Thiophene-diazine molecular semiconductors: Synthesis, structural, electrochemical, optical, and electronic structural properties; Implementation in organic field-effect transistors

Article abstract of DOI:10.1002/chem.200802424

The synthesis, structural, electrochemical, optical, and electronic structure properties of a new azinethiophene semiconductor family are reported and compared to those of analogous oligothiophenes. The new molecules are: 5,5′-bis(6-(thien-2-yl)pyrimid-4-

Full text of DOI:10.1002/chem.200802424

FULL PAPER
DOI: 10.1002/chem.200802424
Thiophene–Diazine Molecular Semiconductors: Synthesis, Structural,
Electrochemical, Optical, and Electronic Structural Properties;
Implementation in Organic Field-Effect Transistors
[
b]
[a]
[a]
[a]
Rocꢀo Ponce Ortiz, Juan Casado, Vꢀctor Hernꢁndez, Juan T. Lꢂpez Navarrete,*
[b]
[
b]
[b]
[b]
Joseph A. Letizia, Mark A. Ratner, Antonio Facchetti,* and Tobin J. Marks*
Abstract: The synthesis, structural,
electrochemical, optical, and electronic
structure properties of a new azine-
thiophene semiconductor family are re-
ported and compared to those of anal-
ogous oligothiophenes. The new mole-
cules are: 5,5’-bis(6-(thien-2-yl)pyri-
mid-4-yl)-2,2’-dithiophene (1), 5,5’-
bis(6-(5-hexylthien-2-yl)pyrimid-4-yl)-
enhances electron affinity. Thin-film
transistors were fabricated with these
materials and evaluated both in
vacuum and in air. We find that al-
though diazine substitution is impor-
tant in tuning oligothiophene orbital
energetics, these oligomers are p-chan-
nel semiconductors and the field-effect
transistor (FET) charge transport prop-
erties are remarkably similar to these
of unsubstituted oligothiophenes. The
combined computational-experimental
analysis of the molecular and thin film
properties indicates that these diazine-
containing oligothiophenes essentially
behave as p-extended bithiophenes. In-
terestingly, despite strong intermolecu-
lar interactions, high solid-state fluores-
cence efficiencies are observed for
these new derivatives. Such emission
characteristics suggest that these mate-
rials behave as more extended p sys-
tems, which should be advantageous in
light-emitting transistors.
2,2’-dithiophene (3), and 5,5’-bis(6-
(
thien-2-yl)pyridazin-3-yl))-2,2’-dithio-
Keywords: density functional calcu-
lations · electrochemistry · p-type
phene (2). Electrochemical experi-
ments demonstrate that introduction of
electron-poor heteroaromatic rings into
the oligothiophene core significantly
mobility
·
semiconductors
·
transistors
Introduction
chemistry community reflects their pronounced chemical/
electrochemical stability, the preparative accessibility of
many thiophene synthons, and the availability of well-devel-
Over the past few years, thiophene-based materials have
emerged as an important class of semiconductors, embracing
chemical structures ranging from small molecules to high
[2]
oped/regioselective ring-ring coupling methodologies. Fur-
thermore, the properties of oligo/polythiophene cores can
be efficiently tuned by substitution at the terminal core posi-
tions as well as by replacing and/or “mixing” some fraction
of the thiophene rings with other (hetero)arenes. As a result
of these molecularly engineered modifications, a wide varie-
ty of electronic and opto-electronic devices having thio-
[1]
molecular weight polymers. The intense research focus
these systems have received from the organic materials
[
a] Prof. J. Casado, Prof. V. Hernꢀndez, Prof. J. T. L o´ pez Navarrete
Department of Physical Chemistry, University of Mꢀlaga
Campus de Teatinos s/n, Mꢀlaga 29071 (Spain)
Fax : (+34)952-132-000
[3]
phene as the key structural unit have been fabricated. Be-
sides the well-developed field of electrically-conducting
[4]
solids, these materials are currently under active investiga-
tion for applications in organic field-effect transistors
E-mail: teodomiro@uma.es
[
b] Dr. R. P. Ortiz, Dr. J. A. Letizia, M. A. Ratner, Prof. A. Facchetti,
Prof. T. J. Marks
[
5]
[6]
[7]
[8]
(OFETs), light-emitting diodes, lasers, sensors, and
[9]
Department of Chemistry, Northwestern University
145 Sheridan Road, Evanston, Illinois 60208-3113 (USA)
photovoltaic cells, as well as for other important technolo-
2
[10,11,12]
gy fields.
For example, an area in which thiophene
Fax : (+1)847-491-2990
core substitution has been profitably exploited is in electro-
luminescent devices, such as lasers and light-emitting transis-
E-mail: a-facchetti@northwestern.edu
t-marks@northwestern.edu
[13]
tors, which require electroactive materials also having ef-
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200802424.
[14]
ficient fluorescent properties.
However, unsubstituted/
Chem. Eur. J. 2009, 15, 5023 – 5039
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5023

[
29]
[30]
alkyl-substituted oligothiophenes generally only exhibit high
tion processes,
in liquid crystal design,
as well as in
[15]
[31]
fluorescence quantum yields in solution.
This limitation
non-linear optical materials. However, there are very few
examples of azine materials in the organic electronics area
and even fewer based on oligomeric structures. For example,
Katz first reported on the FET properties of thiophene-thia-
zole derivatives exhibiting p-channel activity and very large
has been successfully addressed by core-substitution with
[16]
fluorescence-enhancing groups such as phenylene, biphe-
[
17]
[18]
[19]
nylene,
and bis(9,9-dimethylfluorenyl)aminophenylene.
