Substituent effect on the crystal packing and electronic coupling of tetrabenzocoronenes: A structure-property correlation

Article abstract of DOI:10.1039/c4tc00296b

Tetrabenzo[a,d,j,m]coronene (TBC) is a contorted polyaromatic molecule which shows near co-facial π-π stacking in the crystalline state and the high electronic coupling resulting from the packing renders it a potential candidate as a transistor material. Substitution at the periphery perturbs the packing due to steric as well as dipolar interactions and thus changes the electronic coupling between neighbouring molecules. In the light of the high sensitivity of charge mobility toward electronic coupling, a new series of TBC derivatives with substituents at 1-, 2-, 3-, 6-, 7-, 8-positions were designed, synthesized, and characterized. Needle-like single crystals were prepared using the physical vapor transport (PVT) method for these unsymmetrically substituted derivatives and were used for crystal structure analyses as well as the single crystal field-effect transistor (SCFET) device fabrication. The derivatives with fluoro-containing substituents exhibit anti-parallel cofacial or slightly shifted π-π stacking, whereas those with bulky alkyl substituents show skewed and more significantly shifted π-π stacking. A systematic comparison of the crystal packings and the calculated electronic couplings/charge mobilities with the measured SCFET mobilities shows a rough correlation and sheds light on the origin of the large hole-mobility of the SCFET with hexa-fluorinated TBC as the channel material.

Full text of DOI:10.1039/c4tc00296b

Journal of
Materials Chemistry C
View Article Online
View Journal | View Issue
PAPER
Substituent effect on the crystal packing and
electronic coupling of tetrabenzocoronenes: a
Cite this: J. Mater. Chem. C, 2014, 2,
structure–property correlation†
3928
a
b
a
c
a
*
Chi-Hsien Kuo, Ding-Chi Huang, Wei-Tao Peng, Kenta Goto, Ito Chao
ab
*
and Yu-Tai Tao
Tetrabenzo[a,d,j,m]coronene (TBC) is a contorted polyaromatic molecule which shows near co-facial p–p
stacking in the crystalline state and the high electronic coupling resulting from the packing renders it a
potential candidate as a transistor material. Substitution at the periphery perturbs the packing due to
steric as well as dipolar interactions and thus changes the electronic coupling between neighbouring
molecules. In the light of the high sensitivity of charge mobility toward electronic coupling, a new series
of TBC derivatives with substituents at 1-, 2-, 3-, 6-, 7-, 8-positions were designed, synthesized, and
characterized. Needle-like single crystals were prepared using the physical vapor transport (PVT) method
for these unsymmetrically substituted derivatives and were used for crystal structure analyses as well as
the single crystal field-effect transistor (SCFET) device fabrication. The derivatives with fluoro-containing
substituents exhibit anti-parallel cofacial or slightly shifted p–p stacking, whereas those with bulky alkyl
substituents show skewed and more significantly shifted p–p stacking. A systematic comparison of the
crystal packings and the calculated electronic couplings/charge mobilities with the measured SCFET
mobilities shows a rough correlation and sheds light on the origin of the large hole-mobility of the
SCFET with hexa-fluorinated TBC as the channel material.
Received 13th February 2014
Accepted 4th March 2014
DOI: 10.1039/c4tc00296b
www.rsc.org/MaterialsC
transport properties in organic semiconductors, since the afore-
mentioned extrinsic effects associated with polycrystalline lms
Introduction
can be minimized in single crystal devices. A comparison of
properties obtained with single crystal devices would enable a
systematic study on the structure–property correlation, delin-
eating the effect of dielectric surfaces on charge transport.5
Among the p-type SCFETs reported, rubrene yields the highest
reproducible hole mobility of up to 20 cm2 Vꢀ1 sꢀ1, and a
dependence of mobility on the choice of gate dielectric and the
direction of measurement were clearly demonstrated.6 To date,
the compounds which have the highest electron mobility
reported are SCFETs of N,N0-bis(n-alkyl)-(1,7 and 1,6)-dicyano-
perylene-3,4:9,10-bis(dicarboximide) (PDIF-CN2) and 5,7,12,14-
Organic eld-effect transistors (OFETs) have attracted much
attention due to their potential applications in cost effective
exible organic electronics.1 The versatility in structural manip-
ulation also makes OFETs a common approach for studying the
eld-dependence charge transport behaviors. Considerable effort
has been devoted to enhancing the charge carrier mobility (m)
through the design of a novel class of organic semiconductor
materials2 as well as the exploration of new OFET structures via
selection or modication of dielectric materials and electrodes.3
However, a fundamental understanding of the physical parame-
ters affecting the mobility, particularly the correlation between
molecular structure/crystal packing and charge carrier mobility,
is still limited mainly owing to the complexity of the morphology
involved in a thin lm deposited at the substrate surface. On the
other hand, organic single-crystal eld-effect transistors
(SCFETs)4 offer the opportunity to explore the intrinsic charge
tetrachloro-6,13-diazapentacene (TCDAP), both with mobilities of
a few cm2 Vꢀ1 sꢀ1
.
