A windmill-shaped discotic columnar liquid crystal with fast ambipolar charge carrier transport

Article abstract of DOI:10.1021/cm502574d

A study was conducted to demonstrate a windmill-shaped discotic liquid crystal (DLC) with fast ambipolar charge carrier transport. Truxene derivatives of 1,5,9-trialkyltruxene were designed to conduct the investigations under the study. The implementation of this strategy began with the multigram-scale synthesis of hydroxyindanone 2 from the inexpensive dihydrocumarin 1 through a literature procedure. The intermediate 2 was triflated to produce 3 with N,N-bis(trifluoromethylsulfonyl)-phenylamine at room temperature at a high yield up to 70% prior to the alkylation via Suzuki coupling reaction. intermolecular cyclocondensation reactions were performed under mild conditions by treating a synthetic block, 4, with acid to afford the final truxenes of Tru4 and Tru6 at the yields of 63% and 52% after purification by column chromatography.

Full text of DOI:10.1021/cm502574d

Communication
pubs.acs.org/cm
A Windmill-Shaped Discotic Columnar Liquid Crystal with Fast
Ambipolar Charge Carrier Transport
XuYing Liu,† Takayuki Usui,†,‡ and JunIchi Hanna*,†,‡
†Imaging Science and Engineering Laboratory, Tokyo Institute of Technology, 4259 Nagatsuta Midori-ku, Yokohama 226-8503,
Japan
‡CREST, Japan Science and Technology Agency, 4259 Nagatsuta Midori-ku, Yokohama 226-8503, Japan
S
* Supporting Information
arious design concepts have been proposed to arrange the
molecular organization for better charge carrier transport
the columns. Previously, complementary approaches have been
carried out, including polytopic interaction (CPI)-based
discotics9 and gelated DLC molecules with hydrogen-bonded
fibrous aggregates,10 in addition to shortening intercolumnar
space directly by shorter side chains,3,11 which consequently
resulted in higher charge carrier mobility up to 10−2 cm2 V−2
s−1. Recently, it was found that DLC carbazole macrocycles
with inward side chains were able to give rise to strong
intercolumnar coupling due to minimized intercolumnar space,
which could form a three-dimensional pathway for carriers to
avoid traps, eventually leading to mobility of 10−1 cm2 V−2 s−1
simulated using the kinetic Monte Carlo method.12
It is noteworthy that to form homeotropic columnar
mesophase requires the incorporation of complex peripheral
side chains into disclike cores, which actually should be
responsible for defects and molecular misorientation. Herein,
we propose a new molecular design, i.e., the windmill-shaped
molecule based on a disc-like truxene core (Figure 1b), in
which a short monoside chain is adopted at a bay position to
induce the homeotropic columnar mesophase. Generally, the
shorter side chain forms a much narrower intermolecular
distance between adjacent columns, contributing to suppressing
the molecular motions (translational and rotational). Moreover,
what’s more important is to expose the peripheral phenyl ring,
thus leading to less static defects along quasi one-dimensional
columns, while forcing a face-to-face stacking of molecular
cores is expected to maximize intermolecular couplings, thereby
enhancing charge carrier transport within columns.