Organic FETs represent another important area in which
fluorene,
bis(methylphenyl)aminophenylene,
[20]
[
32]
I :I ratios, whereas more recently, Jenekhe and Yama-
on off
[5]
oligo/polythiophenes have been successful, starting from
the pioneering work of Hotta, Garnier, and Katz on oligo-
moto reported promising thienopyrazine-, quinoxaline-, and
thienodithiazole-thiophene co-polymers, again exhibiting p-
channel transport. These interesting results raise the ques-
tion of how key molecular/charge transport properties might
respond to introducing even more strongly electron-accept-
ing heterocycles into such structures. Here we report the
synthesis of a new series of mixed diazine-thiophene oligo-
mers 1–3 (Figure 1), in which the diazine is a pyridazine or
[21]
[33]
thiophenes,
and Tsumara and Assadi on polythio-
[
22]
phenes. Since these initial studies demonstrating p-chan-
nel FET operation, much effort has focused on tuning key
oligo/polythiophene FET properties to optimize carrier mo-
bility, air-stability, and to enable electron- (n-channel) and
[23]
ambipolar transport. These efforts have included the syn-
thesis of variously functional-
ized oligo/polythiophene deriv-
atives, optimization of film
growth processes by means of
specific solvents, and fabrica-
tion of device structures and/or
pairing with semiconductor
properties-enhancing gate di-
[24]
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
electrics.
From a chemical
perspective, molecular structure
modifications offer the promise
of new properties and functions.
For example, n-channel semi-
conductors for OFETs are rare,
and even rarer are those con-
taining ostensibly electron-rich
thiophenes.
Although
the
design rules for n-channel or- Figure 1. Chemical structures of the diazine-thiophene and thiophene oligomers examined in this study.
ganic semiconductors are in
some ways similar to those for
the corresponding p-channel materials, with the exception
that efficient injection of electrons into the LUMO must
occur, the actual realization of new, stable n-channel materi-
als remains a daunting challenge. For example, n-channel
oligothiophenes were realized by a number of us, beginning
pyrimidine unit, and compare and contrast their properties
with those of the unsubstituted oligothiophene cores 4–6.
These nitrogen-containing heterocycles were selected be-
cause of their more positive reduction potentials and greater
electron-withdrawing capacities than other monocyclic pyri-
[5b,c]
[34]
in 2000, by introducing fluorocarbon substituents.
How-
dines and thiazoles,
the ready access to key synthetic
ever, from an electronic structure/redox properties view-
point, n-channel fluorocarbon-substituted oligothiophenes,
especially those having large p-electron cores, exhibit rela-
tively modest properties variations compared to the corre-
sponding p-channel alkyl-substituted parent systems. There-
fore, the origin of the remarkable FET majority carrier sign
inversion upon fluorocarbon substituent introduction and
whether this inversion principally reflects enhanced core
electron affinity, and whether other important factors are
operative, remains unresolved.
building blocks, and the current general research goals in or-
ganic electronics of: 1) Discovering new high-mobility or-
ganic semiconductors, optimally solution-processable; 2) en-
abling new n-channel materials and functions; 3) stabilizing
n- and p-channel operation in ambient.
We show here that the new azine-based targets can be ef-
ficiently synthesized in good yields and characterize them by
a combination of thermal (thermogravimetric analysis and
differential scanning calorimetry), optical spectroscopic
(UV/Vis, photoluminescence, and Raman), and electro-
chemical (cyclic voltammetry) techniques. DFT computa-
tions provide valuable insights into key molecular electronic
structure and thin film properties. In addition, vapor-depos-
ited films of these new molecules are investigated by X-ray
diffraction (XRD) and scanning electron microscopy
(SEM). Finally, thin-film transistors are fabricated and eval-
Nitrogen-containing, electron-poor heterocycles have pre-
viously been incorporated in, and combined with, thio-
phene-based conjugated structures to achieve donor-accept-
[
25,26,27]
or copolymers with reduced band-gaps.
azines have attracted recent attention, owing to their biolog-
In this regard,
[28]
ical properties, their potential applicability in bond-forma-
5024
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 5023 – 5039

Organic Field-Effect Transistors
FULL PAPER
uated both in vacuum and in air. Our results demonstrate
that oligothiophene molecular electronic structure and FET
majority charge transport properties respond in very differ-
ent, and not necessarily intuitive, ways to diazine skeletal in-
troduction.
pounds 1 and 2 were purified by gradient vacuum sublima-
tion, whereas more soluble 3 was purified by recrystallization.
Molecular structure-property relationships
Thermal characterization: Differential scanning calorimetry
(
DSC) was performed on all the new thiophene-diazine ma-
Results and Discussion
terials and reveals that the entire set is thermally stable,
with DSC plots showing no evidence of mesophase forma-
tion before melting (Figure S1). This is surprising consider-
ing the rod-like molecular structures and the established
presence of multiple LC transitions in thiophene homo-olig-
omers and phenylene-thiophene
Synthetic strategies: The new azine-thiophene oligomers 1,
, and 3 were synthesized according to Scheme 1. The start-
ing materials for oligomer realization are readily available,
2
[35]
co-oligomers.
Thermogravi-
metric analysis (TGA) reveals
quantitative sublimation in all
cases (Figure S2), demonstrat-
ing the thermal robustness of
the 2 and 1 cores. Note that this
result stands in contrast to
some alkyl-substituted and
longer unsubstituted oligothio-
[36]
phenes, and despite the very
high melting points (see the Ex-
perimental Section), oligomers
1
–3 sublime quantitatively with-
out significant decomposition.
Intra- and inter-molecular non-
bonded contacts, discussed in
next sections, may also contrib-
ute to this effect.
Electrochemistry: The electro-
chemistry of oligo- and poly-
thiophenes reveals important
aspects of chemical/electronic
structure, charge injection and
storage mechanisms, substituent
Scheme 1. Synthetic Scheme for the new azine-thiophene oligomers.
[36,37,38]
and the general synthetic strategy entails a,w-distannylation
of thiophene and dithiophene, followed by Pd-catalyzed
Stille coupling with the appropriate azine building blocks.
Key pyrimidine intermediate 7 was synthesized in ꢀ70%
effects, and other physical characteristics.
Such studies
have aided thiophene-based conductor development, quanti-
tatively addressing substituent oligomer dimension effects.
