Rubrene-based single-crystal organic semiconductors: Synthesis, electronic structure, and charge-transport properties

Article abstract of DOI:10.1021/cm400736s

Correlations among the molecular structure, crystal structure, electronic structure, and charge-carrier transport phenomena have been derived from six congeners (2-7) of rubrene (1). The congeners were synthesized via a three-step route from known 6,11-dichloro-5,12-tetracenedione. After crystallization, their packing structures were solved using single-crystal X-ray diffraction. Rubrenes 5-7 maintain the orthorhombic features of the parent rubrene (1) in their solid-state packing structures. Control of the packing structure in 5-7 provided the first series of systematically manipulated rubrenes that preserve the π-stacking motif of 1. Density functional theory calculations were performed at the B3LYP/6-31G(d,p) level of theory to evaluate the geometric and electronic structure of each derivative and reveal that key properties of rubrene (1) have been maintained. Intermolecular electronic couplings (transfer integrals) were calculated for each derivative to determine the propensity for charge-carrier transport. For rubrenes 5-7, evaluations of the transfer integrals and periodic electronic structures suggest these derivatives should exhibit transport characteristics equivalent to, or in some cases improved on, those of the parent rubrene (1), as well as the potential for ambipolar behavior. Single-crystal field-effect transistors were fabricated for 5-7, and these derivatives show ambipolar transport as predicted. Although device architecture has yet to be fully optimized, maximum hole (electron) mobilities of 1.54 (0.28) cm ² V^-1 s^-1 were measured for rubrene 5. This work lays a foundation to improve our understanding of charge-carrier transport phenomena in organic single-crystal semiconductors through the correlation of designed molecular and crystallographic changes to electronic and transport properties.

Full text of DOI:10.1021/cm400736s

Article
pubs.acs.org/cm
Rubrene-Based Single-Crystal Organic Semiconductors: Synthesis,
Electronic Structure, and Charge-Transport Properties
Kathryn A. McGarry,† Wei Xie,‡ Christopher Sutton,§ Chad Risko,§ Yanfei Wu,‡ Victor G. Young, Jr.,†
Jean-Luc Bredas,*,§ C. Daniel Frisbie,*,‡ and Christopher J. Douglas*,†
́
†Department of Chemistry, and ‡Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis,
Minnesota 55455, United States
§School of Chemistry and Biochemistry and Center for Organic Photonics and Electronics, Georgia Institute of Technology, Atlanta,
Georgia 30332, United States
S
* Supporting Information
ABSTRACT: Correlations among the molecular structure, crystal
structure, electronic structure, and charge-carrier transport phenom-
ena have been derived from six congeners (2−7) of rubrene (1).
The congeners were synthesized via a three-step route from known
6,11-dichloro-5,12-tetracenedione. After crystallization, their packing
structures were solved using single-crystal X-ray diffraction.
Rubrenes 5−7 maintain the orthorhombic features of the parent
rubrene (1) in their solid-state packing structures. Control of the
packing structure in 5−7 provided the first series of systematically
manipulated rubrenes that preserve the π-stacking motif of 1. Density functional theory calculations were performed at the
B3LYP/6-31G(d,p) level of theory to evaluate the geometric and electronic structure of each derivative and reveal that key
properties of rubrene (1) have been maintained. Intermolecular electronic couplings (transfer integrals) were calculated for each
derivative to determine the propensity for charge-carrier transport. For rubrenes 5−7, evaluations of the transfer integrals and
periodic electronic structures suggest these derivatives should exhibit transport characteristics equivalent to, or in some cases
improved on, those of the parent rubrene (1), as well as the potential for ambipolar behavior. Single-crystal field-effect transistors
were fabricated for 5−7, and these derivatives show ambipolar transport as predicted. Although device architecture has yet to be
fully optimized, maximum hole (electron) mobilities of 1.54 (0.28) cm2 V−1 s−1 were measured for rubrene 5. This work lays a
foundation to improve our understanding of charge-carrier transport phenomena in organic single-crystal semiconductors
through the correlation of designed molecular and crystallographic changes to electronic and transport properties.
KEYWORDS: rubrene derivatives, crystal engineering, electronic band structure, single-crystal field-effect transistors,
ambipolar transport
not yet fully understood. Specifically, the effect of molecular
structure on solid-state packing and resulting charge-carrier
transport in rubrene has not been fully established. Structure−
property relationship studies on rubrene derivatives would
provide some clarity; however, limited studies have examined
both molecular and packing structure in rubrene derivatives,5−7
and we know of only one study that has explored rubrene
derivatives in the SC-OFET architecture.8 We identified the gap
in knowledge needed to relate molecular structure to the
packing structure. While our work was underway, Bergantin
and Moret9 noted that no systematic trend has emerged to
reliably provide rubrenes that pack with a planar tetracene core,
which is needed to relate molecular structure to physical
properties. Our work addresses this gap by using chemical
synthesis to systematically manipulate the chemical structure of
INTRODUCTION
■
Rubrene (1, 5,6,11,12-tetraphenylnaphthacene, Figure 1a) has
intrigued materials chemists for years because of its exemplary
field-effect transistor properties. Notably, room-temperature
hole mobilities on the order of 20 cm2 V−1 s−1 have been
measured for 1 in single-crystal organic field-effect transistors
(SC-OFET).1 These large mobilities have led to extensive
rubrene-focused studies in an effort to better understand
charge-transport phenomena. Measurements have shown that
at room temperature the transport nature of single-crystal 1 is
dominated by band dispersion with a small effective mass,2
which is supported by previous observations of the Hall effect
in rubrene SC-OFETs.3 The charge-carrier mobility in rubrene
also exhibits bandlike temperature dependence, with low-
temperature transport dominated by shallow traps1 and high-
temperature transport recently proposed to be affected mainly
by thermal expansion of the lattice.4
Received: March 5, 2013
Revised: April 29, 2013
Despite the information obtained from these studies, a
number of fundamental aspects governing charge transport are
© XXXX American Chemical Society
A
dx.doi.org/10.1021/cm400736s | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 2. (a) Roll displacement from vector d of π-stacking tetracene
backbones (dark gray), viewed down molecular long-axis. Roll
displacement occurs along the short-axis of the tratracene backbone.
Rubrene exhibits no roll distortion. (b) Pitch displacement from vector
d of π-stacking tetracene backbones (dark gray), viewed down
molecular short-axis. Pitch displacement occurs along the long-axis of
the tetracene backbone. Rubrene exhibits pitch distortion. Hydrogen
atoms omitted for clarity.
Figure 1. (a) Molecular structure of rubrene (1); black arrow indicates
molecular long-axis, and gray arrow indicates molecular short-axis of
the tetracene backbone (core). (b) Crystal structure in the b−c plane;
green arrows indicate direction of π-stacks between tetracene
backbones. (c) Crystal structure in the a−c plane; blue arrows
indicate direction of layers (interlayer distance). Hydrogen atoms
omitted for clarity.
of a charge moving along the b-axis direction, an effect that has
been suggested to suppress transport.22 Lessening the effect of
this polarization, possibly achievable by increasing the distance
between the layers, may therefore be a way to improve the
charge-carrier mobility. Hence, changing the interlayer distance
in a series of derivatives would be worthwhile.
We focused our efforts on two aspects of the crystal
structure: maintaining the stacking/wave function overlap of
the tetracene backbones and expanding the interlayer distance.
