Self-association of bis-dendritic organogelators: The effect of dendritic architecture on multivalent cooperative interactions

Article abstract of DOI:10.1002/chem.200902575

A series of bis-dendritic gelators consisting of a benzamide dendron and an alkyl dendron were synthesized to investigate the dendritic effect on self-assembly. The gelators with a firstgeneration benzamide (benzamide-G1) dendron or a first-generation alkyl (alkyl-G1) dendron formed stable gels in most aromatic solvents, and their self-assembled fibrillar networks were imaged by electron microscopy. The unbranched molecule (G0-G0) or the molecule possessing a second-generation benzamide (benzamide-G2) dendron did not form gels. Differential scanning calorimetry, powder X-ray diffraction, and Fourier transform IR studies revealed that introduction of a dendritic branch strongly affected the molecular packing as well as the strength of intermolecular interactions. Furthermore, concentration-dependent diffusion coefficient measurements and the evaluation of association constants by ¹H NMR spectroscopy indicated that bis-dendritic gelators with a benzamide-G1 dendron possessed high association constants and formed large aggregates, whereas molecules with a single benzamide formed dimers in chloroform. The formation of self-assembled fibrillar networks was driven by the multivalent and cooperative hydrogen bonding observed in the benzamide- G1 dendrons. π- π stacking of aromatic groups and van der Waals interactions between alkyl chains also played roles in the self-assembly process, thus indicating that a spatial balance between two dendrons is important.

Full text of DOI:10.1002/chem.200902575

FULL PAPER
DOI: 10.1002/chem.200902575
Self-Association of Bis-Dendritic Organogelators: The Effect of Dendritic
Architecture on Multivalent Cooperative Interactions
Myungeun Seo, Jung Hak Kim, Jisung Kim, Nojin Park, Jeyoung Park, and
Sang Youl Kim*[a]
Abstract: A series of bis-dendritic gela-
tors consisting of a benzamide dendron
and an alkyl dendron were synthesized
to investigate the dendritic effect on
self-assembly. The gelators with a first-
generation benzamide (benzamide-G1)
diffraction, and Fourier transform IR
studies revealed that introduction of a
dendritic branch strongly affected the
molecular packing as well as the
strength of intermolecular interactions.
Furthermore, concentration-dependent
diffusion coefficient measurements and
the evaluation of association constants
by 1H NMR spectroscopy indicated
that bis-dendritic gelators with a benz-
sociation constants and formed large
aggregates, whereas molecules with a
single benzamide formed dimers in
chloroform. The formation of self-as-
sembled fibrillar networks was driven
by the multivalent and cooperative hy-
drogen bonding observed in the benz-
amide-G1 dendrons. p–p stacking of
aromatic groups and van der Waals in-
teractions between alkyl chains also
played roles in the self-assembly pro-
cess, thus indicating that a spatial bal-
ance between two dendrons is impor-
tant.
dendron or
a first-generation alkyl
(alkyl-G1) dendron formed stable gels
in most aromatic solvents, and their
self-assembled fibrillar networks were
imaged by electron microscopy. The
unbranched molecule (G0-G0) or the
molecule possessing a second-genera-
tion benzamide (benzamide-G2) den-
dron did not form gels. Differential
scanning calorimetry, powder X-ray
AHCTUNGTRENGNaUN mide-G1 dendron possessed high as-
Keywords: benzamide
mers · gels · hydrogen bonds · self-
assembly
·
dendri-
AHCTUNGTRENNUNG
Introduction
Dendritic architectures have also been used for construct-
ing organogelators, which, at low concentrations, immobilize
organic liquids by forming self-assembled fibrillar net-
works.[11] In most cases, hydrogen-bonding moieties intro-
duced at the focal point or along the branches[12–15] of the
dendron complement nonpolar interactions in the bulky
nonpolar dendron to stabilize the network.[16] On the other
hand, Stupp and co-workers pioneered self-assembly of den-
dritic gelators containing hydroxyl groups at the periphery
and demonstrated various functionalities can be embedded
in the architecture.[17]
Introducing hydrogen-bonding moieties at the periphery
of a dendron allows cooperative multivalent bonding be-
tween many proximal moieties. In nature, multivalent inter-
actions between ligands and receptors are frequently ob-
served in supramolecular assemblies[18] because cooperative
binding enhances ligand binding affinity, known as the ꢀden-
dritic effectꢁ.[19] Quantitative descriptions of the self-associa-
tion process are crucial for understanding the dendritic
effect at the molecular level. Because the gelation process
is, in principle, an aggregation process involving a large
number of molecules, it requires models that are more so-
Dendritic architectures provide multiple reactive sites on
the interior and periphery of macromolecules together with
precise control over the number and chemical composition
of these sites.[1] The development of synthetic strategies for
dendritic molecules has focused on supramolecular chemis-
try in which a well-defined molecular structure is necessary
for the design of intermolecular interactions.[2] Various den-
dritic hybrid structures have been described, including den-
dritic wedges,[3–5] dendritic amphiphiles,[6] diblock dendrim-
ers,[7] and dendron-containing block copolymers.[8–10] Globu-
lar dendritic structures have been found useful for the for-
mation of supramolecular assemblies.
[a] Dr. M. Seo, J. H. Kim, J. Kim, N. Park, J. Park, Prof. S. Y. Kim
Department of Chemistry, KAIST
Gwahangno 335, Yuseong-gu, Daejeon 305-701 (Korea)
Fax : (+82)42-350-8177
E-mail: kimsy@kaist.ac.kr
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200902575.
Chem. Eur. J. 2010, 16, 2427 – 2441
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2427

phisticated than the conventional dimerization model used
to describe binding between host and guest molecules. The
indefinite self-association models described by Martin[20]
have been applied to the gelation process in a few reports,[21]
although these models were originally developed to describe
p–p stacking interactions between adjacent DNA base pairs.
To investigate the molecular basis for the dendritic effect
in self-assembly, we synthesized a series of bis-dendritic
molecules consisting of benzamide and alkyl dendrons. The
benzamide moiety, a hydrogen-bonding motif used in crystal
engineering, was selected to investigate the effect of branch-
es on the molecular structure.[22] The hydrogen-bonding pat-
terns of benzamide are described by a combination of two
supramolecular synthons, R22(8) (a face-to-face hydrogen
bond utilizing syn-hydrogen atoms) and C(4) (a side-to-side
hydrogen bond utilizing anti-hydrogen atoms), which yield a
two-dimensional hydrogen-bonded network.[23] A previous
attempt to construct an organogelator with the benzamide
moiety resulted in a gelator with modest gelation properties
due to the tendency of benzamide to form highly crystalline
aggregates.[24] Modifying benzamide at the ortho position
with a trifluoromethyl group improved the solubility while
maintaining the hydrogen-bonding ability.[25] By introducing
the modified benzamide moiety at the dendritic periphery,
we were able to make noncovalently bound polymeric nano-
particles by means of supramolecular chain collapse assisted
by hydrogen bonding.[26]
The association of the dendritic benzamide moieties is ex-
pected to be multivalent and cooperative because they con-
tain multiple hydrogen-bonding sites, each of which forms a
two-dimensional hydrogen bond. The alkyl dendron should
also be important in forming aggregates because it provides
van der Waals interaction between alkyl chains and protects
benzamide hydrogen bonds from the solvent. The spatial
balance between two dendrons may play another role in
self-assembly, by balancing hydrogen bonding and van der
Waals interaction and allowing physical gelation at low con-
centrations.
pared by following the synthetic routes depicted in
Scheme 1.
Previously, we described the unusual crystal growth of 4-
dodecyloxy-2-trifluoromethylbenzamide (G0-G0), which
forms parallelogrammatic pipes by means of a polymorphic
transition.[25] Based on the molecular architecture of G0-G0,
the first-generation benzamide dendron (benzamide-G1)
was synthesized by nucleophilic aromatic substitution
(SNAr) between the 4-nitro-2-trifluoromethylbenzonitrile (1)
and 3,5-dihydroxybenzyl alcohol.[26] The main product re-
sulted from displacement of the nitro group (which was acti-
vated by the electron-withdrawing nitrile and trifluorometh-
yl groups) by phenoxides in the 3,5-dihydroxybenzyl alcohol.
Small quantities of side products consisting of 1- or 3-substi-
tuted compounds were also produced. Hydrolysis of the ni-
trile groups to benzamide groups and esterification of the
benzyl alcohol with 1-dodecanoic acid yielded the desired
compound, G1-G0.
To synthesize the second-generation benzamide dendron
(benzamide-G2), conversion of benzyl alcohol into benzyl
bromide was performed using the convergent approach de-
scribed for the synthesis of Frꢃchet-type benzyl ether den-
drons.[27] Reaction of 4 with carbon tetrabromide and triphe-
nylphosphine in THF produced the desired compound (7).
Benzamide-G2 was obtained by further SN2 reaction of 7
with 3,5-dihydroxybenzyl alcohol and hydrolysis of the ni-
trile groups into benzamide groups. Esterification between
the benzyl alcohol and 1-dodecanoic acid yielded the de-
sired G2-G0.
3,5-Bis(dodecyloxy)benzoic acid,[28] a double-alkyl-tailed
compound joined by a carboxylic acid moiety at the focal
point, was chosen for the construction of G1-G1. In the case
of G0-G1, synthesis began with dimethyl 5-hydroxyisophtha-
late. After introduction of the benzamide group, hydrolysis
of the ester groups liberated the acid moieties, and subse-
quent esterification with 1-dodecanol yielded the desired
compound. Interestingly, during the hydrolysis of the nitrile
group, 3-(4-carbamoyl-3-trifluoromethylphenoxy)-5-methoxy-
carbonylbenzoic acid containing one free carboxylic acid
group was produced as a side product. The compound was
also converted into 5-(4-carbamoyl-3-trifluoromethylphe-
noxy)isophthalic acid (18).
In this study, the self-association of bis-dendritic gelators
was investigated by varying the generation of the dendron
on either side of the bis-dendritic molecules. By employing
various characterization techniques for xerogel and solution
states, we show that the existence of a dendritic branch pro-
duced a physical gel consisting of nanoscopic fibers, but the
gelation properties and intermolecular interactions were
found to be strongly dependent on the generation. The self-
association of the bis-dendritic gelators was fit to the indefi-
nite self-association models, and the results clearly demon-
strated that the dendritic effect plays a key role in the self-
association of bis-dendritic gelators by providing multivalent
and cooperative interactions.
Thermal properties: The thermal transitions of the products
were investigated by differential scanning calorimetry
(DSC) and polarized optical microscopy (POM) equipped
with a heating stage. Figure 1a–e shows DSC thermograms
of the bis-dendritic gelator molecules during the first heat-
ing cycle. Thermograms were recorded at a heating rate of
108Cminꢀ1 under a N2 atmosphere. Thermograms of the
first and second heating/cooling cycles are provided in the
Supporting Information. Table 1 shows a summary of the
transition temperatures and corresponding transition enthal-
pies.
Results and Discussion
According to the previous analysis, G0-G0 possesses two
polymorphic forms, all-trans L and gauche-rich S.[25a] The
DSC thermogram of a solid of G0-G0 from toluene solution
Synthesis: A small library of bis-dendritic molecules, consist-
ing of benzamide dendrons and alkyl dendrons, was pre-
2428
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2427 – 2441

