Rational design of nanofibers and nanorings through complementary hydrogen-bonding interactions of functional π systems

Article abstract of DOI:10.1002/chem.201000839

A simple protocol to create nanofibers and -rings through a rational self-assembly approach is described. Whereas the melamine-oligo(p- phenylenevinylene) conjugate la self-aggregates to form ill-defined nanostructures, conjugate 1b, which possesses an am

Full text of DOI:10.1002/chem.201000839

DOI: 10.1002/chem.201000839
Rational Design of Nanofibers and Nanorings through Complementary
Hydrogen-Bonding Interactions of Functional p Systems
Shiki Yagai,*[a, b] Hiroaki Aonuma,[a] Yoshihiro Kikkawa,[c] Shun Kubota,[a]
Takashi Karatsu,[a] Akihide Kitamura,[a] Sankarapillai Mahesh,[d] and
Ayyappanpillai Ajayaghosh*[d]
8652
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8652 – 8661

FULL PAPER
Abstract: A simple protocol to create
nanofibers and -rings through a ration-
al self-assembly approach is described.
Whereas the melamine–oligo(p-phenyl-
enevinylene) conjugate 1a self-aggre-
gates to form ill-defined nanostruc-
tures, conjugate 1b, which possesses an
amide group as an additional interac-
tive site, self-aggregates to form 1D
nanofibers that induce gelation of the
solvent. AFM and XRD studies have
shown that dimerization through the
melamine–melamine hydrogen-bonding
interaction occurs only for 1b. Upon
complexation with 1/3 equivalents of
cyanuric acid (CA), conjugate 1a pro-
vides well-defined, ring-shaped nano-
structures at micromolar concentra-
tions, which open to form fibrous as-
semblies at submillimolar concentra-
tions and organogels in the millimolar
concentration range. Apparently, the
enhanced aggregation ability of 1a by
CA is a consequence of columnar or-
ganization of the resulting discotic
complex 1a3·CA. In contrast, coaggre-
gation of 1b with CA does not provide
well-defined nanostructures, probably
due to the interference of complemen-
tary hydrogen-bonding interactions by
the amide group.
Keywords: gels · hydrogen bonds ·
nanostructures · oligomers · self-
assembly
Introduction
sequently organize to form twisted helical nanofibers. The
latter OPV derivatives were found to form cylindrical nano-
tubes due to the formation of a cyclic hexamer (rosette)
through double hydrogen bonding between melamine units.
Thus, proper selection of a hydrogen-bonding motif enables
the construction of different types of p-conjugated nano-
structures.
The construction of well-defined nanostructures from p-con-
jugated molecules is a subject of increasing research interest
and underpins supramolecular electronics and photonics.[1]
Whereas p-conjugated molecules have the intrinsic ability to
self-assemble into extended 1D structures through p–p
stacking interactions,[2] more precise control of their dimen-
sionalities might be possible by using directional noncova-
lent interactions.[3] In this context, multiple hydrogen-bond-
ing interactions between heterocyclic compounds are partic-
ularly appealing because of their selectivity and directionali-
ty,[4] offering shape-persistent supramolecular p-conjugated
species that are capable of hierarchically assembling into
well-defined functional nanostructures through p–p stacking
interactions.[5] The elegance of this approach has been exem-
plified by the self-assembly of oligo(p-phenylenevinylene)
(OPV) derivatives equipped with ureidotriazine[6] or mela-
mine moieties.[7] The former OPV derivatives dimerize
through quadruple hydrogen-bonding interactions, and sub-
We have been exploring two-component functional supra-
molecular assemblies based on multiple hydrogen-bonding
interactions[8] due to their capability to produce diverse
supramolecular species by minor structural modifications of
either one of the two components,[9] or by changing their
mixing ratio.[10] Self-assembly and photochemical properties
of merocyanines and perylene bisimides have been success-
fully controlled through complementary triple hydrogen-
bonding interactions between melamine donor–acceptor–
donor (DAD) and imide acceptor–donor–acceptor (ADA)
hydrogen-bonding units. We have recently shown that the
self-assembly of the OPV 1a, capped on one end by a mon-
otopic melamine hydrogen-bonding module and the other
end by a tridodecyloxyphenyl wedge,[11] can be controlled by
complexation with cyanuric acid (CA) or substituted CA de-
rivatives (ddCA or dCA) possessing differing numbers of
ADA hydrogen-bonding sites (Scheme 1).[12] A particularly
salient result was observed for the 3:1 mixture of 1a with
CA; this mixture provided organogels at millimolar concen-
trations as a consequence of columnar organization of the
resulting complex. The self-aggregation of 1a was also evi-
denced from the UV/Vis spectral study, but no morphologi-
cal study has yet been addressed. Herein, we report a ra-
tional approach to control the self-aggregation of the mela-
mine-capped OPV 1a and its coaggregation with CA. To en-
hance the aggregation abilities of 1a and its complex with
CA, we have synthesized 1b, which possesses an amide
group as an additional hydrogen-bonding site.[13] In the pres-
ent study, we underline that the presence of the amide
group in 1b dramatically affects the self- and coaggregation
behavior leading to the controlled formation of rings and
fibers.