Cyclic voltammetry (CV) for oligomers 1–6 was performed
yield by [Pd
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(PPh ) ]-catalyzed cross-coupling of 4,6-dichloro-
under N2 in 0.1 m THF/TBAPF solutions with scanning
3
4
6
ꢁ
1
pyrimidine with 5,5’-bis(tri-n-butylstannyl)-2,2’-dithiophene
in toluene. The final two pyrimidine-thiophene oligomers, 1
and 3, were accessed by means of the coupling of core 7
with 2-(tri-n-butylstannyl)-thiophene and 2-(tri-n-butylstann-
yl)-5-hexylthiophene, respectively. Compounds 1 and 3 were
obtained in ꢀ50 and 70% yields, respectively. Note that be-
cause of the relatively low yields, we attempted to synthe-
rates between 60–150 mVs. Most systems exhibit one or
two reversible and/or quasi-reversible one-electron oxida-
tion and reduction waves within the solvent/electrolyte
window range. Figure 2 shows representative voltammo-
grams whereas the electrochemical data are summarized in
Table 1 below. When the voltammograms are (quasi)reversi-
1
=
2
ble, it is possible to extract formal potentials (E ), as the
midpoints between peak potentials for the forward and re-
verse scans. Unsubstituted oligothiophenes, with the excep-
tion of 6, exhibit two reversible reduction potentials, where-
size 1 through CuCl -promoted dimerization of the lithium
2
salt of 9, however this was unsuccessful.
Pyridazine-thiophene building block 8 was obtained in
[38 39,40]
ꢀ
50% yield by coupling of 3,6-dichloropyridazine with 2-
as reversible oxidations are observed only for 4. ,
Two
(
tri-n-butylstannyl)-thiophene. Compound was subse-
2
single-electron reductions (versus SCE) are observed at
ꢁ1.45/ꢁ1.57 V for 1, at ꢁ1.45/ꢁ1.56 V for 3, and at ꢁ1.48/
ꢁ1.64 V for 2 versus ꢁ1.71/ꢁ1.86 V for 4. Irreversible oxida-
quently synthesized by coupling compound 8 with 5,5’-
bis(tri-n-butylstannyl)-2,2’-dithiophene (ꢀ86% yield). Com-
Chem. Eur. J. 2009, 15, 5023 – 5039
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5025

J. T. Lꢂpez Navarrete, A. Facchetti, T. J. Marks et al.
ꢁACHTUNGTRENNUG
[42]
al negative p charge)
of the former heterocycle. In
marked contrast, for alkyl-/fluorocarbon-substituted oligo-
thiophenes, the electrochemical potential difference be-
1
1
1
=
2
=2
=2
tween subsequent reduction events [DE =jE ꢁE j] does
2
1
not track the increased stabilization of the negative charge.
For equivalent p-delocalized systems, the compression of
1
=
2
DE indicates a reduction of Coulombic repulsion between
excess negative charges. Hence, the value in 4 (0.15 V) is
similar to that in 2 (0.16 V), likely reflecting the identical
para-connection of cores having the same number of p elec-
1
=
2
trons. The DE value is similar in 1 (0.12 V), however, the
1
=
2
DE differences among these three molecules are too small
to clearly ascribe to a particular electronic or structural
characteristic. Moreover, when doubly charged species are
ꢁ
1
Figure 2. Cyclic voltammograms (first scan, v=80 mVs ) in THF on a
0
2
.8 mm diameter glassy carbon electrode for thiophene-diazines 1 (top),
(middle), and 3 (bottom).
[43]
involved, Coulombic repulsions can play a significant role.
Regarding anodic processes, the
same enhanced electron
demand argument or inclusion
of more electronegative
1
/2
a c
Table 1. Anodic (E ), cathodic (E ), and half (E ) potentials [V] versus SCE of compounds 1–6 in dry THF
[
a]
under nitrogen.
N
Compound
Oxidation
Cathodic
Reduction
Anodic
atoms can be invoked to ex-
plain the oxidative events oc-
curring at more positive poten-
Anodic
Half
Cathodic
Half
1
=2
1
1
1
=
2
=2
=2
E
a1
E
a2
E
c1
E
c2
E1
E2
E
c1
E
c2
E
a1
E
a2
E1
E2
1
2
3
4
5
6
1.63
1.87
1.60
ꢁ1.51 ꢁ1.63 ꢁ1.39 ꢁ1.51 ꢁ1.45 ꢁ1.57
ꢁ1.52 ꢁ1.66 ꢁ1.44 ꢁ1.61 ꢁ1.48 ꢁ1.64 tials.
ꢁ1.50 ꢁ1.60 ꢁ1.40 ꢁ1.52 ꢁ1.45 ꢁ1.56
ꢁ1.76 ꢁ1.91 ꢁ1.66 ꢁ1.79 ꢁ1.71 ꢁ1.86
1.02 1.17 0.93 1.08 0.98 1.13
1.40
Optical Properties: Optical ab-
0.81 1.00 1.10
0.77 1.12
ꢁ2.02 ꢁ2.33 ꢁ1.86 ꢁ2.20 ꢁ1.94 ꢁ2.07
sorption and fluorescence emis-
sion spectra of compounds 1–6
were measured in both solution
and as powders/vapor-deposited
thin films (Figure 3) to assess
ꢁ2.60
ꢁ2.24
ꢁ2.42
+
[
a] Referenced to the Fc/Fc couple in THF (0.50 V vs. Ag/AgCl; 0.54 V vs. SCE).
tive features are observed at ꢀ+1.6 V for both pyrimidine-
based systems and at ꢀ+1.87 V for 2, versus reversible oxi-
dations at +0.98/+1.13 V for 4. This solution phase redox
behavior contrasts with the results of thin-film transport
measurements for which it will be seen that electrons are far
less mobile than holes (vide infra), confirming the observa-
tion that OFET charge transport is often dominated by
charge trapping at the semiconductor-dielectric interface
rather than by intrinsic molecular redox properties. There
are, however, other factors that likely influence transport
properties from the perspective of molecular charging such
as internal molecular reorganization energies for cation or
anion generation, that will be discussed in the closing sec-
tion in connection with OFET carrier mobility.
