Gearing of molecular swirls: Columnar packing of nematogenic hexakis(4-alkoxyphenylethynyl)benzene derivatives

Article abstract of DOI:10.1002/chem.201002975

Through molecular design and straightforward synthesis, incorporating an additional alkoxy chain onto various numbers of peripheral phenyls in nematogenic hexakis(4-alkoxyphenylethynyl)benzene was achieved to generate columnar phases with significantly ex

Full text of DOI:10.1002/chem.201002975

DOI: 10.1002/chem.201002975
Gearing of Molecular Swirls: Columnar Packing of Nematogenic Hexakis(4-
alkoxyphenylethynyl)benzene Derivatives
Shern-Long Lee,[b] Hsing-An Lin,[a] Yi-Hui Lin,[a] Hsiu-Hui Chen,[a] Ching-Ting Liao,[a]
Tzu-Ling Lin,[a] Yi-Chen Chu,[b] Hsiu-Fu Hsu,*[a] Chun-hsien Chen,*[b] Jey-Jau Lee,[c]
Wen-Yi Hung,[d] Qu-Yuan Liu,[d] and Chunhung Wu[a]
792
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 792 – 799

FULL PAPER
Abstract: Through molecular design
and straightforward synthesis, incorpo-
rating an additional alkoxy chain onto
various numbers of peripheral phenyls
in nematogenic hexakis(4-alkoxyphen-
chains on every peripheral phenyl,
scanning tunnelling microscopic studies
indicate the molecule adopts a pre-
ferred molecular-swirl geometry by re-
stricting the conformational arrange-
ment of the alkoxy side chains. Cooper-
ative packing of the molecular swirls
by a lock-in mechanism among col-
umns results in a stable helical column
packing evidenced by powder X-ray
diffraction.
Keywords: columnar
suprastruc-
ACHTUNGTRENNUNGylethynyl)benzene was achieved to
tures · helical structures · liquid
crystals · molecular swirls · scanning
tunnelling microscopy
generate columnar phases with signifi-
cantly expanded temperature ranges.
For the compound with two decyloxy
Introduction
side-arm rotation could be restricted and thus the interdisc
interactions were improved by introducing an additional
chain ortho to the original chain of one side-arm in 1, and a
discotic nematic (ND) phase rather than the columnar pack-
ing was observed for 2.[6]
Columnar liquid crystalline materials continuously attract
interest because of their cooperative physical properties that
lead to many technological applications.[1] As for the design
principles of columnar mesogenic materials, the molecular
core planarity and chain density are usually the two main
considerations, which result in columnar suprastructures that
are generated mainly by planar disc-like mesogens with
strong p–p core attractions and assisted by intermolecular
forces, such as hydrogen bonding and dipole–dipole interac-
tions.[2] For example, a columnar packing was developed by
replacing with biphenyls the peripheral phenyls in hexa(phe-
nylethynyl)benzene 1,[3] one of the most studied discotic
nematogens, which would otherwise adopt a nematic meso-
phase due to reduced p–p interactions by the rotational
freedom of the peripheral phenyls.
Recently, Meijer et al. reported oligo(p-phenylene vinyl-
ene) substituted star-like hexaarylbenzene compounds show-
ing columnar packing in liquid crystalline states.[4,5] Scanning
tunnelling microscopic (STM) studies revealed molecular ro-
settes of specific chain “rotation directions” at a solid–liquid
interface. Interdigitation of these chains from adjacent ro-
settes were attributed to the stability of 2D crystal structure,
but was not utilised as the interaction for mesophase forma-
tion. On the other hand, we recently demonstrated that the
Reported herein is a very efficient reaction scheme that
expedites further exploration of the restricted rotation to all
six side-arms of the well-know nematogenic 1, in conjunc-
tion with application of interdigitation of chains with specif-
ic “rotation directions” for the generation of columnar su-
prasturctures. The resulting novel mesogens 5 should exhibit
low melting points and wide mesogenic temperature ranges
due to lateral substitution within the side arms. The rotation
of peripheral phenyls, restricted but not rigid, endows struc-
tural flexibility for 5 to favour a cooperative fashion of
chain arrangement giving a molecular swirl geometry, that
is, a molecular rosette (Figure 1), to minimise intramolecular
steric congestion. Hence, the interdisc p–p attractions are
enhanced. Moreover, along the disc plane, packing of mole-
cules of the same swirl sense, that is, a gearing or lock-in
mechanism of molecular swirls, will be favoured to minimise
steric crowdedness and in-plane interchain van der Waals
(VDW) interactions can be maximised.