[a] Prof. Dr. S. Yagai, H. Aonuma, S. Kubota, Prof. Dr. T. Karatsu,
Prof. Dr. A. Kitamura
Department of Applied Chemistry and Biotechnology
Faculty of Engineering, Chiba University, 1-33 Yayoi-cho
Inage-ku, Chiba 263-8522 (Japan)
Fax : (+81)43-290-3039
E-mail: yagai@faculty.chiba-u.jp
[b] Prof. Dr. S. Yagai
PRESTO (Japan) Science and Technology Agency (JST)
4-1-8 Honcho Kawaguchi, Saitama (Japan)
[c] Dr. Y. Kikkawa
Photonics Research Institute
National Institute of Advanced Industrial Science and Technology
ACHTUNGTRENNUNG(AIST), 1-1-1 Higashi, Tsukuba, Ibaraki 305-8562 (Japan)
[d] S. Mahesh, Dr. A. Ajayaghosh
Photosciences and Photonics Group
Chemical Science and Technology Division
National Institute for Interdisciplinary Science and Technology
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
E-mail: ajayaghosh62@gmail.com
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000839.
Chem. Eur. J. 2010, 16, 8652 – 8661
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8653

S. Yagai, A. Ajayaghosh et al.
carried out at 1008C. After purification by column chroma-
tography, compound 1b was obtained in 45% yield. The
1
chemical structure of 1b was confirmed by H NMR spec-
troscopy and MALDI-TOF and atmospheric pressure chem-
ical ionization (APCI) mass spectrometry.
Optical properties: The UV/Vis and fluorescence spectra of
dilute solutions of 1a and 1b (c=1ꢁ10ꢀ5 m) in chloroform
and methylcyclohexane (MCH) were measured to compare
their self-aggregating behavior. Both 1a and 1b, when dis-
solved in chloroform, show similar absorption and fluores-
cence bands arising from the p–p* transition of the molecu-
larly dissolved OPV chromophores (Figures S1 and S2 in
the Supporting Information).[14] This indicates that the phe-
nylamido substituent of 1b does not affect the intrinsic opti-
cal properties of OPV chromophores.
The OPV 1a in MCH showed the p–p* transition at
lmax =376 nm, which is identical to that observed in chloro-
form (lmax =377 nm), indicating that 1a exists in the molecu-
larly dissolved state (or more precisely, free from p–p stack-
ing interactions) at this concentration (Figure S1 in the Sup-
porting Information). The fluorescence spectrum of the so-
lution in MCH is structured (lem max =415 and 439 nm) when
compared with that of the solution in chloroform (Figure S2
in the Supporting Information), however, no noticeable
change is found for the peak positions, which again illus-
trates the absence of aggregation. Self-aggregation of 1a
starts to occur at around c=5ꢁ10ꢀ4 m, shifting the p–p*
transition to lmax =365 nm.[12] The fluorescence spectrum at
this concentration (see Figure 2a, 0 equiv) does, however,
show only a slight increase in the relative fluorescence in-
tensity of the peak at 439 nm
Scheme 1. Chemical structures of the melamine–oligo(p-phenyleneviny-
lene) conjugates 1 and the complementary cyanurates.
Results and Discussion
Synthesis: Compound 1b was synthesized by the stepwise
introduction of the aniline derivative 3 and dioctylamine
onto the previously reported OPV 4 (Scheme 2).[14] Because
of the relatively low reactivity of 3, the synthesis of 1b was
compared with that of the 1ꢁ
10ꢀ5 m solution (Figure 1). This
observation indicates that the
365 nm absorbing species are
weakly interacting molecular
assemblies,
which
undergo
rapid exchange with molecular-
ly dissolved species.
Interestingly, the p–p* transi-
tion of the amide-functionalized
1b in MCH at c=1ꢁ10ꢀ5 m dis-
played
a large shift towards
lmax =350 nm with an apprecia-
ble hypochromic effect com-
pared with that of 1a
(Figure 1). In the fluorescence
spectrum, a significant decrease
in the peak intensities at l=
415 and 439 nm are observed
when compared with that of 1a,
whereas peaks in the longer
wavelength region (at l=469,
500, and 540 nm) had increases
in their intensities. These spec-
Scheme 2. Synthesis of 1b: a) dodecylamine, pyridine, toluene, 808C, 3 h; b) hydrazine monohydrate, Pd/C,
ethanol, 808C, 1 h; c) 3, diisopropylethylamine, 1,4-dioxane, 1008C, 12 h and then dioctylamine, reflux, 24 h. tral changes are characteristic
8654
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8652 – 8661

Nanofibers and Nanorings of Functional p Systems
FULL PAPER
Figure 1. The UV/Vis (left axis) and fluorescence (right axis; lex
=
349 nm) spectra of 1a (solid line) and 1b (dashed line) in MCH at 208C.
Concentration: 1ꢁ10ꢀ5 m.
of the p–p stacking of OPV chromophores[6a,14,15] and dem-
onstrate that the amide moiety of 1b enhances the self-ag-
gregation propensity of the melamine-capped OPV building
blocks.
CA is a tritopic ADA-type hydrogen-bonding module
that enables the trimerization of the melamine-capped
OPVs through complementary hydrogen bonding.[16] The re-
sulting 3:1 complex of 1 with CA has an expanded p surface,
shifting the aggregation equilibrium towards a p–p stacked
state, compared with 1 alone. We thus investigated the com-
plexation of 1 with CA by using fluorescence spectroscopy
because of its susceptibility to p–p stacking interactions. Fig-
ure 2a shows the fluorescence spectral change of 1a with an
increasing concentration of CA. The concentration of 1a
was set to 5ꢁ10ꢀ4 m to ensure the presence of sufficient hy-
drogen-bonding interactions. Upon increasing the concentra-
tion of CA, the fluorescence peak at l=417 nm derived
from molecularly dissolved OPV chromophores decreased,
and concurrently a new peak at l=463 nm emerged as a
result of p–p stacking of the OPV chromophores. When the
ratio of these fluorescence peak intensities (I463/I417) was
plotted against the molar ratio [CA]/[1a] the maximum was
observed at [CA]/[1a]=0.33 (Figure 2b), suggesting a 3:1
complexation between 1a and CA. The quantitative binding
of 1a to all the ADA binding sites of CA under the present
condition should, however, not be allowed if DAD–ADA
hydrogen bonding is the only intermolecular interaction.[8c,17]
Thus, it is believed that the resulting complexes were further
stabilized by p–p stacking interactions as shown by the ap-
preciable fluorescence spectral changes.
