Synthesis and properties of a triphenylene-butadiynylene macrocycle

Article abstract of DOI:10.1039/c0jm02150d

The synthesis and characterization of a shape-persistent triphenylene-butadiynylene macrocycle formed by intermolecular Glaser-coupling of two "half-rings" and also by intramolecular coupling of the appropriate open dimer, respectively, are described in detail. The investigation of the photophysics has revealed that - compared to its open dimer - the macrocycle is more conjugated in the ground state and less so in the excited state, a result of the diacetylene bending in the macrocycle due to its constrained topology. The macrocycle is decorated with flexible side groups that support its adsorption on highly oriented pyrolytic graphite (HOPG) where a concentration-dependence of the 2D-structure is observed by means of scanning tunnelling microscopy (STM). The flexible side groups also guarantee a high compound solubility even in nonpolar solvents (cyclohexane). However, solvophobic interactions lead to the formation of a tube-like superstructure, as revealed by dynamic light scattering, X-ray scattering and atomic force microscopy.

Full text of DOI:10.1039/c0jm02150d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Synthesis and properties of a triphenylene–butadiynylene macrocycle†‡
a
a
b
b
Henning Wettach, Sigurd Hoger, Debangshu Chaudhuri, John. M. Lupton, Feng Liu,c
*
*
€
Elizabeth M. Lupton, Sergei Tretiak, Guojie Wang, Min Li, Steven De Feyter, Steffen Fischerf
€
and Stephan Forster
c
d
e
e
e
*
f
*
*
Received 6th July 2010, Accepted 23rd August 2010
DOI: 10.1039/c0jm02150d
The synthesis and characterization of a shape-persistent triphenylene–butadiynylene macrocycle
formed by intermolecular Glaser-coupling of two ‘‘half-rings’’ and also by intramolecular coupling of
the appropriate open dimer, respectively, are described in detail. The investigation of the photophysics
has revealed that—compared to its open dimer—the macrocycle is more conjugated in the ground state
and less so in the excited state, a result of the diacetylene bending in the macrocycle due to its
constrained topology. The macrocycle is decorated with flexible side groups that support its adsorption
on highly oriented pyrolytic graphite (HOPG) where a concentration-dependence of the 2D-structure is
observed by means of scanning tunnelling microscopy (STM). The flexible side groups also guarantee
a high compound solubility even in nonpolar solvents (cyclohexane). However, solvophobic
interactions lead to the formation of a tube-like superstructure, as revealed by dynamic light scattering,
X-ray scattering and atomic force microscopy.
liquid crystallinity (LC) when properly decorated with flexible
Introduction
alkyl chains.3,4 In solution, they can form dimers as well as
extended one-dimensional aggregates depending on the elec-
tronic nature of the substituents and, even more important, on
the solvent. Some of these macrocycle aggregates are also stable
in the solid state and have been analyzed by electron microscopy
(EM) and atomic force microscopy (AFM).5 In addition, the
precise control over photophysical properties in combination
with the generic ability of macrocycles and thus aggregates to
accommodate and recognize guest molecules makes them
attractive building blocks for optically active chemical sensors.6
Guest recognition is not limited to one-dimensional superstruc-
tures. Shape-persistent macrocycles can also build up
two-dimensional lattices7 at the air–water or the air–solid and
liquid–solid interface and these assemblates can bind specific
guest molecules.8
Advances in molecular electronics necessitate ever more precise
control over molecular shape to yield specific physical functions.
Such functionality is required both on the intramolecular level,
which governs electronic properties, and with regard to physical
arrangement in the ensemble. Shape-persistent macrocycles with
extended p-electron conjugation are appealing systems to study
this interplay between form and function: the size of the conju-
gated system can be scaled with respect to the overall molecular
size, while maintaining shape; and, as in dendritic systems,
solubility can be tuned independently of intramolecular elec-
tronic structure, ultimately allowing the construction of complex
covalently bonded ‘‘supramolecular’’-like structures.1 In addi-
tion, such shape-persistent structures exhibit a high degree of
isomeric purity, facilitating self-assembly to molecular super-
structures in the melt and in the solid state, in solution and on
surfaces.2 Rigid macrocycles with nanometre dimensions exhibit
Recently, it has also become of interest to incorporate poly-
cyclic aromatic hydrocarbon (PAH) building blocks into the
rigid macrocycle backbone because of their interesting optical
properties and assembly behavior.9 During our own studies we
could show that macrocycles with dibenzonaphthacenes in their
rigid backbone show a higher tendency to aggregate than their
benzene-based analogs and form more stable LC phases.4b,c
However, the dibenzonaphthacenes capable of undergoing
transition metal-catalyzed coupling reactions necessary to build
up the macrocycles could only be obtained in a multi-step
synthesis which involves itself a Pd-catalyzed dehydrohalogena-
tion. Therefore, a methoxy to triflate transformation was
necessary that extended the overall reaction sequence.10 In
contrast, some (halogen) substituted PAHs, and here especially
triphenylenes, are more facile to prepare either by oxidative
coupling or by a Diels–Alder approach. Both reactions do not
interfere with the presence of bromides or iodides.11 Especially
the latter reaction offers appealing building blocks for larger
condensed-phase structures of well-defined electronically active
units.12
a
ꢀ
€
Kekule-Institut fur Organische Chemie und Biochemie, Rheinische
Friedrich-Wilhelms-Universita€t Bonn, Gerhard-Domagk-Str. 1, 53121
Bonn, Germany. E-mail: hoeger@uni-bonn.de
bDepartment of Physics and Astronomy, University of Utah, Salt Lake
City, UT, 84112, USA. E-mail: lupton@physics.utah.edu
cDepartment of Materials Science, University of Utah, Salt Lake City, UT,
84112, USA. E-mail: lupton@eng.utah.edu
dTheoretical Division and Center for Integrated Nanotechnologies
(CINT), Los Alamos National Laboratory (LANL), Los Alamos, NM,
87545, USA
eDepartment of Chemistry, Laboratory of Photochemistry and
Spectroscopy, and Institute for Nanoscale Physics and Chemistry,
Katholieke Universiteit Leuven, Celestijnenlaan 200-F, 3001 Leuven,
Belgium. E-mail: Steven.DeFeyter@chem.kuleuven.be
f
€
€
Universitat Hamburg, Institut fur Physikalische Chemie, Grindelallee 117,
20146 Hamburg, Germany. E-mail: forster@chemie.uni-hamburg.de
† Electronic supplementary information (ESI) available: Detailed
synthesis and characterization of all compounds. Additional ground
state geometry calculations. See DOI: 10.1039/c0jm02150d
‡ This paper is part of a Journal of Materials Chemistry themed issue in
celebration of the 70th birthday of Professor Fred Wudl.
1404 | J. Mater. Chem., 2011, 21, 1404–1415
This journal is ª The Royal Society of Chemistry 2011

View Article Online
Here, we explore the physical and electronic structure of
a family of phenyl-substituted triphenylenes.13 We present the
synthesis and detailed photophysical, quantum chemical, and
structural investigations of macrocycle 1 in solution, at the solid–
liquid interface and in the bulk. We note that this compound is,
effectively, the phenyl-substituted triphenylene analog to Eglin-
ton’s dibenzodehydro[12]annulene 2, which was reported
50 years ago, offering an intriguing paradigm of scalability in
shape-persistent macrocycle synthesis.14–16
(6) gives 7,10-dibromo-2,3-bis-(4-methoxy-phenyl)-1,4-diphenyl-
triphenylene (7). In order to obtain sufficient product solubility,
7 was demethylated and the resulting bis-phenol 8 alkylated with
9 to form the triphenylene 10. Pd-catalyzed twofold Sonoga-
shira–Hagihara coupling reaction with [(3-cyanopropyl)diiso-
propylsilyl]acetylene (CPDIPSA)18 gave 11. The CPDIPS
protecting group exhibits a comparable stability to the triiso-
propylsilyl (TIPS) protecting group. However, the cyanopropyl
substituent enhances the polarity in such a way that the
separation of 11 from the educt 10 and from only partly
reacted side products can be performed rather easily. In
addition to that the high polarity of the CPDIPS group opens
the door for the stepwise synthesis of the desired macrocycle 1.