the effect of azine substitution on oligothiophene absorp-
tion/emission maxima (labs/l ) and the (optical) HOMO–
em
LUMO energy gap. Table 2 collects UV/Vis-PL data for all
compounds in THF solution. The solution absorption spec-
tra of 1–3 exhibit incipient vibronic fingerprints, clearly dis-
tinguishable at lower temperatures (see Figure S3 in the
Supporting Information), whereas in the fluorescence spec-
tra, typical subpeaks are observed even at room tempera-
ture (Figure 3). This finding likely reflects full core planari-
[41]
zation upon electron excitation and relaxation to an S qui-
1
Analysis of the thiophene-diazine half-wave potentials re-
veals interesting trends. Compared to p-isoelectronic 4
having the same number of p electrons with the N and S
1
1
=
2
=2
atoms included, the reduction potentials E1 and E2 of 1–3
1
=
2
are shifted towards substantially more positive values [DE
1
1
=
2
(
4!1–3)=+(0.23-0.26) V, DE (4!1–3)=+ (0.22–0.30) V]
2
in agreement with the very electron-deficient nature of
azines compared to thiophenes. Furthermore, within the dia-
zine-thiophene family, the reduction potentials are shifted to
more negative values on proceeding from pyrimidine to pyr-
Figure 3. Optical absorption (solid lines) and emission (dashed lines)
spectra of compounds 1–6 in A) THF and B) the solid state.
1
1
=
2
=2
azine systems [DE1 (1, 3!2)=ꢁ
A
H
U
G
R
N
U
G
2
5026
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 5023 – 5039

Organic Field-Effect Transistors
FULL PAPER
Table 2. Optical absorption (maximum of absorption, lmax and extinction coefficient at labs, e) and emission corresponding lmax values. The
[
a]
(
maximum of fluorescence, lem and quantum yield, F
f
) data for compounds 1–6 in dry THF and in the solid
shapes of the fluorescence exci-
tation spectra track those of the
absorption spectra. Photolumi-
nescence quantum yields (Ff)
were determined using a qui-
state (50 nm films).
Compound
Solution
Film
D[b]
F
[c]
l
l
D[b]
Eg
op
l
max
e
l
f
f
max
f
ꢁ
1
ꢁ1
[a]
[a]
[a]
[nm]
A
H
U
G
R
N
U
G
]
[nm]
[eV]
[nm]
[nm]
[eV]
[eV]
1
2
3
4
5
6
408
420
409
436
391
304
72700
40100
65500
50000
35000
12800
455
465
455
505*, 539
450*, 478
363
0.31
0.29
0.31
0.39
0.42
0.66
0.38
0.24
0.42
0.46
0.20
0.01
350
344
356
376
336
–
610
590
605
589
520
–
–
–
–
1.19
1.31
–
2.4
2.2
2.7
2.3
2.7
–
[45]
nine sulfate standard, and PL
data are also collected in
[
a]
Table 2. Intrinsically, F is de-
f
termined by the relative rates
of nonradiative and radiative
deactivation. For the present
[
a] Absolute maximum indicated by asterisk, [b] D=lemꢁlabs, [c] Measured using quinine sulfate as the stan-
dard.
series, Ff values in 1–6 are
found to increase quasi-monot-
onically with the number of
noidal emissive state. As will be discussed below, conforma-
rings. This behavior is reminiscent of that found in the a-4–6
tional freedom (rotation about inter-ring CꢁC single bonds)
series, in which F increases in molecules having up to 6
f
in the solution ground electronic state of the azine-contain-
ing molecules is more restricted than in p-isoelectronic 4 in
which the solution ground state is, on average, more twisted.
The splitting periodicity of the vibronic peaks in absorption
rings and then remains constant or even decreases for
[46]
longer oligomers. Fluorescence quantum yields are found
to increase on proceeding from 6 to 4 and then to decrease
in the thiophene-azine systems. The azine-centered HOMO–
LUMO optical transition, with the HOMO localized princi-
pally on the central bithienyl unit, occurs at shorter wave-
lengths than that of 4. This corresponds to an enlargement
of the optical gap or decreased p-conjugation relative to 4.
ꢁ
1
,
is estimated to be ꢀ1400–1500 cm and can be assigned to
collective, totally symmetric oligothiophene core n AHCUTNGRNENUG( C=C)
stretching modes, vibronically coupled to the electronic exci-
tation and hence resonance-enhanced in the Raman spectra
(
see below). As will be discussed below, this result estab-
Consequently, a slight reduction in F is not unexpected on
f
lishes a connection between molecular electronic and vibra-
tional properties.
The large molar absorption coefficients (e, Table 2) indi-
cate dominance in the optical spectra of allowed, conjugated
going from 4 to the azine derivatives. This property is there-
fore midway between those of 6 and 4. Note that owing to
the dominant role of the large sulfur orbital angular mo-
mentum in 6, efficient intersystem crossing strongly decreas-
[47]
core p!p* transitions. As expected, l
values increase
es the fluorescence yield of this short oligomer.
max
across the series for increasing numbers of ring units (6!
The changes in thiophene-diazine conformations to more
planar structures and energy dissipation within the excited
state lifetime is manifested in marked Stokes shifts, the
energy differences between the 0–0 transitions in absorption
5
!1–4). Within the p-isoelectronic systems 1–4, l increas-
abs
es on proceeding from 1 (408 nm) to 2 (420 nm) to 4
(
436 nm). Although the shorter wavelength absorption of 1
[48]
might be explained by reduced p-conjugation, owing to the
azine-thiophene meta linkage, the result for 2 is surprising
since mixing electron-rich and electron-poor rings is a
proven strategy for achieving small-bandgap chromo-
and emission. Since 0–0 transitions are rarely observed in
room temperature solution spectra, it is accepted procedure
to use D=l ꢁl to index the magnitude of the Stokes
em
abs
shift. The observed general trend here is that D decreases
with increasing numbers of heterocyclic rings, and further-
more, when thiophene is replaced by an azine moiety. The
amplitude of this shift generally tracks the lifetime and
[44]
phores.