Figure 2. a) Fluorescence titration of 1a ([1a]=5ꢁ10ꢀ4 m) with CA (0 to
1 equiv) in MCH; lex =349 nm. The spectra were normalized at their
maxima. Arrows indicate the changes upon an increase in the concentra-
tion of CA. b) Plot of the ratio of fluorescence intensities at l=463 and
417 nm versus the [CA]/[1a] ratio. Dotted line indicates [CA]/[1a]=0.33.
formed. These may even behave as terminating species in
the hierarchical organization of 1a3·CA as inferred from the
abrupt increase of the I463/I417 value at [CA]/[1a]=0.33.
For the complex 1a3·CA, the formation of the two stereo-
isomers i and ii, shown in Figure 3a, should be considered
with respect to the binding orientation of the three 1a mole-
cules. The stereoisomer i has a C3-symmetrical structure,
whereas ii is characterized by the presence of a p–p stacked
OPV dimer within the structure. Such a dimeric unit can be
produced by the 2:1 complexation of 1a with the ditopic
dCA (stereoisomer iv of 1a2·dCA in Figure 3b). To obtain
information on the supramolecular structure of complex
1a3·CA, we compared the UV/Vis and fluorescence spectra
of this complex with those of 1a2·dCA (Figure 3c and d). It
has been revealed in a previous study that the 2:1 mixtures
of 1a and dCA in MCH show relatively sharp absorption
spectra with blue-shifted maxima at around l=355 nm over
a wide concentration range ([1a]=1ꢁ10ꢀ6 to 5ꢁ10ꢀ4 m, see
Figure 3c for the spectrum of the 1ꢁ10ꢀ5 m solution).[12] The
large blue shift (l=376!355 nm) and the concentration in-
dependence is consistent with the formation of the stereo-
isomer iv in which a p-stacked OPV dimer is locked by
dCA through triple hydrogen-bonding interactions. In sharp
Of further interest is the reappearance of the fluorescence
peak of the monomeric species at l=417 nm upon the addi-
tion of greater amounts of CA ([CA]/[1a]>0.33, Figure 2).
This finding suggests that the mixing of the two components
with the exact 3:1 molar ratio is a prerequisite for the for-
mation of the complex 1a3·CA. In the presence of greater
amounts of CA, additional hydrogen-bonded complexes,
such as 1a2·CA and 1a·CA, which have less tendency for ag-
gregation through p–p stacking interactions, might be
Chem. Eur. J. 2010, 16, 8652 – 8661
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8655

S. Yagai, A. Ajayaghosh et al.
from such tightly interacting di-
meric OPVs. Thus, it was con-
cluded that complex 1a3·CA
does not form the stereoiso-
meric ii, and the observed spec-
tral features of this complex
emerging in [1a]>5ꢁ10ꢀ4 m are
purely attributed to a hierarchi-
cal stacking of stereoisomer i.
Why the complex 1a3·CA does
not take the stereoisomeric
form ii is most likely to be due
to the effective hierarchical or-
ganization of the C3-symmetri-
cal discotic structure of the ste-
reoisomer i (Figure 4b).
The complexation of 1b with
CA was also investigated with
fluorescence spectroscopy, how-
ever, no notable spectral
changes were observed. Two
possibilities might be suggested
for this result: 1) CA does not
bind to 1b due to the strong
self-aggregation of 1b; 2) CA
binds to 1b to form some kind
of coaggregates in which the
OPV chromophores are fully
p–p stacked as in self-aggregat-
ed 1b. This point is investigated
by using AFM (see below).
AFM analysis of nanostruc-
tures: Solutions of 1a (c=5ꢁ
10ꢀ4 m) and 1b (c=1ꢁ10ꢀ4 m) in
MCH were spin-coated onto
highly oriented pyrolytic graph-
ite (HOPG) and the resulting
nanostructures were visualized
by AFM. The OPV 1a, for
which the UV/Vis data at this
concentration showed the pres-
Figure 3. The possible stereoisomers of complexes a) 1a3·CA and b) 1a2·dCA. c) UV/Vis and d) fluorescence
spectra of 1a (dashed line, [1a]=1ꢁ10ꢀ5 m), 1a2·dCA (dotted line, [1a]=1ꢁ10ꢀ5 m), and 1a3·CA (solid line,
[1a]=5ꢁ10ꢀ4 m) in MCH; lex =349 nm. Different concentrations were chosen for the respective samples (see
the text).
contrast, the absorption maxima of the 3:1 mixtures of 1a
and CA remain at l=376 nm up to 1ꢁ10ꢀ4 m, and shift to
l=363 nm at [1a]=5ꢁ10ꢀ4 m. A further increase in the con-
centration no longer impacts the absorption maximum.
