Columnar liquid crystalline assembly of doubly discotic supermolecules based on tetra-triphenylene-substituted phthalocyanine

Article abstract of DOI:10.1039/b927347f

A series of doubly discotic supermolecules with four triphenylene (Tp) mesogens attached to a phthalocyanine (Pc) core via flexible alkyl spacers were designed and synthesized. The samples were denoted as Pc(Tp)₄. Two alkyl chain lengths (C5 an

Full text of DOI:10.1039/b927347f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/softmatter | Soft Matter
Columnar liquid crystalline assembly of doubly discotic supermolecules
based on tetra-triphenylene-substituted phthalocyanine†
a
Jianjun Miao and Lei Zhu
ab
*
Received 4th January 2010, Accepted 8th February 2010
First published as an Advance Article on the web 29th March 2010
DOI: 10.1039/b927347f
A series of doubly discotic supermolecules with four triphenylene (Tp) mesogens attached to
a phthalocyanine (Pc) core via flexible alkyl spacers were designed and synthesized. The samples were
denoted as Pc(Tp)4. Two alkyl chain lengths (C5 and C12) in the Tp arms and two spacer lengths
(C6 and C10) were used to study columnar liquid crystalline (LC) self-assemblies in Pc(Tp)4 samples.
The mesophase morphology was studied by polarized light microscopy and X-ray diffraction
techniques. When the spacer length was relatively long (C10), a normal hexagonal columnar structure
with the unit cell dimension close to those for the parent Tp or Pc liquid crystals was observed. The Pc
columns packed either randomly or in a hexagonal superlattice in the Tp matrix. When the spacer
length was short (C6), the Tp arms and the Pc core were tightly coupled together, and thus the whole
supermolecules stacked together to form a super-LC column. A column-in-column rectangular
mesophase with unit cell dimensions close to the size of the entire supermolecule was observed.
transition from rectangular columnar for the first generation to
Introduction
hexagonal columnar for the 2–5 generations was observed, and
the discotic Tp mesogens and the dendrimer core microphase
separated into individual columns.
Supermolecules, which are different from polymers, have well-
defined structures and multiple functions. They represent an
important class of advanced functional materials, which often
self-assemble into hierarchical structures from nano- to micro-
scales.1,2 An attractive category of supermolecules is liquid
crystalline (LC) dendrimers or polypedes, because they have
potential for optoelectronic and biological applications.3–8
Several molecular parameters play an important role in the self-
assembly of LC polypedes. They include the number density of
the LC mesogens at periphery, the preferred orientation of
mesogenic units, the spacer length (or the degree of coupling)
between the core and the mesogenic shell, and the rigidity of the
core. Generally, with increasing the number density of the LC
mesogens at periphery, decreasing the spacer length between the
core and the mesogenic shell, and increasing the rigidity of the
core, supramolecular self-assembly morphologies tend to change
from calamitic to columnar, and finally to cubic (or spherical).1
Currently, a large amount of work on LC polypedes has
focused on calamitic LC mesogens, and both calamitic and
columnar mesophase morphologies have been reported.9–36
However, fewer reports are concentrated on discotic LC poly-
pedes, where discotic mesogens are covalently linked to the
periphery of a flexible or rigid central scaffold. In a recent study,
4–64 triphenylene (Tp) mesogens were covalently linked to
a flexible dendrimer core, polypropyleneimine.37 A structural
In addition to flexible central scaffolds, rigid organic and
inorganic molecules were also used as the central cores in discotic
LC polypedes. Rigid organic central cores include a simple
benzene ring,38–41 an n-type anthraquinone,42,43 a p-type Tp,44,45
and a polyaromatic hexa-peri-hexabenzocoronene (HBC),46 to
which a defined number (mostly 6 and sometimes 2, 3, or 4) of
Tp,39–45 pentaalkyne,38 or HBC46 mesogens were attached via
ester, siloxane, or ether bonds. Generally, a normal hexagonal
columnar phase with a unit cell dimension close to that of the
parent Tp or HBC discotic LCs was observed,40–43,45,46 suggesting
that the peripheral Tp or HBC mesogens self-organized into
individual LC columns regardless of the covalent linkages to the
central scaffold. However, with more detailed studies,40,41,45
a disordered superlattice, with the unit cell dimension equivalent
to the size of an entire discotic LC supermolecule, was observed
due to the packing of the central scaffolds. Furthermore, a broad
temperature range for the columnar LC phases was obtained as
compared to the parent Tp LCs.40,41,45 When three flat radial
pentaalkynes were connected to a central benzene core via C11
spacers, columnar nematic phases, instead of a discotic nematic
phase as found in the parent pentaalkyne molecules, were
observed.38
However, discotic LC polypedes with chemically different
peripheral and central polyaromatic/macrocyclic discogens are
rarely studied. In this report, we covalently attached four Tp
mesogens to the four corners of a central phthalocyanine (Pc)
core via alkyl spacers with different lengths, forming a star-sha-
ped doubly discotic LC polypede, Pc(Tp)4. Different from
previous reports on 1,3,5-tritriphenylene benzenetrisamides,40,41
where helical conformation in the hydrogen bonds along the
columns was crucial to their supramolecular self-assembly,
Pc(Tp)4 supermolecules exhibited a reverse trend in the spacer
aPolymer Program, Institute of Materials Science and Department of
Chemical, Materials and Biomolecular Engineering, University of
Connecticut, Storrs, Connecticut, 06269-3136, USA
bDepartment of Macromolecular Science and Engineering, Case Western
Reserve University, Cleveland, Ohio, 44106-7202, USA. E-mail:
lxz121@case.edu
† Electronic Supplementary Information (ESI) available: Tables of
calculated and experimental d-spacings for samples 1–4; MALDI-TOF
MS spectra for samples 1–4. See DOI: 10.1039/b927347f/
2072 | Soft Matter, 2010, 6, 2072–2079
This journal is ª The Royal Society of Chemistry 2010

View Article Online
length-determined self-organization. When the spacer length was
relatively long, a normal hexagonal columnar structure (with the
unit cell dimension close to those for the parent Tp or Pc LCs)
was observed, and the Pc columns packed randomly among the
hexagonal Tp columns. When the spacer length was short, the Tp
arms and the Pc core were tightly coupled together, and the
whole Pc(Tp)4 supermolecules stacked together to form a super-
column. A column-in-column rectangular mesophase with unit
cell dimensions close to the size of the entire supermolecule was
observed.
