Primary amine catalyzed direct asymmetric aldol reaction assisted by water

Article abstract of DOI:10.1016/j.tetasy.2005.02.031

L-Valine was found to be an active catalyst in the asymmetric direct aldol reaction. The aldol reaction of a variety of aromatic aldehydes with acetone was catalyzed by 20 mol % of l-valine at 35°C with the aldol products obtained in moderate to good yields (48-83%) and enantiomeric excesses (42-72%). The reaction was more efficient catalytically with best results observed in the presence of 1 mol equiv of water, with respect to the aldehyde, in either DMSO or DMF as solvent. The effect of water concentration on the reaction rate and enantioselectivity was also investigated. Thus, with increasing water concentration in DMSO there was decreasing enantioselectivity. However, the reaction in the presence of l-phenylalanine showed a lower level of reactivity and enantioselectivity to afford the aldol in 25% with 31% ee. In marked contrast, reaction with l-phenylglycine resulted in the negligible formation of the aldol (<5%). Our results, suggest a new strategy in the design of new bioorganic catalysts for direct asymmetric aldol reactions.

Full text of DOI:10.1016/j.tetasy.2005.02.031

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 1411–1414
Primary amine catalyzed direct asymmetric aldol reaction
assisted by water
Mohamed Amedjkouh*
Organic Chemistry, Department of Chemistry, G o¨ teborg University, SE-412 96Go¨teborg, Sweden
Received 3 December 2004; accepted 28 February 2005
Abstract—L-Valine was found to be an active catalyst in the asymmetric direct aldol reaction. The aldol reaction of a variety of
aromatic aldehydes with acetone was catalyzed by 20 mol % of L-valine at 35 °C with the aldol products obtained in moderate
to good yields (48–83%) and enantiomeric excesses (42–72%). The reaction was more efficient catalytically with best results observed
in the presence of 1 mol equiv of water, with respect to the aldehyde, in either DMSO or DMF as solvent. The effect of water con-
centration on the reaction rate and enantioselectivity was also investigated. Thus, with increasing water concentration in DMSO
there was decreasing enantioselectivity. However, the reaction in the presence of L-phenylalanine showed a lower level of reactivity
and enantioselectivity to afford the aldol in 25% with 31% ee. In marked contrast, reaction with L-phenylglycine resulted in the
negligible formation of the aldol (<5%). Our results, suggest a new strategy in the design of new bioorganic catalysts for direct
asymmetric aldol reactions.
Ó 2005 Elsevier Ltd. All rights reserved.
1
. Introduction
are also thought to be involved in the origin of homochi-
rality in the synthesis of building blocks such as sugars
during the prebiotic world.
5
The asymmetric aldol reaction is one of the most power-
ful methods for the construction of complex chiral
polyol architectures. Early developments in asymmetric
direct aldol reactions catalyzed by heterobimetallic
complexes were reported by Shibasaki et al. and
In a recent report by Wong et al., a hydrogen-bonded
water molecule was found to participate in a proton
relay in the DERA catalyzed effective tautomerization
of acetaldehyde imine to its enamine, allowing for the
1
Trost et al. Recent finding by Barbas et al. and List
6
et al. proved L-proline to be a powerful catalyst
in the asymmetric intramolecular direct aldol reaction
and paved the way for the development of the concept
of small organic molecules as catalysts. However,
screens of catalysts have so far only yielded closely re-
aldol reaction in a hydrophobic pocket. Computational
studies suggested that hydrogen-bond donors, such as
water itself, are able to provide carbonyl compounds
2
7
with two hydrogen bonds. Based on these observations,
we wondered whether water might participate in a pro-
ton relay that could allow for enamine formation involv-
3
lated five-membered proline derivatives. Conversely, it
8
has long been known that except where the secondary
enamine form is stabilized by further conjugation, the
equilibrium is usually almost completely in favor of
the imine form. Despite the thermodynamic instability,
this possibility of enamine formation could be employed
in reactions with electrophilic reagents in process, which
have always required the isolation of the enamine. In
this context there is a need for catalytic direct aldol reac-
tions using aliphatic amino acids. Aliphatic amino acids
ing primary amines in organic solvents. Since the
imine–enamine equilibrium is strongly solvent depen-
dent, these solvents must have high polarity and propen-
sity for hydrogen bond accepting because of their ability
9
to stabilize transition states.