Six rubrene derivatives (Figure 3) were synthesized, crystallized,
rubrene to develop a functional model for crystal engineering.
We then correlate solid-state packing with charge-carrier
transport characteristics of the synthesized rubrenes.
The SC-OFET is an ideal platform for structure−property
relationship studies as the impact of the molecular composition
on the solid-state packing and resulting mobility can be directly
evaluated.10−13 However, these studies hinge on the capacity to
systematically manipulate the solid state of a material. Changing
the molecular structure in a way that results in specific, directed
changes to the crystal structure remains a challenge due to the
unpredictable nature of solid-state packing.14 Further compli-
cating matters is optimization of device architecture to account
for other factors including contact resistance and changes in
absolute ionization energies (e.g., HOMO and LUMO
energies) across different, but related, structures in a class of
molecules. A variety of SC-OFET architectures exist that might
be employed for such studies. As described below, our choice of
top contact PMMA dielectric architecture rather than vacuum
gap is based on device optimization across the new rubrene
derivatives described in this Article. In the anticipation of
overcoming these challenges, knowing what aspects of the
solid-state structure to maintain or change is an appropriate
starting point.
The large charge-carrier mobilities measured for 1 have been
attributed to the packing motif (orthorhombic, Cmca space
group), which exhibits enough spatial overlap of the π-
conjugated tetracene backbone (Figure 1b) so as to lead to
large wave function overlap and significant intermolecular
electronic couplings along the π-stacking direction.15,16
Previous works on other acene-based semiconductors have
also correlated improved performance with increased π-
stacking.17−19 In addition, it has been recognized that solid-
state structures with large roll distortions (Figure 2a) can
destroy π-stacking, while pitch distortions (Figure 2b) tend to
preserve the π-stacks.20 This observation is a key consideration
when designing derivatives that will maintain the beneficial
orbital overlap.
Figure 3. Rubrenes 2−7 synthesized during this work.
and examined via a series of experimental and theoretical
approaches to explore the effects of the molecular structure on
solid-state packing and charge-transport properties. We begin
by first understanding how the substituents influence the
geometric, electronic, and redox properties of the isolated
molecules. We then focus on differences in the crystal packing
and how this modulates the electronic coupling between the
molecules. Finally, we measure SC-OFET mobilities and
correlate the device characteristics with these properties.
The packing motif of 1 also leads to anisotropic charge-
carrier transport, with the mobility along the b-axis about 4
times larger than that along the c-axis.21 The surrounding layers
of π-conjugated molecules down the crystallographic long-axis
(shown in Figure 1c), importantly, can polarize in the presence
EXPERIMENTAL SECTION
■
Synthesis. All reactions were carried out using flame-dried
glassware under a nitrogen or argon atmosphere unless aqueous
solutions were employed as reagents. Tetrahydrofuran (THF) was
dried by distillation from benzophenone/sodium. Dichloromethane
B
dx.doi.org/10.1021/cm400736s | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
W
2L
(CH2Cl2) and isopropanol were degassed by bubbling a stream of
argon through the liquid in a Schlenk flask and then stored and used in
a N2-filled glovebox. All other chemicals were purchased from Acros
Organics or Sigma-Aldrich and used as received. Analytical thin layer
chromatography (TLC) was carried out using 0.25 mm silica plates
from Silicycle. Eluted plates were visualized with UV light. Flash
chromatography was performed using 230−400 mesh (particle size
0.04−0.063 mm) silica gel purchased from Silicycle. See the
Supporting Information for representative experimental procedures,
tabulated data, and spectra of synthetic intermediates and derivatives
2−7.
Crystallography. Compounds were examined by single-crystal X-
ray diffraction using a Bruker SMART Platform CCD diffractometer
using graphite-monochromated Mo Kα radiation (λ = 0.71073 A) by
ω scans at 173(K) for 2,3 and 123(K) for 6 or a Bruker APEX II
Platform CCD diffractometer using graphite-monochromated Mo Kα
radiation (λ = 0.71073 A) by ω scans at 173(K) for 4 and 123(K) for 5
and 7. The structures were solved and refined using the SHELXTL-
PLUS packaged. The structures were solved by direct methods.
Cyclic Voltammetry. Measurements were carried out using a Pine
AFRDE5 bipotentiosat with an analog-to-digital converter (LabJack)
to log the data. Crystals were obtained by physical vapor transport and
dissolved in a solution of 0.1 M tetrabutylammonium perchlorate in
1,2-dichloroethane. The system consisted of a working electrode of
Au, a reference electrode of Ag/AgCl, and a counter electrode of Pt.
All measurements were correlated with ferrocene. A sweep rate of 100
V/s was found to be optimal.
ID(sat) =
μCi(VG − V )2
th
∂ ID
=
W
2L
μCi
∂VG
where ID is the drain current (A), VD is the drain voltage (V), VG is the
gate voltage (V), Vth is the threshold voltage (V), μ is the field-effect
mobility (cm2 V−1 s−1), Ci is the gate insulator specific capacitance (F/
cm2), and W and L are the channel width and length (μm).
RESULTS AND DISCUSSION
■
Synthesis. The most common routes to rubrene include
dimerization of a propargyl chloride,29,30 a one-pot conversion
from a propargyl alcohol,31 or the cycloaddition of an
isobenzofuran and 1,4-naphthoquinone followed by phenyl-
lithium addition and reduction by HI.32,33 Using a different
entry into the latter route described, rubrene derivatives 2−7
were synthesized beginning from known 6,11-dichloro-5,12-
tetracenedione (Scheme 1), which can be prepared on
a
Scheme 1. Synthesis of Rubrene Derivatives
Computational Methodology. Analyses of neutral ground-state
and radical-ion states were carried out using density functional theory
(DFT). The B3LYP23−25 functional and 6-31G(d,p)26 basis set were
used for all calculations. Guassian 09 (revision B.01)27 was used for all
isolated molecule and dimer calculations. Frequency analyses were
performed for the optimized geometries to ensure a minimum had
been reached. Band-structure and density of state (DOS) calculations
were carried out in Crystal 09 at the B3LYP/6-31G(d,p) level of
theory on the crystal structures.28 Uniform 6 × 6 × 6 Monkhorst−
Pack k-point meshes were employed for 1, 5, 6, and 8 × 2 × 4 for 7.
SC-OFET. The morphologies of the vapor-grown crystals for FET
fabrication were measured by a Bruker NanoScope V Multimode
scanning probe microscope operating in tapping mode. The probes
used were silicon nitride (Si3N4) cantilevers with integrated tapping
mode tips fabricated by Mikromasch USA (NSC 16, resonant
frequency 150−190 kHZ, spring constant 40 N/m). The AFM images
were analyzed using the software Gwyddion. High-resolution X-ray
diffraction was carried out in PANalytical X’pert Pro with
monochromatic Cu Kα radiation (wavelength 0.154 nm) at 45 kV
and 40 mA.
Highly p-doped silicon wafers (Silicon Valley Microelectronics.,
Inc., U.S.) with thermally grown 300 nm oxide (SiO2) were used as
substrates for fabricating single-crystal FETs. A thin layer of
poly(methyl methacrylate) (PMMA) (MW = 350 k, Sigma-Aldrich)
was spun cast from anhydrous 1,2-dichloroethane (Sigma-Aldrich) (10
mg/mL solution, 3000 rpm, 60 s) and was then baked at 120 °C for 1
h in a N2-filled glovebox. Final thickness of the PMMA was about 90
nm, determined from ellipsometry (VASE, J. A. Wollam Co., Inc.).