method for Pc syntheses.47,48 A detailed synthetic route is shown
in Scheme 1. Tp phthalonitriles were synthesized from the
coupling of monobromoalkoxy-triphenylene with 4-hydroxy-
1,2-benzenedicarbonitrile according to the literature with certain
modifications.49 In brief, 1.0 g of monobromoalkoxytriphenylene
was added into a solution of 4-hydroxy-1,2-benzenedicarboni-
trile preactivated by K2CO3 in DMF, and the mixture was heated
at 100 ꢀC overnight. The crude product was washed with doubly
distilled water and extracted with dichloromethane three times.
The combined organic phase was dried with anhydrous Na2SO4
and the solvent was evaporated using a rotavapor. The product
was further dried in a vacuum oven to afford Tp phthalonitrile
(yield 83–92%). 0.40 g of Tp phthalonitrile (0.44 mmol for 1,
0.42 mmol for 2, 0.29 mmol for 3, and 0.28 mmol for 4), 0.02 g
(0.15 mmol) of CuCl2, 5 mL of 1-pentanol, and four drops of
DBU were mixed together and refluxed for 24 h. Afterwards,
excess methanol was added to the reaction mixture to precipitate
the target compound. The precipitate was isolated by centrifu-
gation and washed with methanol. The crude product was
further purified by column chromatography (silica gel,
CHCl3:hexane ¼ 2:1 for 1 and 2, CHCl3:hexane ¼ 1:1 for 3
and 4) and recrystallized from acetone twice to give green to blue
solids with yields: 69% for 1, 63% for 2, 19% for 3, and 20% for 4.
Because of the paramagnetic nature of Cu2+ in the Pc core,
quantitative molecular analysis by nuclear magnetic resonance
spectroscopy could not be performed. Instead, high resolution
Experimental
Materials
Monobromoalkoxytriphenylene was synthesized according to
the literature with certain modifications.40 4-Hydroxy-1,2-ben-
zenedicarbonitrile was purchased from TCI America. 1,8-Dia-
zabicyclo[5.4.0]undec-7-ene (DBU) was purchased from
Sigma-Aldrich. All solvents were purchased from Fisher Scien-
tific and used without further purification.
General synthesis procedure for Pc(Tp)4
The synthesis of novel doubly discotic supermolecules based on
a Pc core and four Tp arms was achieved from cyclo-
tetramerization of the precursor dinitriles, which is a common
Scheme 1 Synthesis of Pc(Tp)4 supermolecules with different spacer and alkyl chain lengths.
This journal is ª The Royal Society of Chemistry 2010
Soft Matter, 2010, 6, 2072–2079 | 2073

View Article Online
matrix-assisted laser desorption ionization–time of flight
(MALDI-TOF) mass spectrometry (MS) – was performed using
dithranol as the matrix. MALDI-TOF MS spectra are shown in
Fig. S1–S4†. Sample 1: C228H304CuN8O28, calculated 3668.4 and
found 3666.2. Sample 2: C244H336CuN8O28, calculated 3892.8
and found 3890.6. Sample 3: C368H584CuN8O28, calculated
5632.2 and found 5633.0. Sample 4: C384H616CuN8O28, calcu-
lated 5856.6 and found 5856.9. As we can see, the calculated
values were consistent with the experimental results within
experimental errors. The purity of all samples was checked by
thin layer chromatography and size exclusion chromatography
(SEC, see Fig. 1). Sharp unimodal peaks with polydispersity
indices between 1.01–1.03 were obtained using polystyrene as the
calibration standard.
an Instec HCS410 hot stage. Fluorescence spectroscopy was
performed on a Jobin-Yvon Spex Fluorolog 3-211 spectrofluo-
rometer. Both solution and thin film samples were excited
between 250 and 300 nm, respectively.
Two-dimensional (2D) X-ray diffraction (XRD) experiments
were performed at the synchrotron X-ray beamline X27C at
National Synchrotron Light Source, Brookhaven National
Laboratory. The wavelength of incident X-ray was 0.1371 nm.
The scattering angle was calibrated using silver behenate with the
primary reflection peak at the scattering vector q ¼ (4psinq)/l ¼
1.076 nmꢁ1, where q is the half-scattering angle and l is the
wavelength. Fuji imaging plates were used as detectors for XRD
experiments, and digital images were obtained using a Fuji BAS-
2500 scanner. The typical data requisition time was 1 min. An
Instec HCS410 hot stage equipped with a liquid-nitrogen cooling
accessory was used for temperature-dependant X-ray experi-
ments. One-dimensional (1D) XRD curves were obtained by
integration of the corresponding 2D XRD patterns.