Complete deprotection of the acetylene groups was achieved by
treating 11 with tetrabutylammonium fluoride (TBAF) to give
12.
Compound 12 was oxidatively coupled utilizing Glaser–
Eglinton conditions (CuCl, CuCl2, and pyridine) under pseudo
high-dilution conditions (statistical approach) (Scheme 2a).19
This was performed in such a way that a solution of the bisace-
tylene 12 in pyridine was slowly (96 h) added to a suspension of
the copper catalyst/oxidant in the same solvent. The analysis of
the crude reaction mixture was performed by analytical gel
permeation chromatography (GPC) indicating that the (cyclic)
dimer was the main product of the coupling reaction. It is
interesting to see that here also the cyclodimer is the main
product of the reaction, as it is in the case of the 1,2-diethy-
nylbenzene coupling.14 According to GPC analysis (Fig. 1)
a small amount of the trimer was also formed along with some
oligomeric and polymeric by-products. The removal of the
polymeric fraction can be achieved by column chromatography
on silica gel. However, a separation of cyclodimer and cyclo-
trimer was not possible by this procedure but required separation
by preparative recycling GPC (recGPC, see ESI†).
Results and discussion
Synthesis
The synthesis of the macrocycle building block is displayed in
Scheme 1. 9,10-Phenanthrenequinone (3) was brominated
yielding 3,6-dibromo-9,10-phenanthrenequinone (4).17 Conden-
sation of 4 with 1,3-diphenylacetone provides 3,6-dibromo-
phencyclone (5) as diene precursor of the desired triphenylene
structure.17 Diels–Alder reaction with subsequent CO-extrusion of
3,6-dibromophencyclone (5) and bis-(4-methoxyphenyl)acetylene
Scheme 1 (a) Br2, nitrobenzene, 120 ꢀC, 22 h (70%); (b) 1,3-diphenylacetone, MeOH/KOH, reflux, 30 min (80%); (c) 6, diphenyl ether, 245 ꢀC (55%); (d)
BBr3, ꢁ78 ꢀC to rt (100%); (e) 9, K2CO3, DMF, 65 ꢀC, 96 h (81%); (f) [(3-cyanopropyl)diisopropylsilyl]acetylene (CPDIPSA), CuI, PPh3, PdCl2(PPh3)2,
THF/piperidine, 65 ꢀC, 24 h (72%); and (g) THF, 5% water, TBAF, 0 ꢀC to rt, overnight (52% of 13).
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 1404–1415 | 1405

View Article Online
Scheme 2 (a): (a) CuCl/CuCl2, pyridine, 35 ꢀC, 120 h (53%); (b): (a) CuCl, TMEDA, O2, CH2Cl2, rt, 24 h (91%); (b) TBAF, THF, rt, 1 h (95%); and (c)
CuCl/CuCl2, pyridine, rt, 120 h (79%).
chromatography on silica gel. Acetylene dimerization of two
(mono-protected) half-ring components occurred under stan-
dard Glaser–Hay conditions (CuCl, TMEDA, and air) formed
the dimeric compound 14 and subsequent deprotection of 14
yielded the bisacetylene 15. Cu-catalyzed acetylene coupling
under the same conditions as applied for the dimerization of 12
(CuCl, CuCl2, pyridine, pseudo high-dilution conditions) gave
macrocycle 1 in considerably higher yield than in the statistical
reaction described above (Fig. 1). In this case, the MWD of the
stepwise cyclization shows almost only the desired product peak
and here, isolation of the product can be easily achieved by
simple column chromatography on silica gel. In both cases, 1 was
obtained as a faint yellow solid.20
Self-assembly at the liquid–solid interface
As shortly described in the introduction, the two-dimensional
organization of shape-persistent macrocycles at solid interfaces is
an attractive route towards complex nanostructures by the
bottom-up approach. Several reports on macrocycles on the
HOPG surface have been made,7,8 and also multi-component
structures that include rigid macrocycles and different guest
molecules were published during the past several years.8 In order
to evaluate if 1 is also able to form ordered two-dimensional
structures, we probed the self-assembly of 1 at the 1-phenyl-
octane–HOPG interface by scanning tunnelling microscopy
(STM). Interestingly, the self-assembly behavior of the
compound varies with concentration. At high concentration (2 ꢂ
10ꢁ4 M), molecules are arranged into close-packed lamella in
which the p-conjugated backbones are packed side by side in
a plane parallel to the graphite substrate. Only half of the alkyl
Fig. 1 Molecular weight distribution (MWD) plots of the crude prod-
ucts from the statistical and stepwise cyclization reactions received from
GPC-measurements (vs. polystyrene).
Alternatively, we investigated the stepwise formation of mac-
rocycle 1 (Scheme 2b).14,15 When 11 is treated with equimolar
amounts of TBAF in wet THF (containing 5% water) the reac-
tion rate of the deprotection considerably slows down. Thus, it
was possible to control the deprotection in such a way that the
main product is the mono-deprotected compound 13. By-prod-
ucts are completely deprotected bisacetylene 12 and educt 11.
Due to the great differences in polarity of the compounds con-
taining none, one or two CPDIPS groups, the reaction progress
can be easily monitored by thin layer chromatography (TLC)
and product separation can be easily achieved by simple column
1406 | J. Mater. Chem., 2011, 21, 1404–1415
This journal is ª The Royal Society of Chemistry 2011

View Article Online
chains of the molecule are absorbed on the substrate (Type I).
Yet at low concentrations (below 7 ꢂ 10ꢁ6 M), a two-dimen-
sional porous network forms and molecules are arranged into
a loose-packed lattice structure resulting in voids. Four out of six
alkyl chains on each side of the molecule are absorbed (Type II).
This has a pronounced effect on the molecular density: 0.100 ꢃ
0.004 molecule per nm2 for Type I and 0.050 ꢃ 0.002 molecule
per nm2 for Type II. The interpretation of the concentration
dependent polymorphism is discussed below.
A typical STM image of a monolayer formed upon drop-
casting a 2 ꢂ 10ꢁ4 M solution of 1 in 1-phenyloctane is shown in
Fig. 2. The higher tunnelling efficiency through p-electron-rich
parts makes them appear higher (brighter) than the alkyl chains
which are not well resolved and are located in the dark areas. A
lamellar structure is formed. The distance between two adjacent
molecules along the long lamella axis, a, is 1.3 ꢃ 0.1 nm; the
mean width, b, of a lamella is 7.3 ꢃ 0.1 nm. Molecular modelling
(HyperChem) shows that the p-electron-rich parts can adsorb in
a plane parallel to the substrate. Those alkyl chains which can be
identified adsorb parallel to one of the main symmetry directions
of graphite, highlighting the important role of the substrate in
templating the monolayer growth. As the distance between two
alkyl chains absorbed on graphite is approximately 0.40–0.45
nm,21 only three out of six alkyl chains on each side of the
molecule are absorbed on the substrate, as illustrated in the
schematic model which is superimposed on the STM image in
Fig. 2. The other alkyl chains are most likely directed toward the
solution phase and cannot be observed by STM. The molecular
density measures 0.100 ꢃ 0.004 molecule per nm2 (Table 1). The
unit cell contains one molecule: the plane group is p2. The main
driving forces for the observed molecular arrangement are the
van der Waals interactions between the alkyl side chains and
between the alkyl chains and the substrate.22
Fig. 2 Large-scale (left) and small-scale (right) STM image of a mono-
layer of 1 at the 1-phenyloctane–HOPG interface (2 ꢂ 10ꢁ4 M, Iset ¼ 8
pA, and Vbias ¼ 1 V). A unit cell and molecular models are superimposed
on the STM image (a ¼ 1.3 ꢃ 0.1 nm and b ¼ 7.3 ꢃ 0.1 nm).
alkyl chains aligned along one of the main symmetry axes of
graphite and those in adjacent rows interdigitated. The other
alkyl chains, which are not visible, may possibly be adsorbed in
the cavities between the molecules, or otherwise desorbed. A few
molecular models illustrating the anticipated ordering of the
molecules are superimposed on the STM image. The molecular
density measures only 0.050 ꢃ 0.002 molecule per nm2 (Table 1).
