Fused thiophene semiconductors: Crystal structure-film microstructure transistor performance correlations

Article abstract of DOI:10.1002/adfm.201203439

The molecular packing motifs within crystalline domains should be a key determinant of charge transport in thin-film transistors (TFTs) based on small organic molecules. Despite this implied importance, detailed information about molecular organization in polycrystalline thin films is not available for the vast majority of molecular organic semiconductors. Considering the potential of fused thiophenes as environmentally stable, high-performance semiconductors, it is therefore of interest to investigate their thin film microstructures in relation to the single crystal molecular packing and OTFT performance. Here, the molecular packing motifs of several new benzo[d,d′]thieno[3,2-b;4,5- b′]dithiophene (BTDT) derivatives are studied both in bulk 3D crystals and as thin films by single crystal diffraction and grazing incidence wide angle X-ray scattering (GIWAXS), respectively. The results show that the BTDT derivative thin films can have significantly different molecular packing from their bulk crystals. For phenylbenzo[d,d′]thieno[3,2-b;4,5-b′] dithiophene (P-BTDT), 2-biphenylbenzo[d,d′]thieno-[3,2-b;4,5-b′] dithiophene (Bp-BTDT), 2-naphthalenylbenzo[d,d′]thieno[3,2-b;4,5-b′] dithiophene (Np-BTDT), and bisbenzo[d,d′]thieno[3,2-b;4,5-b′] dithiophene (BBTDT), two lattices co-exist, and are significantly strained versus their single crystal forms. For P-BTDT, the dominance of the more strained lattice relative to the bulk-like lattice likely explains the high carrier mobility. In contrast, poor crystallinity and surface coverage at the dielectric/substrate interface explains the marginal OTFT performance of seemingly similar PF-BTDT films. Copyright

Full text of DOI:10.1002/adfm.201203439

www.afm-journal.de
www.MaterialsViews.com
Fused Thiophene Semiconductors: Crystal Structure–Film
Microstructure Transistor Performance Correlations
Jangdae Youn, Sumit Kewalramani, Jonathan D. Emery, Yanrong Shi, Shiming Zhang,
Hsiu-Chieh Chang, You-jhih Liang, Chia-Ming Yeh, Chieh-Yuan Feng, Hui Huang,
Charlotte Stern, Liang-Hsiang Chen, Jia-Chong Ho, Ming-Chou Chen,*
Michael J. Bedzyk,* Antonio Facchetti,* and Tobin J. Marks*
1. Introduction
The molecular packing motifs within crystalline domains should be a key
Over the past decade, the field of organic
determinant of charge transport in thin-film transistors (TFTs) based on
thin-film transistors (OTFTs) has seen
small organic molecules. Despite this implied importance, detailed infor-
mation about molecular organization in polycrystalline thin films is not
available for the vast majority of molecular organic semiconductors. Con-
sidering the potential of fused thiophenes as environmentally stable, high-
performance semiconductors, it is therefore of interest to investigate their
thin film microstructures in relation to the single crystal molecular packing
and OTFT performance. Here, the molecular packing motifs of several new
benzo[d,d′]thieno[3,2-b;4,5-b′]dithiophene (BTDT) derivatives are studied
both in bulk 3D crystals and as thin films by single crystal diffraction and
grazing incidence wide angle X-ray scattering (GIWAXS), respectively. The
results show that the BTDT derivative thin films can have significantly
different molecular packing from their bulk crystals. For phenylbenzo[d,d′]
thieno[3,2-b;4,5-b′]dithiophene (P-BTDT), 2-biphenylbenzo[d,d′]thieno-[3,2-
b;4,5-b ′]dithiophene (Bp-BTDT), 2-naphthalenylbenzo[d,d′]thieno[3,2-b;4,5-
b′]dithiophene (Np-BTDT), and bisbenzo[d,d′]thieno[3,2-b;4,5-b′]
remarkable progress,[1–3] and a number
of products based on organic electronic
circuitry are already on the verge of com-
mercialization.[4] However, at this exciting
moment, there remain critical challenges
in finding organic semiconductors with
electrical performance, reliability, and
processing windows surpassing amor-
phous silicon. These challenges must
be addressed to propel the widespread
adoption of organic electronics into mass
production. Among several classes of
organic semiconductor materials, acene-
based small molecules,[5] most notably
pentacene,[6,7] have long been of prime
interest due to their excellent charge
transport properties. However, the envi-
ronmental instability of acene-based
organic semiconductors critically limits
their applicability in electronic circuitry.[8]
For this reason, many research groups
have sought to realize new molecular
building blocks that are environmentally
stable, and, at the same time, exhibit effi-
cient charge transport. Fused thiophene
dithiophene (BBTDT), two lattices co-exist, and are significantly strained
versus their single crystal forms. For P-BTDT, the dominance of the more
strained lattice relative to the bulk-like lattice likely explains the high carrier
mobility. In contrast, poor crystallinity and surface coverage at the dielec-
tric/substrate interface explains the marginal OTFT performance of seem-
ingly similar PF-BTDT films.
Dr. J. Youn, Dr. Y. Shi, Dr. S. M. Zhang, Dr. H. Huang, C. Stern,
Prof. A. Facchetti, Prof. T. J. Marks
H.-C. Chang, Y.-J. Liang, C.-M. Yeh, C.-Y. Feng,
Prof. M.-C. Chen
Department of Chemistry and the Materials Research Center
Northwestern University
2145 Sheridan Rd., Evanston, IL 60208-3113, USA
E-mail: a-facchetti@northwestern.edu; t-marks@northwestern.edu
Department of Chemistry
National Central University
Jhong-Li, Taiwan, 32054, R.O.C.
E-mail: mcchen@ncu.edu.tw
Dr. S. Kewalramani, J. D. Emery, Prof. M. J. Bedzyk
Department of Materials Science and Engineering
Department of Physics and Astronomy
Northwestern University
2220 Campus Dr., Evanston, IL 60208-3113, USA
E-mail: bedzyk@northwestern.edu
Dr. L.-H. Chen, Dr. J.-C. Ho
Process Technology Division
Display Technology Center
Industrial Technology Research Institute
Chung Hsing Rd., Chutung, Hsinchu, Taiwan, 31040, R.O.C.
DOI: 10.1002/adfm.201203439
©
3850 wileyonlinelibrary.com
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2013, 23, 3850–3865

www.afm-journal.de
www.MaterialsViews.com
Scheme 1. Examples of fused oligothiophene semiconductors used in TFTs.
derivatives[2,9–14] offer a potential solution to this challenge,
owing to their combination of sufficiently high-lying excited
states and high ionization potentials to suppress photo-oxida-
tion,[11] and bulk single crystal packing motifs which are very
similar to that of pentacene. Among organic semiconductor
materials for p-channel devices, several fused thiophene deriva-
tives with increased numbers of thiophene rings exhibit excel-
lent charge transport performance. For example, DP-DTT,[14]
DP-TTA,[12] and DBTDT[13] exhibit hole mobilities as high as
0.42, 0.14, and 0.51 cm2 V−1 s−1 , respectively (Scheme 1). To
date our laboratory has developed a number of distinctive fused
thiophene molecules for TFT applications. Among them, a
series of fused thiophenes based on benzo[d,d′]thieno[3,2-b;4,5-
b′]dithiophene (BTDT) exhibit good environmental stability
as well as high mobilities. Specifically, the phenyl end-capped
derivative P-BTDT exhibits a carrier mobility as high as
0.70 cm2 V−1 s−1 .[9]
While specific optoelectronic properties of organic semicon-
ductors at the parent molecule level are a necessary factor in
achieving high TFT performance, actual device operation is
affected by many parameters related to film microstructure. For
example, the intrinsic packing of the molecules within the unit
cell,[3,15–17] as well as the size, density, and orientation of crys-
talline domains,[18] the type of domain boundaries,[19] and the
nature of the semiconductor-gate dielectric interface.[20] Specifi-
cally, the local packing of molecules determines intermolecular
orbital overlap and may also influence the crystalline domain
size. Similarly, the connectivity between adjacent domains
influences the paths taken by charge carriers through films.[21]
In particular, the microstructure of the first monolayers above
the dielectric plays a critical role in determining OTFT perfor-
mance because the charge transport is highly localized in the
channel within 3–5 nm of this interface.[21]
have been scrutinized, such assumptions have been proven to
sometimes be inaccurate, and it is now known that molecular
packing in thin films may differ significantly from that in the
bulk crystal structures, especially near the dielectric inter-
face. For example, in thin film form, both tetraceno[2,3-b]thio-
phene[17] and pentacene[16] exhibit a local molecular packing dis-
tinctly different from that in the bulk crystal. These observations
highlight the necessity of obtaining microstructural details for
thin organic films.
In this contribution, the molecular packing motifs of five
newly synthesized BTDT[9] derivatives (Scheme 2), are studied
both in bulk single crystals and in thin films by single crystal
diffraction and grazing incidence wide angle X-ray scattering
(GIWAXS). We first introduce the new BTDT molecules and
describe their thermal, optical, and electrochemical properties.
We then compare the bulk and thin film structures by analyzing
single-crystal X-ray diffraction and GIWAXS data. Lastly, the
effect of thin film microstructure on TFT performance is dis-
cussed. The results indicate that these BTDT derivatives have
different molecular packing in thin films versus the bulk crys-
tals. In the case of P-BTDT, Bp-BTDT, Np-BTDT, and BBTDT, it
will be seen that two types of lattices coexist, and that these are
slightly strained compared to their bulk crystal forms. In con-
trast, for PF-BTDT fi lms, a single lattice is observed, however,
this lattice has no apparent correspondence to the bulk crystal
form. For P-BTDT, which yields the best performing TFTs of
the series, the dominance of the more strained lattice relative to
the bulk-like lattice may explain the excellent charge transport
properties. On the other hand, poor crystallinity and poor sur-
face coverage at the substrate interface explains the poor device
performance of PF-BTDT fi lms.
Unfortunately, with few exceptions,[3,15–17] detailed infor-
mation on the aforementioned issues is unavailable for most
small molecule organic semiconductor films. As a result, dis-
cussions of TFT charge transport properties are often based
on information extrapolated from the bulk single crystal struc-
tures. However, in the few cases where these structural factors
2. Results
2.1. Synthesis
Details of the various BTDT syntheses can be found in the
Supporting Information or previous literature.[9] Briefly, the
Scheme 2. Chemical structures of the benzo[d,d′]thieno[3,2-b;4,5-b′] dithiophene (BTDT) semiconductors examined in this study.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2013, 23, 3850–3865
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3851