7
The charge transport between neighboring organic mole-
cules has been suggested to be determined by the transfer
integral (or electronic coupling) between the molecules, as well
as the reorganization energy involved in neutral/charged state
transformation for that particular molecule.8 It is widely
accepted that a minimization of reorganization energy and a
maximization of the transfer integral is preferred in molecular
design. However, it is also demonstrated by theoretical calcu-
lation that the transfer integral is highly sensitive to the relative
disposition of the neighboring molecules due to overlap of the
nodal regions of the highest occupied molecular orbitals
aInstitute of Chemistry, Academia Sinica, Taipei, Taiwan, Republic of China
bDepartment of Chemistry, National Tsing-Hua University, Hsin-chu, Taiwan, Republic
of China
cInstitute of Materials Chemistry and Engineering (IMCE), Kyushu University,
Fukuoka, 812-8581, Japan
† Electronic supplementary information (ESI) available: Details of synthetic
details and theoretical calculation are available. See DOI: 10.1039/c4tc00296b
3928 | J. Mater. Chem. C, 2014, 2, 3928–3935
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry C
(HOMOs) or the lowest unoccupied molecular orbitals stacking was observed from crystal structure analysis. Theoret-
(LUMOs), so that the highest transfer integral occurs for perfect ical calculations on the hole mobility along the p–p stacking
face-to-face packing9 and diminishes or vanishes for shied direction were carried out and experimental investigations were
p-stacking. Yet the co-facial p-stacking is usually not favored performed based on the SCFET conguration for a systematic
due to quadrupole repulsion for aromatic compounds.10 This is comparison of charge transport properties among various
evidenced by the common herringbone packing motif for most substituted derivatives. The eld-effect mobility was measured
unsubstituted linear acenes and oligothiophenes. Substitution and a correlation with the theoretical results is discussed.
with polar groups or bulky groups at the periphery of the linear
aromatics perturbs the packing for dipolar interactions or steric
interactions in a subtle way that the prediction of the packing of
organic molecules is challenging. The packing in turn affects
the morphology of the crystal that is grown.11
Experimental and computational
details
The series of derivatives of 1 were synthesized in the labora-
tory (the synthetic details are provided in the ESI†). Except for
alkylated compounds 1d and 1f, which were additionally
Contorted polyaromatics are conceived to have a high
tendency to pack face-to-face due to the potential of self-
complementary packing.12 A number of non-planar polycyclic
characterized by NMR, the rest were characterized by mass,
aromatic hydrocarbons have been shown to exhibit cofacial
elemental analyses and X-ray crystallography owing to their
stacking in single crystals.13 1,2,3,4,7,8,9,10-Tetrabenzocor-
poor solubility in common organic solvents. Single crystals
onene (TBC) is one such class of non-planar (or contorted)
were grown in a temperature-gradient copper tube by a vapor-
phase transfer method with argon as the carrier gas.16 The
X-ray diffraction was carried out mostly on a Bruker X8APEX
molecules due to the proximity of several C–H bonds.14
2,7,12,17-Tetraoctyl substituted TBC showed strong long-range
ordered p-stacking in the solution state by forming nanobers,
implying a co-facial packing. We have also demonstrated
recently that the TBC derivatives which bear four uoro groups
(TFTBC) or four chloro groups (TCTBC) at the symmetrical
2,7,12,17-positions exhibit near cofacial p–p stacking, while the
one with four methyl substituents exhibits signicantly shied
p–p stacking. Transistors based on their single crystal exhibit a
high eld-effect mobility from 0.1 to 0.7 cm2 Vꢀ1 sꢀ1 with a good
structure–property correlation in comparison to the theoretical
calculations.15 With the four substituents distributed at the
opposite sides of the molecule, an inevitable steric interaction
determines the relative shis in the otherwise co-facial packing
of neighboring molecules. It is contemplated that with
unsymmetrical substitutions, which generate a molecular
dipole moment, a new attractive molecular interaction and
relaxation of steric repulsion may result in a structural motif
that enhances electronic coupling and thus charge transport.
We report herein the synthesis and structure characteriza-
tion of a new series of TBC derivatives, with substituents at
1-, 2-, 3-, 6-, 7-, 8-positions (1a–h, Fig. 1) so that the molecules
are unsymmetrical. Single crystal structures were determined
for all eight derivatives. Needle-like crystals along the p–p
stacking direction were obtained. Cofacial or shied p–p
˚
X-ray diffractometer with Mo Ka radiation (l ¼ 0.71073 A) and
the structure was solved by SHELX 97 program. For
compounds 1d and 1f, the X-ray data were collected on a
Rigaku RAPID-HR Imaging Plate diffractometer with graphite
˚
monochromated Mo Ka radiation (l ¼ 0.71069 A) and a
Rigaku Saturn724 diffractometer using multi-layer mirror
˚
monochromated Mo-Ka radiation (l ¼ 0.71075 A). The
structure was solved using the direct method technique (SIR
2004 and SHELXL97) and a full-matrix least squares rene-
ment based on F2.