Instrumentation and characterization
Differential scanning calorimetry (DSC) experiments were
carried out on a TA DSC-Q100 instrument. An indium standard
was used for both temperature and enthalpy calibration.
Approximately 3 mg sample was used for the DSC study and the
scanning rate was 10 ꢀC minꢁ1. SEC was performed on a Viscotek
GPCmax VE2001 with quadruple detectors [differential refrac-
tive index (RI), UV-vis, viscometer, and light scattering, but only
the differential RI detector was used in this study]. THF was used
as the solvent and polystyrene standards were used for calibra-
tion. The solvent flow rate was 1.0 mL minꢁ1. MALDI-TOF MS
were carried out on a Bruker Ultraflex III MALDI-TOF/TOF
instrument. Polarized light microscopy (PLM) experiments were
performed using an Olympus BX51P microscope equipped with
Results and discussion
Thermal behavior and phase morphology studied by DSC and
PLM
LC phase transitions in Pc(Tp)4 samples 1–4 were studied by
DSC (Fig. 2). Upon the first cooling, the sample 1 showed
a single transition at 151 ꢀC (24.5 kJ molꢁ1). Upon second
heating, it showed a single isotropization temperature (Ti) at
173 ꢀC (24.9 kJ molꢁ1). In the PLM micrograph for the sample 1
at 165 ꢀC in Fig. 3A, a branched leaf-like texture, which is typical
for hexagonal columnar phases, was observed. When the spacer
Fig. 1 Size exclusion chromatography (SEC) curves for Pc(Tp)4 samples
1–4. A differential refractive index detector was used.
Fig. 2 Differential scanning calorimetry (DSC) first cooling and second
heating curves for Pc(Tp)4 samples 1–4. The scanning rate is 10 ꢀC minꢁ1
.
2074 | Soft Matter, 2010, 6, 2072–2079
This journal is ª The Royal Society of Chemistry 2010

View Article Online
phase. On the basis of a previous study,50 the broad low
temperature transition could be attributed to the C12 alkyl
chain-induced crystallization of entire LC supermolecules. Upon
heating, the crystalline phase melted with a broad peak at 9.2 ꢀC
and a sharp peak at 37.5 ꢀC (total heat of fusion of 152 kJ molꢁ1),
respectively. Finally, the Ti was observed at 166 ꢀC (15 kJ molꢁ1).
The PLM result in Fig. 3C showed fan-like texture at 160 ꢀC for
the sample 3.
Similarly, upon cooling, sample 4 showed a LC formation
ꢀ
peak at 56.5 C (13.4 kJ molꢁ1) and a relatively broad crystalli-
zation peak at 5.8 C (138 kJ molꢁ1). Upon heating, it showed
ꢀ
a broad crystal melting peak at 27.2 ꢀC (148 kJ molꢁ1) and a Ti at
70.7 ꢀC (8.8 kJ molꢁ1), respectively. Again, the high temperature
(e.g., 50 ꢀC) phase should be LC and the low temperature
(e.g., 0 ꢀC) phase should be crystalline. The PLM micrograph for
the sample 4 at room temperature showed a similar texture as
that of the sample 2 with even smaller grains, and kept
unchanged until isotropization (see Fig. 3D). Thermal behaviors
and phase transitions in Pc(Tp)4 samples 1–4 are summarized in
Table 1.
Fig. 3 PLM micrographs of Pc(Tp)4 samples (A) 1 at 165 ꢀC, (B) 2 at
room temperature, (C) 3 at 160 ꢀC, and (D) 4 at room temperature.
length was increased to C10, the sample 2 only showed a glass
transition temperature (Tg) at ꢁ21.2 ꢀC upon cooling. Upon
heating, a weak isotropization peak was observed at 95.7 ꢀC
(8.0 kJ molꢁ1) above the Tg (ꢁ11.2 ꢀC), which could be attributed
to a typical monotropic phase behavior. Obviously, an increase
in the spacer length from C6 to C10 caused the Ti to decrease by
77 ꢀC. Due to the very low liquid crystallinity in the sample 2, its
PLM texture in Fig. 3B only showed an irregular texture with
small grains at room temperature, no matter how long it was
annealed below the Ti.
Liquid crystalline self-assemblies studied by XRD
The mesophase structures in these self-assembled Pc(Tp)4
supermolecules were studied by temperature-dependent 1D and
2D XRD techniques. Fig. 4A shows the 1D XRD profiles at
various temperatures for the sample 1 during the second heating
ꢀ
process after cooling from the isotropic melt. At 25 C, a sharp
reflection peak appeared at q ¼ 3.82 nmꢁ1, together with three
higher order reflections, which had
a q-relationship of
Unlike samples 1 and 2, samples 3 and 4 showed richer mes-
ophase behaviors with two major transitions (see Fig. 2). For
example, upon cooling, sample 3 showed a transition peak at
147 ꢀC (12.1 kJ molꢁ1) and a broad peak at 6.4 ꢀC (135 kJ molꢁ1),
respectively. Judging from the heat of transition, the high
1:O3:O4:O7. This is typical for hexagonal columnar LCs
(see Table S1†). The distance between neighbouring columns was
a ¼ 1.90 nm. Surprisingly, this length was much smaller than the
size of the entire supermolecule (ꢂ5.2 nm), but was close to the
size of either C5-Tp (ꢂ1.7 nm)51 or C6-Pc (ꢂ2.65 nm).52 At an
even lower q, a broad halo was observed at 1.466 nmꢁ1, which
corresponded to a d-spacing of 4.3 nm and was close to the size of
the entire supermolecule with a flat conformation. In the wide-
angle region, a reflection peak was located at 18.0 nmꢁ1
(d-spacing ¼ 0.35 nm), representing an ordered p–p stacking of
the Tp53 and/or Pc52 disks in the LC columns. Fig. 4B shows the
2D XRD pattern for the shear-oriented sample 1. The interdisk
ꢀ
temperature (e.g., 160 C) phase should be a LC phase and the
low temperature (e.g., ꢁ25 ꢀC) phase should be a crystalline
Table 1 Phase transition temperatures (ꢀC, above arrow) and heats of
transition (kJ molꢁ1, below arrow) for samples 1–4a
Sample First cooling
Second heating
151
ƒƒƒƒƒƒ!