Results and Discussion
To investigate the influence of the dialkoxyphenylethynyl
side arms on the mesophase properties in hexaynylbenznes,
a varied number of dialkoxy-containing side arms were in-
corporated as shown in Scheme 1. Stoichiometric control of
1-ethynyl-4-hexyloxybenzene and higher reactivity of iodo
group in 1,3,5-tribromo-2,4,6-triiodobenzene towards Sono-
gashira coupling enabled the successful synthesis of 3 (50%)
and 4 (32%). Subsequent coupling of dialkoxy-containing
side arms to 4 and 3 afforded compounds 5 and 6, respec-
tively. The synthesis of 7 originally suffered from low yields
(ꢀ10%) when diisopropylamine was used as the base and
the solvent. Various conditions were examined for the syn-
thesis of 7 to give optimised yields (ꢀ50%) by using tri-
[a] H.-A. Lin, Y.-H. Lin, Dr. H.-H. Chen, C.-T. Liao, T.-L. Lin,
Prof. H.-F. Hsu, C. Wu
Department of Chemistry
Tamkang University, 251 Taipei (Taiwan)
Fax : (+886) 02-2620-9924
E-mail: hhsu@mail.tku.edu.tw
[b] Dr. S.-L. Lee, Y.-C. Chu, Prof. C.-h. Chen
Department of Chemistry
National Taiwan University, 106 Taipei (Taiwan)
Fax : (+886) 02-2363-6359
E-mail: chhchen@ntu.edu.tw
[c] Dr. J.-J. Lee
National Synchrotron Radiation Research Center of Taiwan
300 Hsinchu (Taiwan)
[d] Prof. W.-Y. Hung, Q.-Y. Liu
AHCTUNGTREGeNNNU thylamine as the base and triphenylphosphine as the auxil-
Institute of Optoelectronic Sciences
National Taiwan Ocean University, 202 Keelung (Taiwan)
lary ligand as detailed in experimental.
The mesogenic properties of 5, 6, and 7b, investigated by
differential scanning calorimeter and optical polarizing mi-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201002975.
Chem. Eur. J. 2011, 17, 792 – 799
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
793

H.-F. Hsu, C.-h. Chen et al.
tion of columnar suprastructures. Compound 6 with three di-
alkoxy-containing side arms possesses the widest liquid crys-
talline range of 1638. The much lower melting points of 2, 5,
6, and 7b are attributed to their lower molecular symmetry
and the lateral position of the added chains.[6,7] Comparing
with the monochain analogue (1a, 143.5–215.98C ND), 7b
with a columnar phase exhibits a melting point 518 lower
and a clearing point 188 higher to give a mesophase range of
698 wider. The generation of columnar suprastructures of 7b
is ascribed to the various aforementioned interactions, in-
cluding the possible intercolumnar, swirl-gearing interac-
tions.
In order to realise the possible swirl conformation as well
as swirl-gearing interactions of 7, compound 7a with but-
AHCTUNGTREGyNNUN loxy chains was prepared in hope to obtain its X-ray
single-crystal structure. However, attempts on growing
single crystals of 7a gave very thin needles that were not
suitable for X-ray structural determination.
The thermal properties of 7a–f are also given in Table 1.
Most of these compounds showed partial decomposition
prior to clearing, except the longer chain analogues 7e and
7 f, which were enantiotropic; this enabled columnar phase
determination from natural mesophase textures, birefringent
dendritic or focal conic (Figure 2). The melting point de-
Figure 1. Top, molecular structures of 1, 1a, 2, and 7; bottom, schematic
molecular geometry representations of compounds 1 and 7, in which co-
operative and non-cooperative fashions of chain rotation in 7 are present-
ed, showing sterically favoured and sterically congested conformations,
respectively.
croscope, are listed in Table 1 together with those of 1a[6]
and 2[6]. As reported previously, replacing one side arm in
1a with a dialkoxy-containing side arm provided a ND phase
with lower transition temperatures and a wider mesophase
range.[6] In contrast, compounds 5, 6, and 7b, with two or
more dialkoxy-containing side arms, exhibit a columnar
phase while keeping the characteristics of a much lower
melting point and wider mesophase range of 2. Hence, two
dialkoxy-containing side arms are sufficient for the genera-
Table 1. Phase behaviors of compounds 1a and 2a ~2 f.[a]
Phase transition behaviour T [8C] (DH [kJmolꢁ1]) Range
1a[b]
Cryst–ND–Isop.d
Cryst–ND–Isop.d
Cryst–Colr–Isop.d.
Cryst–Colr–Isop.d.
Cryst–Colr–Isop.d.
Cryst–Colr–Isop.d.
Cryst–Colr–Isop.d.
