(Thio)amidoindoles and (Thio)amidobenzimidazoles: An investigation of their hydrogen-bonding and organocatalytic properties in the ringopening polymerization of lactide

Article abstract of DOI:10.1002/chem.200902912

The mechanism of the ringopening polymerization (ROP) of lactide catalyzed by two partner hydrogen-bonding organocatalysts was explored. New amidoindoles 4a,c, thioamidoindoles 4b,d, amidobenzimidazoles 5a,c, and thioamidobenzimidazoles 5b,c were synthesi

Full text of DOI:10.1002/chem.200902912

DOI: 10.1002/chem.200902912
ACHTUNGTRENNUNG
Sylvain Koeller,[a] Joji Kadota,[b] Frꢀdꢀric Peruch,[b] Alain Deffieux,[b] Noꢁl Pinaud,[a]
Isabelle Pianet,[a] Stꢀphane Massip,[c] Jean-Michel Lꢀger,[c] Jean-Pierre Desvergne,[a] and
Brigitte Bibal*[a]
4196
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4196 – 4205

FULL PAPER
Abstract: The mechanism of the ring-
opening polymerization (ROP) of lac-
tide catalyzed by two partner hydro-
gen-bonding organocatalysts was ex-
plored. New amidoindoles 4a,c, thio-
of lactide in the presence of a cocata-
lyst, tertiary amine 3a or 3b, which ac-
tivates the growing polymer chain
through hydrogen-bonding. Reactions
were conducted in 2–24 h at 208C; con-
version yields ranged between 22 and
100%. A detailed study of the intermo-
lecular interactions undertaken be-
tween the participating species showed
that, as expected, simultaneous weak
hydrogen bonds do exist to activate the
reagents. Moreover, interactions have
been revealed between the partner cat-
alysts 4/5+3. ROP catalyzed by these
partner activators is thus governed by
multiple dynamic equilibria. The latter
should be judiciously adjusted to fine-
tune the catalytic properties of (thio)-
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
as activators of the monomer. In the
solid state and in solution, compounds
4 and 5 showed a propensity for self-as-
sociation, which was evaluated. (Thio)-
Amides 4 and 5 do catalyze the ROP
Keywords: hydrogen bonds · lacti-
des · organocatalysis · partner cata-
lysts · ring-opening polymerization
AHCTUNGTRENNUNG
Introduction
mild conditions (208C, loading of catalyst below 10 mol%).
This remarkable catalytic activity relies on their ability to
promote hydrogen bonds with the monomer and thus the
smooth increase of its electrophilicity. Additionally, a hydro-
gen-bond-acceptor cocatalyst partner, generally a tertiary
amine, is required to increase the nucleophilicity of the ini-
tiator and the growing polymer chain. These catalytic sys-
tems can be a unique compound such as Takemotoꢁs catalyst
1,[13] or two independent molecules such as thiourea 2+
amine 3 (Scheme 1). Under drastic anhydrous conditions, at
208C for over 24 h, catalyst 1 (5 mol% in dichloromethane,
5 mol% of initiator) allows 97% conversion of lactide into
Organocatalyzed polymerization is an emerging field of re-
search.[1] It represents an elegant alternative to organome-
tallic and enzymatic catalyses, as it allows, within 24 h, the
preparation of polymers with controlled average molar
masses, narrow dispersities, and without any metallic resi-
dues. Organocatalyzed polymerization was essentially devel-
oped for the ring-opening polymerization (ROP) of cyclic
esters, diesters, and carbonates.[2] From a mechanistic point
of view, it has been demonstrated that these ROPs can be
promoted by organocatalysts by means of the activation of
the reagents through temporary covalent bonds or weak in-
teractions. The strategies encompass the activation of the
monomer, that is, the electrophile (using Brønsted acids,[3]
alcohols,[4] 4-dimethylaminopyridine (DMAP) derivatives,[5]
phosphines,[6] N-heterocyclic carbenes[7]); the activation of
the growing polymer chain, that is, the nucleophile (using
a polyACTHNURGTNE(NUG lactide) with a narrow dispersity and controlled
molar masses, as the experimental degree of polymerization
(DP) matches with the theoretical one (moles of monomer
polymerized/moles of initiator). The ROP of lactide was
also efficient over 24 h in the presence of thiourea 2 and co-
catalyst dimethylcyclohexylamine (Me2NCy) 3a or (À)-spar-
teine (Sp) 3b (Scheme 1). Notably, due to the better hydro-
gen-bonding properties of 3b versus 3a, the combination of
2+3b allowed a complete polymerization reaction within
2 h, whereas the time of the reaction in the presence of the
partners 2+3a was 24 h.
basic phosphazenes,[8] 1,8-diazabicyclo
ACHTUNGTRENNUNG
(DBU), and 7-methyl-1,5,7-triazabicycloCAHTUNGTRENNNUG
(MTBD)[9]); or the dual activation of both monomer and
chain end (thiourea derivatives[10] and 1,5,7-triazabicyclo-
ACHTUNGTRENNUNG
[4.4.0]dec-5-ene (TBD)[11]). Among the organocatalysts,
thio
ureas[12] have shown promising results in the access of
A remaining drawback of previously reported procedures
is that satisfying conversion requires extensive and tedious
preparation of dry reagents and performing the reactions in
a glove box.
polymers with controlled molar masses, operating under
[a] Dr. S. Koeller, N. Pinaud, Dr. I. Pianet, Dr. J.-P. Desvergne,
Dr. B. Bibal
Therefore, new hydrogen-bonding organocatalysts are
highly desirable, especially versatile ones. They should be
available through a rapid synthesis and should be efficient
at room temperature under relaxed and less energy-consum-
ing conditions. Our interest in supramolecular chemistry[14]
prompted us to design hydrogen-bonding catalysts for the
ROP of lactide. In this context, amides, which are modular
compounds with easy synthetic access, were anticipated to
be polymerization promoters through hydrogen-bonding.
We recently showed that, in the presence of (À)-sparteine,
an activated amidoindole 4a is an efficient organocatalyst.[15]
We demonstrated that both NH groups from the amide and
the indole moieties participate in the hydrogen-bonding of
Institut des Sciences Molꢂculaires UMR CNRS 5255
Universitꢂ de Bordeaux, 351 cours de la Libꢂration
33405 Talence (France)
Fax : (+33)5-4000-6158
E-mail: b.bibal@ism.u-bordeaux1.fr
[b] Dr. J. Kadota, Dr. F. Peruch, Dr. A. Deffieux
Laboratoire de Chimie des Polymꢃres Organiques
Ecole Nationale Supꢂrieure de Chimie et Physique de Bordeaux
16 avenue Pey-Berland, 33607 Pessac (France)
[c] Dr. S. Massip, Dr. J.-M. Lꢂger
Universitꢂ de Bordeaux, EA 4138 Pharmacochimie
146 rue Lꢂo Saignat, 33076 Bordeaux (France)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200902912.