4
2. Results and discussion
Aliphatic amino acids were initially observed to be inef-
fective in catalyzing the direct aldol reaction, yielding
2
c
less than 10% aldol adduct at room temperature. We
began our investigation by the evaluation of the condi-
tions that would allow the aldol reaction to take place.
*
0
Tel.: +46 31 772 2895; fax: +46 31 772 3840; e-mail: mamou@
chem.gu.se
957-4166/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2005.02.031

1412
M. Amedjkouh / Tetrahedron: Asymmetry 16(2005) 1411–1414
No reaction was observed when reacting 4-nitrobenzal-
dehyde with acetone in the presence of 20 mol % of L-va-
line in pure DMSO. In contrast, when the reaction
mixture was stirred at 35 °C for 24 h, aldol product
was obtained in 30% yield. Also L-valine appeared to
catalyze the aldol reaction at higher temperatures, which
has not been reported previously.
Table 1. Amino acid-catalyzed aldol reaction of acetone and 4-
nitrobenzaldehyde
Amino acid
X
Time (h) Yields (%) Ee aldol (%)
L-Valine
L-Phenylalanine
NO
NO
2
2
2
24
20
20
58
25
<5
35
36
25
37
53
31
—
40
18
12
42
(
D)-Phenylglycine NO
L-Aspartic acid
L-Glutamic acid
L-Histidine
L-Threonine
NO2 24
NO
NO
NO
2
2
2
24
24
24
However, when the reaction was performed in DMSO
containing 1 vol % water and in the presence of
20 mol % L-valine, an acceleration of the reaction was
observed with the aldol product obtained in 58% yield
1
0
and 53% ee after 24 h (Scheme 1 and Table 1).
the reaction with L-phenylglycine resulted in the negligi-
ble formation of the aldol (<5% after 20 h).
Comparison of acidic aspartic and glutamic amino acids
with L-valine resulted both in low reactivity and enantio-
selectivity. An even lower reactivity was observed with
the basic L-histidine. Interestingly, neutral L-threonine
gave comparable results with L-aspartic acid, probably
resulting from hydrogen bonds with the acidic proton
on the side chain. These preliminary results show that
valine is the most suitable primary amine containing
catalyst for the aldol reaction.
O
O
OH
Amino acid 20 mol%
Acetone 33 vol%
H
DMSO-H O (99:1)
2
o
X
35 C
X
1
2
Scheme 1.
With these results in hand, we next tried to evaluate the
reactivity of different aliphatic amino acids under the
same reaction conditions and at the same concentration
of catalyst with respect to substrate. However, the reac-
tion in the presence of L-phenylalanine showed a lower
level of reactivity and enantioselectivity to afford the
aldol in 25% after 20 h with 31% ee. In marked contrast,
The generality of L-valine in catalyzing the direct aldol
reaction was evaluated with a variety of different aro-
matic aldehydes under optimal conditions. The aldol
reaction of aldehydes 1 with acetone was catalyzed by
20 mol % of L-valine at 35 °C with aldol products 2
obtained in good yields (48–83%) and enantiomeric
excesses (42–72%) as shown in Table 2. In general,
Table 2. Direct asymmetric aldol reactions catalyzed by L-valine
O
O
OH
O
L-Valine 20 mol%
Acetone 33 vol%
+
H
DMSO-H O (99:1)
2
o
X
35 C
X
X
1
(R)-2
3
1
1
1
1
1
a, X = H
b, X = Cl
c, X = Br
d, X = NO2
e, X = CN
a
Yields (%)
b
Ee 2 (%)
X
Entry
Solvent–H O (99:1)
2
Time (h)
Aldol 2
Cond. 3
c
H
1
2
3
4
5
6
7
8
9
DMSO
DMF
48
50
51
79
52
68
58
87
82
83
47
42
41
19
41
30
—
—
—
—
46
72
46
63
48
63
24
63
40
63
62
72
c
Cl
DMSO
DMF
48
56
49
62
53
42
62
60
Br
DMSO
DMF
NO
CN
2
DMSO
DMF
DMSO
DMF
10
a
b
c
1
The yields were determined by H NMR spectroscopy.
