Substrate-dependent nonlinear effects in proline-thiourea-catalyzed aldol reactions: Unraveling the role of the thiourea co-catalyst

Article abstract of DOI:10.1002/chem.200902678

Figure Presented It's an affair of four: The study of nonlinear effects in the proline-thioureacatalyzed aldol reaction between acetone and aromatic aldehydes, together with NMR and ESI-MS experiments, shows that the main role of the thiourea co-catalyst is that of promoting both the formation of a soluble Seebach's oxazolidinone intermediate and its conversion into the iminium carboxylate isomer.

Full text of DOI:10.1002/chem.200902678

DOI: 10.1002/chem.200902678
Substrate-Dependent Nonlinear Effects in Proline–Thiourea-Catalyzed
Aldol Reactions: Unraveling the Role of the Thiourea Co-Catalyst
Niama El-Hamdouni,[a, b] Xavier Companyꢀ,[a] Ramon Rios,*[a, c] and Albert Moyano*[a]
Dedicated to Prof. Santiago Olivella on occasion of his 65th birthday
The venerable aldol reaction[1] is enjoying a second youth,
since the re-discovery of its catalysis by proline at the begin-
ning of the century, and the subsequent explosive develop-
ment of enantioselective organocatalysis.[2–5] Even if proline
is an exceptionally good catalyst for this reaction in terms of
its structural simplicity and easy availability, it exhibits two
main drawbacks: a low solubility in nonpolar solvents and a
relatively low reactivity, that requires its use in large
amounts (typically 20–30 mol%) and that leads to the for-
mation of parasitic reaction products especially in the case
of aldol reactions with aromatic aldehydes.[6] It is therefore
not surprising that in the past few years considerable effort
has been devoted to the chemical modification of proline
(or of 4-hydroxyproline) in order to obtain catalytic systems
both with better solubility in common organic media and/or
with higher acidity of the directing acid proton.[3,6b,7] A con-
ceptually attractive alternative strategy to this goal, due to
its simplicity, would be based in the in situ preparation of a
catalytically competent supramolecular assembly from un-
modified proline and a simple hydrogen-bond-donor co-cat-
alyst, taking advantage of the well-known ability of carbox-
ylic acids to participate in hydrogen-bonding networks.[8]
While some precedent for this approach can be found in the
work of Zhou,[9] Zhao,[10] and Clarke,[11] these authors either
use chiral co-catalysts[9,10] or perform extensive chemical
modification of proline.[11,12] In 2006, Miller and co-work-
ers[13] described that the addition of N-methylimidazole to l-
proline enhanced the rate of the Morita–Baylis–Hillman re-
action between methyl vinyl ketone and aromatic aldehydes,
although the resulting adducts were obtained with very low
enantiomeric purity. At the beginning of the present year,
Reis et al.[4] reported that the addition of equimolar quanti-
ties of Schreinerꢀs thiourea 1[14] to proline in nonpolar sol-
vents resulted in the formation of a supramolecular catalyst
that gave excellent conversions and stereoselectivities in the
aldol reaction of cyclohexanone with various aromatic alde-
hydes. In a closely related approach, we have subsequently
demonstrated[5] that the combination of proline with several
kinds of achiral hydrogen-bond donors (o-phenylene dicar-
bamates, 2-(arylamino)benzimidazoles, N.N’-diarylthiour-
eas), using toluene as a solvent, provides a simple solution
to the direct aldol desymmetrization of 4-substituted cyclo-
hexanones, a challenging reaction that is not efficiently cata-
lyzed by proline itself.
To explain their results, Reis et al. proposed the initial for-
mation of a proline–thiourea complex, much more soluble
than proline itself in apolar organic solvents (Figure 1, left),
and suggested that this supramolecular assembly was carried
out to the List–Barbas–Houk transition state,[2b,c,15] improv-
ing the stereoselectivity of the aldol reaction by enhancing
the directing effect of the carboxylic acid group of proline
(Figure 1, right).[4]
While we have demonstrated the formation of stable 1:1
complexes between proline and N,N’-diaryl thioureas (in-
cluding Schreinerꢀs thiourea, 1) in chloroform by UV and
fluorescence spectroscopic techniques;[5] the high dilution
conditions in which these experiments are performed bear
no direct relationship with the concentration of the catalyst
in the actual aldol reactions, which on the other hand are
[a] N. El-Hamdouni, X. Companyꢁ, Dr. R. Rios, Prof. Dr. A. Moyano
Departament de Quꢂmica Orgꢃnica
Universitat de Barcelona, C. Martꢂ i Franquꢄs 1-11
08028-Barcelona, Catalonia (Spain)
Fax : (+34)933397878
E-mail: amoyano@ub.edu
[b] N. El-Hamdouni
Laboratoire de Physico-Chimie des Interfaces et Environnement
Dꢅpartement de Chimie, Facultꢅ des Sciences
Universitꢅ Abdelmalek-Essaadi, BP 2121
93000-Tꢅtouan (Morocco)
[c] Dr. R. Rios
Catalan Institution for Research and Advanced Studies (ICREA)
Pg. Lluꢂs Companys 23, 08018-Barcelona, Catalonia (Spain)
E-mail: rios.ramon@icrea.cat
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200902678.
