The role of Zn2+ in enhancing the rate and stereoselectivity of the aldol reactions catalyzed by the simple prolinamide model

Article abstract of DOI:10.1016/j.tet.2011.07.013

The aldol reaction between acetone and 4-nitrobenzaldehyde catalyzed by single l-prolinamide and its zinc complexes has been studied. An increase in the rate and the stereoselectivity of the reaction has been shown by using zinc derivatives. A mechanistic

Full text of DOI:10.1016/j.tet.2011.07.013

Tetrahedron 67 (2011) 7050e7056
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
The role of Zn2þ in enhancing the rate and stereoselectivity of the aldol reactions
catalyzed by the simple prolinamide model
*
Cecilia Andreu , Gregorio Asensio
ꢀ
ꢀ
ꢀ
ꢁ
ꢁ
ꢀ
Departament de Química Organica, Facultat de Farmacia, Universitat de Valencia, Vicent Andres Estelles s/n. 46100 Burjassot, Valencia, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The aldol reaction between acetone and 4-nitrobenzaldehyde catalyzed by single L-prolinamide and its
Received 15 April 2011
Received in revised form 27 June 2011
Accepted 6 July 2011
zinc complexes has been studied. An increase in the rate and the stereoselectivity of the reaction has
been shown by using zinc derivatives. A mechanistic proposal, based on NMR and ESI studies, has been
put forward to explain the experimental data: zinceprolinamide complexes catalyze the reaction fol-
lowing the general mechanism of stereoselective enamine nucleophilic addition to the acceptor alde-
hyde. Zn2þ prevents the nonspecific base-catalyzed reaction by diminishing the basicity of the amine
nitrogen of prolinamide.
Available online 14 July 2011
Keywords:
Aldol reaction
Lewis acid
Ó 2011 Elsevier Ltd. All rights reserved.
Stereoselectivity
Zinceprolinamide
1. Introduction
Given the difficulty in using enzymes in a large-scale synthesis,
it is desirable to develop simple chemical systems that mimic the
The stereoselective aldol reaction is considered as one of the
most important carbonecarbon bond-forming reactions in organic
action of enzymes by performing highly efficient reactions in water.
In this way, -proline and its derivatives have been employed in
L
synthesis, and is a way to obtain chiral
b
-hydroxy carbonyl
recent years in a wide range of asymmetric organic trans-
formations, especially aldol addition.4 These kinds of catalysts are
known to operate via an enamine mechanism (class I aldolase
mimics).5 However in many cases, high enantioselectivities have
been achieved only in organic solvents, while the presence of water
lowers enantioselectivity.6 Few examples of efficient organo-
catalytic asymmetric aldol reactions in water have been described.7
Most of the proline derivatives used to date are prolinamides.
Some general facts have made these derivatives useful: their syn-
thesis is easy, the presence of a robust amide linkage provides very
stable compounds, and the acidity of the hydrogen of an NH moiety
is enough to activate electrophiles by hydrogen bonding.8 Proli-
namide itself can catalyze aldol condensation though with little
enantioselectivity.9 But the simplicity of this compound makes it
a good model to be used in the optimization of reaction conditions,
that might be improved using more sophisticated derivatives.
Therefore, it was used in our experiments.
Some years ago, Darbre et al. described how the Zn(Pro)2 com-
plex, as well as Zn(Arg)2 and Zn(Lys)2, are able to catalyze the in-
termolecular reaction between aldehydes and ketones in water
with moderate enantioselectivity. These systems are also active in
the self-condensation of glycoaldehyde to form carbohydrates.10
These authors found that the Zn(Pro)2 complex, prepared under
basic conditions, was more effective than proline itself in an
aqueous medium, in terms of both yield and stereoselectivity.
compounds.1
From the green chemistry viewpoint, it is desirable to perform
reactions with a minimum environmental impact, using water as
a reaction solvent and safety catalyst.2 Unfortunately this is not
easy, although nature itself provides a model to be imitated: en-
zymes that catalyze reactions in water under mild conditions with
high efficiency and stereoselectivity. Thus, aldolases are nature’s
catalysts of aldol condensation. According to the mechanism, two
different types of aldolases have been identified and classified. The
first (class I) contains a Lysine residue in the active site, which is
involved in nucleophilic enamine intermediate formation. This
activated donor adds stereoselectively to the aldehyde acceptor.
The second aldolases type (class II) contains a Zn(II) cofactor at the
active site, which is coordinated to His residues. This cation acts by
coordinating to the carbonyl oxygen of the ketone donor, and fa-
cilitates enolate formation. In both types of aldolases, stereo-
selectivity is controlled by the enzyme and does not depend on the
substrate’s structure or stereochemistry, which allows for highly
predictable products.3
* Corresponding author. Tel.: þ34 96 3543048; fax: þ34 96 3544328; e-mail
address: cecilia.andreu@uv.es (C. Andreu).