V
properties, including utilization of various molecular assem-
blies1 and liquid crystalline mesophases.2 Above all, the
columnar mesophases in discotic liquid crystals (DLCs) have
been extensively studied in various molecules such as
triphenylenes,3 phthalocyanines,4 and hexabenzocoronenes,5
in which disc-shaped molecules self-organizing into one-
dimensional molecular columns as conduction channels are
expected to show high mobility. In fact, the existence of
translational and rotational molecular motions6 originated from
electrostatic repulsion and fluctuating flexible side chains (the
length always over 10 Å)7 (Figure 1a) tends to decrease the
transfer integral, which explains the reason for relatively low
charge carrier mobility ranging from 10−4 to 10−2 cm2 V−1 s−1
in columnar mesophases of conventional discotic molecules.8
In order to achieve higher charge carrier mobility over 10−1
cm2 V−1 s−1 (close to that of a-Si) in discotic columnar phases,
one may consider how to suppress the molecular motions in
According to the present strategy, we designed truxene
derivatives of 1,5,9-trialkyltruxene (Scheme 1). The implemen-
tation of this strategy began with the multigram-scale synthesis
of hydroxyindanone 2 from the inexpensive dihydrocumarin 1
through a literature procedure.13 The intermediate 2 was
triflated to produce 3 with N,N-bis(trifluoromethylsulfonyl)-
phenylamine at room temperature at a high yield up to 70%
prior to the alkylation via Suzuki coupling reaction. At last,
intermolecular cyclocondensation reactions were performed
under mild conditions by treating a synthetic block, 4, with acid
to afford the final truxenes of Tru4 and Tru6 at the yields of
63% and 52%, respectively, after purification by column
chromatography. It is worth noting that this synthetic pathway
for the truxene derivative actually is short and efficient, which
Figure 1. Schematic illustration of a conventional DLC peripherally
substituted truxene (a) and a windmill-shaped DLC truxene (b) in
hexagonal columnar mesophases.
Received: July 15, 2014
Revised: September 17, 2014
Published: September 18, 2014
© 2014 American Chemical Society
5437
dx.doi.org/10.1021/cm502574d | Chem. Mater. 2014, 26, 5437−5440

Chemistry of Materials
Communication
texture was still observed at low temperatures, supporting a
liquid crystal glass. When the sample stayed in ambient
atmosphere overnight, it was found to crystallize and show clear
birefringence in the whole visible region.
Scheme 1. Synthetic Route for Windmill Shaped Truxene
Derivatives
a
In order to investigate the microscopic molecular config-
uration in columnar phases, a study of variable-temperature
XRD measurement was conducted in three phases of Tru6, i.e.,
the first mesophase at 160 °C, the second one at 140 °C, and
the glass phase at 30 °C (Figure 3a). Upon cooling at 160 °C a
a
i, AlCl, NaCl, 200 °C; ii, N,N-bis(trifluoromethylsulfonyl)-
phenylamine, Et3N, DCM, 24 h; iii, alkylboronic acid, Pd(dppf)Cl2,
Ag2O, K2CO3, THF; iv, HCl, HOAc, 120 °C.
might help to avoid contamination of chemical impurities that
make it difficult to characterize charge carrier transport by time-
of-flight experiments.
The phase transition behaviors of both truxenes were
investigated by differential scanning calorimetry (DSC) and
polarized optical microscope (POM). Compound Tru4 did not
show clear mesophase transition upon heating (Figure S1a,
Supporting Information), only exhibiting one exothermic peak
at 250 °C, while exhibiting a mesophase from 220 to 209 °C
with a few dendritic textures observed in POM before
crystallization upon cooling (Figure S2a, Supporting Informa-
tion), and this irreversible phase transition is probably relative
to the strong crystallinity of Tru4. Furthermore, the resulting
mesophase was identified to be hexagonal columnar phase with
homeotropic alignment. In the polycrystalline phase, a lot of
cracks were observed (Figure S2b, Supporting Information). In
contrast to Tru4, Tru6 exhibited various thermal behaviors
(Figure S1b, Supporting Information). The DSC plot indicates
that two mesophases appeared in a temperature range from 127
to 175 °C in addition to the crystal phase below 127 °C and
isotropic phase over 175 °C in the heating process, while upon
cooling, the peak (11.36 J/g) at 127 °C is quite smaller than
that (22.08 J/g) in the first heating process, which is usually
identified to be the liquid crystal glass transition.14
Figure 3. XRD patterns of Tru6 in three phases, i.e., hexagonal
columnar phase, rectangular columnar phase, and glass phase (a: small
angle regions; b: wide angle regions), and illustration of columnar
lattice (c: hexagonal columnar lattice and d: rectangular columnar
phase). a, a′, and b′ represent lattice parameters.