1142
ꢆ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 1142 – 1148

COMMUNICATION
to characteristic nonlinear effects (positive for low eeꢀs and
negative for high eeꢀs of the solid proline catalyst, joined by
a plateau reflecting the formation of an eutectic for inter-
mediate eeꢀs of proline) that give a direct and clear-cut indi-
cation of the presence of solid proline as a catalyst source in
the reaction medium.[21] We report now on our initial results
in this subject, which have revealed a hitherto unprecedent-
ed substrate dependence of the nonlinear effects and strong-
ly suggest that the role of the hydrogen-bond-donor co-cata-
lyst is much more complex than previously assumed by Reis
et al.[4]
We selected as benchmark reactions the heterogeneous
proline-catalyzed aldol additions of acetone to 4-nitrobenz-
AHCTUNGTREGaNNNU ldehyde (2a) and to 4-bromobenzaldehyde (2b) (as repre-
sentative highly reactive and less reactive aromatic alde-
hydes, respectively) in a 5:1 toluene–acetone mixture (in
which the catalyst is only sparingly soluble) as the solvent.
The chosen co-catalyst was the N-(3,5-bis (trifluoromethyl)-
Figure 1. Left: Proposed structure of the proline–thiourea complex.
Right: Proposed transition state for the proline-thiourea catalyzed aldol
reaction.
phenyl)-N’-(2,6-diisopropylphenyl) thiourea (3),
a com-
run in less polar hydrocarbon solvents such as toluene and
hexane. Most recently, McQuade and co-workers[16] have
disclosed the results of a study on the effect of bifunctional
urea co-catalysts in the asymmetric a-aminoxylation of alde-
hydes under proline catalysis.
This reaction gives excellent levels of enantioselectivity,
and the authors centered their attention on the reaction
rate, which turns out to be substantially increased upon ad-
dition of the urea co-catalyst. Interestingly enough, simple
NMR experiments showed that the solubility of proline in
the reaction medium (ethyl acetate) was not appreciably en-
hanced by the presence of the urea, and the authors, build-
ing upon previous kinetic studies of Blackmond,[17] proposed
that observed rate enhancement might be due to the inter-
action between Seebachꢀs oxazolidinone intermediate[6] and
the urea, that would increase the rate of enamine formation
(Figure 2).[16]
pound that had given excellent results in the aldol desym-
metrizations studied by us.[5] The reactions with 4-nitrobenz-
AHCTUNGTREGaNNUN ldehyde were performed at ambient temperature with a
10 mol% of proline, and with a 0.25m concentration of alde-
hyde. In the case of 4-bromobenzaldehyde, the reaction rate
was very low with a 10 mol% amount of proline, and the
study was performed with a 30 mol% of the catalyst. After
stirring for 2 h a suspension of the proline catalyst in the tol-
uene/acetone mixture (to ensure the full establishment of
solubility equilibria and therefore avoiding the presence of
disturbing “kinetic conglomerate” effects[22]), the aldehyde
(either 2a or 2b) was added in one portion. The conversion
of the reaction was monitored by 1H NMR spectroscopy,
and the enantiomeric purities of the aldol adducts 4a and
4b were determined by HPLC.[23] The reactions with the
proline–thiourea catalytic assembly were run in a similar
way (10 mol% proline+10 mol% thiourea 3). The results
obtained with enantiopure l-proline are summarized in
Scheme 1.
The presence of thiourea 3 substantially increased the
rate of these reactions. Thus, the aldol addition of 4-bromo-
benzaldehyde (2b) required four days to achieve a 40%
conversion with 0.3 equiv of l-proline as the sole catalyst,
while a similar conversion was reached after only 24 h when
0.1 equivalents of l-proline and 0.1 equivalents of thiourea 3
were used. A similar effect, albeit less pronounced, was ob-
served for the more reactive 4-nitrobenzaldehyde (2a); thus,
a 15% conversion was achieved after 20 h with 0.1 equiva-
lents of l-proline, and a 35% conversion was measured
after the same time with 10 mol% of the l-proline–thiourea
catalytic system. As can be seen in Scheme 1, the proline–
thiourea catalytic combination also afforded sizable increas-
es in the enantiomeric purities of the aldol products (from
60–71% ee in the case of 4a, and from 51–63% ee in the
case of 4b).