Devices were fabricated in a similar fashion as described in
our previous publication.15 In brief, the top-contact, top-gate
SCFET was prepared by placing the single crystal on a glass
substrate, which was modied with a monolayer of n-octade-
cyltrichlorosilane (ODTS). Painted colloidal graphite was used
as the source and drain electrodes. A thin lm of parylene-N
(ꢁ2–4 mm thick) was deposited on top of the crystal in a home-
made reactor as the insulating dielectric. Finally, colloidal
graphite was painted on top of the parylene lm as the gate
electrode. The channel length and width of devices varied
depending on the dimensions of crystals chosen. These
parameters, as well as the thickness of parylene, were deter-
mined for each individual device by using a Veeco Daktak 150
Stylus proler. The electrical characteristics of the devices were
measured in an ambient atmosphere in a dark chamber using a
computer-controlled Agilent 4156C Semiconductor Parameter
Analyzer. The eld-effect mobility of the OFET devices was
calculated from the I–V characteristics in the saturation region
(ISD,sat) according to eqn (1):17
ISD,sat ¼ (WCi/2L)msat(VG ꢀ Vth)2
(1)
where W and L are the channel width and length respectively,
and Ci is the capacitance per unit area of the parylene-N insu-
lator, VG is the gate voltage and Vth is the threshold voltage.
A computational procedure for calculating the charge carrier
Fig. 1 Molecularstructure of substituted tetrabenzo[a,d,j,m]coronene (1). mobility has been elaborated previously.18 Briey, the
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. C, 2014, 2, 3928–3935 | 3929

View Article Online
Journal of Materials Chemistry C
Paper
calculations of internal reorganization energy (l) of 1a–h were
performed with the Gaussian 09 program19 at the level of B3LYP/
6-31G(d,p).8,20 The transfer integral (t) between neighboring
dimer pairs in the crystal structure was calculated with ADF21 at
the PW91PW91/DZP level of theory.22 With l and t, Marcus theory
for self-exchange electron transfer was used to obtain the carrier
transport rate at 300 K. With the rate and the intermolecular
distance obtained from the crystal structure, the carrier diffusion
coefficient was deduced and used in the Einstein relationship to
derive the carrier mobility. Although the equation has certain
limitations such as the neglection of nuclear tunneling effect and
the nonlocal couplings, it can be deemed acceptable for the
purposes of qualitative analysis and molecular design.23 It is
noted that the X-ray coordinates of 1b and 1g are with disorders
for the uorine atoms and the hydrogen atoms at the corre-
sponding positions (e.g., 2,7-positions of 1b and 1,3,6,8-positions
of 1g). Because of the uncertainty of the F/H positions, while
calculating t for these two molecules, the C–F bond length and
C–H bond length at corresponding positions were xed at 1.350
Characterization
The compounds appear yellow (1a–d, 1f) to orange yellow (1e,
1g, and 1h). Except for compounds 1g and 1h, which were not
soluble in any common organic solvents, similar UV-Vis
absorption spectra of compound 1a–f in dichloromethane
exhibiting multiple lmax were obtained in the 300–500 nm
region (Fig. 2). The p-band (p / p* transition) in the long
wavelength region for compound 1a occurs at 423 nm. A small
substituent effect on the absorption maximum can be observed,
with di-chloro (1c), di-methyl (1d), di-triuoromethyl (1e), and
di-t-butyl (1f) substitution giving a red-shi (ꢁ3–4 nm) and di-
uoro (1b) substitution giving blue-shi (ꢁ3 nm) relative to that
of the parent compound. Very weak a-band (n / p*) was
discernible for all six compounds at ꢁ456–458 nm.
The HOMO energy levels for compound 1 were measured
using a photoelectron spectrometer (AC2) in the solid state. The
HOMO levels ranged from 5.17 to 6.09 eV, with electronegative
substituents such as uorine and chlorine shiing the HOMO
level down and a donation group such as methyl and a t-butyl
group shi the HOMO level up. This is in good accordance with
the inductive effect of substituents. The results are summarized
in Table 1.
Thermal gravimetric analyses showed that the series of
compounds have excellent thermal stability with decomposi-
tion temperature (5 to 8% weight loss) at around 470 ꢂC and
above in a nitrogen atmosphere, as illustrated in Fig. 3.
˚
and 0.965 A, respectively. Parallel and antiparallel dimeric rela-
tionships were both considered for 1b and 1g, but they do not
change the magnitude of t signicantly. For 1f and 1h, there are
nonequivalent dimer relationships along the 1-D charge-trans-
porting-stacking direction. Because a charge can only move along
this channel, the smallest t among the dimers will result in a
bottleneck of the 1-D charge transport. This t is used in Table 2 to
give a rough estimate of the charge mobility. For all the calcu-
lation results, see the ESI.†
Crystal structure analyses and theoretical results
Results and discussion
All eight compounds give needle-like crystals. X-ray analyses
revealed a contorted geometry as expected. Similar to our
previous observations, the contorted molecules have two
stereoisomers.15 One has two benzenoids in the right side
pointing in the same direction and opposite to the two in the
le side, as illustrated in Fig. 4a. Compounds 1a–c, 1g, and 1h,
Except for the parent compound 1a, whose synthesis and
characterization have been described in our previous publica-
tion,15 the synthetic approach of unsymmetrical compound
1b–h is summarized in Scheme 1. In brief, Corey–Fuchs24
reaction was rst carried out by treating the commercially
available 10-diphenylmethylene-9(10H)-anthracenone with tri-
phenylphosphine and carbon tetrabromide to afford compound
2. Suzuki–Miyaura25 reaction between 2 and substituted boronic
acid led to the key intermediate bisolen 3b–h, respectively.