HOMO!LUMO excitation rationalizes these observations
vide infra). In contrast to 4, the HOMO of the isoelectronic
Analysis of the orbital composition of the
(
[48]
nitrogen-containing molecules is localized mainly on the
central bithienyl unit, whereas the LUMO is heavily cen-
tered on the vicinal pyrazine or pyrimidine rings. However,
in 4 the HOMO/LUMO topologies involve the entire thio-
phene core, rendering electronic repulsion accompanying
the excitation less pronounced than in 1–3. Thus, the 1–3 vs.
structural reorganization of the excited state, with longer
lifetimes corresponding to increased relative probability of
non-radiative decay and correlating with lower PL quantum
efficiency. Indeed, these trends are in good agreement with
the present quantitative F measurements.
f
op
Solution optical gaps (E ), defined by the 0–0 transition
g
4
lmax differences not only reflect the intrinsic HOMO–
energies, were estimated from the intercept of the normal-
ized optical absorption and emission spectra, regarded as
LUMO gaps but also the nature of the optical excitation.
That is, the HOMO–LUMO excitation in 1–3 represents an
expansion of electron density (localized in the 6 portion for
the HOMO, delocalized over the entire conjugation path-
way for the LUMO), in contrast to the case in homogeneous
oligothiophenes 4–6.
[49]
the mirror of the 0–0 transitions.
Within the thiophene
series, the gaps decrease with extension of the core struc-
ture, from 3.75 eV (6) to 2.89 eV (5) to 2.61 eV (4)
(Table 2). In contrast, when one of the thiophene units is re-
op
placed by a diazine ring, Eg values increase to 2.95–3.04 eV,
Fluorescence emission spectra of compounds 1–6 were
measured in THF by exciting 10 –10 m solutions at the
much larger than those of 5 and 4, and comparable to those
of short phenylene-thiophene oligomers. The present re-
ꢁ
5
ꢁ6
[50]
Chem. Eur. J. 2009, 15, 5023 – 5039
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5027

J. T. Lꢂpez Navarrete, A. Facchetti, T. J. Marks et al.
op
sults for Eg are expectedly similar to those for l , however
abs
the D data may indicate that structural changes (i.e., planari-
zation) after excitation are smaller in the azine derivatives,
an effect in agreement with greater planarization in the so-
lution ground electronic state as a consequence of thio-
phene-azine through-space intramolecular interactions, as
discussed below.
Solid-state optical absorption/PL data for molecules 1–5
are collected in Table 2. The film absorption spectra exhibit
characteristic transitions at high energy (270–280 nm) found
in the spectra of all oligothiophenes and originating in the
[51]
thiophene ring. The position and shape of the high energy
p!p* transitions reflect the interplay of molecular struc-
ture, core length, and solid state packing. In general, a bath-
ochromic shift of this absorption is observed as the core
length increases on going from 5 to 1–4. Compared to solu-
tion values, the maxima of the strongest p–p* absorptions
are shifted to shorter wavelengths in the thin film spectra, as
a result of excitonic interactions between closed-packed
nearest-neighbor molecules. The coupling between the tran-
sition dipoles of molecules at crystallographically nonequi-
valent sites leads, in the case of a rigid infinite lattice, to
Figure 4. Powder FT-Raman spectra (l=1064 nm excitation) of thio-
phene-diazines, 1–3, and unsubstituted oligothiophene compounds, 4–6.
azine compounds along with those of the a,w-dimethyl-sub-
stituted oligothiophenes; DM-6: a,w-dimethylbithiophene,
DM-5: a,w-dimethylquaterthiophene, DM-4: a,w-dimethyl
sexithiophene. DM-6 was selected to compare its electronic
structure to that of the 6 central unit of 1–3. We first consid-
er the most intense Raman bands (Figure 4) originating
from the C=C stretching modes of the central bithienyl frag-
ment. This transition is an important molecular structure fin-
gerprint. The degree of bond length alternation (BLA), esti-
mated as the average sum of the distance differences be-
tween successive C=C/CꢁC bond pairs of each thiophene
[52]
well-known Davydov splitting. When the dipoles are all
parallel, the transition between the ground state and the
lower crystal excited state is strictly forbidden, thus account-
ing for the unique intense peak. In analogy to previous oli-
[53]
gothiophene results,
longest wavelengths–a low energy tail of the intense band
(ꢀ450 nm) in solid compounds 1–3 can be attributed to the
–0 transition of isolated molecules, either located in disor-
the weak unresolved absorption at
ACHTUNGTRENNUNG
0
ring is related to the ring C=C stretching mode associated
[
55]
dered domains or at grain boundaries, where molecular mis-
alignment can lead to weak intermolecular coupling and
minimal splitting of the excited levels. In fact, the absorption
tails of the intense absorptions closely parallel the trends
observed in solution. The shoulders on the highest energy
side of the intense solid state absorptions may result from
with the strongest Raman line.
This Raman transition
ꢁ
1
varies from 1498 to 1477 cm , on going from DM-6 to DM-
4, whereas in the isoelectronic diazine compounds the analo-
ꢁ1
ꢁ1
gous feature is located at 1481 cm in 1 and 1470 cm in 2.
This fall in frequency (relative to DM-6) is in agreement
with the pronounced reduction of the bithiophene B3LYP/6-
31G*-computed BLA values, which vary from 0.051 ꢃ for
DM-6 to 0.028 ꢃ for 1, and 0.031 ꢃ for 2. Note that similar
Raman frequency downshifts are observed for lines primari-
ly associated with the C=C stretching mode located at
[51]
the corresponding C=C vibronic replica.
Film photoluminescence spectra were obtained by lmax ex-
citation, and data are compiled in Table 2. The spectral
shapes and maxima strongly depend on molecular structure,
with most plots exhibiting additional peaks/shoulders. The
spacings between the better-resolved peaks (1300–
ꢁ
1
ꢀ1550 cm .