Total capacitance of the insulator layers was therefore about 8.5 nF/
cm2, which was confirmed by an independent capacitance measure-
ment in a HP 4192A impedance analyzer. In the next step, thin crystals
were laminated on the PMMA and a wetting front was spontaneously
formed, resulting in a clean crystal/insulator interface. Finally, 60 nm
Au layer was thermally evaporated (0.5 Å/s) on the crystals through a
a
Conditions: (a) Ar(R1)B(OH)2, Cs2CO3, Cl2Pd(dppf)2, C6H6, H2O,
dioxane; (b) Ar(R2,R2′)Br, n-BuLi, Et2O; (c) HI, Et2O.
multigram scale in three steps from commercially available
phthaloyl chloride and 1,4-naphthoquinone.34−36 Rubrene 1
was also synthesized via this route as a control. We note that
rubrenes 2 and 4 have been synthesized previously, but crystal
structures have not been reported.32,33
Using an aryl boronic acid, the first substituent set was
introduced via Suzuki coupling, and provided the correspond-
ing 6,11-diaryltetracene-5,12-diones in good-to-high yields
(64−95%). Nucleophilic addition of an aryllithium to the
quinone installed the second substituent set also in good-to-
high yields (65−96%). Reduction of the resulting diol using HI
afforded the desired rubrene derivatives, completing the three-
step sequence. Recrystallization of the crude rubrene product
using slow diffusion of isopropanol:dichloromethane (∼3:1)
provided our desired compounds as crystalline solids in
moderate yields (34−55%) but in sufficient quantities
(0.192−1.24 g). Single-crystal X-ray diffraction (XRD) was
performed on crystals grown from solution to obtain the solid-
state structures of 2−7. This route allowed us to control the
substituents and the symmetry in derivatives via late-stage
introduction of the aryl groups.
stainless steel shadow mask in a home-built thermal evaporator (base
−7
pressure 9 × 10
Torr), forming source and drain contacts. The
highly p-doped substrate served as the gate contact. Transistor
measurements were carried out in a Desert Cryogenics (Lakeshore,
Inc., U.S.) vacuum probe station in a N2-filled glovebox with Keithley
236 and 6517 electrometers and homemade Labview programs. All
measurements were done in dark, at room temperature and
atmospheric pressure. The field-effect mobility is calculated in the
saturation regime (VD > VG − Vth) on the basis of the equation:
Geometric and Electronic Structure. Before turning to
the charge-transport properties, it is of interest to first
understand the influence of the substituents on the (isolated)
geometric and electronic structures. We focus explicitly on how
the substituents on the external phenyl rings influence (i) the
degree of twist (from planarity) of the tetracene core itself and
C
dx.doi.org/10.1021/cm400736s | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
(ii) the out-of-plane torsion of the carbon−carbon bridge
connecting the arylene rings with the tetracene backbone
(Figures 4 and 5). We then use density functional theory
electron-donating methyl in 2, 3, and 4, however, induces a
fairly dramatic change. Methyl substitution in the para-position
(2) on the arylene rings induces a twist of 33.9° (Figure 4a)
and 42.8° for the different molecules in the unit cell, possibly a
result of steric repulsion. Compound 3, which combines meta-
and para-substituted methyl groups, demonstrates a slightly
smaller twist of 30.3° due to a slightly smaller steric strain
(Figure 4b), while 4 has a larger twist of 41.2°, indicating that
steric strain may not be the overriding influence (Figure 4c).
Rubrenes 5, 6, and 7 were synthesized to determine the
effect of an electron-deficient substituent (perfluoromethyl) on
the tetracene core. This substitution pattern induced a planar
tetracene core (Figure 4d−f) in all three derivatives, likely due
to the now attractive interaction between the side phenyls.37,38
Comparing 5 and 7 again suggests that perceived steric
interactions may not be the overriding factor in determining
core planarity. The para-substitution of both the trifluor-
omethyl and the methyl groups in 5 should seemingly increase
the steric interactions and potentially lead to a twisted
structure, while the lack of substitution on two arylene groups
in 7 should show a reduced degree of steric interactions. We
note, however, that both systems exhibit a planar tetracene core
in the crystal structure, indicating that there exists a
combination of factors that determine the planarity of the core.
The other parameter of interest, which is coupled to the
degree of backbone twisting, is the out-of-plane torsion of the
carbon−carbon bond connecting the arylene rings with the
tetracene backbone (Figure 5), a motion that can aid in
reducing steric interactions among neighboring arylene groups;
this out-of-plane parameter is only discussed for the molecules
having a planar backbone in the crystal structure (i.e., 1, 5, 6,
and 7) as it is difficult to define in 2, 3, and 4. For 1, the phenyl
rings move to 14.8° above and below the tetracene plane,
whereas there is a modestly larger shift (14.9° and 18.6°) for 6,
which allows the dipoles of the para-trifluoromethyl and one of
the meta-methyl groups to align. For 5 and 7, twists of ∼13.6°
and ∼13.0° were measured from the crystal structure,
suggesting that having an attractive interaction between the
two side phenyls can lead to a smaller out-of-plane twisting. We
continue our discussion of the side phenyl interactions in the
section on crystal packing.
Figure 4. ORTEP drawings viewed down the molecular long-axis of
(a) 2 (33.9° only), (b) 3, (c) 4, (d) 5, (e) 6, and (f) 7. Thermal
ellipsoids shown at 50% probability; hydrogen atoms omitted for
clarity. Full thermal ellipsoid plots can be found in Supporting
Information, Figure S1.
Figure 5. (a) The torsion angle along the backbone (ϕ) was
determined by the C−C double bonds at either end of the tetracene
backbone (labeled as C1, C2, C3, C4); (b,c) the out-of-plane torsion is
calculated for the constrained backbone (i.e., ϕ = 0) and is defined as
the dihedral angle comprising the center carbon of the tetracene
backbone and two adjacent carbons along the backbone that are
bonded to the substituted arylenes. The torsion is defined such that
the fourth carbon is bridging the arylene and the angle is zero if the
plane of the tetracene bisects the arylene in half.
Having examined the effects of substituents on the geometric
structure, we next delineate how these arylene substituents
modify the electronic structure. For 5−7 (using the molecular
geometries from the crystal), the B3LYP/6-31G(d,p)-deter-
mined HOMO and LUMO are delocalized along the tetracene
core (similar to the HOMO and LUMO of rubrene; see
Supporting Information, Figures S3−S4). The additions of the
electron-withdrawing groups stabilize the HOMO (LUMO) by
(DFT) calculations at the B3LYP/6-31G(d,p) level of theory
and the molecular geometries from the crystals to evaluate the
electronic structure.
Starting with the degree of twisting in the tetracene core, we
find in rubrene (1) that the tetracene core is planar, as was
previously well established.5 Substitution of protons for an
Table 1. B3LYP/6-31G(d,p)-Determined Adiabatic Ionization Potentials (AIP), Electron Affinities (AEA), Intramolecular
Reorganization Energies (λ), and Redox Potentials Determined via Cyclic Voltammetry for 1−7
rubrene
AIP (eV)
λ hole (eV)
AEA (eV)
λ electron (eV)
E1/2ox (V)
E1/2red (V)
Eg (V)
1
2
3
4
5
6
7
5.71
5.56
5.54
5.63
5.93
5.89
5.99
0.082
0.097
0.097
0.090
0.095
0.086
0.085
−1.06
−1.00
−0.96
−1.03
−1.34
−1.35
−1.41
0.098
0.097
0.102
0.098
0.127
0.121
0.124
0.36
0.29
0.29
0.33
0.44
0.43
0.49
−2.03
2.39
a
b
−
−
−
−
−
−
a
b
a
b
−1.89
2.33
a
b
−
−
−1.91
2.39
a
b
Event was not measured. Could not be calculated.