www.afm-journal.de
www.MaterialsViews.com
T a b le 1 . Thermal, optical, and electrochemical properties of BTDT derivatives. All potentials reported are referenced to an Fc+/Fc internal standard
(at +0.6 V).
ΔEgap [eV]
a)
DSC Tm
TGA °C,
UV-Vis λmax
Reduction Potentialb) Oxidation Potentialb)
(UV)a)
3.21
3.22
3.07
3.08
2.74
(DPV)b)
3.37
Compound
P-BTDT
[°C]
[5%]
[nm]
[V]
[V]
269
316
392
299
461
287
247
402
366
471
358
351
369
369
411
–1.98
–1.75
–1.98
–1.96
–1.75
1.39
1.61
1.30
1.31
1.14
PF-BTDT
Bp-BTDT
Np-BTDT
BBTDT
3.36
3.28
3.27
2.89
a) In o-C6H4Cl2; b) In o-C6H4Cl2 at 60 °C.
functionalizations of BTDT molecule were carried out according
to the synthetic protocols shown in Scheme S1 (Supporting
Information). BTDT is first deprotonated with n-BuLi, and then
alkylstannylated to generate BTDT-SnR3. Next, this compound
was straightforwardly coupled with the corresponding aryl bro-
mide to produce P-BTDT, Bp-BTDT, Np-BTDT, PF-BTDT, and
BBTDT in 46%, 85%, 82.5%, 35%, and 45% yield, respectively,
via a Stille coupling protocol. PF-BTDT can be also obtained by
deprotonating BTDT in the presence of C6F6 in 28% yield. All
compounds were characterized by conventional chemical and
physical methodologies.
peaks close to +1.30 V and reduction peaks at –1.98 V. The oxi-
dation potential is shifted to smaller values (easier oxidizing)
than that of P-BTDT (Eox = +1.39 V), which is attributable to
aryl substituent conjugative effects. Similarly, the oxidation
potential of more conjugated BBTDT (Eox = +1.14 V) is shifted
to less positive values, while the reduction potential is displaced
to a less negative value (Ered = –1.75 V). In contrast, both Eox
=
+1.61 V; and Ered = –1.75 V of PF-BTDT are shifted to more
positive values (more difficult oxidizing) than those of P-BTDT
(Eox = +1.39 V; Ered = –1.98V) without a significant difference in
oxidation and reduction potential, consistent with well-known
electron-withdrawing group (EWG) effects. The difference in
oxidation and reduction potential obtained from the DPV data
(Figure 1) rise in the order: BBTDT (2.89 eV) < Bp-BTDT ≈ Np-
BTDT (3.27 eV) < P-BTDT (3.37 eV) ≈ PF- BTDT (3.36 eV), in
good agreement with the values obtained from optical spectros-
copy (Table 1). HOMO energies (EHOMO) were estimated using
equation, EHOMO = –(4.20 + Eox); assuming ferrocene/ferroce-
nium oxidation at 4.8 eV.[23]
2.2. Thermal, Optical, and Electrochemical Properties
of BTDT Derivatives
For the five BTDT derivatives, differential scanning calorim-
etry (DSC) scans reveal sharp endotherms above 269 °C, and
thermogravimetric analysis (TGA) plots demonstrate weight
loss (≈5%) only on heating above 287 °C, except for fluorinated
PF-BTDT, as summarized in Table 1. Lower melting points
and lower weight loss temperatures are observed for the BTDT
derivatives having lower molecular weights. However, PF-BTDT
exhibits a 5% weight loss temperature ≈247 °C, lower than that
of P-BTDT, indicating higher volatility due to the fluorocarbon
substitution.[22] The optical absorption spectra of Bp-BTDT, Np-
BTDT, and BBTDT in o-C6H4Cl2 solution (Supporting Informa-
tion Figure S1) are slightly red-shifted (λmax > 369 nm), relative
to that of the P-BTDT (λmax ≈ 358 nm). In contrast, the absorp-
tion spectrum of PF-BTDT is slightly blue-shifted relative to
P-BTDT. This result is consistent with the observation that
biphenyl, naphthalenyl, and benzo[d,d]thieno[3,2-b;4,5-b]dithio-
phenyl substituted BTDTs have greater electronic delocalization
than does phenyl substituted BTDT. The stability of these com-
pounds against photooxidation was also investigated by moni-
toring the decay of the optical absorption maxima in aerated
o-C6H4Cl2 solutions exposed to a white fluorescent lamp light at
room temperature. No decomposition is observed after 4-5 days
of continuous illumination.
The low HOMO energies for the BTDT derivatives correlate
their UV photooxidative stability. In addition, the fluoroaryl
substitution in PF-BTDT substantially lowers both the HOMO
and LUMO energies versus those of P-BTDT. Note that the
HOMO energies of BTDT derivatives are close to the work
function of a clean fresh gold surface (–5.1 eV), although the
gold work function can be as high as –4.3 eV due to contami-
nants adsorbed under ambient conditions. Nevertheless, it can
be concluded that the present BTDT compounds should have
relatively small hole injection barriers from gold electrodes for
p-type TFT operation.
2.3. Single Crystal Structural Analysis
Before examining the thin film structure of BTDTs, the bulk
single crystal structures are analyzed in order to establish a
referential basis for further discussion. The crystal structures
and lattice parameters obtained from single crystal diffrac-
tion measurements followed by crystallographic refinement
calculations for each of the BTDTs are summarized in Table 2
below. Details of the single crystal diffraction measurements
and crystallographic refinement are summarized in Tables
S1-S5. The local packing parameters known to affect the
Differential pulse voltammograms (DPVs; see Supporting
Information for details) of the BTDT derivatives were recorded
in dichlorobenzene at 60 °C and the results are summarized in
Table 1. The DPVs of Bp-BTDT and Np-BTDT exhibit oxidation
©
3852 wileyonlinelibrary.com
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2013, 23, 3850–3865

www.afm-journal.de
www.MaterialsViews.com
2.3.1. Crystal Structures of P-BTDT, Bp-BTDT, and Np-BTDT
First we note that although Np-BTDT crystallizes in a mono-
clinic lattice, the packing arrangement of the Np-BTDT mole-
cules bears striking resemblance to the packing of P-BTDT and
Bp-BTDT, which crystallize in triclinic lattices. For example, the
lattice parameter a and the lattice parameter b for these three
BTDTs are nearly identical, and the c lattice parameters are
nearly twice the respective molecular lengths (Figure 2). Fur-
thermore, close examination (Supporting Information Table S1–
S5) shows that the unit cells of these compounds contain four
molecules with the nearest neighbors exhibiting a herring-
bone packing motif (Figure 2). The local packing parameters,
such as interplanar distance, pitch angle, roll angle (defined
in Supporting Information Figure S2), and herringbone angle
also reflect the similarity between these three BTDT crystals
and are summarized in Table 3. This packing configuration is
similar to that of pentacene[6] and other high mobility acene
semiconductors.[5]
Figure 1. DPV-derived HOMO and LUMO energy levels of BTDT semi-
conductors compared to the work functions of clean and air-exposed
metallic gold (assuming ferrocene/ferrocenium oxidation at 4.8 eV).
2.3.2. Crystal Structures of BBTDT and PF-BTDT
charge transport are summarized in Table 3. A close examina-
tion of Table 2 shows that despite seeming disparities in crystal
system, P-BTDT, Bp-BTDT, and Np-BTDT are structurally
quite similar. The crystal lattices and molecular arrangements
within the unit cells of BBTDT and PF-BTDT are discussed
separately.
BBTDT crystallizes in a monoclinic lattice (space group P21/a)
with two molecules per unit cell. This structure also exhibits
a herringbone packing motif, similar to those observed in
P-BTDT, Bp-BTDT, and Np-BTDT above, but has unit cell
parameters that are significantly different from those BTDTs
T a b le 2 . Crystal system, space group, and lattice parameters for bulk BTDT crystals at T = 298 K.a) Note that in order to facilitate comparison between
the molecules, the derived lattice vectors a and b for P-BTDT, and Bp-BTDT are exchanged in comparison to the.cif files. Similarly, for BBTDT, the lat-
tice vectors a and c are exchanged. This does not affect the space group designation for the triclinic lattices. For BBTDT, the exchange of the a and c
axes changes the space group from P 21/c to P 21/a, but the unique axis b remains unchanged.
Compound
Crystal system
Space group
a
b
c
α
β
γ
[Å]
[Å]
[Å]
[°]
[°]
[°]
P-BTDT
Triclinic
Triclinic
P1
P1
7.73
7.72
7.64
11.35
8.18
5.96
5.93
5.90
3.94
6.21
31.16
39.35
35.80
21.89
29.04
94.40
89.40
90.00
90.00
90.00
95.20
86.18
95.23
104.00
93.31
90.16
89.87
90.00
90.00
90.00
Bp-BTDT
a)
Np-BTDT
Monoclinic
Monoclinic
Monoclinic
Pc
BBTDT
P21/a
P21/c
PF-BTDT
a)Single crystal diffraction measurement of Np-BTDT was performed only at T = 100 K. At room temperature, the Np-BTDT crystals did not diffract to sufficient resolution
required for successful determination of the single crystal structure. The listed values are obtained by using the lattice parameters obtained at T = 100 K, and by assuming
that the thermal expansion coefficients for Np-BTDT single crystals are equal to the average of the thermal expansion coefficients of P-BTDT, Bp-BTDT, PF-BTDT, and
BBTDT, for which the measurements were performed at both T = 100 K and T = 298 K.
T a b le 3 . Local packing parameters based on single crystal diffraction. Slight deviations from the expected symmetries (see Supporting Information)
are due to approximations in the determination of herringbone, pitch, and roll angles of the BTDT molecules.
Compound
Packing
Herringbone Interplanar Angleb)
Roll Angle
Pitch Angle
Interplanar
Distancec) [Å]
S-S Distance
[Å]
Density
[g/cm3]
Anglea) [°]
[°]
[°]
[°]
P-BTDT
Bp-BTDT
Np-BTDT
BBTDT
Herringbone
Herringbone
Herringbone
Herringbone
Herringbone
49.5
8
65.0
64.5
63.5
27.0
21.5
0
0
2.48
2.48
2.44
3.55
3.49
3.71
3.69
3.62
3.60
3.64
1.502
1.472
1.540
1.714
1.859
49.8
9
49.1
3
0
129.0
135.6
0
0
PF-BTDT
3
8.5
a)The herringbone angle is defined as the interplanar angle for the diagonally aligned molecules. The molecular planes for these calculations are defined as the planes
formed by the sulfur atoms of the BTDT cores; b)The interplanar angle is defined as the angle between the planes of the BTDT cores and the aryl functional groups within
the molecules; c)The interplanar distance is the normal separation of the parallel molecules that are shifted as compared to an idealized face-to-face π-stack (see Sup-
porting Information Figure S2).
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2013, 23, 3850–3865
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3853

www.afm-journal.de
www.MaterialsViews.com
Figure 2. Molecular and crystal structures of P-BTDT (a–c), Bp-BTDT (d–f), and Np-BTDT (g–i) as derived from single-crystal diffraction analysis. All
local packing parameters are reported in Table 3. Red, black, and white colors indicate sulfur, carbon, and hydrogen atoms, respectively. Views from
¯
¯
near the [111] or [111] directions are shown in (b,e,h), and a view nearly parallel to the c-axis highlights the herringbone packing of the molecules (c,f,i).
Red arrows indicate the closest S-S contact distances, while the herringbone angles and angles between the molecules and the [010] lattice direction
(the complement of the roll angle) are shown in (c,f,i).
(Table 2). Furthermore, the interplanar distance for nearest
neighbors of 3.55 Å, the roll angle of 27°, and the herringbone
angle of ≈129° (Figure 3), differ from the corresponding values
of 2.44–2.48 Å, ≈64° and ≈49°, respectively, in the P-BTDT,
Bp-BTDT, and Np-BTDT crystal structures. These distinctions
are relevant because they directly correlate with the degree
of π orbital overlap, which in turn governs the efficiency of
charge transport between neighboring molecules. Specifically,
the interplanar distance and roll angle determine the π orbital
interaction between cofacial parallel neighboring molecules,
and the herringbone angle determines the degree of interaction
between diagonally aligned molecules.
PF-BTDT crystals (Figure 3, Supporting Information
Table S5) belong to the same crystal system (monoclinic
space group P21/c) as BBTDT (Figure 3, Supporting Infor-
mation Table S2). However, the separation between the cen-
troids of the molecules that are related by herringbone sym-
metry is ≈15 Å, which is ≈2.5× that for BBTDT (5.9 Å). This
©
3854 wileyonlinelibrary.com
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2013, 23, 3850–3865