144 (37.7), 216 (0.1)
93.7 (65.72), 195.0
73.5 (47.57), 178
72
101
104
163
95
142
148
145
131
95
2[b]
Figure 2. Optical micrographs of a) 6 at 1838C and b) 7e at 1708C sand-
wiched between glass slides showing dendritic or focal conic textures be-
tween cross polarisers.
5
6
72.9 (40.31), 236
7a (C4)
7b (C6)
7c (C8)
7d (C10) Cryst–Colr–Isop.d.
7e (C12) Cryst–Colr–Iso
137 (66.3), 232 (11.2)
92 (115.3), 234 (11.1)
79 (88.3), 227 (14.8)
67 (118.9), 212 (13.8)
70 (195.7), 201 (14.4)
74 (296.0), 169 (21.9)
creases with increasing chain length and remains at about
708C for derivatives with decyloxy chains or longer. The
clearing temperature also decreases with increasing chain
length, but not as significantly as the melting point. The two
effects in the transition temperatures result in compounds
with medium chain lengths, that is, hexyloxy, octyloxy, and
decyloxy, possessing wider mesophase ranges than the short-
er and longer chain analogues.
7 f (C16)
Cryst–Colr–Isop.d.
[a] Cryst, ND, Colr, Iso, and Isop.d. denote crystalline, discotic nematic, rec-
tangular columnar, isotropic phases, and isotropic liquid with partial de-
composition, respectively. All thermal data were obtained by DSC at a
rate 108C minꢁ1. For 5, 6 and 7a–d, data were obtained from first heating
runs and for 7e and 7 f data were obtained from second heating runs.
[b] See reference [5].
794
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 792 – 799

Molecular Swirls
FULL PAPER
corresponding to the degree of
tunneling probability, were as-
cribed to the p-conjugated
cores versus the saturated
chains.[8–10] Though not every
chain was resolved, possibly
due to some alkyls being float-
ing as reported in the litera-
ture,[10,11] most of them were
recognised to establish the swirl
molecular geometry by the
bending directions of chains rel-
ative to the side-arm axes. As
indicated by the arrows in Fig-
ure 3b and c, the senses of the
molecular swirls are clearly re-
vealed to be counter-clockwise
and clockwise, respectively. The
revolving of the peripheral
phenyl planes on each side-arm
axis is not resolved and hence
the molecules were considered
as 2D swirl discs rather than 3D
helical structures. In contrast to
the enantiomericity for 3D heli-
ces of opposite helical senses, a
2D molecular swirl will super-
impose with its mirror image.
However, when considering the
graphite surface underneath,
the 2D swirls of opposite rota-
tion senses are chiral and enan-
tiomeric to each other, even if
the peripheral phenyl rings are
coplanar with the central ben-
zene.
Scheme 1. Synthesis of compounds 5, 6 and 7a–f.
Single-crystal X-ray structural determination on 7a for
the realisation of possible swirl molecular geometry was not
achieved. Alternatively, to elucidate the development of mo-
lecular swirls for 7, STM (scanning tunneling microscopy)
was employed to unveil possible orientations of the chains
with respect to the core. The results show that these com-
pounds adopted a face-on orientation and hexagonal ar-
rangement on the basal plane of HOPG (highly orientated
pyrolytic graphite). For those with decyloxy groups or short-
er, for example, 7d, the chains were not resolved due to
short alkyls being conformationally unsettled by their rela-
tively weak van der Waals interactions with the substrate.[8]
A representation of the arrangement of alkoxy chains in 7e
is shown in Figure 3; the liquid crystalline performance is in
fact far superior to those with shorter chains. The dotted
line in Figure 3a marks a domain boundary. Details of re-
gimes I and II are shown in Figure 3b and c, respectively.
The molecular cores were well resolved to give bright fea-
tures of a pseudo-hexagon showing six peripheral phenyls
around the central benzene as contrast to the dark charac-
teristics exhibited by the chains. These brightness contrasts,
The STM observations were limited in a quasi-two-dimen-
sional system. The two reverse senses of swirls aggregated
into their own respective domains in which the rotational
senses of the swirls were the same within each domain to
Figure 3. STM images of 7e on HOPG. a) A 40ꢁ28 nm image exhibiting
a domain boundary indicated by the dotted line; 12ꢁ12 nm images in
b) region I and c) region II, respectively, revealing counter-clockwise and
clockwise rotational senses of alkoxy chains (arrows). The interdisc dis-
tance is measured to be 30ꢂ2 ꢂ after the calibration of the piezoelectric
scanner.
Chem. Eur. J. 2011, 17, 792 – 799
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
795

H.-F. Hsu, C.-h. Chen et al.
avoid steric congestions. At the junction of these two do-
mains, a domain boundary with images of blurred disc and
overlapped discs developed. The HOPG surface is known to
influence strongly the organisation of monolayer above.