D
dx.doi.org/10.1021/cm400736s | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
some 0.20−0.29 eV (0.16−0.25 eV) in comparison to 1. For
2−4, the electron-donating groups destabilize the HOMO
(LUMO) by 0.12−0.16 eV (0.07−0.14 eV) as compared to 1.
This suggests that compounds 5−7 should be slightly more
difficult to oxidize, yet easier to reduce, than rubrene.
Conversely, the electron-donating groups for 2−4 should
make these compounds slightly more difficult to reduce, yet
easier to oxidize than rubrene. We note, however, that the
molecules have very similar HOMO−LUMO gaps, indicating
an inductive influence of the substituents on the electronic
structure.
Redox Properties. The oxidation potentials of the rubrene
derivatives were measured in 0.1 M tetrabutylammonium
perchlorate in 1,2-dichloroethane (potential traces can be found
in the Supporting Information, Figures S5−S11). Compounds
2−4 have slightly smaller oxidation potentials as compared to 1
(Table 1), while all three derivatives with trifluoromethyl
groups (5, 6, 7) have slightly larger oxidation potentials as
compared to 1. Reduction potentials for 1, 5, and 7 were able
to be measured and the Eg could then be determined, which
were found to be similar. The redox data are in good agreement
with the theoretical estimates from the evaluation of the
electronic structure.
induces the molecules to adopt the desired planar core
conformation. Conversely, the twisted backbone of derivatives
2−5 might be explained by the increased repulsion of the
electron density of the π-clouds. Intermolecular C−H···F
contacts likely influence the packing structures of 5−7 as well.
Short contacts (less than the sum of the van der Waals radii)
between hydrogen and fluorine were found to occur in all three
derivatives (see Supporting Information, Table S3).46−48 These
contacts likely stabilize the solid-state structure, similar to the
interactions previously observed in a variety of fluoroben-
zenes.49 We will limit our discussion on the influence of
substitution on crystal packing to compounds 1, 5, 6, and 7.
Indeed, the crystal structures for compound 5 (orthorhom-
bic, Pbcm), 6 (orthorhombic, Pbcm), and 7 (orthorhombic,
Pnma) display the planar backbone and preferential stacking of
the π-conjugated cores along one crystallographic axis (a-axis),
similar to the solid-state structure of rubrene, where π-stacking
occurs along the b-axis (Figure 6, left). Upon closer inspection,
The electron affinity (EA) and ionization potential (IP) for
the rubrene derivatives followed a similar trend when computed
through ΔSCF or the DFT analogue of Koopmans’ theorem;
therefore, we will discuss only the former. ΔSCF calculated IPs
and EAs were determined from evaluation of the gas-phase
optimized neutral, radical-cation, and radical-anion structures at
the UB3LYP/6-31G(d,p) level of theory. The adiabatic IP for
2−7 ranges from 5.54 to 5.99 eV (0.17 eV lower than to 0.28
eV larger than that for rubrene (5.71 eV)). The adiabatic EA for
2−7 ranges from −0.96 to −1.41 eV (0.10 eV lower than to
0.35 eV larger than that of rubrene (−1.06 eV)), which for 5−7
is slightly larger than the 0.12−0.14 eV experimentally observed
(Table 1). The adiabatic ionization potential/electron affinity
trends are in good agreement both with the electronic structure
and with electrochemical results. The evaluation of the neutral,
radical-cation, and radical-anion states also allows for evaluation
of the intramolecular reorganization energy, a key parameter in
the semiclassical Marcus theory often applied to the description
of hopping (weak electronic coupling) transport in organic
solids.39,40 The calculated intramolecular reorganization en-
ergies for 2−7 are comparable (although slightly larger) to that
of rubrene (Table 1), revealing that the substituents on the side
phenyls have a minimal impact on the reorganization energy
(should charge localization occur).
Crystal Packing and Intermolecular Electronic Cou-
pling. Compounds 2−7 (Figure 3) were synthesized and
crystallized to demonstrate the effect that substitutions in the
para-/meta-position have on the interlayer distance of the solid
state, provided the π-stacking remains intact. Single-crystal
XRD revealed that 2 (monoclinic, C2/c), 3 (monoclinic, P21),
and 4 (orthorhombic, Pna21) have twisted tetracene backbones
preventing effective π-stacking between molecules in the solid
state, although it is possible other, still unidentified, polymorphs
might be found for these derivatives. We postulate that the
interactions between the side phenyls strongly contribute to the
planarization of the tetracene core. Studies by Siegel et al. have
previously shown that interactions between peri-phenyl groups
are dominated by electrostatics and not charge transfer or
electron donor−acceptor effects.41−45 Therefore, the decreased
electron density of the trifluoromethyl-substituted phenyls
Figure 6. Crystal packing 2D projections showing π-stacking (left)
with measured intermolecular distance indicated by green arrows
(value at left) and expansion of the crystallographic long-axis (right)
with measured interplanar distance between layers indicated by blue
arrows (value at right) of (a) 1, (b) 5, (c) 6, and (d) 7. The
intersection of the crystallographic cell axes does not indicate the unit
cell origin; however, the directions are correct relative to each other.
Hydrogen atoms omitted for clarity.
we find that the interlayer distance has expanded for all three
compounds as compared to 1 (Figure 6, right): (i) 5, which has
substituents in both para-positions, exhibits the largest
interlayer distance increase and the most significant tightening
of the backbone stacking distance (Figure 6b); (ii) 7, having
only two phenyls bearing a para-substituent, exhibits the
smallest increase to the interlayer distance (Figure 6d); and
(iii) 6, having both para- and meta-substituents, has the least
significant tightening of the stacking distance (Figure 6c).
These trends indicate that steric interactions play a large role in
the expansion of the interlayer distance; however, electrostatic
interactions may be the key determinant in the backbone
stacking distance (see below).
One notable disparity and some potential imperfections are
of note on the solid-state packing of these derivatives. Viewed
down the crystallographic long-axis, the layers of parent
rubrene 1 tilt in the same direction, which is the same for
rubrenes 5 and 7 (see Supporting Information, Figure S2). The
layers of derivative 6, on the other hand, alternate the direction
of tilt, which may have an impact on the physical properties of
this compound. As rubrene derivatives, 5−7 may undergo
E
dx.doi.org/10.1021/cm400736s | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
photooxidation similar to 1,32,50 although the presence of the
fluorine atoms on the derivatives may improve their stability in
air.51 To minimize this possibility, all synthetic and device
preparation procedures were performed in the air as little as
possible.
the dipoles align in a parallel fashion. Rubrene 6 is a notable
exception to the trend, however, as increased steric interactions
provided by the meta-methyl groups counteract the attractive
electrostatic forces; these steric interactions, in addition, also
result in 6 being significantly displaced from a cofacial
arrangement (i.e., displaced along the y-direction).