Chem. Eur. J. 2010, 16, 4196 – 4205
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4197

B. Bibal et al.
depth catalytic properties of
new hydrogen-bonding organo-
catalysts for the polymerization
of lactide, in connection with
their aggregative properties.
Results and Discussion
Synthesis of the catalysts: the
straightforward preparation of
new compounds 4 and 5 was
achieved in one or two steps
Scheme 1. Molecular structures of thioureas 1 and 2, tertiary amines 3a,b and amide 4a as hydrogen-bonding
organocatalysts for ring-opening polymerization of lactide.
from
commercial
reagents
(Scheme 3).
lactide, thus allowing its total conversion in 2 h. Moreover,
an X-ray structure of a hydrogen-bonded complex between
a ROP organocatalyst and the monomer was reported for
the first time. To deepen and broaden our investigation, we
decided to explore the behavior of two series of amides and
related thioamides that bear an indole (compounds 4) or a
benzimidazole group (compounds 5), as a second hydrogen-
bond donor (Scheme 2).
Scheme 2. Molecular structures of hydrogen-bonding organocatalysts 4
and 5.
In compounds 4a–d and 5a–d, the global geometry of the
hydrogen-bond donor subunits is different and may have an
impact upon the catalytic properties. Besides, the strength of
Scheme 3. Straightforward synthesis of (thio)amides 4 and 5. DIPEA=
À
diisopropylethylamine.
the hydrogen-donor properties of the NH C=X group could
also be modulated by two factors: 1) the nature of the X
atom, oxygen (amide function), or sulfur (thioamide func-
tion); 2) the electronic density on the phenyl group, from
electron-deficient derivatives (4a,b and 5a,b) to electron-
rich compounds (4c,d and 5c,d). Classically, thioamides are
better hydrogen-bond donors and less subject to self-aggre-
gation. Electron-deficient substituents should increase the
hydrogen-bond donor properties of the corresponding com-
pounds. Finally, the effect of the cocatalyst in these ROPs
will also be evaluated using two tertiary amines 3a or 3b, as
hydrogen-bond acceptors of a different strength.[9b]
The possible disadvantage of this approach is the poten-
tial formation of hydrogen bonds between the partner acti-
vators, that is, 4 or 5 with 3, which could hamper the reac-
tion efficiency. Consequently, a detailed supramolecular
mechanism of ROP involving partner organocatalysts will
be explored. Herein, we report the synthesis and the in-
In detail, amidoindoles 4a and 4c were obtained by con-
densation of the aniline derivatives on the corresponding
acid chloride (freshly prepared), in 90 and 73% yield, re-
spectively.[16] A similar procedure[17] was achieved to synthe-
size amidobenzimidazoles 5a and 5c in 45 and 73% yield,
respectively, from the benzoyl chloride derivatives and 2-
amino-5,6-dimethyl-benzimidazole.[18]
Thioamido-related
compounds were obtained by thionation of the oxo com-
pounds using Lawessonꢁs reagent in toluene heated at
reflux.[19] Molecules 4b and 4d were isolated in 78 and 81%
yield, respectively. Compounds 5b and 5d were also ob-
tained in excellent yields: 97 and 93%, respectively.
1
Interestingly, based upon H NMR spectra in [D6]DMSO
and CDCl3, (thio)amidobenzimidazoles 5a–d appeared to be
in equilibrium between the classical N-(1H-benzo[d]im-
AHCTUNGERTGiNNUN dazol-2-yl) structure and its N-(1,3-dihydro-benzimidazol-2-
4198
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4196 – 4205

Properties of (Thio)Amides
FULL PAPER
ylidene) tautomer (Scheme 4). For instance, all compounds
5 displayed similar H NMR spectra ([D6]DMSO, 4 mm): the
form centrosymmetric dimers (Figure 1), internally tied with
symmetrical hydrogen bonds between the carbonyl group
1
À
two NH groups appeared as a broad signal at d=12.1–
(acceptor) and the H N of the indole (donor) moiety (dis-
Scheme 4. The tautomeric equilibrium of compound 5.
12.5 ppm for amides 5a,b or d=13.4–13.5 ppm for thio-
ACHTUNGTRENNUNGamides 5b–d (see the Supporting Information). In our anal-
yses, spectroscopic data did not allow for the discrimination
between the two forms. As previously reported,[20] this tauto-
meric equilibrium may be favored by hydrogen bonding be-
tween (thio)amidobenzimidazoles 5 or between 5 and the
solvent.
Figure 1. Hydrogen-bonded dimer of compounds 4a in the crystal.
Concerning hydrogen-bonding properties, this tautomeric
equilibrium is anticipated to influence the catalytic proper-
ties of the series 5a–d in ROP reactions, as the solvent of
choice for these polymerizations is dichloromethane.
All catalysts were found to be fairly soluble in dichloro-
methane. In the presence of lactide, this solubility was in-
creased, which allows one to undertake reactions under clas-
sical conditions (208C, 5 mol% with respect to monomer,
that is, 35 mm). This behavior is probably due to specific in-
termolecular interactions between the hydrogen-bonding
catalysts and lactide as a hydrogen-bond acceptor. This is
clearly supported by the weak interactions between reagents
and catalysts (see below), unrevealed so far.
À
À
tance (d) of N H····O=C=2.35 ꢄ). Interestingly, the N H
of the amide spacer is not involved in the dimer but it is
linked through a hydrogen bond to the carbonyl group of
another dimer (d=2.41 ꢄ) that lies in a quasi-parallel plane
(interdimer plane distanceꢀ3.4 ꢄ). A herringbone packing
is formed in the solid with an angle of 878 between the
planes of the dimers (see the Supporting Information).
Compound 4c crystallizes in the monoclinic space group
P21/n (Figure 2). In the solid, the molecules are not totally
X-ray diffraction: Single crystals of compounds 4a, 4c, 5c,
and 5d were grown in dichloromethane and/or chloroform
subjected to slow pentane vapor diffusion. To the best of
our knowledge, very few crystallographic descriptions of
amidoindole,[21] thioamidoindole,[22] and amidobenzimid-
ACHTUNGTRENNUNG
azole[23] have been reported. Notably, the first structure of a
thioamidobenzimidazole (5d) is described herein. Interest-
ingly, in the crystal, the tautomeric N-(1H-benzo[d]imidazol-
2(3H)-ylidene)benzothioamide) form is observed. All mole-
cules adopt centrosymmetric monoclinic space groups. The
crystal lattice is stabilized by strong intermolecular hydro-
gen bonds as usually encountered for amides[24] and thioa-
mides.[25] The experimental hydrogen-bond geometries
(lengths, angles) are in the range of those usually found in
literature.[24d]
Figure 2. Hydrogen-bonded packing of compounds 4c in the crystal.
