Active conformation in amine-thiourea bifunctional organocatalysis preformed by catalyst aggregation

Article abstract of DOI:10.1002/chem.201102701

Self-activation: Takemoto's catalyst gains access to its active conformation by equilibrating between its hydrogen-bonded intra- and intermolecular interactions in apolar aprotic solvents. By destabilization of the inactive monomeric conformations, the extended anti-anti thiourea conformation is preformed in the assembly. On leaving the assembly, this transient conformation has a structural preference to become a catalytically active monomeric species that has the potency for dual activation (see scheme).

Full text of DOI:10.1002/chem.201102701

DOI: 10.1002/chem.201102701
Active Conformation in Amine–Thiourea Bifunctional Organocatalysis
Preformed by Catalyst Aggregation
[
a]
[a]
[b]
[b]
Gꢀbor Tꢀrkꢀnyi,* Pꢁter Kirꢀly, Tibor Soꢂs, and Szilꢀrd Varga
Dedicated to Professor Gꢀbor Pꢀlinkꢀs on the occasion of his 70th birthday
Bifunctionality of amine–thiourea organocatalysts carries
the possibility of intra- and intermolecular interactions via
hydrogen bonding. The key to their function is the simulta-
neous presence of both hydrogen-bond donor and acceptor
sites and the ability to cooperatively and reversibly control
member of the family of successful amine–thiourea organo-
catalysts with numerous examples proving its potential in
[11]
enantioselective CÀC bond forming chemical reactions.
The reaction mechanism studies of 1 rely on the solid-state
conformation of the catalyst available from the X-ray dif-
[1]
[10b]
[12]
their non-covalent interactions. Temperature dependent
solution NMR spectroscopy offers the methodology to
fraction data of the racemate
or its urea analogue. To
the best of our knowledge the solution conformation of the
enantiopure catalyst 1 has not yet been determined. During
the double activation process suggested in the reaction
mechanism studies a well defined (catalytically active) con-
former of the catalyst capable of interacting with both sub-
[2]
probe these reversible interactions at the atomic level. We
have shown for some highly enantioselective thiourea modi-
fied 9-epi-Cinchonas that the hydrogen bonded interactions
of enantiopure organocatalysts favor the formation of ho-
mochiral dimers which are assembled from different confor-
[13]
strates is considered (Figure 1). Because of the difficulties
[3]
[4]
mations. Assessing self-aggregation and determining the
three dimensional solution structure of the self-assembly is
an important step forward in understanding the molecular
recognition properties of bifunctional organocatalytic sys-
[5]
tems. Besides the development of rigid and self-association
[
6]
free catalytic systems, understanding the success of strong-
ly aggregating yet highly enantioselective bifunctional cata-
lysts is a challenging goal.
Recently we found that self-assemblies of thiourea-modi-
fied Cinchona catalysts influence the thermodynamics of the
system by lowering the activation energy of the conforma-
[7]
tion changes of the catalyst. Originally, this effect was at-
[8]
tributed to the CH–p interaction of the quinoline moiety,
[9]
a salient structural feature of Cinchona alkaloids. Here we
[
10]
report the evidence that Takemotoꢀs catalyst
(1)—not
Figure 1. Examples for self-quenched and active monomeric conforma-
tions.
bearing the quinoline part—also undergoes a specific self-as-
sociation process. The interplay between intra- and intermo-
lecular hydrogen bonding of the catalyst offers a so far un-
raveled mechanism of “catalyst self-activation” via induced
change in the conformation. Catalyst 1 is an established
in the direct spectroscopic observation of short living molec-
ular states such as the catalytically active conformation the
experimental evidence for strongly aggregating catalysts has
for long been delayed. The present work was inspired by the
premise that the knowledge of experimentally determined
ground-state conformations in solution helps the completion
of the catalytic cycle that considers catalyst self-associa-
[
a] Dr. G. Tꢁrkꢁnyi, Dr. P. Kirꢁly
Department of NMR Spectroscopy
Institute of Structural Chemistry, Hungarian Academy of Sciences
Pusztaszeri fflt 59-67, Budapest, 1025 (Hungary)
Fax : (+36)1-438-1107
[14]
tion.