www.afm-journal.de
www.MaterialsViews.com
Figure 3. Molecular and crystal structure of BBTDT (a–c) and PF-BTDT (d–f) as derived from single-crystal diffraction analysis. All local packing
parameters are reported in Table 3. Red, black, white, blue colors indicate sulfur, carbon, hydrogen, and fluorine atoms, respectively. For BBTDT, a
¯
view from near the [111] direction is shown in (b) and a view nearly parallel to the c-axis highlights the herringbone packing of the molecules (c). Red
arrows indicate the closest S-S contact distances, while the derived herringbone angles and angles between the molecules and the [010] lattice direc-
¯
tion (the complement of the roll angle) is shown in (c). For PF-BTDT, a view from near the [011] direction is shown in (e) and a view nearly parallel to
the c-axis is shown in (f). Nearest-neighbor molecules are co-facially packed at a distance of 3.64 Å. Herringbone symmetry exists between the pair of
molecules labeled “a” and the pair labeled “b”, but the large separation between these molecules (≈15 Å) effectively eliminates direct charge transport
between them.
effectively eliminates any direct charge transport between these
molecules. In addition, the nearest-neighbor molecules are
co-facially aligned but in an anti-parallel fashion (Figure 3e).
This also significantly reduces the π orbital overlap between
the co-facially aligned molecules. We will return to these struc-
tural distinctions between BTDTs when we discuss the thin
film structures and TFT electronic performance in the sections
which follow.
(Supporting Information Figure S3–S10). Finally, OTS-
functionalized Si/SiO2 substrates were chosen for GIWAXS
measurements because the best OTFT performance is observed
for this surface treatment of the gate dielectric (Table 4). BTDT
films of nominal 3 nm thickness were chosen because the
charge transport in TFTs is generally limited to the first 3–5 nm
at the dielectric-semiconductor interface.[21] In addition,
GIWAXS measurements were performed only on films depos-
ited with the substrate temperature (TD) for which maximum
TFT performance was observed.
The GIWAXS images (Supporting Information Figure S11)
reveal that the scattered intensity is distributed along a series of
qz-extended rods at discrete points on the qxy axis. This implies
that the BTDT fi lms do not consist of crystallites randomly ori-
ented in 3D, but instead have one of the crystallographic axes at
a fixed angle with respect to the surface normal. Furthermore,
because all the diffraction peaks are observed, regardless of the
azimuthal orientation of the sample with respect to the inci-
dent beam, the BTDT fi lms likely consist of crystallites that are
randomly oriented in-plane (2D powders). The qxy positions of
2.4. GIWAXS Analysis of Semiconductor Thin Films
GIWAXS measurements (Supporting Information Figure S11)
were carried out to investigate the effects on the BTDT crystal
structures induced by confinement in thin films. In an effort
to optimize film deposition conditions, three types of sub-
strates, bare Si/SiO2, HMDS-functionalized Si/SiO2, and
OTS-functionalized Si/SiO2, were utilized and the effect of
the growth temperature was assessed by TFT electrical evalu-
ation (Table 4) and out-of-plane X-ray scattering measurements
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2013, 23, 3850–3865
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3855

www.afm-journal.de
www.MaterialsViews.com
T a b le 4 . Derived lattice parameters and molecular orientations within the unit cell for BTDTs in bulk and thin film form. The tilt angle of the mol-
ecules in single crystals is defined as the angle between the direction of the long molecular axis and the normal to the (001) plane.
Molecule
P-BTDT
Medium
a
[Å]
b
[Å]
c
[Å]
Tilt Angle
Herringbone
Angle [°]
α
[°]
β
[°]
γ
[°]
[°]
Bulk
7.73
7.88
7.88
7.72
7.85
7.85
7.75
7.94
7.94
11.35
11.51
11.51
5.96
6.05
6.05
5.93
5.99
5.99
5.92
6.07
6.07
3.94
4.00
4.00
31.06
30.88
31.02
39.35
38.00
38.00
35.91
38.00
34.50
21.89
21.81
21.86
94.4
94.3
92.5
89.4
87.4
86.9
90.0
90.0
90.0
90.0
90.0
90.0
95.2
96.5
99.2
86.2
84.9
82.5
95.2
94.4
96.8
104.0
104.3
104.0
90.2
≈90.0
≈90.0
89.9
14.1
10.9
12.9
3.8
46.9
58.1
18.9
49.8
44.2
19.7
49.1
44.6
84.6
129.0
141.9
161.8
Thin filma)
Thin filmb)
Bulk
Bp-BTDT
Thin filma)
Thin filmb)
Bulk
12.7
9.6
≈90.0
≈90.0
90.0
Np-BTDT
8.8
Thin filma)
Thin filmb)
Bulk
90.0
9.7
90.0
12.7
4.3
BBTDT
90.0
Thin filma)
Thin filmb)
90.0
18.3
13.6
90.0
a)Bulk-like; b)Best-fit.
the diffraction peaks (Figure 4–e) reveal the 2D crystallographic
unit cell in the plane of the substrate. The intensity profiles
of the (h k qz) rods are analyzed separately to determine the
molecular arrangement within the unit cell in the out-of-plane
analysis section.
(Figure 4f–i). Additionally, the absence of (10l) and (01l) reflec-
tions and the simultaneous observation of the (21l) peak in the
GIWAXS data (Figure 4, Supporting Information Figure S11)
are indicative of herringbone molecular packing with p2gg
plane group symmetry.[16,24] This is similar to the case for
the 3D crystals of P-BTDT, Bp-BTDT, Np-BTDT, and BBTDT
where the molecular projections onto the (001) plane reveal a
herringbone motif.[9] The aforementioned close correspond-
ence between the bulk and the thin film structures suggests
that in thin films the BTDT molecules form ordered structures
with their long molecular axes aligned close to the surface
normal, in direct analogy to the molecular arrangement in the
single crystals (see the following section on out-of-plane micro-
structure analysis).
Despite qualitative similarities to the single crystal (001)
plane, note that all in-plane thin film lattices are strained. The
A lattices for Np-BTDT, Bp-BTDT, and BBTDT exhibit a slight
decrease in unit cell area, ΔA ≈ –1% to –2%, when compared to
the (001) plane of their single crystal forms, and the B lattices
for the four BTDTs are expanded by ΔA ≈ 2.7%–4% with respect
to the single crystal (0 0 1) plane. Intuitively, the coexistence of
two lattices suggests a non-uniform film, such that the BTDTs
close to the dielectric surface crystallize in the relatively highly
strained B lattice and the BTDTs in thicker film regions crys-
tallize in the A lattice. However, this explanation is inadequate
to explain the observed compressive strain in the A lattices of
Bp-BTDT, Np-BTDT, and BBTDT. Regardless, it is important
to note that lattice strain can play a critical role in the OTFT
performance. For example, a recent GIWAXS and electrical
characterization study on TIPS-pentacene TFTs showed that a
reduction of ≈7% in the π–π stacking distance correlated with a
≈6-fold increase in carrier mobility.[25]
2.4.1. In-Plane Microstructure Analysis
For P-BTDT, Bp-BTDT, Np-BTDT, and BBTDT fi lms, the qz-pro-
jected 1D intensity profiles along the qxy direction (Figure 4a–d)
indicate that the positions of the diffraction peaks are close to
those expected for the case where the (001) plane corresponding
to their 3D lattices is parallel to the substrate surface. In con-
trast, the PF-BTDT thin films display a crystallographic form
substantially different from the bulk structure. For this reason,
the in-plane lattices for PF-BTDT in thin films will be discussed
separately. First, we focus on the case of P-BTDT, Bp-BTDT,
Np-BTDT, and BBTDT fi lms.
While the P-BTDT, Bp-BTDT, Np-BTDT, and BBTDT fi lms
appear to consist of crystals with the (001) planes parallel to
the substrate surface, for each diffraction peak expected for
the (001) orientation of the crystals, a doublet is observed
(Figure 4a–d and Figure 5). Although a slight splitting may
be expected for (hkl) diffraction peaks that have non-zero h
and k because of the non-orthogonality of the ab and bb axes
(Table 2), this non-orthogonality of the lattice vectors cannot
explain the doublets that are observed close to the expected
positions for (h0l) and (0kl) reflections (Figure 5). This obser-
vation indicates that P-BTDT, Bp-BTDT, Np-BTDT, and
BBTDT crystallize with two distinct structures when confined
to thin films (Figure 4f–i).
It can be shown that for these four BTDTs, the positions of
all the diffraction peaks can be indexed by assuming a coex-
istence of two in-plane rectangular lattices which have lat-
tice parameters close to that for the bulk (001) crystal plane
Unlike the other BTDTs, PF-BTDT exhibits only a single in-
plane lattice (Figure 4j) with observed in-plane lattice param-
eters of a = 7.05 Å, b = 6.78 Å, and γ = 98.7°. Because the
©
3856 wileyonlinelibrary.com
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2013, 23, 3850–3865

www.afm-journal.de
www.MaterialsViews.com
Figure 4. qz-projected 1D intensity profile as a function of qxy for 3 nm P-BTDT (a), Bp-BTDT (b), Np-BTDT (c), BBTDT (d), and PF-BTDT(e) films on
Si/SiO2/OTS. For each, the diffraction peak at qxy ≈ 1.5 Å−1 is due to the packing of the OTS alkyl tails in a 2D hexagonal lattice. All other diffraction
features originate from the crystalline packing of semiconductor molecules. For BBTDT (d), the relatively weak diffraction peaks indicated with the
arrows at qxy ≈ 0.295 and ≈ 0.595 Å−1 correspond to diffraction from the (001) and (002) bulk crystal planes, indicating a small fraction of molecules
that are oriented with their long molecular axis parallel to the substrate surface. The derived 2D unit cell parameters for P-BTDT (f), Bp-BTDT (g),
Np-BTDT (h), BBTDT (i), and PF-BTDT(j) are shown on the right for the bulk single crystal (red dashed polygon) and thin film (solid black polygon)
lattices that comprise the thin film structure. In (f–i) the differences between the observed lattices and the a–b plane corresponding to the bulk single
crystal structures have been exaggerated by 2.5×.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2013, 23, 3850–3865
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3857