The ee was determined by chiral HPLC analysis using a Kromasil TBB-CHI (see Ref. 12).
The ee was determined by chiral GC analysis using a Chrompack Chirasil-CB Dex. The major enantiomer was assigned to be (R) by cross
comparison of racemic aldol products with those reported in the literature using L-proline (Ref. 2c).

M. Amedjkouh / Tetrahedron: Asymmetry 16(2005) 1411–1414
1413
reactions using benzaldehyde, 4-chloro- and 4-bromo-
benzaldehydes 1a–c afforded aldol products together
with considerable amounts of dehydration product 3.
Improved enantioselectivities were observed when reac-
Table 4. Effect of reaction solution concentration on aldol reaction
a
Entry
Solution volume
(mL)
H
2
O
(mmol)
Yield
(%)
Ee
(%)
1
2
3
4
1
2
3
4
0.55
1.1
1.6
2.2
78
58
58
54
42
53
50
48
tions in DMF–H O were performed (99:1) as solvent
2
under similar reaction conditions although appreciable
condensation did still occur. Unfortunately in this case,
the rate of the reaction was much slower than in
a
2
Reactions were run in DMSO–H O (99:1). Acetone (1.0 mL),
DMSO–H O (99:1). Interestingly, 4-nitro- and 4-cyano-
2
4-nitrobenzaldehyde (1.0 mmol), and L-valine (0.2 mmol).
benzaldehydes 1d and 1e reacted smoothly in presence
of valine in DMSO–H O (99:1) to yield aldols 2d and
2
2e as the sole reaction products (Scheme 2). A similar
reactivity was observed in DMF–H O (99:1) although
at a slower rate and lower enantioselectivity.
tively, show clearly the importance of water concentra-
tion. Remarkably, the addition of either more or sub-
stoichiometric amounts of water both resulted in lower
ees, while maintaining good reaction rates.
2
O
L-Valine 20 mol%
Acetone 1 ml
O
OH
From the findings above, it is clear that water plays a
critical role in the reaction. Thus, preliminary NMR
investigations were performed for the purpose of exam-
ining the water effect and for a better understanding of
the efficiency of valine. It is well known that amino acids
can condense with aliphatic and aromatic aldehydes to
form oxazolidinones and other compounds, including
products of self-aldolization. In particular aromatic
aldehydes react with acyclic amino acids in DMSO to
H
DMSO-H O
2
o
3
5 C, 22h
O N
NO2
2
1
d
(R)-2d
Scheme 2.
1
1
produce the corresponding imine.
The effect of water concentration on the reaction rate
and enantioselectivity of the aldol reaction was also
investigated (Table 3). This showed that by increasing
water concentration in DMSO, enantioselectivity de-
creased. Such an observation has previously been
1
In a H NMR study of valine catalyzed aldol reaction of
acetone with 4-nitrobenzaldehyde in DMSO-d –H O
2
6
(99:1), under standard conditions, we observed the coex-
istence of an aldehyde along with a compound display-
2
c
reported for proline catalyzed aldol reactions.
ing a single resonance at d 8.49, and two doublets at d
1
.31 and 8.03 ppm. The latter H NMR peaks were
8
attributed to an imine resulting from the condensation
1
1
Table 3. Effect of water concentration on aldol reaction
of valine with aldehyde as already suggested. Similarly
and in the absence of acetone, the H NMR experiment
1
a
Entry
Amount of H
2
O
Yield (%)
Ee (%)
in DMSO-d –H O (99:1) showed that imine 4 was easily
6
2
vol %
mmol
formed at rt, as outlined in Scheme 3. However,
although oxazolidinone is the major adduct in the case
of proline, we were unable to observe any formation
1
2
3
1
2
4
1.1
2.2
4.4
58
60
60
53
44
42
11
of oxazolidinone 5.
a
Reactions were run using 2 mL solvent. Acetone (1.0 mL), 4-nitro-
benzaldehyde (1.0 mmol), and L-valine (0.2 mmol).