0040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.07.013

C. Andreu, G. Asensio / Tetrahedron 67 (2011) 7050e7056
7051
Interestingly, different enantiomers were found in excess under
both conditions.10a This complex can act as a good class II aldolases
model. Recently, Penhoat et al. performed a broad screening of
different alternative water-compatible Lewis acids cocatalysts with
proline in a neutral medium. They obtained the best results in
stereoselectivity and conversion using zinc chloride.11
Besides, Mlynarsky et al. showed that the addition of Zn(OTf)2
notably improved yield and stereoselectivity in the condensation
between different aldehydes and ketones catalyzed by bis(proli-
namides). This model assumes the formation of an enamine sta-
bilized in water by coordination to zinc at the active site. Zinc
coordinates with both the amide group of the catalyst and the
carbonyl group of the aldehyde, thus enabling condensation in the
chiral surrounding.12
were obtained when other protic solvents, like methanol, were
considered.
35
33
Pde (10%)
31
30
25
20
15
10
5
Zn:Pde 0.5:1 (5%)
Cu:Pde 0.5:1 (20%)
26
24
20
16
14
12
10
5
5
3
2. Results and discussion
0
In our work, we used the aldol reaction between acetone and p-
nitrobenzaldehyde catalyzed by L-prolinamide (Pde from now on-
0
5
10
Water content (%)
30
40
60
20
ward) as a benchmark, and as the simplest model for the proline
derivatives of the prolinamide type (Scheme 1). We considered that
this simple model would also be a good approximation to natural
aldolases, even when stereoselection was always modest, as it
contains both the pyrrolidine ring moiety of proline and the amide
function present in all enzymes. The addition of Zn2þ (zinc acetate)
under neutral conditions and the study of its effect may provide an
easier understanding of how both types of natural aldolases work.
In fact, metalloenzymes also contain a metal at their active site,
which is present on the protein surface with at least one free co-
ordination site on the metal ion. The substrate is activated by the
coordination to it.13 The prolinamide geometry with the pyrrolidine
nitrogen and the amide function fits the coordination sphere of
Zn2þ, forming a five-member ring. This coordination sphere can be
completed with another prolinamide molecule or other donor
species present in the solution.
Fig. 1. Effect of water content on the stereoselectivity of Pde and its complexes. Re-
actions were carried out in acetone as the main solvent and at room temperature until
completion.
Table 1 presents the results obtained when different Zn2þ/Pde
ratios were used. Identical stereoselectivities were reached with
the 1:1 and 0.5:1 stoichiometry (compare entries 4e8, 6e10, and
7e12). No reaction was observed in the absence of Pde or when zinc
acetate was used as the sole catalyst.
Table 1
Effect of the amount of Zn2þ/Cu2þ, temperature, and TFA as an additive
Entry
Conditionsa
% Cat
Time (h)
Yieldb
% eec
1
2
3
4
5
6
7
8
Pde, rt
10
10
10
5
5
5
5
5
5
5
5
5
5
20
20
20
12
12
48
7
24
8
45
3
20
5
30
24
96
48
65
96
100
100
18
100
100
100
80
100
100
100
100
100
50
12
12
37
33
40
33
30
33
42
33
44
30
37
31
30
19
Pde, 0 ꢀC
Pde/TFA 1:1, rt
Zn/Pde 1:1, rt
Zn/Pde/TFA 1:1:1, rt
Zn/Pde 1:1, 0 ꢀC
Zn/Pde 1:1, ꢁ20 ꢀC
Zn/Pde 0.5:1, rt
Zn/Pde/TFA 0.5:1:1, rt
Zn/Pde 0.5:1, 0 ꢀC
Zn/Pde/TFA 0.5:1:1, 0 ꢀC
Zn/Pde 0.5:1, ꢁ20 ꢀC
Zn/Pde/TFA 0.5:1:1,ꢁ20 ꢀC
Cu/Pde 0.5:1, TA
Cu/Pde 0.5:1, 0 ꢀC
Cu/Pde 0.5:1, ꢁ20 ꢀC
Scheme 1.
9
10
11
12
13
14
15
16
In the preliminary experiments, we compared yield and ster-
eoselectivity in the reactions catalyzed by L-prolinamide and its
isolated complexes with Zn2þ (Zn/Pde 0.5:1, or 1:1) and Cu2þ
(Cu/Pde 0.5:1), prepared as described in the Experimental section.
We used these isolated complexes in most experiments. However,
we observed that the results were exactly the same as in the ‘in situ’
preparation of the complexes by adding the stoichiometric
87
80
25
a
Reactions in acetone with 5% distilled water. Complexes prepared using zinc
acetate or copper(II) sulfate were used as a catalyst.
amounts of the metal salt and L-prolinamide to the reaction mix-
b
Determined by NMR.
ture, as described in Mlynarsky et al.12
c
HPLC (IC, Hex/iproh 94/6, 1 ml/min). See the Experimental section for an ab-
solute configuration of the main enantiomer.
Fig. 1 compares the results obtained with Pde, Zn/Pde (0.5:1) or
Cu/Pde (0.5:1) under different solvent conditions. In all cases, ex-
cess enantiomer had an R configuration, which is the stereoisomer
obtained in the proline enamine catalysis, as described in the lit-
erature.8b,c,14 The reactions catalyzed by the ZnePde complex
(0.5:1) were faster than those of the CuePde complex (0.5:1) or Pde
itself. Larger amounts of these must be used for the same conver-
sion, and the stereoselectivity obtained for free Pde was worse. The
best results obtained with the complexes were accomplished in the
presence of 5% distilled water, this being the minimal amount re-
quired to solubilize the complexes. Pde provided the best results
when water was not present at all, but the reaction was very slow
under these conditions. As usually observed, addition of water was
detrimental to stereoselectivity,6 and this is also true when using
the complexes as catalysts, but to a lower extent. Similar results
While no significant differences in stereoselectivity were ob-
served with temperature, the reaction rate dropped at lower tem-
peratures. Moreover in the reaction catalyzed by CuPde (0.5:1),
a reduction in ee took place at the lowest temperature, which is
probably due to poor complex solubility (entries 14e16).