These observations clearly demonstrate that complex
1a3·CA does not feature a p-stacked OPV dimer locked by
CA. Moreover, the complex 1a2·dCA emanates in a more
red-shifted region than complex 1a3·CA (Figure 3d; lem
max =476 nm for 1a2·dCA and 463 nm for 1a3·CA). The
larger Stokes shift of 1a2·dCA (121 nm) than that of 1a3·CA
(100 nm) can be rationalized by the formation of the stereo-
isomer iv in which the two OPV chromophores are strongly
stacked within the complex (Figure 4a). A closer inspection
of the fluorescence spectrum of the complex 1a3·CA re-
vealed that there was no contamination of the fluorescence
ence of p–p stacking, exhibited ill-defined nanostructures
(Figure S3 in the Supporting Information). This observation
indicates that 1a has a low propensity to assemble well-de-
Figure 4. The energy-minimized structures of a) 1a2·dCA (stereoisomeric
form iv) and b) 1a3·CA (stereoisomeric form i) obtained by molecular
mechanical calculations.
8656
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8652 – 8661

Nanofibers and Nanorings of Functional p Systems
FULL PAPER
fined nanostructures and hence nondirectional association
occurs. On the contrary, well-defined fibrillar morphologies
were visualized for 1b (Figure 5a and b). The height of the
elementary rod-shaped nanostructure is 8 nm (Figure 5c, d),
twice the molecular length of 1b with extended alkyl chains
(see Figure 9 f below). Thus, it is likely that 1b forms dimers
by hydrogen-bonding interactions between melamine moie-
ties[7a,18] and the resulting dimers stack to form nanofibers
that are stabilized by hydrogen bonding between amide
groups (Scheme 3a). FTIR spectroscopy of thin films of 1b
exhibited an amide I band at 1631 cmꢀ1, confirming the for-
mation of the hydrogen-bonding interaction. The dimeriza-
tion of 1b is further confirmed by the XRD study of the
bulk material (see below).
Solutions of the melamine-capped OPVs (c=2ꢁ10ꢀ4 m) in
MCH containing 1/3 equivalents of CA were also spin-
coated onto HOPG and imaged by AFM. Remarkably, the
samples prepared from the mixtures of 1a and CA showed
agglomerated and isolated ring-shaped nanostructures (Fig-
ure 6a). Similar ring-shaped nanostructures have recently
been discovered for small amphiphilic molecular building
blocks[19] and several hydrogen-bonded disc-shaped assem-
blies (rosettes) by our group.[20] High-resolution AFM imag-
ing of the nanorings revealed the size range of 20–50 nm
with an average height of 3–4 nm. In sharp contrast to 1a,
CA had a negative effect on the self-assembly of the amide-
functionalized 1b. AFM images of the 3:1 mixture of 1b
Figure 5. AFM a) height and b) phase images of self-aggregated 1b spin-
cast from a solution in MCH (c=1.0ꢁ10ꢀ4 m) onto HOPG. c) Magnified
AFM height image of elementary nanofibers of 1b and d) cross-sectional
analysis along the white line in c). z scale=30 nm in a) and 12 nm in c).
and CA exhibited only amorphous structures. The fibrous
nanostructures of 1b completely disappeared upon the addi-
tion of CA (Figure S4 in the Supporting Information). This
observation indicates that 1b binds with CA, but the result-
ing coaggregates lack the ability to organize into well-de-
Scheme 3. Schematic representation of the a) self-aggregation of 1b and b) coaggregation of 1a with cyanuric acid (CA).
Chem. Eur. J. 2010, 16, 8652 – 8661
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8657

S. Yagai, A. Ajayaghosh et al.
DLS analysis of nanostructures: To support the formation of
the nanostructures in solution, solutions of self- and coag-
gregates in MCH were assessed by dynamic light scattering
(DLS). No analyzable particle was detected for the solutions
of 1a and 1b3·CA in the concentration ranging from 1ꢁ10ꢀ4
to 5ꢁ10ꢀ4 m, which is consistent with the results of the AFM
studies. For a 1ꢁ10ꢀ4 m solution of the amide-functionalized
1b, however, extended assemblies with hydrodynamic diam-
eters (DH) close to 1 mm were detected (top data in Fig-
ure 8a). Interestingly, the average DH of the assemblies
Figure 6. a–c) AFM height image of 1a3·CA spin-coated from a solution
in MCH (c=2.0ꢁ10ꢀ4 m) and d) cross-sectional analysis along the white
line in c). z scale=6 nm in a) and 5 nm in b).
fined nanostructures. Nonspecific hydrogen bonding be-
tween CA and the amide group of 1b might occur to form
various hydrogen-bonded coaggregates.
Figure 7. a) AFM height image of methylcyclohexane gel of 1a3·CA. z
scale=10 nm. Inset shows a picture of the gel. b) AFM height image of
1a3·CA spin-cast from a solution in MCH (c=5.0ꢁ10ꢀ4 m). z scale=
6 nm.
Gelation: Gelation of solvents by self-assembled, low-mo-
lecular-weight compounds is a good indicator of the forma-
tion of elongated fibrous nanostructures.[21] Furthermore,
the formation of an organogel as a result of supramolecular
organization of a small functional molecule, such as an
OPV, allows the formation of optically and electronically
active soft materials.[1,22] Appropriate amounts of 1a, 1b,
and their 3:1 mixtures with CA were dissolved in MCH with
heating, and the resulting homogeneous solutions were
cooled to room temperature to examine the gelation behav-
ior. Amide-functionalized 1b instantly formed transparent
gels with a concentration above 2ꢁ10ꢀ3 m. This observation
is in good agreement with the formation of 1D nanostruc-
tures visualized by AFM (Figure 5). In contrast, neither gels
nor viscous liquids were formed for 1a when homogeneous
solutions (ꢁ5ꢁ10ꢀ3 m) were kept for several days. This ob-
servation is also consistent with the formation of ill-defined
nanostructures revealed by AFM (Figure S3 in the Support-
ing Information).