E-mail: gtarkanyi@chemres.hu
We determined and interpreted the thermodynamically
[
b] Dr. T. Soós, S. Varga
stable conformations of the pure catalyst 1 in solution by an-
Department of Synthetic Organic Chemistry
Institute of Biomolecular Chemistry, Hungarian Academy of Sciences
Pusztaszeri fflt 59-67, Budapest, 1025 (Hungary)
1
alyzing the H NMR spectra in apolar aprotic solvents
(
Scheme 1). The method relies on the success of limiting the
chemical exchange between the different molecular (mono-
meric, dimeric) states by significantly lowering the tempera-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201102701.
1918
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1918 – 1922

COMMUNICATION
1
Scheme 1. Proposed role of catalyst-aggregation in the catalytic cycle.
The active monomeric conformation m-III is preformed as d-III in the
self-assembly. Thiourea conformation is defined by the S-C-N-H torsion
angle (syn 08, anti 1808).
Figure 2. Low temperature H NMR (163 K) of enantiopure (bottom)
and racemic (1) (top). The notations refer to the N-H protons (A and B)
in conformations (I, II, III) of the monomeric (m) and homo- and hetero-
chiral dimeric (d, D) states.
ture. Under these conditions the individual molecular
states—which also exist around room temperature but their
intramolecular hydrogen bonding. As a reinforcement of its
capacity towards non-covalent intermolecular interactions
the monomeric m-III conformer is not a populated ground
state conformation of 1. Instead, we identified this confor-
[2]
signals are time-averaged—can be analyzed. Figure 2 dis-
plays the results suggesting a remarkable spectral and struc-
tural complexity for such a small sized molecule. There are
four sets of signals detected for the enantiopure form of the
catalyst corresponding to two monomeric (m) and an asym-
metric homochiral dimeric (d) state. The three abundant
conformations (I, II and III) are markedly different and
prefer either intramolecular or intermolecular hydrogen
bonding (Scheme 1). Interpreting their relation from the cat-
alytic point of view allows us to draw important conclusions.
Conformer m-I exhibits a strong intramolecular hydrogen
bond which involves the more acidic thiourea hydrogen
mer as
a constituent of the self-assembly (d-III in
Scheme 1).
There are a few noteworthy spectroscopic features of the
distinguished conformation d-III which proves its identity.
The first is that the chemical shift order of the thiourea pro-
tons is just the opposite (dNA–H > dNB–H) to what is character-
istic to all the rest of the conformers. Secondly, d-III repre-
1
1
sents the only conformation where H, H spatial proximity
is detected between the two thiourea protons (N H and
B
N H) in the NOESY spectrum. Thirdly, the extended anti–
A
(
N -H) interacting with the basic nitrogen. Since both the
anti thiourea conformation D-III also appears in the hetero-
chiral dimer (D) measured for the racemate which confirms
the mode of interaction between conformers II and III (see
Scheme 1, top and Supporting Information).
B
acidic and basic moieties are blocked, this energetically low-
lying conformation is sterically unsuitable for bifunctional
catalysis. Conformer m-II is formed with a different intra-
molecular hydrogen bond involving the N -H hydrogen. Its
These experimental observations allowed us to refine the
general scheme of the bifunctional catalytic cycle by adding
the role of catalyst self-association. Scheme 1 depicts the
process where the catalytically active conformation m-III is
complexed by one of the stable, intramolecularly hydrogen
bonded conformations of the catalyst itself (m-II). We recall
that conformer d-III exhibits no intramolecular hydrogen
bond, its interactions are solely of intermolecular type.