www.afm-journal.de
www.MaterialsViews.com
vertically standing-up molecules (Supporting
Information Figure S12). Because the 2D
¯
basis for the (
) plane (a = 6.82 Å, b =
3120
6.38 Å, and γ = 28.4°) is substantially dif-
ferent from the observed in-plane lattice for
PF-BTDT thin films, the thin film structure
has no apparent correspondence to any of
the crystallographic planes of the bulk lattice.
The lack of similarity between the thin film
and the single crystal lattices likely arises
because, for the dense molecular packing in
single crystals, the neighboring PF-BTDTs
are anti-parallel with their fluoro-aryl groups
aligned in opposite directions. It is possible
that in thin films the nearest-neighbor PF-
BTDT molecules are parallel due to the pref-
erential affinity of the fluoro-aryl groups to
either the substrate/film or, more likely, the
film/air interface. This observation may also
explain the apparent expansion of the thin
film in-plane unit cell as compared to 2D
¯
unit cells for single crystal (001) and (
lattice planes.
)
3120
2.4.2. Out-of-plane Microstructure Analysis
In order to gain further insights into the
packing of BTDTs in thin films, we extract
the lattice parameter c, the crystallographic
angles α and β, and the orientation of mol-
ecules within the unit cell, by analyzing the
intensity profile along (h k qz) rods under a
simplified model for the molecular shapes
(Supporting Information Figure S13). Briefly,
Figure 5. Splitting of the (20l) diffraction peak for P-BTDT, Bp-BTDT, Np-BTDT, and BBTDT.
The peaks indexed as (20l)A and (20l)B are from the slightly strained (Lattice A) and the highly
strained (Lattice B) structures. The constituent components of the doublets are shown in red
and the sum of the two components is in black.
BTDT molecules are treated as paralellopipeds of uniform
electron density and dimensions 5.5 × 2.1 × L Å3, where L is the
length of the molecule along the long molecular axis (Figure 2,3),
5.5 Å represents the approximate length along the short molec-
ular axis, and 2.1 Å is roughly the diameter of the sulfur atom.
The centroids of the molecules were fixed at the same frac-
tional coordinates as those in the single crystals. The molecule
is allowed to rotate about the three orthogonal axes passing
through the center of mass. In this model, the symmetry equiv-
alent molecules in the single crystal forms are constrained to
rotate by the same amounts. The films are modeled to consist of
regions that are 1, 2 or 3 unit cells thick because the BTDT fi lms
are expected to have regions of varying thicknesses (Figure 7).
The mathematical formalism is described in the Experimental
Section. Fits are performed only for B lattices for P-BTDT, Bp-
BTDT, Np-BTDT, and BBTDT because the intensity profile
along the rods for A lattices is qualitatively similar to those
for B lattices. Further analysis of PF-BTDT was not performed
because the simplfied model gives equally satisfactory mulitple
solutions that correspond to disparate molecular arrangements
(for example, see Figure 6), making it impossible to identify
the true molecular packing in the absence of a single crystal
reference (Figure 6). For the case of PF-BTDT, we do not have
such a basis because in thin film form the molecular packing
is distinctively different from that in the bulk (Figure 4j). The
in-plane lattice was derived from indexing only the four clearly
visible diffraction peaks, the validity of the assignment was also
examined by measuring 50 nm films deposited at the same
temperature and flux (data not shown). The 50 nm films show
10 diffraction peaks corresponding to an in-plane lattice with
parameters; a = 7.24Å, b = 6.93 Å, and γ = 99.6°. Although the
lattice for the 50 nm films is expanded (ΔA ≈ 4.7%) versus the
2D lattice of the 3nm films, it is also an oblique lattice with
nearly the same angle γ. This observation clearly demonstrates
that unlike the other BTDTs, the in-plane lattice for PF-BTDT
does not correspond to the (001) plane of the single crystal,
which for PF-BTDT is defined by ab = 8.12 Å, bb = 6.21 Å, and
γ = 90°.
In order to assess the molecular orientation of the PF-BTDT
molecules within the unit cells, out-of-plane XRD measure-
ments were performed on 50 nm thick films (Supporting
Information Figure S6). The XRD plots reveal an interlayer
spacing along the interface normal of 16.1 Å, which is com-
parable to the expected length L = 15.2 Å for PF-BTDT along
its long molecular axis. This observation implies that the
PF-BTDT molecules, like other BTDTs (see out-of-plane anal-
ysis), are oriented with their long axes along the interface normal.
If the 3D lattice for thin film structure were identical to that of
¯
the PF-BTDT single crystals, the crystallographic plane (
)
3120
would be parallel to the substrate surface for the case of
©
3858 wileyonlinelibrary.com
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2013, 23, 3850–3865

www.afm-journal.de
www.MaterialsViews.com
simultaneous fits for the (11qz), (20qz) and
(21qz) rods correponding to the B lattices of
P-BTDT, Bp-BTDT, Np-BTDT, and BBTDT
are shown in Figure 6, and simulation results
are summarized in Table 4.
For each BTDT derivative, the three data
sets were fit simultaneously with the same
set of parameters (except for the scaling
parameters, see Supporting Information).
For P-BTDT, Np-BTDT, Bp-BTDT as well as
BBTDT, satisfactory fits were achieved under
two conditions: when the herringbone angle
was constrained to be near a bulk-like value,
and when it was far from it (Table 4). For
P-BTDT, Bp-BTDT, and Np-BTDT, the three
scaling factors for the bulk-like fits were
nearly identical, however, for the best fits
they differed by more than an order of mag-
nitude. In contrast, for BBTDT, the scaling
factor for the (20qz) rod was an order of
magnitude greater than those for the (11qz)
and (21qz) rods for both the bulk-like and
the best fits. Physically, the scaling param-
eters are related to the number of unit cells
or the crystal dimensions in the (110), (200),
and (210) directions. For the case of BBTDT,
AFM images (Figure 7d) show that the films
consist of elongated crystals/polycrystalline
aggregates. Based on the above analysis of
the rod profiles, we suggest that the BBTDT
crystals grow and/or aggregate preferentially
along (200) direction in the substrate plane.
For the case of P-BTDT, Bp-BTDT, and
Figure 6. Intensity distribution along the three (h k qz) rods along with model fits for P-BTDT
(a), Bp-BTDT (b), Np-BTDT (c), and BBTDT (d). The (20qz) and (21qz) rods are offset for clarity.
Note that for BBTDT the sharp intensity modulations at qz ≈ 0.85 Å−1 and 1.7 Å−1 observed
along the (20qz) rod arise from reflections from a minority of large crystals oriented along
an axis other than (001). These reflections can be seen to be extraneous to the (20qz) rod in
Figure 4d.
Figure 7. Sub-monolayer (3 nm) film morphologies of BTDT films on the Si/SiO2/OTS substrates. a) Bp-BTDT grown at 80 °C, b) Np-BTDT grown at
80 °C, c) PF-BTDT grown at 40 °C, and d) BBTDT grown at 110 °C (tapping mode, topography, 10 μm × 10 μm).
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2013, 23, 3850–3865
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3859

www.afm-journal.de
www.MaterialsViews.com
Np-BTDT, no such preferential growth direction is observable
in the AFM images (Figure 7a,b), which show nearly uniform
film morphologies, and also in the GIWAXS measurements
(Figure 4, Supporting Information Figure S11), which show
that the coherence lengths along the (110), (200) and (210) are
very similar. These observations suggest that for P-BTDT, Bp-
BTDT, and Np-BTDT, the bulk-like fits obtained with similar
scaling factors for the three rod profiles likely represent the
correct molecular arrangements within the unit cell. For these
BTDTs, the best fit and the bulk-like fit are qualitatively equally
good representations of the rod profiles. Therefore, it seems
reasonable to suggest that the molecular arrangements in all
the four BTDTs are very similar to those in their single crystal
forms.
Visual inspection of the intensity profiles along the (h k qz)
rods suggests that all BTDT fi lms are more than one unit cell
thick, especially BBTDT fi lms, where the (20qz) rod exhibits
strong intensity modulations which extend up to at least
qz ≈ 2.3 Å−1 . Indeed, analysis of the rod profiles reveals that for
BBTDT, ≈50% of the film is 3 unit cells (u.c.) thick, and 30%
is 2 unit cells thick. In contrast, only 20% of the P-BTDT fi lm
is 2 unit cells thick. For Bp-BTDT and Np-BTDT (Figure 6),
the (h k qz) rods are relatively featureless. For these cases, fits
to the data reveal that more than ≈90% of the film is 1 unit
cell thick. These observations are qualitatively consistent with
the AFM measurements on 3 nm BTDT fi lms (Figure 7),
where for BBTDT, a non-uniform film composed of vertically
extended aggregates is observed while highly uniform coverage
is observed for the Bp-BTDT and Np-BTDT fi lms. Note that
the fit values for the fractional coverage by films of different
thicknesses are robust, and do not vary by more than 5% upon
changes in simulation starting conditions. For P-BTDT, Np-
BTDT, Bp-BTDT, and BBTDT, the fit values of lattice constant
c, the crystallographic angles α and β, and the extracted tilts
of the molecules with respect to the surface normal (Table 4)
are also very robust and reveal that the lattice parameter c and
the angles α and β are very close to their single crystal values.
Finally, for all four BTDTs, the molecules were found to align
with their long axes within a tilt of ≈10–20° from the surface
normal.
on-to-off current ratio (Ion/Ioff), were extracted using standard
procedures.[26] The results are summarized in Table 5, and rep-
resentative I–V plots are presented in Figure 8 and Supporting
Information Figure S14.
In general, the P-BTDT, Bp-BTDT, Np-BTDT, and BBTDT
films exhibit good charge transport characteristics: Bp-BTDT
(μ = 0.11 cm2 V−1 s−1 , max), BBTDT (μ = 0.15 cm2 V−1 s−1 ,
max), Np-BTDT (μ = 0.079 cm2 V−1 s−1 , max), and previously
reported P-BTDT devices also exhibit excellent performance,
μ = 0.70 cm2 V−1 s−1 , on OTS substrates at an elevated growth
temperature, TD = 40 °C. In marked contrast to these results,
PF-BTDT exhibits considerably lower device performance (μ =
6.2 × 10−6 cm2 V−1 s−1 , max) relative to other BTDT materials.
As will be discussed below, these differences in device per-
formance are convincingly related to the thin film molecular
packing structures based on the GIWAXS analysis discussed in
the previous section.
3. Discussion
The GIWAXS analyses of the P-BTDT, Bp-BTDT, Np-BTDT,
and BBTDT thin films reveal that the molecules are arranged
in crystalline structures that are similar to, but not identical to,
their bulk crystal forms. Note that in the thin films the crystal-
lites do not form 3D powders, but are oriented with the (001)
planes parallel to the substrate surface. The (001) plane is a
high packing density plane in the single crystal forms of these
BTDTs. This observation may partially explain the high carrier
mobilities observed in BTDT TFTs because, all other factors
being equal, charge transport efficiency is expected to increase
with compressed inter-molecular separation,[27] reflecting
enhanced π orbital overlap between neighboring molecules.
Furthermore, for the present crystallographic orientations, the
molecules are aligned essentially along the substrate normal
(Table 5). This edge-on BTDT orientation ensures that the direc-
tion of π orbital overlap between neighboring molecules is suit-
ably aligned to the source–drain direction of TFT charge flow.
The aforementioned four BTDTs have similar bulk and thin
film crystal lattices, but there are also subtle differences. Two
lattices with slightly different strains relative to the single crystal
form are observed in each case. Note that the lattice parameters
for the B crystal forms of P-BTDT, Np-BTDT, and Bp-BTDT are
very similar, and differ by less than 1%. Therefore, the differ-
ences in mobilities observed in their TFTs cannot be explained
on the basis of intermolecular separation alone. However, the
fractional coverage for the two lattices differs from one BTDT
to another as shown in Figure 5. For example, Gaussian fits to
the (20l) diffraction peaks for P-BTDT show that the integrated
intensity ratio for that reflection from the B crystal to that from
the A form, IB/IA is ≈2.33. In contrast, for Bp-BTDT, Np-BTDT,
and BBTDT, IB/IA is 0.64, 0.85, and 0.97, respectively. Recall
that the B lattices are more strained than the A lattices. Because
the strains are expected to arise from the substrate-molecule
interactions, it may be expected that the B crystals comprise
the layer closest to the substrate. Therefore, for crystalline films
having the same thickness and similar lattice parameters, such
as for P-BTDT, Np-BTDT and Bp-BTDT, a higher IB/IA likely
implies a more uniform first layer, which is possibly related to
2.5. Organic Thin-Film Transistor (OTFT) Characterization
Thin-film BTDT transistors were fabricated in bottom gate-
top contact configurations. Highly doped p-type (100) silicon
wafers were used as gate electrodes as well as substrates, and
300 nm thermally grown SiO2 on the Si was used as the gate
insulator. The organic semiconductor thin films (50 nm) were
vapor-deposited onto the Si/SiO2, HMDS-treated Si/SiO2, OTS-
treated Si/SiO2 substrates. Next, 50 nm gold source and drain
electrodes were vapor-deposited at 2 × 10−6 Torr through a
shadow mask in a high vacuum growth chamber. Devices were
fabricated with two sets of device configurations, either with a
channel length of 50 μm and a channel width of 2000 μm, or
a channel length of 100 μm and channel width of 5000 μm.
Current–voltage (I–V) transfer and output plots were meas-
ured for each device in air. Key device performance parameters,
such as field-effect mobility (μ), threshold voltage (VT), and
©
3860 wileyonlinelibrary.com
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2013, 23, 3850–3865