Remarkably, the gelation abilities of 1a and 1b are re-
versed in the presence of 1/3 equivalents of CA. When a ho-
mogeneous solution of the 3:1 mixture of 1a (c=5ꢁ10ꢀ3 m)
and CA in MCH was kept at room temperature, a transpar-
ent gel was formed after one day. The AFM observation of
a dried gel did not show any fibrous structures typically ob-
served for low-molecular-weight gels (Figure 7a). Subse-
quently, the gel was diluted to c=5ꢁ10ꢀ4 m. The AFM
images of the spin-cast solution exhibited short wormlike
structures (Figure 7b), which are considered to be inter-
mediate species between the gel-forming extended columnar
structures and the nanorings formed in solutions at lower
concentration (Scheme 3b). These structures were <300 nm
in length and 10–30 nm in width.
showed a time-dependent increase that exceeded 2 mm after
50 min (from top to bottom in Figure 8a). This is an indica-
tion of the presence of nonequilibrated extended supra-
molecular assemblies.
In contrast, solutions of 1a3·CA exhibited equilibrated,
time-independent DLS results. For a 1ꢁ10ꢀ4 m solution, as-
semblies for which the DH values were 60–90 nm were de-
tected in addition to larger assemblies around 300–400 nm
(Figure 8b). The former assemblies are believed to be the
ring-shaped nanostructures observed by AFM. The smaller
sizes of the rings observed by AFM compared with those
from the DLS studies might be due to shrinkage of the
sample during the evaporation of the solvent.[20a] The larger
assemblies are, on the other hand, considered to be open-
ended fibrous assemblies that are not detected by AFM.
Upon increasing the concentration to 5ꢁ10ꢀ4 m, the smaller
assemblies clearly decreased, and the larger assemblies were
the predominantly existing species (Figure 8c). These results
agree well with the result of AFM studies and clearly dem-
onstrate the concentration-dependent formation of closed
and open structures of columnar stacks of the discotic supra-
molecular species 1a3·CA.[20b]
Packing structures of OPV self-aggregates: To gain a deeper
insight into the dramatically different self-aggregation abili-
ties of 1a and 1b, solid-state packing structures of these
OPVs were investigated with polarized optical microscopy
(POM), differential scanning calorimetry (DSC), and XRD
techniques. The POM and DSC observations revealed that
1a exhibits shearable birefringent mesophases between 71
(2.5 kJmolꢀ1) and 858C (25.4 kJmolꢀ1), and 1b between 53
8658
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8652 – 8661

Nanofibers and Nanorings of Functional p Systems
FULL PAPER
tion of 1b might be due to the
stabilization of the resulting
dimers by hierarchical organiza-
tion through two-point hydro-
gen-bonding interactions be-
tween amide functionalities and
p–p
stacking
interaction
(Scheme 3a).
Conclusion
By the exploitation of the exist-
ing knowledge of the self-as-
sembly behavior of OPVs in
conjunction with the multiple
hydrogen-bonding
melamine
moiety, we could rationally
design molecular systems that
Figure 8. Time-dependent (from top to bottom) changes of the hydrodynamic diameter (DH) distributions ob-
tained by ten DLS measurements of solutions in MCH: a) 1b (1ꢁ10ꢀ4 m), b) 1a3·CA (1ꢁ10ꢀ4 m), and
c) 1a3·CA (5ꢁ10ꢀ4 m). Data acquisition was started 10 min after the preparation of sample solutions and were self- and coassemble into supra-
taken at 5 min intervals.
molecular architectures with
distinct morphologies. Thus, we
(1.1 kJmolꢀ1) and 1588C (10.3 kJmolꢀ1). In the temperature
ranges stated, the observed textures were easily broken by
shearing. Above the temperature ranges, both compounds
exist as isotropic liquids. The considerably different clearing
temperatures suggest that the intermolecular interactions in-
volved in the two mesophases are dramatically different.
Upon cooling the isotropic liquids, shearable birefringent
textures were observed for 1a and 1b (Figure 9a and b).
These textures are considerably different from those of the
columnar liquid crystalline mesophases of quadruply hydro-
gen-bonded OPV dimers.[23] The more well-defined fibrous
texture of 1b than that of 1a is indicative of a higher degree
of molecular ordering in the former mesophase.
The XRD pattern of 1a in the mesophase showed broad
reflections at 45.3, 22.6, and 17.8 ꢂ (Figure 9c), demonstrat-
ing the formation of a lamellar structure with interlayer
spacings around 45 ꢂ. Since the molecular length of 1a with
a fully extended conformation is around 58 ꢂ, a multilamel-
lar packing of 1a with interdigitation of alkyl chains is
strongly suggested (Figure 9e).[24] This observation excludes
the possibility of dimerization of 1a by melamine–melamine
hydrogen-bonding interactions. In contrast, compound 1b
showed much sharper XRD peaks at 73.6, 36.3, and 25.0 ꢂ
with several unidentifiable peaks, indicating the formation
of a more ordered multilamellar structure (Figure 9d). Inter-
estingly, the interlayer spacing of 1b (73.6 ꢂ) is much longer
than that of 1a. Significantly, different interlayer spacings
between the two compounds cannot be explained by the dif-
ference in their molecular lengths (1a=58 and 1b=65 ꢂ).