Protonated chiral prolinamide derivatives are usually more ef-
fective than the free base for stereoselective aldol additions since
protonation diminishes the basic strength of the amine catalyst,
favoring iminium ion catalysis.15 In our case, we made similar ob-
servations. Table 1 and Fig. 2 show the results when 1 equiv of
trifluoroacetic acid, relative to prolinamide, was added under
standard conditions (5% water) at different temperatures. In all

7052
C. Andreu, G. Asensio / Tetrahedron 67 (2011) 7050e7056
cases, the use of protonated prolinamide resulted in slower re-
actions, but better stereoselectivity (entries 1e5, and 8e13). Thus,
the greatest increase occurred when prolinamide trifluoroacetate
salt was used instead of free Pde.
described to show two prolines, trans in relation to each other, and
coordinated to the Zn atom via their nitrogen and carboxylic oxy-
gen atoms. A fifth coordination site of the Zn atom is occupied by
a symmetry-related O atom of a neighboring proline molecule,
generating an infinite polymeric chain.16a Moreover, the complex
between zinc chloride and
conditions (Dichlorobis(DL-proline-
D
,
L
-proline, prepared under neutral
42
k
O)zinc(II)), shows Zn2þ ion
45
40
37
tetrahedrally coordinated with two halogenides and two O atoms
of two negatively charged carboxylates as donors. Hydrogen
bonding links the molecule in linear chains. Halogenides and car-
bonyl oxygens act as acceptors of hydrogen bonds, which are do-
nated by the positively charged ammonium groups.16b
40
33
33
35
30
25
20
15
10
5
Although knowledge of the complex structure in the solid state
is of interest, knowledge of the nature of the species in solution
under the reaction conditions could be even more interesting. The
analysis of the complexes by 1H NMR indicates new species
formation.
12
The 1H NMR spectra of the complex in methanol presented
a downfield displacement in the signals of the pyrrolidine ring
hydrogens. The shift of protons more closely to the nitrogen atom
was particularly significant, especially the double doublet signal
0
Pde Pde:TFA
(1:1)
Zn:Pde Zn:Pde:TFA
Zn:Pde Zn:Pde:TFA
(0.5:1)
(0.5:1:1)
(1:1)
(1:1:1)
corresponding to HeCa (Fig. 4). No significant differences in Pde
Fig. 2. Effect of TFA on stereoselectivity.
signals were observed in the spectra when a ratio of Zn/Pde 0.5:1 or
1:1 was used, which did not allow us to determine the nature and
stoichiometry of the formed species.
Regarding yield and ee, the same trend for 4-nitrobenzaldehyde
was observed for the 2- and 3-regioisomers used as substrates in
the aldol condensation with acetone (Fig. 3). Reactions with the
trifluoroacetate salts of the complexes were the most stereo-
selective. The greatest ee increase was also observed when com-
paring free prolinamide and its trifluoroacetate salt.
Zn:Pde 0.5:1 (5%)
Pde (10%)
60
50
40
30
20
10
2-nitrobenzaldehyde
51
3-nitrobenzaldehyde
47
45
45
45
40
23
Fig. 4. 1H NMR in methanol of Pde and its complexes with 0.5 or 1 equiv of zinc acetate
dihydrate.
20
The similar stereoselectivity obtained when using the tri-
fluoroacetate salt of prolinamide (PdeHþ) and the trifluoroacetate
salts of the zinc complexes (Table 1) may indicate that complexes
are not formed when the nitrogen of the amine is protonated by the
acid, being in these cases protonated prolinamide the catalytic
species. To confirm this hypothesis, the spectra of Pde/TFA and Zn/
Pde/TFA were recorded in methanol and in DMSO-d6. Both species
exhibited a downfield displacement in the pyrrolidine moiety sig-
nals in relation to prolinamide, especially for the former (Fig. 5).
The changes in the amide/ammonium area in the spectra recorded
in DMSO were highly significant: signal at 3.1 ppm (NeH) dis-
appeared in the protonated prolinamide, and two new signals
appeared at 8.3 and 9.4 ppm (NeH ammonium). These signals were
erased after the addition of zinc acetate, and a new broad signal
increased again at 3.1 ppm (NeH and water from the zinc acetate
dihydrate). These observations allow us to conclude that, in the
presence of Zn2þ, N of pyrrolidine coordinates to it, and that the
equilibrium among protonated Pde, free Pde, and Pde coordinated
to the ion must be displaced mostly to the later species. Similar
0
Pde Pde:TFA Zn:Pde Zn:Pde:TFA
Pde:TFA
Zn:Pde:TFA
Zn:Pde
(0.5:1)
Pde
(1:1)
(0.5:1)
(0.5:1:1)
(1:1)
(0.5:1:1)
Fig. 3. Reaction with 2 and 3-nitrobenzaldehyde (in acetone with 5% water).