Figure 2. Possible interaction by hydrogen bonding between a Seebach
oxazolidinone and a urea (adapted from reference [16]).
These findings cast serious doubts upon the extent and
relevance of proline solubilization by the thiourea co-cata-
lyst in the aldol reactions explored by Reis et al.[4] and by
us,[5] and we felt that this important issue deserved more de-
tailed consideration. We envisaged that the comparative
study of nonlinear effects[18] in the heterogeneous proline-
catalyzed aldol reaction, both in the absence and in the
presence of a thiourea co-catalyst, could shed some light on
this problem. As demonstrated in the pioneering contribu-
tions of Hayashi[19] and Blackmond,[20] the higher solubility
of enantiopure proline with respect to the racemic gives rise
We then proceeded to the study of nonlinear effects, by
running the reactions in the same conditions described
above, but using freshly prepared, carefully dried solid pro-
Chem. Eur. J. 2010, 16, 1142 – 1148
ꢆ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1143

A. Moyano, R. Rios et al.
of the graph, showing that a eu-
tectic mixture of proline was
still present in the reaction
medium. The nonlinear effects
are therefore dependent on the
structure of the substrate,
a
phenomenon that has been
rarely observed in homogene-
ous catalytic asymmetric reac-
tions[24] and that, to the best of
our knowledge, is totally unpre-
cedented for partially soluble
catalysts.
In the light of these findings,
we reexamined the solubility
behavior
[D8]toluene
[D8]toluene/[D6]acetone in the
of
proline
and in
in
5:1
Scheme 1. Aldol reactions of acetone with benzaldehyde derivatives 2a and 2b catalyzed by l-proline and by
a proline–thiourea complex.
line samples of different enantiomeric purities. The results
of this study are summarized in Figures 3 (for aldehyde 2a)
and 4 (for aldehyde 2b), in which we show the usual plots
Figure 3. Influence of the thiourea co-catalyst (10 mol%) in the nonlinear
effects in the proline-catalyzed (10 mol% of proline) aldol reaction be-
~
&
tween acetone and 4-nitrobenzaldehyde (2a) ( without urea,
with
urea).
Figure 4. Influence of the thiourea co-catalyst (10 mol%) on the nonlin-
ear effects in the proline-catalyzed (10 mol%) aldol reaction between
acetone and 4-bromo benzaldehyde 2b with (bottom) and without (top)
of catalyst% ee versus product% ee, both in the absence
and in the presence of thiourea 3. The points in each graphic
are the mean values of at least two duplicate experiments.
The results obtained for 4-nitrobenzaldehyde 2a were
clear-cut, and seemed to support a complete solubilization
of proline by the thiourea 3 in the reaction medium. In
effect, in its absence a characteristic nonlinear graph was ob-
tained (i.e., the enantiomeric purity of the aldol remained
essentially constant at 30% ee between 40 and 70% ee of
the catalyst), indicating that the enantiomeric purity of the
proline in solution was mainly governed by the formation of
a eutectic mixture. On the other hand, the use of the pre-
formed 1:1 proline–thiourea complex as the catalyst led to a
linear graph, pointing towards the complete solubilization of
proline (Figure 3). We were utterly surprised to find, howev-
er, that completely different results were obtained for 4-bro-
mobenzaldehyde 2b (Figure 4). In this case, the addition of
the thiourea co-catalyst only marginally changed the shape
thiourea. The experiments without thiourea were performed with
30 mol% of proline.
a
presence of Schreinerꢀs thiourea 1 (or of thiourea 3) by
1
means of H NMR spectroscopy, at the concentrations used
in the actual aldol reactions. In the absence of acetone, and
starting from a 1:1 thiourea/proline mixture, no signals cor-
responding to proline could be observed in the solution.[25]
On the other hand, in the toluene/acetone mixture two pro-
line-derived species could be observed, in an approximate
3:1 ratio (Figure 5, top). The minor compound was identi-
fied as the oxazolidinone 5, and the major compound as the
open zwitterionic iminium species 6, by comparison of their
spectral data (chemical shift and multiplicities of the C2- and
C5-proline protons) with those reported by Seebach for the
corresponding
cyclohexanone-derived
species
in
1144
www.chemeurj.org
ꢆ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 1142 – 1148

Organocatalysis
COMMUNICATION
1
lution, and H NMR spectra were recorded after a few mi-
nutes. In the case of 2a, some new species were clearly seen,
albeit in minor quantities, and, more interestingly, the total
amount of proline solubilization was higher than 87%
(Figure 6, top). If we assume that the new signals also corre-
spond to proline-derived compounds (see below), a practi-
cally complete solubilization of the initial proline has taken
place.