The crystal of 4a belongs to the space group P21/a in
which the molecules are packed in layers. The layers are
separated by a distance of approximately 2.9 ꢄ, which corre-
sponds to F···F interactions between the CF3 substituents sit-
uated at the extremities of the layers. In the layers, the two
aromatic parts of the molecules are non-coplanar (angle be-
tween the two aromatic parts: 16.7(3)8). Compounds 4a
flat (angle equal to 15.9(1)8 between the two aromatic
parts). They are packed in quasi-coplanar dimers in which
the molecules are linked by two strong symmetrical hydro-
À
gen bonds that involve the N H (donor) of the indole
À
moiety and the carbonyl group acceptor (d(N H····O=C)=
À
2.04 ꢄ). Here again, the N H of the amide spacer does not
Chem. Eur. J. 2010, 16, 4196 – 4205
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4199

B. Bibal et al.
participate in the cohesion of the centrosymmetric dimer; it
interacts with the oxygen of the methoxy group of the
À
À
neighboring dimer (d(N H····O CH3)=2.24 ꢄ). These
dimers are stacked in a herringbone mode and form an
angle of 708.
Amidobenzimidazole 5c crystallizes in the monoclinic
space group P21/a (Figure 3). The asymmetric unit includes
two independent molecules of 5c and one disordered mole-
Figure 4. Hydrogen-bonded dimer of compound 5d in the crystal.
atoms of the benzimidazole hold a hydrogen atom. In the
molecular structure, the five-atom ring of the benzimidazole
is symmetrical, as shown by the bond lengths (see the Sup-
À
porting Information), and the N C bond, which links the
benzimidazole moiety and the (thio)amide, is shortened
compared with that of 5c (1.369(4) and 1.380(4) ꢄ in 5c and
1.346(3) ꢄ in 5d, respectively). Within 5d, the thiobenzim-
AHCTUNGTREGiNNNU dazole moiety forms an angle of 38.6(1)8 with the anisyl
ring. The molecules are organized in centrosymmetric
À
dimers that involve one of the N H units of the imidazole
À
ring and the nitrogen of the spacer (d
T
À
whereas the other N H and the C=S thiocarbonyl group are
not implicated in such directional hydrogen bonds. Several
short contacts also participate in the cohesion of the solid
(see the Supporting Information).
Figure 3. Hydrogen-bond packing of compounds 5c in the crystal.
Inspection of the molecular packing of these new (thio)-
AHCTUNGTREGaNNNU mides shows that the latter could be classed in two catego-
cule of solvent on a site partially occupied by molecules of
chloroform (80%) and dichloromethane (20%). In this
asymmetric unit, both independent molecules are not flat, as
the two aromatic parts of the molecules form an angle of
36.2(1) and 42.7(1)8 for the two molecules, respectively. Of
note, the amide link is situated in the benzimidazole plane.
These two independent mole-
ries (Table 1). Concerning 4a and 4c, the centrosymmetric
À
hydrogen-bonded dimers only involve the C=O and the N
À
H of the indole moiety, whereas the N H of the linker
(amide) is not directly implicated. In 5c and 5d crystals, the
À
N H of the amide spacer is strongly involved in the dimer
(or ribbonlike) cohesion. Additionally, 5d presents an imine
cules are not in the same plane
and displayed an average twist-
ed angle of 278 (taking benzim-
ACHTUNGTRENNUNGidACHTUNGTRENNUNGazole carbons plus C=O to
build each average plane).
These two compounds are asso-
ciated by hydrogen bonds that
Table 1. Hydrogen bond lengths [ꢄ] and angles [8] in monocrystals 4a, 4c, 5c, and 5d.
[a]
À
À
À
N Hamide···O=C
N Harom···O=C
Other hydrogen bonds
aD H···A
4a
4c
2.41
–
2.35
2.04
–
2.24
143.9, 147.6
161.4, 161.4
(NHamide···OMe)
2.03, 2.08
5c
5d
–
–
2.22, 2.21
–
149.9, 139.9,
158.9, 151.8
172.1
À
(N Harom···N)
1.97
À
involve only the N H of the
À
(N Harom···N=C)
amide spacer and the nitrogen
[a] Respective angles of the hydrogen bonds, from column 2 to column 4.
lone pair of the benzimidazole
moiety (d=2.03 and 2.08 ꢄ, re-
spectively). The carbonyl groups are symmetrically hydro-
tautomeric form in the dimer. Finally, if these dimeric forms
À
gen-bonded with the N H of benzimidazole group that be-
also exist in solution (see above), in 4a and 4c the amido
À
longs to another asymmetric unit (d=2.21 and 2.22 ꢄ), as
indicated in Figure 3. Finally, these hydrogen-bonded mole-
cules form infinite ribbons in the crystals in which the C=O
N H can be available for extra hydrogen-bonded linkage
such as the one necessary for the activation of the lactide. In
contrast, hydrogen-bond donors in 5c and 5d are embedded
within the aggregates and should not be as available as
those present in 4a and 4c. This would rationalize the exper-
imental data presented hereafter: the relative catalytic effi-
ciency of the compounds and/or the preferential molecular
conformations experienced in solution.
À
and N H groups are all involved.
The crystal of 5d belongs to the centrosymmetric mono-
clinic space group C2/c (Figure 4). In the crystal, this thio-
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
4200
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4196 – 4205

Properties of (Thio)Amides
FULL PAPER
Molecular interactions in solution: Based on the observa-
tions in the solid state, we decided to investigate the propen-
sity of the catalysts to self-aggregate in solution. Titrations
experimental ROP conditions (i.e., 35 mm), approximately
60–80% of (thio)benzimidazoles 5 are aggregated. There-
fore, a complementary geometry of hydrogen-bond acceptor
and donor within (thio)amides 5 probably favors dimeriza-
tion, as previously shown in the solid state. This result has
obviously to be taken into account for the evaluation of the
catalytic activity.
1
monitored by H NMR (CDCl3) spectroscopy were conduct-
ed on compounds 2 (as a model), 4, and 5 in the 0.5–35 mm
concentration range. NMR spectra showed concentration-
dependent chemical shifts, except for compounds 4b and 4d
(see the Supporting Information). The variations mainly
concern the amide NH groups and, to a lesser extent, some
aromatic or aliphatic protons. These observations were inter-
preted as the formation of hydrogen-bonded complexes be-
Catalytic properties in ROP reactions: Molecules 4 and 5
were tested as organocatalysts in the ROP of lactide, chosen
as a model monomer. Classical conditions were employed:
concentration of lactide at 0.7m in dichloromethane, 208C,
using biphenylmethanol as the initiator (5 mol%) and a co-
catalyst 3a or 3b (5 mol%) as an activator of the initiator/
polymer growing chain (Table 3). All experiments were con-
ducted under anhydrous conditions (dry reagents and sol-
vents) and in the presence of 4 ꢄ molecular sieves to trap
residual water, a competitor of the catalysts in the hydro-
gen-bonding process. An independent experiment proved
that the activated or nonactivated molecular sieves had no
catalytic effect upon reactions. Time reactions were pro-
grammed between 24 and 72 h to evaluate the kinetics.