Self-Association of Bis-Dendritic Organogelators
FULL PAPER
Scheme 1. Synthetic routes to the bis-dendritic molecules. a) 1-Dodecanol, NaH, THF, reflux, 15 h. b) KOH, H2O, EtOH, 608C, 10 h. c) 3,5-Dihydroxy-
benzyl alcohol, K2CO3, DMSO, RT, 10 h. d) K2CO3, H2O2 (aq), DMSO, RT, 2 h. e) 1-Dodecanoic acid, EDC, DMAP, THF, RT, 4 h. f) CBr4, Ph3P, THF,
RT, 30 min. g) 3,5-Dihydroxybenzyl alcohol, K2CO3, acetone, reflux, 48 h. h) H2SO4 (cat.), MeOH, reflux, 20 h. i) 1-Bromododecane, K2CO3, acetone,
reflux, 48 h. j) KOH, EtOH, reflux, 2 h. k) 5, EDC, DMAP, THF, RT, 4 h. l) Dimethyl 5-hydroxyisophthalate, K2CO3, DMSO, RT, 3 h. m) LiOH, THF/
MeOH/H2O 6:3:1, RT, 4 h. n) 1-Dodecanol, EDC, DMAP, DMF, RT, 18 h. (EDC=1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride,
DMAP=4-dimethylaminopyridine.)
(Figure 1d) shows that the com-
pound possesses an
S poly-
morphic form, an intercalated
bilayer lamellar structure that
consists of stronger face-to-face
hydrogen bonds between benz-
A
a
more
gauche-rich configuration of
alkyl chains than the L poly-
morphic form. It seems that the
melting transition at 1038C cor-
responds to the cleavage of hy-
drogen bonds between benz-
AHCTUNGTERGaNNUN mide groups and the premelt-
ing transition at 908C corre-
sponds to the transition be-
tween alkyl-chain configura-
tions. No morphological change
was observed between 90 and
1038C.
Figure 1. DSC thermograms of the bis-dendritic gelator molecules a) G1-G1, b) G2-G0, c) G1-G0, d) G0-G0,
and e) G0-G1 during the first heating cycle. Thermograms were recorded at a heating rate of 108Cminꢀ1 in a
N2 atmosphere. POM images of compounds f) G1-G1, g) G1-G0, and h) G0-G1 showing liquid-crystalline tex-
tures. Scale bars represent 20 mm.
Compared to the thermal be-
havior of G0-G0, G1-G0
Chem. Eur. J. 2010, 16, 2427 – 2441
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2429

S. Y. Kim et al.
Table 1. Thermal properties of the bis-dendritic gelators Gn-Gm.[a]
and crystallization was also observed at 498C on cooling.
During the second heating cycle, G0-G1 showed two endo-
thermic transitions consisting of premelting (718C) and
melting (888C), which were very similar to the transitions in
the thermogram of G1-G1 during the first heating cycle.
Similar liquid crystalline textures were also observed be-
tween the temperatures (Figure 1h). These results suggest
that introduction of a branch on either side of the dendri-
mer broke the symmetry of crystalline G0-G0 and promoted
anisotropic intermolecular interactions. Whereas an increase
in the number of benzamide groups enhanced the intermo-
lecular association to a certain extent, reflected in an in-
crease in melting transition temperature, an appropriate
number of alkyl chains was also important for stabilizing hy-
drogen bonds.
Entry 1st heating
1st cooling
2nd heating
2nd cooling
G1-
G1
G2-
G0
G1-
G0
124.26 (28.78)
138.39 (52.96)
n.d.[b]
90.60 (ꢀ26.59)
136.85 (35.90)
n.d.[b]
n.d.[b]
n.d.[b]
n.d.[b]
n.d.[b]
121.42 (21.69) n.d.[b]
n.d.[b]
n.d.[b]
G0-
G0
G0-
G1
90.26 (35.71)
102.50 (91.34)
79.81 (1.79)
102.67 (92.81)
71.41 (29.28)
88.21 (16.57)
79.20 (ꢀ84.97)
78.17 (ꢀ84.06)
49.20 (ꢀ49.83)
91.65 (64.70) 49.20 (ꢀ48.50)
[a] Temperatures are expressed in 8C. Transition enthalpies [Jgꢀ1] are
shown in parentheses. Positive and negative values represent endother-
mic and exothermic transitions, respectively. [b] n.d.: not determined.
showed a very weak and broad transition at 1218C that was
classified as a melting transition (Figure 1c). When observed
by POM, G1-G0 showed a morphological change to a
liquid-crystalline state depicted in Figure 1g. The texture
persisted as the sample was cooled to room temperature.
However, the sample melted completely above 1258C and
remained amorphous when subjected to further temperature
changes. These results suggest that the initial alkyl-chain
transition was followed by cleavage of hydrogen bonds be-
tween benzamide groups. An increase in the melting transi-
tion temperature with a decrease in the corresponding tran-
Gelation properties: All bis-dendritic molecules were sub-
jected to gelation tests in a variety of organic solvents. Each
compound was combined with solvent in a screw-capped
sample tube, and heated until the solid fully dissolved. The
solution was then cooled to room temperature, and gelation
was monitored by inverting the sample tube to see whether
the solution flows or not. The observed gelation behavior of
each solution is summarized in Table 2.
Table 2. Gelation of organic solvents by the bis-dendritic gelators Gn-
Gm.[a]
sition enthalpy suggested that the benzACHTNUTRGNEUNGamide-G1 dendron
increased the amount of intermolecular interactions but de-
creased the crystallinity.
G0-G0
G1-G0
G2-G0
G1-G1
G0-G1
n-hexane
benzene
toluene
P
S
S
S
S
S
S
S
S
S
I
P
G
G
G
S
S
S
S
S
I
P
P
P
P
I
G
G
G
G
G
G
S
S
S
S
G
S
S
S
S
S
S
S
S
S
POM imaging indicated that G2-G0 became isotropic at
928C. However, G2-G0 did not show any detectable thermal
transitions in the DSC thermograms during the first and
second heating/cooling cycles (Figure 1b). This result implies
that the benzamide groups in the benzamide-G2 dendron of
G2-G0 do not interact cooperatively and do not increase the
strength of intermolecular interactions even though the den-
dron has four potential interaction sites nearby. The benz-
amide groups in the benzamide-G2 dendron appeared to
have different orientations due to steric repulsion, and the
intermolecular association of the benzamide-G2 dendron
seemed to involve several molecules. Thus, the melting tran-
sition temperature of G2-G0 mainly reflects monomeric dis-
sociation of benzamide groups.
In Figure 1a, G1-G1 was characterized by the highest
melting transition temperature in the series, 1398C, which
suggests the importance of spatial balance in the molecular
structure. The melting transition was accompanied by anoth-
er premelting transition at 1258C, and POM analysis indicat-
ed that a liquid-crystalline state similar to that observed for
G1-G0 appeared during the premelting transition tempera-
ture (Figure 1g). The state became isotropic above 1408C.
During the second heating cycle, the compound showed an
exothermic transition corresponding to crystallization at
918C and an endothermic transition corresponding to melt-
ing at 1378C.
xylene
chlorobenzene
1-butanol
chloroform
ethyl acetate
acetonitrile
DMSO
I
S
S
S
[a] Tested at 3 mgmLꢀ1 concentration. G: gel. P: precipitation. S: solu-
tion. I: insoluble.
Compound G0-G0 showed good solubility in a range of
organic solvents with the exception of n-hexane, in which
crystallization occurred. While G0-G0 showed no gelation in
any solvent tested, compounds possessing a single branch on
the benzamide dendron (G1-G0), the alkyl dendron (G0-
G1), or both dendrons (G1-G1) formed a gel in most non-
polar solvents. Compound G1-G0 gelled in the aromatic sol-
vents toluene, ethylbenzene, and chlorobenzene but was in-
soluble in benzene or n-hexane. Compound G1-G0 dis-
solved in the polar solvents such as ethyl acetate and
DMSO presumably due to the disruption of hydrogen bond-
ing. However, G2-G0 showed much poorer solubility in aro-
matic solvents, in which precipitation occurred when the
heated solution was cooled to room temperature. No gela-
tion properties were observed for G2-G0. Again, these
trends in gelation properties indicate that nonspecific hydro-
The melting transition temperature of G0-G1 was ob-
served at 928C during the first heating cycle (Figure 1e),
2430
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2427 – 2441

Self-Association of Bis-Dendritic Organogelators
FULL PAPER
gen bonding in G2-G0 involved a number of molecules. As
a result, insoluble aggregates were produced because the
relatively small alkyl groups could not provide the necessary
solubility in nonpolar solvents.
Compound G1-G1 showed not only enhanced solubility
relative to G1-G0, but also underwent gelation in the widest
range of solvents, including aromatic solvents, n-hexane, and
1-butanol. The origin of the wide gelation window of G1-G1
was attributed to the well-matched spatial balance between
the benzamide dendron and the alkyl dendron, which coop-
eratively provided strong intermolecular interactions con-
sisting of hydrogen bonding, p–p stacking, and van der
Waals interactions between alkyl chains.
Compound G0-G1 showed the best solubility in the series
because the relatively large alkyl dendron seemed to im-
prove solubility in organic solvents. It only formed a stable
gel in n-hexane, presumably due to the favorable packing of
the alkyl-G1 dendron. It should be noted that the com-
pounds possessing benzamide-G1 dendrons formed gels in
aromatic solvents, and the compounds possessing alkyl-G1
dendrons formed gels in n-hexane.
The microscopic structure of the gel was investigated by
field-emission SEM (FE-SEM). Figure 2 shows representa-
tive FE-SEM images of the xerogel of G1-G0, G1-G1, and
G0-G1 formed in various solvents (additional SEM and
TEM images are presented in the Supporting Information).
These images clearly show that well-defined nanofibers
were formed by self-assembly of the bis-dendritic gelators
that contained at least one branch on one side of the bis-
dendritic gelator, which yielded a physically cross-linked fi-
brillar network in three dimensions. The diameter of the
nanofibers ranged from several dozen to more than a hun-
dred nanometers, which suggests that they are actually bun-
dles of smaller fibers. Interestingly, the self-assembled struc-
tures of G1-G1 shown in Figure 2d–f look like ribbons
rather than fibers: they are straight and rarely twisted. Com-
pared with G1-G0 and G0-G1, the spatially well-balanced
molecular structure of G1-G1 seemed to provide more fa-
vorable packing of alkyl chains, strengthening and straight-
ening the self-assembled structure while reducing the curva-
ture. An SEM image of solid G0-G1 obtained by evapora-
tion of toluene also revealed a highly entangled fibrillar net-
work. This observation implied that self-association of G0-
G1 also occurs in toluene, similar to G1-G0 and G1-G1, al-
though the concentration at which the microscopic structure
appears has not been determined. Compound G0-G0, ob-
tained by evaporation of toluene, only produced crystalline
solids as depicted in Figure 2i. It seems that the introduction
of a branch was essential for producing one-dimensional
growth of self-assembled structures instead of three-dimen-
sional growth of crystalline solids. The G2-G0 precipitate
formed in toluene showed a two-dimensional sheetlike mor-
phology on the micrometer scale, as shown in Figure 2j.
Analysis of self-assembled structures: The dependence of
the self-association properties on dendron generation for
the bis-dendritic gelator molecules was further investigated
for G1-G1, G1-G0, G0-G0, and G0-G1 in toluene, but not
for G2-G0 because it failed to form self-assembled struc-
tures in comparable conditions. Figure 3 shows powder
Figure 2. FE-SEM images of the xerogel of G1-G0, G1-G1, and G0-G1
formed in various solvents (a–g), and solids of G0-G1, G0-G0, and G2-
G0 obtained by evaporation of a solution drop on a Si wafer. a),b) G1-
G0 in toluene. c) G1-G0 in xylene. d) G1-G1 in benzene. e) G1-G1 in tol-
uene. f) G1-G1 in 1-butanol. g) G0-G1 in n-hexane. h) G0-G1 in toluene.
i) G0-G0 in toluene. j) G2-G0 in chlorobenzene.
Figure 3. Powder XRD patterns of the compounds obtained from tolu-
ene. Patterns were normalized to the intensities of the first peaks. a) Xe-
rogel of G1-G1. b) Xerogel of G1-G0. c) Solid of G0-G0 obtained by
evaporation. d) Solid of G0-G1 obtained by evaporation.
Chem. Eur. J. 2010, 16, 2427 – 2441
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2431