Oxidative dehydrogenation of 3 gave a mixture of half-cyclized
products which, without isolation, were converted into fully
cyclized products 1 by Scholl cyclization using FeCl3 as the
catalyst.26 All products were puried by vacuum-sublimation
and obtained in good yield.
Fig. 2 UV-Vis spectra for 1a–f.
Table 1 HOMO energy levels of tetrabenzocoronenes derivatives
Scheme 1 Synthesis of tetrabenzocoronenes with unsymmetrical
substitutions. Reagents and conditions: (I) CBr4, PPh3, 80 ꢂC; (II)
boronic acid, K2CO3, Pd(PPh3)4, toluene, ethanol/H2O; (III) I2,
propylene oxide, benzene, hn; (IV) FeCl3, CH2Cl2.
1a
1b
1c
1d
1e
1f
1g
1h
HOMO (ꢃ0.1 eV)
5.4
5.6
5.5
5.5
6.1
5.2
5.6
5.8
3930 | J. Mater. Chem. C, 2014, 2, 3928–3935
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry C
which bear compact substituents (H, F, and Cl) adopt this
geometry. In contrast, compounds 1d–f, which bear bulky
substituents (CH3, CF3, t-butyl, respectively), have the two
benzenoids in each side pointing in the opposite direction
(Fig. 4b).
Among the compounds with small substituents, 1a, 1b, 1c,
and 1g pack nearly face-to-face with p–p stacking distances of
˚
˚
3.64–3.66 A and centroid-to-centroid distances of 3.75–3.77 A
(Fig. 5a and b and Table 2). The spatial arrangement of crystal
packing for minimizing the steric repulsion and maximizing
intermolecular interactions leads to a shi of molecular
packing between layers, which can be described with displace-
ments along the long (dl) and short (ds) molecular axes. For the
parent compound 1a, dihalogenated 1b, 1c, and tetra-
Fig. 3 Thermal gravimetric analyses of 1.
˚
uorinated 1g, the dl values are between 0.01 and 0.19 A, and
˚
the ds values are between 0.86 and 0.96 A. This pattern of
p-stacking with a more signicant shi along the short axis
than the long axis is similar to that of 2,7,12,17-tetrauoro- and
tetrachloro-TBCs we reported previously (TFTBC and TCTBC in
Table 2).15 Compared to the perfectly face-to-face sandwich
packing, shiing along the short molecular axis can effectively
alleviate the H/H repulsions between the upward/downward
pointing hydrogen atoms at the 4,5,14,15-postions of two con-
torted TBC molecules. Take the parent compound (1a) as an
example, if two 1a molecules were held in the sandwich packing
Fig. 4 (a) The structure observed for TBC with small substituents (1a–
c, 1g, and 1h). (b) The structure observed for TBC with bulky substit-
uents (1d–f).
˚
position with the p–p distance found in crystal (3.66 A), close
˚
H/H contacts would be found (2.02 A, as shown in Fig. 6).
Whereas in the crystal structure the distance between the same
˚
pair of hydrogen atoms is 2.642 A, which is larger than the sum
˚
of the van der Waals radii (2.4 A). In contrast, compound hex-
auorinated 1h exhibits shied p–p stacking mainly along the
˚
long molecular axis with its dl value as large as 1.96 A and a
˚
rather small ds value (0.24 A; Fig. 5c; Table 2). Shiing along the
long molecular axis also can avoid the H/H repulsions just
mentioned. The possibility of an additional electrostatic driving
force for shiing along the long axis is implied by the molecular
electrostatic potential (MEP) of 1h, which features an electro-
positive face and an electronegative edge due to extensive
uorination (Fig. 7a). By shiing along the long axis, the elec-
tropositive face can interact with the electronegative face of the
antiparallel molecule in the next layer, and the electronegative
uorine edge along the long molecular axis can interact with the
electropositive hydrogen edge in the next layer (Fig. 7b).
It is noted that compounds 1b and 1g have disorders in their
crystal structures, probably due to packing in an anti-parallel
manner between layers (Fig. 5a). This anti-parallel packing
appears in another uoro-substituted compound 1h, but
without crystal disorder. The dichloro-substituted 1c, however,
packs parallel between layers probably due to Cl/Cl and Cl/H
interactions (Fig. 5b). Comparing the crystal structures of
unsymmetrical 1b, 1c and 1g to those of symmetrical
Fig. 5 (a) Crystal packing of 1b viewed from the top of the molecules
(b axis) and down the long molecular axis (a axis), respectively. Because
of disorders in the X-ray coordinates, half-occupancies of hydrogen
and fluorine atoms are found at the 2,7,12,17-positions. (b) Crystal substituted 2,7,12,17-tetrachloro-TBC (TCTBC) and 2,7,12,17-
tetrauoro-TBC (TFTBC),15 one would notice that the dipole–
packing of 1c viewed from the top of the molecules (b axis) and down
the long molecular axis (c axis), respectively. (c) Crystal packing of 1h
viewed down the long molecular axis (c axis) and from the top of the
molecules (a axis), respectively and (d) crystal packing of 1f viewed
from the top of the molecules (a axis) and down the b axis,
dipole interaction is not a determining factor for crystal engi-
neering of TBCs toward close p-stacking for stronger coupling.