Therefore, the 2D ordering of 7e influenced by the graphite
surface underneath can be expected. However, reverse
sense swirls aggregating into respective domains, resulting in
domain separation, cannot be fully justified by HOPG sur-
face induction. These observations support the proposition
of the in-plane intermolecular correlations favouring the
same swirl sense due to steric considerations.
The columnar mesophases of 5, 6, and 7a–f were probed
by X-ray diffraction (XRD) studies and the results are tabu-
lated in Table 2. All compounds show a broad halo at about
4.5 ꢂ of molten chains indicating mesophase formation.
Most compounds show signals indexed to d11, d20, d02, and
d31, indicative of a rectangular arrangement of columns. For
compounds 5, 7d, and 7e, only the halos d11, and d20 were
observed. However, a rectangular columnar phase was as-
signed due to their similar small-angle-signal patterns with
respect to those of 7a and 7b and their similar optical tex-
tures. In the columnar phases of 5 and 6, the intercolumnar
distances (dintercol) of about 24.1 ꢂ, one half of the rectangle
diagonal, are significantly longer than the interdisc distance
of 21.3 ꢂ for the nematic phase of 1. For 1, the interdisc dis-
tance is slightly longer the rigid core size, indicating the
chains were overlapping with adjacent molecular core. For 5
and 6, adjacent molecules were spaced out from each other
to avoid interdisc chain/core overlap for the generation of
columnar suprastructures. With the same chain length, com-
pounds 5, 6, and 7b exhibit similar intercolumnar distances,
indicating dintercol is primarily dependent on the dialkoxy-
containing side arms. The values of dintercol of 7 do not corre-
spond to, and are significantly shorter than, the calculated
in-plane interdisc distances with full interdigitation of
chains. As an example, when the chains are fully extended
along the axis of the attached side arm for 7e, fully interdi-
gitated all-trans chains between two adjacent columns will
give a dintercol of approximately 35.7 ꢂ and the value from
XRD is 30.7 ꢂ.
The significant discrepancy between experimental and cal-
culated intercolumnar distances can be realised by detailed
analysis of the XRD results versus possible molecular pack-
ing fashions in the columnar phase. From the full interdigita-
tion model, shortening dintercol by further shifting the chains
toward adjacent discs will destroy columnar packing. Hence,
the shorter experimental dintercol can only come from a non-
fully-extended chain configuration, in-plane bending of
chain axis from the axis of attached side arm, or out-of-
plane bending of chain axis from the core plane. Non-fully-
extended chain configuration can be excluded by the ap-
proximately regular increment of each C2 lengthening of
chains. Both the in-plane and out-of-plane chain bendings
can be expected due to the bent ether linkage. The out-of-
plane chain bending model can be ruled out, since the calcu-
lated dintercol values, obtained by summing up the trans-O-to-
O distance (19.1 ꢂ) and cos(q) of chain length at various
angle qs, do not correlate with the experimental values. For
in-plane chain bending, there is no simple mathematic equa-
tion for the estimation of intercolumnar distances for 7 with
different bending angles, since the hexagonal cores also turn
with different bending angles. Simulation of molecular pack-
ing has been used for the estimation of intercolumnar dis-
tances with various bending angles, 30, 60, 77, and 908, for
7e. The dintercol for the bending angle of 778 was simulated
for the comparison with the respective STM value. As the
in-plane bending angle increases, the intercolumnar distance
decreases, from 35.7 ꢂ for 308 to 26.5 ꢂ for 908 as the two
limiting cases (Figure 4). The estimated intercolumnar dis-
tance of 30.2 ꢂ with an in-plane bending of 778 is similar to
the values of 30.4 and 30.0 ꢂ determined by XRD and STM
investigations, respectively. Similar simulations were also
performed for the shorter and longer chain analogues 7a
and 7 f with an in-plane bending angle of 778. For 7 f, the
dintercol found by XRD is longer than that of calculated value,
which implies a bending angle smaller than 778. For 7a, the
dintercol value determined by XRD is significantly shorter
than that by packing simulation, which may arise from a sig-
Table 2. Powder X-ray diffraction data of compounds 7a–f.