The formation of the imine could be regarded as a par-
asitic equilibrium, which results in the poisoning of the
catalyst thus causing lower reaction rates (Scheme 3).
At the same time, water effects shift this equilibrium to
the formation of the aldehyde and regeneration of the
amino acids that still allow for catalysis. This might also
explain why phenylalanine was still producing aldol
adduct besides the imine, although in a much lower rate
when compared to valine. Conversely, the lower enan-
tioselectivity of phenylalanine may be attributed to
More intriguing is the fact that when performing the
aldol reaction in the presence of water at different
concentrations, it appears that 1.1 mmol, that is,
1
mol equiv of water with respect to aldehyde gave the
best ee (Table 3, entry 1). Similar results were observed
when varying the concentration of reagents (Table 4).
Comparison of entries 1 and 2 in Tables 3 and 4, respec-
iPr
HN
O
iPr
O
O
L-Valine
O
R
H
N
OH
DMSO-H O (99:1)
2
o
Ar
Ar
3
5 C
1
4
5
Scheme 3.

1414
M. Amedjkouh / Tetrahedron: Asymmetry 16(2005) 1411–1414
racemization during imine formation as previously
(g) List, B. Synlett 2001, 1675; (h) List, B. Tetrahedron
002, 58, 5573.
. For proline based catalyst see: (a) Saito, S.; Nakadai, M.;
Yamamoto, H. Synlett 2001, 1245; (b) Nakadai, M.; Saito,
S.; Yamamoto, H. Tetrahedron 2002, 58, 8167; (c) Kofoed,
J.; Nielsen, J.; Reymond, J. L. Bioorg. Med. Chem. Lett.
1
1a
2
reported. Although not observed, the carbinolamine,
enamine, and hydrate could also be formed at lower
concentrations.
3
Furthermore, participation of water at low concentr-
ation, that is, hydrophobic conditions, can be character-
ized in providing a proton shuffle through a hydrogen
bond network that might enable imine–enamine tauto-
merization and aldehyde activation. This situation is
facilitated by polar and hydrogen bond acceptors
DMSO or DMF.
2
003, 13, 2445–2447; (d) Tang, Z.; Jiang, F.; Yu, L.-T.;
Cui, X.; Gong, L.-Z.; Mi, A.-Q.; Jiang, Y.-Z.; Wu, Y.-D.
J. Am. Chem. Soc. 2003, 125, 5262–5263; (e) Tang, Z.;
Jiang, F.; Cui, X.; Gong, L.-Z.; Mi, A.-Q.; Jiang, Y.-Z.;
Wu, Y.-D. PNAS 2004, 101, 5755–5760; (f) Cobb, A. J.
A.; Shawn, D. M.; Ley, S. V. Synlett 2004, 558–560; (g)
Berkessel, A.; Koch, B.; Lex, J. Adv. Synth. Catal. 2004,
3
46, 1141–1146; (h) Cheong, P. H.-Y.; Houk, K. N.;
Warrier, J. S.; Hanessian, S. Adv. Synth. Catal. 2004, 346,
111–1115.
1
3. Conclusion
4
. For reviews see: (a) dÕAngelo, J.; Desma e¨ le, D.; Dumas,
F.; Guingant, A. Tetrahedron: Asymmetry 1992, 459–505;
In summary, we have demonstrated that water partici-
pates in the efficient catalysis of aldol reactions using
primary amino groups of amino acids. This suggests a
new strategy in the design of new bioorganic catalysts
for direct asymmetric aldol reactions. The results herein
suggest that water plays a more intricate role. Further
investigations of the solution structures in order to
identify the factors that facilitate the imine–enamine
tautomerization of primary amines, are currently in
progress in our laboratory.
(
Hickmott, P. W. Tetrahedron 1982, 38, 3363–3446.
5. Pizzarioell, S.; Weber, A. L. Science 2004, 303, 1151.
6. Heine, A.; DeSantis, G.; Luz, J. G.; Mitchell, M.; Wong,
C.-H.; Wilson, I. A. Science 2001, 294, 369–374.
b) Hickmott, P. W. Tetrahedron 1982, 38, 1975–2050; (c)
7
. Severance, D. L.; Jorgensen, W. L. J. Am. Chem. Soc.