Leaving the assembly this extended d-III conformer carries
the structural preference for becoming a catalytically active
A
anti–syn thiourea conformation separates the acidic (N -H)
B
and basic (:NMe ) moieties in space making them unavaila-
2
ble for the reactants at the same time. It can however effec-
tively form intermolecular interactions as a hydrogen-bond
donor, which predestinates its role in the self-assembled di-
meric state. Conformers m-II and d-II are essentially the
same which fulfills a promoter function between the mono-
meric and dimeric states.
The catalytically active conformation m-III is unique by
its extended anti–anti thiourea conformation and exhibits no
°
monomeric species [m-III] . The reactions proceed with [m-
Chem. Eur. J. 2012, 18, 1918 – 1922
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1919

G. Tꢁrkꢁnyi et al.
°
III] if favorable collisions occur with the reactants. There
are two general consequences of this process. On the one
hand the active conformation is frequently available from
the self-assembly. On the second hand regulation occurs
when the self-assembled state counterbalances product in-
hibition (binding). The catalytic cycle is maintained as long
as the product is displaced by the competing m-II. The pro-
cess inherently exists for 1 and becomes dominant in apolar
aprotic solvents. Self-activation and quenching of the active
conformation is brought to balance (buffering effect) ac-
cording to the reaction conditions.
To prove the relevance of the above catalytic cycle under
reaction conditions we analyzed the spectra of 1 in the pres-
1
9
[15]
Figure 4. Low-temperature F NMR of the Michael addition reaction in
]dichloromethane. Initial concentrations are (2): 270 mm, (3): 270 mm,
1): 27 mm. Note that the signals of the d-II/d-III conformers are time
averaged in the self-assembly at À208C.
ence of reactants and product. The classical Michael addi-
tion of ethyl malonate to trans-b-nitrostyrene was selected
for demonstration (Scheme 2) as this reaction gives good
productivity and enantioselectivity (90%<ee) in apolar
aprotic solvents (toluene, dichloromethane) with 10 mol%
[
(
D
2
[
10b]
1
1
.
mations already known and assigned from H NMR. Confor-
mer m-I has a sharp F resonance which is the least affected
1
9
by the chemical exchange in this temperature range. By rais-
19
ing the temperature the population of m-II drops and its
F
signal coalesces with the two asymmetric halves d-II and d-
III. This picture correlates well with the cycle shown in
Scheme 1 as m-II has the weaker hydrogen bond and can
easily form the assembly without significant change of its
conformation.
Scheme 2. Asymmetric Michael addition of ethyl-malonate to trans-b-ni-
trostyrene.
By raising the concentration of the catalyst we observed
the growth of the population of the dimeric state which veri-
fied the assignments of Figure 4 making a further connection
between the interpretation of the À908C experiments and
those apply for the reaction conditions (for details see Sup-
porting Information).
1
9
[4b,16]
Figure 3 shows the F-DOSY NMR
of the reaction
mixture 2+3+1 measured in dichloromethane at À208C.
19
There are two F resonances detected corresponding to the
monomeric (m) and dimeric (d) states of the catalyst (1)
proving the presence of the catalyst-assembly in the reaction
mixture. In this experiment, resonance (d) representing the
larger hydrodynamic radius correlates to the smaller (D=
Finally, the solvent dependence of the reaction productivi-
1
9
ty was compared and related to the F NMR spectroscopic
picture shown Figure 5. Our in situ NMR kinetic experi-
ments (see Supporting Information) have confirmed that the
reaction proceeds faster in toluene than in dichloromethane
À10
2
À1
2
.13”10 m s ) diffusion coefficient according to the
Stokes–Einstein equation.
two asymmetric halves of the dimeric assembly (d-II and d-
III) are in the fast exchange we verified the assignments of
the F NMR by additional temperature- and concentration
dependent experiments. Figure 4 demonstrates how the
F NMR of the same reaction mixture changed with the
temperature. The observation of four F peaks at À908C is
in line with the presence of m-I, m-II and d-II/d-III confor-
[16b]
Because the CF groups of the
3
[
10b]
at À208C, while it slows down considerably in THF.