peak appearing at 23.3° corresponds to the average core-to-core
correlation of 3.80 Å, which is assigned to a highly ordered
hexagonal columnar phase, combined with the intense low
angle peak indexed to (100) that is attributed to a typical
hexagonal unit cell with the parameter of a = 19.2 Å (Figure
3c). In particular, the molecular diameter of Tru6 is calculated
to be around 20 Å, which is a good agreement with the small
lattice parameter. This indicates that Tru6 molecules in
mesophases tend to arrange tightly to compress intermolecular
space, so that they sit in a restricted lattice. As expected, the
minimum distance among adjacent columns was estimated to
be less than 5 Å (core to core) as schematically illustrated in
Figure 1b, which is quite narrow compared with that of peri-
substituted DLC truxene (Figure 1a),11b,c which is so close to a
typical π−π interaction distance of 3.5 to 4 Å in a column.
Tru6 exhibited another columnar phase when cooled down
to 144 °C. Obviously, it is an ordered mesophase because of the
appearance of the peak (23.5°) at a wide angle region of the
XRD pattern at 140 °C. At the low angle region of XRD
pattern, the (100) face in the hexagonal lattice split into (200)
and (110), which is typical for an ordered rectangular columnar
phase (Colro) (Figure 3d). Indeed, the slight split supports that
the phase transition from Colho phase to Colro just induced a
trivial lattice variation. And also, the π−π interaction distance
can be determined to be 3.75 Å, which is a little smaller than
that in Colho phase, originated from a tilted molecular
conformation in Colro. This typical Colro lattice remained in a
Through observation of POM images of Tru6, its large
dendritic textures up to several hundred micrometers (Figure
2a) clearly indicated the presence of a highly ordered hexagonal
columnar phase in homeotropic alignment.15 Moreover, the
dendritic textures turned to be leaf-like with large domains
(Figure 2b) in the second liquid crystal phase, probably due to
the appearance of a higher order structure, and finally this
Figure 2. Normal and polarized optical microscope images of Tru6 in
different phases; a and b were taken without crossed polarizers, while
the insets in a and b were with crossed polarizers. All scale bars
represent 100 μm. The blue arrows point out defects.
5438
dx.doi.org/10.1021/cm502574d | Chem. Mater. 2014, 26, 5437−5440

Chemistry of Materials
Communication
room temperature range, forming a glass phase, in which the
peak at the wide angle region of the XRD pattern at 30 °C
presented a right shift indicating a π−π interaction distance of
3.70 Å narrower than those in the two mesophases (Figure 3b).
To characterize charge carrier transport in densely packed
columns formed in the windmill-shaped Tru6, we measured
transient photocurrents by TOF experiments.16 Samples of
Tru6 were capillary-filled into a 9.3 μm thick liquid crystal cell
with ITO electrodes at 180 °C in its isotropic phase. As shown
in Figure 4a,b, well-defined nondispersive transient photo-
respectively, as is often reported in the liquid crystalline
phases.19 This behavior can be well explained by a narrow
distribution of the density of states less than 60 meV
comparable to kT, where T is the temperature for TOF
experiments20 and is attributed to less disordered molecular
alignment in the column and a small dipole moment of Tru6.
Author: This glass state was metastable and turned to be
polycrystalline when the cell was kept at room temperature
overnight. In the resulting polycrystalline film, non-dispersive
photocurrent could be still observed (Figure S5, Supporting
Information) and enables us to determine that charge carrier
mobilities in the polycrystalline state are 0.2 and 0.17 cm2 V−1
s−1 for hole and electron, respectively, which are 2 orders of
magnitude higher than those of DLC alkyl-substituted-
benzocoronene derivatives.8d,21 They also hardly depended
on neither temperature (Figure 5) nor electric field (Figure S6,
Supporting Information) and were invariable for several weeks
(Figure S7, Supporting Information). Polycrystalline Tru4 also
showed high ambipolar mobility around 0.3 cm2 V−1 s−1
(Figure S8, Supporting Information), which is the highest in
all truxene derivatives.11b,c It is well-known that the non-
dispersive charge transport in polycrystalline thin films is rarely
observed in nonliquid crystalline organic materials because
defective grain boundaries cause deep trap states for charge
carriers. Therefore, the fast charge transport in the truxenes is
explained by preferential formation of grain boundaries along
with the columns22 and/or possible charge transport among the
columns thanks to the small intercolumnar space of 5 Å in the
present polycrystalline films.