The unit cell contains one molecule and the plane group is p2.
Upon decreasing the concentration, such low-density networks
are observed exclusively. Typically, the voids can host solvent
molecules, which are too mobile to be resolved.
The formation and structure of self-assembled monolayers at
the liquid–solid interface depend on the mutual interplay
between molecule–molecule interactions, molecule–solvent
interactions and molecule–substrate interactions.23 The reported
concentration dependency of 1 is in line with recent studies on
concentration dependent ‘‘polymorphism’’ for physisorbed
monolayers at the liquid–solid interface. The densest polymorph
is formed at the higher concentrations and self-assembly at the
Upon decreasing the concentration, a new pattern is identified.
For instance, at a concentration of 7 ꢂ 10ꢁ6 M, a loose-packed
lattice structure is the dominant polymorph (Fig. 3 (left)). The
densely packed lamellar structure is the minority phase. Fig. 3
(right) is a typical high-resolution STM image of the open lattice
structure, recorded at even lower concentration (7 ꢂ 10ꢁ8 M).
The unit cell vectors a and b are 3.8 ꢃ 0.1 and 5.8 ꢃ 0.1 nm,
respectively. The angle between a and b, a, measures 62 ꢃ 2ꢀ. The
increased distance between two molecules allows four alkyl
chains at each side to be adsorbed in all trans-conformation on
graphite, which is confirmed by STM imaging (Fig. 3 (right)). In
this STM image, the alkyl chains are clearly identified. These
lower concentrations results in
a
porous network.24 The
adsorption–desorption equilibrium at the liquid–solid interface
determines the surface coverage ratio between the dense and the
open network. The concentration dependency of the self-
assembling can be understood to arise from the different stability
and molecular density of the two polymorphs.
Table 1 Unit cell parameters and molecular density of monolayers of 1
determined by STM at the liquid–solid interface at various concentra-
tions
Compound 1 in
phenyloctane,
concentration/ꢂ10ꢁ6
Molecular
density/nmꢁ2
M
a/nm
b/nm
a/ꢀ
200
7a
0.07
1.3 ꢃ 0.1 7.3 ꢃ 0.1 90 ꢃ 2 0.100 ꢃ 0.004
3.8 ꢃ 0.1 5.7 ꢃ 0.1 62 ꢃ 2 0.050 ꢃ 0.002
3.8 ꢃ 0.1 5.8 ꢃ 0.1 65 ꢃ 2 0.050 ꢃ 0.002
Fig. 3 STM image of a monolayer of 1 at the 1-phenyloctane–HOPG
interface. Left: 7 ꢂ 10ꢁ6 M, Iset ¼ 8 pA, and Vbias ¼ 1 V. Right: 7 ꢂ 10ꢁ8
M, Iset ¼ 6 pA, and Vbias ¼ 1 V. A unit cell and molecular models are
superimposed on the STM image.
a
Unit cell parameters refer to open lattice structure.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 1404–1415 | 1407

View Article Online
Steady-state and time-resolved photoluminescence
image of its absorption spectrum; however, the same is not true
for 15. Such exceptions to the mirror image rule are known to
result from either a structural rearrangement in the excited state
or formation of lower energy excimers involving different tri-
phenylene units.27 In the absence of any red-shifted emission
feature in the PL spectra of 15, excimer formation can be safely
ruled out.28 A departure from the mirror image rule therefore
might be a consequence of a structural rearrangement in the
excited state. We speculate that owing to the molecular rigidity, 1
is far less likely to undergo any appreciable structural reorgani-
zation in the excited state, and therefore the mirror image rule
holds good in its case.
The open dimer 15 and the macrocycle 1 offer an interesting
ground to explore elementary aspects of delocalization in large
p-electron systems.16 For example, does twisting or bending have
a greater impact on the energetics of the p-system? This question
is crucial to understanding the microscopic nature of excited
states in materials for organic electronics, such as conjugated
polymers. Subtle single-molecule spectroscopic techniques have
been able to identify a vast diversity of both intramolecular
bending and twisting occurring within the conjugated chain.25,26
It is not immediately obvious which of the two model
compounds, the open dimer 15 or the macrocycle 1, supports
greater delocalization. In order to address these questions, we
combined optical and quantum chemical investigations of the
To test this hypothesis we applied first principles Hartree Fock
calculations to determine the ground state molecular configura-
tions, and then used configuration interaction singles (CIS) to
relax the excited state geometry. Vertical transition energies were
determined with the ZINDO Hamiltonian and all calculations
were performed using the Gaussian program package.29 We
found that indeed the ground state of the open dimer 15 is
twisted, with a dihedral angle of 41ꢀ, whereas the macrocycle 1
ensured the coplanarity of the triphenylene moieties (Fig. 5). The
energy difference between the twisted and planar configurations
is, however, marginal (0.0008 eV), and so the twisted confor-
mation may not be exclusive in the ground state. For the ground
state configurations we determine bright states at 3.54 eV and
3.82 eV for the open dimer which we compare to the absorption
peaks at 3.37 eV and 3.79 eV respectively, and for the macrocycle
we find a bright state at 3.55 eV which we associate with the peak
at 3.37 eV. The lower energy transition for the open bridge is
HOMO–LUMO dominated, and the transition for the macro-
cycle has contributions from the HOMO ꢁ 1 and LUMO + 1 as
well as the HOMO and LUMO. Fig. 5 shows the computed
conformations and frontier orbital plots of 1 and 15. By exam-
ining the spatial extent of the HOMO and LUMO, we find that
the orbitals reach over the ring formed by the two diacetylene
bridges in 1, thus increasing the degree of p-delocalization.
Further details of the ground state geometries are given in the
ESI†; it is found that the triphenylene moiety in 15 is slightly bent
around a central axis, whereas a subtle asymmetric twisting
occurs in 1.
ꢁ
macrocycle 1 vis-a-vis its open dimer 15 to allow us to charac-
terize the extent of p-electron delocalization. The difference
between 1 and 15 is principally due to the increased p-delocal-
ization in the macrocycle in the ground state. This increase is
reversed in the excited state due to strong structural relaxation.
Fig. 4 shows the UV-Vis absorption and steady state photo-
luminescence (PL) spectra of the monomer 12 and the two
dimers, 15 and 1, in solution. For the discussion of the effect of
dimerization and cyclization on the optical spectra we bear in
mind that the absorption wavelength relates to the ground state
electronic conformation whereas emission occurs following
structural relaxation in the excited state. The lowest energy
transition of 12 appears at 4.0 eV. Upon dimerization, it shifts to
lower energies. However, the shift is slightly greater in the case of
the macrocycle 1, suggesting that the degree of p-delocalization
is higher in 1 in the ground state. Though significantly strained,
the pair of arylene bridges in 1 enforces a higher degree of
coplanarity between the two triphenylene units, and thus
potentially facilitates efficient delocalization throughout the
molecule. In addition to the spectral shifts, important informa-
tion about the excited state behavior can also be obtained from
the spectral shapes. We find that the PL spectrum of 1 is a mirror
On determination of the excited state configurations (lower
panel in Fig. 5) we indeed find that 15 planarizes to a dihedral
angle of 8ꢀ and we find that the planar structure is 1.58 eV more
stable than the twisted configuration in the excited state. Inter-
estingly, compound 1 also relaxes with a loss of symmetry of the
bridges. We determined transition energies of 2.65 eV for the
open dimer and 2.55 eV for the macrocycle which we compare to
the lowest energy peaks of the emission spectra at 2.80 eV and
2.64 eV, respectively. In our calculations it therefore appears that
we can reproduce the configurational changes which give rise to
the reversal of peak order on going from absorption to emission
of 15 and 1, respectively. Our calculations support the assump-
tion that the significant difference between the absorption and
emission spectrum of the open dimer is indeed due to a more
dramatic change in the excited state configuration when
compared to 1. We note, however, that interpretation of the
excited state transition energies requires a precise analysis of the
relative oscillator strengths of the individual transitions, which
we leave to a future detailed analysis. Suffice to say that our
Fig. 4 Absorption (solid) and photoluminescence (dotted) spectra of the
monomer 12 (black), open dimer 15 (red) and the macrocycle 1 (blue) in
chloroform solution.