The
1
1
9
F NMR spectra which were analyzed in conjunction with
H NMR to assign the F resonances reflect this finding
1
9
19
well. Toluene and dichloromethane were found to be rich in
the catalyst self-assembly (d) whereas no assembly was
found in THF owing to the disruption of hydrogen bonding
and the different nature of solvation effects.
19
1
9
In summary, the conformational equilibrium of 1 as dem-
onstrated for the reaction of ethyl malonate and trans-b-ni-
trostyrene highlights the dynamic nature of small molecular
catalysis. Although organocatalysis lacks the potency of
enzyme catalyzed reactions the equilibrium of significantly
different conformations is mimicking the regulation mecha-
nism of enzymes. Just like proteins are considered as dynam-
ically active assemblies where the internal motions are
[17]
closely linked to the function, the conformational flexibili-
ty of organocatalyst 1 helps to synergize the amine and thio-
urea functionalities. Stereoselective rate acceleration is ach-
ieved by the optimal spatial arrangement of the relevant
1
9
[16]
Figure 3. Low-temperature F-DOSY of the Michael addition reaction
mixture shown in Scheme 2 at time t=12 h in [D ]dichloromethane.
2
1920
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1918 – 1922

Amine–Thiourea Bifunctional Organocatalysis
COMMUNICATION
mension and 256 complex data points in the F1 dimension. Spectral
widths of 16 kHz were used in both dimensions. Relaxation delay time
1
s and mixing time 100 ms were used. Data were multiplied by Gaussian
weighting functions and zero filled to 8192”2048 matrix.
Acknowledgements
1
9
Figure 5. Low-temperature (À908C) F NMR of the pure catalyst (1) in
various deuterated solvents. The signals of the d-II/d-III conformers are
split as a result of the slower exchange at the lower temperature. The
presence of the catalyst assembly correlates well with reactivity.
The support of the Hungarian GVOP 3.2.1.-2004-04-0210/3.0 project and
OTKA K-69086 grant are gratefully acknowledged.
Keywords: bifunctional catalysts · molecular recognition ·
organocatalysis · self-association
acidic and basic groups of the catalyst with respect to the re-
actants during the conformational switch.
Clearly, there is a link between catalyst self-association
and successful bifunctional amine–thiourea catalysis with 1.
Using low-temperature NMR spectroscopy our work has re-
vealed the complex dynamics of Takemotoꢀs catalyst 1 in so-
lution. It was found that the catalyst gains inherent access to
its active extended thiourea conformation, by equilibrating
between its own hydrogen bonded intra- and intermolecular
interactions. By destabilization of the ground state sterically
inactive conformations it is the self-assembly which pro-
motes the reactions in apolar aprotic solvents rationalizing
the general need for the high catalyst loading. Enantioselec-
tivity of 1 in the CÀC forming reactions does not suffer as
[1] a) P. Vachal, E. N. Jacobsen, J. Am. Chem. Soc. 2002, 124, 10012–
1
0014; b) P. R. Schreiner, Chem. Soc. Rev. 2003, 32, 289–296; c) A.
Berkessel, H. Grçger, Asymmetric Organocatalysis, Wiley-VCH,
Weinheim, 2005; d) C. A. Hunter, H. R. Anderson, Angew. Chem.
2
009, 121, 7624–7636; Angew. Chem. Int. Ed. 2009, 48, 7488–7499;
e) Z. Zhiguo, P. R. Schreiner, Chem. Soc. Rev. 2009, 38, 1187–1198.
2] A. D. Bain, Prog. Nucl. Magn. Reson. Spectrosc. 2003, 43, 63–103.
3] G. Tꢁrkꢁnyi, P. Kirꢁly, Sz. Varga, B. Vakulya, T. Soós, Chem. Eur. J.
[
[
2
008, 14, 6078–6086.