The new species appearing in Figure 6 (top) was identified
as the diastereomeric pair (cis/trans) of the well-known “par-
asitic” bicyclic compound 7 arising from the 1,3-dipolar cy-
cloaddition between the azomethine ylide species derived
from the proline–aldehyde iminium adduct and free aldehy-
de.[5,6b] This identification is also supported by the corre-
sponding ESI-MS spectrum (in nondeuterated solvents),
which exhibits in the positive mode an enhanced intensity
for the peak at m/z 156 and a new peak at m/z 356 corre-
sponding to 7+H.[28] No traces of the oxazolidinone precur-
sor derived from proline and 2a,[6b] or of the corresponding
iminium species, were present in the NMR spectrum, al-
though a relatively weak peak at m/z 205, corresponding to
the protonated azomethyne ylide intermediate (8+H+; see
Scheme 2), was present in the ESI-MS spectrum (positive
mode). Contrary to this, in the case of 2b, no new signals
were apparent and the solubilization percentage of proline
remained essentially constant (60%, Figure 6, bottom).
Figure 5. Top: 1H NMR spectrum (300 MHz) of a solution of l-proline
(0.025 mmolmLÀ1
)
and of thiourea
[D8]toluene/[D6]acetone, after 2 h of stirring at room temperature.
Bottom: 1H NMR spectrum (300 MHz) of
solution of l-proline
(0.025 mmolmLÀ1 (0.012 mmolmLÀ1
and of thiourea in 5:1
1
(0.025 mmolmLÀ1
)
in 5:1
a
)
1
)
[D8]toluene/[D6]acetone, after 2 h of stirring at room temperature. See
Scheme 2 for the structures of compounds 5 and 6.
[D6]benzene.[6a,26] We assume that both compounds are com-
plexed with the thiourea, the NH protons of which exhibited
a low-field shift indicative of hydrogen-bonding.[4,8] The esti-
mated solubilization percentage of proline was 65%, accord-
ing to the thiourea/ACHTUNGTRENNUNG(5+6) ratio calculated from the integrat-
ed 1H NMR spectrum. Further support for this structural
identification was provided by an ESI-MS spectrum of a 1:1
proline–thiourea mixture in 5:1 toluene–acetone; in the pos-
itive mode a signal (m/z 156) for 5+H (or 6+H) was ob-
served. In the negative ESI mode, signals (m/z 499, m/z 999)
for the dissociated thiourea 1 and for its dimer with another
molecule of thiourea could be identified.[27] On the other
hand, no peaks ascribable to free proline or to a proline–
thioACHTUNGTRENNUNGurea complex could be found in the ESI-MS spectra. It
is also interesting to observe that when the thiourea/proline
ratio was decreased to 1:2, the total proline solubilization
percentage diminished only slightly to 57%, but the imini-
um/oxazolidinone ratio dropped from about 3:1 to about
1.5:1 (Figure 5, bottom). It is therefore clear that the
amount of iminium zwitterionic species 6 increases with the
amount of thiourea present in the medium. This is in com-
plete accordance with the results obtained by Seebach, who
observed that the presence of acids of decreasing pKa gener-
ated increasing amounts of iminium derivative from the oxa-
zolidinone.[6a]
Figure 6. Top: 1H NMR spectrum (300 MHz) of a solution of l-proline
(0.025 mmolmLÀ1
)
and of thiourea
1
(0.025 mmolmLÀ1
)
in 5:1
[D8]toluene/[D6]acetone, after 2 h of stirring at room temperature, and
addition of aldehyde 2a (1 equivalent with respect to proline). Bottom:
1H NMR spectrum (300 MHz) of
a
solution of l-proline
(0.025 mmolmLÀ1
)
and of thiourea
1
(0.025 mmolmLÀ1
)
in 5:1
[D8]toluene/[D6]acetone, after 2 h of stirring at room temperature, and
addition of aldehyde 2b (1 equiv with respect to proline). See Scheme 2
for the structures of compounds 5, 6, and 7.
Next, aldehydes 2a and 2b (one molar equivalent with re-
spect to proline) were added to the 1:1 proline–thiourea so-
Chem. Eur. J. 2010, 16, 1142 – 1148
ꢆ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1145

A. Moyano, R. Rios et al.
part of the “catalytic” effect of
the thiourea in the aldol reac-
tion can therefore be ascribed
to the formation of the ace-
tone-derived
species
5+6,
which are the direct precursors
of the key enamine–aldehyde
intermediate 9[16] in solution.