T
Phase
d spacing [ꢂ]
Miller
Lattice
dntercol
[8C]
obsd
calcd
indices constants [ꢂ]
[ꢂ]
5
6
120
100
Colrd
Colrd
22.4
18.2
4.5
22.7
19.7
14.0
11.8
11.4
4.5
20.9
17.4
12.9
10.6
10.4
4.3
23.2
19.7
14.4
11.9
9.8
22.4
18.2
11
20
alkyl
11
20
02
31
40
alkyl
11
20
02
31
22
alkyl
11
20
02
31
40
alkyl
11
20
02
alkyl
11
20
alkyl
11
20
alkyl
11
20
31
40
a=36.48
b=28.30
23.8
22.7
19.7
13.9
11.9
11.4
a=39.38
b=27.85
24.3
7a (C4)
7b (C6)
7c (C8)
180
180
190
Colrd
Colrd
Colrd
20.9
17.4
13.0
10.6
10.5
–
23.2
19.7
14.3
11.9
9.9
a=34.9
b=26.1
21.8
24.4
26.8
a=39.4
b=28.6
4.4
–
25.4
21.9
15.5
4.3
28.2
25.5
4.7
28.4
25.5
4.7
32.3
30.0
17.7
15.0
4.6
25.4
21.9
15.6
–
28.2
25.0
–
28.4
25.5
–
32.3
30.0
17.7
15.0
–
a=43.7
b=31.1
7d (C10) 133
7e (C12) 180
7 f (C16) 150
Colrd
Colrd
Colrd
a=50.0
b=34.1
30.3
30.7
36.6
a=51.0
b=34.2
a=60.0
b=38.3
alkyl
796
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 792 – 799

Molecular Swirls
FULL PAPER
correlation length of 4.6 ꢂ, a pair of diffuse reflections, cor-
responding to a long p–p stacking distance of peripheral
phenyls at 3.9 ꢂ with an averaged tilting angle of 528,[4] ap-
pears at both sides of the meridian as shown in Figure 5b.
The fact that the meridian-centred alkyl halos are perpen-
dicular to the equator-centred intercolumnar reflections
proves the out-of-plane chain-bending model to be invalid
as described previously. At middle angles, a very weak dif-
fuse reflection at 7.2 ꢂ was detected, which cannot be deriv-
ative signals of the rectangular lattice due to its very diffuse
nature. Apparent alignment of this scattering could not be
unambiguously determined due to its very low intensity;
however, the azimuthal scan of this scattering revealed the
reflection centred along meridian rather than along the
Figure 4. Simulated two-molecule packing patterns of 7e with chain
bending angles of a) 308, b) 608, c) 778, and d) 908. For clarity, only one
chain is shown for each molecule.
nificantly larger bending angle.
It should be noted that devia-
tion from the cooperative fash-
ion of in-plane chain bending
model, one, two, or three chains
bending to the opposite direc-
tion with respect to bending di-
rections of other chains, will
not give the experimentally ob-
served in-plane rectangular lat-
tices and will lead to significant
longer intercolumnar distances
and result in the experimentally
observed in-plane rectangular
lattice.
X-ray diffraction investiga-
tions on aligned 7e prepared by
filament extrusion[12] were uti-
lised for the realisation of its
possible helical packing. For
comparison, Figure 5a and
b
show the XRD investigations
on the unaligned sample of 7e.
Figure 5c shows the 2D XRD
pattern of an extruded filament
of 7e. For this aligned sample,
in the small-angle regime,
a
pair of sharp and intense arcs
along the equator, correspond-
ing to intercolumnar correla-
tions, is found. Each arc con-
sists of two overlapped reflec-
tions of d11 and d20. Since the
filament was placed vertically,
these small-angle equatorial re-
flections correspond to the
inter-columnar order. In the
wide angle regime, in addition
Figure 5. Structural investigations of 7e by a) the 2D PXRD pattern of unaligned sample at 1808C (inset: 2D
PXRD of small angle scatterings), b) respective intensity/2q plot of a), c) the 2D PXRD pattern of a sample
aligned by filament extrusion at 1058C (inset: 2D PXRD of small angle scatterings), d) the azimuthal integra-
tion of the scattering intensity corresponding to the p-stacking distance, and e) the schematic illustration of
the intramolecular arrangement of 7e within a helical column (* indicates a reference point on each disc to
to the broad alkyl halo with a better illustrate the helical conformation within a column).
Chem. Eur. J. 2011, 17, 792 – 799
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
797

H.-F. Hsu, C.-h. Chen et al.
equator, indicative of a modulation of electronic density
along columns (Figure 5d).[12,13] The distance is about twice
that of the p–p stacking distance, indicating that every third
disc within a column is rotationally correlated. With a C6 ro-
tational axis along the disc normal of 7 with a cooperative
in-plane chain bending, such a correlation can be realised by
rotating each successive stacked disc by 3608/6/2=158 about
the C6 axis to give a helical stacking with every other disc to
be rotationally eclipsed (Figure 5e). Thus, the helical pitch
can determined to be 7.2ꢁ6=43.2 ꢂ.