First of all, unsymmetrical 1g packs in an anti-parallel manner
respectively.
as one would expect by considering an ideal arrangement of
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. C, 2014, 2, 3928–3935 | 3931

View Article Online
Journal of Materials Chemistry C
Paper
Table 2 Intermolecular distances along the cell axis relevant to p-stacking (daxis), p–p stacking distances measured with the use of molecular
mean planes (dp–p), molecular centroid-to-centroid distances (dc–c), short molecular axis shifts (ds), long molecular axis shifts (dl) in the crystal
structures and the calculated reorganization energies (l+), electronic couplings (t+), and mobilities (m+) for hole transport
l+ (meV)
t+ (meV)
m+ (cm2 Vꢀ1 sꢀ1
)
˚
˚
˚
˚
˚
Compounds
daxis (A)
dp–p (A)
dc–c (A)
ds (A)
dl (A)
1a
1b
1c
1d
1e
1f
1g
1h
3.77
3.77
3.77
3.74
3.65
3.84
3.75
3.58
3.72
3.75
3.66
3.66
3.66
3.64
—
—
—
3.65
3.58
3.61
3.63
—
3.77
3.77
3.77
4.23
4.00
6.39
3.75
4.06
3.72
3.75
4.39
0.87
0.87
0.96
—
—
—
0.86
0.24
0.84
0.76
—
0.01
0.19
0.17
—
—
—
0.01
1.96
0.34
0.55
—
132
151
143
134
148
136
135
147
169
154
136
41
22
22
7
35
40
41
59
41
51
20
0.58
0.13
0.15
0.02
0.33
0.56
0.56
0.91
0.36
0.68
0.13
TFTBC
TCTBC
TMTBC
Therefore, the crystal structure of all three compounds can still
be categorized as 1-D packing. It is thus not surprising that
needle-like crystals are found for 1d–1f.
Some selected cell parameters are listed in Table 3. The
theoretical results of reorganization energies (l+), electronic
couplings (t+) and charge transfer mobility (m+) for hole trans-
port are shown in Table 2.
Fig. 6 Close H/H contacts found in a pair of 1a molecules stacked
face-to-face with the p–p distance found in the crystal structure of 1a.
Because the substituents are attached to positions with small
wave function coefficients in the HOMO, the effect of substit-
uents on l+ is relatively small for 1a–h (132 to 151 meV) in
comparison to that for TFTBC, TCTBC, and 2,7,12,17-tetra-
methyl-TBC (TMTBC) (136 to 169 meV).15 Small variations in l+
in a series of compounds could facilitate the structure–property
correlation between crystal packing and mobility. For example,
it is difficult to deduce the packing-mobility relationship
between pentacene and peruoropentacene because their
reorganization energies differ by ca. 100 meV.27 The large
overlapping molecular surface along the p-stacking direction
Fig. 7 (a) Molecular electrostatic potential of 1h and (b) dimer of 1h
with the top molecule shown in a ball and stick model and arranged in (b axis for 1a–c, 1e, and 1g; a axis for 1f and 1h; c axis for 1d)
the same orientation as that shown in (a) and the antiparallel bottom
molecule shown in wireframe.
leads to 1-D packing and charge transporting. Therefore, we
employ the unit length of the cell axis relevant to p-stacking to
deduce the intermolecular distance (daxis), which is then used in
calculating the mobility of the charge carrier.28 The largest
mobility thus obtained is 0.91 cm2 Vꢀ1 sꢀ1 for hexauoro
molecular dipole moments, but it does not show a shorter p/p
distance than that of symmetrical TFTBC (Table 2). Secondly
derivative 1h. As pointed out previously, the uniqueness of 1h is
and more importantly, dichloro 1c packs in a parallel arrange-
ment between layers, which is not favorable in terms of the
alignment of molecular dipole moments.
˚
that it has a large dl (1.96 A), but has the smallest shi ds
˚
(0.24 A). The other compounds that have relatively large
With larger substituents than 1a–1c, methyl-substituted 1d,
triuoromethyl-substituted 1e, and t-butyl-substituted 1f
(Fig. 5d) do not pack in a parallel manner. Because the molec-
ular planes of neighboring molecules in the p stacking direc-
tion are unparalleled to each other, it is not possible to obtain
dl, ds and p/p distances for these compounds. For 1d and 1f,
there are dimers in which the molecules rotating 90 degrees
between layers in the crystal structures (e.g., perpendicular
dimers 2–3 in Fig. 5d), while for 1e the rotation is 180 degrees
and molecules shi along the long molecular axis. Along the
p-stacking direction, molecules in any given dimer pair stack
with a signicant surface area overlapping with each other.