Particularly significant also are the discrepancies between
the Raman line positions in
ꢁ
1
ꢁ1
1
500 cm ) suggest coupling with excited state vibrational
1
(1481 cm ) and
2
ꢁ
1
modes, probably thiophene C=C stretches along the quinoi-
(1470 cm ), and the opposite behavior of the corresponding
computed BLAs (0.028 and 0.031 ꢃ, respectively). The as-
signment of ring BLAs is based on the ring character of the
stretching vibrations. However, as will be shown later, intra-
molecular S···N interactions favor polyene-like inter-ring
conjugation which should be evaluated not only as intra-ring
distances but also as CC bonds connecting both internal thi-
ophenes (note that the intra-ring C=C/CꢁC modes will be
strongly coupled to CꢁC inter-ring stretching transitions).
op
dal molecular backbone. Extraction of film Eg parameters
is less straightforward, owing to the multiple absorption/
op
emission transitions. However, reasonable estimates of Eg
values can be obtained from the onset of the absorption (at
op
1
0% of the maximum). Similar to the solution trends, Eg
values decrease as the core conjugation length increases and
are larger for the azine-substituted systems. In general, film
op
Eg values are smaller than the corresponding solution
values by 0.1–0.4 eV. A similar trend is observed for oligo-
These distances decrease somewhat on proceeding from 1
and 3 to 2, supporting the evolution of p–conjugation in
accord with the aforementioned falling energy gaps and the
concurrent downshifting of the most intense Raman bands.
Analyzing these effects further, Figure 5 shows the FT-
[54]
thiophenes.
Molecular structural features from vibrational spectra:
Figure 4 shows Raman spectra of the present thiophene-
5028
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 5023 – 5039

Organic Field-Effect Transistors
FULL PAPER
Figure 5. l=1064 nm FT-Raman spectra of 1 in A,B) the solid state and
C) in a CH Cl solution. The Raman spectra of the solid samples were
2 2
taken with a l=514.4 nm laser excitation. The band with an asterisk de-
notes a solvent vibration.
Raman spectra of 1 as a powder and as a thin film deposited
on a Si/SiO2 substrate (likely containing both amorphous
and crystalline forms, see below) as well as in the solution
phase. In all cases, the energies of the most intense Raman
transitions are very similar, indicating a strong tendency for
1
to planarize, even in solution. Because of the lack of
strong intermolecular interactions in solution, this observa-
tion corroborates the existence of substantial intramolecular
[56]
S···N attractions favoring planar conformations. The pres-
ent hypothesis concerning weak S···N interactions in thio-
phene-azine derivatives is supported by literature crystallo-
graphic data on similar thiophene-azine molecules (Figure 6;
see Figure 8 below for a sketch), which indicate short solid
state S···N distances, similar to those argued here for 1, 3,
and 2. Such interactions are also evident in the computa-
tional results discussed below for our compounds. The accu-
racy of theoretical molecular geometries can be tested in
one of the reference systems in Figure 6, for instance the ex-
perimental S····N distance of 2.93 ꢃ is theoretically repro-
duced as 2.95 ꢃ at the B3LYP/6-31G** level.
Figure 6. A) X ray crystal structures of molecules similar to the present
thiophene-azines exhibiting short intraACTHUNGTRENNUNG
[
57a,c,d]
(
distances are in ꢃ).
case is also shown.
The optimized geometries of molecules 1–6 provide metri-
cal parameters for comparison with the experimental dimen-
sions from single-crystal X-ray diffraction.
marizes molecular core lengths (maximum ring carbon-ring
carbon distance along the axis of the oligothiophene core)
and maximum lengths (the greatest extent of the molecule,
including the side-chains) illustrating, for the oligothio-
phenes, the excellent agreement between computed molecu-
[57]
[67]
Table 3 sum-
[68]
Electronic structure computation: Theoretical modeling of
thiophene-diazine electronic structures was performed in
parallel with the present experimental work. The parent oli-
gothiophenes were previously studied as polythiophene
models and as important representative p-electron organic
lar geometries and experimental crystal structures.
The
lengths of the oligothiophene cores are very similar for vari-
ous ring substitutions, indicating that the rigidity of the con-
jugated cores parallels the long molecular axes, with the
common all-anti heterocycle conformation prevalent in all
oligomers. Each additional thiophene unit is computed to in-
crease the length of the core by 3.91 ꢃ on average (vs.
[58]
materials. The utility of such modeling has advanced with
increased reliability of DFT methods, which now provide ac-
[59]
[68]
curate estimates of molecular geometries, including dihe-
3.87 ꢃ from experiment). Similar effects are also observed
[60]
[61]
dral angles and rotational barriers, dipole moments, as
well as electronic structure properties such as electron affin-
in computed dipole moment trends for the series 1–6
(Table 3). These largely symmetric molecules have modest
[
62,63]
[64]
[69]
ity,
vibrational frequencies.
ionization potential, band gaps and ground state
dipole moments.
[65,66]
To ensure the most meaningful
Important information on HOMO and LUMO energies
derives from analysis of the experimental electrochemical
and optical data, supplemented by theoretical calculations.
Table 4 summarizes electrochemical, optical, and computed
comparison, the entire 1–6 series was analyzed using the
same computational methodology (see the Computational
methods for details).
Chem. Eur. J. 2009, 15, 5023 – 5039
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5029

J. T. Lꢂpez Navarrete, A. Facchetti, T. J. Marks et al.
Table 3. Computed metrical parameters for molecules 1–6 derived from DFT//B3LYP/6-31G** optimized co- genvalues compare favorably
ordinates.
with experimental ionization
[
a,b]
Compound
Length
Avg. Dihedral
Angle [8]
potentials and slightly less well
m [D]
[b]
Core [ꢃ]
Maximum [ꢃ]
[62,63]
with electron affinities.
The
1
2
3
4
5
6
9.83
9.83
9.83
25.87
26.75
38.82
26.59 (26.14)
18.75 (18.17)
10.90 (10.61, 10.90)
0
0
0
0
0
principal disparity between
computed and electrochemical
HOMO/LUMO values doubt-
less reflects the differences in
environments (solvation), and
0.584
0.023
0.020
0.022
25.51 (25.29, 25.31)
17.67 (17.55, 17.48)
9.83 (9.76, 9.72)
0.5 (0.5, 0.8)
0.5
1.7 (0.0)
[
a] Core length refers to the maximum distance from ring carbon to ring carbon atom along the thiophene previous work has shown that
long axis excluding the lateral hexyl groups in 3, whereas the maximum length is the greatest extent of the parameters such as polarizabili-
molecule, including side chains. Both lengths include standard van der Waals radii for carbon (1.70 ꢃ), and hy-
drogen atoms (1.20 ꢃ). [b] Numbers in parentheses indicate experimental values from X-ray crystal structures
ty and cavity radius can be used
to linearly adjust (by means of
(
see Ref. [67]).