1408 | J. Mater. Chem., 2011, 21, 1404–1415
This journal is ª The Royal Society of Chemistry 2011

View Article Online
Fig. 5 Configurations and frontier orbital plots for the open dimer 15 and macrocycle 1. The upper panel shows the ground state configuration in which
15 is twisted about the diacetylene bridge, and the frontier orbitals are delocalized over both bridges in 1. The lower panel shows the configuration after
excitation in which 15 planarizes, and the bridges in 1 are no longer symmetrical resulting in a reduction in the strength of conjugation over the longer
bridge.
present results lead us to speculate that bending of the p-electron
system may promote overlap between px and py orbitals, leading
to an internal conversion to the lowest energy (dark) state thus
promoting non-radiative decay and lowering quantum yield.
This internal conversion following Kasha’s rule could be reduced
in the open dimer, impeding non-radiative decay (see ESI†).
A better understanding of the excited state geometry of these
compounds can be developed from PL lifetime (s) measurements.
Fig. 6 presents the time resolved PL decay of the three
compounds. All three exhibit a single-exponential decay. There
does not appear to be any systematic correlation between s and
the degree of delocalization in the ground state. The monomer 12
shows the longest-lived PL, with a lifetime of ꢄ8.2 ns. Among the
dimers, the less delocalized 15 exhibits a faster PL decay than the
more delocalized 1. However, we observe a direct correlation
between the PL decay rates and the corresponding PL quantum
efficiencies (F). s and F are related to the radiative (kr) and non-
radiative (knr) decay rate constants by sꢁ1 ¼ kr + knr and F ¼ s ꢂ
kr. The values of kr and knr calculated from the experimentally
determined s and F values are given in Table 2. It is interesting to
note that the introduction of a second diacetylene bridge in 1
influences the radiative and non-radiative decay channels in an
opposite manner. As we move from the flexible open dimer 15 to
the more rigid macrocycle 1, kr decreases 2.7-fold. On the other
hand, knr increases by a factor of 2. This trend may initially
appear surprising, for one could intuitively expect the more
planar macrocyclic structure to have greater oscillator strength
and thus larger kr. We note that a similar dependence of the non-
radiative decay rate on the dihedral angle was recently reported
for p-quaterphenyl derivatives which were assigned to systematic
changes in electron–vibrational coupling due to shifting of
ground and excited state potential surfaces.30 It was shown that
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 1404–1415 | 1409

View Article Online
an increased delocalization in the ground state (smaller dihedral
angles) led to a faster non-radiative decay of the excited singlet
state, and vice versa. The ratio, kr/knr, can be related to the
conjugation length (Ap*) in the excited singlet state by the
31
˚
expression, Ap* ¼ ln (kr/knr) + 4.6 (A). Using this expression,
we calculate the excited state conjugation length for 1 and 15 as
given in Table 2. Following optical excitation, the open dimer 15
undergoes significant structural reorganization in the excited
state to become coplanar. The structurally rigid macrocycle 1, on
the other hand, does not undergo much structural change upon
excitation. Therefore, in the excited state, where the two triphe-
nylene units are coplanar in both 1 and 15, a difference in the
extent of delocalization is possibly governed by the strain in
the diacetylene bridges (bending), rather than twisting of the
p-system. From the quantum chemistry calculations we deter-
mine that in the excited state the bridges in 1 are not exactly the
˚
same length with a slight difference of 0.03 A between them,
resulting in the reduction of the strength of conjugation over the
longer bridge as seen in Fig. 5. The shorter bridge in the mac-
˚
rocycle 1 has the same absolute length of 6.66 A as that in 15, but
Fig. 6 Photoluminescence intensity decay of the monomer 12, open
dimer (15) and macrocycle (1).
whereas in 15 the bridge is linear, in 1 the distance between tri-
˚
phenylene moieties is 0.30 A less. As a consequence of this
compression, the bridges in 1 are bent. We therefore conclude
that strain in the bridge reduces the effective conjugation, thus
increasing non-radiative decay. Increased p-conjugation in the
excited state leads to a faster radiative decay, and consequently
a higher quantum efficiency for 15. To summarize, bending of the
arylene bridges has a greater impact on conjugation in the excited
state than in the ground state where torsion dominates; the
macrocycle 1 is more conjugated in the ground state than the
open dimer 15, but less so in the excited state.
Aggregation
The aggregation of shape-persistent phenylene ethynylene mac-
rocycles has been extensively investigated during the past
decade.5 It is generally accepted that electron accepting groups
enhance the self-association while electron donating or bulky
substituents disfavor the aggregation. Apart from approaches
that use an alternating arrangement of electron-rich and elec-
tron-deficient aromatics, or aromatics and perfluoro aromatics,
solvent effects play a highly important role in the aggregation
process. The concentration dependence of the pattern formation
of 1 on HOPG motivated us to investigate if an assembly of the
molecules could also be observed in solution.
Fig. 7 Solution NMR-spectra of macrocycle 1 in dichloromethane (6.7
ꢂ 10ꢁ4 M) and cyclohexane (6.7 ꢂ 10ꢁ4 M).
representing a first indication of the self-association of 1.
However, a determination of the aggregation constant by
concentration dependent NMR-chemical shift variations was not
possible.
Fig. 7 shows the proton NMR data of 1 in dichloromethane
and cyclohexane solution, respectively. In the good solvent
dichloromethane, macrocycle 1 shows sharp proton signals in the
aromatic and the aliphatic region as well. In cyclohexane, which
is assumed to be a poor solvent for the polycyclic aromatic, the
proton signals in the aromatic region flattens considerably,
Therefore, dynamic light scattering (DLS) experiments were
performed with 1 in cyclohexane and dichloromethane solutions
to further investigate the aggregation behavior in different
solvents (Fig. 8).5d,32 The CONTIN analysis33 of the autocorre-
lation function at a scattering angle of 90ꢀ in dichloromethane
solution showed one process corresponding to a hydrodynamic
radius (Rh) of approximately 1.7 nm. From the cyclohexane
solution, analysis of the autocorrelation function provides
a hydrodynamic radius about 3.6 nm, confirming the first indi-
cation of self-association received from the corresponding
proton NMR-spectra. However, the aggregate size is not suffi-
ciently large that extended tubular superstructures can be
proposed. This result is coincident with the fact that the cyclo-
hexane solutions do not exhibit a high viscosity, nor do they
Table 2 Overview of photophysical parameters of the open dimer 15
and macrocyle 1 (see text for definitions)
˚
Sample
labs/nm s/ns
F
kr/nsꢁ1 knr/nsꢁ1 kr/knr Ap*/A
Compound15 395
Compound 1 403
1.7 0.86 0.5
2.9 0.54 0.19
0.08
0.16
6.3
1.2
6.4
4.8
1410 | J. Mater. Chem., 2011, 21, 1404–1415
This journal is ª The Royal Society of Chemistry 2011

View Article Online
show any birefringence. At present, we attribute this behavior to
the phenyl rings in the 1- and 4-position which disfavor a strong
aggregation of the macrocycles even if the compounds are
twisted by 90ꢀ within a column due to the strong entropy loss.34
In order to obtain more information on the self-assembly of 1 in
cyclohexane, we performed small-angle X-ray scattering experi-
ments. The solutions were investigated at a concentration of 10 g
Lꢁ1. The measured scattering curve is shown in Fig. 9. The scat-
tering curve can best be described by a core/shell-cylinder with
a high electron density aromatic core and a low electron density
alkyl shell. The solid line in Fig. 9 is a fit to the data using the form
factor of a core/shell-cylinder with a homogeneous core and an
inhomogeneous shell with a brush-like density profile of the form
rꢁb. The form factor P(q) for such a core/shell-cylinder is given by:
P(q) ¼ hF2(q)ihF2t(q)i
(1)
k
Fig. 9 Small-angle X-ray scattering curve of 1 measured in cyclohexane
at a concentration of 2.2 ꢂ 10ꢁ3 M. The solid line is a fit to the data using
a model of a core/shell-cylinder consisting of a high electron density
aromatic core and a low electron density alkyl shell.
where F (q) is the longitudinal scattering amplitude parallel to
k
the cylinder axis. Ft(q) is the contribution from the cross-section
of the cylinder given by:
FtðqÞ ¼
ꢀ
ꢁ
1
r
rpbꢁ2
Fð0; Rc; qÞ ꢁ
Fðb; Rc; qÞ þ
Fðb; Rm; qÞ
2
ð2 ꢁ bÞ
ð2 ꢁ bÞ
a cylinder core radius Rc ¼ 2.7 nm, an outer radius of Rm ¼ 3.9
nm, an electron density ratio of r ¼ ꢁ0.56 and an exponent for
the radial density profile of b ¼ 0.93. The negative value of the
density ratio indicates that the aromatic core has a higher and the
alkyl shell a lower electron density compared to the solvent
cyclohexane. The value of b is close to the value of b ¼ 1 expected
for stiff elongated alkyl chains extending from a cylindrical core.