Table 3 Selected cell parameters of single crystals of 1
1a
1b
1c
1d
1e
1f
1g
1h
˚
a (A)
16.25
3.77
18.72
28.19 22.60 32.24 16.64 11.53 16.62
7.15
˚
b (A)
3.77
21.91 28.62 7.48
90 90 90
b (deg.) 106.55 97.00 97.39 90
g (deg.) 90 90 90 90
3.77
20.37 7.30
18.24 3.75
12.24
14.93
74.98
˚
c (A)
22.90 23.46 19.26
a (deg.) 90
90
87.11 90
97.95 81.25 105.60 86.09
90
77.89 90
73.56
ꢀ
P1
ꢀ
Space
group
P21/n
C2/c P21/n Pbca P21/n P1
P21/n
3932 | J. Mater. Chem. C, 2014, 2, 3928–3935
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry C
predicted mobilities are 1a, 1f, and 1g. Compounds 1a and 1g
have very similar dl and ds values, whereas 1f has a perpendic-
ular dimer relationship in the crystal structures.29
SCFET device properties
Crystallites grown using the vapor phase transfer method
were carefully examined under an optical microscope for their
integrity and chosen for device fabrication. The single crystal
eld-effect transistors were prepared in the same fashion for
all compounds, which exhibited typical p-type behavior. Five
to nine samples for each compound were prepared and
measured. The analyzed device characteristics are summa-
rized in Table 4. The averaged values of hole-mobility of 1h,
1f, 1a, and 1d are noticeably larger than those of 1c, 1e, 1g,
and 1b, and computational results gave plausible trends
except for 1d and 1g. The exact source of discrepancy in the
measured and calculated mobility is unclear. A worth-noting
observation is that 1b and 1g exhibit the lowest mobility (both
0.06 cm2 Vꢀ1 sꢀ1) among these compounds and both
compounds are with an uncertainty of uorine positions in
the crystal structures. For compound 1a, a negligible decrease
of the averaged hole-mobility with a minimal deviation
compared to our previous publication is obtained.15 The
typical output and transfer characteristics for compound 1d
are shown in Fig. 8.
Fig. 9 The highest occupied molecular orbital of hexafluorinated 1h.
mostly distributed along the short axis, but not the long axis
(see Fig. 9 for HOMO of 1h). As a result, shiing along the
long axis can still result in high t+. In Table 2, compounds
with the largest t+ values (49 meV for 1h and 51 meV for
TCTBC) have larger dl than other compounds. Finally, it
should be pointed out that TBCs with small substituents tend
to pack with molecular planes parallel between layers, but
TBCs with larger substituents (CH3, CF3, and t-butyl) tend to
pack with their molecular planes unparallel to each other.
Therefore, large substituents at the right positions could be
the stepping stones to design high performance TBCs with
2-D packing.
As predicted by theoretical calculations, 1h exhibits the
highest value of hole-mobility among these compounds with
mobility exceeding 1 cm2 Vꢀ1 sꢀ1. Data in Table 2 show that
the large mobility originates from the high electronic
coupling value, which can in turn be rationalized by TBCs'
HOMO distribution. The nodal planes of HOMOs of TBCs are
Conclusion
In conclusion, a new series of non-planar and fused aromatic
compounds, namely tetrabenzo[a,d,j,m]coronenes with
unsymmetrical substitution were synthesized, characterized,
and their single crystals were prepared for X-ray analyses as
well as SCFET device fabrication. The strong dipole repulsion
of the uoro-containing derivatives leads to molecular
packing in an anti-parallel manner, whereas the bulky alkyl
substituents cause the molecules between layers to shi
signicantly in crystal packing. The subtle structural pertur-
bation resulting from the substitution lead to variation in the
transfer integral as well as reorganization energy and thus the
mobility, as calculated using the hopping model. The calcu-
lated data seem to be in reasonable agreement with the
experimental measurements. Among these, the SCFET with
hexa-uorinated 1h as the channel material and parylene-N as
the gate insulator gave a hole-mobility as high as 1.19 cm2 Vꢀ1
sꢀ1. The result suggests that the perturbation caused by
substituents in terms of shis in the packing may have
different effects on the transfer integral, depending on the
distribution of the HOMO orbital.
Table 4 SCFET performance of tetrabenzocoronene derivatives
Mobility
Average
mobility (mavg
On/off
ratio
Threshold
voltage (Vth, V)
(m, cm2 Vꢀ1 sꢀ1
)
)
1a
1b
1c
1d
1e
1f
0.35–0.44
0.027–0.10
0.10–0.17
0.23–0.30
0.09–0.16
0.35–0.64
0.044–0.090
0.842–1.19
0.38 ꢃ 0.030
0.060 ꢃ 0.025
0.13 ꢃ 0.027
0.26 ꢃ 0.026
0.11 ꢃ 0.0030
0.45 ꢃ 0.12
104–105
103–105
102–104
103–105
104–105
103–105
102–104
103–105
(ꢀ26)–(ꢀ41)
(ꢀ17)–(ꢀ31)
(ꢀ7)–(ꢀ32)
(ꢀ7)–(ꢀ34)
(ꢀ18)–(ꢀ38)
(ꢀ10)–(ꢀ18)
(ꢀ21)–(ꢀ28)
(ꢀ30)–(ꢀ35)
1g
1h
0.061 ꢃ 0.014
1.06 ꢃ 0.13
Acknowledgements
The authors wish to thank the National Science Council, Taiwan
and Academia Sinica for the nancial support of the work. They
also thank the National Center for High-Performance
Computing and the Computing Center of Academia Sinica for
providing computational resources.
