Direction of kinetically versus thermodynamically controlled organocatalysis and its application in chemoenzymatic synthesis

Article abstract of DOI:10.1002/anie.201008042

The catalyst loading in the narrow range between 0.5 and 10 mol% determines whether an enantioselective organocatalytic aldol reaction proceeds under kinetic or thermodynamic control; high conversions and ee values can be achieved with low catalyst loadin

Full text of DOI:10.1002/anie.201008042

Communications
DOI: 10.1002/anie.201008042
Organocatalysis
Direction of Kinetically versus Thermodynamically Controlled
Organocatalysis and Its Application in Chemoenzymatic Synthesis**
Giuseppe Rulli, Nongnaphat Duangdee, Katrin Baer, Werner Hummel, Albrecht Berkessel,* and
Harald Grçger*
Recent developments in the field of (enantioselective)
organocatalysis have established it as a broadly applicable
and efficient synthetic tool for the preparation of many types
of enantiomerically enriched and enantiomerically pure
molecules.[1] In these syntheses, organocatalysts are typically
used in amounts of 1 to 20 mol%.[1,2] In general it is assumed
that the enantioselective reactions proceed under kinetic
control when the amount of catalyst used is within this range.
Accordingly, the applied amount of catalyst indicates the
degree of catalyst activity, and the catalyst amount can be
adjusted in order to optimize the reaction rate and the overall
conversion. Although organocatalytic reactions are in general
assumed to be kinetically controlled within this range of
catalyst loadings, it is in principle possible to switch from a
kinetically controlled to a thermodynamically controlled
regime even within this narrow range of catalyst loadings
and within typical reaction times. Herein we report such an
example in which the switch from kinetic to thermodynamic
control occurs through a variation of catalyst loading in a
narrow range between 0.5 and 10 mol%. Since the trans-
formations reported here can be carried out in water, this also
allows new efficient applications for chemoenzymatic one-pot
multistep syntheses in aqueous reaction media.[3]
ature as we had aimed at a combination with enzymatic
syntheses, and in this connection we used a loading of 5 mol%
of the organocatalyst 3.[5] Under these reaction conditions the
aldol reaction of 2 (9 equiv) with 1 proceeded with an
enantioselectivity of 70% ee (Scheme 1).[6]
Scheme 1. Organocatalytic aldol reaction in an organic reaction
medium.
With a view toward chemoenzymatic one-pot syntheses in
aqueous medium we also have been interested in conducting
the aldol reaction in this reaction medium. Accordingly we
tested this transformation in aqueous NaCl. We observed that
with the same catalyst amount (5 mol%), the reaction gave
significantly lower enantioselectivity after 48 h, leading to the
formation of the desired product (S)-4 with only 47% ee
(Figure 1). An even more surprising result was obtained in an
experiment with 10 mol% of the organocatalyst (R,R)-3,
which led to a complete loss of enantioselectivity (0% ee). To
determine the reasons for this drastic decrease of enantiose-
lectivity, we first studied the effect of lower catalyst amounts.
Interestingly, when the catalyst amount was lowered to
1.0 mol%, the enantioselectivity of the reaction continuously
improved (Figure 1). For example, in the presence of
1.0 mol% of the catalyst a high, greatly improved enantiose-
lectivity of 91% ee was achieved at a product-based con-
version of 90% (95% overall conversion). A further increase
of the enantioselectivity up to 93% ee at a product-based
conversion of 92% (95% overall conversion) was obtained
upon further decrease of the catalyst amount to 0.5 mol%.
The surprising significant negative influence of increased
catalyst amounts of 5 and 10 mol% on the enantioselectivity
of the organocatalytic aldol reaction at a reaction time of 48 h
is interesting,[7] since many organocatalytic reactions with
related catalyst systems are carried out with similar or even
higher amounts of catalyst. Thus, reducing the catalyst loading
(which also would be advantageous from an economic
perspective) might be an option for the optimization and
improvement of enantioselectivity of such organocatalytic
As a model reaction we chose the aldol reaction of
acetone (2) with 3-chlorobenzaldehyde (1) in the presence of
the organocatalyst 3, which was developed by Singh et al.[4] In
previous work, we conducted such reactions at room temper-
[*] G. Rulli, K. Baer, Prof. Dr. H. Grçger[+]
Department Chemie und Pharmazie
Universitꢀt Erlangen-Nꢁrnberg
Henkestrasse 42, 91054 Erlangen (Germany)
Prof. Dr. W. Hummel
Institut fꢁr Molekulare Enzymtechnologie der Heinrich-Heine-
Universitꢀt Dꢁsseldorf, Forschungszentrum Jꢁlich
Stetternicher Forst, 52426 Jꢁlich (Germany)
N. Duangdee, Prof. Dr. A. Berkessel
Department fꢁr Chemie, Universitꢀt zu Kçln
Greinstrasse 4, 50939 Kçln (Germany)
E-mail: berkessel@uni-koeln.de
[+] Current address: Fakultꢀt fꢁr Chemie, Universitꢀt Bielefeld
Universitꢀtsstrasse 25, 33615 Bielefeld (Germany)
E-mail: harald.groeger@uni-bielefeld.de
[**] We thank Evonik-Degussa GmbH, Amano Enzymes Inc., and
Oriental Yeast Company Ltd. Japan for generous support with
chemicals and the Deutsche Forschungsgemeinschaft (DFG) for
generous support within the priority programme SPP 1179
“Organokatalyse” (BE 998/11-1, GR 3461/2-1).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201008042.
7944
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 7944 –7947

