Visualization of various supramolecular assemblies of oligo(para- phenylenevinylene)-melamine and perylene bisimide

Article abstract of DOI:10.1002/chem.200800760

A melamine derivative has been covalently equipped with two oligo(para-phenylenevinylene) (OPV) chromophores. This procedure yields a bifunctional molecule with two hydrogen-bonding arrays available for complementary binding to perylene bisimide derivatives. Depending on the solvent, hydrogen-bonded trimers, tetramers, and dimers on a graphite surface are observed for pure OPV-melamine by using scanning tunneling microscopy (STM). Upon the addition of perylene bisimide, linear tapes of perylene bisimide, 12-membered rosettes that consist of alternating hydrogenbonded OPV-melamine and perylene bisimide moieties are visualized. These results provide direct evidence for the possible modes of hydrogen bonding within a supramolecular co-assembly in solution. Subsequently, the optical properties of pure OPV-melamine and co-assemblies with a perylene bisimide derivative were characterized in solution. In an apolar solvent, OPV-melamine self-assembles into chiral superstructures. Disassembly into molecularly dissolved species is reversibly controlled by concentration and tempera-ture. Complementary hydrogen bonding to a perylene bisimide derivative in an apolar solvent yields multicomponent, π-stacked dye assemblies of enhanced stability that are characterized by fluorescence quenching of the constituent chromophores. Titration experiments reveal that a mixture of hydrogen-bonded oligomers is present in solution, rather than a single discrete assembly. The solution experiments are consistent with the STM results, which revealed various supramolecular assemblies. Our system is likely not to be optimally programmed to obtain a discrete co-assembled structure in quantitative yield.

Full text of DOI:10.1002/chem.200800760

FULL PAPER
DOI: 10.1002/chem.200800760
Visualization of Various Supramolecular Assemblies of
Oligo(para-phenylenevinylene)–Melamine and Perylene Bisimide
Freek J. M. Hoeben,[a] Jian Zhang,[b] Cameron C. Lee,[a] Maarten J. Pouderoijen,[a]
Martin Wolffs,[a] Frank Würthner,[c] Albertus P. H. J. Schenning,*[a] E. W. Meijer,*[a] and
Steven De Feyter*[b]
Abstract: A melamine derivative has
been covalently equipped with two oli-
go(para-phenylenevinylene)(OPV)
results provide direct evidence for the
possible modes of hydrogen bonding
within a supramolecular co-assembly in
solution. Subsequently, the optical
properties of pure OPV–melamine and
co-assemblies with a perylene bisimide
derivative were characterized in solu-
tion. In an apolar solvent, OPV–mela-
mine self-assembles into chiral super-
structures. Disassembly into molecular-
ly dissolved species is reversibly con-
trolled by concentration and tempera-
ture.
Complementary
hydrogen
bonding to a perylene bisimide deriva-
tive in an apolar solvent yields multi-
component, p-stacked dye assemblies
of enhanced stability that are charac-
terized by fluorescence quenching of
the constituent chromophores. Titra-
tion experiments reveal that a mixture
of hydrogen-bonded oligomers is pres-
ent in solution, rather than a single dis-
crete assembly. The solution experi-
ments are consistent with the STM re-
sults, which revealed various supra-
molecular assemblies. Our system is
likely not to be optimally programmed
to obtain a discrete co-assembled struc-
ture in quantitative yield.
chromophores. This procedure yields a
bifunctional molecule with two hydro-
gen-bonding arrays available for com-
plementary binding to perylene bisi-
mide derivatives. Depending on the
solvent, hydrogen-bonded trimers, tet-
ramers, and dimers on a graphite sur-
face are observed for pure OPV–mela-
mine by using scanning tunneling mi-
croscopy (STM). Upon the addition of
perylene bisimide, linear tapes of pery-
lene bisimide, 12-membered rosettes
that consist of alternating hydrogen-
bonded OPV–melamine and perylene
bisimide moieties are visualized. These
Keywords: chirality · donor–accept-
or systems
scanning probe microscopy
self-assembly
·
hydrogen bonds
·
·
Introduction
[a] Dr. F. J. M. Hoeben, Dr. C. C. Lee, M. J. Pouderoijen, M. Wolffs,
Dr. A. P. H. J. Schenning, Prof. Dr. E. W. Meijer
Laboratory of Macromolecular and Organic Chemistry Eindhoven
University of Technology
The design of functional p-conjugated structures for imple-
mentation in the field of organic electronics has been shown
to be a promising strategy.[1] Our previous work focused on
the combination of oligo(para-phenylenevinylene)(OPV)
and perylene bisimide (PBI)dyes and yielded a variety of
multicomponent supramolecular architectures based on p–p
stacking and hydrogen-bonding interactions.[2] The electron-
ic nature of the constituting OPV (p-type)and PBI (n-type)
units ensured that the resulting co-assemblies were generally
characterized by efficient electron transfer from the OPV to
the PBI species. Eventually, such an approach may result in
efficient components for solar cells by preorganizing the
system for charge separation.[3] Furthermore, the quest for
well-defined donor–acceptor co-assemblies yields valuable
noncovalent model systems for fundamental charge-transfer
studies in addition to their covalent counterparts.[4]
P.O. Box 513, 5600 MB Eindhoven (The Netherlands)
Fax : (+31)40-245-1036
E-mail: a.p.h.j.schenning@tue.nl
e.w.meijer@tue.nl
[b] Dr. J. Zhang, Dr. S. De Feyter
Laboratory of Photochemistry and Spectroscopy
Institute of Nanoscale Physics and Chemistry
Katholieke Universiteit Leuven
Celestijnenlaan 200-F, 3001 Leuven (Belgium)
E-mail: steven.defeyter@chem.kuleuven.be
[c] Prof. Dr. F. Würthner
Institut für Organische Chemie
Universität Würzburg
Am Hubland, D-97074 Würzburg (Germany)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
Chem. Eur. J. 2008, 14, 8579 – 8589
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8579

Complementary
hydrogen-
bonding interactions are ideally
suited for creating well-defined
nanostructures as a result of
their high specificity and direc-
tionality.[5] Pioneering studies
on the melamine/barbituric (cy-
anuric)acid hydrogen-bonding
couple by the groups of White-
sides and Reinhoudt revealed
the existence of beautiful struc-
tural motifs, such as linear
tapes, crinkled tapes, six-mem-
bered rosettes, and double ro-
settes.[6] The successful combi-
nation of synthetic accessibility
and directional hydrogen bond-
ing of both moieties has rapidly
Scheme 1. Synthesis of (OPV4)2M: a)diisopropylethylamine, dioxane, 100 8C, 120 h (15%). PBI is used for co-
assembly with (OPV4)2M by hydrogen bonding and p–p stacking. R*=(S)-2-methylbutyl.
transformed this molecular recognition principle into a
flourishing area of supramolecular chemistry.[7] Using a com-
bination of complementary hydrogen bonding to melamine
derivatives and p–p stacking interactions, the groups of
ACHTREmUNG esoscopic assemblies of, respectively, naphthalene bisimide
and PBI derivatives in methylcyclohexane (MCH). Recently,
Wasielewski and co-workers showed the formation of cylin-
drical nanostructures that consist of hydrogen-bonded build-
ing blocks comprised of covalently attached melamine and
ACHTREUNG
Kimizuka[8] and Würthner[9] have formed well-defined
Scheme 2. Possible structures of stoichiometric co-assemblies of (OPV4)2M and PBI. A linear tapes shown here and six-membered rosette shown over.