2.1. NMR and ESI-MS measurements
The above results show that Zn2þ plays a role in the mechanism
of the reaction being produced, this being faster and more stereo-
selective in almost all cases, even when adding increased amounts
of water. It would be very interesting to know the catalytic complex
structure but, unfortunately, suitable crystals for an X-ray analysis
could not be obtained for either zinc complex (Zn/Pde 0.5:1, Zn/Pde
1:1), or for the copper complex (Cu/Pde 0.5:1).
The complex between zinc oxide and
basic conditions (Bis( -prolinato-N,O)zinc(II)), was previously
L-proline, prepared under
L

C. Andreu, G. Asensio / Tetrahedron 67 (2011) 7050e7056
7053
Fig. 5. (a) 1H NMR in methanol and (b) in DMSO-d6 of Pde, Pde with 1 equiv of trifluoroacetic acid, Pde with 1 equiv of both trifluoroacetic acid and 0.5/1 equiv of zinc acetate, and
Pde with zinc acetate.
behavior was observed in the experiments carried out in methanol
when studying the signal corresponding to H
and 13C NMR spectra of Pde showed the presence of two species,
but only the signals corresponding to one of them remained several
hours later (Fig. 6).
a
.
Electrospray ionization mass spectrometry (ESI-MS) has proved
to be a useful tool in the qualitative study of a solution at equilib-
rium.17 Application of the positive ion mode (ESIþ) allowed us to
detect all the participating species. Stoichiometry could be de-
termined directly from their m/z values and the isotopic pattern of
peaks. Each peak’s relative intensity should not be considered
quantitatively if we believe that the sensitivity of this technique to
detect different species depends on its ability to undergo
ionization.
The ESI-MS spectrum corresponding to the relation Zn/Pde 1:1
in methanol shows an equilibrium of different complexes between
prolinamide and Zn2þ (Supplementary data). A peak at m/z 237
corresponding to a Zneprolinamide complex (1:1) with one acetate
equivalent, and the peaks at m/z 291, and the most intense one at
351, were, respectively, expected for those complexes with one Zn
and two prolinamides (0.5:1), and with one additional acetate ion
in the second case. The ESI-MS spectrum corresponding to a Zn/Pde
0.5:1 ratio in methanol solution was very similar, but the peak at m/
z 237 appeared with very low intensity. Even though this is
a qualitative technique, the behavior of all these species must be
similar under ionization conditions. The fact that the peak at m/z
351, was the most intense one in all cases indicates that the stoi-
chiometry for the most stable complex is two molecules of Pde and
one of zinc ion (ZnPde2).
Fig. 6. 1H NMR in acetone/water (85:15) of Pde (at different times) and Zn/Pde 0.5:1.
In order to clarify the identity of this species, we must consider
all the possible intermediates that might form between Pde and
acetone. Most mechanistic studies performed in the aldol reaction
use Pro as catalyst.5,18 To date very few use prolinamide de-
rivatives.19 In all cases, an enamine-based mechanism is accepted if
compared with the reactions catalyzed by Class I aldolase enzymes
(Scheme 3). The addition of proline to acetone provides different
kinds of intermediates whose stability depends on the reaction
The ESI-MS corresponding to Zn/Pde/TFA 1:1:1 shows the
presence of some 1:1 (m/z 195, 237,) and 1:2 (m/z 291, 323, 351)
complexes between Zn and Pde. These data agree with the NMR
experiments shown in Fig. 5, indicating the major preference of the
N of pyrrolidine to be coordinated by zinc (Scheme 2).
Scheme 2.
A series of similar experiments was carried out using deuterated
acetone as a solvent. The zinc complexes of prolinamide were in-
soluble in pure acetone, and could not be dissolved in the solvents
mixture used in our reactions containing 5% water, at least at the
concentration required for the analysis by 13C NMR. Therefore,
a mixture of 85:15 acetone/water had to be used. Initially, the 1H
conditions. The first intermediate is a hemiaminal that affords an
iminium carboxylate after dehydratation, which is in equilibrium
with an oxazolidinone or an enamine, considered the key in-
termediate in the addition to the acceptor aldehyde.5 The enamine
had never been detected in situ in different NMR studies before
Gschwind et al. described it as an intermediate in the self-

7054
C. Andreu, G. Asensio / Tetrahedron 67 (2011) 7050e7056
aldolization of propionaldehyde catalyzed by proline in DMSO
(water content less than 15%).20 In fact, theoretical studies done in
both the gas phase and solution have shown a lack of stability for
this intermediate.18 Oxazolidinone, considered a parasitic5c,18 spe-
cies in the asymmetric aldol reaction, is usually described as the
main compound when using aprotic solvents. Equilibrium shifts to
the iminium intermediate in the presence of protic solvents.18a On
ꢁ
hemiaminal formation easier (Scheme 4). The dehydration of this
species is faster too with the complex because zinc, acting as a Lewis
acid, could coordinate to the O of the hydroxyl group of the hemi-
aminal, thus favoring its dehydration and nucleophilic enamine
formation. Zn2þ coordination to the O of the acceptor aldehyde
makes it more electrophilic and also increases the reaction rate.