Figure 1. Effect of catalyst loading on the selectivity and conversion of
the organocatalytic aldol reaction in an aqueous reaction medium.
Figure 2. Dependence of the enantioselectivity of the aldol reaction on
the reaction time.
reactions, which so far proceed with low to moderate
enantioselectivity.
In our search for an explanation of this phenomenon we
considered and evaluated the following three conceivable
possibilities: 1) catalyst impurities, which effect a nonselective
aldol reaction and which play a more significant role at higher
catalyst amounts; 2) concentration dependence of the cata-
lytic effects, such that dimeric or oligomeric complexes of 3
act as the catalytic species; 3) thermodynamic rather than
kinetic control of the organocatalytic aldol reaction, which
results from the rapidly reached reaction equilibrium.
In order to scrutinize explanation option (1), we synthe-
sized the organocatalyst (S,S)-3 by two different routes. Since
we obtained comparable results with both catalyst samples
(which additionally show different degrees of purity), this
explanation was discarded (for details, see the Supporting
Information).[8] The plausibility of explanation options (2)
and (3) was evaluated by investigating the rate of the
formation of product (S)-4 and the change in its enantiomeric
excess as a function of the reaction time. In the case of
explanation option (2), the enantioselectivity should be
independent of the reaction time, whereas in case of
explanation option (3) the enantiomeric excess of product
(S)-4 should be depleted as a result of an initial kinetically
controlled reaction and subsequent formation of the thermo-
dynamic equilibrium induced by the catalyst. Interestingly,
exactly this result was found when we used a high catalyst
amount of 5 mol%: the enantioselectivity decreased signifi-
cantly from 90% ee after 2 min to 47% ee after 48 h
(Figure 2). In contrast, the enantiomeric excess remained
nearly unchanged over a period of 48 h when a low catalyst
amount of 0.5 mol% was used. This indicates kinetic control
of the reaction within this reaction time.
cally controlled reaction at an increased catalyst amount of
5 mol% according to explanation option (3) was also
confirmed by the observed reaction rates (Figure 3). Whereas
in the presence of 0.5 mol% of (R,R)-3 the formation of the
product (S)-4 proceeded with a reaction course typical of
kinetically controlled reactions, the reaction in the presence
of 5 mol% of (R,R)-3 was so rapid that significant product
formation was observed even after a short reaction time
(<30 min) with, for example, 58% product-based conversion
(61% overall conversion) after 2 min. This goes along with
the expectation that thermodynamic control of the reaction is
reached relatively rapidly. A further, major advantage from
the perspective of process technology when this reaction is
conducted with lower amounts of catalyst is the slowing down
of side-product formation (caused, for example, by elimina-
tion of water from the aldol adduct). In our case, side products
accounted for a proportion of 16% after 48 h when 5 mol%
of catalyst (R,R)-3 was used (Figure 3). Accordingly, in the
presence of 0.5 mol% of the catalyst (R,R)-3, the product-
based conversion could be enhanced up to 92%, at a
decreased side-product proportion of < 5%.
Carrying out the aldol reaction in water under kinetic
control with low amounts of the catalyst also offers interesting
perspectives for its combination with biotransformations in
one-pot multistep processes in an aqueous reaction medium.
As a “proof of concept” for such a process, we linked the
organocatalytic aldol reaction described above with a sub-
sequent enzymatic reduction of the formed aldol adduct (R)-4
(Scheme 2). After addition of R- or S-enantioselective alcohol
dehydrogenases (ADH), the respective 1,3-diols of type 5
were obtained with excellent diastereoselectivities and enan-
tioselectivities. For the chemoenzymatic one-pot synthesis of
the anti diastereomer (1R,3S)-5 in the presence of the
This change from a kinetically controlled reaction at
0.5 mol% of (R,R)-3 to a (predominantly) thermodynami-
Angew. Chem. Int. Ed. 2011, 50, 7944 –7947
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7945