8580
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 8579 – 8589

Co-assembly of Oligo(para-phenylenevinylene)–Melamine and Perylene Bisimide
FULL PAPER
PBI.[10] Additionally, Champness, Beton, and co-workers re-
ported that PBI and melamine form two-dimensional hydro-
gen-bonded networks when the two molecules self-assemble
on a silicon surface.[11] The individual cells of the thus-
formed honeycomb structure were composed of 12-mem-
bered rosettes, as expected based on the geometry of the
molecules. Inspired by these hydrogen-bonded systems, the
covalent connection of OPV derivatives to a melamine unit
is an intriguing option as this behavior may yield a new type
of donor–acceptor assembly upon binding PBI derivatives.
Herein, the complexity in donor–acceptor co-assembly is
enhanced by increasing the OPV density with respect to the
systems studied previously.[2a,b] We explore the formation of
the structural motif and subsequent self-assembly behavior
at the liquid–solid interface and in solution for a novel mela-
mine-appended OPV derivative and a complementary PBI
derivative. Substitution of melamine with two OPV units re-
sults in a chromophore with two identical hydrogen-bonding
arrays. This process yields bifunctional (OPV4)2M
(Scheme 1), which can bind up to two PBI molecules,[12] thus
giving rise to extended or cyclic structures with potentially
interesting morphological and optical properties. Supra-
molecular donor–acceptor oligomers and cyclic hydrogen-
bonded structures (Scheme 2)are observed on graphite by
using scanning tunneling microscopy (STM). The ring-like
structures[13] represent well-organized, densely packed chro-
mophore ensembles. Ultimately, we show that although a
number of interesting aggregate structures are observed, it
is not possible to obtain a discrete supramolecular architec-
ture in quantitative yield either in the solid state or in solu-
tion; the system, presumably, lacks a driving force to bias it
towards the formation of a single aggregate type.[5]
Results and Discussion
Synthesis and characterization: Compound (OPV4)2M was
readily synthesized by employing the difference in reactivity
of the three electrophilic sites upon substitution of cyanuric
chloride (Scheme 1). Reaction with ammonia at À58C re-
Chem. Eur. J. 2008, 14, 8579 – 8589
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8581

A. Schenning, E. W. Meijer, S. De Feyter et al.
sulted in the formation of the monoaminated adduct[14] and
subsequent reaction with the previously reported enantio-
[15]
merically pure OPV4NH2 yielded the desired bifunctional
target compound. Pure (OPV4)2M was obtained as an
orange solid in a poor yield of 15% (as a result of the rela-
tively low reactivity of OPV4NH2)after column chromatog-
raphy and preparative size-exclusion chromatography; char-
acterization was carried out with 1H NMR, 13C NMR, and
IR spectroscopic and mass-spectrometric analysis. The syn-
thesis of tetra(tert-butylphenoxy)perylene bisimide has been
reported previously.[12]
STM analysis: As a prelude to the solution-based studies,
STM analysis at the liquid–solid interface was used to
obtain information on the structure of the possible co-as-
semblies.[16] To explore the optimum conditions under which
bicomponent hydrogen-bonded supramolecular structures
form, the 2D self-assembly of (OPV4)2M in 1-phenyloctane,
1,2,4-trichlorobenzene, and tetradecane was explored first.
No effort was made to investigate PBI at the liquid–solid in-
terface as this compound failed to form regular 2D lattices
in previous studies.[17] The above mentioned solvents were
chosen as they dissolve the compounds of interest, have low
vapor pressures (a prerequisite for measurements at the
liquid–solid interface without a closed cell), are electro-
chemically inert under the experimental conditions, and
show a low tendency toward self-assembly on surfaces at
room temperature.[18] In a typical experiment, a drop of sol-
vent containing the compound(s)of interest was applied to
a freshly cleaved surface of highly oriented pyrolytic graph-
ite (HOPG). Upon spontaneous monolayer formation, STM
images were acquired in the variable current mode by scan-
ning with the STM tip immersed in the solution.
Figure 1. STM images showing monolayers of (OPV4)2M on HOPG
physisorbed from a)1-phenyloctane (image size: 40”40 nm 2, Iset
=
0.48 nA,
V
set =À0.486 V), b) tetradecane (image size: 48.5”48.5 nm2,
I
set =0.56 nA, Vset =À0.66 ; the arrows indicate vacancies in the 2D lat-
tice), and c) 1,2,4-trichlorobenzene (image size: 45.5”45.5 nm2, Iset =0.7
nA, Vset =À1.2 V). d)–f) Tentative molecular models for the repetitive
units of the self-assembled structures observed in (a)–(c), respectively.
Interestingly, (OPV4)2M shows a range of different con-
formations and hydrogen-bonding modes at the liquid–solid
interface. Typical STM images of physisorbed monolayers in
the three solvents used are shown in Figure 1. The bright
strips correspond to the conjugated backbone of the OPVs
and indicate that the phenyl rings are adsorbed flatly on the
surface. The self-assembled monolayer is disordered at the
1-phenyloctane/HOPG interface. Cyclic structures of
dimers, trimers, and tetramers can be distinguished (Fig-
ure 1a; tetramer and trimer models in Figure 1d), but no
preferential binding modes or conformations are obvious.
Hydrogen bonding between the melamine units is likely but
the precise interaction mode can not be distinguished. Do-
mains that consist of hydrogen-bonded dimers are observed
with molecules that adopt mainly a small V-shape conforma-
tion at the tetradecane/HOPG interface (Figure 1b). The
point defects (missing V-shape structures)indicate that a V-
shape structure corresponds to one molecule (indicated by
arrows in Figure 1b).
(OPV4)2M is attributed to molecular conformational effects
that most likely occur as the result of rotation around the
À
single C N bond that links the OPV moiety to the s-triazine
ring. The domain sizes of the monolayers correlate to the
degree of monolayer order. Thus, the well-ordered monolay-
ers at the 1,2,4-trichlorobenzene/HOPG interface are the
largest and cover areas as large as 40000 nm2. The domains
are typically smaller than 2500 nm2 at the tetradecane/
HOPG interface, and disordered regions often appear be-
tween the ordered domains. The most disordered monolayer
(amorphous)is found at the 1-phenyloctane/HOPG inter-
face. In all cases, hydrogen bonding is presumably involved
in stabilizing the supramolecular structures. Because of the
disorder present at the 1-phenyloctane/HOPG interface, the
exact interaction mode between the melamine units is un-
clear. The images suggest that the melamine units interact
through two hydrogen bonds at the tetradecane/HOPG and
1,2,4-trichlorobenzene/HOPG interfaces, thus leading to the
formation of dimers (Figure 1e and Figure 1f).