[72]
a Kamlet–Taft relationship)
computed ionization potentials
and electron affinities for solva-
tion to allow comparisons with electrochemical data.
Table 4. Comparison of electrochemical, optical, and computed HOMO–
[73]
LUMO energy gap (E
molecules 1–6.
g
) and absolute HOMO and LUMO energies for
Finally, the energy gaps also demonstrate excellent overall
agreement between electrochemical and optical data and
the computed DFT orbital energies, although the theoretical
gaps predicted from vertical transitions are somewhat great-
Compound
E [eV]
E
g
[eV]
[
a]
Experimental
Theoretical
CV[b]
op[c]
th[d]
Eg
Eg
Eg
HOMO LUMO HOMO LUMO
CV
op
1
2
3
4
5
6
ꢁ6.64 ꢁ3.39
ꢁ6.64 ꢁ3.36
ꢁ6.47 ꢁ3.39
ꢁ5.82 ꢁ3.13
ꢁ5.62
ꢁ5.26
ꢁ5.52
ꢁ4.79
ꢁ4.95
ꢁ5.47
ꢁ2.45
ꢁ2.32
ꢁ2.35
ꢁ2.18
ꢁ1.93
ꢁ1.25
3.05 3.04 3.17
3.28 2.95 2.94
3.08 3.04 3.17
2.69 2.61 2.61
2.89 3.02
er in energy than the Eg or Eg data in Table 4, as expect-
ed. Assuming the validity of the Koopmans approach, the
slight increase in the reduction potential for 2 relative to 1 is
consistent with the LUMO destabilization observed in
Figure 7. In the case of 2 (with two vicinal N atoms) only
one of the two N atoms participates in the LUMO, whereas
in 1, both heteroatoms contribute at the expense of one vici-
nal carbon. The greater electronegativity of N vs. C depress-
es the LUMO and renders reduction more favorable. From
the CV experiments (Table 1) it is found that 2 and 1 oxida-
tion processes occur at more positive potentials (0.2–0.6 V)
than in 5 and 4, where electrochemical generation of the
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(ꢁ5.79) ꢁ2.90
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(ꢁ6.22) ꢁ2.42
3.75 4.22
[
a] LUMO energy estimated from the relationship: LUMO (eV)=
ꢁ
4.84 eVꢁeE1/2,Red
. HOMO energy estimated from the relationship:
CV
HOMO (eV)=LUMOꢁE . The values reported in parenthesis are esti-
g
CV
mated using the optical gap. [b] From electrochemical data: Eg
=
1
1
=
2
=2
E1 ꢁE
. [c] From optical spectroscopic data. [d] From DFT compu-
1, Red
, Ox
tation.
energy gaps (see Figure 7) for
compounds 1,2 and 4–6. The
first important observation is
the excellent agreement be-
CV
op
tween solution Eg and Eg
data, with differences of only
ꢀ
0.1–0.2 eV for almost all mol-
ecules. Since these are two in-
dependently derived measure-
ments, the good agreement sup-
ports the accuracy of the re-
sults. Table 4 also summarizes
the experimental and computed
LUMO and HOMO energies,
calculated from the first reduc-
1
=
2
tion potentials (E ) and from
1
the onset of oxidation, respec-
[70]
tively. Although HOMO and
LUMO eigenvalues from DFT
methods cannot be formally
taken as either rigorous ioniza-
tion potentials or electron affin-
[71]
ities,
previous work has
Figure 7. DFT//B3LYP/6-31G** derived topologies and energies of the indicated frontier MOs for molecules
shown that B3LYP-derived ei- 1–6.
5030
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 5023 – 5039

Organic Field-Effect Transistors
FULL PAPER
corresponding cationic species is observed at 1.1–1.13 V.
Indeed, the thiophene-azine redox parameters are energeti-
cally similar to those of 6. This can be rationalized by con-
sidering: i) computed thiophene-azine HOMO levels are
rather close in energy to those of bithiophene (see Figure 7).
This similarity is accentuated in the pyrimidyl derivative
owing to cross-conjugation: ii) The HOMO wavefunctions
are primarily centered on the bithienyl fragment: iii) Al-
though thiophene-azine derivative p-conjugation is greater
than in 6, the aforementioned S···N interactions will moder-
ate the thienyl sulfur (which becomes more electropositive)
electron donor capacity. In this regard, it appears that elec-
tron wavefunction extension and S···N coupling effects
largely cancel each other.
The thiophene-azine derivatives all exhibit a strong solid
state optical absorption band at 410-420 nm (see Figure 3)
with characteristic vibronic components on both sides of the
maximum. According to the present time-dependent (TD)-
DFT calculations, these intense bands correspond to one-
electron HOMO!LUMO excitations. For 1, the absorption
maximum is computed at 433 nm with an oscillator strength
of f=1.62, which is by far the most intense of all the excita-
tions. In contrast to 6, the observation of vibronic replicas in
these mixed co-oligomers can be explained by the nature of
the azine N-electron lone pairs which can engage in hydro-
gen-bonding with neighboring molecules (see Figure S4 in
the Supporting Information), thus conferring extra rigidity
on the molecular skeleton and, as a result, the observed vi-
bronic structure. On the other hand, this type of interaction
may favor head-to-tail or lateral intermolecular interactions
instead of co-facial p-stacking, which may be unfavorable
for charge transport. For the same reasons, one expects non-
negligible intramolecular S···N interactions. Two facts sup-
port this: i) The sum of the S and N atomic radii is 3.350 ꢃ,
although the present calculations estimate these distances to
be significantly shorter, 2.922 ꢃ in 1, and 2.916 ꢃ in 2
Figure 8. A) B3LYP/6-31G** computed rotational barriers for TPm and
TPr as models of 1 and 2. Molecule 6 is also shown for comparison.