Moreover, this interaction fixes aldehyde in such a way that the
nucleophilic Re attack is favored for steric considerations. Although
less effective, the interaction with aldehyde could also be performed
by hydrogen bonds forming with the amide hydrogens when using
Pde as a catalyst. When the concentration of water (or of any other
protic solvent) in the reaction medium increases, it competes with
the other hand, the studies carried out by Gryko and Moran using L-
prolinethioamides and aromatic prolinamides, respectively, have
shown the formation of a cyclic imidazolidinone as the main in-
termediate, which is stable under different conditions, and even
has been isolated and fully characterized (Scheme 3).19
Scheme 3. Intermediates proposed for proline (1, XR¼OH) and prolinamine derivatives (1, XR¼NHR) catalyzed asymmetric aldol condensation.
The 13C NMR spectra of Pde in the mixture used (acetone/water
85:15) shows a signal at 76 ppm, corresponding to a quaternary
carbon bonded to oxygen and nitrogen atoms. This product, and in
analogy to the studies described before, has been identified as the
imidazolidinone derived from the condensation between prolina-
mide and acetone (Supplementary data). Those signals corre-
sponding to methyl-d3 carbons could not be observed in acetone-d6
due to DeC coupling. Therefore, in order to fully characterize this
intermediate, it was synthesized in a flask by dissolving in acetone
(acetone-h6) Pde for 20 h. Evaporation of the solvent yielded
a white solid whose spectrum recorded in CDCl3 showed two clear
singlets corresponding to the diastereotopic CH3 signals.
Zn2þ or with the hydrogens of the amide group in the ZnPde or Pde
catalyzed reactions, respectively, by breaking down the chiral en-
vironment for enantioselective addition.
Finally, the results in Table 1 suggest the enhancement of the
€
stereoselectivity by using Bronsted acid (TFA) as additive. This
The 1H and 13C NMR spectra of the Zn2þePde complex in ace-
tone-d6 (Fig. 6) indicate the absence of the imidazolidinone in-
termediate and
a downfield displacement of those signals
corresponding to Pde.
The ESI-MS spectra in acetone solution were also recorded. The
base peak at m/z 155 corresponds to one of the imine/enamine or
imidazolidinone intermediates. Peaks at m/z 237 and m/z 291 were
also present in those samples corresponding to the complexes.
Scheme 4. Mechanism proposed for the cross aldol reaction between 4-
nitrobenzaldehyde and acetone catalyzed by the zinc complex of prolinamide (only
one Pde is shown for clarity).
2.2. Mechanistic considerations
observation can be easily explained by the fact that protonation of
The results observed in our study are consistent with Darbre’s
experiments using Zn(Pro)2.10 For Pde, the NMR and ESI-MS ex-
periments indicate the formation of complexes whose majority
stoichiometry Zn/Pde in solution is 0.5:1 (ZnPde2). To check if the
reaction catalyzed by the complex occurs through the formation of
an imineeenamine type intermediate, a solution of ZnPde2 and
1.5 equiv of acetone in methanol was treated with NaBH4.10e The
isolation of N-isopropyl prolinamide allowed us to test this hy-
pothesis (Supplementary data). Therefore, we can conclude that
the catalyst acts as Class I aldolase.
Reactions with free Pde proved to be slower and less stereo-
selective than those with the zinc complexes. This fact indicates that
Zn2þ plays a role in the reaction rate and in stereoselection, and that
the model is also analogous to Class II aldolases.11 Zn2þ may accel-
erate the process by acting as a Lewis acid because it coordinates the
ketone, approaches the N of the pyrrolidine ring and makes
€
the amine prevents its action as Bronsted base in the intermolecular
condensation that would lead to the racemic aldol. So, in acidic
conditions the enantioselective nucleophilic attack is the mecha-
nism favored.12a,15 The competitive reaction as Bronsted base seems
€
to be more important when Pde, rather than its zinc complexes, is
used as a catalyst, and stereoselectivity increases further by its
protonation (Fig. 2). In the zinc complexes, the basicity of nitrogen
has already diminished because of ion coordination. Addition of
€
Bronsted acid to Pde allows the attainment of similar ee than with
ZnePdecomplexes, althoughthereactionisfasterinthesecond case.
The interesting mechanistic studies conducted by Gryko and
ꢁ
Moran with prolinamide derivatives as catalysts for the aldol re-
action has allowed them to verify that the imidazolidinone in-
termediate could be responsible for the reduction in yield and
stereoselectivity of the aldol product.19 They have observed that the

C. Andreu, G. Asensio / Tetrahedron 67 (2011) 7050e7056
7055
€
addition of Bronsted acid inhibits their formation, being able to
respectively, collected by vacuum filtration and dried (70%, 60% and
improve the stereochemical results.
96% yields, respectively).
The conditions we used (addition of TFA or Zn2þ to Pde) could
suppress the formation of the imidazolidinone intermediate, and as
result lead to an increase of the stereoselectivity and yield in re-
lation to reactions catalyzed with Pde. In fact, as shown in Fig. 6,
imidazolidinone intermediate is observed only in the spectrum of
Pde in acetone, but not in that of the ZnPde2.
We are currently conducting experiments that allow us to learn
more about the role of this intermediate imidazolidinone and the
possible inhibition of its formation through the use of complexes of
Zn as catalyst.