S. Y. Kim et al.
XRD patterns of the xerogels of G1-G1 and G1-G0 in tolu-
ene, and the solids of G0-G0 and G0-G1 obtained by evapo-
ration of toluene.
The powder XRD patterns of G1-G1, G1-G0, and G0-G1
show characteristics of self-assembled materials consisting of
lamellar structures with different layer spacings, evident in
the sharp peaks that appear at small angles and the broad
peaks at wide angles. Compared with these XRD patterns,
crystalline G0-G0 shows more refined signals accompanied
by higher-order reflections. The diffraction pattern of G1-
G0 was interpreted as an intercalated bilayer lamellar struc-
ture with the layer spacing of d=2.05 nm. However, this
spacing is quite small considering that the extended length
of the molecule is around 2.5 nm. Therefore it appears that
the alkyl chains that filled the void between the relatively
large benzamide dendrons adopted the highly disordered
configuration. The intensity of the XRD pattern was rela-
tively weak, indicating poor periodicity in the self-assembled
structure and supporting the presumed presence of disorder
among alkyl groups.
In contrast with G1-G0, G1-G1 showed much more re-
fined signals with strong intensities that reflected balance in
the molecular structure of G1-G1. Moreover, a peak at 2.98
and following higher-order reflections indicated the much
longer lamellar spacing of d=3.09 nm than that of G1-G0.
Introduction of the alkyl-G1 dendron therefore provided
better packing of the alkyl groups and increased van der
Waals interactions. The straight and rarely twisted morphol-
ogy of G1-G1 compared with the fibrous G1-G0 could be
also attributed to balance in the molecular structure.
The effect of the alkyl-G1 dendron is also evident in G0-
G1, which showed an XRD peak at 3.08 corresponding to
the layer spacing of d=2.97 nm. The observation that layer
spacing of G0-G1 is slightly smaller than G1-G1 but much
larger than G1-G0 suggests that the packing of alkyl chains
determines the periodicity of the bilayer structure. XRD
patterns of G0-G1 obtained from a sample in toluene and a
xerogel of n-hexane were indistinguishable, which indicated
that they possessed identical self-assembled structures. In a
dilute solution in toluene, however, hydrogen bonding be-
tween single benzamide groups of G0-G1 may be too weak
to compensate for the entropy loss of alkyl chains, and mac-
roscopic gelation may not occur.
Figure 4. FTIR spectra of the compounds obtained from toluene evapora-
tion. a),e) Xerogel of G1-G1. b),f) Xerogel of G1-G0. c),g) Solid of G0-
G0 obtained by evaporation. d),h) Solid of G0-G1 obtained by evapora-
tion.
3400 cmꢀ1 and symmetric ns
ACTHNUTRGENNU(G NH2) stretching at approximate-
To gain insight into the driving force for gelation, FTIR
studies were conducted. Figure 4 shows the characteristic IR
absorption of G1-G1 and G1-G0 xerogels in toluene, and
G0-G0 and G0-G1 solids obtained by evaporation of the tol-
ly 3200 cmꢀ1.[29] Analysis of G0-G0 in toluene again confirms
the presence of the S polymorphic form as suggested by the
DSC and powder XRD study.[25] In contrast with G0-G0, the
compounds with benzamide-G1 dendrons (G1-G1 and G1-
uene. The spectra in Figure 4a–d show n
bration frequencies at 3500–3100 cmꢀ1 and n
frequencies at 3000–2800 cmꢀ1, and the spectra in Figure 4e–
G
G0) showed much broader and redshifted nasACTHNGUETRN(NUG NH2) stretch-
E
ing frequencies, which indicated the presence of strong hy-
[28]
ꢀ
drogen bonds between N H and C=O. This observation,
h show n
G
also supported by broadened bands of amide I and amide II
shown at 1700–1550 cmꢀ1 in G1-G1 and G1-G0, suggests
that hydrogen bonding is the driving force for gelation. It
1750–1700 cmꢀ1; amide I for n
ACTHNUTRGNE(NUG C=O) stretching and amide II
ꢀ
ꢀ
for N H in-plane bending coupled to a C N stretching of
benzamides at 1700–1550 cmꢀ1; and d
(CH2) scissoring defor-
was noticed that frequencies of nsACTHNGUTERN(NUG NH2) stretching were
mations at approximately 1470 cmꢀ1.[29,30]
blueshifted to 3196 (G1-G0) and 3199 cmꢀ1 (G1-G1) com-
pared to those of G0-G0 (3182 cmꢀ1), presumably due to the
looser molecular packing in G1 series. These values are
Benzamides are known to exhibit two bands correspond-
ing to asymmetric nas
ACHTUNGTRENNUNG
2432
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2427 – 2441

Self-Association of Bis-Dendritic Organogelators
FULL PAPER
comparable to those found in the L polymorphic form of
G0-G0 (3197 cmꢀ1).[25a]
A vibration band corresponding to nasACHTNUGTRNEUNG(NH2) stretching of
G0-G1 also shows a redshifted shoulder, but the frequency
of the main band appears at 3394 cmꢀ1, indicating the weak-
ness of hydrogen bonds in G0-G1. The same tendency is ob-
Stokes equation predicts that a diffusion coefficient is inver-
sely proportional to the hydrodynamic radius of a species in
solution, the diffusion coefficients obtained can be used to
estimate the degree of self-association in solution. Figure 5a
shows measured diffusion coefficients of the bis-dendritic
gelators.
served for the frequency of nsACHTNUGRTENUNG(NH2) stretching, which was
blueshifted (3201 cmꢀ1) compared to the corresponding
band in G0-G0. It seems that G0-G1 does not provide opti-
mal geometry for hydrogen bonding between benzamide
groups. The bulky alkyl-G1 dendron is large enough to
induce steric hindrance between the benzamide groups and
the favorable packing of alkyl chains is thermodynamically
more important than the benzamide hydrogen bonding for
stabilizing the self-assembled structure.
Analysis of bands corresponding to the methylene vibra-
tions provides more information on the packing of alkyl
groups. The CH2 asymmetric nasACTHNUTRGNEUNG(CH2) and symmetric ns-
ACHTUNGTRENNUNG(CH2) stretching vibrations of alkyl groups generally show
strong absorptions around 2918–2928 cmꢀ1 and 2850–
2860 cmꢀ1, respectively.[30] Inclusion of gauche conformations
in the alkyl group results in a blueshift in bands at frequen-
cies 2923–2928 and 2852–2860 cmꢀ1. In the series, G1-G0
shows the highest frequency values for nas
(CH2) (2926 cmꢀ1)
ACHTUNGTRENNUNG
and ns
ACHTUNGTRENNUNG
ing that the spatially unbalanced alkyl chains are highly dis-
ordered, as suggested by powder XRD study. This trend is
also evident in dACHTUNGTRENNUNG(CH2) scissoring deformation frequencies,
for which inclusion of gauche conformations results in a red-
shift because G1-G0 exhibits the lowest frequency, at
1456 cmꢀ1. Apparently, the ordering of alkyl chains in G0-
G1 is comparable to G1-G1 and even G0-G0, based on the
band frequencies of nasACHTNUGRTEN(NNGU CH2), nsAHCTUNRTGENN(GUN CH2), and dCAHTNUGTRNEN(UGN CH2) vibra-
tions. This observation highlights the importance of alkyl
chains packing in the self-assembled structure. Again, an
FTIR spectrum of G0-G1 obtained from the n-hexane xero-
gel was identical to the spectrum shown in Figure 4d and h.
Examination of self-assembly in solution: In addition to
characterizing the self-assembly of bis-dendritic gelators in
the xerogel state, self-assembly behavior of bis-dendritic ge-
lators was examined directly in solution using concentra-
tion-dependent pulsed-field gradient spin-echo (PGSE)
1
NMR spectroscopy[31] and concentration-dependent H solu-
Figure 5. a) Measured diffusion coefficients of the bis-dendritic gelators.
b) Viscosity corrected and molecular-weight weighted diffusion coeffi-
cients of the bis-dendritic gelators. Measurements were performed in
&
tion NMR spectroscopy measurements.[21,32] Although deu-
terated toluene would have been the ideal solvent, chemical
shift overlap between residual solvent protons and the
amide and aromatic protons of the compound proved to be
problematic. Deuterated chloroform (CDCl3) was therefore
used because it provides a moderately nonpolar environ-
ment and does not interrupt hydrogen bonding. All the bis-
dendritic gelator molecules except for G2-G0 were soluble
in CDCl3 and provided clear and distinct chemical shifts in
the concentration ranges of 1–100 mm.
*
*
CDCl3 at 258C. : G1-G1; &: G1-G0; : G0-G1; : G0-G0.
The relationship between the diffusion coefficient and the
concentration is clearly different for the G1 and G0 series.
In the G1 series, diffusion coefficients varied from 5.8ꢄ10ꢀ10
to 3.1ꢄ10ꢀ10 m2 sꢀ1 for G1-G1, and from 7.7ꢄ10ꢀ10 to 4.4ꢄ
10ꢀ10 m2 sꢀ1 for G1-G0. In contrast, the change of diffusion
coefficient was much smaller in G0 series, ranging from
6.1ꢄ10ꢀ10 to 4.8ꢄ10ꢀ10 m2 sꢀ1 for G0-G1 and from 9.0ꢄ10ꢀ10
to 7.9ꢄ10ꢀ10 m2 sꢀ1 for G0-G0. The large difference between
By employing PGSE NMR spectroscopy techniques, dif-
fusion coefficients of the bis-dendritic gelators were deter-
mined at several concentrations. Because the Einstein–
Chem. Eur. J. 2010, 16, 2427 – 2441
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2433

S. Y. Kim et al.
the G1 and G0 series suggests that the association modes of
the two series are different.
G1 at various concentrations. Peak assignments for each
molecule are also given. In the spectra, signals 1–3 appear
consistently at d=7–8 ppm and were assigned to the aro-
matic protons in the benzamide phenyl group. Other aro-
matic protons in the “branching” phenyl rings were also as-
signed for each generation of dendrons. Aliphatic protons
on the carbon adjacent to the ethereal oxygen or on the car-
bonyl carbon appeared at chemical shifts of dꢁ4 ppm or d
ꢁ2.3 ppm, respectively. Below d=2 ppm, peaks were as-
signed to aliphatic protons in the alkyl chains.
To clarify the effect of a dendritic branch on benzamide
dendron self-association, the diffusion coefficients shown in
Figure 5a were corrected by solution viscosity determined
by the diffusion coefficient of residual chloroform. In addi-
tion, diffusion coefficients were normalized by the cubed
root of the bis-dendritic gelator molecular weight to remove
the effect of molecular volume (assuming that density is uni-
form across compounds and that molecular volume is pro-
portional to molecular weight). Again, the slopes shown in
Figure 5b clearly indicate that G1-G1 and G1-G0 form large
aggregates at high concentrations, whereas G0-G1 and G0-
G0 appear to form dimeric species. These results imply that
the dendritic branch in the benzamide dendron is key to the
indefinite self-association and the formation of self-assem-
bled fibrillar networks.
The amide protons in the benzamide group exhibited two
signals (NH1 and NH2) indicating restricted rotation of the
ꢀ
NH2 group around its N CO bond due to hydrogen bond-
ing.[33] By considering that benzamides form two-dimension-
al hydrogen-bonding networks consisting of face-to-face hy-
drogen bonds by means of Hsyn atoms and side-to-side hy-
drogen bonding by means of Hanti atoms, the peak shifted
further downfield (NH1) was assigned to Hsyn and the up-
field peak (NH2) was assigned to Hanti. As the concentration
increased, NH1 and NH2 shifted further downfield, and the
distance between NH1 and NH2 increased. The downfield
shift indicates proton deshielding by interaction with hydro-
Association constants of the bis-dendritic gelators were
evaluated by concentration-dependent solution 1H NMR
spectroscopy experiments in deuterated chloroform.
Figure 6 shows representative 1H NMR spectra of the bis-
dendritic gelator molecules G1-G1, G1-G0, G0-G0, and G0-
Figure 6. 1H NMR spectra of the bis-dendritic gelator molecules at various concentrations (300 MHz, CDCl3, 258C). a) G1-G1. b) G1-G0. c) G0-G0.
d) G0-G1.
2434
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2427 – 2441