In summary, we synthesized novel windmill-shaped discotic
liquid crystalline truxenes, i.e., 1,5,9-trialkyltruxene substituted
with alkyl chains at bay-positions of truxene core. These
truxenes showed highly ordered columnar phases with
homeotropic alignment upon cooling, in which a very narrow
intercolumnar space of 5 Å was estimated. The resulting
columnar structure was quite stable and so densely packed that
molecular motions in mesophases were thought to be confined.
Interestingly, these DLC materials showed fast ambipolar
carrier mobility over 0.1 cm2 V−1 s−1 in both columnar and
crystalline phases in spite of a relatively small core size of
truxene. These results suggest the windmill-shaped DLC
molecules with short alkyl substituents at bay positions give
densely packed columnar phases, which are advantageous over
conventional DLC molecules substituted with peripheral side
chains for fast charge carrier transport.3−5 This new molecular
design strategy is expected to be applied in various π-
conjugated systems, like phthalocyanine, tetrabenzoporphyrin,
and trithiatruxene, which will make discotic liquid crystals more
promising as organic semiconductors.
Figure 4. Typical transient photocurrents of Tru6 at 165 °C in time-
of-flight experiment; a: positive charge transport; b: negative charge
transport.
currents for positive and negative charges were observed in the
Colho phases of Tru6 at 165 °C. Each photocurrent exhibited a
clear shoulder, which indicates the transit time.
It is noted that mobility of hole and electron is almost
constant irrespective of columnar and glassy phase, even if the
phase structures are changed slightly as discussed above. In
mesophases, the mobility depended on neither temperature nor
electric field (Figure 5 and Figure S5, Supporting Information),
ASSOCIATED CONTENT
■
S
* Supporting Information
Details of synthesis and characterization of Tru4 and Tru6,
NMR, DSC, POM, and the results of transient photocurrents
through time-of-flight technique. This material is available free
of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: hanna@isl.titech.ac.jp.
Notes
Figure 5. Temperature dependence of charge transport mobility in
Tru6; the electric field was 1.1 × 104 V cm−1.
The authors declare no competing financial interest.
5439
dx.doi.org/10.1021/cm502574d | Chem. Mater. 2014, 26, 5437−5440

Chemistry of Materials
Communication
(b) Kajitani, T.; Suna, Y.; Kosaka, A.; Osawa, T.; Fujikawa, S.; Takata,
M.; Aida, T. J. Am. Chem. Soc. 2013, 135, 14564.
(16) Iino, H.; Hanna, J. I.; Haarer, D. Phys. Rev. B 2005, 72, 193203.
(17) Iino, H.; Hanna, J. I.; Bushby, R. J.; Movaghar, B.; Whitaker, B.
J.; Cook, M. J. Appl. Phys. Lett. 2005, 87, 132102.
ACKNOWLEDGMENTS
■
We thank Prof. Dr. H. Iino, Ms. Y. Takayashiki, and Dr. A.
Ohno for very useful discussion. This work was partly
supported by CREST under the JST Strategic Basic Research
Programs sponsored by Japan Science Agency.
(18) Liu, X. Y.; Usui, T.; Iino, H.; Hanna, J. I. J. Mater. Chem. C 2013,
1, 8186.