When, in the presence of 3b, the conversion of the mono-
mer became over 50%, the crude polylactides were charac-
À
tween complementary X=C NH moieties. Experimental
data were fitted with a model of dimerization (except com-
pound 2, which oligomerizes) and the corresponding binding
constants Kdimer are reported in Table 2.
Table 2. Maximum variation of the chemical shifts (dmax in ppm) for com-
pounds 2, 4a, 4c, and 5a–d and the corresponding constants of dimeriza-
tion (Kdimer in mÀ1) in CDCl3.[a,b]
2
4a[c]
4c
5a
5b
5c
5d
Ddmax
+0.51
+0.18
4
+0.27
5
À0.66
+1.11
66
À0.25
À0.24
[c]
Kdimer
6[d]
226
195
190
[a] Monitoring protons at 208C in the 0.5–35.0 mm concentration range.
[b] Estimated error Æ15%. [c] Measured in CD2Cl2. [d] Data fitted with
an isodesmic model.
1
terized by the average molar mass determined by H NMR
spectroscopy, M
nACHTUNGTRENN(UNG NMR), and size-exclusion chromatography
(SEC; MnA(SEC)). The index of dispersity (PDI) was also de-
CTHUNGTRENNUNG
Taken as a reference catalyst in the ROP of lactide, thio-
termined from the SEC analysis.
urea 2 showed moderate aggregative properties (Kaggreg
=
Under these conditions, over 24 h the catalytic property
of the individual components was evaluated with prelimina-
ry tests. Reactions achieved in the presence of the initiator
and in the absence of organocatalyst 3 and 4/5 gave no con-
6mÀ1) that had not been previously reported. Amidoindoles
4a and 4c form dimers with a small association constant
(Kdimer =4 and 5mÀ1, respectively). Notably, thioamidoin-
doles 4b and 4d do not aggre-
Table 3. ROP conditions and properties of the polyCAHTUNGTRENNUNG
gate (Ddmax <0.05 ppm). This
behavior is expected for thio-
carbonyls because the latter are
less efficient hydrogen-bond ac-
ceptors than their oxo deriva-
tives.[26] Interestingly, even if
the values of the self-associa-
tion constants are weak (4–
6mÀ1), approximately 15–20%
of dimer is present in solution
under the ROP conditions
(concentration of catalyst=
35 mm).
Benzimidazoles 5 display a
different behavior. All proton
NMR spectroscopic signals
were largely affected by in-
creasing the concentration.
Compounds 5 form tighter hy-
Entry
Catalyst
(5 mol%)
t
Conversion [%][b]
M
G
M
A
E
M
A
E
PDI[d]
(3b)
G
[h]
3a
3b
ACHTUNGTRENNUNG
1
2
1
24
24
95
46
100
–
3070
3070
3210
5150
1.08
[e]
[e]
[e]
1+H2O
5 mol%
1+H2O
50 mol%
2
–
–
–
3
24
0
0
3070
–
–
–
4
5
6
24
48
24
85
95
72
100
–
100
3070
2920
3070
2633
5420
1.10
1.06
1.07
2
4a
2778[f]
3070
3970[f]
4460
AHCTUNGTRENNUNG
7
8
9
4a
4b
4b
4c
4d
5a
5a
5a
5b
5c
5d
48
24
48
24
24
24
48
72
24
24
24
95
43
63
53
54
23
35
49
20
24
28
2920
1630
1810
2270
2180
1010
1720
2550
950
3498[f]
1480
1769
2489
2489
760
4190[f]
1907
2119
2850
3200
–
1720
2576
–
1.06[f]
1.05
1.05
1.07
1.07
–
1.06
1.06
–
53
59
74
71
33
56
83
31
20
28
10
11
12
13
14
15
16
17
n.d.[g]
n.d.[g]
904
616
904
610
860
–
–
–
–
drogen-bonded dimers (Kdimer
=
66–226mÀ1) than those of com-
pounds 4 (Kdimer ꢁ5mÀ1). Only
thio derivative 5b is associated
[a] Conditions: lactide 0.7m in CH2Cl2, catalyst 4 or 5 (5 mol%), tertiary amine 3 (5 mol%), biphenylmethanol
as initiator (5 mol%), 4 ꢄ molecular sieves, 208C. [b] Determined by 1H NMR spectroscopy. [c] Theoretical
molar mass when conversion was 100%. [d] Determined by size-exclusion chromatography. [e] Not measured
when conversion was lower than 50%. [f] Experiment achieved in the presence of amine 3a. [g] n.d.=not de-
in a looser manner. Under the termined.
Chem. Eur. J. 2010, 16, 4196 – 4205
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4201

B. Bibal et al.
version of the monomer. The ROP of lactide conducted in
the presence of the initiator plus organocatalyst 4/5, without
partner amine 3, resulted in a conversion lower than 10%.
Polymerization in the presence of the initiator plus amine 3,
without organocatalyst 4 or 5, allowed the conversion of ap-
proximately 10% of lactide with 3a and 15–20% with 3b.
These results demonstrated that individual ROP partners
have a low catalytic efficiency.
24 h) and, secondly, the same quantity of lactide was added.
After 24 h, the conversion was total and the resulting poly-
lactide had similar characteristics (M
(theor)=5944 gmolÀ1,
nACHTUNGTRENNUNG
MnACHTUTGNRENNUNG
that a controlled process is indeed occurring in these orga-
nocatalyzed ROP reactions.
Catalysts 4+3 (Table 3, entries 6–11) were more efficient
than compounds 5+3 (entries 12–17) and provided poly-
Table 3 shows the conditions of the ROP experiments in
the presence of different organocatalysts 1, 2, 4, or 5 and a
cocatalyst 3 and the characterization of the resulting poly-
lactides. Molar masses determined by NMR spectroscopy
ACHTUNGTREN(NGNU lactides) in 24 h, with a conversion ranging from 43 to
100%. Additionally, the catalytic activity of (thio)amides
4a–d was slightly influenced by the hydrogen-bond acceptor
character of amine 3 (entries 6, 8, 10, and 11). In the pres-
ence of the better hydrogen-bond activator 3b, the percent-
age of conversion was higher: 71–100% compared to 53–
72% when 3a was employed as a cocatalyst. Notably, quan-
titative conversion with 4a was obtained in 2 h in the pres-
ence of 3b (entry 6 indicates the conversion after 2 h).
These observations can be ascribable to the better hydro-
gen-bond donor character of amine 3b.[9b,10] This assumption
will be checked by the measurement of the corresponding
association constants (see below). For partner catalysts 4a–
d+3a, conversion was significantly increased when reaction
time was extended to 48 h (entries 5, 7, and 9), thereby indi-
cating that polymerization is still going on and that chain
ends are not deactivated. This phenomenon was already ob-
served with thiourea 2 (entry 5). Kinetics of these ROP re-
actions can be slow in some cases and it is dependent upon
the activation efficiency of both partner catalysts.