www.afm-journal.de
www.MaterialsViews.com
T a b le 5 . TFT device performance data for BTDT-based semiconductor films.
a)
a)
Compound
P-BTDT[b]
TD
[°C]
Substrate
VTa)
[V]
Ion/Ioff
μave
μmax
[cm2 V−1 s−1
]
[cm2 V−1 s−1
]
(2.8 1.0) × 107
(1.4 0.5) × 107
(3.3 0.6) × 107
(6.3 2.0) × 106
(2.6 1.0) × 107
(1.2 0.1) × 108
–
25
40
25
50
25
80
25
80
25
80
110
Bare SiO2
HMDS
OTS
0.044
0.041
0.042 0.002
0.040 0.001
0.069 0.001
0.021 0.002
0.049 0.005
0.70 0.02
–20
1
2
2
1
1
1
–23
–29
–20
–24
–41
0.070
Bare SiO2
HMDS
OTS
0.022
0.054
0.72
PF-BTDT
Bp-BTDT
Np-BTDT
BBTDT
Bare SiO2
HMDS
OTS
NAc)
–
–
(5.1 0.8) × 10−7
(1.3 0.1) × 10−6
NA c)
5.9 × 10−7
1.4 × 10−6
–
–51
–40
2
1
14
13
3
1
Bare SiO2
HMDS
OTS
–
–
(4.3 1.6) × 10−6
(7.6 1.9) × 10−7
0.0034 0.0006
0.0077 0.0004
(5.3 1.8) × 10−4
0.0062 0.0009
0.013 0.001
0.11 0.01
6.2 × 10−6
9.0 × 10−7
0.0042
0.0081
7.2 × 10−4
0.0068
0.014
–65
–37
–35
–33
–37
–13
–13
–16
–15
–24
–38
–26
–17
–15
–3
4
4
3
1
4
1
2
3
2
2
6
1
2
7
59 35
11
2
(4.5 0.1) × 105
(2.7 0.8) × 106
(1.2 0.1) × 105
(7.7 0.1) × 105
(1.5 1.4) × 106
(3.1 3.8) × 106
(5.8 2.2) × 106
(9.8 3.3) × 106
(8.3 1.8) × 107
(2.7 0.1) × 104
(4.5 2.6) × 106
(8.6 5.1) × 107
(7.0 3.0) × 103
(5.1 0.2) × 104
(1.3 0.6) × 105
(1.5 0.4) × 103
(1.6 1.0) × 105
(3.1 0.1) × 106
(9.3 7.5) × 102
(4.0 3.1) × 104
(2.7 2.3) × 106
Bare SiO2
HMDS
OTS
Bare SiO2
HMDS
OTS
0.11
Bare SiO2
HMDS
OTS
0.011
0.011 0.001
0.015 0.001
0.054 0.008
(2.9 0.1) × 10−4
0.074 0.024
0.072 0.008
0.017 0.002
0.021 0.003
0.041 0.004
0.026 0.004
0.073 0.012
0.071 0.014
0.044 0.010
0.040 0.023
0.13 0.02
0.015
0.062
2.9 × 10−4
0.088
Bare SiO2
HMDS
OTS
0.079
Bare SiO2
HMDS
OTS
0.018
1
0.023
–10
–18
–4
2
4
0.044
Bare SiO2
HMDS
OTS
0.028
1
0.084
–9
1
0.088
–13
–6
4
Bare SiO2
HMDS
OTS
0.051
1
1
1
0.072
–9
0.15
–9
a)Average mobilities are obtained for at least 5 devices; b)The values for P-BTDT are from ref. [9a]; c)Not active.
the higher mobility observed for P-BTDT versus Np-BTDT and
Bp-BTDT (Table 4).
the X-ray trends and the AFM-derived film topography is not
immediately obvious, and a possible extension of the current
work would be to examine the relative abundance of the A and
the B lattices as a function of film thickness.
BBTDT differs from P-BTDT, Np-BTDT, and Bp-BTDT
with respect to film uniformity. For a nominal film thickness
of 3 nm (≈1.5 unit cells), analysis of the (hkqz) intensity pro-
files (Figure 6C) show that the dominant BBTDT component
is 3 unit cells thick, implying inhomogeneous films comprised
of vertically extended aggregates separated by empty spaces.
This interpretation is qualitatively confirmed by the AFM
Note here that while the Bragg rod line shape analysis for
Bp-BTDT and Np-BTDT indicates that the films are largely one
unit cell thick, meaning that the A and the B lattices should
coexist within the same layer, the AFM images (Figure 7) show
that the films are more inhomogenous than that. Therefore, the
intuitive scenario, in which the degree of interaction between
the semiconducting molecules and the substrate gives rise to
distinct strain states of the A and the B lattices, may still be
valid. However, the exact origin of these differences between
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2013, 23, 3850–3865
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3861

www.afm-journal.de
www.MaterialsViews.com
Figure 8. Representative transfer plots for OTFTs fabricated from Bp-BTDT, Np-BTDT, and BBTDT films grown on OTS-coated substrates: a) Bp-BTDT,
transfer plot; p-type, μ = 0.11 cm2 V−1 s−1 , VT = –18 V, and Ion/Ioff = 3.5 × 105 in air. Substrate temperature = 80 °C, channel length = 50 μm, and channel
width = 2000 μm. b) Np-BTDT, transfer plot; p-type, μ = 0.079 cm2 V−1 s−1 , VT = –18 V, and Ion/Ioff = 3.7 × 107 in air. Substrate temperature = 80 °C,
channel length = 100 μm, and channel width = 5000 μm. c) BBTDT, transfer plot; p-type, μ = 0.15 cm2 V−1 s−1 , VT = –9 V, and Ion/Ioff = 3.0 × 106 in air.
Substrate temperature = 110 °C, channel length = 50 μm, and channel width = 2000 μm.
at TD = 25 °C to 0.15 cm2 V−1 s−1 at TD = 110 °C on OTS sub-
data (Figure 7), thereby suggesting a plausible reason for the
reduced carrier mobility observed for BBTDT versus P-BTDT,
despite the fact that the larger BBTDT π-core relative to the
other BTDT molecules is expected, all other things being equal,
to increase the electronic interaction between π orbitals of adja-
cent molecules,[22] and to decrease the Marcus reorganization
energy.[27,28]
strates. These trends are in excellent accord with the out-of-
plane X-ray diffraction data on the 50 nm thick semiconductor
films (Supporting Information Figure S3,S9). As described
in the TFT performance summary in Table 5, the enhance-
ment in film crystallinity is best reflected in the results for
Bp-BTDT and BBTDT. In both the cases, the intensity of crys-
talline reflections increases and the peaks become sharper as
the film growth temperature increases from 25 °C to 80 °C to
110 °C (Supporting Information Figure S3,S9), implying that
the size and population density of crystalline domains increases
in films deposited on hotter substrates. Notably, the Bp-BTDT
films on OTS substrates strongly exhibit this trend in the sense
that not only the peak intensities increase but also a number
of new peaks appear in the samples grown at 80 °C relative to
25 °C (Supporting Information Figure S4). Compared with the
other molecules, PF-BTDT fi lms exhibit less obvious growth
condition-mobility trends. In fact, the intensity of crystalline
peaks for films grown at 50 °C is somewhat less than for those
grown at 25 °C (Supporting Information Figure S5). This crys-
tallinity increase is marked given the grain size growth in the
50 nm thick semiconductor films (Supporting Information
Figure S15–S18).
The case of PF-BTDT is unique in that the thin film structure
does not correlate with the bulk single crystal structure. Never-
theless, note that the unit cell area for the surface-projected PF-
BTDT lattice differs by only 4% in comparison to the (001) crys-
tallographic plane of the single crystal. Furthermore, in thin
films, out-of-plane XRD measurements show that the PF-BTDT
molecules are aligned with their long molecular axes along sur-
face normal. If the local molecular packing of PF-BTDT were
identical to that in the single crystal, there would be little inter-
action between the π orbitals of adjacent molecules because the
adjacent molecules are aligned in an anti-parallel orientation
about their long molecular axes in the single crystal. Although
the PF-BTDT molecular packing in thin films does not adhere
to the packing structure of the single crystal form, note that
the GIWAXS measurements highlight the poor crystallinity
of the PF-BTDT fi lms. Specifically, for 3 nm films, the most
intense reflections from PF-BTDT are only twice as strong as
the OTS peak (Figure 4e). In marked contrast, the strong reflec-
tions from the other BTDT films are ≈5× more intense than
those from the underlying OTS SAM. Furthermore, PF-BTDT
forms vertically extended aggregates as in the case of BBTDT
(Figure 7c), which reduces the surface coverage of these rela-
tively thick films in the charge transport layer. Taken together,
the above factors convincingly explain why the PF-BTDT fi lms
yield the poorest performing OTFTs.
As suggested in the thermal and photo-oxidation evaluations
of these materials, the present BTDT semiconductors exhibit
excellent environmental stability in TFT performance. After
two months of shelf-life testing, the Bp-BTDT devices fabricated
on OTS substrates at 80 °C retain device performance with
μ = 0.16 cm2 V−1 s−1 (max) and Ion/Ioff = 6.4 × 104. In the case
of BBTDT, the devices fabricated on OTS substrates at 110 °C
exhibit μ = 0.17 cm2 V−1 s−1 (max) and Ion/Ioff = 7.5 × 104 after
two months. Previously we showed that P-BTDT fi lms exhibit
good device performance with μ = 0.3 cm2 V−1 s−1 and Ion/Ioff
=
Other than packing parameters within crystalline domains,
the film growth conditions also impact device performance
(Table 5). BTDT fi lms exhibit enhanced charge carrier mobility
as the growth temperature (TD) is increased. Among the pre-
sent materials, Bp-BTDT and BBTDT exhibit the most dramatic
changes in mobility. The mobility of Bp-BTDT-based OTFTs
increases from 7.2 × 10−4 cm2 V−1 s−1 (max) at TD = 25 °C to
0.11 cm2 V−1 s−1 (max) at TD = 80 °C, and the mobility of the
BBTDT-based devices increases from 0.044 cm2 V−1 s−1 (max)
107 after two months of shelf-life testing.[9] The origin of slightly
increased mobilities in the Bp-BTDT and BBTDT devices may
reflect additional charge carriers generated by O2 diffusion
into the films. When an organic TFT is exposed to air or O2,
holes are doped into trap states at the semiconductor–SiO2 gate
dielectric interface because O2 is electron acceptor. This hole
filling of trap states can sometimes lead to higher field-effect
mobilities.[29]
©
3862 wileyonlinelibrary.com
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2013, 23, 3850–3865