173
ƒƒƒƒƒƒ!
195
ƒƒƒƒƒƒ!
1
2
3
I
I
I
I
Col
Colho
Col
I
ho
hd
24:5
24:9
ꢁ21:2
ꢁ11:2
95:7
g
g
Col
I
ƒƒƒƒƒƒ!
ƒƒƒƒƒƒ!
37:5
ƒƒƒƒƒƒ!
166
ho
8:0
147
ƒƒƒƒƒƒ!
12:1
6:0
ƒƒƒƒƒƒ!
Col
Col
Cr Cr
Col
Col
I
ƒƒƒƒƒƒ!
27:2
ƒƒƒƒƒƒ!
148
ƒƒƒƒƒƒ!
ro
ro
7:8
15
56:5
ƒƒƒƒƒƒ!
13:4
5:8
ƒƒƒƒƒƒ!
138
70:7
ƒƒƒƒƒƒ!
8:8
4
Cr Cr
I
oo
oo
a
Cr ¼ crystal; g ¼ glass; Colho ¼ordered hexagonal columnar phase;
Colhd
¼
disordered hexagonal columnar phase; Colro
¼
ordered
Fig. 4 (A) Temperature-dependent 1D XRD profiles and (B) 2D XRD
pattern onshear-oriented Pc(Tp)4 sample1 at 25 ꢀC. The insetin (A) shows
rectangular columnar phase; Coloo ¼ ordered oblique columnar phase;
I ¼ isotropic.
an additional magnified (ꢃ10) profile at 25 ꢀC between 6 and 12 nmꢁ1
.
This journal is ª The Royal Society of Chemistry 2010
Soft Matter, 2010, 6, 2072–2079 | 2075

View Article Online
reflection at 18.0 nmꢁ1 (0.35 nm) oriented parallel to the shear
direction, while the strong hexagonal reflection at 3.82 nmꢁ1
(1.90 nm) oriented orthogonal to the shear direction. This XRD
pattern indicated that the columns conformed to the shear
direction and the Tp and Pc disks oriented perpendicular to the
column axes.
These microphase-separated Tp and Pc columns further self-
assembled into a mixed hexagonal columnar phase, as shown in
the left panel of Fig. 5. Assuming the core diameters of Tp and Pc
were 0.81 and 1.32 nm and the interdisk (both Tp and Pc)
distance was 0.35 nm, the electron densities for the Tp and
Pc columns were 649 and 616 electrons nmꢁ3, respectively. These
close electron densities resulted in the fact that the incident X-ray
could not clearly distinguish between the Tp and Pc columns.
Therefore, the major reflection was seen at 3.82 nmꢁ1 and thus
the hexagonal lattice was small (a ¼ 1.9 nm, see the left panel in
Fig. 5). However, the average distance among Pc columns could
still be evidenced by the halo at 1.466 nmꢁ1 because of a small
electron density difference (5%) between Tp and Pc columns.
This halo at low angles suggested that the packing of Pc columns
among the Tp column were random (see the left panel in Fig. 5).
This result is similar to the XRD observations in previous reports
for 1,3,5-tristriphenylene benzenetrisamide supermolecules with
relatively short spacers.40,41 However, we prefer not to use the
term ‘‘superlattice’’ as in ref. 40 and 41, because no lattice existed
among randomly packed Pc columns.
At 173 ꢀC, the hexagonal columnar phase melted, as evidenced
by the disappearance of the interdisk reflection at 17.4 nmꢁ1 at
145 ꢀC. Meanwhile, the sharp reflection at 3.82 nmꢁ1 trans-
formed into a broad halo at 3.92 nmꢁ1, representing the average
distance (1.6 nm) among randomly packed Tp and Pc discotic
molecules. Intriguingly, immediately after isotropization at
173 ꢀC, another sharp reflection peak at 1.44 nmꢁ1 developed
from the halo at 1.466 nmꢁ1, together with a higher order
reflection at 2.45 nmꢁ1. The q-relationship for these two reflec-
tions was 1:O3, suggesting a super-hexagonal lattice with a larger
unit cell dimension of a ¼ 5.04 nm (see Table S2†), which was
nearly the same as the overall molecular size of the sample 1.
Because no interdisk reflection at 17.4 nmꢁ1 was observed at
This peculiar XRD profile for the sample 1 could be explained
by the very close electron densities between Tp and Pc columns in
the bulk. First of all, we need to prove that Tp arms and Pc cores
self-organized into microphase separated columns, instead of
mixing together in one LC column. If Tp and Pc mixed together
in a LC column, Foster energy transfer54 would be facilitated by
€
the excimer formation and the fluorescence spectrum would shift
to much higher wavelengths.55 Fig. 6 shows fluorescence spectra
for C5-Tp in CH2Cl2, sample 1 in CH2Cl2, and sample 1 thin film
excited at 250 nm, respectively. As we can see, the fluorescence
spectra for C5-Tp and sample 1 in CH2Cl2 were nearly the same,
with the main peak centred at 383 nm. For the sample 1 thin film,
a broad fluorescence band for the Tp moiety appeared at 391 nm,
almost the same as those for C5-Tp and sample 1 in CH2Cl2. This
result suggested that the energy absorbed by Tp did not transfer
to Pc. Therefore, Tp and Pc should form microphase separated
LC columnar domains in the solid state.