(M
value (M CHTUNGTRENNUNG
(NMR)) are in good agreement with the theoretical
nA(theor)), whereas molar weights determined by
SEC are overestimated, as they are calculated versus poly-
styrene standards.
Under our conditions, in the presence of molecular sieves,
thioureas 1 and 2 allowed a very good conversion (85–
100%; Table 3, entries 1, 4, and 5), especially when the co-
ACHTUNGTRENNUNGcatalyst was 3b, which is recognized as a better hydrogen-
bond acceptor than 3a.[9b] Moreover, the polylactides with a
narrow dispersity and predictable average number molar
mass were obtained in reasonable time (24 h). The catalytic
efficiency of 1 and 2 under our conditions is thus identical
to that previously reported.[10]
The inhibiting properties of water in ROP triggered by
hydrogen-bonding catalysts were demonstrated in entries 2
and 3 of Table 3. Indeed, an increase of the quantities of
water (5 and 50 mol% with respect to lactide) in the reac-
tion medium resulted in the progressive decrease of the per-
centage of monomer conversion (95, 46, and 0%, respective-
ly). Reactions were therefore systematically run in the pres-
ence of 4 ꢄ molecular sieves (five beads in 1 mL of reaction
medium).
Each compound 4/5 was found to be active and allowed
the conversion of lactide in 24 h with 20 to 100% yield. As
expected, the ROP of lactide appeared to be strongly depen-
dent upon the nature of the partner organocatalysts. Nota-
bly, all polymers had a narrow dispersity (polydispersity
index (PDI)=1.05–1.07) coupled with an experimental
Nevertheless, additional parameters have to be taken into
account to better understand the structure–activity relation-
ships. Concerning the hydrogen-bonding catalysts 4, the cat-
alytic activity of the NH group appeared to be dependent
upon two main structural factors: the electron density of its
phenyl substituent and the nature of the linker onto the
indole substituent (C=O or C=S). Compound 4a with an
electron-withdrawing group is the most active catalyst of the
series (Table 3, entry 6, 72–100% conversion). Its corre-
sponding electron-donating derivative 4c and its thioamide
derivative 4b are less active and allow for 53–74%
(entry 10) and 43–53% conversion (entry 8), respectively.
These observations can be explained by two opposite phe-
nomena: in the structure of 4c, the amido NH is less acidic
than in 4a and thus it is a worse hydrogen-bond donor than
4a, whereas in compound 4b, the acidity of NH is largely in-
À
number average molar mass M
nACHTUNGRTEN(NUNG NMR), which is in agree-
ment with the theoretical (MnA(theor)) assuming the forma-
CTHUNGTRENNUNG
tion of one chain per initiator molecule (Table 3). Variation
of the [monomer]/[initiator]([M]/[I]) ratio from 20 to 50 and
100 led to narrowly dispersed polymers (PDI=1.12 and
1.11, respectively), again with masses that match the theo-
creased (in acetonitrile, pKaACHTUNTRGENN(UG S=C NH)=11–13 and pKaACHTUNGTRENNUNG(O=
[26]
À
C NH)=17)
and thus might be partly inhibited by the
retical ones (M
ly, to be compared to their
G
basic species present in the solution, that is, the cocatalyst 3
(see below). Interestingly, catalyst 4c was as efficient as its
thio derivative 4d. In this case, due to the electron-rich aro-
matic substituent, the NH group is a poorer hydrogen-bond
donor whatever the nature of the C=X bond.
M CHTUNGTRENNUNG
14584 gmolÀ1; see the Supporting Information). Besides,
13C NMR spectroscopy indicates that the polymers exhibit a
fully isotactic structure, thereby suggesting the absence of
side transesterification reactions that would yield racemiza-
tion (see the Supporting Information). Additionally, a chain-
extension experiment was successfully realized: firstly, the
polymerization of lactide was conducted under standard
Given a 24 h reaction time, (thio)amidobenzimidazoles
5a–d were moderate catalysts (20–33% conversion; Table 3,
entries 12, 15–17) whatever the nature of the aromatic sub-
stituent and the nature of the cocatalyst 3. When the reac-
tion time was increased to 48 or 72 h (entries 13 and 14), the
activity of hydrogen-bonding catalyst 5a was increased, es-
conditions (Table 3, entry 6: 100% conversion, M CHTUNTGREN(NUNG theor)=
nA
3064 gmolÀ1, M (SEC)=3070 gmolÀ1, and PDI=1.07, after
nACHTUNGTRENNUNG
4202
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4196 – 4205

Properties of (Thio)Amides
FULL PAPER
Table 4. Binding constants K1–3 (CDCl3, mÀ1) between the molecules in-
pecially in the presence of cocatalyst 3b (35 and 49% con-
version with 3a and 56 and 83% conversion with 3b). The
kinetics of catalysts 5 is slower than that observed for com-
pounds 4. This observation can be rationalized by the strong
dimerization of 5, which embeds the active NH group.
volved in the ROP of lactide.[a,b]
Host/guest
Lactide
3a
3b
2
K1 =2
–
K1 =2
K1 =3
K3 =2
K2 =2[d]
K3 =6
K3 <1
K3 =6
ROH[c]
4a
K2 =2[d]
K3 =14
K3 =57
4b
Supramolecular insight into the ROP reactions: Partner or-
ganocatalysts have been poorly studied in the ROP reac-
tions. In this field of research, to the best of our knowledge,
only one binding constant of 39mÀ1 was measured in C6D6
between thiourea 2 and valerolactone,[9b] which represents a
rough model for the ROP of lactide in dichloromethane. In
a first approach, we can speculate that these hydrogen-
bonding compounds 4/5 and 3, respectively, activate the
[a] Titrations were monitored by 1H NMR spectroscopy at 208C, using a
2 mm concentration of host 2, 4a, or 4b and a 2–25 mm of guest lactide
3a, 3b, or ROH. [b] Estimated error Æ15%. [c] ROH is the initiator 4-
biphenylmethanol. [d] Host concentration was 20 mm.
At first, the expected interactions between cocatalysts and
reagents were investigated to prove the supramolecular acti-
vation (framed equilibria in Scheme 5). Indeed, catalysts 2,
4a, and 4b are weakly associated to the lactide in (1:1) stoi-
chiometry (K1 ꢀ2–3mÀ1). For (thio)amidoindoles 4a and 4b,
both NH groups from the (thio)amide and the indole moiet-
ies were involved in the complexation of lactide (see the
Supporting Information). As expected from hydrogen-bond-
donating properties, activated amide groups (Ddꢀ0.7 ppm)
were more affected than NH from the indole (Ddꢀ0.07–
0.2 ppm). These observations are in favor of the participa-
tion of both NH groups in the complexation of the C=O
group from the lactide.