www.afm-journal.de
www.MaterialsViews.com
and then purified by gradient sublimation at pressures of ∼ 10−5 Torr at
270°C, giving a bright-yellow solid, 417 mg; yield, 85%. Mp: 392 °C. This
material was insufficiently soluble to obtain a useful 1H and 13C NMR
spectrum. Anal. Calcd for C24H14S3: C, 72.32; H, 3.54; Found: C, 72.25;
H, 3.63. HRMS (EI, m/z) calcd.: 398.0254 (M+). Found: 398.0258. The
molecular structure of Bp-BTDT has been confirmed by X-ray diffraction.
Synthesis of 2-Naphthalenebenzo[d,d′]thieno[3,2-b;4,5-b′]dithiophene
(Np-BTDT): Similar to above Bp-BTDT synthetic procedure,
2-bromonaphthalene (279 mg, 1.35 mmol) was used instead of
(4-bromophenyl)benzene. After simliar work up, the desired product
was purified by gradient sublimation at pressures of ∼ 10−5 Torr at
260 °C, giving a bright-yellow solid, 376 mg; yield, 82.4%. Mp: 299 °C.
1H NMR (300 MHz, DMSO-d6): δ 8.29 (d, J = 6.3 Hz, 1H), 8.23 (s, 1H),
8.12 (d, J = 7.8 Hz, 1H), 7.97 (m, 5H), 7.52 (m, 4H). This material was
insufficiently soluble to obtain a useful 13C NMR spectrum. Anal. Calcd
for C22H12S3: C, 70.93; H, 3.25;Found: C, 70.81; H, 3.34. HRMS (EI, m/z)
calcd.: 372.0101 (M+). Found: 372.0098.
4. Conclusions
In this contribution, we report a newly synthesized series of
BTDT molecules and characterize their thin film microstruc-
tures and electronic properties. X-ray studies on the BTDT thin
film structures in conjunction with single crystal data provide
a qualitative understanding of how the structural aspects of
the molecular solids influence the charge transport properties
in thin-film transistor architecture. In the case of P-BTDT, Bp-
BTDT, Np-BTDT, and BBTDT, two types of in-plane lattices co-
exist, and the dominance of a more strained lattice relative to
the bulk-like lattice may be associated with the excellent charge
transport properties of P-BTDT fi lms. On the other hand, poor
in-plane crystallinity and surface coverage at the dielectric inter-
facial layer explain the marginal device performance of PF-
BTDT fi lms. We believe that this study will serve as a structural
guide for the future development of thiophene - based organic
semiconductors.
Synthesis
of
2-Pentafluorophenylbenzo[d,d′]thieno[3,2-b;4,5-b′]
dithiophene (PF-BTDT): Similar to above Bp-BTDT synthetic procedure,
bromopentafluorobezene (0.17 mL, 1.36 mmol) was used instead of
(4-bromophenyl)benzene. After similar work up, the desired product
was purified by gradient sublimation at pressures of ≈10−5 Torr at
260 °C, giving a bright-yellow solid, 165 mg; yield, 35%. Mp: 316 °C.
1H NMR (500 MHz, CDCl3): δ7.89 (d, J = 8.0 Hz, 1H), 7.87 (d, J = 8.0
Hz, 1H), 7.79 (s, 1H), 7.47 (t, J = 8.0 Hz, 1H). 7.40 (t, J = 8.0 Hz, 1H).
This material was insufficiently soluble to obtain a useful 13C NMR
spectrum.19F NMR (282 MHz, CDCl3): δ -139, -154, -161. Anal. Calcd for
C18H5F5S3: C, 52.42; H, 1.22; Found: C, 52.35; H, 1.34. HRMS (EI, m/z)
calcd.: 411.9474 (M+). Found: 411.9476.
5. Experimental Section
Materials and Methods: All chemicals and solvents were of reagent
grade and were obtained from Aldrich, Arco, or TCI Chemical Co.
Reaction Solvents (toluene, benzene, ether, and THF) were distilled
under nitrogen from sodium/benzophenone ketyl, and halogenated
solvents were distilled from CaH2. 1H and 13C NMR spectra were
recorded on a Bruker 500 or a 300 instrument. Chemical shifts for 1H
and 13C NMR spectra were referenced to solvent signals. Differential
scanning calorimetry (DSC) was carried out on a Mettler DSC 822
instrument, calibrated with a pure indium sample at a scan rate of 10 K/
min. Thermogravimetric analysis (TGA) was performed on a Perkin Elmer
TGA-7 thermal analysis system using dry nitrogen as the carrier gas at a
flow rate of 40 mL/min. The UV–Vis absorption and fluorescence spectra
were obtained using JASCO V-530 and Hitachi F-4500 spectrometers,
respectively, and all spectra were measured in the indicated solvents
at room temperature. IR spectra were obtained using a JASCO FT/
IR-4100 spectrometer. Differential pulse voltammetry experiments were
performed with a CH Instruments model CHI621C Electrochemical
Analyzer. All measurements were carried out at the temperature
indicated with a conventional three-electrode configuration consisting of
a platinum disk working electrode, an auxiliary platinum wire electrode,
and a non-aqueous Ag reference electrode. The supporting electrolyte
was 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6) in a
specified dry solvent. All potentials reported are referenced to an Fc+/
Fc internal standard (at +0.6 V). Elemental analyses were performed on
a Heraeus CHN-O-Rapid elemental analyzer. Mass spectrometric data
were obtained with a JMS-700 HRMS instrument. Prime grade silicon
wafers (p+-Si) with ≈300 nm ( 5%) thermally grown oxide (from Montco
Silicon) were used as device substrates. Benzo[d,d′]thieno[3,2-b;4,5-b′]
dithiophene and bisbenzo[d,d′]thieno[3,2-b;4,5-b′] dithiophene (BBTDT)
were prepared according to the literature.[9]
Single Crystal Structure Determination: Crystals suitable for X-ray
diffraction were crystallized from a hot trimethylbenzene solution. Then
the single crystals were mounted in oil (Infineum V8512) on a glass
fiber under a nitrogen cold stream at 100(2)K or at room temperature.
X-Ray diffraction data were collected on Kappa diffractometer, equipped
with Cu I μS source and an APEX II CCD detector. Data were collected,
integrated and corrected for decay and Lp effects using Bruker APEX II
software. Final unit cell parameters were obtained through a refinement
of all observed reflections during data integration. A Face-indexed
absorption correction was performed via XPREP. The structure was
solved and refined using the SHELXTL suite of software. All of the non-
hydrogen atoms were refined anisotropically. Hydrogen atoms attached
to carbon atoms were fixed at calculated positions and refined using
a riding mode. Selected crystallographic information and additional
refinement details are provided in Supporting Information Table S2
and S5.
Thin Film Structure Characterization: Grazing incidence wide angle
X-ray scattering (GIWAXS) measurements on BTDT thin films were
carried out at room temperature at beamline 33BM-C of the Argonne
National Laboratory’s Advanced Photon Source (APS) using X-rays
of wavelength λ = 0.826 Å and a MAR 345 image plate area detector.
The incident beam size was 2.0 (h) x 0.05 (v) mm2 and the incident
flux was 5 × 1010 photons/s. The out-of-plane XRD data were collected
on an in-house 18 kW Rigaku ATXG diffractometer using a multilayer
parabolic mirror, NaI scintillation detector and X-rays of wavelength λ =
1.542 Å.
Synthesis
of
2-Biphenylbenzo[d,d′]thieno[3,2-b;4,5-b′]dithiophene
Analysis of Bragg Rod Intensity Profiles: Analysis of the (h k qz)
rods was completed for each BTDT by performing simultaneous
fits of background subtracted line profiles of (11qz), (02qz) and
(Bp-BTDT): Under nitrogen and anhydrous conditions at 0 °C, 2.5 M n-
BuLi (0.49 mL in hexanes, 1.23 mmol) was slowly added to a 20 mL
THF solution of BTDT (304 mg, 1.23 mmol) and the mixture was stirred
for 40 min. Next, tri-n-butyltinchloride (0.38 mL, 1.35 mmol) was added
and the mixture was stirred for 30 min at this temperature, then warmed
to room temperature and stirred overnight. After simple filtration under
nitrogen, THF was removed under vacuum and 30 mL toluene was
added. The toluene solution was then transferred to a (4-bromophenyl)
benzene(316 mg, 1.35 mmol) and tetrakis(triphenylphosphine)
palladium (57mg, 0.05 mmol) toluene (30 mL) solution and was refluxed
at 140 °C for 2 days. After cooling back to room tempertaure, the desired
solid product was collected by filtration, washed with hexanes and ether,
(12qz) rods. The intensity along the rod was calculated using
ꢀ
(
)
(
(
))
2, where t1, t2 and
0i
Ihk qz = Nhk t1 + P2t2 + P3 1 − t1 − t2
|
fi eiq r
|
t3 are the fractions of the surface covered iby 1, 2 and 3 unit cell thick
regions. P2 and P3 are the lattice sums over two and three unit cells
2
along the surface normal, specifically, P2 = 1 − ei 2q c
1 − ei q c
,
2
P3 = 1 − ei 3q c
1 − ei q c
.
The scattering vector
is given by
qꢀ
q = h a∗ − az∗ + k b∗ − bz∗ + qz, where az and bz are the surface
*
*
normal components of the reciprocal lattice vectors a∗ and b∗, and
depend upon the crystallographic angles α and β. The form factor
fi
ꢁ
of the ith molecule in the unit cell is given by
, where
3
fi = ei q r d ri
i
V
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2013, 23, 3850–3865
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3863