992, 114, 10966–10968.
1
8
. During the course of the present work two new reports
using water have been published, see: (a) Tanaka, F.;
Thayumanavan, R.; Mase, N.; Barbas, C. F., III. Tetra-
hedron Lett. 2004, 45, 325–328; (b) Torii, H.; Nakadai, M.;
Ishihara, K.; Saito, S.; Yamamoto, H. Angew. Chem., Int.
Ed. 2004, 43, 1983–1986.
Acknowledgements
9. (a) Reichardt, C. Solvents and Solvent Effects in Organic
Chemistry, 3rd ed.; Wiley-VCH: Weinheim, 2003; (b)
Clark, R. A.; Parker, D. C. J. Am. Chem. Soc. 1971, 93,
M.A. is grateful to Prof. Per Ahlberg for support and
encouragement. The author thanks Prof. Stig Allenmark
for help with HPLC chromatography.
7
ation, the amount of enamine being 4.7% (CD OD),
257–7261; (c) Polar solvents were found to favor enamiz-
3
1
4.6% (CD
see: Quast, H.; Heublein, A. Chem. Ber. 1975, 108, 2574–
579.
10. Typical procedure for aldol reaction: Amino acid
0.2 mmol) and aldehyde 1 (1.0 mmol) were added to a
2 2 6
Cl ) 33% (dioxane), and 37.8% (DMSO-d );
2
References
(
1
. (a) Yamada, Y. M. A.; Yoshikawa, N.; Sasai, H.;
Shibasaki, M. Angew. Chem. 1997, 109, 1942; Angew.
Chem., Int. Ed. 1997, 36, 1871–1873; (b) Yoshikawa, N.;
Yamada, Y. M. A.; Das, J.; Sasai, H.; Shibasaki, M.
J. Am. Chem. Soc. 1999, 121, 4168–4178; (c) Trost, B. M.;
Ito, H. J. Am. Chem. Soc. 2000, 122, 12003–12004; (d)
Trost, B. M.; Ito, H.; Silcoff, E. R. J. Am. Chem. Soc.
solution of acetone (1 mL) and solvent (2 mL, with
1 vol % H O in stock solution). The non-homogenous
mixture was stirred at 35 °C for the given time. Samples
(100 lL) were withdrawn, quenched with saturated aque-
2
4
ous NH Cl (200 lL) and then extracted with EtOAc
(1 mL). Conversions were determined using both NMR
and HPLC. Enantiomeric excess was determined by chiral
HPLC (Kromasil TBB-CHI) and comparison with pure
racemic aldol adduct.
2001, 123, 3367–3368.
2
. (a) List, B.; Lerner, R. A.; Barbas, C. F., III. J. Am. Chem.
Soc. 2000, 122, 2396; (b) Notz, W.; List, B. J. Am. Chem.
Soc. 2000, 122, 7386; (c) Sakthivel, K.; Notz, W.; Bui, T.;
Barbas, C. F., III. J. Am. Chem. Soc. 2001, 123, 5260; (d)
List, B.; Pojarliev, P.; Castello, C. Org. Lett. 2001, 3, 573;
11. (a) Orsini, F.; Pelizzoni, F.; Forte, M.; Sisti, M.; Bombieri,
G.; Benetello, F. J. Heterocycl. Chem. 1989, 26, 837–841;
(b) Polonski, T. Tetrahedron 1985, 41, 603–609; (c) List,
B.; Hoang, L.; Martin, H. J. Proc. Natl. Acad. Sci. U.S.A.
2004, 101, 5839–5842.
(
e) C o´ rdova, A.; Notz, W.; Barbas, C. F., III. J. Org.
Chem. 2002, 67, 301; (f) C o´ rdova, A.; Notz, W.; Barbas,
C. F., III. Chem. Commun. 2002, 3204; For reviews, see:
12. Allenmark, S. G.; Andersson, S.; M o¨ ller, P.; Sanchez, D.
Chirality 1995, 7, 248–256.