[
4] a) H. S. Rho, S. H. Oh, J. W. Lee, J. Y. Lee, J. Chin, C. E. Song,
Chem. Commun. 2008, 10, 1208–1210; b) S. H. Oh, H. S. Rho, J. W.
Lee, J. E. Lee, S. H. Youk, J. Chin, C. E. Song, Angew. Chem. 2008,
1
20, 7990–7993; Angew. Chem. Int. Ed. 2008, 47, 7872–7875;
c) H. B. Jang, H. S. Rho, J. S. Oh, E. H. Nam, S. E. Park, H. Y. Bae,
C. E. Song, Org. Biomol. Chem. 2010, 8, 3918–3922.
long as the reaction runs under kinetic control and only one
distinguished conformation of the catalyst exhibits bifunc-
[
[
5] a) E. N. Jacobsen, D. W. C. MacMillan, Proc. Natl. Acad. Sci. USA
2010, 107, 20618–20619; b) R. R. Knowles, E. N. Jacobsen, Proc.
Natl. Acad. Sci. USA 2010, 107, 20678–20685; c) H. M. Li, X. F. Liu,
F. H. Wu, L. A. Tang, L. Deng, Proc. Natl. Acad. Sci. USA 2010,
[18]
tionality towards the reactants.
Rate acceleration is ex-
pected from the presence and equilibrium break-up of the
self-assembly as the catalytically active m-III conformation
is not a populated conformation of the system. The mea-
sured solvent dependent productivity of the ethyl malonate
trans-b-nitrostyrene Michael addition reaction shows a cor-
relation with the presence of the catalyst self-assembly. The
above catalyst self-activation model is the first example for
the consideration of catalyst self-association in the mecha-
nistic picture of amine–thiourea bifunctional organocataly-
sis.
1
07, 20625–20629.
6] J. W. Lee, T. H. Ryu, J. S. Oh, H. Y. Bae, H. Bin Jang, C. E. Song,
Chem. Commun. 2009, 46, 7224–7226.
[7] P. Kirꢁly, T. Soós, Sz. Varga, B. Vakulya, G. Tꢁrkꢁnyi, Magn. Reson.
Chem. 2010, 48, 13–19.
[
8] a) M. Nishio, Y. Umezawa, K. Honda, S. Tsuboyama, H. Suezawa,
CrystEngComm 2009, 11, 1757–1788; b) M. Nishio, Phys. Chem.
Chem. Phys. 2011, 13, 13873–13900; c) M. B. Schmid, K. Zeitler,
R. M. Gschwind, Chem. Sci. 2011, 2, 1793–1803.
[
9] a) S. J. Connon, Chem. Commun. 2008, 2499–2510; b) T. Marcelli,
H. Hiemstra, Synthesis 2010, 8, 1229–1279.
[
[
10] a) T. Okino, Y. Hoashi, Y. Takemoto, J. Am. Chem. Soc. 2003, 125,
2672–12673; b) T. Okino, Y. Hoashi, T. Furukawa, X. Xu, Y. Take-
1
moto, J. Am. Chem. Soc. 2005, 127, 119–125.
11] a) T. Okino, S. Nakamura, T. Furukawa, Y. Takemoto, Org. Lett.
Experimental Section
2
004, 6, 625–627; b) Y. Hoashi, T. Yabuta, Y. Takemoto, Tetrahe-
dron Lett. 2004, 45, 9185–9188; c) Y. Hoashi, T. Okino, Y. Takemo-
to, Angew. Chem. 2005, 117, 4100–4103; Angew. Chem. Int. Ed.
2005, 44, 4032–4035; d) A. G. Doyle, E. N. Jacobsen, Chem. Rev.
2007, 107, 5713–5743; e) A. Dondoni, A. Massi, Angew. Chem.