The core/shell-structure is schematically shown in Fig. 10. Many
other models such as spheres and disks with core/shell-structure
were also investigated, but the core/shell-cylindrical model gave
by far the best agreement with the experimental data.
ꢀ
ꢁ
1
r
rpbꢁ2
ꢁ
þ
2
ð2 ꢁ bÞ ð2 ꢁ bÞ
ꢀ
ꢁ
2 ꢁ b
4 ꢁ b
q2R2
with Fðb; R; qÞ ¼ 1F2
; 1;
;
ꢁ
being the
2
2
4
hypergeometric function where r is the core/shell electron density
ratio and p ¼ Rc/Rm the ratio of core-radius to outer radius of the
cylinder. The longitudinal scattering amplitude is given by
ꢂ
ꢃ
2
Â
Ã
2
sinðqL=2Þ
qL=2
The small-angle X-ray scattering data therefore suggest that in
cyclohexane 1 self-assembles by stacking into short cylinders (as
suggested by the DLS measurements). The value of the core
radius and outer radius is in good agreement with the dimensions
of a cylindrical aggregate with the aromatic groups forming the
core and the alkyl chains forming the shell. It is also in good
agreement with the structures observed in the STM measure-
ments and the AFM investigations (see below).
FjjðqÞ ¼
Si qL ꢁ
with Si[z] being the sine-
qL
integral. The averages h.i are calculated over a Schulz–Zimm-
type distribution of radii. The details of the calculations have
been described previously.35
The form factor P(q) (eqn (1)) was fitted to the measured
scattering curve using a Simplex-algorithm. As a result we obtain
If 1 assembles into cylindrical structures in dilute solution, it
could be expected that the cylinders assemble into a hexagonal
arrangement upon evaporation of the solvent. For such a bulk
sample we obtained the SAXS-curve as shown in Fig. 11.
We observe three rather broad maxima at q ¼ 1.25 nmꢁ1
,
q ¼ 2.4 nmꢁ1, and q ¼ 3.3 nmꢁ1 together with a weak reflection at
q ¼ 3.7 nmꢁ1. This is an unusual set of reflections, indicating
lower dimensional, i.e. one- or two-dimensional packing of
objects. A scattering curve that captures the main features of the
experimental curve is shown by the solid line in Fig. 11. It is
calculated for the model given in Fig. 10b, where homogeneous
cylinders are packed in a hexagonal array. The positions of the
corresponding reflections are indicated. The solid line is
calculated as:
I(q) ¼ I0P(q)S(q)
(2)
Fig. 8 CONTIN-fit of DLS-derived relaxation time distribution of 1 in
dichloromethane (3.1 ꢂ 10ꢁ4 M) and cyclohexane (3.1 ꢂ 10ꢁ4 M).
S(q) ¼1 + b(q)(Z0(q) ꢁ 1)G(q)
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 1404–1415 | 1411

View Article Online
Fig. 10 (a) Density profile of the core/shell-cylinders observed in dilute
solution in cyclohexane. The core radius is Rc ¼ 2.7 nm and the outer
radius is Rm ¼ 3.9 nm. The density profile in the shell decays as rꢁ0.92 (b)
hexagonal arrangement of cylinders observed in the dry state. The
cylinder radius Rhex ¼ 2.4 nm and the distance between two cylinders is
dhex ¼ 5.8 nm.
where I0 is a proportionality factor related to primary beam
intensity, scattering contrast and number density, P(q) is the
form factor of a cylinder, S(q) is the structure factor given by
a set of reflections related to the lattice factor Z0(q), the Debye–
Waller factor G(q), and a fluctuation term b(q). Details of the
calculation are given elsewhere.36
We observe the first five reflections (10), (11), (20), (21), and
(30) of a hexagonal lattice, where the (11) and (20) reflections
appear to be merged into one broad peak. The cylinder radius is
Rhex ¼ 2.4 nm and the distance between two cylinders is dhex
5.8 nm with the expected size reduction, i.e. Rhex < Rc and dhex
¼
<
2Rm when going from the solvent swollen state to the dry state.
This suggests that the cylindrical aggregates are sufficiently
stable to assemble into a regular hexagonal lattice in the bulk
state.
Fig. 12 Tapping-mode topography AFM image and 2D-FFT graph
(inset) of 1 deposited from chloroform on HOPG. The center-to-center
distance between adjacent fibers is about 13.8 nm (top). Tapping-mode
topography AFM image and 2D-FFT graph (inset) of 1 deposited from
cyclohexane on HOPG. The center-to-center distance between adjacent
rows is about 8.7 nm (bottom).
The self-assembly of 1 into tubular aggregates was additionally
investigated by means of atomic force microscopy (AFM). As
shown before, STM at a liquid–solid interface only reveals
monolayer structures or in some exceptional cases, multilayer
formation. However, to detect and identify self-assembled
objects formed upon drop-casting and solvent evaporation,
AFM is a better tool. It turns out that 1 forms fiber-like features
on HOPG when deposited from chloroform and cyclohexane
following solvent evaporation, as shown in Fig. 12. The initial
concentration of both solutions was 3.3 ꢂ 10ꢁ5 M. The ‘‘fibers’’
are longer when deposited from chloroform than from cyclo-
hexane, which could be due to the influence of the solvent.37 For
instance, the surface tensile force (or contact angle), the solvent
evaporation rate, and the specific molecule–solvent interactions
can affect the assembling process.
The spots in the two-dimensional-fast Fourier transform
(2D-FFT) graphs of AFM images recorded on samples prepared
from chloroform show a roughly hexagonal order (Fig. 12).
However, the spots are broad which suggests a distribution of
orientations around 60ꢀ. The spots in the 2D-FFT graphs from
AFM images recorded on samples prepared from cyclohexane
Fig. 11 Scattering curve of the sample in the dry state. The scattering
curve can be described by a hexagonal array of cylinders with relatively
small ordered domain sizes leading to the observed considerable peak
broadening, as illustrated by the calculated curve (solid line).
1412 | J. Mater. Chem., 2011, 21, 1404–1415
This journal is ª The Royal Society of Chemistry 2011

View Article Online
are more distinct, and 12 spots can be recognized in the 2D-FFT
graphs of the lower panel in Fig. 12. Basically, these 2D-FFT
graphs show two sets of hexagonal patterns, rotated with respect
to each other by about 22 ꢃ 1ꢀ.
Nevertheless, to approach the conditions used for the scanning
tunneling microscopy experiments, upon drop casting a 1 ꢂ 10ꢁ4
M solution of 1 in phenyloctane on the basal plane of graphite,
and after evaporation of the solvent over 22 hours, the film
formation was investigated. In certain areas fiber-like features
were observed, which often appeared to be part of dense
networks as shown in the right image of Fig. 14. The width of
a fiber is 31.6 ꢃ 3.3 nm which is much larger than that formed in
chloroform and cyclohexane, indicating a more complicated
stacking mode. In addition, 1 tends to grow structures devel-
oping into the third dimension (3D) in some domains as pointed
out by the green arrows in the left image of Fig. 14. Samples
prepared at a lower concentration show patches or islands
forming (not shown). The height of such films was measured to
be 1.0 ꢃ 0.2 nm, which indicates that in those regions 1 is
oriented with its macrocycle unit perpendicular to the substrate.