ꢀ
173 C, this phase should be determined as a disordered super-
hexagonal lattice, and the corresponding phase transformation is
depicted in Fig. 5. Based on temperature-dependent XRD
experiments, this disordered super-hexagonal lattice persisted
until 195 ꢀC and completely disappeared above 200 ꢀC. However,
this transition involꢀved too little heat to be detected by DSC
(see Fig. 2). At 220 C, three halos were observed at 1.61, 3.88,
and 13.13 nmꢁ1, respectively, representing average distances
Fig. 5 Schematic representation of the phase transformation from
a normal hexagonal lattice into a super-hexagonal lattice for the sample 1
upon melting at 173 ꢀC. The grey circles represent triphenylene columns
and the black circles represent phthalocyanine columns.
Fig. 7 (A) Temperature-dependent 1D XRD profiles and (B) 2D XRD
pattern onshear-oriented Pc(Tp)4 sample2 at 25 ꢀC. The insetin (A) shows
Fig. 6 Fluorescence spectra for C5Tp in CH2Cl2, sample 1 in CH2Cl2,
and sample 1 thin film excited at 250 nm, respectively.
an additional magnified (ꢃ10) profile at 50 ꢀC between 5 and 10 nmꢁ1
.
2076 | Soft Matter, 2010, 6, 2072–2079
This journal is ª The Royal Society of Chemistry 2010

View Article Online
among randomly packed Pc (3.90 nm), Tp (1.62 nm), and alkyl
chains (0.48 nm) in the isotropic melt.
0.35 nm. In the 2D XRD pattern in Fig. 8B, the (10)/(01)
reflection oriented orthogonal to the shear direction, while the
interdisk reflection was parallel to the shear direction. Again, this
indicated that the Tp and Pc disks were perpendicular to the
column axꢀes.
Temperature-dependent 1D XRD profiles for the sample 2
during the second heating process are shown in Fig. 7A. A
ꢀ
similar phase behavior as for sample 1 was observed. At 50 C,
reflections from a normal hexagonal columnar lattice were
observed at 3.56 and 6.15 nmꢁ1, respectively, with a q-relation-
ship being 1:O3 (see Table S3†). The unit cell dimension for this
small hexagonal lattice was a ¼ 2.04 nm, slightly larger than that
(a ¼ 1.90 nm) for sample 1 because of an increase in the spacer
length from C6 to C10. Again, the interdisk distance was
observed at 17.6 nmꢁ1 (0.36 nm) in the wide-angle region. From
the 2D XRD pattern in Fig. 6B, the same orientation of the
molecular disks with respect to the column axes as in sample 1
was observed; the Tp and Pc disks were perpendicular to the LC
column axes. Differing from sample 1, no disordered super-
hexagonal columnar phase was observed for the sample 2 upon
isotropization at 95.7 ꢀC. Three halos were observed at 1.36,
3.56, and 13.10 nmꢁ1, respectively, representing average distances
among randomly packed Pc (4.62 nm), Tp (1.76 nm), and alkyl
chains (0.48 nm). The difference between samples 1 and 2 could
be attributed to the longer spacer length in the sample 2.
At ꢁ50 C, entire supermolecules of the sample 4 crystallized
due to the crystallization of C12 alkyl chains. However, the
major feature of the oblique columnar packing did not change, as
ꢀ
seen in Fig. 8A. After heating to 125 C, the oblique columnar
phase disappeared. Three halos were observed at 1.39, 2.93, and
13.32 nmꢁ1, corresponding to average distances among randomly
packed Pc (4.53 nm), Tp (2.15 nm), and alkyl chains (0.47 nm),
respectively.
The sample 3 displayed a completely different mesophase
structure from those of samples 1, 2, and 4. This is evidenced
from the temperature-dependent 1D XRD profiles during the
The temperature-dependent 1D XRD profiles during the
second heating process for sample 4 are shown in Fig. 8A. At
50 ꢀC, a sharp reflection was seen at 2.89 nmꢁ1 with several higher
order reflections in the low-angle region. Again, this reflection
corresponded to a d-spacing of 2.17 nm, which was close to the
sizes of C12-Tp (ca. 2.5 nm)56 and C10-Pc (ca. 3.15 nm).52
However, the higher order reflections could not fit into a hexag-
onal lattice as in samples 1 and 2. Instead, they fitted to an
oblique lattice with unit cell dimensions: a ¼ b ¼ 2.19 nm and
g ¼ 97ꢀ (see Table S4†), and the corresponding Miller indices
were labelled in Fig. 8A. This result indicated that Tp and Pc
columns assembled (or mixed) into an oblique lattice, rather than
a hexagonal lattice as shown in the left panel of Fig. 5.
Comparing with sample 2, this difference is clearly a result of an
increased alkyl arm length (C12) for the sample 4. At even lower
q values, a broad halo was seen at 1.31 nmꢁ1, corresponding to
the average distance (4.80 nm) among randomly packed Pc
columns. Note that this distance was close to the entire super-
molecule size. In the wide-angle region, the interdisk reflection
was seen at 17. 9 nmꢁ1, which corresponded to a d-spacing of
Fig. 9 (A) Temperature-dependent 1D XRD profiles and (B) 2D XRD
pattern on the shear-oriented Pc(Tp)4 sample 3 at 100 ꢀC.