On the other hand, the present
experiments do not rule out the
possibility of a hydrogen-bond-
ing stabilization of the transi-
tion state by the thiourea co-
catalyst (see Figure 1, right), as
previously proposed by Reis
et al.[4] and by us.[5]
In summary, we have dis-
closed, from a study of nonlin-
ear effects in the proline–thio-
urea-catalyzed aldol reaction
between acetone and 4-substi-
tuted benzaldehydes in dry tol-
uene, an unprecedented de-
pendence of these effects on
the nature of the aromatic alde-
hyde. These results strongly
suggest that the main role of
the thiourea co-catalyst (at
least in hydrocarbon solvents)
is not that of generating a solu-
ble proline–thiourea hydrogen-
Scheme 2. Proposed mechanism for the generation and fate of proline-derived soluble species in the proline–
thiourea-catalyzed intermolecular aldol reaction in hydrocarbon solvents.
bonded species,
a conclusion
that is supported both by
1
These simple NMR and ESI-MS experiments are therefore
completely consistent with the results of the study of nonlin-
ear effects. Although more work is clearly needed at this
point, we suggest, on the basis of the data presented above,
a possible explanation of the role of thiourea in the proline-
catalyzed aldol reactions studied by Reis et al.[4] and by us[5]
that can be summarized as follows (Scheme 2).
The addition of one molar equivalent of thiourea to a sus-
pension of proline in toluene (or hexane), in the presence of
excess ketone, leads to the partial solubilization of proline
as an approximate 1:3 mixture of the bicyclic oxazolidinone
5 and its open zwitterionic counterpart 6. The equilibrium
between solid proline and these soluble proline-derived spe-
cies can be further shifted to the right upon addition of the
aldehyde, especially in the case of strongly electron-deficient
aldehydes, such as 4-nitrobenzaldehyde (2a), leading to the
formation of the bicyclic 1-oxapyrrolizidine 7 by a multistep
process that involves conversion into the aldehyde-derived
oxazolidinone,[29] decarboxylation of the corresponding zwit-
terionic monocyclic species, and 1,3-dipolar cycloaddition of
the resulting azomethyne ylide 8 with another molecule of
aldehyde 2a. Although this process reduces the amount of
proline, the concentration of the zwitterionic species 6 is in-
creased (see Supporting Information). A non-negligeable
simple H NMR and ESI-MS experiments. In fact, the hy-
drogen-bonded interaction between proline and thiourea ap-
pears to be operative only in the presence of the ketone re-
agent, promoting both the formation of the ketone–oxazoli-
dinone intermediate 5 and its equilibration with the “open”
zwitterionic structure 6. These species, after conversion to
the proline–ketone enamine, could in fact be the true cata-
lysts of the aldol reaction in a mechanistic cycle similar to
that proposed by Seebach[6a] and Vilarrasa,[6c] that runs in
the absence of water. The nature of the aromatic aldehyde
is also very relevant for the extent of proline solubilization,
through its partial conversion to the “parasitic” 1-oxapyrro-
lizidine species. Further experimental and theoretical work
on these issues is currently underway in our laboratory.
Experimental Section
Experimental procedure for the study of nonlinear effects
1) In the absence of thiourea: An ordinary vial equipped with a magnetic
stirring bar was charged with dry proline (6 mg, 0.05 mmol for 2a; 17 mg,
0.15 mmol for 2b) and 5:1 toluene-acetone (2 mL), and stirred for 2 h at
room temperature. Then, the aldehyde 2a/b (0.50 mmol) was added in
one portion. After 20 h (2a) or 10 days (2b), the conversion was mea-
sured by NMR spectroscopy of an aliquot of the reaction mixture and
1146
www.chemeurj.org
ꢆ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 1142 – 1148

Organocatalysis
COMMUNICATION
the enantiomeric purity of the aldol 4a/b was determined by chiral
15101; c) C. Isart, J. Burꢅs, J. Vilarrasa, Tetrahedron Lett. 2008, 49,
5414–5418.
HPLC.[23]
[7] For reviews, see: a) P. Melchiorre, M. Marigo, A. Carlone, G. Barto-
li, Angew. Chem. Int. Ed. 2008, 120, 6232–6265; Angew. Chem. Int.