The electronic couplings suggest that a charge carrier is
expected to preferentially proceed along the b-axis for 1 and the
a-axes in 5, 6, and 7. Full (periodic) electronic band structure
calculations were carried out for the crystal structures of 1, 5, 6,
and 7 to provide a further prediction of the transport properties
along the high-symmetry points in reciprocal space. The results
of the electronic coupling are corroborated by the three-
dimensional band structure (Figure 8) and computed effective
masses (see Supporting Information, Table S7).
We now turn to how these changes in the packing influence
the intermolecular electronic coupling, a second key parameter
in terms of understanding charge-carrier mobility in these
systems. Methods previously discussed in detail52,53 were used
to evaluate the transfer integral (electronic coupling), t, for π-
stacked dimers of molecules 1, 5, 6, and 7, and reveal electronic
couplings in the newly synthesized molecules that are slightly
larger than that of rubrene. The largest t’s are obtained, as one
might expect, for the molecules along the π-stacking direction:
the b-axis in 1 (100 meV for holes and 53 meV for electrons)
and the a-axis in 5 (134, 82 meV), 6 (95, 79 meV), and 7 (126,
71 meV). Small couplings are noted for the edge-to-face dimers
along the b-axis for 5, 6, and 7 (<20 meV). This trend is not
surprising as the intermolecular wave function overlap is
expected to be larger for the dimers with substantial overlap
between the tetracene backbones and in closer proximity to one
another. Indeed, the change in coupling along the different axes
suggests these molecules will display the anisotropic mobility
characteristic of rubrene.
To further understand the impact of intermolecular
separation (DAB) on the transfer integral, a model tetracene
backbone was displaced from the DAB value of 3.68 Å found in
the crystal structure of 1 in the direction normal to the parallel
planes of the tetracene backbone in increments of 0.01 Å from
3.48 to 3.88 Å (i.e., displaced along the z-direction, Figure 7).
Figure 8. Electronic band structures along the Y → Γ direction for (a)
1 and the X → Γ direction for (b) 5, (c) 6, and (d) 7. The conduction
band (CB) and valence band (VB) are labeled.
For 5, 6, and 7, the top of the valence band and bottom of
conduction band are found at the Γ-point, indicating a direct
band gap in these materials. The π−π coupling in the a-
direction is represented by the X-direction in the Brillouin zone
and has a bandwidth of 410 (360) for 5, 610 (360) for 6, and
580 (320) meV for 7 for the valence (conduction) bands.
These values are approximately 4t, as one might expect from a
simple tight-binding approximation. Negligible band disper-
sions were observed in the Y(0,1,0) and Z(0,0,1) directions of
the Brillouin zone (corresponding to b and c components
(short molecular axis)), respectively, which again suggests these
derivatives will exhibit anisotropy in mobility similar to that of
rubrene 1. Therefore, the band structure and intermolecular
electronic coupling calculations in total suggest that 5, 6, and 7
should have similar, or even in some cases improved, intrinsic
charge-carrier properties relative to 1.
Figure 7. B3LYP/6-31G(d,p) calculated tH (black) and tL (blue) for 1
⧫
■
▲
●
( ), 5 ( ), 6 ( ), and 7 ( ) overlaid on the tH (black line) and tL
(blue line) for model tetracene dimer displaced from 3.48 to 3.88 Å.
The right axis displays the dipole moments for isolated 1, 5, 6, and 7
(empty symbols) taken from the crystal structure.
Structural Characterization and Device Measure-
ments of SC-OFETs. For device fabrication, single crystals
were grown via horizontal physical vapor transport using
ultrapure Ar as the carrier gas.55 Because of the twisted core in
rubrenes 2−4, our discussion will focus on 5, 6, and 7, for
which thin and plate-like crystals (a few hundred nanometers in
thickness, Figure 9a) were collected in the growth tube after
about 2 days at 250 °C. No further purification of the source
materials was performed. Before device studies, as-grown
crystals of compounds 5, 6, and 7 were structurally
characterized by high-resolution XRD and AFM techniques
to confirm the solid-state packing matched the previous single-
crystal analysis (Figure 9b,c).
The intermolecular electronic coupling decays exponentially as
one might expect,54 and reveals that the changes in coupling
across the series are predominantly affected by changes in DAB.
Figure 7 also shows the dipole moments that point along the
long-axis of the tetracene core for each molecule in the crystal-
structure geometry. The dipole moment magnitudes correlate
well with DAB, that is, larger dipole moments lead to smaller π-
stacking distances as compared to 1, and exposes an important
contribution to how changing the nature of the molecular
electrostatics via functionalization affects the molecular packing.
Note that as one moves along the backbone-stacking direction
F
dx.doi.org/10.1021/cm400736s | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
As a check of our synthetic route, we first tested 1, both
commercial and synthetic material, in the vacuum gap
architecture (lamination-type bottom contact, polydimethylsi-
loxane (PDMS) substrates), which is known to provide the
highest mobilities of rubrene.1,56 This positive control experi-
ment showed that both materials performed the same, within
error (room temperature mobility of 14 cm2 V−1 s−1, see
Supporting Information, Figure S16(b) for synthetic rubrene
and Figure S19(b) for commercial rubrene). These measure-
ments also suggest that our purification protocol for synthetic
rubrenes, mixed solvent recrystallization followed by a single
sublimation, provides high-purity crystals suitable for transport
measurements. Initial studies of our derivatives 5−7 in the
same vacuum gap architecture were unsatisfactory (see the
discussions below), after which we turned to an alternate device
structure in which the top-contact strategy was employed.
Single-crystal FETs were constructed with top-contact,
bottom-gate geometry with PMMA as the dielectric layer and
Au as the source/drain contacts (Figure 10a). Source/drain
electrodes were patterned such that the channel was aligned
Figure 9. (a) Optical image of a rubrene 5 single crystal. Difference in
color results from different crystal thickness. (b) High-resolution wide-
angle XRD patterns for rubrene 1, 5, 6, and 7. (Left) 2θ−ω scans for
(002) or (020) peaks. The intermolecular layer spacings of each crystal
determined from the peak positions are also labeled. (Right) Rocking
curves (ω scan) for each crystal. Full-width-at-half-maximum
(FWHM) is about 0.03°. Note that all diffraction plots are shifted
vertically for a clear comparison. (c) Atomic-force microscopy (AFM)
height image for 5. The inset shows the height profile of molecular
steps. Images of rubrenes 6 and 7 can be found in the Supporting
Information.
Figure 9b shows the results of wide-angle 2θ−ω coupled
scan (left plot) and ω-scan rocking curve (right plot) for single
crystals of 5, 6, and 7; 1 is included for comparison. Only the
first-order peaks ((002) for 5 and 6, (020) for 7) are shown for
clarity; however, other high order peaks in the same family can
also be repeatedly observed. The peak position shift, from
which the d-spacing of the (002) or (020) planes (i.e., the
layer−layer distance in the long-axis of unit cell) is determined,
indicates the interlayer expansion of 5, 6, and 7. These values
are consistent with the results measured from single-crystal
XRD in Figure 6, suggesting that the vapor-grown crystals
maintain their as-synthesized orthorhombic structure. The full-
width-at-half-maximum (FWHM) of the rocking curve analysis
for different crystals is about 0.02−0.03°, indicating those
crystals preserve high-quality well-aligned surface planes. The
AFM height image of 5 in Figure 9c reveals molecular steps,
with the step height corresponding to the interlayer spacing,
reaffirming the crystal’s high quality necessary for good charge
transport.