www.afm-journal.de
www.MaterialsViews.com
AFOSR (grants FA9550-08-1-0331 and FA9550-11-10275), by the
National Science Council, Taiwan, Republic of China (grants NSC100-
2628-M-008-004, NSC101-2113-M-008-005), and by the Industrial
Technology Research Institute of Taiwan (grant B361A51500). Use of
the Advanced Photon Source, 33BM-C, was supported by the U.S. DOE,
Office of Science, Office of Basic Energy Sciences, under Contract No.
DE-AC02-06CH11357.
(
)
(
)
(
)
( )
ri are the components of the
l
j,l
(ri )j
=
Rx η1 Ry η2 Rz η3
l
position vector
and Rx, Ry and Rz are the conventional rotation
ri
matrices. The position of the centroids of the ith molecule is r0i and Nhk
is a scale factor. Because the simulations span a large parameter space,
a number of solutions (local minima) can be obtained depending on
the starting values of the fitting parameters. Best-fits were obtained by
running the simulations for at least 20 different starting conditions.
OTFT Fabrication: TFTs were fabricated in bottom gate-top contact
configuration. Highly doped p-type (100) silicon wafers (<0.004 Ω cm)
were used as gate electrodes as well as substrates, and 300 nm SiO2
thermally grown on Si was used as the gate insulator. The unit area
capacitance was taken to be 10 nF cm−2 . The substrate surface was
treated with octadecatrichlorosilane (OTS) and hexamethyldisilazane
(HMDS) purchased from Sigma-Aldrich Chemical Co. A few drops
of HMDS were loaded inside a self-assembly chamber under an N2
blanket. The SiO2/Si substrates were exposed to this atmosphere for at
least 7.0 days to give a hydrophobic surface. After HMDS deposition,
the advancing aqueous contact angle is 95°. OTS-modified substrates
were fabricated by immersing Si/SiO2 substrates in 3.0 mM hexane
solutions of the silane reagent for 1 hour after leaving the solution in air
under 55–60% humidity for 10 h. After OTS deposition, the substrates
were sonicated with hexane, acetone, and, ethanol. The contact angle
of a water drop on OTS SAM is 104°. Semiconductor thin films (50 nm)
were next vapor-deposited onto the Si/SiO2 held at predetermined
temperatures of 25 °C, 80 °C for Bp-BTDT and Np-BTDT, 25 °C, 80 °C,
and 110 °C for BBTDT with a deposition rate of 0.3 Å/s at 6 × 10−6 Torr,
employing a high-vacuum deposition chamber (Denton Vacuum, Inc.,
USA). Gold source and drain electrodes (50 nm) were vapor-deposited
at 2 × 10−6 Torr through a shadow mask in the vacuum deposition
chamber. Devices were fabricated with two sets of device configuration,
a channel length of 50 μm and a channel width of 2000 μm as well as a
channel length of 100 μm and channel width of 5000 μm.
Received: November 21, 2012
Revised: December 19, 2012
Published online: March 13, 2013
[1] a) Y. G. Wen, Y. Q. Liu, Y. L. Guo, G. Yu, W. P. Hu, Chem. Rev.
2011, 111, 3358; b) M. J. Spijkman, K. Myny, E. C. P. Smits,
P. Heremans, P. W. M. Blom, D. M. de Leeuw, Adv. Mater. 2011,
23, 3231; c) W. Shao, H. L. Dong, L. Jiang, W. P. Hu, Chem. Sci.
2011, 2, 590; d) X. W. Zhan, A. Facchetti, S. Barlow, T. J. Marks,
M.A.Ratner,M.R.Wasielewski,S.R.Marder,Adv.Mater.2011,23,268;
e) A. A. Virkar, S. Mannsfeld, Z. A. Bao, N. Stingelin, Adv. Mater.
2010, 22, 3857; f) H. Sirringhaus, M. Bird, T. Richards, N. Zhao,
Adv. Mater. 2010, 22, 3893; g) W. H. Lee, J. H. Cho, K. Cho, J. Mater.
Chem. 2010, 20, 2549; h) G. Gelinck, P. Heremans, K. Nomoto,
T. D. Anthopoulos, Adv. Mater. 2010, 22, 3778; i) T. J. Marks, MRS
Bull. 2010, 35, 1018; j) M. E. Roberts, A. N. Sokolov, Z. N. Bao,
J. Mater. Chem. 2009, 19, 3351; k) M. L. Tang, S. C. B. Mannsfeld,
Y. S. Sun, H. A. Becerril, Z. N. Bao, J. Am. Chem. Soc. 2009, 131,
882; l) N. Kobayashi, M. Sasaki, K. Nomoto, Chem. Mater. 2009,
21, 552; m) M. L. Tang, A. D. Reichardt, N. Miyaki, R. M. Stolten-
berg, Z. Bao, J. Am. Chem. Soc. 2008, 130, 6064; n) S. Subramanian,
S. K. Park, S. R. Parkin, V. Podzorov, T. N. Jackson, J. E. Anthony,
J. Am. Chem. Soc. 2008, 130, 2706; o) Y. Didane, G. H. Mehl,
A. Kumagai, N. Yoshimoto, C. Videlot-Ackermann, H. Brisset,
J. Am. Chem. Soc. 2008, 130, 17681; p) A. R. Murphy, J. M. J. Frechet,
Chem. Rev. 2007, 107, 1066; q) C. Kim, A. Facchetti, T. J. Marks, Sci-
ence 2007, 318, 76; r) B. A. Jones, A. Facchetti, M. R. Wasielewski,
T. J. Marks, J. Am. Chem. Soc. 2007, 129, 15259.
OTFT Characterization: I–V plots of device performance were
measured under vacuum, and transfer and output plots were recorded
for each device. The I–V) characteristics of the devices were measured
using a Keithley 6430 subfemtoammeter and a Keithly 2400 source
meter, operated by a local Labview program and GPIB communication.
Key device parameters, such as charge carrier mobility (μ) and on-to-off
current ratio (Ion/Ioff), were extracted from the source-drain current (ISD
)
versus source-gate voltage (VSG) characteristics employing standard
procedures. Mobilities were obtained from the formula defined by the
saturation regime in transfer plots, μ = 2ISDL /(CiW (VSG − VT)2), where
ISD is the source-drain current, VSG is source-gate voltage, L is the
channel length, W is the channel width, Ci is the capacitance of the gate
dielectric, VT is the threshold voltage. Threshold voltage was obtained
from x intercept of VSG vs. ISD1/2 plots.
[2] a) H. Yan, Z. H. Chen, Y. Zheng, C. Newman, J. R. Quinn, F. Dotz,
M. Kastler, A. Facchetti, Nature 2009, 457, 679686; b) H. Usta,
A. FacchettiAcc. Chem. Res 2011, 44, 501.
[3] J. E. Anthony, Chem. Rev. 2006, 106, 5028.
[4] A. C. Arias, J. D. MacKenzie, I. McCulloch, J. Rivnay, A. Salleo, Chem.
Rev. 2010, 110, 3.
Film Morphology Measurements: AFM measurements were performed
using a JEOL-Microscope (JEOL Ltd. Japan) and Dimension Icon
Scanning Probe Microscope (Veeco, USA) in the tapping mode.
Crystallographic data (excluding structure factors) for the structures
reported in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication nos. CCDC
916020-916028. Copies of the data can be obtained free of charge from
the CCDC via www.ccdc.cam.ac.uk/data_request/cif.
[5] a) D. J. Gundlach, J. E. Royer, S. K. Park, S. Subramanian,
O. D. Jurchescu, B. H. Hamadani, A. J. Moad, R. J. Kline,
L. C. Teague, O. Kirillov, C. A. Richter, J. G. Kushmerick, L. J. Richter,
S. R. Parkin, T. N. Jackson, J. E. Anthony, Nat. Mater. 2008, 7, 216;
b) J. M. Choi, J. Lee, D. K. Hwang, J. H. Kim, S. Im, E. Kim, Appl.
Phys. Lett. 2006, 88, 082904; c) C. Santato, I. Manunza, A. Bonfiglio,
F. Cicoira, P. Cosseddu, R. Zamboni, M. Muccini, Appl. Phys. Lett.
2005, 86, 163511; d) J. Reynaert, D. Cheyns, D. Janssen, R. Muller,
V. I. Arkhipov, J. Genoe, G. Borghs, P. Heremans, J. Appl. Phys. 2005,
97, 061301; e) Q. Miao, M. Lefenfeld, T. Q. Nguyen, T. Siegrist,
C. Kloc, C. Nuckolls, Adv. Mater. 2005, 17, 407; f) F. Cicoira,
C. Santato, F. Dinelli, M. Murgia, M. A. Loi, F. Biscarini,
R. Zamboni, P. Heremans, M. Muccini, Adv. Funct. Mater. 2005,
15, 375; g) K. Ito, T. Suzuki, Y. Sakamoto, D. Kubota, Y. Inoue,
F.Sato,S.Tokito,Angew.Chem.Int.Ed.2003,42,1159;h)D.J.Gundlach,
J. A. Nichols, L. Zhou, T. N. Jackson, Appl. Phys. Lett. 2002, 80,
2925.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Acknowledgements
[6] a) C. Kim, P. Y. Huang, J. W. Jhuang, M. C. Chen, J. C. Ho, T. S. Hu,
J. Y. Yan, L. H. Chen, G. H. Lee, A. Facchetti, T. J. Marks, Org.
Electron. 2010, 11, 1363; b) Y. N. Li, Y. L. Wu, P. Liu, Z. Prostran,
S. Gardner, B. S. Ong, Chem. Mater. 2007, 19, 418; c) D. H. Kim,
This research, including use of the J.B. Cohen X-ray Diffraction
Facility, was supported by the NSF MRSEC program (grant DMR-
1121262) at the Materials Research Center of Northwestern U., by
©
3864 wileyonlinelibrary.com
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2013, 23, 3850–3865