Fig. 8 (a) Output and (b) transfer characteristics of 1d.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. C, 2014, 2, 3928–3935 | 3933

View Article Online
Journal of Materials Chemistry C
Paper
12 S. Xiao, M. Myers, Q. Miao, S. Sanaur, K. Pang,
M. L. Steigerwald and C. Nuckolls, Angew. Chem., Int. Ed.,
2005, 44, 7390–7394.
13 (a) M. M. Islam, F. Valiyev, H. Lu, M. Kuo, I. Chao and
Y.-T. Tao, Chem. Commun., 2011, 47, 2008–2010; (b)
I. Oonishi, S. Fujisawa, J. Aoki, Y. Ohashi and Y. Sasada,
Bull. Chem. Soc. Jpn., 1986, 59, 2233–2238; (c) S. Fujisawa,
I. Oonishi, J. Aoki, Y. Ohashi and Y. Sasada, Bull. Chem.
Soc. Jpn., 1982, 55, 3424–3428.
Notes and references
1 (a) G. Gelinck, P. Heremans, K. K. Nomoto and
T. D. Anthopoulos, Adv. Mater., 2010, 22, 3778–3798; (b)
M. Muccini, Nat. Mater., 2006, 5, 605–613; (c)
C. D. Dimitrakopoulos and P. R. L. Malenfant, Adv. Mater.,
2002, 14, 99–117; (d) H. Dong, C. Wang and W. Hu, Chem.
Commun., 2010, 46, 5211–5222; (e) H. Klauk, Chem. Soc.
Rev., 2010, 39, 2643–2666; (f) A. R. Murphy and
J. M. J. Frchet, Chem. Rev., 2007, 107, 1066–1096; (g)
M. E. Roberts, A. N. Sokolov and Z. Bao, J. Mater. Chem.,
2009, 19, 3351–3363.
14 X. Zhang, X. Jiang, K. Zhang, L. Mao, J. Luo, C. Chi,
H. S. O. Chan and J. Wu, J. Org. Chem., 2010, 75, 8069–
8077.
15 S. Pola, C.-H. Kuo, W.-T. Peng, M. M. Islam, I. Chao and
Y. T. Tao, Chem. Mater., 2012, 24, 2566–2571.
2 C. Wang, H. Dong, W. Hu, Y. Liu and D. Zhu, Chem. Rev.,
2012, 112, 2208–2267.
16 V. Podzorov, V. M. Pudalov and M. E. Gershenson, Appl. Phys.
Lett., 2003, 82, 1739–1741.
3 C. A. Di, Y. Q. Liu, G. Yu and D. B. Zhu, Acc. Chem. Res., 2009,
42, 1573–1583.
17 S. M. Sze, Semiconductor Device Physics and Technology,
Wiley, New York, 2nd edn, 2001, ch. 6, p. 169.
18 M.-Y. Kuo, H.-Y. Chen and I. Chao, Chem.–Eur. J., 2007, 13,
4750–4758.
4 (a) V. Podzorov, MRS Bull., 2013, 38, 15–24; (b) L. Jiang,
H. Dong and W. Hu, J. Mater. Chem., 2010, 20, 4994–5007;
(c) C. Reese and Z. N. Bao, J. Mater. Chem., 2006, 16, 329–
333; (d) R. W. I. de Boer, M. E. Gershenson, A. F. Morpurgo
and V. Podzorov, Phys. Status Solidi, 2004, 201, 1302–1331;
(e) M. E. Gershenson, V. Podzorov and A. F. Morpurgo, Rev.
Mod. Phys., 2006, 78, 973–989; (f) J. Takeya, C. Goldmann,
S. Haas, K. P. Pernstich, B. Ketterer and B. Batlogg, J. Appl.
Phys., 2003, 94, 5800–5804; (g) R. W. I. de Boer,
T. M. Klapwijk and A. F. Morpurgo, Appl. Phys. Lett., 2003,
83, 4345–4347.
5 (a) D. Boudinet, M. Benwadih, S. Altazin, J.-M. Verilhac, E. De
Vito, C. Serbutoviez, G. Horowitz and A. Facchetti, J. Am.
Chem. Soc., 2011, 133, 9968–9971; (b) M. M. Islam, S. Pola
and Y.-T. Tao, ACS Appl. Mater. Interfaces, 2011, 3, 2136–2141.
6 (a) V. C. Sundar, J. Zaumseil, V. Podzorov, E. Menard,
R. L. Willett, T. Someya, M. E. Gershenson and
J. A. Rogers, Science, 2004, 303, 1644–1646; (b) J. Takeya,
M. Yamagishi, Y. Tominari, R. Hirahara, Y. Nakazawa,
T. Nishikawa, T. Kawase, T. Shimoda and S. Ogawa, Appl.
Phys. Lett., 2007, 90, 102120.
19 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino,
G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida,
T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,
V. N. Staroverov, R. Kobayashi, J. Normand,
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
¨
J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox,
Gaussian 09, A.1 Revision, Gaussian Inc., Wallingford CT,
2009.
7 (a) A. S. Molinari, H. Alves, Z. Chen, A. Facchetti and
A. F. Morpurgo, J. Am. Chem. Soc., 2009, 131, 2462–2463;
(b) N. A. Minder, S. Ono, Z. Chen, A. Facchetti and
A. F. Morpurgo, Adv. Mater., 2012, 24, 503–508; (c)
M. M. Islam, S. Pola and Y.-T. Tao, Chem. Commun., 2011,
47, 6356–6358.