Highly ordered 2D monolayers consisting of hydrogen-
bonded dimers were obtained at the 1,2,4-trichlorobenzene/
HOPG interface (Figure 1c). The molecules in the monolay-
ers adopt a big V-shape conformation with well-separated
OPV moieties. The difference in the appearance of
The above-mentioned results highlight the importance of
surface confinement and the solvent on the self-assembly
process at the liquid–solid interface. Stacking between
8582
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 8579 – 8589

Co-assembly of Oligo(para-phenylenevinylene)–Melamine and Perylene Bisimide
FULL PAPER
solute molecules at the interface is unlikely because mole-
cule–graphite p–p interactions are expected to overrule in-
termolecular p–p interactions. Furthermore, the surface
itself can exert (geometrical)constraints on the ways in
which the molecules adsorb through molecule–substrate in-
teractions. In addition, solvent–substrate interactions can
affect the way in which molecules initially pack onto the sur-
face. These aspects may be the origin of the differences in
the solvent-dependent surface-confined assemblies. The co-
assembly of (OPV4)2M and PBI at the liquid–solid interface
was also assessed by using the three solvents mentioned. To
successfully image (OPV4)2M/PBI complexes at the liquid–
solid interface, a large excess of PBI had to be used, typical-
ly in a 1:40 ratio. This ratio is in line with our previous study
on bicomponent complexes of PBI on surfaces, in which it
was observed that increasing the concentration of the
weakly adsorbing PBI led to an increase of heterocomplex
content on the surface.[17] Despite this high ratio, no multi-
component hydrogen-bonded complexes were observed at
the 1,2,4-trichlorobenzene/HOPG interface upon mixing
(OPV4)2M and PBI in various ratios. Instead, only domains
containing (OPV4)2M were observed, and the molecular or-
dering was identical to that obtained upon physisorption
from a monocomponent solution.
With 1-phenyloctane and tetradecane, however, it was
Figure 2. STM images showing co-assemblies of (OPV4)2M and PBI
physisorbed on HOPG from a)1-phenyloctane (image size: 13.0”
13.2 nm2, Iset =0.62 nA, Vset =À0.42 V), b) 1-phenyloctane (image size:
12.3”19.7 nm2, Iset =0.63 nA, Vset =À0.996 V), and c) tetradecane (image
size: 21.4”28.2 nm2, Iset =0.46 nA, Vset =À0.402 V; a unit cell is indicat-
ed). d)–f) The corresponding tentative models for the structures observed
in (a)–(c).
possible to visualize the heterocomplexes. The types of 2D
self-assembled patterns formed after deposition of a drop of
a solution of (OPV4)2M/PBI in 1-phenyloctane on HOPG
are shown in Figure 2a,b. The monolayers imaged for the
mixed assemblies are largely disordered. However, occasion-
ally 2D tape-like supramolecular oligomers (Figure 2a)and
12-membered rosettes (Figure 2b)are observed. Pairs of
bright OPV rods with V-shaped conformations are visible
with faint rod-like structures that link their vertices (Fig-
ure 2a). Although the features are faint, the rods could cor-
respond to PBI molecules as the measured lengths are in
good agreement with those predicted (1.13 nm; see the
model in Figure 2d). Under the specific imaging conditions,
cyclic structures were observed sporadically, but the full ro-
sette proved to be very unstable and quickly became partial
or disappeared. The rosettes are composed of six bright and
V-shaped (OPV4)2M molecules, whereas the PBI molecules
are not clearly visible (Figure 2b). The measured head-to-
head distance between opposite V-shaped structures in a ro-
sette is 3.6–3.7 nm, which corresponds nicely to the observed
distance in the model presented in Figure 2e. Based on this
model, the inner radius of the 12-membered supramolecular
rosette is estimated to be r=0.89 nm.
quence, only one of the two triple hydrogen-bonding sites of
each (OPV4)2M molecule is involved in hydrogen bonding
(Figure 2f).
Bias-dependent STM imaging was utilized to identify the
PBI units on the surface.[19] At high negative sample bias
(electrons tunneling from the substrate into the tip), the
OPV units appear brighter than the PBI moieties because
the tunneling process is more efficient through OPV than
through PBI, whereas at less negative bias voltages both the
OPV and PBI units are visible. Figure 3a,b are STM images
obtained sequentially at the same location, thus revealing a
tape-like structure at the 1-phenyloctane/HOPG interface.
These images differ in the value of the negative sample bias.
In Figure 3a, recorded at Vset =À1.21 V, the OPV rods
appear fuzzy and no PBI molecules are visible. In Figure 3b,
recorded at Vset =À0.914 V, V-shaped (OPV4)2M molecules
are observed with relatively bright features that appear be-
tween them, which we attribute to PBI.
Similar to the situation in 1-phenyloctane, mixtures of
(OPV4)2M and PBI in tetradecane gave rise to mixed mono-
layers. In this solvent, only half as many PBI molecules
were present in the lattice, and instead of tapes and rosettes,
trimers consisting of two (OPV4)2M molecules hydrogen
bonded to a central PBI molecule were found (Figure 2c).
The trimers are shifted with respect to each other, and as
such it is not geometrically possible for another PBI mole-
cule to link the two trimers to form a tape. As a conse-
At the appropriate bias voltage (Vset =À0.572 V), the PBI
molecules in partial rosettes were also revealed (Figure 4).
An OPV mismatch (indicated by the arrow)appeared as a
result of one of the (OPV4)2M molecules not existing in the
required V-shaped conformation. In this case, it seems that
À
rotation about the N C bond linking one of the OPV units
Chem. Eur. J. 2008, 14, 8579 – 8589
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8583

A. Schenning, E. W. Meijer, S. De Feyter et al.
Figure 3. Bias-dependent imaging of the same area (image size: 11.3”
18.7 nm2)of an alternating tape structure of (OPV4) 2M and PBI on
HOPG physisorbed from 1-phenyloctane. a) Iset =0.57 nA, Vset =À1.21 V;
b) Iset =0.57 nA, Vset =À0.914 V. The arrows in both images indicate iden-
tical sites, which correspond to a PBI molecule.
Figure 4. STM image of a partial rosette with an OPV mismatch (indi-
cated by the arrow)on HOPG physisorbed from 1-phenyloctane (image
size: 13.3”11.7 nm, Iset =0.62 nA, Vset =À0.572 V).
Figure 5. a)UV/vis, normalized PL, and CD spectra for (OPV4) 2M in
CHCl3 (1.4”10À5 m, dashed line)and MCH (7.5”10 À5 m, solid line), the
value of lexc is the respective value of lmax. b)Concentration-dependent
UV/vis spectra of (OPV4)2M in MCH at room temperature; the arrows
indicate increasing concentration. The inset shows the fraction of aggre-
gated species (aagg)as a function of concentration, determined by nor-
malizing the measured absorption at l=500 nm. Line to guide the eye.