Fig. 10 Schematic representations of the supramolecular self-assembly
of Pc(Tp)4 supermolecules 1–4 with short and long spacers, respectively.
2D unit cells are shown with dotted lines.
Fig. 8 (A) Temperature-dependent 1D XRD profiles and (B) 2D XRD
pattern on shear-oriented Pc(Tp)4 sample 4 at 40 ꢀC.
This journal is ª The Royal Society of Chemistry 2010
Soft Matter, 2010, 6, 2072–2079 | 2077

View Article Online
second heating in Fig. 9A. At 50 and 130 ꢀC, many sharp
reflections were observed in the low-angle region. To solve the
LC structure, a 2D XRD experiment was performed on the
shear-oriented sample 3, and the results are shown in Fig. 9B. In
this 2D XRD pattern, all low-angle reflections were oriented on
the equator, suggesting a 2D columnar LC phase in the sample.
Both the 1D profile and 2D patterns could be fitted with a 2D
rectangular symmetry with unit cell dimensions: a ¼ 5.03 nm and
b ¼ 4.30 nm (see Table S5†), and Miller indices were assigned in
both figures. Note that these dimensions were close to the size
(ꢂ5 nm) of the entire supermolecules. This result is similar to
those observed for 1,3,5-tristriphenylene benzenetrisamide
supermolecules with relatively long spacers,40,41 where rectan-
gular and oblique columnar phases were observed. At high
angles, the interdisk reflection was seen at 17.7 nmꢁ1 (0.35 nm) in
the 1D profile and on the meridian in the 2D pattern, which was
perpendicular to the (10) reflection on the equator. Therefore, the
Tp and Pc disks again were oriented perpendicular to the LC
columns.
At ꢁ10 ꢀC, entire supermolecules crystallized as induced by
the C12 alkyl chain crystallization. Although XRD line widths
slightly increased, the fundamental rectangular structure was
preserved in the sample. After isotropization at 170 ꢀC, three
broad halos were observed at 1.54, 3.25, and 13.0 nmꢁ1, corre-
sponding to average distances among randomly packed Pc
(4.08 nm), Tp (1.93 nm), and alkyl chains (0.48 nm), respectively.
Fig. 10 summarizes the supramolecular self-assembly of
Pc(Tp)4 supermolecules with short and long spacers, respectively.
When the spacer (C6) is shorter than the Tp C12 alkyl arms, the
Tp arms are tightly attached to the Pc core, and thus the
supermolecules adopt a disk-like overall molecular shape when
they stack parallel together. Finally, a column-in-column nano-
structure is obtained; four Tp columns inside a super-column
formed by the entire supermolecules (see the top panel of
Fig. 10). When the spacer is no shorter than the alkyl arms in the
Tps such as in samples 1, 2, and 4, Tp arms and Pc cores are
decoupled to a great extent. Therefore, ordered hexagonal or
rectangular structures with a random arrangement of the Pc
cores are obtained (see the bottom panel of Fig. 10).
oblique (or rectangular) columnar phases with unit cell dimen-
sions close to the entire supermolecular sizes were observed. This
is evidenced by the disappearance of the ordered interdisk
reflections around 2q ꢂ 25ꢀ for long spacer length samples
(see XRD profiles for T-5-Hex and T-6-Hex samples in Fig. 3
of ref. 41).
Conclusions
In summary, a series of doubly discotic Pc(Tp)4 supermolecules
were designed and successfully synthesized. Columnar LC phases
were observed for all samples. PLM and XRD techniques were
employed to study the mesophase morphology. When the spacer
length was relatively long, a normal hexagonal columnar struc-
ture with the unit cell dimension close to those for the parent Tp
or Pc LCs were observed for samples 1, 2, and 4, and the Pc
columns packed randomly among the hexagonal Tp columns
(see the left panel in Fig. 5). Intriguingly, a hexagonal Pc
superlattice was formed immediately after the isotropization of
the Tp columns in the sample 1 at 173 ꢀC (see the right panel in
Fig. 5). When the spacer length was short, the Tp arms and the Pc
core were tightly coupled together, and the whole Pc(Tp)4
supermolecules stack together to form
a super-column.
A
column-in-column rectangular mesophase with unit cell
dimensions close to the size of the entire supermolecule was
observed.
Acknowledgements
This work was supported by NSF CAREER Award DMR-
0348724, DuPont Young Professor Grant, and 3M Nontenured
Faculty Award. The synchrotron X-ray experiments were carried
out at the National Synchrotron Light Source, Brookhaven
National Laboratory, supported by the U.S. Department of
Energy. Assistance from Drs. Lixia Rong, Jie Zhu, and Prof.
Benjamin Hsiao at State University of New York at Stony Brook
for synchrotron X-ray experiments is greatly acknowledged.
References
Our observations seem to be contradictory to those observed
for hydrogen-bond stabilized 1,3,5-tristriphenylene benzene-
trisamide supermolecules,40,41 where normal hexagonal columnar
phases were observed for short spacers (C3 and C4), while
oblique or rectangular superlattice columnar phases were
observed for long spacers (C5 and C6) as compared to the alkyl
chain length (C6) in the Tp arms. We speculate that the helical
conformation of the hydrogen-bonds in the 1,3,5-benzene-
trisamides may be sensitive to the length spacers between the Tp
arms and the benzene core. When the spacer length is relatively
short (C3 and C4), the helical conformation in hydrogen bonds
may be quite open and thus Tp arms could stack together to form
LC columns. This is evidenced by the ordered interdisk reflec-
tions at 2q ꢂ 25ꢀ for short spacer length samples (see XRD
profiles for T-3-Hex and T-4-Hex samples in Fig. 3 of ref. 41).