4.2.1. Pde. 1H NMR (MeOD): 1.70e1.83 (m, 3H), 2.00e2.20 (m, 1H),
2.89e2.93 (m, 1H), 2.90e3.10 (m, 1H), 3.60e3.70 (m, 1H). 13C NMR
(MeOD): 27. 04 (t), 32.33 (t), 48.03 (t), 61.39 (d), 179.89 (s).
4.2.2. Imidazolidinone Pdeeacetone. 1H NMR (CDCl3): 1.43 (s, 3H),
1.45 (s, 3H), 1.60e2.02 (m, 3H), 2.06e2.29 (m, 1H), 2.56e2.66 (m,
1H), 2.88e3.02 (m, 1H), 3.92 (1H, dd, 9.76 Hz, 4.75 Hz), 6.67 (1H, br
s).13C NMR (CDCl3): 23.63 (q), 25.65 (t), 25.82 (t), 30.72 (q), 48.69
(t), 63.98 (d), 74.84 (s).
HRMS (EIþ) calcd for C8H14N2O 154.1106, found 154.1104. Anal.
Calcd for C8H14N2O: C 62.33, H 9.09, N 18.18. Found C 62.20, H 9.16,
N 18.10. Mp 140e142 ꢀC.
3. Conclusions
In short, we have studied the possible role of zinc ion in aldol
condensations catalyzed by the simple Pde model. We conclude
that this catalyzes the reaction following the general mechanism of
enamine nucleophilic addition to the acceptor aldehyde. Zn2þ not
only prevents the base-catalyzed reaction by diminishing the ba-
sicity of the amine nitrogen of Pde, but also acts by favoring en-
amine formation in an aqueous medium. Besides, the Lewis acid
character of Zn2þ makes the interaction with the acceptor aldehyde
more effective. Exclusion of water from the reaction center could
also avoid its coordination with ion, thus making it more acidic, and
could prevent its coordination to the aldehyde. The stereo-
selectivity achieved with this model is modest, but can be improved
by using other derivatives in which the ion is wrapped in a more
hydrophobic environment.
4.2.3. ZnPde2. 1H NMR (MeOD): 1.75e1.95 (m, 3H), 2.33e2.45
(m, 1H), 3.00e3.20 (m, 2H), 3.95e4.05 (m, 1H). 13C NMR (MeOD):
27.44 (t), 32.36 (t), 48.62 (t), 60.16 (d), 180.27 (s). ESI-MSþ (MeOD):
351, 291, 236.
4.2.4. CuPde2. ESI-MSþ (MeOD): 290.
See Supplementary data.
4.3. Typical aldol condensation procedure
The corresponding aldehyde (0.165 mmol, 25 mg) and catalyst
(5% of the zinc complex or 10% of Pde) were added to a mixture of
acetoneewater (2.375e0.125 ml) in a capped vial. The resulting
mixture was stirred at the indicated temperature and monitored by
TLC. The reaction was quenched by addition of saturated ammo-
nium chloride solution and evaporation of acetone in vacuum.
Then, water was extracted with methylene chloride and the organic
phase was dried over sodium sulfate. Crude material was employed
to determine the yield by NMR and the ee by HPLC using a chiral
stationary phase.
4. Experimental
4.1. General
All the commercially available reagents were purchased from
Aldrich. Reactions were monitored by thin layer chromatography
(TLC) on Merck silica plates 60 F254. Flash chromatography was
performed on Merck silica gel (60-particle size: 0.040e0.063 mm).
NMR spectra were recorded with Bruker DRX 300 spectrometers in
different deuterated solvents. Chemical shifts are reported in parts
per million in relation to the residual solvent peaks. Absolute
configurations were determined by comparing with the optical
rotations reported in the literature, and these were performed in
a PerkineElmer 241 Polarimeter using a Na lamp. Electrospray
ionization mass (ESI-MS) spectra were recorded in an Ion Trap
Esquire 3000þMass Spectrometer (Bruker), coupled with an HPLC
Reaction products were purified previously by column chro-
matography (hexane/ethyl acetate, 4:1) and characterized by NMR
and chiral HPLC. Data were consistent with those described in the
literature.8b,c
4.3.1. 4-Hydroxy-4-(p-nitrophenyl)-butan-2-one. 1H NMR (CDCl3):
2.21 (s, 3H), 2.84 (m, 2H), 3.60 (s, 1H), 5.26 (m, 1H), 7.53 (d, 8.8 Hz,
2H), 8.20 (d, 8.8 Hz, 2H). ee was determined by HPLC with a Chir-
alpak IC column 94/6 hexane/2-propanol, 254 nm, 1.0 mL/min,
tR¼31.6 (minor, S), tR¼33.5 min (major, R).
Agilent 1100. Samples were analyzed by direct infusion at 240 ml/h.
Positive-ion conditions: ES capillary voltage, 4100 V; Skimmer,
40 V; Cap Exit,135 V; Trap Drive, 37; Oct RF,138.6 Vpp; Lens 1, ꢁ5 V;
Lens 2, ꢁ60 V; Nebulizer, 10 psi; Dry Gas, 5 l/min; Dry Ta, 350 ꢀC.
Scanning was carried out from 80 to 600 m/z.