(19) (a) Pisula, W.; Tomovic, Z.; Simpson, C.; Kastler, M.; Pakula,
REFERENCES
■
T.; Mullen, K. Chem. Mater. 2005, 17, 4296. (b) Sergeyev, S.; Pisula,
̈
(1) (a) Prasanthkumar, S.; Saeki, A.; Seki, S.; Ajayaghosh, A. J. Am.
Chem. Soc. 2010, 132, 8866. (b) Lee, O. P.; Yiu, A. T.; Beaujuge, P. M.;
W.; Geerts, Y. H. Chem. Soc. Rev. 2007, 36, 1902. (c) Feng, X.;
Marcon, V.; Pisula, W.; Hansen, M. R.; Kirkpatrick, J.; Grozema, F.;
́
Woo, C. H.; Holcombe, T. W.; Millstone, J. E.; Frechet, J. M. Adv.
Andrienko, D.; Kremer, K.; Mullen, K. Nat. Mater. 2009, 8, 421.
̈
Mater. 2011, 23, 5359. (c) Takacs, C. J.; Sun, Y.; Welch, G. C.; Perez,
L. A.; Liu, X.; Wen, W.; Heeger, A. J. J. Am. Chem. Soc. 2012, 134,
16597.
(20) Ohno, A.; Hanna, J. I. Appl. Phys. Lett. 2003, 82, 751.
(21) Kastler, M.; Laquai, F.; Mullen, K.; Wegner, G. Appl. Phys. Lett.
̈
2006, 89, 252103.
(2) (a) Kim, B. G.; Jeong, E. J.; Chung, J. W.; Seo, S.; Koo, B.; Kim, J.
Nat. Mater. 2013, 12, 659. (b) Liu, X. Y.; Usui, T.; Hanna, J. I.
Chem.Eur. J. 2014, DOI: 10.1002/chem. 201403472. (c) Iino, H.;
́ ́
Hanna, J. I. Adv. Mater. 2011, 23, 1748. (d) Ide, J.; Mereau, R.;
Ducasse, L.; Castet, F.; Bock, H.; Olivier, Y.; Cornil, J.; Beljonne, D.;
Davino, G.; Roscioni, O. M.; Zannoni, L. M. J. Am. Chem. Soc. 2014,
136, 2911.
(22) Iino, H.; Takayashiki, Y.; Hanna, J. I.; Bushby, R. J. J. Appl. Phys.
2005, 44, 1310.
(3) (a) Adam, D.; Schuhmacher, P.; Simmerer, J.; Haussling, L.;
̈
Siemensmeyer, K.; Etzbachi, K. H.; Haarer, D. Nature 1994, 371, 141.
(b) Osawa, T.; Kajitani, T.; Hashizume, D.; Ohsumi, H.; Sasaki, S.;
Takata, M.; Aida, T. Angew. Chem. 2012, 124, 8114.
(4) (a) Cook, M. J.; Daniel, M. F.; Harrison, K. J.; McKeown, N. B.;
Thomson, A. J. J. Chem. Soc., Chem. Comm. 1987, 14, 1086.
(b) Gbabode, G.; Dumont, N.; Quist, F.; Schweicher, G.; Moser, A.;
Viville, P.; Geerts, Y. H. Adv. Mater. 2012, 24, 658.
(5) (a) Schmidt-Mende, L.; Fechtenkotter, A.; Mullen, K.; Moons,
̈
̈
E.; Friend, R. H.; MacKenzie, J. D. Science 2001, 293, 1119. (b) Van de
Craats, A. M.; Warman, J. M.; Fechtenkotter, A.; Brand, J. D.;
̈
Harbison, M. A.; Mullen, K. Adv. Mater. 1999, 11, 1469.
̈
(6) (a) Lemaur, V.; Da Silva Filho, D. A.; Coropceanu, V.; Lehmann,
M.; Geerts, Y.; Piris, J.; Cornil, J. J. Am. Chem. Soc. 2004, 126, 3271.