When the spacer length (C5 and C6) is comparable to the alkyl
chain length (C6) in the Tp, well-defined helical conformation in
the hydrogen bonds prevents the ordered p–p stacking of the Tp
arms. Instead, the whole supermolecules stack together with
helical hydrogen bonds to form a super-column, and thus
1 I. M. Saez and J. W. Goodby, J. Mater. Chem., 2005, 15, 26–40.
2 T. Kato, N. Mizoshita and K. Kishimoto, Angew. Chem., Int. Ed.,
2006, 45, 38–68.
3 V. Percec, M. Glodde, T. K. Bera, Y. Miura, I. Shiyanovskaya,
K. D. Singer, V. S. K. Balagurusamy, P. A. Heiney, I. Schnell,
A. Rapp, H. W. Spiess, S. D. Hudson and H. Duan, Nature, 2002,
419, 384–387.
4 V. Percec, A. E. Dulcey, V. S. K. Balagurusamy, Y. Miura,
J. Smidrkal, M. Peterca, S. Nummelin, U. Edlund, S. D. Hudson,
P. A. Heiney, H. Duan, S. N. Magonov and S. A. Vinogradov,
Nature, 2004, 430, 764–768.
5 V. Percec, A. E. Dulcey, M. Peterca, M. Ilies, M. J. Sienkowska and
P. A. Heiney, J. Am. Chem. Soc., 2005, 127, 17902–17909.
6 V. Percec, M. Glodde, M. Peterca, A. Rapp, I. Schnell, H. W. Spiess,
T. K. Bera, Y. Miura, V. S. K. Balagurusamy, E. Aqad and
P. A. Heiney, Chem.–Eur. J., 2006, 12, 6298–6314.
7 V. Percec, A. E. Dulcey, M. Peterca, M. Ilies, S. Nummelin,
M. J. Sienkowska and P. A. Heiney, Proc. Natl. Acad. Sci. U. S. A.,
2006, 103, 2518–2523.
8 M. Peterca, V. Percec, A. E. Dulcey, S. Nummelin, S. Korey, M. Ilies
and P. A. Heiney, J. Am. Chem. Soc., 2006, 128, 6713–6720.
9 V. Percec, P. W. Chu, G. Ungar and J. Zhou, J. Am. Chem. Soc., 1995,
117, 11441–11454.
10 K. Lorenz, D. Holter, B. Stuhn, R. Mulhaupt and H. Frey, Adv.
Mater., 1996, 8, 414–416.
2078 | Soft Matter, 2010, 6, 2072–2079
This journal is ª The Royal Society of Chemistry 2010

View Article Online
11 M. Baars, S. H. M. Sontjens, H. M. Fischer, H. W. I. Peerlings and
E. W. Meijer, Chem.–Eur. J., 1998, 4, 2456–2466.
12 J. Barbera, M. Marcos and J. L. Serrano, Chem.–Eur. J., 1999, 5,
1834–1840.
13 M. Marcos, R. Gimenez, J. L. Serrano, B. Donnio, B. Heinrich and
D. Guillon, Chem.–Eur. J., 2001, 7, 1006–1013.
34 P. K. Karahaliou, P. H. J. Kouwer, T. Meyer, G. H. Mehl and
D. J. Photinos, Soft Matter, 2007, 3, 857–865.
35 P. K. Karahaliou, P. H. J. Kouwer, T. Meyer, G. H. Mehl and
D. J. Photinos, J. Phys. Chem. B, 2008, 112, 6550–6556.
36 Q. Pan, X. Chen, X. Fan, Z. Shen and Q. Zhou, J. Mater. Chem.,
2008, 18, 3481–3488.
14 B. Donnio, J. Barbera, R. Gimenez, D. Guillon, M. Marcos and
J. L. Serrano, Macromolecules, 2002, 35, 370–381.
15 S. A. Ponomarenko, N. I. Boiko, V. P. Shibaev, R. M. Richardson,
I. J. Whitehouse, E. A. Rebrov and A. M. Muzafarov,
Macromolecules, 2000, 33, 5549–5558.
16 J. M. Rueff, J. Barbera, M. Marcos, A. Omenat, R. Martin-Rapun,
B. Donnio, D. Guillon and J. L. Serrano, Chem. Mater., 2006, 18,
249–254.
17 J. H. Cameron, A. Facher, G. Lattermann and S. Diele, Adv. Mater.,
1997, 9, 398–403.
18 K. U. Jeong, A. J. Jing, B. Mansdorf, M. J. Graham, D. K. Yang,
F. W. Harris and S. Z. D. Cheng, Chem. Mater., 2007, 19, 2921–2923.
19 C. T. Imrie, Z. Lu, S. J. Picken and Z. Yildirim, Chem. Commun.,
2007, 1245–1247.
37 M. D. McKenna, J. Barbera, M. Marcos and J. L. Serrano, J. Am.
Chem. Soc., 2005, 127, 619–625.
38 A. Grafe, D. Janietz, T. Frese and J. H. Wendorff, Chem. Mater.,
2005, 17, 4979–4984.