High performance liquid chromatography (HPLC) was per-
formed with a Merck Hitachi Lachrom system. For the analytical
4.3.2. 4-Hydroxy-4-(o-nitrophenyl)-butan-2-one. 1H NMR (CDCl3):
2.16(s, 3H), 2.63(dd,17.8Hz, 9.4Hz,1H), 3.09(dd,17.8Hz, 2.0Hz), 3.90(s,
1H), 5.60 (m, 1H), 7.40 (m, 1H), 7.61 (m, 1H), 7.86 (m, 1H). ee was de-
termined by HPLC with a Chiralpak IC column 85/15 hexane/2-propanol,
254 nm, 1.0 mL/min, tR¼24.8 (major, R), tR¼27.0 min (minor, S).
work, a Chiralpak IC column (5
the solvent mixture indicated in each case. The wavelength for
detection was fixed at 254 nm.
m
m, 250ꢂ4.6 mm ID) was used with
4.3.3. 4-Hydroxy-4-(m-nitrophenyl)-butan-2-one. 1H NMR (CDCl3):
2.23 (s, 3H), 2.90 (m, 2H), 3.59 (s, 1H), 5.30 (m, 1H), 7.53 (m, 1H),
7.70 (m, 1H), 8.24 (m, 1H). ee was determined by HPLC with
a Chiralpak IC column 80/20 hexane/2-propanol, 254 nm, 1.0 mL/
min, tR¼12.6 min (major, R), tR¼13.6 (minor, S).
4.2. Isolation of complexes
Complexes Zn2þ/prolinamide 1:2 (ZnPde2), 1:1 (ZnPde) and
Cu2þ/prolinamide 1:2 (CuPde2) were prepared by mixing a solution
of neutral Pde (1.75 mmol, 200 mg) in tetrahydrofurane with a so-
lution of either zinc acetate dihydrate (0.87 or 0.435 mmol) or
cupper nitrate trihydrate (0.87 mmol) in the minimum amount of
this solvent in which these salts are soluble. The mixture was
stirred for 30 min, and the white and blue precipitates were,
Acknowledgements
ꢁ
This work has been supported by Spanish Direccion General de
ꢁ
ꢁ
Investigacion Científica y Tecnica Consolider Ingenio 2010 (CSD
2007-00006), (CTQ 2007-65720). We gratefully acknowledge SCSIE
ꢀ
(Universitat de Valencia) for access to instrumental facilities.

7056
C. Andreu, G. Asensio / Tetrahedron 67 (2011) 7050e7056
ꢁ
ꢁ
8. (a) Guillena, G.; Najera, C.; Ramon, D. J. Tetrahedron: Asymmetry 2007, 18, 2249;
(b) Russo, A.; Botta, G.; Lattanzi, A. Tetrahedron 2007, 63, 11886; (c) 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.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tet.2011.07.013.
9. Thang, Z.; Jiang, F.; Cui, X.; Gong, L. Z.; Mi, A.-Q.; Jiang, Y. Z.; Wu, Y.-D. Proc. Natl.
Acad. Sci. U.S.A. 2004, 101, 5755.
10. (a) Darbre, T.; Machuqueiro, M. Chem. Commun. 2003, 1090; (b) Kofoed, J.;
Reymond, J. L.; Darbre, T. Org. Biomol. Chem. 2005, 3, 1850; (c) Kofoed, J.;
Machuqueiro, M.; Reymond, J.-L.; Darbre, T. Chem. Commun. 2004, 1540; (d)
References and notes
ꢁ
ꢁ
Fernandez-Lopez, R.; Kofoed, J.; Machuqueiro, M.; Darbre, T. Eur. J. Org. Chem.
2005, 5268; (e) Kofoed, J.; Darbre, T.; Reymond, J.-L. Chem. Commun. 2006,
1482.
1. Modern Aldol Reactions; Marhwald, R., Ed.; Wiley-VCH: Weinheim, 2004.
2. Organic Synthesis in Water; Grieco, P. A., Ed.; Blackie Academic & Professional:
London, 1988.
3. (a) Fessner, W.-D. In Stereoselective Biocatalysis; Patel, R. N., Ed.; Marcel Dekker:
New York, NY, 2000; p 239; (b) Machajewski, T. D.; Wong, C.-H. Angew. Chem.,
Int. Ed. 2000, 39, 1352.
11. Penhoat, M.; Barbry, D.; Rolando, C. Tetrahedron Lett. 2011, 52, 159.
12. (a) Paradowska, J.; Stodulski, M.; Mlynarski, J. Adv. Synth. Catal. 2007, 349, 1041;
(b) Paradowska, J.; Stodulski, M.; Mlynarski, J. Angew. Chem., Int. Ed. 2009, 48,
4288.
13. (a) Kimura, E. Pure Appl. Chem. 1993, 65, 355; (b) Mareque Rivas, J. C.; Salvagni,
E.; Prabaharan, R.; Torres Martin de Rosales, R.; Parsons, S. Dalton Trans. 2004,
172; (c) Trzaskowski, B.; Adamowicz, L.; Deymier, P. A. J. Biol. Inorg. Chem. 2008,
13, 133.
14. (a) List, B.; Lerner, R. A.; Barbas, C. F., III. J. Am. Chem. Soc. 2000, 122, 2395; (b)
List, B.; Notz, W. J. J. Am. Chem. Soc. 2000, 122, 7386; (c) Northrup, A. B.;
MacMillan, D. W. C. J. Am. Chem. Soc. 2002, 124, 6798.