Self-Association of Bis-Dendritic Organogelators
FULL PAPER
gen acceptors (carbonyl oxygen in this case), consistent with
an increase in intermolecular hydrogen bonding among
benzACHTUNGTRENNUNGamide groups. Notably, the extent of the downfield shift
model (G1 series) or the DIM model (G0 series). Figure 7c
and d show fits of the G1 series using the coEK model.
Table 3 shows a summary of the association constants and
fit parameters of the bis-dendritic gelators. Both models fit
well to all data sets for the benzamide-G1 series, but com-
parison of two models based on Akaikeꢁs information crite-
was much larger in molecules containing a benzamide-G1
dendron than in molecules containing a single benzamide,
which indicated that the presence of the dendron enhanced
the hydrogen bonding. NH1 and NH2 signals in molecules
containing a single benzamide merged at a concentration of
approximately 10 mm but split when the concentration was
further decreased. Due to merging, concentration-dependent
chemical shifts of NH2 for G0-G0 and G0-G1 yielded scat-
tered data points.
Table 3. Association constants of the bis-dendritic gelators Gn-Gm.[a]
G1-G1 NH1 G1-G1 NH2 G1-G0 NH1 G1-G0 NH2
EK model
KE [mꢀ1
Pa [ppm]
]
24
5.73
28
5.80
22
5.78
16
5.81
The measured dependences of NH1 and NH2 NMR spec-
troscopy shifts on bis-dendritic gelator concentration were
fit using association models described by Martin.[20] Based
on the results of diffusion coefficient measurements,
Equal K (EK) and cooperative Equal K (coEK) models
were applied to the G1-G1 and G1-G0 data as indefinite
self-association models, and a dimeric (DIM) model was ap-
plied to the G0-G1 and G0-G0 data. Whereas the DIM
model describes the fraction of dimeric aggregates and non-
aggregated molecules in the solution with association con-
stant K2, the coEK model accounts for nearest-neighbor in-
teractions and represents indefinite association with KE =
K2/1=K3 =K4 =…=Kn (1=K2/KE). If 1=1, the coEK
model converges to the EK model, which does not differen-
tiate between the nucleation and growth processes (isodes-
mic growth).[20,34] Figure 7a and b show concentration-de-
Px [ppm]
7.87
6.42
7.82
6.54
fitting error[b]
3ꢄ10ꢀ3
3ꢄ10ꢀ3
1.2ꢄ10ꢀ2
3ꢄ10ꢀ3
coEK model
KE [mꢀ1
]
49
0.240
17
5.84
38
1.00
55
5.78
6.37
3ꢄ10ꢀ3
39
0.431
17
39
0.860
34
5.79
1
K2 [mꢀ1
]
Pa [ppm]
5.83
Px [ppm]
7.31
7.41
6.36
fitting error[b]
7ꢄ10ꢀ3
1ꢄ10ꢀ2
4ꢄ10ꢀ3
G0-G0 NH1 G0-G0 NH2 G0-G1 NH1 G0-G1 NH2
DIM model
K2 [mꢀ1
]
2
5.64
3
5.72
5
5.71
6
5.77
Pa [ppm]
Pl [ppm]
8.34
6.15
7.22
6.18
fitting error[b]
3ꢄ10ꢀ3
3ꢄ10ꢀ3
1ꢄ10ꢀ3
1ꢄ10ꢀ3
[a] For definitions of the symbols, see the Experimental Section. [b] The
m
P
2
fitting error was calculated as jðresidualÞij , in which the residual is de-
fined by the difference between the experimental and the model fit.
1
pendent H NMR spectroscopy shifts of NH1 and NH2 for
G1-G1, G1-G0, G0-G0, and G0-G1, fitted to either the EK
n¼1
rion method and the extra sum-
of-squares F test indicated that
the simpler EK model (with 3
parameters) is more likely to be
correct than the more compli-
cated coEK model (with 4 pa-
rameters).[35] These results sug-
gest that the self-assembly of
the benzamide-G1 dendrons
follows the isodesmic growth
mechanism and the strength of
secondary interactions between
monomers in the polymer
chains is unaffected by the
length of the chain.[34] The benz-
amide association constants es-
timated for the benzACHTNUTGRNEUNGamide-G1
dendrons were in the range of
16–28mꢀ1 determined by the
EK model. These values are
comparable to association con-
stants reported for toluene ge-
lators in chloroform,[21c] and, as
suggested previously, the associ-
ation constants are expected to
be much higher in toluene be-
Figure 7. Concentration-dependent 1H NMR spectroscopy shifts of NH1 and NH2 of the bis-dendritic gelator
molecules (300 MHz, CDCl3). a) NH1. b) NH2. Measured data points are shown as squares (G1 series) and
circles (G0 series), and fits are shown as lines (G1 series, EK model) and dots (G0 series, DIM model). c) Fits
of G1-G1 based on the coEK model. d) Fits of G1-G0 based on the coEK model.
Chem. Eur. J. 2010, 16, 2427 – 2441
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2435

S. Y. Kim et al.
cause chloroform provides
a
considerably more polar envi-
ronment than toluene. Observa-
tion of larger association con-
stants for G1-G1 than for G1-
G0 implies that stabilization of
hydrogen bonding in nonpolar
environments by the sterically
bulky alkyl-G1 dendron is also
important for self-association.
Association constants of the
G0 series corresponding to di-
merization were much smaller
than those of the G1 series, and
ranged from 2–6mꢀ1. Consider-
ing that molecules consisting of
the benzamide-G1 dendron
possess twice as many benza-
mide groups at a given concen-
tration, the association con-
stants of the G0 series are still
smaller than half of those of the
G1 series. Therefore, introduc-
tion of the benzamide-G1 den-
dron significantly enhances the
strength and cooperativity of
hydrogen bonding. For the
Figure 8. Concentration-dependent 1H NMR spectroscopy shifts of protons in G1-G1 (300 MHz, CDCl3).
a) H1. b) H4. c) H5. d) H8.
system discussed here, this effect was explained by a reduc-
tion in entropy loss during successive association processes
of the benzamide-G1 dendron. When one of the benzamide
moieties in the dendron participates in aggregation by
means of hydrogen bonding, the next association by means
of hydrogen bonding will be more facile due to the restric-
tion on translational degrees of freedom. On the other
hand, dissociation of a single benzamide from the aggregate
will not be entropically favorable because it does not result
in complete dissociation from the aggregate.[36] Again, asso-
ciation constants for G0-G1 were larger than those of G0-
G0.
The aromatic and benzylic protons also showed a depend-
ence of chemical shift on concentration. Peaks 1–3, corre-
sponding to aromatic protons in the benzamide phenyl
group, shifted upfield as the concentration increased. This
shift was attributed to increased shielding of protons from
p–p stacking between proximal aromatic rings.[20,37] Peak 4
for the benzamide-G1 dendrons (G1-G1 and G1-G0) also
shifted upfield, because the H4 proton is located between
two aromatic rings in the benzamide-G1 dendron and expe-
riences the increased shielding upon aggregation.
These results suggest that the branching phenyl rings
behave differently from the benzamide phenyl rings for the
molecules containing a benzamide-G1 dendron. A possible
explanation for the different behavior between types of aro-
matics is that a branching phenyl ring is nearly perpendicu-
lar to the benzamide phenyl ring and tends to form separate
weak stacking systems with the aid of van der Waals interac-
tions between the alkyl chains. Thus, the mechanism of gela-
tion in G1-G1 and G1-G0 can be described as association of
the benzamide-G1 dendron by means of hydrogen bonding
and p–p stacking, and continued growth of aggregates by
the participation of branching phenyl rings and alkyl chains
through p–p stacking and van der Waals interactions.
This mechanism also explains why G1-G1 forms the most
stable gels in a variety of solvents, because G1-G1 achieves
the strongest intermolecular interactions by spatially balanc-
ing hydrogen bonding, p–p stacking, and van der Waals in-
teractions. Compound G1-G0 has weaker p–p stacking and
van der Waals interactions and its gelation relies to a great
extent on hydrogen bonding, which restricts the range of
solvents conducive to gelation. On the contrary, G0-G1
relies on mainly van der Waals interactions for gelation, so
stable gels form only in n-hexane, in which the interaction
between alkyl chains is maximized. Compound G0-G0 may
have a balanced molecular structure, but its intermolecular
interactions are weak in solution and crystallization is fa-
vored over gelation in the solid state. Figure 9 summarizes
the conclusion that self-assembly of the bis-dendritic gela-
tors can be reliably tuned by adjusting the generation
number of each dendron.
In contrast, the H5 and H6 protons in G1-G1 and G1-G0
show only a slight shift downfield at low concentrations,
reach maximal chemical shifts at 20–25 mm, and then shift
upfield at higher concentrations. This trend is also observed
for the H7 and H8 protons in G1-G1 but not for H5 and H6
protons in G0-G1. Figure 8 shows the concentration-depen-
1
dent H NMR spectroscopy shifts of H1, H4, H5, and H8
protons in G1-G1.
2436
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2427 – 2441

