
,10
,10
69
73
49
66
70
86
88
84
95
97
95
7
92
n.de
n.d
90
89
91
94
94
97
98
90
96
97
96
9
10
11
12
13
14
15f
a
PhCH3
PhCH3
PhCH3
PhCH3
Unless specified otherwise, 20 mol% 15.TFA was used as the
catalyst, the concentration of aldehyde is 0.1 M, and v/v of acetone/
solvent is 1/4. Isolated yield based on the aldehyde. The ee values
b
c
were determined by HPLC and the configuration was assigned as R
by comparison of retention time. No TFA was added. Not
d
e
A survey of solvents with 20 mol% catalyst 15.TFA revealed
that most of the low boiling-point solvents with relatively low
polarity are friendly to this catalytic system (Table 2). Consistently
high ee values were obtained with all of the solvents listed in the
table except DMF and DMSO, in either of which the reaction was
extremely sluggish (entries 3 and 4). While the highest ee of 94%
was obtained in either Et2O or THF at room temperature (entries
8 and 9), the yield in the latter was slightly better than in the
former. Up to 98% ee was achieved at 220 uC in THF with 88%
isolated yield. Toluene turned out to be the best solvent to suppress
the side reactions and in the meantime maintain high reactivity and
enantioselectivity (entries 12–15). With this solvent, lowering the
temperature from room temperature to 0 uC increased the ee of the
product from 90% to 96% and more importantly, completely
suppressed the side reactions, affording 95% isolated yield (entry
13). Lowering the temperature further to 220 uC had little effect
on the ee whereas the reaction rate was strikingly decreased (entry
14). Under the optimal reaction condition, namely 20 mol%
catalyst 15.TFA, at 0 uC and with toluene as solvent, the model
reaction was completed in 7 hours, affording 95% yield and 96% ee.
Decreasing the loading of the catalyst to 5% merely slowed down
the reaction to some extent but didn’t sacrifice either the yield or
the enantioselectivity (entry 15).
f
detected. 5 mol% catalyst was used.
hydrazide catalyst seems to be substrate-specific, and its applica-
tion scope is limited to relatively electron-deficient aromatic
aldehydes.
We also investigated some other ketone substrates (Scheme 1).
With the catalysis of 15.TFA, cyclohexanone gave a dr of 92 : 8 for
the anti/syn products; with an ee of 92% for the major anti isomer.
Cyclopentanone yielded a dr of 25 : 75 for the anti/syn products;
with ee value of 67% and 74%, respectively.
Theoretical calculations have been performed to understand the
high enantioselectivity with catalyst 15.TFA.8 As shown in Fig. 3,
in the enamine intermediate, the optimal CO–N–N–CO dihedral
angle was found to be nearly perpendicular. This is consistent with
the observations from early computational studies on diacylhy-
drazines9 and from our X-ray analyses of the single crystal of 12.10
When the aldehyde was added, a simple rotation of the protonated
pyrrolidine (up left) could allow its NH to form a strong hydrogen
bond with the hydrazide NH and the aldehyde (see TS1 and TS2).
The phenyl of the N-benzylic group in the enamine is reaching out
towards the reaction center to avoid the steric interaction with the
two carbonyl groups in the back. In TS1, which leads to the
Other aldehydes were also examined to expand the substrate
scope of catalyst 15.TFA under the optimal condition. As shown
in Table 3, excellent enantioselectivities and yields were consistently
obtained for the aromatic aldehydes with an electron-withdrawing
group (entries 1–5). However, for the aromatic aldehydes with an
electron-donating group or without any substitution, the reactions
were much slower and less than 20% of isolated yields were
obtained after 72 hours under the optimal conditions (entry 6 and
7). Interestingly, the ee values of the products remained at high
levels. On the other hand, this catalytic system was proven to be
totally ineffective for aliphatic aldehydes. Thus the present
Table 3 Aldol reactions catalyzed by 15.TFAa
Entry Product
R
Time (h) Yield (%)b
ee (%)c
1
2
3
4
5
6
7
a
16a
16b
16c
16d
16e
16f
16g
4-NO2Ph
2-NO2Ph
3-NO2Ph
4-CNPh
2-ClPh
7
5
5
21
24
72
72
95
97
86
97
75
15
17
96
96
96
92
96
87
90
4-MePh
Ph
The concentration of aldehyde is 0.1 M, and v/v of acetone/toluene
b
c
is 1/4. Isolated yield based on the aldehyde. The ee values were
determined by HPLC and the configuration was assigned as R by
comparison of retention time.