Clearly, the self-assembly of these macrocycles is not only
concentration dependent, but shows in addition to a solvent
dependency, also interface dependent phenomena. The fact that
the STM and AFM experiments reveal different kinds of nano-
structures might be surprising initially despite the fact that
samples were prepared in a similar way (concentration, substrate,
and solvent). Note though that the AFM samples were investi-
gated at the air/solid interface upon evaporation of the solvent,
while the STM experiments were carried out at the liquid/solid
interface. So, there is an important concentration effect on the
final nanostructures. In addition, one should take into account
the difference in the way STM and AFM are probing the
surfaces. In STM, objects larger than a few Angstrom in height
will be pushed away by the tip while such objects can be easily
revealed by AFM.
In both cases, the 2D-FFT graphs clearly indicate the impor-
tance of substrate–molecule interactions in affecting the growth
direction of the fibers. This also suggests that the fibers are not
merely deposited from solution but that they are formed at the
liquid–solid interface. The two subsets in the 2D-FFT graphs for
AFM images of samples deposited from cyclohexane are a clear
signature that the fibers are not aligned along one of the
symmetry directions of graphite. Most likely, the fiber’s long axis
runs at an angle of 11 ꢃ 1ꢀ with respect to a symmetry direction
of graphite.
The center-to-center distance between adjacent fibers is quite
different: about 13.8 nm from a chloroform solution and 8.7 nm
from a cyclohexane solution, respectively. The molecular
dimensions of 1 are displayed in Fig. 13. The calculated size
corresponds well with the results from the STM investigations
and with the observed fiber size in solution.38
In cyclohexane, we suggest that the fibers are formed from
parallel organization of the tubular 1D stacks. Surprisingly, also
a coincidence with the results from the light scattering and from
the X-ray scattering in terms of the restricted cylinder length is
found. On the other hand, the structures formed from chloro-
form are much harder to interpret and at present it is not clear if
dimers of the tubular aggregates or even trimers of those
aggregates are observed under the conditions we applied for the
AFM measurements. Dimers and trimers of aggregates have
previously been also observed on similar systems.39
Conclusion
In summary, we have synthesized a shape-persistent tripheny-
lene–butadiynylene macrocycle by two different routes using the
well-known Glaser-coupling as cyclization reaction. The mac-
rocycle adsorbs on HOPG to form well-ordered 2D-structures,
however, the density of the packing depends on the concentra-
tion of the supernatant phenyloctane solution. The photo-
physical investigation of the macrocycle and its open precursor
has revealed that the macrocycle is more conjugated in the
ground state and less so in the excited state, a result of the
diacetylene bending in the macrocycle due to its constrained
topology. The flexible side groups also guarantee a high
compound solubility even in nonpolar solvents (cyclohexane).
However, solvophobic interactions lead to the formation of tube-
like superstructures that were identified by scattering methods as
well as by AFM.
Fig. 13 Calculated molecular dimensions of 1 (HyperChem).
Acknowledgements
Financial support by the Deutsche Forschungsgemeinschaft
(DFG), the SFB 624, the MIWFT in the frame of the consortium
ENQUETE, the Volkswagen Foundation (JML and SH), K.U.
Leuven (GOA), Fund for Scientific Research—Flanders
(F.W.O.) and the Belgian Federal Science Policy Office through
IAP-6/27 is gratefully acknowledged. EML and JML
Fig. 14 Tapping-mode topography AFM image of 1 deposited from
1-phenyloctane on HOPG. The width of the fiber in the right image is
31.6 ꢃ 3.3 nm.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 1404–1415 | 1413

View Article Online
Chem. Soc., 2006, 128, 3502; (i) S.-H. Jung, W. Pisula,
acknowledge support through a LANL-CINT user project. JML
is a David and Lucile Packard Foundation Fellow.
€
€
A. Rouhanipour, H. J. Rader, J. Jacob and K. Mullen, Angew.
Chem., 2006, 118, 4801; (j) K. Tahara, C. A. Johnson, II, T. Fujita,
M. Sonoda, F. C. De Schryver, S. De Feyter, M. M. Haley and
Y. Tobe, Langmuir, 2007, 23, 10190; (k) M. Schmittel, V. Kalsani,
Notes and references
€
C. Michel, P. Mal, H. Ammon, F. Jackel and J. P. Rabe, Chem.–
Eur. J., 2007, 13, 6223; (l) S. Lei, K. Tahara, F. C. De Schryver,
M. Van der Auweraer, Y. Tobe and S. De Feyter, Angew. Chem.,
2008, 120, 3006; (m) S.-S. Jester, N. Shabelina, S. M. Le Blanc and
1 As it holds also for linear rigid structures, shape-persistent cyclic
structures have to be decorated with side groups that guarantee
processability, either from solution or from the melt: (a) W. Kern,
M. Seibel and H. O. Wirth, Makromol. Chem., 1959, 29, 164; (b)
€
S. Hoger, Angew. Chem., 2010, 122, 6237.
€
8 (a) G.-B. Pan, X.-H. Cheng, S. Hoger and W. Freyland, J. Am. Chem.
Soc., 2006, 128, 4218; (b) E. Mena-Osteritz and P. Bauerle, Adv.
Mater., 2006, 18, 447; (c) S. Furukawa, K. Tahara, F. C. De
Schryver, M. van der Auweraer, Y. Tobe and S. De Feyter, Angew.
€
Chem., 2007, 119, 2889; (d) K. Tahara, S. Lei, D. Mossinger,
H. Kozuma, K. Inukai, M. Van der Auwaerar, F. C. De Schryver,
€
H. A. Staab, H. Braunling and K. Schneider, Chem. Ber., 1968,
101, 879.
2 (a) J. S. Moore, Acc. Chem. Res., 1997, 30, 402; (b) S. Hoger, J. Polym.
Sci., Part A: Polym. Chem., 1999, 37, 2685; (c) M. M. Haley, J. J. Pak
and S. C. Brand, Top. Curr. Chem., 1999, 201, 81; (d) C. Grave and
€
€
€
A. D. Schluter, Eur. J. Org. Chem., 2002, 3075; (e) Y. Yamaguchi
€
S. Hoger, Y. Tobe and S. De Feyter, Chem. Commun., 2008, 3897;
and Z. Yoshida, Chem.–Eur. J., 2003, 9, 5430; (f) J. A. Marsden,
G. J. Palmer and M. M. Haley, Eur. J. Org. Chem., 2003, 2355; (g)
(e) K. Tahara, S. Lei, W. Mamdouh, Y. Yamaguchi, T. Ichikawa,
H. Uji-i, M. Sonoda, K. Kirose, F. C. De Schryver, S. De Feyter
and Y. Tobe, J. Am. Chem. Soc., 2008, 130, 6666; (f) S. Lei,
€
K. Tahara, X. Feng, S. Furukawa, F. C. De Schryver, K. Mullen,
Y. Tobe and S. De Feyter, J. Am. Chem. Soc., 2008, 130, 7119; (g)
€
D. Zhao and J. S. Moore, Chem. Commun., 2003, 807; (h) S. Hoger,
Chem.–Eur. J., 2004, 10, 1320; (i) W. Zhang and J. S. Moore,
Angew. Chem., 2006, 118, 4524.
3 (a) J. Zhang and J. S. Moore, J. Am. Chem. Soc., 1994, 116, 2655; (b)
O. Mindyuk, M. R. Stetzer, P. A. Heiney, J. C. Nelson and
J. S. Moore, Adv. Mater., 1998, 10, 1363; (c) M. Fischer, G. Lieser,
A. Rapp, I. Schnell, W. Mamdouh, S. De Feyter, F. C. De Schryver
€
B. Schmaltz, A. Rouhanipour, H. J. Rader, W. Pisula and
K. Mullen, Angew. Chem., 2009, 121, 734; (h) T. Chen, G.-B. Pan,
€
€
H. Wettach, M. Frtitzsche, S. Hoger, L.-J. Wan, H.-B. Yang,
B. H. Nortrop and P. J. Stang, J. Am. Chem. Soc., 2010, 132, 1328.
9 (a) W. J. Liu, Y. Zhou, Q.-F. Zhou, Y. Ma and J. Pei, Org. Lett., 2008,
€
and S. Hoger, J. Am. Chem. Soc., 2004, 126, 214; (d) S. H. Seo,
H. Seyler, J. O. Peters, T. V. Jones, T. H. Kim, J. Y. Chang and
ꢀ
10, 2123; (b) J. M. W. Chan, J. R. Tischler, S. E. Kooi, V. Bulovic and
G. N. Tew, J. Am. Chem. Soc., 2006, 128, 9264; (e) W. Pisula,
€
T. M. Swager, J. Am. Chem. Soc., 2009, 131, 5659; (c) D. Wang,
J. F. Hsu, M. Bagui, V. Dusevich, Y. Wang, Y. Liu, A. J. Holder
and Z. Peng, Tetrahedron Lett., 2009, 50, 2147; (d) C. Chou,
D. Wang, J. F. Hsu, Y. Liu and Z. Peng, Synth. Met., 2009, 159,
1657.