Self-Association of Bis-Dendritic Organogelators
FULL PAPER
as a light source. FTIR spectra were
recorded on a Bruker EQUINOX-55
spectrometer using KBr pellets.
Compound 6 (G1-G0): Compound 5
(0.4 g, 0.778 mmol), 1-dodecanoic acid
(0.467 g, 2.333 mmol), and DMAP
(0.285 g, 2.333 mol) were dissolved in
THF (40 mL) and placed in a 100 mL
one-necked round-bottomed flask.
The reaction mixture was stirred at
08C for 30 min. EDC (0.285 g,
2.333 mmol) was added into the flask
and stirred for an additional 30 min.
The reaction mixture was warmed up
to room temperature and was stirred
for 3 h. Then, the reaction mixture was
filtered, washed with 0.1m HCl (aq),
0.1m NaHCO3, and brine. The organic
layer was dried and evaporated in a
rotary evaporator. The excess dodeca-
noic acid was removed by precipita-
tion in n-hexane (0.491 g, 88.3%).
M.p. 1258C; 1H NMR (400 MHz,
Figure 9. Relationship between the structure and properties of the bis-dendritic gelator molecules.
CDCl3, 258C): d=7.56 (d, 1J
(H,H)=2.3 Hz, 2H; ArH), 7.41 (dd, 1J
2H; ArH), 6.84 (d, 1J
(H,H)=2.0 Hz, 2H; ArH), 6.74 (s, 2H; NH of
ACHTUNGTRENNUNG
Conclusion
U
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
We prepared a series of bis-dendritic gelator molecules con-
sisting of benzamide and alkyl dendrons to study the rela-
tionship between macroscopic gelation (self-assembled fi-
brillar networks) and the molecular-level intermolecular in-
teractions. The generation-dependent gelation behavior re-
ported here demonstrates that multivalent cooperativity
among peripherally interacting groups in dendritic architec-
tures strongly affects the nature of intermolecular associa-
tion. Investigation of the influence of alkyl chains revealed
that spatial balance in the molecular structure is also very
important for balancing the intermolecular interactions and
for allowing gelation in a range of solvents.
amide), 6.44 (t, 1J
ACTHNUGRTENUNG(H,H)=2.3 Hz, 1H; OH), 6.00 (s, 2H; another NH of
amide), 5.06 (s, 2H; CH2OH), 2.34 (t, 1J
N
1
1.61 (t, J
ACHTUNGTRENNUNG
(H,H)=7.0 Hz, 2H; C=OCH2CH2), 1.24 (m, 16H), 0.85 ppm (t,
1J
173.4,
2
1
N
U
ACHTUNGTRENNUNG
AHCTUNGERTG(NNUN C,F)=5.1 Hz), 64.8, 34.1, 31.8, 29.5–29.1, 24.8, 22.6, 14.0 ppm; ESIMS:
m/z: calcd: 697.2707; found: 697.2707 [M+]; calcd: 719.2526; found:
719.5310 [M+Na+] ; elemental analysis calcd (%) for C35H38F6N2O6: C
60.34, H 5.50, F 16.36, N 4.02, O 13.78; found: C 61.88, H 5.49, N 3.86.
Compound 7: Compound 4 (1.216 g, 2.091 mmol), carbon tetrabromide
(2.022 g, 6.097 mmol), and triphenylphosphine (1.600 g, 6.100 mmol) were
placed in a 100 mL three-necked round-bottomed flask. Under nitrogen
purging, THF (50 mL) was added into the flask. The reaction mixture
was stirred for 30 min at room temperature. Then the mixture was
poured into the excess water, and then extracted with ethyl acetate
(EA). The organic layer was dried and evaporated in a rotary evaporator.
The desired product was purified by flash column chromatography (EA/
hexane (Hx) 1:3 to 1:2) to yield a yellow solid (0.4 g, 35.3%). M.p.
Experimental Section
General: All starting materials were obtained from commercial suppliers
and were used without further purification. Syntheses of 1, 2, 3, 4, and 5
were published elsewhere.[25a,26] Compound 14[28] was synthesized follow-
ing reported procedures. Synthesized compounds were structurally veri-
fied by NMR spectroscopy. MgSO4 was used to dry all organic solutions
during workup procedures. Analytical TLC was performed on Kieselgel
60 F-254 precoated TLC plates. Silica 60 (230–400 mesh) from Merck
was used for flash chromatography. The 1H and 13C NMR spectra were
recorded on a Bruker Fourier Transform AVANCE 400 spectrometer.
Chemical shifts were expressed in part per million using residual solvent
protons. High-resolution mass spectra (HRMS) of the compounds were
obtained on a VG Autospec Ultima mass spectrometer using the electro-
ionization method and on a Bruker MicroTOF-Q tandem mass spectrom-
eter using the electrospray ionization method. DSC studies were per-
formed on a TA Q100 calorimeter. Samples were measured in aluminum
1
1
<408C; H NMR (400 MHz, CDCl3, 258C): d=7.80 (d, J
2H; ArH), 7.37 (d, 1J(H,H)=2.5 Hz, 2H; ArH), 7.21 (dd, 1J
8.6 Hz, 2J(H,H)=2.5 Hz, 2H; ArH), 7.00 (d, 1J
(H,H)=2.2 Hz, 2H;
ArH), 6.73 (t, 1J
(H,H)=2.2 Hz, 1H; ArH), 4.42 ppm (s, 2H; CH2Br);
13C NMR (100 MHz, CDCl3, 258C): d=160.2, 155.9, 142.7, 136.9, 135.3
(q, 2J(C,F)=33.0 Hz), 121.8 (q, 1J
(C,F)=272.7 Hz), 120.5, 116.5 (q,
3J
ACHTUNGTRENNUNG
G
ACHTUNGTRENNUNG
U
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
AHCTUNGTREG(NNUN C,F)=4.8 Hz), 115.0, 111.9, 104.4, 31.0 ppm; ESIMS: m/z: calcd:
540.9981; found: 540.9988 [M+H+]; elemental analysis calcd (%) for
C23H11BrF6N2O2: C 51.04, H 2.05, Br 14.76, F 21.06, N 5.18, O 5.91;
found: C 47.03, H 1.83, N 4.48.
Compound 8: Compound 7 (0.506 g, 0.935 mmol), 3,5-dihydroxybenzyl
alcohol (0.066 g, 0.468 mmol), potassium carbonate (0.161 g, 1.169 mmol),
and [18]crown-6 (0.024 g, 0.092 mmol) were placed in a 25 mL three-
necked round-bottomed flask and stirred. Acetone (5 mL) was added
into the flask and the reaction mixture was refluxed under nitrogen purg-
ing for 24 h. Then the reaction mixture was cooled to room temperature,
filtered, and concentrated in a rotary evaporator. The mixture was dis-
solved again in EA, washed with water and brine, and dried. The organic
layer was evaporated in a rotary evaporator. The desired product was pu-
rified by flash column chromatography (EA/Hx 1:1) to yield a yellow vis-
cous liquid (0.365 g, 36.8%). M.p. 658C; 1H NMR (400 MHz, CDCl3,
hermetic cells under a N2 atmosphere at a heating rate of 108Cminꢀ1
.
POM studies were performed on a Nikon Eclipse ME600 microscope
equipped with a Mettler-Toledo FP82 hot stage and an FP90 central pro-
cessor. SEM studies were performed on a Philips XL30S FE-SEM instru-
ment for gold-sputtered samples. TEM studies were performed on an
EM 912 OMEGA TEM instrument for samples on a carbon-coated
copper TEM grid without staining. Powder XRD patterns were obtained
1
1
with
a
Rigaku D/MAX-RB diffractometer using
a
scan speed of
258C): d=7.76 (d, J
G
ACHTUNGTREN(NUNG H,H)=2.4 Hz,
0.058minꢀ1 and with sampling width of 0.018, using CuKa (l=0.154 nm)
4H; ArH), 7.19 (dd, 1J
E
ACHTUNGTRENNUNG
Chem. Eur. J. 2010, 16, 2427 – 2441
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2437

S. Y. Kim et al.
(d, 1J
G
E
102.9, 52.7 ppm; EIMS: m/z: calcd: 379.0668; found: 379.0667 [M+]; ele-
mental analysis calcd (%) for C18H12F3NO5: C 57.00, H 3.19, F 15.03, N
3.69, O 21.09; found: C 56.95, H 3.22, N 3.65.
6.58 (d, 1J(H,H)=2.1 Hz, 2H; ArH), 6.42 (t, 1J
ArH), 5.05 (s, 4H; CH2O), 4.61 ppm (s, 2H; CH2OH); 13C NMR
(100 MHz, CDCl3, 258C): d=160.2, 155.9, 142.7, 136.9, 135.3 (q,
Compound 17: Compound 16 (0.5 g, 1.318 mmol) was dissolved in
DMSO (3 mL) and placed in a 10 mL one-necked round-bottomed flask.
At 08C, potassium carbonate (0.2 g, 1.447 mmol) and 30% H2O2 (aq)
(0.9 mL) were added into the flask. The reaction mixture was warmed up
to room temperature and additional DMSO (5 mL) was added. After
21 h, the reaction mixture was poured into water (100 mL). The white
solid was filtered and dried under vacuum (0.182 g, 34.7%). M.p. 1868C;
1H NMR (400 MHz, [D6]DMSO, 258C): d=8.35 (s, 1H; ArH), 7.98 (s,
1H; NH of amide), 7.79 (s, 1H; another NH of amide), 7.60 (d,
2J
ACHTUNGTRENNUNG(C,F)=33.0 Hz), 121.8 (q, JACHTUNRTGEGUNN(N C,F)=272.7 Hz), 120.5, 116.5 (q, JACHTNGUTREN(NUGN C,F)=
3
4.8 Hz), 115.0, 111.9, 104.4, 31.0 ppm; ESIMS: m/z: calcd: 1061.1839;
found: 1061.1907 [M+H+]; elemental analysis calcd for C53H28F12N4O7: C
60.01, H 2.66, F 21.49, N 5.28, O 10.56; found: C 59.77, H 2.92, N 4.91.
Compound 9: Compound 9 was synthesized from 8 (0.147 g, 0.139 mmol)
using the same protocol as that for 5 (0.125 g, 89.4%). M.p. 1208C;
1H NMR (400 MHz, [D6]DMSO, 258C): d=7.91 (s, 2H; NH of amide),
1
7.56 (s, 2H; another NH of amide), 7.55 (s, 2H; ArH), 7.37 (d, J
A
1J
8.4 Hz, 2J
(100 MHz, [D6]DMSO, 258C): d=168.3, 164.6, 156.5, 156.2, 132.8, 132.3,
130.9, 127.8 (q, 2J(C,F)=32.1 Hz), 126.4, 124.4, 123.1 (q, 1J
(C,F)=
272.3 Hz), 122.3, 117.2 (q, J(C,F)=4.5 Hz), 52.7 ppm; EIMS: m/z: calcd:
A
ACHTUNGTRENNUNG(H,H)=
2.0 Hz, 2H; ArH), 7.32 (dd, 1J
ArH), 6.86 (d, 1J
(H,H)=1.5 Hz, 2H; ArH), 6.73 (s, 1H; ArH), 5.36 (t,
1J(H,H)=5.7 Hz, 1H; OH), 4.48 ppm (d, 1J
(H,H)=5.5 Hz, 2H;
CH2OH); 13C NMR (100 MHz, [D6]DMSO, 258C): d=168.5, 156.9, 147.6,
132.0, 130.8, 127.8 (q, 2J(C,F)=32 Hz), 127.3, 123.2 (q, 1J
(C,F)=
AHCTUNGTRENNUNG
G
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
3
AHCTUNGTRENNUNG
397.077; found: 3397.0774 [M+]; elemental analysis calcd (%) for
C18H14F3NO6: C 54.42, H 3.55, F 14.35, N 3.53, O 24.16; found: C 54.41,
H 3.54, N 3.51.
272.2 Hz), 121.7, 116.2, 112.6, 108.8, 62.1 ppm; ESIMS: m/z: calcd:
1133.2262; found: 1133.2408 [M+H+]; elemental analysis calcd (%) for
C53H35F12N4O11: C 56.19, H 3.20, F 20.12, N 4.95, O 15.54; found: C 54.52,
H 3.12, N 4.90.
The filtrate was acidified with 10% HCl and filtered to yield another
white solid, 3-(4-carbamoyl-3-trifluoromethylphenoxy)-5-methoxycarbo-
nylbenzoic acid (0.123 g, 44.8%). M.p. 2558C; 1H NMR (400 MHz,
[D6]DMSO, 258C): d=13.6 (br, 1H; COOH), 8.25 (s, 1H; ArH), 7.99 (s,
1H; NH of amide), 7.74 (s, 1H; ArH), 7.72 (s, 1H; another NH of
Compound 10 (G2-G0): Compound 10 was synthesized from 9 (5 g,
14.07 mmol) using the same protocol as that for 6 (1.11 g, 21.1%). M.p.
888C; 1H NMR (400 MHz, [D6]DMSO, 258C): d=7.89 (s, 4H; NH of
amide), 7.57 (s, 4H; ArH), 7.55 (s, 4H; another NH of amide), 7.39 (d,
amide), 7.60 (d, 1J
1H; ArH), 7.39 (dd, 1J
3.86 ppm (s, 3H; CH3); 13C NMR (100 MHz, [D6]DMSO, 258C): d=
168.3, 165.7, 164.8, 156.5, 156.2, 133.6, 132.8, 132.2, 128.0 (q, 2J
(C,F)=
(C,F)=272.5 Hz), 123.0, 122.4, 119.1,
A
ACHTUNGTREN(NUNG H,H)=2.3 Hz,
1J
2.2 Hz, 4H; ArH), 7.00 (d, 1J
1J(H,H)=1.9 Hz, 2H; ArH), 6.59 (d, 1J
1H; ArH), 5.09 (s, 4H; G1-CH2O), 4.98 (s, 2H; G2-CH2O), 2.31 (t,
1J(H,H)=7.3 Hz, 2H; C=OCH2), 1.50 (t, 1J
(H,H)=6.5 Hz, 2H; C=
OCH2CH2), 1.18 (m, 16H), 0.81 ppm (t, 1J
(H,H)=6.5 Hz, 3H; CH3);
13C NMR (100 MHz, [D6]DMSO, 258C): d=172.7, 168.4, 159.1, 156.9,
156.6, 141.6, 138.8, 132.1, 130.7, 127.9 (q, 2J
(C,F)=31.9 Hz), 123.1 (q,
1J(C,F)=272.2 Hz), 121.6, 116.2 (q, 3J
(C,F)=5.1 Hz), 113.8, 109.5, 106.8,
A
T
ACHTUNGTRNE(NUNG H,H)=
A
ACHTUNGTRENNUNG
T
ACHTUNGTRENNUNG
32.0 Hz), 125.3, 123.4, 123.1 (q, 1J
ACHTUNGTRENNUNG
N
ACHTUNGTRENNUNG
117.3, 52.7 ppm; EIMS: m/z: calcd: 383.0617; found: 383.0616 [M+]; ele-
mental analysis calcd (%) for C17H12F3NO6: C 53.27, H 3.16, F 14.87, N
3.65, O 25.05; found: C 53.15, H 3.13, N 3.64.
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
T
Compound 18: Compound 17 (0.1 g, 0.252 mmol) was dissolved in a sol-
vent mixture of THF (6 mL), methanol (3 mL), and water (1 mL). Lithi-
um hydroxide (0.030 g, 1.258 mmol) was added at 08C and stirred for 4 h.
The mixture was warmed up to room temperature, and mixed with
excess of water. The white suspension was acidified with 10% HCl (aq)
and kept in a refrigerator overnight. The white solid was filtered and
dried under vacuum (0.086 g, 92.5%). M.p. 1208C; 1H NMR (400 MHz,
[D6]DMSO, 258C): d=13.2 (br, 2H; COOH), 8.35 (s, 1H; ArH), 7.99 (s,
1H; NH of amide), 7.73 (s, 1H; another NH of amide), 7.66 (d,
101.3, 68.3, 64.9, 33.4, 31.3, 28.9–28.4, 24.4, 22.1, 13.9 ppm; ESIMS: m/z:
calcd: 1337.3752; found: 1337.3682 [M+Na+]; elemental analysis calcd
(%) for C65H58F12N4O12: C 59.36, H 4.45, F 17.33, N 4.26, O 14.60; found:
C 58.44, H 5.26, N 4.12.
Compound 15 (G1-G1): Compound 15 was synthesized from 6 (0.2 g,
14.07 mmol) and 14 using the same protocol as that for 6 (0.219 g,
56.9%). M.p. 1388C; 1H NMR (400 MHz, CDCl3, 258C): d=7.54 (d,
1J
(dt, 1J
2.3 Hz, 2H; ArH), 6.63 (t, 1J
of amide), 6.31 (t, 1J
(H,H)=2.2 Hz, 1H; ArH), 6.07 (s, 2H; another NH
of amide), 5.30 (s, 2H; CH2OC=O), 3.94 (t, 1J
(H,H)=6.5 Hz, 4H; O=
(H,H)=7.8 Hz, 4H; O=COCH2CH2), 1.42 (m,
(H,H)=8.0 Hz, 4H; C=OCH2CH2CH2), 1.24 (m, 32H), 0.86 ppm (t,
(H,H)=6.6 Hz, 6H; CH3); 13C NMR (100 MHz, CDCl3, 258C): d=
T
1J
(d, 1J
2J(H,H)=2.3 Hz, 1H; ArH); 13C NMR (100 MHz, [D6]DMSO, 258C):
d=168.5, 165.9, 156.5, 156.3, 133.5, 132.8, 131.0, 128.1 (q, 2J
(C,F)=
32.0 Hz), 125.5, 123.8, 123.2 (q, 1J
(C,F)=272.4 Hz), 122.5, 117.4 ppm (q,
3J(C,F)=4.8 Hz); EIMS: m/z: calcd: 369.0460; found: 369.0464 [M+]; el-
A
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
COCH2), 1.75 (m, 1J
AHCTUNGTRENNUNG
1J
1J
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
emental analysis calcd (%) for C16H10F3NO6: C 52.04, H 2.73, F 15.44, N
3.79, O 26.00; found: C 49.59, H 2.95, N 3.61.
ACHTUNGTRENNUNG
169.5, 166.1, 160.2, 157.8, 157.4, 140.6, 131.3, 130.7, 130.5, 129.4 (q,
2J(C,F)=32.7 Hz), 122.9 (q, 1J
(C,F)=272.2 Hz), 122.1, 117.5, 113.9, 1082,
G
Using the same protocol, 3-(4-carbamoyl-3-trifluoromethylphenoxy)-5-
107.9, 106.6, 68.4, 65.5, 31.9, 29.6–29.2, 26.0, 22.7, 14.1 ppm; ESIMS: m/z:
calcd: 987.4953; found: 987.4980 [M+H+]; elemental analysis calcd (%)
for C54H68F6N2O8: C 65.70, H 6.94, F 11.55, N 2.84, O 12.97; found: C
62.83, H 5.92, N 3.65.
methoxycarbonyl benzoic acid was hydrolyzed to also give compound 18.
Compound 19 (G0-G1): Compound 18 (0.06 g, 0.162 mmol), 1-dodecanol
(0.09 mL, 0.406 mmol), and DMAP (0.040 g, 0.325 mol) were dissolved in
N,N-dimethyl formamide (DMF, 12 mL) and placed in a 25 mL one-
necked round-bottomed flask. The reaction mixture was stirred at 08C
for 30 min. EDC (0.062 g, 0.325 mmol) was added into the flask and
stirred for an additional 30 min. The reaction mixture was warmed up to
room temperature and stirred for 18 h. Then the reaction mixture was fil-
tered, washed with 0.1m HCl (aq), 0.1m NaHCO3, and brine. The organic
layer was dried and evaporated in a rotary evaporator. The crude product
was purified by column chromatography using n-hexane as the eluent
(0.078 g, 68.2%). M.p. 928C; 1H NMR (400 MHz, CDCl3, 258C): d=8.48
Compound 16: Compound 1 (1.028 g, 4.758 mmol) and dimethyl 3-hy-
droxyisophthlate (1 g, 4.758 mmol) were dissolved in DMSO (12 mL) and
placed in a 25 mL one-necked round-bottomed flask. Potassium carbon-
ate (0.658 g, 4.758 mmol) was added into the flask and the solution was
stirred at room temperature for 3 h. Then the solution was poured into
water (150 mL), and filtered to obtained a white solid that was dried
under vacuum (1.481 g, 79.9%). M.p. 1338C; 1H NMR (400 MHz,
[D6]DMSO, 258C): d=8.33 (s, 1H; ArH), 8.14 (d, 1J
ArH), 7.93 (s, 2H; ArH), 7.66 (d, 1J
(H,H)=2.0 Hz, 1H; ArH), 7.53 (s,
1H; ArH), 7.40 (dd, 1J(H,H)=8.6 Hz, 2J
(H,H)=2.3 Hz, 1H; ArH),
3.87 ppm (s, 6H; CH3); 13C NMR (100 MHz, [D6]DMSO, 258C): d=
ACHTUNGTREN(NGNU H,H)=8.6 Hz, 1H;
T
(s, 1H; ArH), 7.84 (s, 2H; ArH), 7.59 (d, 1J
7.32 (d, 1J(H,H)=2.3 Hz, 1H; ArH), 7.14 (dd, 1J
2J(H,H)=2.2 Hz, 1H; ArH), 5.90 (d, 1J
(H,H)=79.5 Hz, 2H; NH of
amide), 4.32 (t, 1J(H,H)=6.7 Hz, 4H; O=COCH2), 1.75 (m, 1J
(H,H)=
7.0 Hz, 4H; O=COCH2CH2), 1.24 (m, 32H), 0.85 ppm (t, 1J
(H,H)=
ACHTUNGTRENNUNG
A
U
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
164.5, 160.4, 154.6, 137.9, 133.2 (q, 2J
N
A
ACHTUNGTRENNUNG
122.0 (q, 1J
G
G
ACHTUNGTRENNUNG
2438
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2427 – 2441