20 For calculation details and performance of this theory level
on internal reorganization energy, see (a) N. E. Gruhn,
D. A. da Silva Filho, T. G. Bill, M. Malagoli, V. Coropceanu,
´
A. Kahn and J. L. Bredas, J. Am. Chem. Soc., 2002, 124,
8 V. Coropceanu, J. Cornil, D. A. da Silva, Y. Olivier, R. Silbey
and J. L. Bredas, Chem. Rev., 2007, 107, 926–952.
9 O. Kwon, V. Coropceanu, N. E. Gruhn, J. C. Durivage,
J. G. Laquindanum, H. E. Katz, J. Cornil and J. L. Bredas,
J. Chem. Phys., 2004, 120, 8186–8194.
7918–7919; (b) M. Malagoli, V. Coropceanu, D. A. da Silva
and J. L. Bredas, J. Chem. Phys., 2004, 120, 7490–7496; (c)
Y.-C. Chang and I. Chao, J. Phys. Chem. Lett., 2010, 1, 116–
121.
21 G. te Velde, F. M. Bickelhaupt, E. J. Baerends, C. Fonseca
Guerra, S. J. A. van Gisbergen, J. G. Snijders and T. Ziegler,
J. Comput. Chem., 2001, 22, 931–967.
22 (a) K. Senthilkumar, F. C. Grozema, F. M. Bickelhaupt and
L. D. A. Siebbeles, J. Chem. Phys., 2003, 119, 9809–9817; (b)
P. Prins, K. Senthilkumar, F. C. Grozema, P. Jonkheijm,
A. P. H. J. Schenning, E. W. Meijer and L. D. A. Siebbeles,
J. Phys. Chem. B, 2005, 109, 18267–18274.
10 E. V. Anslyn and D. A. Dougherty, Modern Physical Organic
Chemistry, University Science Book, Sausalito, CA, 2006.
11 (a) H. Moon, R. Zeis, E. J. Borkent, C. Besnard, A. J. Lovinger,
T. Siegrist, C. Kloc and Z. N. Bao, J. Am. Chem. Soc., 2004, 126,
15322–15323; (b) J. E. Anthony, J. S. Brooks, D. L. Eaton and
S. R. Parkin, J. Am. Chem. Soc., 2001, 123, 9482–9483; (c)
J. E. Anthony, D. L. Eaton and S. R. Parkin, Org. Lett., 2002,
4, 15–18.
3934 | J. Mater. Chem. C, 2014, 2, 3928–3935
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry C
23 (a) A. N. Sokolov, S. Atahan-Evrenk, R. Mondal, 26 (a) T. S. Navale, K. Thakur and R. Rathore, Org. Lett., 2011,
´
H. B. Akkerman, R. S. Sanchez-Carrera, S. Granados-Focil,
13, 1634–1637; (b) K. N. Plunkett, K. Godula, C. Nuckolls,
N. Tremblay, A. C. Whalley and S. Xiao, Org. Lett., 2009, 11,
2225–2228.
J. Schrier, S. C. B. Mannsfeld, A. P. Zoombelt, Z. Bao and
A. Aspuru-Guzik, Nat. Commun., 2011, 2, 437; (b) V. Stehr,
J. Pster, R. F. Fink, B. Engels and C. Deibel, Phys. Rev. B: 27 H. Y. Chen and I. Chao, Chem. Phys. Lett., 2005, 401, 539–
Condens. Matter Mater. Phys., 2011, 83, 155208; (c) Y. Song, 545.
C. Di, X. Yang, S. Li, W. Xu, Y. Liu, L. Yang, Z. Shuai, 28 In our previous study on TBCs, dc–c was used to calculate
D. Zhang and D. Zhu, J. Am. Chem. Soc., 2006, 128, 15940–
15941; (d) W. Q. Deng and W. A. Goddard III, J. Phys.
Chem. B, 2004, 108, 8614–8621; (e) H. Li, J.-L. Bredas and
C. Lennartz, J. Chem. Phys., 2007, 126, 164704; (f) S. E. Koh,
to the mobility. The only compound affected by
replacing dc–c with daxis is TMTBC, whose mobility
changed from 0.18 to 0.13 cm2 Vꢀ1 sꢀ1 as shown in
Table 2.
´
C. Risko, D. A. da Silva, O. Kwon, A. Facchetti, J.-L. Bredas, 29 Along the p-stacking direction of the crystal structure, all
T. J. Marks and M. A. Ratner, Adv. Funct. Mater., 2008, 18,
dimers are perpendicular to 1d and there is only one t+
value for dimeric electronic coupling. For 1f, the
dimeric relationship along the p-stacking direction is
more complicated as shown in Fig. 5d. The t+ values for
332–340.
24 P. M. Donovan and L. T. Scott, J. Am. Chem. Soc., 2004, 126,
3108–3112.
0
25 N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457–
dimer1–2, dimer2–3, and dimer3–1 are 56, 40, and 79 meV,
2483.
respectively.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. C, 2014, 2, 3928–3935 | 3935