Int.=intensity.
to the melamine core results in a conformational isomer
that is not capable of completing the rosette for geometric
reasons. This behavior shows that the conformational flexi-
bility of (OPV4)2M hampers the ability of these molecules
to form discrete rosette structures with PBI. Moreover, this
flexibility is likely to influence the self-assembly behavior in
solution as well.
and bathochromically shifted to lem =522 nm and lem
=
547 nm. These combined features are typically observed for
OPV aggregation into H-type assemblies.[21] A concentra-
tion-dependent UV/vis spectroscopic study shows that the
self-assembly of (OPV4)2M in MCH occurs at concentra-
tions higher than 10À5 m (Figure 5b, inset).
Concomitant with the observed spectral changes upon
self-assembly, a clear Cotton effect arises for (OPV4)2M
that coincides with the observed UV/vis absorption bands,
thus indicating that the chiral information stored in the side
chains directs the aggregation process into chiral architec-
tures. No linear dichroism was observed at these wave-
lengths and thus we are confident that the measured CD
signal is not the result of macroscopic alignment artifacts.[22]
Interestingly, the effect is largely negative, not bisignate as
is usually observed for self-assembled OPVs,[20,23] thus indi-
Optical properties of (OPV4)2M in solution: The optical
properties of pure (OPV4)2M were examined in CHCl3 and
MCH (Figure 5). In CHCl3 (1.4”10À5 m), the p–p* transition
of the OPV backbone is observed at lmax =436 nm (e=1.4”
105 LmolÀ1 cmÀ1), which corresponds to the position found
for previously reported similar OPV systems.[20] The com-
pound exhibits bright fluorescence at its main emission
wavelength of lem =502 nm and its first vibrational progres-
sion at lem =531 nm. Upon dissolving the molecule in MCH
at an elevated concentration (7.5”10À5 m), the absorption
maximum exhibits a hypsochromic shift to lmax =419 nm ac-
companied by the appearance of a red-shifted shoulder at
l=500 nm. Moreover, the fluorescence is strongly quenched
8584
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 8579 – 8589

Co-assembly of Oligo(para-phenylenevinylene)–Melamine and Perylene Bisimide
FULL PAPER
cating a lack of exciton coupling in the (OPV4)2M structures
formed. The (OPV4)2M moiety is similar in structure to a
previously reported OPV4 species equipped with a diamino
s-triazine hydrogen-bonding array (OPV4T),[24] the latter of
which may be regarded a mono-OPV-substituted melamine
derivative. Both (OPV4)2M and OPV4T form face-to-face
packed assemblies in MCH; however, supramolecular chiral-
ity is expressed in a different fashion for both molecules.
Obviously, the ability of the bifunctional molecule to rotate
around the bonds centered by the nitrogen atoms that con-
nect the OPV molecules to the s-triazine ring yields various
chiral conformations (also see the STM data). Moreover, it
is well known that the phenyl rings on a disubstituted mela-
mine unit can be highly twisted, as found for 1:1 co-crystals
with barbituric acid derivatives.[25]
Temperature-dependent UV/vis, photoluminescence (PL),
and circular dichroism (CD)measurements were performed
on (OPV4)2M in MCH (7.5”10À5 m; Figure 6). The spectra
exhibited fully reversible shifts upon heating that are consis-
tent with disassembly to a molecularly dissolved state. At
temperatures above about 248C, the optical properties even-
tually saturated at values similar to those in CHCl3, thus
yielding evidence that the molecules are present as dissolved
species. This behavior indicates relatively weak assemblies
relative to the systems studied previously,[20,24] which is also
revealed in the concentration-dependent measurements. The
variation of the CD spectra with increasing temperature in-
dicates that an intermediate state is present, as the shape of
the spectrum completely changes during cooling (Figure 6c,
inset).
Co-assemblies with PBI that display fluorescence quench-
ing: To investigate the feasibility of mixed-assembly forma-
tion in solution, mixing experiments were performed in
MCH (Figure 7). Studies were performed in dilute solution
(<10À5 m)to ensure complete solubility of each component
in MCH. Upon mixing of (OPV4)2M and PBI in various
Figure 6. Temperature-dependent a)UV/vis, b)PL, and c)CD spectra for
(OPV4)2M in MCH (7.5”10À5 m). The arrows indicate a temperature in-
crease from 5 to 408C. The inset in (b)shows the fraction of aggregated
species as a function of temperature, determined by normalizing the mea-
sured UV/vis (l=500 nm, squares), PL (l=522 nm, circles), and CD (l=
390 nm, triangles)data. The inset in (c)shows the CD intensity at l=
420 nm as a function of temperature. Lines to guide the eye. Abs.=ab-
sorption.
ratios, distinct shifts in the UV/vis spectra occur that are
generally characteristic for the co-assembly of OPV and PBI
chromophores into J-type assemblies (Figure 7a,b,d).[2b]
Complexation was also monitored using fluorescence spec-
troscopy (Figure 7c,f). After addition of one equivalent of
(OPV4)2M to PBI (5.0”10À6 m), PBI fluorescence at lem
=
590 nm is almost completely quenched and very little OPV
emission is observed (Figure 7c). OPV emission is observed
for (OPV4)2M/PBI ratios greater than 1:1, thus indicating
that disassembled (OPV4)2M is present in solution (pure
(OPV4)2M is molecularly dissolved at this concentration).
Similarly, in the reverse titration experiment, (OPV4)2M
fluorescence at lem =491 nm is completely quenched after
the addition of one equivalent of PBI, and very little PBI
transfer from (OPV4)2M to PBI, as observed for closely re-
lated systems.[2b] However, we were not able to distinguish
between the formation of a charge-separated state or energy
transfer to a nonemissive state from femtosecond transient
absorption measurements (data not shown).
The lack of fluorescence at 1:1 stoichiometries suggests
that strong binding occurs between (OPV4)2M and PBI.
However, if hydrogen bonding is the only intermolecular in-
teraction responsible for the assembly of the two chromo-
phores, full binding should not be achieved at micromolar
concentrations as the association constant for triple hydro-
gen bonding between melamine and PBI is at the most k=
105 mÀ1 in MCH.[9a,12] Even when the possibility of cycle for-
emission is observed when the PBI/
ACHTREUNG
(Figure 7f). PBI emission is observed at PBI/AHCTREUNG
ratios of >1:1, thus indicating that disassembled PBI is pres-
ent in solution.
Near-complete fluorescence quenching for both chromo-
phores in a 1:1 mixed assembly is an indication of electron
Chem. Eur. J. 2008, 14, 8579 – 8589
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8585

A. Schenning, E. W. Meijer, S. De Feyter et al.
ings, plus a lack of clear iso-
sbestic points, indicate that the
structure of the most abundant
PBI-containing species in solu-
tion changes more than once
while varying the (OPV4)2M/
PBI ratio, thus suggesting that a
single favored oligomeric as-
sembly (e.g., a 12-membered
[27]
rosette)does not exist.