Initiator ROH does not self-aggregate (see the Supporting
Information) and it interacts with the cocatalysts 3a and 3b
to form weak complexes (K2 =2mÀ1). These values corrobo-
rate the activation of the monomer and the initiator/growing
chain by tertiary amines 3, but the two interactions ap-
peared to be in the same order, contrary to the expected hy-
drogen-bond-donating classification (3a would be a lesser
hydrogen-bond donor than 3b).[9b,10b] Interestingly, the ob-
served difference in the catalyt-
ACHTUNGTRENNUNGmonoACHTUNGTRENNUNGmer and the initiator/growing chain in a similar
manner to the thiourea organocatalysts (Scheme 5, framed
equilibria). However, we have already demonstrated that or-
ganocatalysts 4 and 5 can self-aggregate in a dimeric form.
Thus an extra equilibrium, Kdimer, is present in the reaction
medium and should be considered in the mechanism
(Scheme 5).
Unexpected results reported in Table 3 (entry 8) prompt-
ed us to investigate the interactions between the different
reagents using the model catalysts 2, 4a, and 4b. Titrations
monitored by 1H NMR spectroscopy were achieved in
CDCl3 to demonstrate intermolecular interactions between
the multiple components of the reaction (Table 4): 1) cata-
lyst and lactide (equilibrium K1), 2) initiator and cocatalyst
3 (equilibrium K2), 3) catalyst and cocatalyst 3 (equilibrium
K3). Knowing that molecules 2–4 self-aggregate, the host
concentration was fixed at 4–8 mm to minimize this side
phenomenon.
ic power of amine 3a and
strained diamine 3b might rely
on a more complex hydrogen-
bonding system. Further inves-
tigations will be carried out.
Broadening the scope of the
possible hydrogen bonds, we
found that the catalysts 2, 4a,
and 4b do also interact with the
cocatalyst 3 within an undesired
hydrogen-bonding equilibrium
(K3), which could hamper the
ROP. Titrations concerning
ACHUTNGRENNUtG hioACHTUNTGRNEuNUGN rea 2 showed that this cat-
alyst interacts with the cocata-
lysts 3a and 3b. The corre-
sponding binding constants K3
are 2 and 6mÀ1, respectively,
and are in the same range as
the other hydrogen-bonding
constants K1 and K2. For 4a
and 4b, host–guest titrations
showed that NH groups (Dd
ꢀ0.06–0.9 ppm) as well as sev-
Scheme 5. Multiple hydrogen-bonding equilibria involving the reagents and catalysts 4+3 during the ring-
opening polymerization of lactide. (A similar process is postulated for catalysts 5.)
Chem. Eur. J. 2010, 16, 4196 – 4205
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4203

B. Bibal et al.
eral aromatic or aliphatic protons were largely affected (Dd
ꢀ0.1–0.5 ppm) by these interactions. In this equilibrium, hy-
drogen bonding probably occurs between the NH of the
(thio)amides 4 and the nitrogen atom of 3. In the presence
of amine 3a, the association constants K3 that involve cata-
lysts 4 were in the same range (ꢁ6mÀ1) as those found be-
tween the catalysts and their desired guests (K1 and K2),
whereas in the presence of 3b these K3 values were larger
(ꢀ14–57mÀ1). This result showed that partner catalysts can
self-inhibit through undesired hydrogen bonding. Neverthe-
less, as all supramolecular species are in equilibrium, the
progress of the ROP reaction appeared mainly driven by
the balance between the four equilibria present in solution
(Kdimer, K1–K3).
A remarkable value was noticed between organocatalyst
4b and (À)-sparteine 3b, which bind much more strongly
(K3 ꢀ57mÀ1). This result can explain the poorer catalytic
properties of thioamide 4b in the presence of the basic
amine 3b relative to those obtained with the cocatalyst 3a,
which is a weaker base (Table 3, entries 8 and 9).
Experimental Section
General preparation of the amidoindoles: A mixture of indole-2-carbox-
ylic acid (483 mg, 3.0 mmol) and thionyl chloride (2.2 mL, 30 mmol) in
CH2Cl2 (10 mL) was heated at reflux for 1 h under nitrogen and evapo-
rated to dryness. The residue was dried under vacuum for 1 h and dis-
solved in CH2Cl2 (10 mL), and a solution of the corresponding aniline
(2.5 mmol) and N,N-diisopropylethylamine (517 mg, 4.0 mmol) in CH2Cl2
(5 mL) was added dropwise. The resulting mixture was stirred at room
temperature for 15 min and heated at reflux for 16 h. After extraction
and concentration in vacuo, the crude solid was purified by column chro-
matography.
General preparation of the amidobenzimidazoles: A solution of the cor-
responding benzoyl chloride (3.3 mmol) in THF (5 mL) was added drop-
wise to
a solution of 2-amino-5,6-dimethylbenzimidazole (484 mg,
3.0 mmol) and N,N-diisopropylethylamine (1.0 mL, 5.7 mmol) in THF
(15 mL). The mixture was stirred at room temperature for 16 h under ni-
trogen atmosphere and evaporated to dryness. The residue was extracted
and eventually purified by column chromatography.
General preparation of the thioamidoindoles and thioamidobenzimid-
AHCTUNGERTGaNNUN zoles: A solution of the corresponding amidobenzimidazole or amido-
benzimidazole (0.5 mmol) and Lawessonꢁs reagent (0.5 mmol) in toluene
(15 mL) was heated at reflux for 16 h under nitrogen and evaporated to
dryness. The residue was purified by column chromatography.
Intermolecular interactions do exist between the different
hydrogen-bond donor and acceptors in solution. The dynam-
ic equilibria are favorably shifted towards the formation of
the polylactide in a controlled manner (predictable Mn and
narrow PDI), especially when the organocatalyst interacts
poorly with itself or with its cocatalyst.
NMR spectroscopic titrations: Deuterated solutions were freshly pre-
pared and dried in the presence of 4 ꢄ molecular sieves. Association con-
stants between host and guest (K1–K3) as well as the dimerization binding
constants (Kdimer) were determined using titrations monitored by
1H NMR spectroscopy (host signals) in CDCl3. For determination of K1–
K3, a solution (100 mL) of host (ꢀ20 mm) was introduced in each NMR
spectroscopic tube (12 to 15 experiments per titration). Increasing ali-
quots of guest stock solution (ꢀ70 mm) were added and the total volume
(500 mL) was adjusted with CDCl3. The titration data (Dd ppm versus
guest concentration) were fitted using the nonlinear curve-fitting proce-
dure with a (1:1) binding equation using the WinEqNMRprogram.[27]
Concerning Kdimer evaluation, a stock solution of host (ꢀ30 mm) in
CDCl3 was used to prepare the diluted NMR spectroscopy tubes (12 to
15) required for each titration. The titration data (Dd ppm versus host
concentration) were fitted with a dimerization model using Excel.[28] For
thiourea 2, the isodesmic aggregation model was more accurate.[29]
Conclusion
Activated (thio)amides that bear a second hydrogen-bond
donor group, 4 and 5, were straightforwardly synthesized
and tested as organocatalysts for the ROP of lactide. Hydro-
gen-bonding networks within crystal structures of com-
pounds 4a, 4b, 5c, and 5d revealed tight dimers, which, in
some cases, embedded the NH amide group necessary for
catalysis. Notably, the first X-ray structure of thioamidoben-
zimidazole tautomer 5d is described. Investigation of the
self-association properties in solution proved that molecules
4 and 5 form weak and strong dimers, respectively. In the
case of series 5, these self-associations probably decrease
the availability of the active NH group for the monomer ac-
tivation, whatever the nature of cocatalyst 3. In the efficient
series 4, we showed that the catalytic power of the NH
group is better when the aromatic substituent is electrodefi-
cient and when NH is an amide versus thioamide.