15. (a) Roberts, R. D.; Ferran, H. E.; Gula, J. M.; Spencer, T. A. J. Am. Chem. Soc. 1980,
102, 7054; (b) Chimni, S. S.; Mahajan, D.; Babu, V. V. S. Tetrahedron Lett. 2005,
46, 5617; (c) Chimni, S. S.; Singh, S.; Mahajan, D. Tetrahedron: Asymmetry 2008,
19, 2276; (d) Chimni, S. S.; Mahajan, D. Tetrahedron: Asymmetry 2006, 17, 2108;
(e) Gryko, D.; Zimnicka, M.; Lipinski, R. J. Org. Chem. 2006, 17, 964.
16. (a) Ng, C.-H.; Fun, H.-K.; Teo, S.-B.; Teoh, S.-G.; Chinnakali, K. Acta Crystallogr.,
Sect. C 1995, 51, 244; (b) Lutz, M.; Bakker, R. Acta Crystallogr., Sect. C 2003, 59, 18.
17. Di Marco, V. B.; Bombi, G. G. Mass Spectrom. Rev. 2006, 25, 347.
18. (a) Seebach, D.; Beck, A. K.; Badine, D. M.; Limbach, M.; Schenmoser, A.;
Treasurywala, A. M.; Hobi, R. Helv. Chim. Acta 2007, 90, 425; (b) Hu, Y. F.; Lu, X.
Chinese J. Struct. Chem. 2008, 5, 547.
4. (a) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2004, 43, 5138; (b) List, B.
Tetrahedron 2002, 58, 5573; (c) Sakthivel, K.; Notz, V.; Bui, T.; Barbas, C. F., III. J.
Am. Chem. Soc. 2001, 123, 5260; (d) Jarvo, E. R.; Miller, S. J. Tetrahedron 2002, 58,
2481; (e) Kotsuki, H.; Ikishima, H.; Okuyama, A. Heterocycles 2008, 75, 493; (f)
Stamatis, F.; Christoforos, K.; Evaggelia, T.; Kokotos, G. Eur. J. Org. Chem. 2011, 7,
1310.
5. (a) Hoang, L.; Bahmanyar, S.; Houk, K. M.; List, B. J. Am. Chem. Soc. 2003, 125, 16;
(b) Bahmanyar, S.; Houk, K. N. J. Am. Chem. Soc. 2001, 123, 11273; (c) List, B.;
Hoang, L.; Martin, H. J. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5839.
6. (a) Dickerson, T. J.; Janda, K. D. J. Am. Chem. Soc. 2002, 124, 3220; (b) Ramachary,
D. B.; Chowdari, N. S.; Barbas, C. F., III. Tetrahedron Lett. 2003, 44, 1923.
7. (a) Mase, N.; Nakai, Y.; Ohara, N.; Yoda, H.; Takabe, K.; Tanaka, F.; Barbas, C. F., III.
J. Am. Chem. Soc. 2006, 128, 734; (b) Huang, J.; Zhang, X.; Armstrong, D. W.
Angew. Chem., Int. Ed. 2007, 46, 9073; (c) Chen, Y.; Reymnond, J.-L. J. Org. Chem.
1995, 60, 6970; (d) Hayashi, Y.; Sumiya, T.; Takahashi, J.; Gotoh, H.; Urushima, T.;
Shoji, M. Angew. Chem., Int. Ed. 2006, 45, 958; (e) Hayashi, Y.; Aratake, S.; Okano,
T.; Takahashi, J.; Sumiya, T.; Shoji, M. Angew. Chem., Int. Ed. 2006, 45, 5527; (f)
Tzeng, Z.-H.; Chen, H.-Y.; Huang, C.-T.; Chen, K. Tetrahedron Lett. 2008, 49, 4134;
(g) Jia, Y. M.; Wu, F.-C.; Ma, X.; Zhu, G.-J.; Da, C.-S. Tetrahedron Lett. 2009, 50,
3059; (h) Zhang, S.-P.; Fu, X.-K.; Fu, S.-D. Tetrahedron Lett. 2009, 50, 1173; (i)
Giacalone, F.; Gruttadauria, M.; Agrigento, P.; Lo Meo, P.; Noto, R. Eur. J. Org.
19. (a) Gryko, D.; Lipinski, R. Eur. J. Org. Chem. 2006, 3864; (b) Gryko, D.; Zimnicka,
ꢁ
M.; Lipinski, R. J. Org. Chem. 2007, 72, 964; (c) Fuentes de Arriba, A. L.; Simon, L.;
ꢁ
~
ꢁ
Raposo, C.; Alcazar, V.; Sanz, F.; Muniz, F. M.; Moran, J. R. Org. Biomol. Chem.
ꢁ
ꢁ
2010, 8, 2979.
Chem. 2010, 29, 5696; (j) Pedrosa, R.; Andres, J. M.; Manzano, R.; Roman, D.;
Tellez, S. Org. Biomol. Chem. 2011, 9, 935.
ꢁ
20. Schmid, M. B.; Zeitler, K.; Gschwind, R. M. Angew. Chem., Int. Ed. 2010, 49, 4997.