Ed. 2008, 47, 6138–6171, especially section 5.1.; b) M. Gruttadauria,
F. Giacalone, R. Noto, Adv. Synth. Catal. 2009, 351, 33–57, sec-
tions 2.2–2.4. For recent examples, see: c) Y. Hayashi, T. Urushima,
S. Aratake, T. Okano, K. Obi, Org. Lett. 2008, 10, 21–24; d) D.
Font, S. Salayero, A. Bastero, C. Jimeno, M. A. Pericꢃs, Org. Lett.
2008, 10, 337–340; e) L. Zu, H. Xie, H. Li, J. Wang, W. Wang, Org.
Lett. 2008, 10, 1211–1214; f) G. Guillena, M. del Carmen Hita, C.
Nꢇjera, S. F. Viꢁzquez, J. Org. Chem. 2008, 73, 5933–5943; g) F.
Giacalone, M. Gruttadauria, P. Lo Meo, S. Riela, R. Noto, Adv.
Synth. Catal. 2008, 350, 2747–2760; h) H. Yang, R. G. Carter, Org.
Lett. 2008, 10, 4649–4652; i) J.-R. Chen, X.-L. An, X.-Y. Zhu, X.-F.
Wang, W.-J. Xiao, J. Org. Chem. 2008, 73, 6006–6009; j) J. N. Moor-
thy, S. Saha, Eur. J. Org. Chem. 2009, 739–748; k) H. Yang, R. G.
Carter, J. Org. Chem. 2009, 74, 2246–2249; l) M. Lombardo, S.
Easwar, F. Pasi, C. Trombini, Adv. Synth. Catal. 2009, 351, 276–282.
[8] For recent studies on the interaction of hydrogen-bond donors with
carboxylic acids, see: a) F. M. MuÇiz, V. A. Montero, A. L. Fuentes
de Arriba, L. Simꢁn, C. Raposo, J. R. Morꢇn, Tetrahedron Lett.
2008, 49, 5050–5052; b) Y. Willener, K. M. Joly, C. J. Moody,
J. H. R. Tucker, J. Org. Chem. 2008, 73, 1225–1233.
b) In the presence of thiourea: An ordinary vial equipped with a magnetic
stirring bar was charged with dry proline (6 mg, 0.05 mmol), thiourea 3
(22 mg, 0.05 mmol), and 5:1 toluene–acetone (2 mL), and stirred for 2 h
at room temperature. Then, the aldehyde 2a/b (0.50 mmol) was added in
one portion. After 20 h, the conversion was measured by NMR spectros-
copy of an aliquot of the reaction mixture and the enantiomeric purity of
the aldol 4a/b was determined by chiral HPLC.[23]
Acknowledgements
We thank the Spanish Ministry of Science and Education for financial
support (Project AYA2006-15648-C02-01). N.E.-H. is grateful to the
Spanish Ministry of Foreign Affairs and Cooperation for a pre-doctoral
fellowship (MAEC-AECID Program). We also thank Dr. Jordi Burꢅs
(DQO-UB) and Francisco Cꢇrdenas (SCT-UB) for their invaluable help
and advice.
Keywords: aldol
reaction
·
nonlinear
effects
·
organocatalysis · proline · thiourea
[9] Y. Zhou, Z, Shan, J. Org. Chem. 2006, 71, 9510–9512.
[10] T. Mandal, C.-G. Zhao, Angew. Chem. 2008, 120, 7828–7831;
Angew. Chem. Int. Ed. 2008, 47, 7714–7717.
[11] M. L. Clarke, J. A. Fuentes, Angew. Chem. 2007, 119, 948–951;
Angew. Chem. Int. Ed. 2007, 46, 930–933.
[12] For the beneficial effects of hydrogen-bond-donating co-catalysts on
the intermolecular aldehyde-enone Michael reaction catalyzed by a
MacMillan oxazolidinone, see: T. J. Peelen, Y. Chi, S. H. Gellman, J.
Am. Chem. Soc. 2005, 127, 11598–11599.
[1] a) R. Kane, J. Prakt. Chem. 1838, 15, 129; b) C. A. Wurtz, Compt.
Rend. 1872, 178, 1361–1367; c) A. T. Nielsen, W. J. Houlihan, Org.
React. 1968, 16, 1–438; d) T. Mukaiyama, Org. React. 1982, 28, 203–
331; e) H. J. M. Gijsen, L. Qiao, W. Fitz, C.-H. Wong, Chem. Rev.
1996, 96, 443–473.
[2] a) B. List, R. A. Lerner, C. F. Barbas III, J. Am. Chem. Soc. 2000,
122, 2395–2396; b) W. Notz, B. List, J. Am. Chem. Soc. 2000, 122,
7386–7387; c) K. Sakthivel, W. Notz, T. Bui, C. F. Barbas III, J. Am.