2008, 120, 4716–4739; Angew. Chem. Int. Ed. 2008, 47, 4638–4660;
f) B. Vakulya, Sz. Varga, T. Soós, J. Org. Chem. 2008, 73, 3475–
3480; g) A. Zea, G. Valero, A. N. R. Alba, A. Moyano, R. Rios,
Adv. Synth. Catal. 2010, 352, 1102–1106; h) C. Yu, Y. Zhang, A.
Song, Y Ji, W. Wang, Chem. Eur. J. 2011, 17, 770–774.
[
10]
Compound 1 was prepared according to literature procedure. Spectro-
[
10b]
scopic characterization of compound 4 has been reported earlier.
Compounds 2 and 3 were purchased from Sigma–Aldrich. NMR experi-
ments were carried out on a 600 MHz (for H) Varian NMR System spec-
trometer using a 5 mm direct detection N, P/
Chemical shifts are referenced to residual solvent signals (CD
.32 ppm, d =54.0 ppm). Deuterated (99.98 atom%) solvents were pur-
chased from CIL. CFCl was purchased from Sigma–Aldrich. The cryo-
genic solvent was CD Cl /CFCl 1:4 v/v. The assignment of the NMR res-
onances was aided by the analysis of through-bond ( H-gDQFCOSY,
H, C-gHSQC) and through-space ( H-NOESY) correlations measured
by using the VnmrJ 2.2C software. The DgsteSL_cc sequence was used in
In low-temperature
D experiments 4096 complex data points were acquired in the F2 di-
1
1
5
31
1
19
ACHTUNGTRENNUNG
2
2
H
=
5
C
3
2
2
3
[12] A. Berkessel, F. Cleemann, S. Mukherjee, T. N. Müller, J. Lex,
Angew. Chem. 2005, 117, 817–821; Angew. Chem. Int. Ed. 2005, 44,
807–811.
[13] a) Y. Takemoto, Org. Biomol. Chem. 2005, 3, 4299–4306; b) T. Ino-
kuma, Y. Hoashi, Y. Takemoto, J. Am. Chem. Soc. 2006, 128, 9413–
9419; c) A. Hamza, G. Schubert, T. Soós, I. Pꢁpai, J. Am. Chem.
1
1
13
1
1
9
[16]
convection compensated F-DOSY experiments.
2
Chem. Eur. J. 2012, 18, 1918 – 1922
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1921

G. Tꢁrkꢁnyi et al.
Soc. 2006, 128, 13151–13160; d) K. Yoshida, T. Inokuma, K. Takasu,
Y. Takemoto, Molecules 2010, 15, 8305–8326.
14] D. Monge, K. L. Jensen, P. T. Franke, L. Lykke, K. A. Jørgensen,
Chem. Eur. J. 2010, 16, 9478–9484.
[17] a) P. K. Agarwal, J. Am. Chem. Soc. 2005, 127, 15248–15256;
b) D. D. Boehr, H. J. Dyson, P. E. Wright, Chem. Rev. 2006, 106,
3055–3079; c) Y. Sohtome, B. Shin, N. Horitsugi, K. Noguchi, K.
Nagasawa, Chem. Asian J. 2011, 6, 2463–2470.
[
[
[
15] We thank one of the referees for the suggestion to demonstrate the
self-assembly in the Michael addition reaction.
16] a) A. Jerschow, N. Müller, J. Magn. Reson. 1997, 125, 372–375; b) T.
Megyes, H. Jude, T. Grósz, I. Bakó, T. Radnai, G. Tꢁrkꢁnyi, G.
Pꢁlinkꢁs, P. J. Stang, J. Am. Chem. Soc. 2005, 127, 10731–10738.
[18] G. Rulli, N. Duangdee, K. Baer, W. Hummel, A. Berkessel, H.
Grçger, Angew. Chem. 2011, 123, 8092–8095; Angew. Chem. Int.
Ed. 2011, 50, 7944–7947.
Received: August 29, 2011
Revised: December 1, 2011
Published online: January 20, 2012
1922
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2012, 18, 1918 – 1922