We demonstrated that the hydrogen-bonding partner cata-
lysts do play their role by activating the monomer and the
polymer growing chain, respectively. Notably, for the first
time, undesirable hydrogen bonds between the partner cata-
lysts were highlighted. Some associations were strong and
thus inhibit the catalytic power of the new (thio)amides. In
a general manner, as hydrogen-bonding catalysts may be in-
volved in multiple equilibria in the reaction medium, bind-
ing partners should be judiciously chosen to tune the out-
come of the polymerization.
Typical experimental procedure for ROP reactions: Under nitrogen and
in a dry Schlenk tube, the organocatalyst (35 mmol), the initiator (4-bi-
phenylmethanol, 35 mmol), lactide (700 mmol), 4 ꢄ molecular sieves
(5 beads), dry dichloromethane (1 mL), and amine 3 as a cocatalyst
(35 mmol) were successively introduced. The reaction mixture was stirred
at 208C under nitrogen for the indicated period. The reaction mixture
was filtered and concentrated in vacuo. Conversion was determined by
1H NMR spectroscopy and integrating the signals of the methine proton
in both the residual lactide and the polymer. Polymer molar masses and
the dispersity index were measured by size-exclusion chromatography
(SEC) using a PL-GPC50 Plus apparatus equipped with RI and UV de-
tectors and Tosoh G4000HXL, G3000HXL, and G2000HXL columns
(eluent: THF, flow rate 1.0 mLminÀ1, temperature: 408C, calibrated with
polystyrene standards).
X-ray structural analysis: Detailed crystal structures, cell parameters, and
R values for 4a, 4c, 5c, and 5d are reported in the Supporting Informa-
tion. CCDC-749820 (5c), 749821 (4c), 749822 (5d), and 749823 (4a) con-
tain the supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallographic
Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
4204
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4196 – 4205

Properties of (Thio)Amides
FULL PAPER
Mukherjee, T. N. Mꢆller, J. Lex, Angew. Chem. 2005, 117, 817–821;
Angew. Chem. Int. Ed. 2005, 44, 807–811; c) M. S. Taylor, E. N. Ja-
cobsen, Angew. Chem. 2006, 118, 1550–1573; Angew. Chem. Int. Ed.
2006, 45, 1520–1543; d) S. J. Connon, Chem. Eur. J. 2006, 12, 5418–
5427.
Acknowledgements
The Agence Nationale de la Recherche (ANR) and the Chimie des Pro-
cꢂdꢂs et du Developpement Durable Program (ANR-07-CP2D-15) are
gratefully acknowledged for funding (S.K. and J. K. Fellowships).
[13] a) T. Okino, Y. Hoashi, Y. Takemoto, J. Am. Chem. Soc. 2003, 125,
12672–12673; b) Y. Hoashi, T. Okino, Y. Takemoto, Angew. Chem.
2005, 117, 4100–4103; Angew. Chem. Int. Ed. 2005, 44, 4032–4035.
[14] a) C. Givelet, B. Tinant, L. Van Meervelt, T. Buffeteau, N. March-
and-Geneste, B. Bibal, J. Org. Chem. 2009, 74, 652–659; b) M.-L.
Dumartin, C. Givelet, P. Meyrand, B. Bibal, I. Gosse, Org. Biomol.
Chem. 2009, 7, 2725–2728; c) C. Givelet, T. Buffeteau, F. Arnaud-
Neu, V. Hubscher-Bruder, B. Bibal, J. Org. Chem. 2009, 74, 5059–
5062.
[15] S. Koeller, J. Kadota, A. Deffieux, F. Peruch, S. Massip, J.-M. Lꢂger,
J.-P. Desvergne, B. Bibal, J. Am. Chem. Soc. 2009, 131, 15088–
15089.
[16] Adapted from: C. Kuehm-Caubꢃre, P. Caubꢃre, B. Jamart-Grꢂgoire,
A. Negre-Salvayre, D. Bonnefont-Rousselot, J.-G. Bizot-Espiard, B.
Pfeiffer, D.-H. Caignard, B. Guardiola-Lemaꢇtre, P. Renard, J. Med.
Chem. 1997, 40, 1201–1210.
[1] a) F. Nederberg, E. F. Connor, M. Mçller, T. Glauser, J. L. Hedrick,
Angew. Chem. 2001, 113, 2784–2787; Angew. Chem. Int. Ed. 2001,
40, 2712–2715; b) N. E. Kamber, W. Jeong, R. M. Waymouth, R. C.
Pratt, B. G. G. Lohmeijer, J. L. Hedrick, Chem. Rev. 2007, 107,
5813–5840; c) D. Bourissou, S. Moebs-Sanchez, B. Martin-Vaca, C.
R. Chim. 2007, 10, 775–794; d) A. P. Dove, Chem. Commun. 2008,
6446–6470.
[2] a) F. Nederberg, B. G. G. Lohmeijer, F. Leibfarth, R. C. Pratt, J.
Choi, A. P. Dove, R. M. Waymouth, J. L. Hedrick, Biomacromole-
cules 2007, 8, 153–160; b) R. C. Pratt, F. Nederberg, R. M. Way-
mouth, J. L. Hedrick, Chem. Commun. 2008, 114–116.
[3] a) H. R. Kricheldorf, I. Kreiser, Makromol. Chem. 1987, 188, 1861–
1873; b) F. Sanda, H. Sanada, Y. Shibasaki, T. Endo, Macromole-
cules 2002, 35, 680–683; c) X. Lou, C. Detrembleur, R. Jꢂrꢅme,
Macromolecules 2002, 35, 1190–1195; d) P. V. Persson, J. Schrçder,
K. Wickholm, E. Hedenstrçm, T. Iversen, Macromolecules 2004, 37,
5889–5893; e) S. Gazeau-Bureau, D. Delcroix, B. Martin-Vaca, D.
Bourissou, C. Navarro, S. Magnet, Macromolecules 2008, 41, 3782–
3784.
[4] O. Coulembier, D. P. Sanders, A. Nelson, A. N. Hollenbeck, H. W.
Horn, J. E. Rice, M. Fujiwara, P. Dubois, J. L. Hedrick, Angew.
Chem. 2009, 121, 5272–5275; Angew. Chem. Int. Ed. 2009, 48, 5170–
5173.