39 N. C. Maliszewskyj, P. A. Heiney, J. Y. Josefowicz, T. Plesnivy,
H. Ringsdorf and P. Schuhmacher, Langmuir, 1995, 11, 1666–1674.
40 I. Paraschiv, M. Giesbers, B. van Lagen, F. C. Grozema,
R. D. Abellon, L. D. A. Siebbeles, A. T. M. Marcelis, H. Zuilhof
and E. J. R. Sudholter, Chem. Mater., 2006, 18, 968–974.
41 I. Paraschiv, K. de Lange, M. Giesbers, B. van Lagen, F. C. Grozema,
R. D. Abellon, L. D. A. Siebbeles, E. J. R. Sudholter, H. Zuilhof and
A. T. M. Marcelis, J. Mater. Chem., 2008, 18, 5475–5481.
42 H. K. Bisoyi and S. Kumar, Tetrahedron Lett., 2008, 49, 3628–3631.
43 S. K. Gupta, V. A. Raghunathan and S. Kumar, New J. Chem., 2009,
33, 112–118.
20 A. M. Levelut and K. Moriya, Liq. Cryst., 1996, 20, 119–124.
21 K. Moriya, T. Suzuki, S. Yano and S. Miyajima, J. Phys. Chem. B,
2001, 105, 7920–7927.
44 T. Plesnivy, H. Ringsdorf, P. Schuhmacher, U. Nuetz and S. Diele,
Liq. Cryst., 1995, 18, 185–190.
22 J. Barbera, M. Bardaji, J. Jimenez, A. Laguna, M. P. Martinez,
L. Oriol, J. L. Serrano and I. Zaragozano, J. Am. Chem. Soc., 2005,
127, 8994–9002.
23 J. Barbera, J. Jimenez, A. Laguna, L. Oriol, S. Perez and
J. L. Serrano, Chem. Mater., 2006, 18, 5437–5445.
24 B. Dardel, R. Deschenaux, M. Even and E. Serrano, Macromolecules,
1999, 32, 5193–5198.
25 T. Chuard, R. Deschenaux, A. Hirsch and H. Schonberger, Chem.
Commun., 1999, 2103–2104.
26 D. Felder-Flesch, L. Rupnicki, C. Bourgogne, B. Donnio and
D. Guillon, J. Mater. Chem., 2006, 16, 304–309.
27 G. H. Mehl and J. W. Goodby, Angew. Chem., Int. Ed. Engl., 1996,
35, 2641–2643.
28 J. W. Goodby, G. H. Mehl, I. M. Saez, R. P. Tuffin, G. Mackenzie,
R. Auzely-Velty, T. Benvegnu and D. Plusquellec, Chem. Commun.,
1998, 2057–2070.
29 R. Elsasser, G. H. Mehl, J. W. Goodby and D. J. Photinos, Chem.
Commun., 2000, 851–852.
30 I. M. Saez and J. W. Goodby, J. Mater. Chem., 2001, 11, 2845–2851.
31 I. M. Saez, J. W. Goodby and R. M. Richardson, Chem.–Eur. J.,
2001, 7, 2758–2764.
32 R. M. Laine, C. Zhang, A. Sellinger and L. Viculis, Appl. Organomet.
Chem., 1998, 12, 715–723.
45 A. Zelcer, B. Donnio, C. Bourgogne, F. D. Cukiernik and D. Guillon,
Chem. Mater., 2007, 19, 1992–2006.
46 L. Zhi, J. Wu and K. Mullen, Org. Lett., 2005, 7, 5761–5764.
47 R. I. Gearba, M. Lehmann, J. Levin, D. A. Ivanov, M. H. J. Koch,
J. Barbera, M. G. Debije, J. Piris and Y. H. Geerts, Adv. Mater.,
2003, 15, 1614–1618.
48 J. Tant, Y. H. Geerts, M. Lehmann, V. De Cupere, G. Zucchi,
B. W. Laursen, T. Bjornholm, V. Lemaur, V. Marcq, A. Burquel,
E. Hennebicq, F. Gardebien, P. Viville, D. Beljonne, R. Lazzaroni
and J. Cornil, J. Phys. Chem. B, 2006, 110, 3449–3449.
49 G. J. Clarkson, N. B. Mckeown and K. E. Treacher, J. Chem. Soc.,
Perkin Trans. 1, 1995, 1, 1817–1823.
50 J. Miao and L. Zhu, Chem. Mater., 2010, 22, 197–206.
51 L. Cui, J. P. Collet, G. Xu and L. Zhu, Chem. Mater., 2006, 18, 3503–3512.
52 J. F. Van der Pol, E. Neeleman, J. W. Zwikker, R. J. M. Nolte,
W. Drenth, J. Aerts, R. Visser and S. J. Picken, Liq. Cryst., 2006,
33, 1378–1387.
53 T. Wang, D. Yan, J. Luo, E. Zhou, O. Karthaus and H. Ringsdorf,
Liq. Cryst., 1997, 23, 869–878.
€
54 T. Foster, Ann. Phys., 1948, 437, 55–75.
55 A. Hayer, V. de Halleux, A. Kohler, A. El-Garoughy, E. W. Meijer,
J. Barbera, J. Tant, J. Levin, M. Lehmann, J. Gierschner, J. Cornil
and Y. H. Geerts, J. Phys. Chem. B, 2006, 110, 7653–7659.
56 P. Wu, Q. Zeng, S. Xu, C. Wang, S. Yin and C. L. Bai,
ChemPhysChem, 2001, 2, 750–754.
33 C. Zhang, T. J. Bunning and R. M. Laine, Chem. Mater., 2001, 13,
3653–3662.
This journal is ª The Royal Society of Chemistry 2010
Soft Matter, 2010, 6, 2072–2079 | 2079