A re-
verse titration of (OPV4)2M
with PBI provided similar re-
sults (Figure 7d). Additionally,
when monitoring the absorb-
ance at a single wavelength,
curves with single or multiple
kinks are obtained, and the
PBI/ACHTRE(UNG OPV4)2M ratio at which a
kink occurs depends on the
chosen wavelength. For exam-
ple (Figure 7e), when titrating
(OPV4)2M with PBI, the OPV
absorbance
at
l=443 nm
(mainly assembled OPV wave-
length)displays a kink near
0.5 equivalents of PBI. When
monitoring the PBI absorbance
at l=591 nm (assembled PBI
wavelength), however, a kink is
found near 1.0 equivalents of
PBI. This evidence is further
support for the presence of
multiple-assembly stoichiome-
tries and demonstrates that
each chromophore shows a dif-
ferent sensitivity to changes in
the aggregate structure.
Figure 7. a,b)UV/vis and c)PL ( lexc =350 nm)data showing the spectral changes upon addition of 0–2 equiva-
lents of (OPV4)2M to a 5.0”10À6 m solution of PBI in MCH at room temperature; b)an enlargement of (a).
d,e)UV/vis and f)PL ( lexc =350 nm)data of the spectral changes upon addition of 0–2 equivalents of PBI to a
5.1”10À6 m solution of (OPV4)2M in MCH at room temperature; e)how the absorbances at l=443 and
591 nm vary during the titration. The insets of (a)and (d)show how lmax of PBI varies. Quenching of PBI
(lexc =535 nm)and (OPV4) 2M (lexc =350 nm)fluorescence is shown in the insets of (c)and (f,) respectively;
the dotted line in each inset demonstrates that the quenching curves are concave upward.
Proof of the lack of a single
self-assembled product is pro-
vided by the fluorescence titra-
tion experiments as well. When
PBI is titrated with (OPV4)2M,
mation is considered,[26] rough calculations predict that there
should be significant quantities of disassembled (and thus
the PBI emission intensity versus (OPV4)2M/PBI ratio
curve is concave upward rather than linear, thus showing
that PBI fluorescence quenching occurs faster than one
would expect for the quantitative formation of only a 1:1 ag-
gregate (Figure 7c, inset). When (OPV4)2M is titrated with
PBI, again OPV fluorescence decreases faster than one
would expect if a single 1:1 aggregate was being formed in a
quantitative manner (Figure 7f, inset). If it is assumed that
PBI or (OPV4)2M must be intimately incorporated into a
co-assembly to have its fluorescence quenched, then these
pieces of evidence strongly suggest that complex stoichiome-
tries other than 1:1 are obtained with this system.
fluorescent)PBI and (OPV4) M remaining in solution. The
2
stronger than anticipated binding between (OPV4)2M and
PBI implies that in addition to hydrogen bonding other in-
teractions, such as p–p stacking and solvophobic effects,
contribute to the overall stability of the co-assembly in
MCH. A combination of hydrogen bonding and p–p stack-
ing is consistent with the UV/vis spectral changes, and simi-
lar observations have been made in related systems.[2b,12]
Interestingly, the position of the maximum of the PBI ab-
sorption changes more than once during the titration (Fig-
ure 7a, inset). For example, lmax =564 nm when (OPV4)2M/
PBI is 0, lmax =591 nm when (OPV4)2M/PBI is 0.75–1.25:1,
and lmax =602 nm when (OPV4)2M/PBI is 2:1. These find-
The aggregation behavior was further studied in tempera-
ture-dependent optical studies performed on a 1:1 mixture
of (OPV4)2M and PBI (1.4”10À5 m; Figure 8). At high tem-
8586
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 8579 – 8589

Co-assembly of Oligo(para-phenylenevinylene)–Melamine and Perylene Bisimide
FULL PAPER
peratures, the spectra display features expected for molecu-
larly dissolved (OPV4)2M and PBI. Therefore, the UV/vis
spectrum is a summation of the contributions from both
compounds. Upon self-assembly at lower temperatures,
(OPV4)2M/PBI hydrogen-bonded oligomers there is a left-
handed, helical stacking of the OPV moieties. Also, PBI
seems to exhibit a similar bisignate Cotton effect with a
zero-crossing at l=597 nm. However, the weakness of the
measured signal in combination with the significant baseline
drift with increasing temperature troubles a conclusive inter-
pretation.
From the temperature-dependent UV/vis spectra, the
transition from co-assemblies to molecularly dissolved spe-
cies is detected at around 668C (1.4”10À5 m; Figure 8a), thus
indicating enhanced stability with respect to previously stud-
ied systems.[2b] This finding was further corroborated by a
concentration-dependent UV/vis study on the 1:1
(OPV4)2M/PBI complex, in which the system showed little
change in aggregated character when diluted from 10À4 m to
2.5”10À7 m.
clear bathochromic shifts of the OPV absorption from lmax
=
424 to 438 nm and the PBI absorption from lmax =559 to
595 nm indicate co-assembly of the building blocks into a
mixed J-type aggregate (Figure 8a).[2b] Concomitantly, the
fluorescence of (OPV4)2M and PBI is completely quenched,
both upon predominant OPV excitation at lexc =417 nm and
selective PBI excitation at lexc =568 nm (Figure 8b). Surpris-
ingly, a bisignate Cotton effect with a zero-crossing at l=
429 nm is recovered for the OPV chromophores upon co-as-
sembly (Figure 8c). This behavior indicates that in the
The combination of hydrogen bonding and p–p stacking
in the aggregates complicates the determination of a clear
molecular picture. A mechanism consistent with our obser-
vations is proposed: In the absence of stacking, based on
our STM results, we expect an equilibrating mixture of mon-
omers and linear/cyclic oligomers of varying lengths. The
main demands for the specific formation of cyclic hydrogen-
bonded structures, namely, high binding constant and highly
preorganized monomeric subunits, are both not fulfilled. We
propose that steric effects that select between cyclization
and linear polymerization (although the mechanistic inter-
pretations differ)[26,28] will not play a major role in the
(OPV4)2M/PBI system as a consequence of the large dis-
tance that the PBI unit spans. In addition, the multiple con-
formational stereoisomers possible for (OPV4)2M (which
À
result from rotation around the N C bonds that link the
OPV units to the s-triazine ring)and PBI (which result from
interconversion between the two axially chiral forms of the
highly substituted, and thus nonplanar, PBI derivative)[2b]
may hinder cyclization and favor the formation of linear
structures when the appropriate combination of geometries
for the interacting subunits is not attained. Ultimately, we
propose that these conditions give rise to an oligomer mix-
ture of a more or less statistical nature, which is expressed
both in solution (in which the oligomers are p stacked)and
on HOPG (in which the oligomers are absorbed to the sur-
face). In principle, a multitude of different structures can be
envisaged (Scheme 3).