Communications
Scheme 2. Combination of organocatalysis and biocatalysis in a one-
pot synthesis in an aqueous reaction medium.
narrow, synthetically widely used range of catalyst amounts.
Since such syntheses can furthermore be carried out in water,
applications to chemoenzymatic one-pot multistep synthesis
in aqueous reaction medium are possible. By means of such a
combination with a stereoselective biocatalytic reduction, a
one-pot synthesis of 1,3-diols of type 5 has been realized with
diastereomeric ratios of > 25:1 and excellent enantioselectiv-
ities of 99% ee.
Figure 3. Dependence of the conversion of the aldol reaction on the
reaction time.
Received: December 20, 2010
Revised: April 8, 2011
Published online: July 8, 2011
organocatalyst (S,S)-3 and the S-alcohol dehydrogenase from
Rhodococcus sp.,
a product-based conversion of 89%
(> 95% overall conversion) at a diastereomeric ratio of
> 25:1 (anti/syn) and an enantioselectivity of 99% ee was
achieved (Scheme 2). The corresponding syn diastereomer
(1R,3R)-5 was obtained in a one-pot synthesis in the presence
of organocatalyst (S,S)-3 and the R-alcohol dehydrogenase
from Lactobacillus kefir with a product-based conversion of
72% (> 95% overall conversion) at a diastereomeric ratio of
> 25:1 (syn/anti) and with an enantioselectivity of 99% ee. To
the best of our knowledge, these are the first examples of a
combination of an enantioselective organocatalytic reaction
with a subsequent biotransformation, which proceed as one-
pot syntheses and completely in aqueous reaction medium
(Scheme 2).[9]
In conclusion, we report an enantioselective organocata-
lytic transformation in which kinetic versus thermodynamic
control has been directed through variation of the catalyst
amount in a range of 0.5 to 10 mol%. Notably, high
enantioselectivities were obtained when low catalyst amounts
were used. Although the change from kinetic control to
thermodynamic control with increasing amounts of catalyst is
well known, it is, however, noteworthy that this change has
been observed in a very significant fashion within such a
Keywords: aldol reactions · chemoenzymatic synthesis ·
organocatalysis · synthetic methods
.
[1] Overviews: a) A. Berkessel, H. Grçger, Asymmetric Organo-
catalysis, Wiley-VCH, Weinheim, 2005; b) B. List, J. W. Yang,
Science 2006, 313, 1584 – 1586; c) Enantioselective Organocatal-
ysis (Ed.: P. I. Dalko), Wiley-VCH, Weinheim, 2007; d) P.
Melchiorre, M. Marigo, A. Carlone, G. Bartoli, Angew. Chem.
2008, 120, 6232 – 6265; Angew. Chem. Int. Ed. 2008, 47, 6138 –
6171.
[2] In some cases organocatalytic reactions have been carried out
successfully also in the presence of very low catalyst amounts of
< 1 mol%; for selected examples, see: a) T. Kano, O. Tokuda, K.
Maruoka, Tetrahedron Lett. 2006, 47, 7423 – 7426; b) M. Lom-
bardo, S. Easwar, F. Pasi, C. Trombini, Adv. Synth. Catal. 2009,
351, 276 – 282; c) F. Rodrꢀguez, C. Bolm, J. Org. Chem. 2006, 71,
2888 – 2891; d) M. Rueping, A. P. Antonchick, T. Theissmann,
Angew. Chem. 2006, 118, 6903 – 6907; Angew. Chem. Int. Ed. 2006,
45, 6751 – 6755; e) M. Wiesner, G. Upert, G. Angelici, H.
Wennemers, J. Am. Chem. Soc. 2010, 132, 6 – 7; f) S. Zhu, S. Yu,
D. Ma, Angew. Chem. 2008, 120, 555 – 558; Angew. Chem. Int. Ed.
2008, 47, 545 – 548.
7946 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 7944 –7947