(b) Haverkate, L. A.; Zbiri, M.; Johnson, M. R.; Deme, B.; Mulder, F.
M.; Kearley, G. J. J. Phys. Chem. B 2011, 115, 13809. (c) Hansen, M.
R.; Feng, X.; Macho, V.; Mullen, K.; Spiess, H. W.; Floudas, G. Phys.
̈
Rev. Lett. 2011, 107, 257801. (d) Haverkate, L. A.; Zbiri, M.; Johnson,
M. R.; Deme, B.; Mulder, F. M.; Kearley, G. J. J. Phys. Chem. B 2012,
116, 3908.
(7) (a) Elmahdy, M. M.; Dou, X.; Mondeshki, M.; Floudas, G.; Butt,
H. J.; Spiess, H. W.; Mullen, K. J. Am. Chem. Soc. 2008, 130, 5311.
̈
(b) Mulder, F. M.; Stride, J.; Picken, S. J.; Kouwer, P. H.; De Haas, M.
P.; Siebbeles, L. D.; Kearley, G. J. J. Am. Chem. Soc. 2003, 125, 3860.
(8) (a) Bushby, R. J. Mater. Chem. 1999, 9, 2081. (b) Tatum, L. A.;
Johnson, C. J.; Fernando, A. A.; Ruch, B. C.; Barakoti, K. K.; Alpuche-
Aviles, M. A.; King, B. T. Chem. Sci. 2012, 3, 3261. (c) Jankowiak, A.;
́
Pociecha, D.; Szczytko, J.; Monobe, H.; Kaszynski, P. J. Am. Chem. Soc.
2012, 134, 2465. (d) Xiao, Q.; Sakurai, T.; Fukino, T.; Akaike, K.;
Honsho, Y.; Saeki, A.; Aida, T. J. Am. Chem. Soc. 2013, 135, 18268.
(9) Kreouzis, T.; Scott, K.; Donovan, K. J.; Boden, N.; Bushby, R. J.;
Lozman, O. R.; Liu, Q. Chem. Phys. 2000, 262, 489.
(10) Mizoshita, N.; Monobe, H.; Inoue, M.; Ukon, M.; Watanabe, T.;
Shimizu, Y.; Kato, T. Chem. Commun. 2002, 5, 428.
(11) (a) Simmerer, J.; Glusen, B.; Paulus, W.; Kettner, A.;
̈
Schuhmacher, P.; Adam, D.; Haarer, D. Adv. Mater. 1996, 8, 815.
(b) Isoda, K.; Yasuda, T.; Kato, T. Chem. Asian J. 2009, 4, 1619.
(c) Zhao, K. Q.; Chen, C.; Monobe, H.; Hu, P.; Wang, B. Q.; Shimizu,
Y. Chem. Commun. 2011, 47, 6290.
(12) (a) Vehoff, T.; Baumeier, B.; Andrienko, D. J. Chem. Phys. 2010,
133, 134901. (b) Jung, S. H.; Pisula, W.; Rouhanipour, A.; Rader, H. J.;
̈
Jacob, J.; Mullen, K. Angew. Chem., Int. Ed. 2006, 45, 4685.
̈
(13) Onyango, E. O.; Jacobi, P. A. J. Org. Chem. 2012, 77, 7411.
(14) (a) Wu, J.; Usui, T.; Hanna, J. I. J. Mater. Chem. 2011, 21, 8045.
(b) Lee, K. M.; Smith, M. L.; Koerner, H.; Tabiryan, N.; Vaia, R. A.;
Bunning, T. J.; White, T. J. Adv. Funct. Mater. 2011, 21, 2913.
(15) (a) Schweicher, G.; Gbabode, G.; Quist, F.; Debever, O.;
Dumont, N.; Sergeyev, S.; Geerts, Y. H. Chem. Mater. 2009, 21, 5867.
5440
dx.doi.org/10.1021/cm502574d | Chem. Mater. 2014, 26, 5437−5440