The charge transport properties of 7c and 7e were investi-
gated by the time-of-flight (TOF) transient-photocurrent
technique.[1,14] The samples for the TOF measurements were
prepared by dissolving appropriate weight ratios (up to
25 wt%) of the compounds in chloroform and then dip-coat-
ing the solutions onto an ITO substrate to form amorphous
films. Figure 6a presents the representative TOF transient
for holes of 7c, revealing dispersive transport behaviours.
Conclusion
In summary, through a straightforward and efficient synthe-
sis, compounds 7 adopt a preferred molecular swirl geome-
try by restricting the conformational arrangement of alkoxy
side chains, rather than inventing novel aromatic core struc-
tures for the generation of columnar suprastructures . Coop-
erative packing of the molecular swirls by a lock-in mecha-
nism among columns results in a stable columnar meso-
phase, the temperature range of which becomes twice that
of the prototypical 1. The molecular design may be applied
to other discotic nematogens for pursuing columnar packing
motifs in further optoelectronic applications in which or-
dered disc packing along and perpendicular to the molecular
plane is required.
Experimental Section
All chemicals and solvents were reagent grades (Aldrich Chemical Co.)
and used without further purification. 1H and 13C NMR spectra were re-
corded on a Bruker AC-300 spectrometer. Chemical shifts are reported
in ppm relative to residual CHCl3 (1H: d=7.26 ppm; 13C: d=77.0 ppm).
Multiplicities are given as s (singlet), d (doublet), t (triplet), q (quartet),
and m (multiplet). Absorption spectra were recorded with a Jasco V-550
spectrometer. Photoluminescence spectra were recorded with a Hitachi
F-2500 fluorescence spectrophotometer. Differential scanning calorime-
try (DSC) was measured on a Perkin–Elmer Pyris1 with heating and
cooling rates of 5 and 108C minꢁ1. Polarizing optical microscopy (POM)
was carried out on a Zeiss Axio Imager A1m with a Mettler FP90/
FP82HT hot stage system. X-ray powder diffraction data were collected
on the wiggler beam line BL17 A of the National Synchrotron Radiation
Figure 6. a) Hole transient photocurrent signals for 7c (thickness: 2.9 mm)
at E=2.4ꢁ105 Vcmꢁ1. Inset: double-logarithmic plots for determination
of the transit times. b) Hole mobilities plotted as a function of the square
root of the electric field.
Research Center (NSRRC; Taiwan), using a triangular bent Si
ACHTUNGTRENNUNG(111)
monochromator and a wavelength of 1.3271 ꢂ. The sample in a 1 mm ca-
AHCTUNGTRENNUNG
pillary or an extruded filament was mounted on the Huber 5020 diffrac-
tometer. The BL17 A beamline was equipped with an air-stream heater
and the temperature controller was programmed by a PC with a PID
feedback system. Scanning tunneling microscopy (STM) experiments
were carried out with a PicoScan 4500 controller (Agilent Technologies)
and using commercially available Pt/Ir tips (PT, Nanotips, Veeco Metrol-
ogy Group/Digital Instruments). The substrate was HOPG (SPI, ZYH)
and the images of bare HOPG were employed for the piezoelectric cali-
bration of the x and y displacement per volt. The samples for STM imag-
ing were prepared by placing on HOPG a 50 mL aliquot of 7 (1 mg) dis-
solved in phenyloctane (1 mL). Typical imaging conditions of bias voltage
and tunneling current were 0.1–1.0 V and 0.01–0.20 nA, respectively. The
images were subjected to minimal flattening to reduce noise (WSxM,
Nanotech Electronica). For the TOF measurements, the samples were
mounted in a cryostat under vacuum (ca. 10ꢁ3 Torr). A pulsed nitrogen
tunable dye laser was used as the excitation light source (to match the
absorption of organic films) through the semitransparent electrode (ITO)
induced photogeneration of a thin sheet of excess carriers. Under an ap-
plied dc bias, the transient photocurrent was swept across the bulk of the
organic film toward the collection electrode (Ag), and then recorded
with a digital storage oscilloscope. Depending on the polarity of the ap-
plied bias, selected carriers (holes or electrons) are swept across the
sample with a transit time of tT. With the applied bias V and the sample
thickness D, the applied electric field E is V/D, and the carrier mobility is
The carrier transit time (tT) needed for determining carrier
mobilities can be evaluated from the intersection point of
two asymptotes in the double-logarithmic representation of
the TOF transient. Figure 6b displays the field dependence
of the carrier mobility of 7c and the hole mobility ranged
from 1.5ꢁ10ꢁ4 to 2ꢁ10ꢁ4 cm2 Vꢁ1 sꢁ1 for fields varying from
2.8ꢁ105 to 4.6ꢁ105 Vcmꢁ1. The field dependence of hole
mobility follows the nearly universal Poole–Frenkel relation-
ship[15]: m/exp
ACHTUNGTRENNUNG
(bE1/2), in which b is the Poole–Frenkel factor
and E1/2 is the square root of the electric field, which as-
sumes that charge transport within amorphous organic
solids by means of a hopping mechanism associated with en-
ergetic disorder of localised hopping sites and intermolecu-
lar positional disorder. For 7e, the TOF transient displays
highly dispersive hole transport behavior and hence the car-
rier transit time could not be determined. The increase in
the length of alkyl chains elongates the carrier hopping dis-
tance between molecules, leading to carrier trapping. Con-
sidering the non-fused molecular core of 7, the obtained fair
hole mobility may arise from the high intermolecular corre-
lations in the columnar phase.