Conclusion
Co-assembly of an OPV–melamine moiety with a perylene
bisimide derivative has been investigated. Bias-dependent
STM studies reveal various assembly modes at the 1-phenyl-
octane/HOPG interface, which result in the formation of oli-
gomeric tapes, 12-membered rosettes, and partial rosettes
consisting of alternating hydrogen-bonded chromophores.
Additionally, optical techniques demonstrate that OPV–mel-
amine forms stable self-assembled heterocomplexes with
perylene bisimide in solution. Titration experiments suggest
Figure 8. Temperature-dependent a)UV/vis, b)PL ( lexc =417 nm), and
c)CD spectra of a 1:1 mixture of (OPV4) 2M with PBI in MCH (1.4”
10À5 m). The arrows indicate a temperature increase from 10 to 908C. The
inset in (b)shows PBI fluorescence when selectively excited at lexc
=
568 nm.
Chem. Eur. J. 2008, 14, 8579 – 8589
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8587

A. Schenning, E. W. Meijer, S. De Feyter et al.
[15]
amino-4,6-dichloro-1,3,5-triazine,[14] and OPV4NH2 were prepared ac-
cording to reported procedures.
Synthesis of (OPV4)2M: 2-Amino-4,6-dichloro-1,3,5-triazine (47 mg,
0.29 mmol)was dissolved in freshly distilled dioxane (2 mL)and the solu-
tion was heated to 408C. Half the amount of a solution of OPV4NH2
(768 mg, 0.59 mmol, 2.1 equiv)and diisopropylethylamine (78 mg,
0.61 mmol, 2.1 equiv)in freshly distilled dioxane (16 mL)warmed to
608C was added dropwise. After stirring at 408C for 30 min, the other
half of the OPV4NH2 solution was added dropwise, and the reaction mix-
ture was heated to reflux for 120 h. After cooling down to room tempera-
ture, the product was precipitated in MeOH and the orange precipitate
was further purified using column chromatography on silica gel (eluent
gradient: CH2Cl2 to 10% ethanol in CH2Cl2). Final purification using
Scheme 3. a)Monomeric building blocks PBI (far left)and (OPV4) 2M in
three of its possible conformational isomers; b)hydrogen bonding in 1:2
and 2:1 complexes; c)stacking possibilities for 1:2 and 2:1 complexes;
d)a 12-membered hydrogen-bonded rosette; e)a linear hydrogen-
bonded tape; f)stacked linear hydrogen-bonded tapes; g)stacked 12-
membered hydrogen-bonded rosettes. According to the spectroscopic evi-
dence, all stacked structures should, in principle, be J-type assemblies.
AHCTREUNG
C
ACHTREUNG
AHCTREUNG
CTHREUNG
A
ACHTREUNG
A
ACHTREUNG
AHCTREUNG
AHCTREUNG
that multiple aggregate stoichiometries exist for the mixed
system that depend on the ratio of OPV–melamine to pery-
lene bisimide. The origin of the disorder exhibited by the
mixed system, which is expressed both on HOPG and in so-
lution, most likely stems (amongst other factors)from the
inherent conformational flexibility and lack of preorganiza-
tion contained in the OPV and perylene building blocks.
This behavior results in an insufficient energetic preference
for one specific self-assembly mode amongst the possible
cyclic and linear supramolecular motifs. It was not possible
to obtain a specific discrete co-assembly in quantitative
yield by adjusting common variables, such as solvent, tem-
perature, or concentrations. Our results show that for the
construction of functional multicomponent systems preorga-
nization of the building blocks become increasingly impor-
tant when using common self-assembly tools.
CH), 7.13 (s, 2H, ArH), 7.15 (s, 2H, ArH), 7.16 (d, 2H, CH=CH), 7.21
(s, 4H, CH=CH, CH=CH), 7.28 (br, 2H, NH), 7.43 (d, 2H, CH=CH),
7.46 (d, 2H, CH=CH), 7.52 (d, 4H, ArH), 7.55 (s, 4H, ArH), 7.62 ppm
(d, 4H, ArH); 13C NMR (CDCl3): d=11.66, 11.73, 11.8, 14.4, 17.0, 17.11,
17.15, 22.9, 26.4, 26.62, 26.65, 29.62, 29.65, 29.70, 29.92, 29.97, 29.99,
30.01, 30.6, 32.18, 32.20, 35.2, 35.3, 35.4, 69.3, 73.8, 74.3, 74.4, 74.7, 74.8,
105.3, 109.8, 110.1, 110.7, 110.9, 120.9, 122.6, 122.8, 122.9, 127.0, 127.1,
127.2, 127.66, 127.72, 128.3, 128.8, 133.51, 133.53, 138.1, 138.4, 151.28,
151.32, 151.4, 151.5, 153.5, 164.9, 167.3 ppm; IR (UATR): n˜ =3406, 3320,
3174, 3061, 2957, 2922, 2875, 2853, 1598, 1571, 1497, 1464, 1420, 1386,
1340, 1314, 1302, 1241, 1201, 1153, 1116, 1045, 1010, 962, 916, 852, 808,
773, 722, 698 cmÀ1; MALDI-TOF MS (Mr =2685.10): m/z 2686.02 [M]+.
STM analysis: All the STM experiments were performed at room tem-
perature under ambient conditions. The STM images were obtained at
the liquid–solid interface with a Discoverer scanning tunneling micro-
scope (Topometrix Inc., Santa Barbara, CA)along with an external
pulse/function generator (model HP 8111 A). The STM tips were electro-
chemically etched from Pt/Ir wire (80:20; diameter: 0.2 nm)in a solution
of 2n KOH/6n NaCN in water. Prior to imaging, (OPV4)2M and PBI
were dissolved in 1-phenyloctane (Aldrich)in concentrations of approxi-
mately 0.005 and 0.08 mgmLÀ1, respectively. A drop of the solution was
applied onto a freshly cleaved surface of HOPG (grade ZYB; Advanced
Ceramics Inc., Cleveland, OH). The STM images were acquired in the
variable current mode by scanning the STM tip immersed in solution at a
negative sample bias (electrons tunnel from the sample to the tip). The
measured tunneling currents are converted into a gray scale: black
(white)refers to a low (high)measured tunneling current. After the suc-
cessful imaging of the monolayer, an atomically resolved image of the
graphite substrate was recorded at exactly the same location with identi-
cal scanning parameters except for the sample bias. The images were cor-
rected for scanner drift by using SPIP software (Image Metrology ApS)
with graphite as the calibration grid. Only images containing a small drift
were used for analysis.
Titration experiments: Stock solutions of (OPV4)2M (5.1”10À6 m)and
PBI (5.0”10À6 m)were prepared in spectrophotometric grade methylcy-
clohexane (MCH; heating was required to ensure full dissolution of the
compounds in this solvent). Stock solutions of (OPV4)2M (7.0”10À5 m)
and PBI (6.8”10À5 m)were also prepared in dichloromethane. PBI was ti-
trated with (OPV4)2M as follows: the (OPV4)2M stock solution in di-
chloromethane (0–72.5 mL)was added to a silanized vial (Supelco)with a
syringe, and the solvent was then evaporated under a stream of nitrogen.