M. Kastler, C. Yang, V. Enkelmann and K. Mullen, Chem.–Asian
J., 2007, 2, 51; (f) H. Shimura, M. Yoshio and T. Kato, Org.
Biomol. Chem., 2009, 7, 3205.
€
4 (a) S. Hoger, V. Enkelmann, K. Bonrad and C. Tschierske, Angew.
Chem., 2000, 112, 3256; (b) S. Hoger, X. H. Cheng,
€
10 Functionalized phenyl dibenzo[fg,op]napthacenes were prepared from
the corresponding 1,3-di(2-bromophenyl)-2,5-diarylbenzenes via
A.-D. Ramminger, V. Enkelmann, A. Rapp, M. Mondeshki and
€
I. Schnell, Angew. Chem., 2005, 117, 2862; (c) S. Hoger, J. Weber,
€
Pd(0)-catalyzed dehydrohalogenation: (a) X. H. Cheng, S. Hoger
and D. Fenske, Org. Lett., 2003, 5, 2587; (b) X. Cheng, A. Ver
A. Leppert and V. Enkelmann, Beilstein J. Org. Chem., 2008, 4, 1.
5 (a) J. Zhang and J. S. Moore, J. Am. Chem. Soc., 1992, 114, 9701; (b)
A. S. Shetty, J. Zhang and J. S. Moore, J. Am. Chem. Soc., 1996, 118,
1019; (c) Y. Tobe, A. Nagano, K. Kawabata, M. Sonoda and
K. Naemura, Org. Lett., 2000, 2, 3265; (d) S. Rosselli, A.-
€
Heyen, W. Mamdouh, H. Uji-I, F. De Schryver, S. Hoger and
S. De Feyter, Langmuir, 2007, 23, 1281, The tetraphenylbenzenes
were obtained from the condensation of pyrylium salts with phenyl
acetates as described by Zimmermann and Fischer: T. Zimmermann
and G. W. Fischer, J. Prakt. Chem., 1987, 329, 975.
€
D. Ramminger, T. Wagner, B. Silier, S. Wiegand, W. Haußler,
G. Lieser, V. Scheumann and S. Hoger, Angew. Chem., 2001, 113,
€
€
11 (a) W. Dilthey, S. Henkels and A. Schafer, Ber. Dtsch. Chem. Ges. B,
3234; (e) K. Nakamura, H. Okubo and M. Yamaguchi, Org. Lett.,
2001, 3, 1097; (f) Y. Tobe, N. Utsumi, K. Kawabata, A. Nagano,
K. Adachi, S. Araki, M. Sonoda, K. Hirose and K. Nakamura,
J. Am. Chem. Soc., 2002, 124, 5350; (g) D. Zhao and J. S. Moore,
J. Org. Chem., 2002, 67, 3548; (h) S. Rosselli, A.-D. Ramminger,
1938, 71, 974; (b) R. A. Pascal, Jr, D. Van Engen, B. Kahr and
W. D. McMillan, J. Org. Chem., 1988, 53, 1687; (c) W. Ried and
V. B. Saxena, Justus Liebigs Ann. Chem., 1970, 739, 159.
12 (a) H. M. Duong, M. Bendikov, D. Steiger, Q. Zhang, G. Sonmez,
J. Yamada and F. Wudl, Org. Lett., 2003, 5, 4433; (b) R. A. Pascal,
Chem. Rev., 2006, 106, 4809; (c) M. Saleh, M. Baumgarten,
€
T. Wagner, G. Lieser and S. Hoger, Chem.–Eur. J., 2003, 9, 3481;
(i) Y. Saiki, H. Sugiura, K. Nakamura, M. Yamaguchi, T. Hoshi
and J. Anzai, J. Am. Chem. Soc., 2003, 125, 9268; (j) C. T. L. Ma
and M. J. MacLachlan, Angew. Chem., 2005, 117, 4250; (k) L. Shu
and M. Mayor, Chem. Commun., 2006, 4134; (l) H. Enozawa,
M. Hasegawa, D. Takamatsu, K. Fukui and M. Iyoda, Org. Lett.,
2006, 8, 1917; (m) C. Couet and M. Biesalski, Soft Matter, 2006, 2,
1005; (n) P. Flores, P. Guadarrama, E. Ramos and S. Fomine,
J. Phys. Chem. A, 2008, 112, 3996; (o) J.-K. Kim, E. Lee, M.-
C. Kim, E. Sim and M. Lee, J. Am. Chem. Soc., 2009, 131, 17768.
6 T. Naddo, Y. Che, W. Zhang, K. Balakrishnan, X. Yang, M. Yen,
J. Zhao, J. S. Moore and J. Zang, J. Am. Chem. Soc., 2007, 129, 6978.
7 (a) J. Kromer, I. Rios-Carreras, G. Fuhrmann, C. Musch,
M. Wunderlin, T. Debaerdemacker, E. Mena-Osteritz and
€
A. Mavrinskiy, T. Schafer and K. Mullen, Macromolecules, 2010,
€
43, 137.
13 For triphenylene based electrooptical active compounds see, e.g.: (a)
S.-i. Kawano, C. Yang, M. Ribas, S. Baluschev, M. Baumgarten
€
and K. Mullen, Macromolecules, 2008, 41, 7933; (b) M. Saleh,
Y.-S. Park, M. Baumgarten, J.-J. Kim and K. Mullen, Macromol.
€
Rapid Commun., 2009, 30, 1279; (c) T. Qin, G. Zhou, H. Scheiber,
R. E. Bauer, M. Baumgarten, C. E. Anson, E. J. W. List and
€
K. Mullen, Angew. Chem., 2008, 120, 8416; (d) H. Wettach,
S. S. Jester, A. Colsmann, U. Lemmer, N. Rehmann, K. Meerholz
€
and S. Hoger, Synth. Met., 2010, 160, 691.
14 (a) G. Eglinton and A. R. Galbraith, Proc. Chem. Soc., London, 1957,
350; (b) O. M. Behr, G. Eglinton and R. A. Raphael, Chem. Ind., 1959,
699; (c) O. M. Behr, G. Eglinton, A. R. Galbraith and R. A. Raphael,
J. Chem. Soc., 1960, 3614; (d) Q. Zhou, P. J. Carroll and
T. M. Swager, J. Org. Chem., 1994, 59, 1249.
€
P. Bauerle, Angew. Chem., 2000, 112, 3623; (b) S. Hoger,
K. Bonrad, A. Mourran, U. Beginn and M. Moller, J. Am. Chem.
€
€
€
Soc., 2001, 123, 5651; (c) D. Borissov, A. Ziegler, S. Hoger and
W. Freyland, Langmuir, 2004, 20, 2781; (d) C. Grave, D. Lentz,
15 (a) S. Ott and R. Faust, Chem. Commun., 2004, 388; (b) T. Nishinaga,
N. Nodera, Y. Miyata and K. Komatsu, J. Org. Chem., 2002, 67,
6091; (c) U. H. F. Bunz and V. Enkelmann, Chem.–Eur. J., 1999, 5,
263.