then given by m=D/ACHTNUTRGNENUG ACHTUGNTNER(NUNG VtT), from which the carrier transit time
(tTE)=D2/
(tT) can be extracted from the intersection point of two asymptotes to the
plateau and the tail sections in double-logarithmic plots. Mass spectra
were obtained on Finnegan MAT-95XL and elemental analyses were car-
ried out on a Heraeus CHN-O-Rapid Analyzer at the NSC Regional In-
strumental Center at National Chiao Tung University, Hsinchu (Taiwan)
and at National Cheng Kung University, Tainan (Taiwan).
798
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 792 – 799

Molecular Swirls
FULL PAPER
Hexakis(3,4-dialkoxyphenylethynyl)benzene (7): 1,2-Dialkoxy-4-ethynyl-
benzene (2.172 mmol) in triethylamine (8 mL) wasa added dropwise to a
mixture of trans-dichlorobis(triphenylphosphine)palladium(II) (13 mg,
0.019 mmol), copper iodide (8 mg, 0.04 mmol), triphenylphosphine
(10 mg, 0.038 mmol), hexabromobenzene (100 mg, 0.181 mmol), triethyl-
Praefcke, Chimia 1987, 41, 196–198; S. Kumar, S. K. Varshney,
Angew. Chem. 2000, 112, 3270–3272; Angew. Chem. Int. Ed. 2000,
39, 3140–3142; K. Praefcke, B. Kohne, D. Singer, Angew. Chem.
1990, 102, 200–202; Angew. Chem. Int. Ed. Engl. 1990, 29, 177–179;
K. Praefcke, D. Singer, B. Gꢄndogan, K. Gutbier, M. Langner, Ber.
Bunsen-Ges. 1993, 97, 1358–1361.
ACHTUNGTRENNUNGamine (8 mL), and tetrahydrofuran (8 mL) under a nitrogen atmosphere
ˇ
´
under reflux. The mixture was stirred under reflux for 12 h. After cool-
ing, dichloromethane (100 mL) was added. The mixture was washed with
saturated NH4Cl(aq) (40 mL), 4n HCl(aq) (40 mLꢁ3) and H2O (50 mLꢁ
3) and dried over MgSO4. After removal of the solvents, the residue was
purified by crystallisation from hot ethyl acetate. When necessary, the
product was further purified by column chromatography (SiO2, n-hexane/
dichloromethane 5:1) and crystallised from n-hexane to yield 7 as a
yellow powder (yield 8–59%).
Data for 7e: Yield: 21%. 1H NMR (CDCl3, 300 MHz): d=7.21 (dd, J=
8.4, 1.8 Hz, 6H), 7.12 (d, J=1.5 Hz, 6H), 6.81 (d, J=8.4 Hz, 6H), 4.01 (t,
J=6.5 Hz, 12H), 3.83 (t, J=6.5 Hz, 12H), 1.84 (m, 12H), 1.74 (m, 12H),
1.47 (m, 24H), 1.34 (m, 192H), 0.89 ppm (m, 36H); 13C NMR (CDCl3,
75 MHz): d=150.1, 148.9, 126.9, 125.0, 116.7, 115.5, 113.1, 99.4, 86.3,
69.2, 69.1, 31.93, 29.77, 29.71, 29.65, 29.56, 29.48, 29.39, 29.36, 29.29,
[4] Z. Tomovic, J. van Dongen, S. J. George, H. Xu, W. Pisula, P. Le-
clꢅre, M. M. J. Smulders, S. D. Feyter, E. W. Meijer, A. P. H. J.
Schenning, J. Am. Chem. Soc. 2007, 129, 16190–16196.
ˇ
[5] H. Xu, A. Minoia, Z. Tomovic, R. Lazzaroni, E. W. Meijer,
A. P. H. J. Schenning, S. D. Feyter, ACS Nano 2009, 3, 1016–1024.
[6] S.-C. Chien, H.-H. Chen, H.-C. Chen, Y.-L. Yang, H.-F. Hsu, T.-L.
Shih, J.-J. Lee, Adv. Funct. Mater. 2007, 17, 1896–1902.