www.afm-journal.de
www.MaterialsViews.com
D. Y. Lee, H. S. Lee, W. H. Lee, Y. H. Kim, J. I. Han, K. Cho, Adv.
Mater. 2007, 19, 678.
Funct. Mater. 2006, 16, 1859; f) S. Pratontep, F. Nuesch, L. Zuppiroli,
M. Brinkmann, Phys. Rev. B 2005, 72, 115101; g) G. Horowitz, Adv.
Funct. Mater. 2003, 13, 53; h) J. W. Orton, M. J. Powell, Rep. Prog.
Phys. 1980, 43, 1263.
[7] a) C. R. Swartz, S. R. Parkin, J. E. Bullock, J. E. Anthony, A. C. Mayer,
G. G. Malliaras, Org. Lett. 2005, 7, 3163; b) T. W. Kelley, P. F. Baude,
C. Gerlach, D. E. Ender, D. Muyres, M. A. Haase, D. E. Vogel,
S. D. Theiss, Chem. Mater. 2004, 16, 4413; c) H. Meng, M. Bendikov,
G. Mitchell, R. Helgeson, F. Wudl, Z. Bao, T. Siegrist, C. Kloc,
C. H. Chen, Adv. Mater. 2003, 15, 1090; d) T. W. Kelley,
L. D. Boardman, T. D. Dunbar, D. V. Muyres, M. J. Pellerite,
T. Y. P. Smith, J. Phys. Chem. B 2003, 107, 5877; e) A. R. Volkel,
R. A. Street, D. Knipp, Phys. Rev. B 2002, 66; f) H. Klauk, M. Halik,
U. Zschieschang, G. Schmid, W. Radlik, W. Weber, J. Appl. Phys.
2002, 92, 5259; g) C. D. Dimitrakopoulos, P. R. L. Malenfant, Adv.
Mater. 2002, 14, 99; h) F. Heringdorf, M. C. Reuter, R. M. Tromp,
Nature 2001, 412, 517; i) S. F. Nelson, Y. Y. Lin, D. J. Gundlach,
T. N. Jackson, Appl. Phys. Lett. 1998, 72, 1854; j) D. J. Gundlach,
Y. Y. Lin, T. N. Jackson, S. F. Nelson, D. G. Schlom, IEEE Electron
Device Lett. 1997, 18, 87.
[8] a) H. Yamada, Y. Yamashita, M. Kikuchi, H. Watanabe, T. Okujima,
H. Uno, T. Ogawa, K. Ohara, N. Ono, Chem.-Eur. J. 2005, 11, 6212;
b) P. Coppo, S. G. Yeates, Adv. Mater. 2005, 17, 3001; c) A. Maliakal,
K. Raghavachari, H. Katz, E. Chandross, T. Siegrist, Chem. Mater.
2004, 16, 4980.
[9] a) J. Youn, M. C. Chen, Y. J. Liang, H. Huang, R. P. Ortiz, C. Kim,
C. Stern, T. S. Hu, L. H. Chen, J. Y. Yan, A. Facchetti, T. J. Marks,
Chem. Mater. 2010, 22, 5031; b) P.-Y. Huang, L.-H. Chen, C. Kim,
H.-C. Chang, Y.-J. Liang, C.-Y. Feng, C.-M. Yeh, J.-C. Ho, C.-C. Lee,
M.-C. Chen, ACS Appl. Mater. Interfaces 2012, 4, 6992.
[19] a) J. Rivnay, L. H. Jimison, J. E. Northrup, M. F. Toney, R. Noriega,
S. F. Lu, T. J. Marks, A. Facchetti, A. Salleo, Nat. Mater. 2009, 8,
952; b) M. Tello, M. Chiesa, C. M. Duffy, H. Sirringhaus, Adv. Funct.
Mater. 2008, 18, 3907; c) G. E. Pike, C. H. Seager, J. Appl. Phys. 1979,
50, 3414.
[20] a) U. Zschieschang, F. Ante, M. Schlorholz, M. Schmidt, K. Kern,
H. Klauk, Adv. Mater. 2010, 22, 4489; b) M. Marchl, M. Edler,
A. Haase, A. Fian, G. Trimmel, T. Griesser, B. Stadlober, E. Zojer,
Adv. Mater. 2010, 22, 5361; c) K. C. Liao, A. G. Ismail, L. Kreplak,
J. Schwartz, I. G. Hill, Adv. Mater. 2010, 22, 3081; d) U. Kraft,
U. Zschieschang, F. Ante, D. Kalblein, C. Kamella, K. Amsharov,
M. Jansen, K. Kern, E. Weber, H. Klauk, J. Mater. Chem. 2010, 20,
6416; e) A. Virkar, S. Mannsfeld, J. H. Oh, M. F. Toney, Y. H. Tan,
G. Y. Liu, J. C. Scott, R. Miller, Z. Bao, Adv. Funct. Mater. 2009,
19, 1962; f) Y. Ito, A. A. Virkar, S. Mannsfeld, J. H. Oh, M. Toney,
J. Locklin, Z. A. Bao, J. Am. Chem. Soc. 2009, 131, 9396;
g) S. A. DiBenedetto, A. Facchetti, M. A. Ratner, T. J. Marks, Adv.
Mater. 2009, 21, 1407; h) H. S. Lee, D. H. Kim, J. H. Cho, M. Hwang,
Y. Jang, K. Cho, J. Am. Chem. Soc. 2008, 130, 10556; i) Organic
Field-Effect Transistors, (Ed: J. L. Zhenan Bao), CRC Press Taylor and
Francis Group 2007; j) H. C. Yang, T. J. Shin, M. M. Ling, K. Cho,
C. Y. Ryu, Z. N. Bao, J. Am. Chem. Soc. 2005, 127, 11542;
k) L. L. Chua, J. Zaumseil, J. F. Chang, E. C. W. Ou, P. K. H. Ho,
H. Sirringhaus, R. H. Friend, Nature 2005, 434, 194; l) G. Witte,
C. Woll, J. Mater. Res. 2004, 19, 1889; m) K. P. Pernstich,
S. Haas, D. Oberhoff, C. Goldmann, D. J. Gundlach, B. Batlogg,
A. N. Rashid, G. Schitter, J. Appl. Phys. 2004, 96, 6431;
n) S. Kobayashi, T. Nishikawa, T. Takenobu, S. Mori, T. Shimoda,
T. Mitani, H. Shimotani, N. Yoshimoto, S. Ogawa, Y. Iwasa, Nat.
Mater. 2004, 3, 317; o) Organic thin-film transistors with field-
effect mobility, Vol. 57th Annual, 164., (Ed: D. J. N. Gundlach,
C.-C. Jackson, T. N. Jackson), Device Research Conference Digest
1999.
[10] a) Y. Liu, C. A. Di, C. Y. Du, Y. Q. Liu, K. Lu, W. F. Qiu, G. Yu,
Chem.-Eur. J. 2010, 16, 2231; b) C. Kim, M. C. Chen, Y. J. Chiang,
Y. J. Guo, J. Youn, H. Huang, Y. J. Liang, Y. J. Lin, Y. W. Huang,
T. S. Hu, G. H. Lee, A. Facchetti, T. J. Marks, Org. Electron. 2010, 11,
801; c) L. Tan, L. Zhang, X. Jiang, X. D. Yang, L. J. Wang, Z. Wang,
L. Q. Li, W. P. Hu, Z. G. Shuai, L. Li, D. B. Zhu, Adv. Funct. Mater.
2009, 19, 272; d) P. Gao, D. Beckmann, H. N. Tsao, X. L. Feng,
V. Enkelmann, M. Baumgarten, W. Pisula, K. Mullen, Adv. Mater.
2009, 21, 213.
[11] S. Shinamura, E. Miyazaki, K. Takimiya, J. Org. Chem. 2010, 75,
1228; T. Yamamoto, K. Takimiya, J. Am. Chem. Soc. 2007, 129, 2224.
[12] Y. Liu, Y. Wang, W. P. Wu, Y. Q. Liu, H. X. Xi, L. M. Wang, W. F. Qiu,
K. Lu, C. Y. Du, G. Yu, Adv. Funct. Mater. 2009, 19, 772.
[13] J. H. Gao, R. J. Li, L. Q. Li, Q. Meng, H. Jiang, H. X. Li, W. P. Hu,
Adv. Mater. 2007, 19, 3008.
[14] Y. M. Sun, Y. W. Ma, Y. Q. Liu, Y. Y. Lin, Z. Y. Wang, Y. Wang,
C. G. Di, K. Xiao, X. M. Chen, W. F. Qiu, B. Zhang, G. Yu, W. P. Hu,
D. B. Zhu, Adv. Funct. Mater. 2006, 16, 426.
[15] a) S. C. B. Mannsfeld, M. L. Tang, Z. A. Bao, Adv. Mater. 2011,
23, 127; Y. Quan, S. C. B. Mannsfeld, M. L. Tang, M. Roberts,
M. F. Toney, D. M. DeLongchamp, Z. Bao, Chem. Mater. 2008, 20,
2763; b) T. N. Krauss, E. Barrena, X. N. Zhang, D. G. de Oteyza,
J. Major, V. Dehm, F. Wurthner, L. P. Cavalcanti, H. Dosch,
Langmuir 2008, 24, 12742.
[21] A. Dodabalapur, L. Torsi, H. E. Katz, Science 1995, 268, 270.
[22] a) J. Youn, P.-Y. Huang, Y.-W. Huang, M.-C. Chen, Y.-J. Lin,
H. Huang, R. Ponce Ortiz, C. Stern, M.-C. Chung, C.-Y. Feng,
L.-H. Chen, A. Facchetti, T. J. Marks, Adv. Funct. Mater. 2012, 22, 48;
b) M.-C. Chen, Y.-J. Chiang, C. Kim, Y.-J. Guo, S.-Y. Chen, Y.-J. Liang,
Y.-W. Huang, T.-S. Hu, G.-H. Lee, A. Facchetti, T. J. Marks, Chem.
Commun. 2009, 14, 1846.
[23] a) S. J. Liu, Q. Zhao, Q. L. Fan, W. Huang, Eur. J. Inorg. Chem.
2008, 2177; J. R. Quinn, F. W. Foss, L. Venkataraman, R. Breslow,
J. Am. Chem. Soc. 2007, 129, 12376; b) F. I. Wu, H. J. Su, C. F. Shu,
L. Y. Luo, W. G. Diau, C. H. Cheng, J. P. Duan, G. H. Lee, J. Mater.
Chem. 2005, 15, 1035.
[24] M. K. Durbin, A. G. Richter, C. J. Yu, J. Kmetko, J. M. Bai, P. Dutta,
Phys. Rev. E 1998, 58, 7686.
[25] G. Giri, E. Verploegen, S. C. B. Mannsfeld, S. Atahan-Evrenk,
D. H. Kim, S. Y. Lee, H. A. Becerril, A. Aspuru-Guzik, M. F. Toney,
Z. A. Bao, Nature 2011, 480, 504.
[16] S. C. B. Mannsfeld, A. Virkar, C. Reese, M. F. Toney, Z. N. Bao, Adv.
Mater. 2009, 21, 2294.
[17] Q. Yuan, S. C. B. Mannsfeld, M. L. Tang, M. F. Toney, J. Luening,
Z. A. Bao, J. Am. Chem. Soc. 2008, 130, 3502.
[26] Physics of Semiconductor Devices, (Ed: S. M. Sze), John Wiley &
Sons, New York 1981.
[18] a) B. Watts, T. Schuettfort, C. R. McNeill, Adv. Funct. Mater. 2011,
21, 1122; b) V. Kalihari, E. B. Tadmor, G. Haugstad, C. D. Frisbie,
Adv. Mater. 2008, 20, 4033; c) B. Stadlober, U. Haas, H. Maresch,
A. Haase, Phys. Rev. B 2006, 74, 165302; d) Y. B. Song, C. A. Di,
X. D. Yang, S. P. Li, W. Xu, Y. Q. Liu, L. M. Yang, Z. G. Shuai,
D. Q. Zhang, D. B. Zhu, J. Am. Chem. Soc. 2006, 128, 15940;
e) H. S. Lee, D. H. Kim, J. H. Cho, Y. D. Park, J. S. Kim, K. Cho, Adv.
[27] G. R. Hutchison, M. A. Ratner, T. J. Marks, J. Am. Chem. Soc. 2005,
127, 16866.
[28] R. P. Ortiz, J. Casado, V. Hernandez, J. T. L. Navarrete, J. A. Letizia,
M. A. Ratner, A. Facchetti, T. J. Marks, Chem.-Eur. J. 2009, 15,
5023.
[29] H. Okamoto, N. Kawasaki, Y. Kaji, Y. Kubozono, A. Fujiwara,
M. Yamaji, J. Am. Chem. Soc. 2008, 130, 10470.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2013, 23, 3850–3865
2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3865