Combined with the formation of fibrous nanostructures with
the width of 8 nm (Figure 5), it is strongly suggested that 1b
dimerizes by hydrogen-bonding interactions between the
melamine moieties, and the resulting dimer is the building
block for the extended self-assembled architectures (Fig-
ure 9 f). The occurrence of dimerization in the self-aggrega-
could demonstrate that the melamine-capped OPV 1a self-
aggregated to form ill-defined structures, and coaggregated
with CA at low concentrations to form nanorings and at
high concentrations to form open structures, leading to the
gelation of solvents. On the other hand, the OPV 1b, which
has an additional amide group, self-assembled to form nano-
fibers, leading to gelation of solvents. The coassembly of CA
with 1b resulted in ill-defined structures with gelation abili-
ties. By using spectroscopic, DLS, and AFM analyses, it was
revealed that the amide group strongly enhances the self-ag-
gregation of the melamine-capped OPV building blocks.
XRD in the liquid-crystalline state suggested that dimeriza-
tion by melamine–melamine hydrogen-bonding interactions
occur only for the amide derivative, thereby enhancing its
self-aggregation ability. The poor aggregation ability of the
melamine-capped OPV lacking the amide group could be
improved by the addition of 1/3 equivalents of CA due to
the formation of C3-symmetrical hydrogen-bonded discs.
Such discs possessing three OPV p systems hierarchically or-
ganized in a nonpolar solvent to form a ring-shaped mor-
phology at micromolar concentrations and gel-forming ex-
tended columnar structures at millimolar concentrations.
Thus, the present study demonstrates the possibility of con-
structing defined supramolecular architectures with mela-
mine-capped p-conjugated molecules. We are currently ap-
plying the present molecular design strategy to the creation
of various p-conjugated functional architectures with de-
fined size and shape.
Acknowledgements
This work was partially supported by a Grant-in-Aid for Scientific Re-
search from the Japan Society for the Promotion of Science (JSPS). S.Y.,
S.M., A.A., and A.K. thank the Department of Science and Technology,
Chem. Eur. J. 2010, 16, 8652 – 8661
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8659

S. Yagai, A. Ajayaghosh et al.
[2] C. A. Hunter, J. K. M. Sanders, J.
Am. Chem. Soc. 1990, 112, 5525–
5534.
[3] a) G. M. Whitesides, J. P. Mathias,
C. T. Seto, Science 1991, 254,
1312–1319; b) J.-M. Lehn, Supra-
molecular Chemistry: Concepts
and Perspectives, VCH, Wein-
heim, 1995; c) T. Kato, Struct.
Bonding (Berlin) 2000, 96, 95–
146.
[4] a) G. M. Whitesides, E. E. Sima-
nek, J. P. Mathias, C. T. Seto, D.
Chin, M. Mammen, D. M.
Gordon, Acc. Chem. Res. 1995,
28, 37–44; b) S. C. Zimmerman,
P. S. Corbin, Struct. Bonding
(Berlin)
2000,
96,
63–94;
c) M. W. Peczuh, A. D. Hamilton,
Chem. Rev. 2000, 100, 2479–
2494; d) L. J. Prins, D. N. Rein-
houdt, P. Timmerman, Angew.
Chem. 2001, 113, 2446–2492;
Angew. Chem. Int. Ed. 2001, 40,
2382–2426; e) R. P. Sijbesma,
E. W. Meijer, Chem. Commun.
2003, 5–16; f) S. Lena, S. Ma-
siero, S. Pieraccini, G. P. Spada,
Chem. Eur. J. 2009, 15, 7792–
7806; g) U. Drechsler, B. Erdo-
gan, V. M. Rotello, Chem. Eur. J.
2004, 10, 5570–5579; h) S. Yagai,
J. Photochem. Photobiol. C 2006,
7, 164–182.
[5] a) H. M. Keizer, R. P. Sijbesma,
Chem. Soc. Rev. 2005, 34, 226–
234; b) T. Kato, T. Yasuda, Y. Ka-
mikawa, M. Yoshio, Chem.
Commun. 2009, 729–739; c) T.
Kato, N. Mizoshita, K. Kishimoto,
Angew. Chem. 2006, 118, 44–74;
Angew. Chem. Int. Ed. 2006, 45,
38–68.
[6] a) P. Jonkheijm, F. J. M. Hoeben,
R. Kleppinger, J. Van Herrikhuy-
zen, A. P. H. J. Schenning, E. W.
Meijer, J. Am. Chem. Soc. 2003,
125, 15941–15949; b) P. Jonk-
heijm, P. P. A. M. van der Schoot,
Figure 9. Polarized optical micrographs (a, b), XRD patterns (c, d), and proposed lamellar packing structures
(e, f) of 1a (a, c, e) at 758C and 1b (b, d, f) at 1758C upon cooling from the isotropic melt (cooling rate=18C
minꢀ1). The peaks designated by asterisk in d) are unidentified peaks.
A. P. H. J.
Schenning,
E. W.
Meijer, Science 2006, 313, 80–83;
c) S. J. George, Z. Tomovic,
M. M. J. Smulders, T. F. A. d. Greef, P. E. L. G. Leclere, E. W.
Meijer, A. P. H. J. Schenning, Angew. Chem. 2007, 119, 8354–8359;
Angew. Chem. Int. Ed. 2007, 46, 8206–8211.
New Delhi and the Japan Society for the Promotion of Science, Japan for
financial support. S. Mahesh thanks the University Grants Commission,
New Delhi for a fellowship. We thank Dr. Yasuo Norikane of the Nation-
al Institute of Advanced Industrial Science and Technology (AIST), Tsu-
kuba, for the measurement of MALDI-TOF MS spectrometry.