[17] Adapted from: J. Sluka, J. Danek, P. Bedrnik, Z. Budesinsky, Col-
lect. Czech. Chem. Commun. 1981, 46, 2703–2708.
[18] N. J. Leonard, D. Y. Curtin, K. M. Beck, J. Am. Chem. Soc. 1947, 69,
2459–2461.
[19] a) M. Jesberger, T. P. Davis, L. Barner, Synthesis 2003, 1929–1958;
b) T. Ozturk, E. Ertas, O. Mert, Chem. Rev. 2007, 107, 5210–5218.
[20] a) E. V. Nosova, G. N. Lipunova, A. A. Laeva, V. N. Charushin,
Russ. J. Org. Chem. 2005, 41, 1671–1677; b) J. P. Powers, S. Li, J. C.
Jaen, J. Liu, N. P. C. Walker, Z. Wang, H. Wesche, Bioorg. Med.
Chem. Lett. 2006, 16, 2842–2845.
[21] a) S. Ianelli, M. Nardelli, D. Belletti, B. Jamart-Grꢂgoire, C. Cau-
bꢃre, P. Caubꢃre, Acta Crystallogr. Sect. C 1995, 51, 1341–1345;
b) G. W. Bates, M. E. Light, M. Albrecht, P. A. Gale, J. Org. Chem.
2007, 72, 8921–8927.
[22] a) J. Li, D.-Q. Shi, W.-X. Chen, Heterocycles 1997, 45, 2381–2384;
b) L. Jing, F. He-Liang, S. Jie, C. Wei-Xing, Acta Crystallogr. Sect. C
1997, 53, 320–322.
[23] a) B. Yang, Y.-F. Xia, S. Bi, S.-S. Zhang, Acta Crystallogr. Sect. E
2006, 62, o4200–o4202; b) M. G. Fisher, P. A. Gale, M. E. Light,
New J. Chem. 2007, 31, 1583–1584.
[24] a) L. Leiserowitz, G. M. J. Schmidt, J. Chem. Soc. A 1969, 2372–
2382; b) L. Leiserowitz, M. Tuval, Acta Crystallogr. Sect. B 1978, 34,
1230–1247; c) P. Dauber, A. T. Haggler, Acc. Chem. Res. 1980, 13,
105–112; d) R. Taylor, O. Kennard, W. Versichel, Acta Crystallogr.
Sect. B 1984, 40, 280–288; e) M. C. Etter, Acc. Chem. Res. 1990, 23,
120–126.
[25] a) C. N. Sundaresan, S. Dixit, P. Venugopalan, J. Mol. Struct. 2004,
693, 205–209; b) G. Pavlovic, V. Tralic-Kulenovic, M. Vinkovic, D.
Vikic-Topic, I. Matanovic, Z. Popovic, Struct. Chem. 2006, 17, 275–
285.
[26] a) T. Sifferlen, M. Rueping, K. Gademann, B. Jaun, D. Seebach,
Helv. Chim. Acta 1999, 82, 2067–2093; b) R. Frank, M. Jakob, F.
Thunecke, G. Fischer, M. Schutkowski, Angew. Chem. 2000, 112,
1163–1165; Angew. Chem. Int. Ed. 2000, 39, 1120–1122.
[27] J. M. Hynes, J. Chem. Soc. Dalton Trans. 1993, 311–312.
[28] P. R. Mitchell, H. Sigel, Eur. J. Biochem. 1978, 88, 149–154.
[29] T. H. Lilley, H. Linsdell, A. Maestre, J. Chem. Soc. Faraday Trans.
1992, 88, 2865–2870.
[5] a) F. Nederberg, E. F. Connor, T. Glauser, J. L. Hedrick, Chem.
Commun. 2001, 2066–2067; b) reference [1a].
[6] M. Myers, E. F. Connor, T. Glauser, A. Mock, G. Nyce, J. L. He-
drick, J. Polym. Sci. Polym. Chem. Ed. 2002, 40, 844–851.
[7] a) E. F. Connor, G. W. Nyce, M. Myers, A. Mçck, J. L. Hedrick, J.
Am. Chem. Soc. 2002, 124, 914–915; b) O. Coulembier, B. G. G.
Lohmeijer, A. P. Dove, R. C. Pratt, L. Mespouille, D. A. Culkin, S. J.
Benight, P. Dubois, R. M. Waymouth, J. L. Hedrick, Macromolecules
2006, 39, 5617–5628; c) A. P. Dove, R. C. Pratt, B. G. G. Lohmeijer,
D. A. Culkin, E. C. Hagberg, G. W. Nyce, R. M. Waymouth, J. L.
Hedrick, Polymer 2006, 47, 4018–4025; d) W. Jeong, J. L. Hedrick,
R. M. Waymouth, J. Am. Chem. Soc. 2007, 129, 8414–8415.
[8] a) L. Zhang, F. Nederberg, R. C. Pratt, R. M. Waymouth, J. L. He-
drick, C. G. Wade, Macromolecules 2007, 40, 4154–4158; b) L.
Zhang, F. Nederberg, J. M. Messman, R. C. Pratt, J. L. Hedrick,
C. G. Wade, J. Am. Chem. Soc. 2007, 129, 12610–12611.
[9] a) R. C. Pratt, B. G. G. Lohmeijer, D. A. Long, R. M. Waymouth,
J. L. Hedrick, J. Am. Chem. Soc. 2006, 128, 4556–4557; b) B. G. G.
Lohmeijer, R. C. Pratt, F. Leibfarth, J. W. Logan, D. A. Long, A. P.
Dove, F. Nederberg, J. Choi, C. G. Wade, R. M. Waymouth, J. L. He-
drick, Macromolecules 2006, 39, 8574–8583; c) X. Sun, J. P. Gao,
Z. Y. Wang, J. Am. Chem. Soc. 2008, 130, 8130–8131.
[10] a) A. P. Dove, R. C. Pratt, B. G. G. Lohmeijer, R. M. Waymouth,
J. L. Hedrick, J. Am. Chem. Soc. 2005, 127, 13798–13799; b) R. C.
Pratt, B. G. G. Lohmeijer, D. A. Long, P. N. P. Lundberg, A. P. Dove,
H. Li, C. G. Wade, R. M. Waymouth, J. L. Hedrick, Macromolecules
2006, 39, 7863–7871.
[11] A. Chuma, H. W. Horn, W. C. Swope, R. C. Pratt, L. Zhang, B. G. G.
Lohmeijer, C. G. Wade, R. M. Waymouth, J. L. Hedrick, J. E. Rice,
J. Am. Chem. Soc. 2008, 130, 6749–6754.
[12] Recent reviews about bifunctional thioureas: a) Y. Takemoto, Org.
Biomol. Chem. 2005, 3, 4299–4306; b) A. Berkessel, F. Cleemann, S.
Received: October 21, 2009
Published online: March 16, 2010
Chem. Eur. J. 2010, 16, 4196 – 4205
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4205