[7] H.-F. Hsu, H.-C. Chen, C.-H. Kuo, B.-C. Wang, H.-T. Chiu, J. Mater.
Chem. 2005, 15, 4854–4861; H.-F. Hsu, C.-H. Kuo, C.-F. Chen, Y.-H.
Lin, L.-Y. Huang, C.-H. Chen, K.-C. Cheng, H.-H. Chen, Chem.
Mater. 2004, 16, 2379–2385.
[8] Q. Chen, T. Chen, G.-B. Pan, H.-J. Yan, W.-G. Song, L.-J. Wan, Z.-T.
Li, Z.-H. Wang, B. Shang, L.-F. Yuan, J.-L. Yang, Proc. Natl. Acad.
Sci. USA 2008, 105, 16849–16854; X. Xu, H. L. J. Yin, Y. Zhou, J.
Li, J. Pei, K. Wu, J. Phys. Chem. C 2009, 113, 8844–8852.
[9] R. Lazzaroni, A. Calderone, J. L. Brꢆdas, J. P. Rabe, J. Chem. Phys.
1997, 107, 99–105; a) P. Samorꢇ, A. Fechtenkçtter, F. Jꢃckel, T.
Bçhme, K. Mꢄllen, J. P. Rabe, J. Am. Chem. Soc. 2001, 123, 11462–
11467; F. Jꢃckel, M. Ai, J. Wu, K. Mꢄllen, J. P. Rabe, J. Am. Chem.
Soc. 2005, 127, 14580–14581; S. Xu, Q. Zeng, J. Lu, C. Wang, L.-J.
Wan, C.-L. Bai, Surf. Sci. 2003, 538, 219–232.
26.13, 26.07, 22.68, 14.01 ppm; MS (FAB+): m/z calcd for C198H318O12
:
2890.64; found 2890.56; elemental analysis calcd (%) for C198H318O12: C
82.27, H 11.09; found: C 82.27, H 11.08.
Acknowledgements
[10] N. Katsonis, J. Vicario, T. Kudernac, J. Visser, M. M. Pollard, B. L.
Feringa, J. Am. Chem. Soc. 2006, 128, 15537–15541; K. Tahara, S.
Furukawa, H. Uji-i, T. Uchino, T. Ichikawa, J. Zhang, W. Mamdouh,
M. Sonoda, F. C. D. Schryver, S. D. Feyter, Y. Tobe, J. Am. Chem.
Soc. 2006, 128, 16613–16625.
We thank the National Science Council of Taiwan (NSC 96-2628M-032
and NSC 99-2628M-002-013) for supporting this work.
[11] Y. Kaneda, M. E. Stawasz, D. L. Sampson, B. A. Parkinson, Lang-
muir 2001, 17, 6185–6195.
[1] S. Laschat, A. Baro, N. Steinke, F. Giesselmann, C. Hꢃgele, G.
Scalia, R. Judele, E. Kapatsina, S. Sauer, A. Schreivogel, M. Tosoni,
Angew. Chem. 2007, 119, 4916–4973; Angew. Chem. Int. Ed. 2007,
46, 4832–4887.
[2] T. Kato, N. Mizoshita, K. Kishimoto, Angew. Chem. 2006, 118, 44–
74; Angew. Chem. Int. Ed. 2006, 45, 38–68; I. M. Saez, J. W.
Goodby, J. Mater. Chem. 2005, 15, 26–40; C. Tschierske, Annu. Rep.
Prog. Chem. Sect. C 2001, 97, 191–267; V. Percec in Handbook of
Liquid Crystal Research (Eds.: P. J. Collings, J. S. Patel), Oxford Uni-
versity Press, New York, 1997.
ˇ
´
[12] W. Pisula, Z. Tomovic, C. Simpson, M. Kastler, T. Pakula, K.
Mꢄllen, Chem. Mater. 2005, 17, 4296–4303.
[13] J. Barberꢈ, E. Cavero, M. Lehmann, J.-L. Serrano, T. Sierra, J. T.
Vꢈzquez, J. Am. Chem. Soc. 2003, 125, 4527–4533.
[14] P. M. Borsenberger, D. S. Weiss, Organic Photoreceptors for Imaging
Systems, Marcel Dekker, NewYork, 1993.
[15] H. Bꢃssler, Phys. Status Solidi B 1993, 175, 15.
[3] M. Ebert, D. A. Jungbauer, R. Kleppinger, J. H. Wendorff, B.
Kohne, K. Praefcke, Liq. Cryst. 1989, 4, 53–67; B. Kohne, K.
Received: October 15, 2010
Published online: January 5, 2011
Chem. Eur. J. 2011, 17, 792 – 799
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
799