[7] P. Jonkheijm, A. Miura, M. Zdanowska, F. J. M. Hoeben, S. De Feyt-
er, A. P. H. J. Schenning, F. C. De Schryver, E. W. Meijer, Angew.
Chem. 2004, 116, 76–80; Angew. Chem. Int. Ed. 2004, 43, 74–78.
[8] For examples of two-component functional dye assemblies based on
multiple hydrogen-bonding interactions, see: a) M. Kotera, J. M.
Lehn, J. P. Vigneron, J. Chem. Soc. Chem. Commun. 1994, 197–199;
b) N. Kimizuka, T. Kawasaki, K. Hirata, T. Kunitake, J. Am. Chem.
Soc. 1995, 117, 6360–6361; c) F. Wꢃrthner, C. Thalacker, A. Sautter,
W. Schartl, W. Ibach, O. Hollricher, Chem. Eur. J. 2000, 6, 3871–
3886; d) F. Wꢃrthner, S. Yao, B. Heise, C. Tschierske, Chem.
Commun. 2001, 2260–2261; e) L. J. Prins, C. Thalacker, F. Wꢃrthner,
P. Timmerman, D. N. Reinhoudt, Proc. Natl. Acad. Sci. USA 2001,
98, 10042–10045; f) A. P. H. J. Schenning, V. H. Jeroen, P. Jonk-
heijm, Z. Chen, F. Wꢃrthner, E. W. Meijer, J. Am. Chem. Soc. 2002,
[1] a) F. Wꢃrthner, Chem. Commun. 2004, 1564–1579; b) A. F. J. M.
Hoeben, P. Jonkheijm, E. W. Meijer, A. P. H. J. Schenning, Chem.
Rev. 2005, 105, 1491–1546; c) J. W. W. Pisula, K. Mꢃllen, Chem. Rev.
2007, 107, 718–747; d) A. Ajayaghosh, V. K. Praveen, Acc. Chem.
Res. 2007, 40, 644–656; e) L. Zang, Y. Che, J. S. Moore, Acc. Chem.
Res. 2009, 42, 1596–1608; f) V. Percec, M. Glodde, T. K. Bera, Y.
Miura, I. Shiyanovskaya, K. D. Singer, V. S. K. Balagurusamy, P. A.
Heiney, I. Schnell, A. Rapp, H. W. Spiess, S. D. Hudson, H. Duan,
Nature 2002, 417, 384–387; g) J. P. Hill, W. Jin, A. Kosaka, T. Fu-
kushima, H. Ichihara, T. Shimomura, K. Ito, T. Hashizume, N. Ishii,
T. Aida, Science 2004, 304, 1481–1483.
8660
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8652 – 8661

Nanofibers and Nanorings of Functional p Systems
FULL PAPER
124, 10252–10253; g) F. Wꢃrthner, S. Yao, J. Org. Chem. 2003, 68,
8943–8949; h) F. J. M. Hoeben, M. J. Pouderoijen, A. P. H. J. Schen-
ning, E. W. Meijer, Org. Biomol. Chem. 2006, 4, 4460–4462; i) H.
Nakade, B. J. Jordan, H. Xu, G. Han, S. Srivastava, R. R. Arvizo, G.
Cooke, V. M. Rotello, J. Am. Chem. Soc. 2006, 128, 14924–14929;
j) A. Llanes-Pallas, C.-A. Palma, L. Piot, A. Belbakra, A. Listorti,
M. Prato, P. Samori, N. Armaroli, D. Bonifazi, J. Am. Chem. Soc.
2009, 131, 509–520; k) K. Yoosaf, A. Belbakra, N. Armaroli, A.
Llanes-Pallas, D. Bonifazi, Chem. Commun. 2009, 2830–2832.
[9] a) S. Yagai, M. Higashi, T. Karatsu, A. Kitamura, Chem. Mater.
2004, 16, 3582–3585; b) S. Yagai, M. Higashi, T. Karatsu, A. Kita-
mura, Chem. Mater. 2005, 17, 4392–4398; c) S. Yagai, M. Higashi, T.
Karatsu, A. Kitamura, Chem. Commun. 2006, 1500–1502; d) S.
Yagai, T. Kinoshita, M. Higashi, K. Kishikawa, T. Nakanishi, T. Kar-
atsu, A. Kitamura, J. Am. Chem. Soc. 2007, 129, 13277–13287; e) S.
Yagai, Y. Monma, N. Kawauchi, T. Karatsu, A. Kitamura, Org. Lett.
2007, 9, 1137–1140; f) T. Seki, S. Yagai, T. Karatsu, A. Kitamura, J.
Org. Chem. 2008, 73, 3328–3335; g) T. Seki, S. Yagai, T. Karatsu, A.
Kitamura, Chem. Lett. 2008, 37, 764–765; h) S. Yagai, T. Kinoshita,
Y. Kikkawa, T. Karatsu, A. Kitamura, Y. Honsho, S. Seki, Chem.
Eur. J. 2009, 15, 9320–9324.
[10] a) S. Yagai, T. Seki, T. Karatsu, A. Kitamura, F. Wꢃrthner, Angew.
Chem. 2008, 120, 3415–3419; Angew. Chem. Int. Ed. 2008, 47, 3367–
3371; b) S. Yagai, S. Hamamura, H. Wang, V. Stepanenko, T. Seki,
K. Unoike, Y. Kikkawa, T. Karatsu, A. Kitamura, F. Wꢃrthner, Org.
Biomol. Chem. 2009, 7, 3926–3929; c) S. Mahesh, R. Thirumalai, S.