Figure 10. (a) Device structure (bottom gate, top contact) of SC-
OFET with PMMA dielectric. (b,c) Transfer characteristics (ID−VG)
for (b) 1 and (c) 5 SC-OFET gated with PMMA. (Left axis, open
circles) Drain current (ID) plotted in log-scale. (Right axis, solid lines)
Square-root of drain current (ID) in linear-scale. Device dimensions:
for 1, W = 300 μm, L = 500 μm; for 5, W = 700 μm, L = 500 μm. Plots
for rubrenes 6 and 7 can be found in the Supporting Information.
G
dx.doi.org/10.1021/cm400736s | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Table 2. Summary of μ (cm2 V−1 s−1) and Vth (V) SC-OFET Data for 1, 5−7 in PMMA Gated, Top Contact Architecture
rubrene
μh, avg (sdev)
Vth,h, avg (sdev)
μh, max
Vth,h
μe, avg (sdev)
Vth,e, avg (sdev)
μe, max
Vth,e
1
5
6
7
0.43 (0.27)
0.42 (0.40)
0.074 (0.02)
0.25 (0.19)
−4.1 (5.2)
−25.8 (4.3)
−22.9 (6.9)
−25.5 (4.4)
0.83
1.54
0.10
0.63
0
−25
−23
−25
0.0038 (0.0035)
0.091 (0.07)
0.008 (0.005)
0.12 (0.09)
79.4 (18.5)
69.0 (23.5)
68.7 (10.8)
73.5 (7.2)
0.0077
0.28
110
90
0.013
0.22
78
82
along the most favorable direction for transport in the crystal,
along the b-axis for 1, and along the a-axis for 5−7. All devices
were fabricated in air, but electrical measurements were taken
in the dark at atmospheric pressure in a N2-filled glovebox.
Rubrene 1 was first measured in this architecture (Figure
10b). Hole accumulation occurred with negligible hysteresis
when a negative gate voltage (VG) was applied (at a drain
voltage (VD) of −60 V), resulting in a hole mobility of 0.83 cm2
V−1 s−1 and a threshold voltage (Vth) very close to 0 V.
Although much smaller than the benchmark mobility of 20 cm2
V−1 s−1 seen in vacuum-gap devices, this hole mobility agrees
with previously reported values measured in similar device
structures (with PMMA gate dielectric).57,58 Conversely,
electrons were barely accumulated in 1 even as a large positive
VG was applied (electron mobility 0.0077 cm2 V−1 s−1). This
small electron mobility likely results from the fabrication of
these devices in air, which allows electrons to be trapped by
water and oxygen.59 The large hysteresis in the electron
accumulation sweep (at VD = 60 V) and large electron
threshold voltage (Vth,e = 110 V) are also indicative of electrons
being trapped. Alternatively, diminished electron mobility may
be due to the large electron injection barrier present caused by
the misalignment of the Au work function (5.1 eV) and rubrene
(1) LUMO (3.2 eV).57,58 The measurements on 1 provide a
point of comparison for our studies.
For devices built from compounds 5, 6, and 7, the transfer
curves exhibit the V-shape characteristic of ambipolar transport,
with one arm indicating electron transport and the other
indicating hole transport (5, Figure 10c; 6 and 7, see
Supporting Information, section I). For negative (positive)
drain bias, only holes (electrons) were accumulated in the
semiconductor channel at the largest negative (positive) gate
voltages, which are referred to as the unipolar regimes. Despite
using symmetric Au contacts and exposure of the crystals to air,
5−7 demonstrate fairly well-balanced and large hole and
electron mobilities, which compare well with the results for 1
(Table 2). Although compound 5 had negligible hysteresis in
the transfer curve for both hole and electron transport, large
threshold voltages (Vth,h = −25 V and Vth,e = 90 V) imply that a
significant number of traps still exist at the crystal/insulator
interface, which may be explained by photooxidation of the
molecules at the crystal surface.60,61 Compounds 6 and 7 had
similar hole/electron threshold voltages as compared to 5 (for
6, Vth,h = −23 V and Vth,e = 78 V; for 7, Vth,h = −25 V and Vth,e
= 82 V), but the hysteresis was more pronounced. Never-
theless, the observed mobilities are comparable to the best
ambipolar wide band gap molecular crystals to date,62,63
suggesting these derivatives have maintained important solid-
state properties necessary for charge transport.
be explained in part by the larger electron affinities of 5−7
versus 1, which lower the injection barriers for electrons in the
derivatives by about 0.2−0.3 eV, regardless of the possible
presence of interface dipoles that may change the barrier height.
The energy stabilization of the electron affinity (LUMO) in 5−
7 also facilitates the operation of electron transport even
though all devices were exposed to the air during fabrication.
From the perspective of intermolecular electronic coupling,
electron mobilities were expected to be improved in the
derivatives because the transfer integrals of 5−7 are about 20−
30 meV larger than that of 1 (Figure 11) due to the closer
stacking of the tetracene backbones and enhanced wave
function overlap.
Figure 11. Correlation of hole and electron mobilities of rubrenes 1, 5,
6, and 7 with their respective calculated transfer integral values. These
mobilities are measured in a bottom gate (Au), top contact (Au)
structure with PMMA dielectric layer, as shown in Figure 8. The red
■
●
“ ” stand for maximum values, while the “ ” are the average values
with one standard deviation.
The hole mobility of 5−7 was also expected to be
comparable to or even exceed the hole mobility of 1 (for
holes, t is 100, 134, 95, and 126 meV for rubrene, 1, 5, 6, and
7), provided the electronic coupling is a dominant factor in
determining the carrier mobility. Indeed, as seen from Figure
11, the hole mobilities roughly trend with the transfer integrals
for all four rubrene compounds in this particular structure with
PMMA insulator. We note that the broad distribution of
mobility is inherently due to the staggered top-contact structure
we employed, in which an extra access resistance is introduced
and mobility is thus strongly dependent on the crystal
thickness.64 However, it is difficult to conclude at this point
whether the interlayer expansion, as discussed before, also plays
a role in controlling the hole mobility. These unexceptional
mobility values, although comparable to the literature,
prompted us to examine the best derivative in a different
architecture.
Given the established fact that single-crystal mobility is
strongly dependent on the dielectric constant of the gate
insulator,56 we posited that the hole mobility of compound 5
will increase in a vacuum-gap architecture, wherein the intrinsic
mobility of the crystal can be potentially measured (the low
reproducibility of electron transport with vacuum-gap precludes
a systematic study of dielectric material dependence). We
To understand the resulting transport properties of these
derivatives, we now correlate the measurements to the
previously discussed electronic properties. Compounds 5−7
have improved injection efficiencies for electrons from the high
workfunction Au contact over 1, resulting in appreciable
electron mobilities and improved threshold voltages. This can
H
dx.doi.org/10.1021/cm400736s | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
addressed this problem by making both bottom-contact and
top-contact vacuum-gap single-crystal FETs (bottom-gate)
based on rubrene 1 and 5 (Supporting Information, section
J). Rubrene 1 exhibited ideal transistor behaviors with
negligible hysteresis and large hole mobility, regardless of the
contact geometry (15.9 cm2 V−1 s−1 for bottom-contact device
and 9.56 cm2 V−1 s−1 for top-contact device). However,
vacuum-gap device performance of rubrene 5 is far from
satisfactory. In the bottom-contact structure, devices required
application of large VD and VG to carry charge, resulting in low
hole mobility ∼0.03 cm2 V−1 s−1 and low device reproducibility.