Fig. 2 Structure of the optimised catalyst.
216 | Chem. Commun., 2006, 215–217
This journal is ß The Royal Society of Chemistry 2006

View Online
study revealed that both the CO–N–N–CO group and protonated
prolyl group are important for the reactivity of the catalyst and the
benzyl substituent has certain contribution to the stereoselectivity.
We are grateful for financial supports from Natural Science
Foundation of China (Project 20402014, 20225312) and from the
Chinese Academy of Sciences (Hundreds of Talents Program).
Scheme 1 Aldol reactions of 4-nitrobenzaldehyde with some ketones
catalyzed by 15.TFA.
Notes and references
1 For the pioneering findings, see: (a) B. List, R. A. Lerner and
C. F. Barbas, III, J. Am. Chem. Soc., 2000, 122, 2395; (b) K. Sakthivel,
W. Notz, T. Bui and C. F. Barbas, III, J. Am. Chem. Soc., 2001, 123,
5260; (c) W. Notz and B. List, J. Am. Chem. Soc., 2000, 122, 7386.
2 For leading references, see: (a) A. B. Northrup and D. W. C. MacMillan,
Science, 2004, 305, 1752; (b) A. Co´rdova, W. Notz and C. F. Barbas, III,
J. Org. Chem., 2002, 67, 301; (c) A. Bøgevig, N. Kumaragurubaran and
K.A. Jørgensen,Chem.Commun.,2002,620;(d)A.Co´rdova,M. Engqvist,
I. Ibrahem, J. Casas and H. Sunde´n, Chem. Commun., 2005, 2047; (e)
D. Ender and C. Grondal, Angew. Chem., Int. Ed., 2005, 44, 2.
3 For recent reviews, see: (a) B. List, Tetrahedron, 2002, 58, 5573; (b)
B. Alcaide and P. Almendros, Angew. Chem., Int. Ed., 2003, 42, 858; (c)
B. List, Acc. Chem. Res., 2004, 37, 548; (d) W. Notz, F. Tanaka and
C. F. Barbas, III, Acc. Chem. Res., 2004, 37, 580; (e) U. Kazmaier,
Angew. Chem., Int. Ed., 2005, 44, 2186.
4 (a) Z. Tang, F. Jiang, L. T. Yu, X. Cui, L. Z. Gong, A. Q. Mi,
Y. Z. Jiang and Y.-D. Wu, J. Am. Chem. Soc., 2003, 125, 5262; (b)
Z. Tang, F. Jiang, X. Cui, L. Z. Gong, A.-Q. Mi, Y. Z. Jiang and
Y.-D. Wu, Proc. Natl. Acad. Sci. USA, 2004, 101, 5755.
5 For the other types of proline-derived organocatalysts effective for the
asymmetric direct aldol reactions, see: (a) H. Torii, M. Nakadai,
K. Ishihara, S. Saito and H. Yamamoto, Angew. Chem., Int. Ed., 2004,
43, 1983; (b) S. Saito and H. Yamamoto, Acc. Chem. Res., 2004, 37,
570; (c) A. Hartikaa and P. Arvidsson, Tetrahedron: Asymmetry, 2004,
15, 1831; (d) H. J. Martin and B. List, Synlett, 2003, 12, 1901; (e)
J. Kofoed and J. L. Reymond, Bioorg. Med. Chem. Lett., 2003, 13,
2445; (f) N. Halland, A. Braunton, S. Bachmann, M. Marigo and
K. A. Jørgensen, J. Am. Chem. Soc., 2004, 126, 4790; (g) E. Lacoste,
Y. Landais, K. Schenk, J. B. Verlhac and J. M. Vincent, Tetrahedron
Lett., 2004, 45, 8035; (h) A. J. A. Cobb, D. M. Shaw, D. A. Longbottom,
J. B. Gold and S. V. Ley, Org. Biomol. Chem., 2005, 3, 84; (i)
P. Krattiger, R. Kovasy, J. D. Revell, S. Ivan and H. Wennemers, Org.
Lett., 2005, 7, 1101; (j) A. Berkessel, B. Koch and J. Lex, Adv. Synth.