Yagai, A. Kitamura, A. Ajayaghosh, Chem. Commun. 2009, 5984–
5986; d) T. Tazawa, S. Yagai, Y. Kikkawa, T. Karatsu, A. Kitamura,
A. Ajayaghosh, Chem. Commun. 2010, 46, 1076–1078.
[11] B. M. Rosen, C. J. Wilson, D. A. Wilson, M. Peterca, M. R. Imam, V.
Percec, Chem. Rev. 2009, 109, 6275–6540.
[12] S. Yagai, S. Kubota, K. Unoike, T. Karatsu, A. Kitamura, Chem.
Commun. 2008, 4466–4468.
A. Ajayaghosh, Chem. Eur. J. 2005, 11, 3217–3227; c) B. W. Mess-
more, J. F. Hulvat, E. D. Sone, S. I. Stupp, J. Am. Chem. Soc. 2004,
126, 14452–14458; d) J. F. Hulvat, M. Sofos, K. Tajima, S. I. Stupp, J.
Am. Chem. Soc. 2005, 127, 366–372; e) K. Tajima, L.-S. Li, S. I.
Stupp, J. Am. Chem. Soc. 2006, 128, 5488–5495.
[16] M. Arduini, M. Crego-Calama, P. Timmerman, D. N. Reinhoudt, J.
Org. Chem. 2003, 68, 1097–1106.
[17] F. J. M. Hoeben, J. Zhang, C. C. Lee, M. J. Pouderoijen, M. Wolffs,
F. Wꢃrthner, A. P. H. J. Schenning, E. W. Meijer, S. De Feyter,
Chem. Eur. J. 2008, 14, 8579–8589.
[18] a) J. Barberꢄ, L. Puig, J. L. Serrano, T. Sierra, Chem. Mater. 2004,
16, 3308–3317; b) J. Barberꢄ, L. Puig, P. Romero, J. L. Serrano, T.
Sierra, J. Am. Chem. Soc. 2005, 127, 458–464.
[19] J.-K. Kim, E. Lee, Z. Huang, M. Lee, J. Am. Chem. Soc. 2006, 128,
14022–14023.
[20] a) S. Yagai, S. Mahesh, Y. Kikkawa, K. Unoike, T. Karatsu, A. Kita-
mura, A. Ajayaghosh, Angew. Chem. 2008, 120, 4769–4772; Angew.
Chem. Int. Ed. 2008, 47, 4691–4694; b) S. Yagai, S. Kubota, H.
Saito, K. Unoike, T. Karatsua, A. Kitamura, A. Ajayaghosh, M. Ka-
nesato, Y. Kikkawa, J. Am. Chem. Soc. 2009, 131, 5408–5410.
[21] a) P. Terech, R. G. Weiss, Chem. Rev. 1997, 97, 3133–3160; b) J. H.
Van Esch, B. L. Feringa, Angew. Chem. 2000, 112, 2351–2354;
Angew. Chem. Int. Ed. 2000, 39, 2263–2266; c) L. A. Estroff, A. D.
Hamilton, Chem. Rev. 2004, 104, 1201–1218; d) N. M. Sangeetha, U.
Maitra, Chem. Soc. Rev. 2005, 34, 821–836; e) T. Ishi-i, S. Shinkai,
Top. Curr. Chem. 2005, 258, 119–160.
[22] a) A. Ajayaghosh, S. J. George, A. P. H. J. Schenning, Top. Curr.
Chem. 2005, 258, 83–118; b) A. Ajayaghosh, C. Vijayakumar, V. K.
Praveen, S. S. Babu, R. Varghese, J. Am. Chem. Soc. 2006, 128,
7174–7175; c) A. Ajayaghosh, V. K. Praveen, C. Vijayakumar, S. J.
George, Angew. Chem. 2007, 119, 6376–6381; Angew. Chem. Int.
Ed. 2007, 46, 6260–6265; d) A. Ajayaghosh, V. K. Praveen, S. Srini-
vasan, R. Varghese, Adv. Mater. 2007, 19, 411–415; e) A. Ajaya-
ghosh, V. K. Praveen, C. Vijayakumar, Chem. Soc. Rev. 2008, 37,
109–122.
[13] a) Y. Kobayashi, Y. Matsunaga, Bull. Chem. Soc. Jpn. 1987, 60,
3515–3518; b) A. J. Wilson, M. Masuda, R. P. Sijbesma, E. W.
Meijer, Angew. Chem. 2005, 117, 2315–2319; Angew. Chem. Int. Ed.
2005, 44, 2275–2279; c) F. Wꢃrthner, C. Bauer, V. Stepanenko, S.
Yagai, Adv. Mater. 2008, 20, 1695–1698.
[23] W. Pisula, Z. Tomovic, M. Wegner, R. Graf, B. Maarten, J. Pouderoi-
jen, E. W. Meijer, A. P. H. J. Schenning, J. Mater. Chem. 2008, 18,
2968–2977.
[14] S. Yagai, S. Kubota, T. Iwashima, K. Kishikawa, T. Nakanishi, T.
Karatsu, A. Kitamura, Chem. Eur. J. 2008, 14, 5246–5257.
[24] The antiparallel packing of 1a within a layer may be another pro-
posed structure.
[15] a) J.-F. Eckert, J.-F. Nicoud, J.-F. Nierengarten, S.-G. Liu, L. Eche-
goyen, F. Barigelletti, N. Armaroli, L. Ouali, V. Krasnikov, G. Had-
ziioannou, J. Am. Chem. Soc. 2000, 122, 7467–7479; b) S. J. George,
Received: April 3, 2010
Published online: July 7, 2010
Chem. Eur. J. 2010, 16, 8652 – 8661
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8661