Top-contact devices performed substantially better; yet the
highest mobility was only 0.66 cm2 V−1 s−1. These unsuccessful
efforts in achieving higher hole mobility of rubrene 5 in
vacuum-gap architecture seem to suggest other factors besides
the electronic coupling need to be taken into account when
single-crystal mobility is measured.
One such important factor is the charge injection efficiency
from Au to rubrenes 5−7 as compared to rubrene 1. While the
substituents on the arylenes lead to packing motifs that increase
the electronic coupling for holes and/or electrons, they also
influence the ionization potential and electron affinity through
inductive effects that in turn impact hole/electron injection
from the same metal contact.59 As we previously mentioned,
the stabilization of both HOMO and LUMO (while
maintaining the bandgap) in rubrenes 5−7 is advantageous
for electron transport but increases the hole injection barrier
when symmetric Au contacts are used. The presence of a large
hole injection barrier and large contact resistance from Au may
ultimately limit the hole transport in 5−7 such that the hole
mobility measured in the current device structure as well as in
vacuum-gap structure is significantly lower than the intrinsic
values. SC-OFETs with more suitable electrodes and/or
contact modifications may provide a better route to
investigating the intrinsic hole and electron-transport mobilities
in these new derivatives. Detailed studies on contact resistance
and device architecture optimization will be published else-
where. Despite the encountered challenges, we have been able
to demonstrate a series of rubrenes in which the fine-tuning of
molecular and crystal structure was shown to have a large
impact on the electronic structure and the charge-transport
properties.
solid state. Such efforts, if successful, could help create a clear
path toward enhanced organic semiconductors for use in FETs.
Our work will continue to blend synthesis, theoretical
modeling, and device studies to explore how molecular design
can be used to tune the intrinsic molecular properties. Future
investigations will closely examine steric and electrostatic effects
that impact molecular packing configurations, which in turn
influence the macroscopic material properties and device design
(e.g., choice of electrodes and device architecture). Such studies
will lead to the creation of improved small-molecule acenes for
use in OFETs.
ASSOCIATED CONTENT
* Supporting Information
■
S
Detailed synthetic experimental procedures; full character-
ization data of all rubrene derivatives; NMR spectra of all
rubrene derivatives and intermediates; crystal parameters for
2−7; ORTEP drawings for 2−7; HOMO−LUMO energies for
2−7; potential traces for 2−7; full band structures from
periodic DFT calculations, including effective masses; transfer
integrals, molecular orbital distributions, and eigenvalues for 1,
5, 6, and 7; theoretical and crystallographic geometric
structures; IPs, EAs, and intramolecular reorganization energies
for 1−7; AFM images and FET performance of 6 and 7; and
details on comparison of 1 and 5 device performance in
vacuum-gap structure along with a compiled CIF of all rubrene
derivatives. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: jean-luc.bredas@chemistry.gatech.edu (J.-L.B.);
frisbie@umn.edu (C.D.F.); cdouglas@umn.edu (C.J.D.).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We would like to thank Gregory Rhode for X-ray assistance,
Sean Murray for MS analysis, and Alexandr Fonari for
providing the code to determine the effective masses. Parts of
this work were carried out in the Characterization Facility,
University of Minnesota, which receives partial support from
NSF through the MRSEC program. The work at Minnesota
and Georgia Tech was primarily supported by the National
Science Foundation under the NSF MRSEC Program, Award
Numbers DMR-0819885 and DMR-1006566, as well as 3M
and DuPont.
CONCLUSION
■
We have discovered a synthetic route to effect late-stage
manipulation of the molecular structure of rubrene so that the
solid-state packing arrangements maintain the beneficial wave
function overlap of the conjugated backbones (π-stacking)
while at the same time result in an increase in the interlayer
distance. Our strategy using alkyl and fluoroalkyl substituents
provides rubrene derivatives with systematically controlled
crystal packing structures. A fine balance between steric and
electrostatic effects allows us to manipulate not only the
molecular packing configurations, but also the molecular
conformation in the solid state (i.e., tetracene backbone
planarity), a route that potentially could be applied to other
small-molecule acenes for OFET studies. The π-stacking
present in 5−7 leads to enhanced electronic couplings versus
the parent rubrene 1. These enhanced electronic couplings
manifest themselves in SC-OFETs, although the mobility values
have yet to reach the best reported for rubrene. The work
herein begins to scratch the surface on the use of molecular
design to influence the electronic properties of the crystalline
REFERENCES
■
(1) Podzorov, V.; Menard, E.; Borissov, A.; Kiryukhin, V.; Rogers, J.
A.; Gershenson, M. E. Phys. Rev. Lett. 2004, 93, 086602.
(2) Machida, S.; Nakayma, Y.; Duhm, S.; Xin, Q.; Funakoshi, A.;
Ogawa, N.; Kera, S.; Ueno, N.; Ishii, H. Phys. Rev. Lett. 2010, 104,
156401.
(3) Podzorov, V.; Menard, E.; Rogers, J. A.; Gershenson, M. E. Phys.
Rev. Lett. 2005, 95, 226601.
(4) Li, Y.; Coropceanu, V.; Bred
3325−3329.
́
as, J.-L. J. Phys. Chem. Lett. 2012, 3,
(5) Paraskar, A. S.; Reddy, A. R.; Patra, A.; Wijsboom, Y. H.; Gidron,
O.; Shimon, L. J. W.; Leitus, G.; Bendikov, M. Chem.−Eur. J. 2008, 14,
10639−10647.
(6) Schuck, G.; Haas, S.; Stassen, A. F.; Berens, U.; Batlogg, B. Acta
Crystallogr. 2007, E63, o2894.
I
dx.doi.org/10.1021/cm400736s | Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
(7) Schuck, G.; Haas, S.; Stassen, A. F.; Kirner, H.-J.; Batlogg, B. Acta
Crystallogr. 2007, E63, o2893.
(8) Haas, S.; Stassen, A. F.; Schuck, G.; Pernstich, K. P.; Gundlach, D.
J.; Batlogg, B.; Berens, U.; Kirner, H.-J. Phys. Rev. B 2007, 76, 115203−
6.
(9) Bergantin, S.; Moret, M. Cryst. Growth Des. 2012, 12, 6035−6041.
(10) Reese, C.; Bao, Z. Mater. Today 2007, 10, 20−27.
(11) Braga, D.; Horowitz, G. Adv. Mater. 2009, 21, 1473−1486.
(12) Gershenson, M. E.; Podzorov, V.; Morpurgo, A. F. Rev. Mod.
Phys. 2006, 78, 973−989.
(36) Yagodkin, E.; Xia, Y.; Kalihari, V.; Frisbie, C. D.; Douglas, C. J. J.
Phys. Chem. C 2009, 113, 16544−16548.
(37) Williams, J. H. Acc. Chem. Res. 1993, 26, 593−598.
(38) Okamoto, T.; Nakahara, K.; Saeki, A.; Seki, S.; Oh, J. H.;
Akkerman, H. B.; Bao, Z.; Matsuo, Y. Chem. Mater. 2011, 23, 1646−
1649.
(39) Marcus, R. A. J. Chem. Phys. 1956, 24, 966−978.
(40) Barbara, P. F.; Meyer, T. J.; Ratner, M. A. J. Phys. Chem. 1996,
100, 13148−13168.
(41) Cozzi, F.; Ponzini, F.; Annunziata, R.; Cinquini, M.; Siegel, J. S.
Angew. Chem., Int. Ed. Engl. 1995, 34, 1019−1020.