[3] Reviews about chemoenzymatic one-pot syntheses: a) A. Brug-
gink, R. Schoevaart, T. Kieboom, Org. Process Res. Dev. 2003, 7,
622 – 640; b) H. Pellissier, Tetrahedron 2008, 64, 1563 – 1601.
[4] a) M. Raj, Vishnumaya, S. K. Ginotra, V. K. Singh, Org. Lett.
2006, 8, 4097 – 4099; b) V. Maya, M. Raj, V. K. Singh, Org. Lett.
2007, 9, 2593 – 2595; c) M. Raj, V. Maya, V. K. Singh, J. Org.
Chem. 2009, 74, 4289 – 4297; d) Review: M. Raj, V. K. Singh,
Chem. Commun. 2009, 6687 – 6703.
[5] K. Baer, M. Kraußer, E. Burda, W. Hummel, A. Berkessel, H.
Grçger, Angew. Chem. 2009, 121, 9519 – 9522; Angew. Chem. Int.
Ed. 2009, 48, 9355 – 9358.
[6] With respect to expanding the substrate range, in this work we
used 3-chlorobenzaldehyde (1) as an aldehyde component instead
of 4-chlorobenzaldehyde which was used in our previous work
(see Ref. [5]).
observed (e.g., for the aldol reaction with 3-chlorobenzaldehyde
at room temperature: 0.5 mol%: 97% ee after 5 h; 10 mol%:
86% ee after 2 h). An influence of the thermodynamic control on
the asymmetric amplification of a proline-catalyzed aldol reaction
has been additionally reported in: M. Klussmann, H. Iwamura,
S. P. Mathew, D. H. Wells, U. Pandya, A. Armstrong, D. G.
Blackmond, Nature 2006, 441, 621 – 623.
[8] In addition, organocatalysts of this type have been mostly
postulated as monomeric species so far (see Ref. [4c]), which
also is in contrast to explanation option (2). An example of a
dimeric organocatalyst is, however, described in: S. Rho, S. H. Oh,
J. W. Lee, J. Y. Lee, J. Chin, C. E. Song, Chem. Commun. 2008,
1208 – 1210.
[9] The racemization of a-amino esters with achiral aldehydes as
organocatalysts in chemoenzymatic dynamic-kinetic resolutions
in aqueous reaction medium has already been reported. For
pioneering work, see: a) T. Riermeier, U. Dingerdissen, P. Gross,
W. Holla, M. Beller, D. Schichl, DE19955283, 2001; b) S.-T. Chen,
W.-H. Huang, K.-T. Wang, J. Org. Chem. 1994, 59, 7580 – 7581.
[7] This result is in accordance with the work of Singh et al. (see
Ref. [4]), who reported higher enantioselectivities at a decreased
amount of 0.5 mol% of organocatalyst (S,S)-3, although such a
significant discrepancy in the enantioselectivities was not
Angew. Chem. Int. Ed. 2011, 50, 7944 –7947
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7947