Self-Association of Bis-Dendritic Organogelators
FULL PAPER
6.4 Hz, 6H; CH3); 13C NMR (100 MHz, CDCl3, 258C): d=168.8, 165.0,
ogy of Materials funded by the Ministry of Knowledge Economy, Repub-
lic of Korea.
158.0, 155.8, 133.2, 131.0, 130.0, 129.8 (q, 2J
ACHTUNGTRENNUNG
122.9 (q, 1J
N
ACHTUNGTRENNUNG
31.9, 29.6–29.2, 28.6, 25.9, 22.8, 22.7, 14.1 ppm; ESIMS: m/z: calcd:
706.4289; found: 706.4319 [M+H+]; elemental analysis calcd (%) for
C40H58F3NO6: C 68.06, H 8.28, F 8.07, N 1.98, O 13.60; found: C 68.62, H
8.49, N 1.93.
[1] a) G. R. Newkome, C. N. Moorefield, F. Vçgtle, Dendrimers and
Dendrons, Wiley-VCH, Weinheim, 2001; b) D. A. Tomalia, A. M.
Naylor, W. A. Goddard III, Angew. Chem. 1990, 102, 119–157;
Angew. Chem. Int. Ed. Engl. 1990, 29, 138–175; c) O. A. Matthews,
A. N. Shipway, J. F. Stoddart, Prog. Polym. Sci. 1998, 23, 1–56;
d) M. Fischer, F. Vçgtle, Angew. Chem. 1999, 111, 934–955; Angew.
Chem. Int. Ed. 1999, 38, 884–905; e) A. W. Bosman, H. M. Janssen,
E. W. Meijer, Chem. Rev. 1999, 99, 1665–1688; f) K. Inoue, Prog.
Polym. Sci. 2000, 25, 453–571; g) R. Esfand, D. A. Tomalia, Drug
Discovery Today 2001, 6, 427–436; h) S. M. Grayson, J. M. J. Frꢃ-
chet, Chem. Rev. 2001, 101, 3819–3867; i) S. Hecht, J. M. J. Frꢃchet,
Angew. Chem. 2001, 113, 76–94; Angew. Chem. Int. Ed. 2001, 40,
74–91; j) L. H. Bryant, Jr., J. W. M. Bulte, Focus on Biotechnology
2002, 7, 47–69; k) C. C. Lee, J. A. MacKay, J. M. J. Frꢃchet, F. C.
Szoka, Nat. Biotechnol. 2005, 23, 1517–1526; l) W.-D. Jang, K. M. K.
Selim, C.-H. Lee, I.-K. Kang, Prog. Polym. Sci. 2009, 34, 1–23.
[2] a) F. Zeng, S. C. Zimmerman, Chem. Rev. 1997, 97, 1681–1712; b) T.
Emrick, J. M. J. Frꢃchet, Curr. Opin. Colloid Interface Sci. 1999, 4,
15–23; c) F. Vçgtle, S. Gestermann, R. Hesse, H. Schwierz, B. Wind-
isch, Prog. Polym. Sci. 2000, 25, 987–1041; d) G. M. Dykes, J. Chem.
Technol. Biotechnol. 2001, 76, 903–918.
PGSE NMR spectroscopy: The diffusion coefficient measurements were
carried out using
a Bruker AVANCE 400 MHz NMR spectrometer
(equipped with a 5 mm Bruker QNP probe with an actively shielded z
gradient coil) at 258C. Diffusion coefficients were extracted from a series
of 1H NMR spectra measured by a stimulated echo sequence using bipo-
lar gradient (STEbp) pulse sequences as a function of gradient amplitude.
The diffusion time (D) was 100 ms, the length of the gradient (d) was
2.6 ms and the delay time (d1) was 30 s. At a given concentration, 16
data points with different gradient strengths ranging from 1 to 51 Gcmꢀ1
were obtained. Using the attenuated intensities of the peak shown at d=
1.27 ppm (assigned to methylene groups in the alkyl chain), diffusion co-
efficients were calculated by using SimFit[38] in Bruker XWINNMR 3.0.
Obtained diffusion coefficients were calibrated by measuring the diffu-
sion coefficient of residual chloroform in each sample solution (Dsolution
)
and comparing it to the chloroform diffusion coefficient in pure CDCl3
(Dpure). This allowed an estimation of the solution viscosity of the solu-
tion to be made. Equation (1) was used for this calibration:
Dcorrelated ¼ Dobtained ꢂ ðDpure=Dsolution
Þ
ð1Þ
[3] a) V. Percec, P. Chu, G. Ungar, J. Zhou, J. Am. Chem. Soc. 1995,
117, 11441–11454; V. Percec, G. Johansson, G. Ungar, J. Zhou, J.
Am. Chem. Soc. 1996, 118, 9855–9866; V. Percec, C-.H. Ahn, B.
Barboiu, J. Am. Chem. Soc. 1997, 119, 12978–12979; S. D. Hudson,
H.-T. Jung, V. Percec, W.-D. Cho, G. Johansson, G. Ungar, V. S. K.
Balagurusamy, Science 1997, 278, 449–452; V. Percec, C.-H. Ahn, G.
Ungar, D. J. P. Yeardley, M. Mçller, S. S. Sheiko, Nature 1998, 391,
161–164; V. Percec, W.-D. Cho, G. Ungar, D. J. P. Yeardley, Chem.
Eur. J. 2002, 8, 2011–2025; J. A. A. W. Elemans, M. J. Boerakker,
S. J. Holder, A. E. Rowan, W.-D. Cho, V. Percec, R. J. M. Nolte,
Proc. Natl. Acad. Sci. USA 2002, 99, 5093–5098; V. Percec, M.
Glodde, G. Johansson, V. S. K. Balagurusamy, P. A. Heiney, Angew.
Chem. 2003, 115, 4474–4478; Angew. Chem. Int. Ed. 2003, 42, 4338–
4342; b) V. Percec, A. E. Dulcey, V. S. K. Balagurusamy, Y. Miura, J.
Smidrkal, M. Peterca, S. Nummelin, U. Edlund, S. D. Hudson, P. A.
Heiney, H. Duan, S. N. Magonov, S. A. Vinogradov, Nature 2004,
430, 764–768; V. Percec, A. E. Dulcey, M. Peterca, P. Adelman, R.
Samant, V. S. K. Balagurusamy, P. A. Heiney, J. Am. Chem. Soc.
2007, 129, 5992–6002; V. Percec, J. Smidrkal, M. Peterca, C. M.
Mitchell, S. Nummelin, A. E. Dulcey, M. J. Sienkowska, P. A.
Heiney, Chem. Eur. J. 2007, 13, 3989–4007; V. Percec, M. Peterca,
A. E. Dulcey, M. R. Imam, S. D. Hudson, S. Nummelin, P. Adelman,
P. A. Heiney, J. Am. Chem. Soc. 2008, 130, 13079–13094; c) V.
Percec, J. G. Rudick, M. Peterca, M. E. Yurchenko, J. Smidrkal,
P. A. Heiney, Chem. Eur. J. 2008, 14, 3355–3362; d) V. Percec, M. R.
Imam, M. Peterca, D. A. Wilson, P. A. Heiney, J. Am. Chem. Soc.
2009, 131, 1294–1304.
Normalization of the diffusion coefficients by the molecular volume of
the bis-dendritic gelators was conducted by multiplying the values by the
cubed root of the molecular weights of the gelators.
Evaluation of association constants: NMR spectroscopy chemical shifts
of the bis-dendritic gelators were obtained on
a Bruker AVANCE
300 MHz NMR spectrometer. Based on the models described by
Martin,[20] association constants for the DIM model were evaluated by
using a nonlinear curve fitting module in Origin 7.0[39] with Equation (2):
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ꢀ1 þ 1 þ 8K2CT
ð2Þ
P ¼ Pl þ ðPa ꢀ PlÞ
4K2CT
in which P, CT, Pa, and Pl denote the observed chemical shift of the cor-
responding proton, total concentration in solution, the chemical shift of
the proton in the monomeric state, and the chemical shift of the proton
in the dimeric state, respectively. K2, Pa, and Pl were optimized by using
the Levenberg–Marquardt iterative method. Association constants for
the EK model were evaluated using this equation, setting KE =2K2. As-
sociation constants for the coEK model were evaluated by using a
custom Python routine provided in the Supporting Information. The
chemical shifts of protons on the periphery of aggregates were assumed
to be given by Equation (3):
Pl ¼ ðPa þ PxÞ=2
ð3Þ
[4] a) Y. Wang, F. Zeng, S. C. Zimmerman, Tetrahedron Lett. 1997, 38,
5459–5462; P. S. Corbin, L. J. Lawless, Z. Li, Y. Ma, M. J. Witmer,
S. C. Zimmerman, Proc. Natl. Acad. Sci. USA 2002, 99, 5099–5104;
Y. Ma, S. V. Kolotuchin, S. C. Zimmerman, J. Am. Chem. Soc. 2002,
124, 13757–13769; b) H.-A. Klok, J. J. Hwang, J. D. Hartgerink, S. I.
Stupp, Macromolecules 2002, 35, 6101–6111; E. R. Zubarev, S. I.
Stupp, J. Am. Chem. Soc. 2002, 124, 5762–5773; S. Lecommandoux,
H.-A. Klok, M. Sayar, S. I. Stupp, J. Polym. Sci. Polym. Chem. Ed.
2003, 41, 3501–3518; J. C. Stendahl, L. Li, R. C. Claussen, S. I.
Stupp, Biomaterials 2004, 25, 5847–5856; c) C.-H. Wong, H.-F.
Chow, S.-K. Hui, K.-H. Sze, Org. Lett. 2006, 8, 1811–1814.
in which Px denotes the chemical shift of the proton on the interior of ag-
gregates. Optimization of KE, 1, Pa, and Px was performed using
Python 2.5[40] with the SciPy and Elefant modules.
Full DSC thermograms of the bis-dendritic gelator molecules, SEM and
TEM images of the gels, and the Python script for fitting based on the
coEK model can be found in the Supporting Information.
Acknowledgements
[5] a) M. Lee, Y.-S. Jeong, B.-K. Cho, N.-K. Oh, W.-C. Zin, Chem. Eur.
J. 2002, 8, 876–883; C.-J. Jang, J.-H. Ryu, J.-D. Lee, D. Sohn, M.
Lee, Chem. Mater. 2004, 16, 4226–4231; J. Holzmueller, K. L.
Genson, Y. Park, Y.-S. Yoo, M.-H. Park, M. Lee, V. Tsukruk, Lang-
muir 2005, 21, 6392–6398; b) B.-K. Cho, A. Jain, S. Mahajan, H.
Ow, S. M. Gruner, U. Wiesner, J. Am. Chem. Soc. 2004, 126, 4070–
M.S. thanks Dr. T. F. A. de Greef for invaluable advice for NMR spec-
troscopy experiments and Dr. M. Jin for the Python script used in the
study. This work was supported by the National Research Foundation
(NRF) through ERC (R11-2007-050-04001-0) and NRL (R0A-2008-000-
20121-0) programs and by Fundamental R&D program for Core Technol-
Chem. Eur. J. 2010, 16, 2427 – 2441
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2439