16 For phthalocyaninodehydroannulenes and subphthalocyaninodehy-
droannulenes see, e.g.: (a) M. J. Cook and M. J. Heeney,
€
€
A. Schafer, P. Samori, J. P. Rabe, P. Franke and A. D. Schluter,
J. Am. Chem. Soc., 2003, 125, 6907; (e) M. Fischer, G. Lieser,
A. Rapp, I. Schnell, W. Mamdouh, S. De Feyter, F. C. De Schryver
€
and S. Hoger, J. Am. Chem. Soc., 2004, 126, 214; (f) V. Kalsani,
H. Ammon, F. Jackel, J. P. Rabe and M. Schmittel, Chem.–Eur. J.,
2004, 10, 5841; (g) A. Ziegler, W. Mamdouh, A. V. Heyen,
M. Surin, H. Uji-i, M. M. S. Abdel-Mottaleb, F. C. De Schryver,
€
ꢀ
Chem.–Eur. J., 2006, 6, 3958; (b) E. M. Garcıa-Frutos,
F. Fernandez-Lazaro, E. M. Maya, P. Vazquez and T. Torres,
ꢀ
ꢀ
ꢀ
€
S. De Feyter, R. Lazzaroni and S. Hoger, Chem. Mater., 2005, 17,
J. Org. Chem., 2000, 65, 6841; (c) R. S. Iglesias, C. G. Claessens,
ꢀ
M. A. Harranz and T. Torres, Org. Lett., 2007, 9, 5381.
5670; (h) S. Furukawa, H. Uji-i, K. Tahara, T. Ichikawa,
M. Sonoda, F. C. De Schryver, Y. Tobe and S. De Feyter, J. Am.
1414 | J. Mater. Chem., 2011, 21, 1404–1415
This journal is ª The Royal Society of Chemistry 2011

View Article Online
17 R. Callahan, K. Marshall, R. Rothchild and K. Rosmarion, Chem.
Educ., 2001, 6, 227.
27 S. A. Jenekhe and J. A. Osaheni, Science, 1994, 265, 765.
28 J. M. Lupton, I. D. W. Samuel, R. Beavington, P. L. Burn and
€
€
18 G. Gaefke and S. Hoger, Synthesis, 2008, 14, 2155.
H. Bassler, Adv. Mater., 2001, 13, 258.
19 (a) C. Glaser, Ber. Dtsch. Chem. Ges., 1869, 2, 422; (b) F. Sondheimer
and R. Wolovsky, J. Am. Chem. Soc., 1962, 84, 260; (c) F. Sondheimer,
Acc. Chem. Res., 1972, 3, 81; (d) D. O’Krongly, S. R. Denmeade,
M. Y. Chiang and R. Breslow, J. Am. Chem. Soc., 1985, 107, 5544;
29 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr, T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi,
V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth,
P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz,
Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov,
G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin,
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaussian 03,
Revision C.02, Gaussian, Inc., Wallingford, CT, 2004.
€
€
(e) K. Kadei and F. Vogtle, Chem. Ber., 1991, 124, 909; (f) S. Hoger
and V. Enkelmann, Angew. Chem., 1995, 107, 2917; (g) P. Siemsen,
R. C. Livingston and F. Diederich, Angew. Chem., 2000, 112, 2740.
1
20 It is interesting to note that the H-NMR signal of the intraannular
proton of 1 is remarkably downfield shifted (by about 0.7 ppm)
compared to the non-cyclic precursors 12 or 15 (14). The
extraannular triphenylene protons ortho to the acetylene
(diacetylene) moiety are also shifted downfield, however, only by
about 0.3 ppm. In the case of a paratropic ring current due to the
formation of a formally (antiaromatic) 28 e-system we would expect
an upfield shift for the latter protons. Therefore, we ascribe the
observed NMR shift to a sterical compression of the protons inside
1 upon cyclization (as shown in the ESI†) (c.f.: B. F. Plummer,
Polycyclic Aromat. Compd., 1994, 4, 25).
21 (a) G. C. McGonical, R. H. Bernhardt and D. J. Thomson, Appl.
Phys. Lett., 1990, 57, 28; (b) J. P. Rabe and S. Buchholz, Science,
1991, 253, 424; (c) K. Morishige, Y. Takami and Y. Yokota, Phys.
Rev. B: Condens. Matter, 1993, 48, 8277.
22 X. H. Qiu, C. Wang, Q. D. Zeng, B. Xu, S. X. Yin, H. N. Wang,
S. D. Xu and C. L. Bai, J. Am. Chem. Soc., 2000, 122, 5550.
23 (a) W. Mamdouh, H. Uji-i, J. S. Ladislaw, A. E. Dulcey, V. Percec,
F. C. De Schryver and S. De Feyter, J. Am. Chem. Soc., 2006, 128,
317; (b) S. De Feyter and F. C. De Schryver, J. Phys. Chem. B,
2005, 109, 4290; (c) J. P. Rabe, S. Buchholz and L. Askadskaya,
Synth. Met., 1993, 54, 339; (d) D. M. Cyr, B. Venkararaman and
G. W. Flynn, Chem. Mater., 1996, 8, 1600; (e) L. Piot, D. Bonifazi
and P. Samori, Adv. Funct. Mater., 2007, 17, 3689.
30 B. D. Allen, A. C. Benniston, A. Harriman, I. Llarena and
C. A. Sams, J. Phys. Chem. A, 2007, 111, 2641.
31 Y. Yamaguchi, Y. Matsubara, T. Ochi, T. Wakamiya and
Z. Yoshida, J. Am. Chem. Soc., 2008, 130, 13867.
€
32 (a) S. Klyatskaya, N. Dingenouts, C. Rosenauer, B. Muller and
S. Hoger, J. Am. Chem. Soc., 2006, 128, 3150; (b) K.-S. Moon,
€
E. Lee and M. Lee, Chem. Commun., 2008, 3061.
33 S. W. Provencer, Comput. Phys. Commun., 1982, 27, 229.
€
34 X. Feng, W. PisulaandK. Mullen, J. Am. Chem. Soc., 2007, 129, 14116.
35 S. Forster and C. Burger, Macromolecules, 1998, 31, 879.
€
€
36 S. Forster, A. Timmann, M. Konrad, C. Schellbach, A. Meyer,
S. S. Funari, P. Mulvaney and R. Knott, J. Phys. Chem. B, 2005,
109, 1347.
37 (a) Y. J. Lin, Q. Cai, Q. F. Li, L. W. Xue, R. G. Yin and X. P. Yang,
J. Appl. Polym. Sci., 2010, 115, 3393; (b) X. L. Feng, W. Pisula,
24 (a) C. A. Palma, M. Bonini, A. Llanes-Pallas, T. Breiner, M. Prato,
ꢀ
D. Bonifazi and P. Samorı, Chem. Commun., 2008, 5289; (b) S. Lei,
K. Tahara, F. C. De Schryver, M. V. d. Auweraer, Y. Tobe and
S. De Feyter, Angew. Chem., 2008, 120, 3006; (c) L. Kampschulte,
T. L. Werblowsky, R. S. K. Kishore, M. Schmittel, W. M. Heckl
and M. Lackinger, J. Am. Chem. Soc., 2008, 130, 8502; (d)
K. Tahara, S. Okuhata, J. Adisoejoso, S. Lei, T. Fujita, S. De
Feyter and Y. Tobe, J. Am. Chem. Soc., 2009, 131, 17583.
25 K. Becker, E. Da Como, J. Feldmann, F. Scheliga, E. T. Csanyi,
S. Tretiak and J. M. Lupton, J. Phys. Chem. B, 2008, 112, 4859.
26 E. Da Como, K. Becker, J. Feldmann and J. M. Lupton, Nano Lett.,
2007, 7, 2993.
€
T. Kudern, D. Q. Wu, L. J. Zhi, S. De Feyter and K. Mullen,
J. Am. Chem. Soc., 2009, 131, 4439; (c) Y. L. Yang and C. Wang,
Curr. Opin. Colloid Interface Sci., 2009, 14, 135.
38 In the bulk the cylinder–cylinder distance of 5.8 nm is smaller
compared to the AFM-images, because on the substrate the alkyl-
chains are nearly completely stretched due to adsorption to the
interface.
39 N. Dingenouts, S. Klyatakaya, S. Rosenfeldt, M. Ballauff and
€
S. Hoger, Macromolecules, 2009, 42, 5900.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 1404–1415 | 1415