Catal., 2004, 346, 1141.
6 Side reactions, especially dehydration of the product, is a common
problem encountered with proline and most of its derivatives catalyzed
intermolecular aldol reactions.
7 Addition of 0.5 equivalent of TFA significantly decreased both the
reactivity and enantioselectivity. Addition of 1.5 equivalent of TFA also
significantlydecreasedthereactivity,butdidn’taffecttheenantioselectivity.
8 All calculations were performed with the Gaussian03 program.
GAUSSIAN 03 (Revision C.02), Gaussian, Inc., Wallingford, CT,
2004. See ESI for full reference{.
Fig. 3 The optimized structures of the enamine intermediate (in two
views) and the transition states. The geometries were optimized with
the HF/6-31G* method. The relative energies (kcal mol21) are from the
B3LYP/6-311+G** method and the IEFPCM13 solvation model. The
activation enthalpies are in ( ) and the activation Gibbs free energies in [ ].
formation of the major product observed experimentally, this
phenyl group is far away from the phenyl group of the aldehyde.
However, in TS2 this phenyl group has to turn back to make space
for the phenyl group of benzaldehyde substrate, causing the
structure to be less stable than TS1 by about 3.3 kcal mol21. When
the N-benzyl group is replaced by a hydrogen to model the
situation of catalyst 12, a much smaller preference of TS1 over
TS2 (DDG = 1.2 kcal mol21) is predicted by our calculations (see
Supporting Information{) because of the absence of the above
mentioned steric interaction. It should be noted that the secondary
amine of the prolyl on the right-hand side of 15 should also be able
to form enamine with acetone.11 However, the catalytic effect of 15
in the aldol reactions was unambiguously proven not to result
from the enamine formation with this prolyl but with the one on
the left-hand side by the fact that N-benzylation of the former only
slightly affects the reactivity and enantioselectivity12 whereas
N-benzylation of the latter totally deactivates the catalyst.
9 C. H. Reynolds and R. E. Hormann, J. Am. Chem. Soc., 1996, 118, 9395.
10 Crystal data for 12: C10H18N4O2 (286 K). M = 226.28, orthorhombic,
˚
˚
˚
space group P212121, a = 9.705 (1) A, b = 7.660 (1) A, c = 7.818 (1) A,
a = b = c = 90u, V = 581.23 (17) A , Z = 2, rcalc = 1.293 mg m23
,
3
˚
absorption coefficient = 0.093 mm21, F (000) = 244, total reflections
collected 939, unique 823 (Rint = 0.0095), final R indices [I > 2s(1)] were
R1 = 0.0351, wR2 = 0.0751, GOF = 0.963. CCDC 281902. For
crystallographic data in CIF or other electronic format see DOI:
10.1039/b511992h.
11 Both prolyl units of 15 were found to be iso-propylated when 15.TFA
was treated with excess acetone and NaBH(OAc)3.
12 In the model reaction of p-nitrobenzaldehyde and acetone with the
corresponding Catalyst.TFA, 79% of yield and 93% of ee were obtained
under the same optimal reaction conditions.
In conclusion, we have developed proline hydrazides as highly
enantioselective catalysts for the asymmetric direct intermolecular
aldol reactions. Catalyst 15.TFA catalyzed the reactions of
aromatic aldehydes and acetone with 87–96% ee. A theoretical
13 E. Cance`s, B. Mennucci and J. Tomasi, J. Chem. Phys., 1997, 107, 3032.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 215–217 | 217