S. Y. Kim et al.
4071; c) S. Coco, C. Cordovilla, B. Donnio, P. Espinet, M. J. Garcꢅa-
Casas, D. Guillon, Chem. Eur. J. 2008, 14, 3544–3552.
[12] a) W.-D. Jang, T. Aida, Macromolecules 2003, 36, 8461–8469; b) V.
Percec, M. Peterca, M. E. Yurchenko, J. G. Rudick, P. A. Heiney,
Chem. Eur. J. 2008, 14, 909–918; c) J. H. Kim, M. Seo, Y. J. Kim,
S. Y. Kim, Langmuir 2009, 25, 1761–1766; d) Y. Chen, Y. Lv, Y.
Han, B. Zhu, F. Zhang, Z. Bo, C.-Y. Liu, Langmuir 2009, 25, 8548–
8555.
[13] a) M. McWatt, G.-J. Boons, Euro. J. Org. Chem. 2001, 2535–2545;
b) C. Kim, K. T. Kim, Y. Chang, J. Am. Chem. Soc. 2001, 123, 5586–
5587; H. S. Ko, C. Park, S. M. Lee, H. H. Song, C. Kim, Chem.
Mater. 2004, 16, 3872–3876; c) H.-F. Chow, J. Zhang, Chem. Eur. J.
2005, 11, 5817–5831; H.-F. Chow, J. Zhang, Tetrahedron 2005, 61,
11279–11290.
[14] a) Y. Kamikawa, T. Kato, Langmuir 2007, 23, 274–278; b) Y. Ji,
Y. F. Luo, X.-.R. Jia, E.-Q. Chen, Y. Huang, C. Ye, B.-B. Wang, Q.-F.
Zhou, Y. Wei, Angew. Chem. 2005, 117, 6179–6183; Angew. Chem.
Int. Ed. 2005, 44, 6025–6029; W.-S. Li, X.-R. Jia, B.-B. Wang, Y. Ji,
Y. Wei, Tetrahedron 2007, 63, 8794–8800; M. Gao, G.-C. Kuang, X.-
R. Jia, W.-S. Li, Y. Li, Y. Wei, Tetrahedron Lett. 2008, 49, 6182–
6187; c) M. Yang, Z. Zhang, F. Yuan, W. Wang, S. Hess, K. Lien-
kamp, I. Lieberwirth, G. Wegner, Chem. Eur. J. 2008, 14, 3330–
3337; Z. Zhang, M. Yang, X. Zhang, L. Zhang, B. Liu, P. Zheng, W.
Wang, Chem. Eur. J. 2009, 15, 2352–2361.
[6] a) A. P. H. J. Schenning, C. Elissen-Romꢆn, J.-W. Weener,
M
.W. P. L. Baars, S. J. van der Gaast, E. W. Meijer, J. Am. Chem. Soc.
1998, 120, 8199–8208; b) K. Ariga, T. Urakawa, A. Michiue, Y.
Sasaki, J. Kikuchi, Langmuir 2000, 16, 9147–9150; K. Ariga, T. Ura-
kawa, A. Michiue, J. Kikuchi, Langmuir 2004, 20, 6762–6769;
c) A. R. Schmitzer, S. Franceschi, E. Perez, I. Rico-Lattes, A. Lattes,
L. Thion, M. Erard, C. Vidal, J. Am. Chem. Soc. 2001, 123, 5956–
5961; d) D. Joester, M. Losson, R. Pugin, H. Heinzelmann, E.
Walter, H. P. Merkle, F. Diederich, Angew. Chem. 2003, 115, 1524–
1528; Angew. Chem. Int. Ed. 2003, 42, 1486–1490; e) S. V. Aathima-
nikandan, E. N. Savariar, S. Thayumanavan, J. Am. Chem. Soc. 2005,
127, 14922–14929; A. Klaikherd, B. S. Sandanaraj, D. R. Vutukuri,
S. Thayumanavan, J. Am. Chem. Soc. 2006, 128, 9231–9237; A. V.
Ambade, S. V. Aathimanikandan, D. van der Poll, S. Thayumanavan,
J. Org. Chem. 2007, 72, 8167–8174; f) K. T. Kim, M. A. Winnik, I.
Manners, Soft Matter 2006, 2, 957–965; g) B.-B. Wang, W.-S. Li, X.-
R. Jia, M. Gao, L. Jiang, Y. Wei, J. Polym. Sci. Polym. Chem. Ed.
2008, 46, 4584–4593; h) C. N. Urbani, D. E. Lonsdale, C. A. Bell,
M. R. Whittaker, M. J. Monteiro, J. Polym. Sci. Polym. Chem. Ed.
2008, 46, 1533–1547.
[15] a) K. S. Partridge, D. K. Smith, G. M. Dykes, P. T. McGrail, Chem.
Commun. 2001, 319–320; A. R. Hirst, D. K. Smith, M. C. Feiters,
H. P. M. Geurts, A. C. Wright, J. Am. Chem. Soc. 2003, 125, 9010–
9011; A. R. Hirst, D. K. Smith, M. C. Feiters, H. P. M. Geurts, Lang-
muir 2004, 20, 7070–7077; A. R. Hirst, D. K. Smith, M. C. Feiters,
H. P. M. Geurts, Chem. Eur. J. 2004, 10, 5901–5910; A. R. Hirst,
D. K. Smith, J. P. Harrington, Chem. Eur. J. 2005, 11, 6552–6559;
b) C. S. Love, A. R. Hirst, V. Chechik, D. K. Smith, I. Ashworth, C.
Brennan, Langmuir 2004, 20, 6580–6585; J. G. Hardy, A. R. Hirst,
D. K. Smith, C. Brennan, I. Ashworth, Chem. Commun. 2005, 385–
387; B. Huang, A. R. Hirst, D. K. Smith, V. Castelletto, I. W.
Hamley, J. Am. Chem. Soc. 2005, 127, 7130–7139; c) A. R. Hirst, B.
Huang, V. Castelletto, I. W. Hamley, D. K. Smith, Chem. Eur. J.
2007, 13, 2180–2188.
[7] a) J.-F. Nierengarten, J.-F. Eckert, Y. Rio, M. del Pilar Carreon, J.-L.
Gallani, D. Guillon, J. Am. Chem. Soc. 2001, 123, 9743–9748; I.
Bury, B. Heinrich, C. Bourgogne, D. Guillon, B. Donnio, Chem. Eur.
J. 2006, 12, 8396–8413; I. Bury, B. Donnio, J.-L. Gallani, D. Guillon,
Langmuir 2007, 23, 619–625; b) A. Kimoto, J.-S. Cho, M. Higuchi,
K. Yamamoto, Macromolecules 2004, 37, 5531–5537; N. R. Luman,
M. W. Grinstaff, Org. Lett. 2005, 7, 4863–4866.
[8] a) I. Gitsov; K. L. Wooley, C. J. Hawker, P. T. Ivanova, J. M. J. Frꢃ-
chet, Macromolecules 1993, 26, 5621–5627; K. L. Wooley, C. J.
Hawker, P. T. Ivanova, J. M. J. Frꢃchet, Macromolecules 1993, 26,
5621–5627; M. R. Leduc, C. J. Hawker, J. Dao, J. M. J. Frꢃchet, J.
Am. Chem. Soc. 1996, 118, 11111–11118; J. L. Mynar, A. P. Good-
win, J. A. Cohen, Y. Ma, G. R. Fleming, J. M. J. Frꢃchet, Chem.
Commun. 2007, 2081–2082; b) D. J. Pochan, L. Pakstis, E. Huang,
C. Hawker, R. Vestberg, Macromolecules 2002, 35, 9239–9242; c) I.
Gitsov, C. Zhu, J. Am. Chem. Soc. 2003, 125, 11228–11234; K. R.
Lambrych, I. Gitsov, Macromolecules 2003, 36, 1068–1074; I.
Gitsov, A. Simonyan, N. G. Vladimirov, J. Polym. Sci. Polym. Chem.
Ed. 2007, 45, 5136–5148.
[9] a) J. Iyer, K. A. Fleming, P. T. Hammond, Macromolecules 1998, 31,
8757–8765; M. A. Johnson, J. Iyer, P. T. Hammond, Macromolecules
2004, 37, 2490–2501; P. M. Nguyen, P. T. Hammond, Langmuir 2006,
22, 7825–7832; b) L. Tian, P. Nguyen, P. T. Hammond, Chem.
Commun. 2006, 3489–3491; L. Tian, P. T. Hammond, Chem. Mater.
2006, 18, 3976–3984; c) H.-I. Lee, J. A. Lee, Z. Poon, P. T. Ham-
mond, Chem. Commun. 2008, 3726–3728.
[10] a) B.-K. Cho, A. Jain, S. M. Gruner, U. Wiesner, Science 2004, 305,
1598–1601; B.-K. Cho, A. Jain, S. M. Gruner, U. Wiesner, Chem.
Commun. 2005, 2143–2145; Y.-W. Chung, J.-K. Lee, W.-C. Zin, B.-
K. Cho, J. Am. Chem. Soc. 2008, 130, 7139–7147; E. Lee, B.-I. Lee,
S.-H. Kim, J.-K. Lee, W.-C. Zin, B.-K. Cho, Macromolecules 2009,
42, 4134–4140; b) T. M. Chapman, G. L. Hillyer, E. J. Mahan, K. A.
Shaffer, J. Am. Chem. Soc. 1994, 116, 11195–11196; c) Z. Ge, D.
Chen, J. Zhang, J. Rao, J. Yin, D. Wang, X. Wan, W. Shi, S. Liu, J.
Polym. Sci. Polym. Chem. Ed. 2007, 45, 1432–1445; d) A. Sousa-
Herves, E. Fernandez-Megia, R. Riguera, Chem. Commun. 2008,
3136–3138; e) A. Harada, K. Nakanishi, S. Ichimura, C. Kojima, K.
Kono, J. Polym. Sci. Polym. Chem. Ed. 2009, 47, 1217–1223; f) S.-J.
Hsieh, C.-C. Wang, C.-Y. Chen, Macromolecules 2009, 42, 4787–
4794.
[16] a) Q. Liu, H. Zang, S. Yin, L. Wu, C. Shao, Z. Su, Polymer 2007, 48,
3759–3770; b) Y. Feng, Z.-T. Liu, J. Liu, Y.-M. He, Q.-Y. Zheng, Q.-
H. Fan, J. Am. Chem. Soc. 2009, 131, 7950–7951.
[17] a) E. R. Zubarev, M. U. Pralle, E. D. Sone, S. I. Stupp, J. Am. Chem.
Soc. 2001, 123, 4105–4106; E. R. Zubarev, M. U. Pralle, E. D. Sone,
S. I. Stupp, Adv. Mater. 2002, 14, 198–203; E. D. Sone, E. R. Zubar-
ev, S. I. Stupp, Small 2005, 1, 694–697; E. R. Zubarev, E. D. Sone,
S. I. Stupp, Chem. Eur. J. 2006, 12, 7313–7327; b) B. W. Messmore,
J. F. Hulvat, E. D. Sone, S. I. Stupp, J. Am. Chem. Soc. 2004, 126,
14452–14458; B. W. Messmore, P. A. Sukerkar, S. I. Stupp, J. Am.
Chem. Soc. 2005, 127, 7992–7993; c) J. C. Stendahl, L. Li, E. R. Zu-
barev, Y.-R. Chen, S. I. Stupp, Adv. Mater. 2002, 14, 1540–1543.
[18] a) M. Mammen, S. K. Choi, G. M. Whitesides, Angew. Chem. 1998,
110, 2908–2953; Angew. Chem. Int. Ed. 1998, 37, 2754–2794; b) A.
Varki, Glycobiology 1993, 3, 97–130.
[19] a) C. Valꢃrio, J.-L. Fillaut, J. Ruiz, J. Guittard, J.-C. Blais, D. Astruc,
J. Am. Chem. Soc. 1997, 119, 2588–2589; b) H.-F. Chow, C.-F.
Leung, G.-X. Wang, Y.-Y. Yang, C. R. Chim. 2003, 6, 735–745; c) S.
Hong, P. R. Leroueil, I. J. Majoros, B. G. Orr, J. R. Baker, Jr.,
M. M. B. Holl, Chem. Biol. 2007, 14, 107–115; d) T. Kehat, M. Port-
noy, Chem. Commun. 2007, 2823–2825; e) G. Zaupa, P. Scrimin,
L. J. Prins, J. Am. Chem. Soc. 2008, 130, 5699–5709.
[20] R. B. Martin, Chem. Rev. 1996, 96, 3043–3064.
[21] a) B. Bai, H. Wang, H. Xin, F. Zhang, B. Long, X. Zhang, S. Qu, M.
Li, New J. Chem. 2007, 31, 401–408; S. Qu, M. Li, Tetrahedron 2008,
64, 10890–10895; b) J. E. A. Webb, M. J. Crossley, P. Turner, P. Thor-
darson, J. Am. Chem. Soc. 2007, 129, 7155–7162; c) A. R. Hirst,
I. A. Coates, T. R. Boucheteau, J. F. Miravet, B. Escuder, V. Castel-
letto, I. W. Hamley, D. K. Smith, J. Am. Chem. Soc. 2008, 130, 9113–
9121.
[22] A. Nanjia, G. R. Desiraju, Top. Curr. Chem. 1998, 198, 58–95.
[23] a) M. C. Etter, Acc. Chem. Res. 1990, 23, 120–126; b) M. C. Etter, J.
Phys. Chem. 1991, 95, 4601–4610.
[11] a) J. H. Fuhrhop, W. Helfrich, Chem. Rev. 1993, 93, 1565–1582; b) P.
Terech, R. G. Weiss, Chem. Rev. 1997, 97, 3133–3159; c) O. Gron-
wald, E. Snip, S. Shinkai, Curr. Opin. Colloid Interface Sci. 2002, 7,
148–156; d) A. R. Hirst, D. K. Smith, Top. Curr. Chem. 2005, 256,
237–273; D. K. Smith, Chem. Commun. 2006, 34–44; D. K. Smith,
Adv. Mater. 2006, 18, 2773–2778.
2440
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 2427 – 2441