Chem. Soc. 2001, 123, 5260–5267; d) B. List, P. Pojarliev, C. Costello,
Org. Lett. 2001, 3, 573–576.
[3] See, among others, the following reviews: a) Modern Adol Reac-
tions, Vol. 1–2 (Ed.: R. Mahrwald), Wiley-VCH, Weinheim, 2004;
b) B. Schetter, R. Mahrwald, Angew. Chem. 2006, 118, 7668–7687;
Angew. Chem. Int. Ed. 2006, 45, 7506–7525; c) G. Guillena, C.
Nꢇjera, D. J. Ramꢁn, Tetrahedron: Asymmetry 2007, 18, 2249–2293;
d) L. M. Geary, P. G. Hultin, Tetrahedron: Asymmetry 2009, 20, 131–
173. For some recent papers on organocatalytic asymmetric aldol re-
actions, see: e) S. Luo, H. Xu, L. Zhang, J. Li, J.-P. Cheng, Org. Lett.
2008, 10, 653–656; f) Y. Hayashi, T. Itoh, S. Aratake, H. Ishikawa,
Angew. Chem. 2008, 120, 2112–2114; Angew. Chem. Int. Ed. 2008,
47, 2082–2084; g) J. Liu, Z. Yang, Z. Wang, F. Wang, X. Chen, X.
Liu, X. Feng, Z. Su, C. Hu, J. Am. Chem. Soc. 2008, 130, 5654–
5655; h) D. Hazelard, H. Ishikawa, D. Hashizume, H. Koshino, Y.
Hayashi, Org. Lett. 2008, 10, 1445–1448; i) S. S. V. Ramasastry, K.
Albertshofer, N. Utsumi, C. F. Barbas III, Org. Lett. 2008, 10, 1621–
1624; j) S. Luo, H. Xu, L. Chen, J.-P. Cheng, Org. Lett. 2008, 10,
1775–1778; k) C. L. Chandler, B. List, J. Am. Chem. Soc. 2008, 130,
6737–6739; l) Y. Hayashi, S. Samanta, T. Itoh, H. Ishikawa, Org.
Lett. 2008, 10, 5581–5583; m) K. Nakayama, K. Maruoka, J. Am.
Chem. Soc. 2008, 130, 17666–17667; n) X. Wu, X. Ma, Z. Ye, S.
Qian, G. Zhao, Adv. Synth. Catal. 2009, 351, 158–162; o) J. Li, S.
Luo, J.-P. Cheng, J. Org. Chem. 2009, 74, 1747–1750.
[13] M. M. Vasbinder, J. E. Imbriglio, S. J. Miller, Tetrahedron 2006, 62,
11450–11459.
[14] A. Kotke, P. R. Schreiner, Tetrahedron 2005, 61, 434–439.
[15] a) S. Bahmanyar, K. N. Houk, J. Am. Chem. Soc. 2001, 123, 11273–
11283; b) L. Hoang, S. Bahmanyar, K. N. Houk, B. List, J. Am.
Chem. Soc. 2003, 125, 16–17; c) C. Allemann, R. Gordillo, F. R.
Clemente, P. H.-Y. Cheong, K. N. Houk, Acc. Chem. Res. 2004, 37,
558–569; d) C. F. Barbas III, Angew. Chem. 2008, 120, 44–50;
Angew. Chem. Int. Ed. 2008, 47, 42–47.
[16] S. L. Poe, A. R. Bogdan, B. P. Mason, J. L. Steinbacher, S. M.
Opalka, D. T. McQuade, J. Org. Chem. 2009, 74, 1574–1580.
[17] a) S. P. Mathew, H. Iwamura, D. G. Blackmond, Angew. Chem. 2004,
116, 3379–3383; Angew. Chem. Int. Ed. 2004, 43, 3317–3321; b) H.
Iwamura, S. P. Mathew, D. G. Blackmond, J. Am. Chem. Soc. 2004,
126, 11770–11771.
[18] For a recent authoritative review on nonlinear effects in asymmetric
catalysis, that includes a section on nonlinear effects through partial
solubility of the catalyst, see: a) T. Satyanarayana, S. Abraham,
H. B. Kagan, Angew. Chem. 2009, 121, 464–503; Angew. Chem. Int.
Ed. 2009, 48, 456–494. For a short review on the solubility behavior
of proline and its implications in organocatalysis, see: b) R. M. Kel-
logg, Angew. Chem. 2007, 119, 498–502; Angew. Chem. Int. Ed.
2007, 46, 494–497.
[19] Y. Hayashi, M. Matsuzawa, J. Yamaguchi, S. Yonehara, Y. Matsumo-
to, M. Shoji, D. Hashizume, H. Koshino, Angew. Chem. 2006, 118,
4709–4713; Angew. Chem. Int. Ed. 2006, 45, 4593–4597.