(42) Cozzi, F.; Siegel, J. S. Pure Appl. Chem. 1995, 67, 683−689.
(43) Hunter, C. A. Angew. Chem., Int. Ed. Engl. 1993, 32, 1584−1586.
(44) Cozzi, F.; Cinquini, M.; Annuziata, R.; Siegel, J. S. J. Am. Chem.
Soc. 1993, 115, 5330−5331.
(13) Boer, D.; I, R. W.; Gershenson, M. E.; Morpurgo, A. F.;
Podzorov, V. Phys. Status Solidi A 2004, 201, 1302−1331.
(14) Mas-Torrent, M.; Rovira, C. Chem. Rev. 2011, 111, 4833−4856.
(15) Filho, D. A.; da, S.; Kim, E.-G.; Bred
17, 1072−1076.
(16) Bredas, J. L.; Calbert, J. P.; Filho, D. S.; A, D.; Cornil, J. Proc.
́
as, J.-L. Adv. Mater. 2005,
(45) Cozzi, F.; Cinquini, M.; Annunziata, R.; Dwyer, T.; Siegel, J. S. J.
Am. Chem. Soc. 1992, 114, 5729−5733.
́
Natl. Acad. Sci. U.S.A. 2002, 99, 5804−5809.
(46) Desiraju, G. R.; Parthasarathy, R. J. Am. Chem. Soc. 1989, 111,
8725−8726.
(17) Anthony, J. E.; Eaton, D. L.; Parkin, S. R. Org. Lett. 2002, 4, 15−
18.
(47) Rowland, R. S.; Taylor, R. J. Phys. Chem. 1996, 100, 7384−7391.
(48) Bondi, A. J. Phys. Chem. 1964, 68, 441−451.
(18) Anthony, J. E.; Brooks, J. S.; Eaton, D. L.; Parkin, S. R. J. Am.
Chem. Soc. 2001, 123, 9482−9483.
(49) Thalladi, V. R.; Weiss, H.-C.; Blaser, D.; Boese, R.; Nangia, A.;
̈
(19) Miao, Q.; Chi, X.; Xiao, S.; Zeis, R.; Lefenfeld, M.; Siegrist, T.;
Steigerwald, M. L.; Nuckolls, C. J. Am. Chem. Soc. 2006, 128, 1340−
1345.
Desiraju, G. R. J. Am. Chem. Soc. 1998, 120, 8702−8710.
(50) Nardello, V.; Marti, M.-J.; Pierlot, C.; Aubry, J.-M. J. Chem. Educ.
1999, 76, 1285.
(20) Curtis, M. D.; Cao, J.; Kampf, J. W. J. Am. Chem. Soc. 2004, 126,
(51) Uttiya, S.; Raimondo, L.; Campione, M.; Miozzo, L.; Yassar, A.;
Moret, M.; Fumagalli, E.; Borghesi, A.; Sassella, A. Synth. Met. 2012,
161, 2603−2606.
4318−4328.
(21) Sundar, V. C.; Zaumseil, J.; Podzorov, V.; Menard, E.; Willett, R.
L.; Someya, T.; Gershenson, M. E.; Rogers, J. A. Science 2004, 303,
1644−1646.
(52) Valeev, E. F.; Coropceanu, V.; Da Silva Filho, D. A.; Salman, S.;
Bred
(53) Coropceanu, V.; Cornil, J.; Da Silva Filho, D. A.; Olivier, Y.;
Silbey, R.; Bredas, J.-L. Chem. Rev. 2007, 107, 926−952.
́
as, J.-L. J. Am. Chem. Soc. 2006, 128, 9882−9886.
(22) Minder, N. A.; Ono, S.; Chen, Z.; Facchetti, A.; Morpurgo, A. F.
Adv. Mater. 2012, 24, 503−508.
́
(23) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(24) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
(25) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623−11627.
(54) Pope, M. Electronic Processes in Organic Crystals and Polymers,
2nd ed.; Oxford University Press: New York, 1999.
(55) Laudise, R.; Kloc, C.; Simpkins, P.; Siegrist, T. J. Cryst. Growth
1998, 187, 449−454.
(26) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56,
2257−2261.
(56) Hulea, I. N.; Fratini, S.; Xie, H.; Mulder, C. L.; Iossad, N. N.;
Rastelli, G.; Ciuchi, S.; Morpurgo, A. F. Nat. Mater. 2006, 5, 982−986.
(57) Takahashi, T.; Takenobu, T.; Takeya, J.; Iwasa, Y. Appl. Phys.
Lett. 2006, 88, 033505.
(58) Takenobu, T.; Takahashi, T.; Takeya, J.; Iwasa, Y. Appl. Phys.
Lett. 2007, 90, 013507.
(59) Zaumseil, J.; Sirringhaus, H. Chem. Rev. 2007, 107, 1296−1323.
(60) Najafov, H.; Mastrogiovanni, D.; Garfunkel, E.; Feldman, L. C.;
Podzorov, V. Adv. Mater. 2011, 23, 981−985.
(61) Mitrofanov, O.; Lang, D. V.; Kloc, C.; Wikberg, J. M.; Siegrist,
T.; So, W.-Y.; Sergent, M. A.; Ramirez, A. P. Phys. Rev. Lett. 2006, 97,
166601.
(62) Tang, Q.; Tong, Y.; Li, H.; Ji, Z.; Li, L.; Hu, W.; Liu, Y.; Zhu, D.
Adv. Mater. 2008, 20, 1511−1515.
(63) Bisri, S. Z.; Takenobu, T.; Yomogida, Y.; Shimotani, H.; Yamao,
T.; Hotta, S.; Iwasa, Y. Adv. Funct. Mater. 2009, 19, 1728−1735.
(64) Bao, Z.; Locklin, J. J. Organic Field-Effect Transistors; CRC Press:
Boca Raton, FL, 2007.
(27) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Guassian 09,
revision B.01; Gaussian, Inc.: Wallingford, CT, 2009.
(28) Dovesi, R.; Orlando, R.; Civalleri, B.; Roetti, C.; Saunders, V. R.;
Zicovich-Wilson, C. M. Z. Kristallogr. 2005, 220, 571−573.
(29) Moureu, C.; Dufraisse, C.; Dean, P. M. C. R. Hebd. Seances Acad.
Sci. 1926, 182, 1440−1443.
(30) Wittig, G.; Waldi, D. J. Prakt. Chem. 1942, 160, 242−244.
(31) Braga, D.; Jaafari, A.; Miozzo, L.; Moret, M.; Rizzato, S.;
Papagni, A.; Yassar, A. Eur. J. Org. Chem. 2011, 2011, 4160−4169.
(32) Badger, G. M.; Pearce, R. S.; Rodda, H. J.; Walker, I. S. J. Chem.
Soc. 1954, 3151.
(33) Dodge, J.; Bain, J. D.; Chamberlin, A. R. J. Org. Chem. 1990, 55,
4190−4198.
(34) Balodis, K. A.; Medne, R. S.; Neiland, O. Y.; Kozlova, L. M.;
Klyavinya, Z. P.; Mazheika, I. B.; Gaukhman, A. P. Zh. Org. Khim.
1985, 21, 2423−2427.
(35) Yagodkin, E.; McGarry, K. A.; Douglas, C. J. Org. Prep. Proced.
Int. 2011, 43, 360−363.
J
dx.doi.org/10.1021/cm400736s | Chem. Mater. XXXX, XXX, XXX−XXX