Self-Association of Bis-Dendritic Organogelators
FULL PAPER
[24] U. Beginn, S. Sheiko, M. Mçller, Macromol. Chem. Phys. 2000, 201,
1008–1015.
terod, J. Chem. Soc. Perkin Trans. 1 1998, 1019–1025; b) W. Zhang,
S. O. Gonzalez, E. Simanek, Macromolecules 2002, 35, 9015–9021.
[33] J. Zhu, J.-B. Lin, Y.-X. Xu, X.-B. Shao, X.-K. Jiang, Z.-T. Li, J. Am.
Chem. Soc. 2006, 128, 12307–12313.
[34] a) D. Zhao, J. S. Moore, Org. Biomol. Chem. 2003, 1, 3471–3491;
b) T. F. A. de Greef, E. W. Meijer, Nature 2008, 453, 171–173;
c) C. A. Hunter, H. L. Anderson, Angew. Chem. 2009, 121, 7624–
7636; Angew. Chem. Int. Ed. 2009, 48, 7488–7499.
[35] Comparison of two models was performed by GraphPad QuickCalcs
provided on the website (http://www.graphpad.com/quickcalcs/
AIC1.cfm). Refer to H. J. Motulsky, A. Christopoulos, Fitting
Models to Biological Data Using Linear and Nonlinear Regression:
A Practical Guide to Curve Fitting, GraphPad Software Inc., San
Diego, CA, 2003.
[25] a) M. Seo, G. Seo, S. Y. Kim, Angew. Chem. 2006, 118, 6454–6458;
Angew. Chem. Int. Ed. 2006, 45, 6306–6310; b) M. Seo, J. H. Kim,
G. Seo, C.-H. Shin, S. Y. Kim, Chem. Eur. J. 2009, 15, 612–622.
[26] M. Seo, B. J. Beck, J. M. J. Paulusse, C. J. Hawker, S. Y. Kim, Macro-
molecules 2008, 41, 6413–6418.
[27] C. J. Hawker, J. M. J. Frꢃchet, J. Am. Chem. Soc. 1990, 112, 7638–
7647.
[28] a) M. Seredyuk, A. B. Gaspar, V. Ksenofontov, S. Reiman, Y. Galya-
metdinov, W. Haase, E. Rentschler, P. Gꢇtlich, Chem. Mater. 2006,
18, 2513–2519; b) C. Rçger, M. G. Muller, M. Lysetska, Y. Milosla-
vina, A. R. Holzwarth, F. Wꢇrthner, J. Am. Chem. Soc. 2006, 128,
6542–6543.
[36] a) A. Mulder, J. Huskens, D. N. Reinhoudt, Org. Biomol. Chem.
2004, 2, 3409–3424; b) J. W. Steed, J. L. Atwood, Supramolecular
Chemistry, Wiley, New York, 2005, p. 10; c) L. Baldini, A. Casnati,
F. Sansone, R. Ungaro, Chem. Soc. Rev. 2007, 36, 254–266.
[37] J. Wu, A. Fechtenktter, J. Gauss, M. D. Watson, M. Kastler, C. Fech-
tenktter, M. Wagner, K. Mꢇllen, J. Am. Chem. Soc. 2004, 126,
11311–11321.
[38] SimFit integrated in XWINNMR 3.0, Bruker Biospin GmbH, Rein-
stetten.
[39] Origin 7.0, OriginLab, Northampton, MA.
[29] a) B. C. Smith, Infrared Spectral Interpretation: A Systematic Ap-
proach, CRC Press, Hamilton, 1999, p. 128; b) P. G. Puranik, K. V.
Ramiah, Nature 1961, 191-192, 796–797; c) A. V. Iogansen, G. A.
Kurkchi, L. A. Dement’eva, J. Struct. Chem. 1978, 18, 589–595;
A. V. Iogansen, G. A. Kurkchi, L. A. Dement’eva, J. Struct. Chem.
1978, 18, 595–599.
[30] M. Masuda, V. Vill, T. Shimizu, J. Am. Chem. Soc. 2000, 122, 12327–
12333.
[31] a) H. Ihm, J. S. Ahn, M. S. Lah, Y. H. Ko, K. Paek, Org. Lett. 2004,
6, 3893–3896; b) M. S. Kaucher, Y.-F. Lam, S. Pieraccini, G. Gottar-
elli, J. T. Davis, Chem. Eur. J. 2005, 11, 164.
[40] Python 2.5, G. Von Rossum, Python Software Foundation, http://
www.python.org.
[32] For evaluation of association constants of dendrimers, see a) F. Os-
terod, A. Kraft, Chem. Commun. 1997, 1435–1436; A. Kraft, F. Os-
Received: September 18, 2009
Published online: January 11, 2010
Chem. Eur. J. 2010, 16, 2427 – 2441
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2441