[20] a) M. Klussmann, H. Iwamura, S. P. Matthew, D. H. Wells, Jr., U.
Pandya, A. Armstrong, D. G. Blackmond, Nature 2006, 441, 621–
623; b) M. Klussmann, T. Izumi, A. J. P. White, A. Armstrong, D. G.
Blackmond, J. Am. Chem. Soc. 2007, 129, 7657–7660.
[4] ꢈ. Reis, S. Eymur, B. Reis, A. S. Demir, Chem. Commun. 2009,
1088–1090; please note that the absolute configurations of the aldol
adducts depicted in this paper are systematically opposite to the
actual ones.
[21] For the study of nonlinear effects in a proline-catalyzed Mannich re-
action, see: G. Valero, A.-N. Balaguer, A. Moyano, R. Rios, Tetrahe-
dron Lett. 2008, 49, 6559–6562.
[22] M. Klussmann, S. P. Mathew, H. Iwamura, D. H. Wells, Jr., A. Arm-
strong, D. G. Blackmond, Angew. Chem. 2006, 118, 8157–8160;
Angew. Chem. Int. Ed. 2006, 45, 7989–7992.
[5] X. Companyꢁ, G. Valero, L. Crovetto, A. Moyano, R. Rios, Chem.
Eur. J. 2009, 15, 6564–6568.
[6] a) D. Seebach, A. K. Beck, D. M. Badine, M. Limbach, A. Eschen-
moser, A. M. Treasurywala, R. Hobi, W. Prikoszovich, B. Linder,
Helv. Chim. Acta 2007, 90, 425–471; b) N. Zotova, A. Franzke, A.
Armstrong, D. G. Blackmond, J. Am. Chem. Soc. 2007, 129, 15100–
Chem. Eur. J. 2010, 16, 1142 – 1148
ꢆ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1147

A. Moyano, R. Rios et al.
[23] Conditions for the chiral HPLC analyses. 4a: Chiralcel IC column,
80:20 hexane/2-propanol, F=1 mLminÀ1, l=254 nm, retention time
(R)=9.13 min, (S)=9.54 min. 4b: Chiralcelꢉ IC column, 99:1
nally, a solution of oxazolidinone [D6]5 was independently generated
by exchange from SolPro (obtained from l-proline and pivalalde-
hyde in benzene by adaptation of the procedure described by Vilar-
rasa et al.[6c]) with [D6]acetone; upon addition of thiourea 1 to this
solution, equilibration of 5 with the zwitterionic species 6 was ob-
served (see the Supporting Information).
hexane/2-propanol, F=1 mLminÀ1
(R)=57.4 min, (S)=60.8 min.
, l=262 nm, retention time
[24] a) Y. K. Chen, A. M. Costa, P. J. Walsh, J. Am. Chem. Soc. 2001,
123, 5378–5379; b) P. C. Bulman Page, S. M. Allin, S. J. Maddocks,
M. R. Elsegood, J. Chem. Soc. Perkin Trans. 1 2002, 2827–2832.
[25] Reis et al.[4] have observed by 1H NMR the partial solubilization of
proline by Schreinerꢀs thiourea 1 in chloroform, a solvent that is
much more polar than those (hexane, toluene) used by them in the
aldol reactions.
[27] For the use of ESI-MS in the characterization of key intermediates
in organocatalytic reactions, see: W. Schrader, P. P. Handayani, J.
Zhou, B. List, Angew. Chem. 2009, 121, 1491–1494; Angew. Chem.
Int. Ed. 2009, 48, 1463–1466, and references therein.
[28] It is worth noting that in the absence of acetone, the formation of 7
from proline, thiourea 1 (or 3) and aldehyde 2a in [D8]toluene takes
place at a much lower rate; see reference [5].
[29] The ready interconversion between aldehyde- and ketone-derived
proline oxazolidinones has been nicely demonstrated by Vilarrasa
and co-workers; see reference [6c].
[26] Substitution of [D6]acetone by acetone led to a 1:3.8 mixture of the
nondeuterated species 5 and 6. On the other hand, the 1H NMR
spectrum of
a 1:1 proline/thiourea 1 solution (0.025m) in 5:1
[D8]toluene/pivalaldehyde showed a completely different set of sig-
nals, corresponding to the formation of the expected bicyclic oxazo-
lidinone (SolPro)[6c] as the sole soluble proline-derived species. Fi-
Received: September 29, 2009
Published online: December 22, 2009
1148
www.chemeurj.org
ꢆ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 1142 – 1148