The PBI stock solution in MCH (0.5 mL)was added by syringe to the
(OPV4)2M residue. The vial was capped and shaken at room temperature
in the dark for 12 h. (OPV4)2M was titrated with PBI using the same
method. UV/vis and photoluminescence spectra were recorded for the
solutions in a 2-mm quartz cuvette at room temperature. The PL meas-
Experimental Section
General: 1H and 13C NMR spectra were recorded by using a 400 MHz
NMR (Varian Mercury, 400 MHz for 1H NMR and 100 MHz for
13C NMR)or a 300 MHz NMR (Varian Gemini, 300 MHz for 1H NMR
and 75 MHz for 13C NMR)spectrometer. In the 1H and 13C NMR spectra,
chemical shifts are reported in ppm downfield from tetramethylsilane
(TMS). IR spectra were recorded by using a Perkin–Elmer 1600 FT-IR.
Matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF)
mass-spectrometric analysis was performed by using a PerSeptive Biosys-
tems Voyager-DE PRO spectrometer. Elemental analysis was carried out
by using a Perkin–Elmer 2400. UV/vis spectra were recorded by using a
Perkin–Elmer Lambda 40 spectrometer. Generally, fluorescence spectra
were recorded by using a Perkin–Elmer LS50B luminescence spectrome-
ter, but the fluorescence spectra in the titration experiments were record-
ed by using an Edinburgh Instruments luminescence spectrometer. CD
spectra were recorded by using a JASCO J-600 spectropolarimeter (sensi-
tivity, time constant, and scan-rate were chosen appropriately).
Materials: All the solvents were of analytical reagent quality unless speci-
fied further. Dichloromethane was freshly distilled over potassium/
sodium, dioxane was freshly distilled over LiAlH4, and N,N-dimethylfor-
mamide (DMF)was dried over 4 Š molecular sieves. Reagents were pur-
chased from Acros or Aldrich and were used without further purification.
Bio-Beads SX-1 were obtained from Bio-Rad Laboratories. PBI,[12] 2-
8588
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 8579 – 8589

Co-assembly of Oligo(para-phenylenevinylene)–Melamine and Perylene Bisimide
FULL PAPER
urements were collected in the front-face mode with excitation occurring
at either lexc =350 nm (for OPV and PBI excitation)or lexc =535 nm (for
PBI excitation).
Chem. 2008, 120, 3415–3419; Angew. Chem. Int. Ed. 2008, 47, 3367–
3371.
[8] N. Kimizuka, T. Kawasaki, K. Hirata, T. Kunitake, J. Am. Chem.
Soc. 1995, 117, 6360–6361.
[9] a)F. Würthner, C. Thalacker, A. Sautter, Adv. Mater. 1999, 11, 754–
758; b)C. Thalacker, F. Würthner, Adv. Funct. Mater. 2002, 12, 209–
218.
[10] L. E. Sinks, B. Rybtchinski, M. Iimura, B. A. Jones, A. J. Goshe,
X. B. Zuo, D. M. Tiede, X. Y. Li, M. R. Wasielewski, Chem. Mater.
2005, 17, 6295–6303.
[11] J. A. Theobald, N. S. Oxtoby, M. A. Phillips, N. R. Champness, P. H.
Beton, Nature 2003, 424, 1029–1031.
[12] F. Würthner, C. Thalacker, A. Sautter, W. Schartl, W. Ibach, O. Holl-
richer, Chem. Eur. J. 2000, 6, 3871–3886.
[13] a)W. Kühlbrandt, Nature 1995, 374, 497–498; b)G. McDermott,
S. M. Prince, A. A. Freer, A. M. Hawthornthwaite-Lawless, M. Z.
Papiz, R. J. Cogdell, N. W. Isaacs, Nature 1995, 374, 517–521.
[14] a)O. Diels, Ber. Dtsch. Chem. Ges. 1899, 32, 691–702; b)H. S.
Mosher, F. C. Whitmore, J. Am. Chem. Soc. 1945, 67, 662–664.
[15] E. Peeters, P. A. van Hal, S. C. J. Meskers, R. A. J. Janssen, E. W.
Meijer, Chem. Eur. J. 2002, 8, 4470–4474.
[16] a)A. Marchenko, N. Katsonis, D. Fichou, C. Aubert, M. Malacria, J.
Am. Chem. Soc. 2002, 124, 9998–9999; b)J. W. Zhao, K. Uosaki,
NanoLett. 2002, 2, 137–140; c)K. E. Plass, A. J. Matzger, Chem.
Commun. 2006, 3486–3488.
Acknowledgement
The authors from Eindhoven thank the Council of Chemical Sciences of
the Netherlands Organization for Scientific Research (NWO-CW). The
authors from Leuven thank the Federal Science Policy through IUAP-V-
03 and the Institute for the Promotion of Innovation by Science and
Technology in Flanders (IWT). Support from the Fund for Scientific Re-
search-Flanders (FWO)and K.U. Leuven is also appreciated. Dr.
ACHTREUNGDaniela Popescu and Dr. Philippe Leclꢁre are acknowledged for the
SEM and AFM measurements, respectively.
[1] a)F. J. M. Hoeben, P. Jonkheijm, E. W. Meijer, A. P. H. J. Schenning,
Chem. Rev. 2005, 105, 1491–1546; b)P. Jonkheijm, P. P. A. M.
van der Schoot, A. P. H. J. Schenning, E. W. Meijer, Science 2006,
313, 80–83; c)P. Jonkheijm, P. J. K. J. van Duren, M. Kemerink,
R. A. J. Janssen, A. P. H. J. Schenning, E. W. Meijer, Macromolecules
2006, 39, 784–788.
[2] a)A. P. H. J. Schenning, J. van Herrikhuyzen, P. Jonkheijm, Z. Chen,
F. Würthner, E. W. Meijer, J. Am. Chem. Soc. 2002, 124, 10252–
10253; b)F. Würthner, Z. Chen, F. J. M. Hoeben, P. Osswald, C. C.
You, P. Jonkheijm, J. van Herrikhuyzen, A. P. H. J. Schenning,
P. P. A. M. van der Schoot, E. W. Meijer, E. H. A. Beckers, S. C. J.
Meskers, R. A. J. Janssen, J. Am. Chem. Soc. 2004, 126, 10611–
10618; c)E. H. A. Beckers, P. Jonkheijm, A. P. H. J. Schenning,
S. C. J. Meskers, R. A. J. Janssen, ChemPhysChem 2005, 6, 2029–
2031; d)J. Zhang, F. J. M. Hoeben, M. J. Pouderoijen, A. P. H. J.
Schenning, E. W. Meijer, F. C. De Schryver, S. De Feyter, Chem.
Eur. J. 2006, 12, 9046–9055.
[17] S. De Feyter, A. Miura, S. Yao, Z. Chen, F. Würthner, P. Jonkheijm,
A. P. H. J. Schenning, E. W. Meijer, F. C. De Schryver, NanoLett.
2005, 5, 77–81.
[18] a)B. Venkataraman, J. J. Breen, G. W. Flynn, J. Phys. Chem. 1995,
99, 6608–6619; b)C. J. Li, Q. D. Zeng, C. Wang, L. J. Wan, S. L. Xu,
C. R. Wang, C. L. Bai, J. Phys. Chem. B 2003, 107, 747–750; c)M.
Lackinger, S. Griessl, W. A. Heckl, M. Hietschold, G. W. Flynn,
Langmuir 2005, 21, 4984–4988; d)W. Mamdouh, H. Uji-i, J. S. Ladi-
slaw, A. E. Dulcey, V. Percec, F. C. De Schryver, S. De Feyter, J.
Am. Chem. Soc. 2006, 128, 317–325; e)L. Kampschulte, M. Lack-
inger, A. K. Maier, R. S. K. Kishore, S. Griessl, M. Schmittel, W. M.
Heckl, J. Phys. Chem. B 2006, 110, 10829–10836; f)X. Shao, X. C.
Luo, X. Q. Hu, K. Wu, J. Phys. Chem. B 2006, 110, 1288–1293.
[19] A. Miura, Z. Chen, H. Uji-i, S. De Feyter, M. Zdanowska, P. Jonk-
heijm, A. P. H. J. Schenning, E. W. Meijer, F. Würthner, F. C. De
Schryver, J. Am. Chem. Soc. 2003, 125, 14968–14969.
[20] P. Jonkheijm, F. J. M. Hoeben, R. Kleppinger, J. van Herrikhuyzen,
A. P. H. J. Schenning, E. W. Meijer, J. Am. Chem. Soc. 2003, 125,
15941–15949.
[21] E. Peeters, A. M. Ramos, S. C. J. Meskers, R. A. J. Janssen, J. Chem.
Phys. 2000, 112, 9445–9454.
[22] M. Wolffs, S. J. George, Z. Tomovic, S. C. J. Meskers, A. P. H. J.
Schenning, E. W. Meijer, Angew. Chem. 2007, 119, 8351–8353;
Angew. Chem. Int. Ed. 2007, 46, 8203–8205.
[23] E. H. A. Beckers, Z. Chen, S. C. J. Meskers, P. Jonkheijm, A. P. H. J.
Schenning, X. Q. Li, P. Osswald, F. Würthner, R. A. J. Janssen, J.
Phys. Chem. B 2006, 110, 16967–16978.
[3] C. H. Huang, N. D. McClenaghan, A. Kuhn, J. W. Hofstraat, D. M.
Bassani, Org. Lett. 2005, 7, 3409–3412.
[4] a)E. H. A. Beckers, S. C. J. Meskers, A. P. H. J. Schenning, Z. Chen,
F. Würthner, R. A. J. Janssen, J. Phys. Chem. A 2004, 108, 6933–
6937; b)E. H. A. Beckers, S. C. J. Meskers, A. P. H. J. Schenning, Z.
Chen, F. Würthner, P. Marsal, D. Beljonne, J. Cornil, R. A. J. Jans-
sen, J. Am. Chem. Soc. 2006, 128, 649–657.
[5] L. J. Prins, D. N. Reinhoudt, P. Timmerman, Angew. Chem. 2001,
113, 2446–2492; Angew. Chem. Int. Ed. 2001, 40, 2382–2426.
[6] a)J. A. Zerkowski, C. T. Seto, G. M. Whitesides, J. Am. Chem. Soc.
1992, 114, 5473–5475; b)G. M. Whitesides, E. E. Simanek, J. P. Ma-
thias, C. T. Seto, D. N. Chin, M. Mammen, D. M. Gordon, Acc.
Chem. Res. 1995, 28, 37–44; c)R. H. Vreekamp, J. P. M. van Duyn-
hoven, M. Hubert, W. Verboom, D. N. Reinhoudt, Angew. Chem.
1996, 108, 1306–1309; Angew. Chem. Int. Ed. Engl. 1996, 35, 1215–
1218.
[7] a)K. Hanabusa, T. Miki, Y. Taguchi, T. Koyama, H. Shirai, J. Chem.
Soc. Chem. Commun. 1993, 1382–1384; b)H. Koyano, P. Bissel, K.
Yoshihara, K. Ariga, T. Kunitake, Langmuir 1997, 13, 5426–5432;
c)E. A. Archer, A. E. Sochia, M. J. Krische, Chem. Eur. J. 2001, 7,
2059–2065; d)P. Timmerman, L. J. Prins, Eur. J. Org. Chem. 2001,
3191–3205; e)P. W. Zhu, H. Kang, A. Facchetti, G. Evmenenko, P.
Dutta, T. J. Marks, J. Am. Chem. Soc. 2003, 125, 11496–11497;
f)N. D. McClenaghan, C. Absalon, D. M. Bassani, J. Am. Chem.
Soc. 2003, 125, 13004–13005; g)S. Keinan, M. A. Ratner, T. J.
Marks, Chem. Mater. 2004, 16, 1848–1854; h)S. Yagai, M. Higashi,
T. Karatsu, A. Kitamura, Chem. Mater. 2004, 16, 3582–3585; i)S.
Yagai, T. Kinoshita, M. Higashi, K. Kishikawa, T. Nakanishi, T. Kar-
atsu, A. Kitamura, J. Am. Chem. Soc. 2007, 129, 13277; j)T. Seki, S.
Yagai, T. Karatsu, A. Kitamura, J. Org. Chem. 2008, 39, 3328–3335;
k)S. Yagai, T. Seki, T. Karatsu, A. Kitamura, F. Würthner, Angew.
[24] 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.
[25] a)J. A. Zerkowski, C. T. Seto, D. A. Wierda, G. M. Whitesides, J.
Am. Chem. Soc. 1990, 112, 9025–9026; b)J. A. Zerkowski, J. C.
MacDonald, G. M. Whitesides, Chem. Mater. 1997, 9, 1933–1941.
[26] A. G. Bielejewska, C. E. Marjo, L. J. Prins, P. Timmerman, F.
de Jong, D. N. Reinhoudt, J. Am. Chem. Soc. 2001, 123, 7518–7533.
[27] a)C. T. Seto, G. M. Whitesides, J. Am. Chem. Soc. 1993, 115, 905–
916; b)S. Yagai, T. Nakajima, T. Karatsu, K. Saitow, A. Kitamura, J.
Am. Chem. Soc. 2004, 126, 11500–11508.
[28] J. P. Mathias, E. E. Simanek, G. M. Whitesides, J. Am. Chem. Soc.
1994, 116, 4326–4340.
Received: April 22, 2008
Published online: August 1, 2008
Chem. Eur. J. 2008, 14, 8579 – 8589
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
8589