B) Theoretical predicted distances in ꢃ;. Ring BLAs are shown under-
lined (ꢃ).
(
Figure 7,8); and ii) the optimized geometry of 2 predicts
the C=C bonds of the pyridazyl and thiophene connected
groups (i.e., Kekule resonance structures) to be in a syn
conformation. It is well established that, in the absence of
additional effects, the trans isomers are generally more
p-interaction between the aromatic thiophenes and the pyri-
midyl groups). This observation is based on the cross-conju-
gated character of the pyrimidyl groups relative to bithio-
phene for this orbital. For the pyridazyl analogue, the inter-
action with the bithiophene core leads to linear p-conjuga-
tion, and results in the highest HOMO energy within the
series. In contrast, the LUMO wavefunction displays a
bonding interaction between the thienyl core and the pyri-
midyl moieties, leading to an effective inter-ring C=C/CꢁC
[74]
stable. Figure 8 also shows the computed B3LYP/6-31G**
rotational barriers about the thiophene-pyrimidine (TPm)
and thiophene-pyridazine (TPr) bonds (i.e., as models of 1
and 2) which reveal that the syn (i.e., 1808) conformation is
the most stable. The rotational barrier for 6 is also shown
for comparative purposes. From Figure 8, note that the rota-
tion barriers for the syn-anti interconversion are larger in
the diazine oligomers than in 6, in agreement with the great-
er rigidity, owing to the through-space S···N interactions.
Furthermore, these interactions are also responsible for sta-
bilizing the syn isomer in the diazine-thiophene derivatives,
whereas the anti disposition is invariably preferred in unsub-
stituted oligothiophenes.
conjugation in the central bithiophene which extends to-
wards the outermost groups. As a result, the LUMO energy
is more stabilized than its HOMO counterpart. Further sub-
stitution of the above molecular fragments with two external
thiophene rings does not alter the HOMO/LUMO wave-
functions, causing only slight modifications of the energies.
As a whole, the 6!1 transition results in a net reduction of
the HOMO/LUMO gap in agreement with the observed re-
At the intramolecular level, substitution of 6 with two pyr-
imidyl groups (see Figure 7) slightly stabilizes the HOMO,
likely owing to inductive effects (note that there is minimal
duction of the optical gap associated with the HOMO–
op
g
LUMO photoexcitation and E . The 1!2 progression
Chem. Eur. J. 2009, 15, 5023 – 5039
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5031

J. T. Lꢂpez Navarrete, A. Facchetti, T. J. Marks et al.
causes an additional narrowing of the HOMO–LUMO
energy gap, owing to simultaneous HOMO and LUMO de-
stabilization, more pronounced for the former MO. The
greater destabilization of the HOMO on going from 1!2
can be rationalized as follows: i) Inductive effects are parti-
ally removed as a result of the NꢁN bond; ii) electrostatic
The maxima of the emission bands for molecules 1–6
follow the same trend as the theoretical S !S transition en-
1
0
ergies. Theoretical and experimental Stokes shifts also ex-
hibit the same trends. From 1 to 3, the fluorescence quan-
tum yields slightly increase, as do the associated oscillator
strengths. Also, for the homologous oligothiophene series,
S !S oscillator strengths and fluorescence quantum yields
coupling cannot be excluded in the pyridazyl compound
1
0
(
e.g., lone pair repulsion within the N=N moiety may be mi-
behave similarly. For 6, discrepancies between experiment
and theory are observed, likely due to efficient intersystem
crossing, not considered in the calculations, and which has
tigated by the S···N interactions), which would favor rigidity
and overall p-electron conjugation; iii) strengthening of
these S···N interactions (see Figure 8) could decouple the S
electron pair from the thiophene 4n+2 ring aromaticity,
thus favoring inter-ring conjugation. These concurrent ef-
fects are evident in the optical spectra through a red-shift of
the lowest energy absorption transition on going from 1!2.
As already mentioned, upon excitation of the HOMO–
LUMO transition, the emission spectra of the longer oligo-
thiophenes (i.e., 3–6 units) exhibit vibronic features reflect-
[75]
been reported for the shortest member. From 1 to 2, Ff
decreases (0.38 and 0.24), whereas the computation predicts
an increase of the oscillator strength (1.68 and 1.99).
The above comparisons establish that a main route of re-
laxation from the S excited states in larger oligothiophenes
1
is a radiative process, involving dipole–dipole coupling be-
tween the two electronic states. This is important since the
photophysical properties of aromatic oligothiophenes are af-
fected by other fluorescence quenching processes, namely
intersystem crossing for short chain molecules such as men-
tioned above for 6. In this regard, the fluorescence quantum
yield of 2 deviates from the theoretical prediction and is
likely affected by intersystem crossing. To some extent the
para connectivity of the pyridazyl moiety more efficiently
involves N atom participation in the p-electron system, and
heteroatoms always favor non-radiative decay routes such as
ing rigidity of the quinoidal emitting S state. For these ex-
1
cited states, the additional conjugation of the 6 fragment
with the external six-membered groups accounts for the vi-
bronic peaks in the thiophene-diazine derivatives (note that
6
does not exhibit a vibronic progression). Fluorescence
quantum yields greatly increase upon conjugation of the
central bithiophene fragment with the diazine groups, likely
owed to the decreasing S atom role. Moreover, the solid
state emission spectra are red-shifted nearly by 100 nm vs.
the solution spectra, indicating strong solid state intermolec-
ular interactions. Comparison with bithiophene in the solid
state corroborates the importance of the intra/intermolecu-
lar contacts characteristic of diazines. Note that 1 displays a
greater fluorescence quantum yield than 2, likely owed to
the meta!para change in connectivity with the bithienyl
fragment.
[76]
intersystem crossing or charge transfer processes. One ad-
ditional reason is evident in Figure 9 in which S state geo-
1
metrical distortions are shown. Note that, evolving from S0
to S , 2 always exhibits smaller CꢁC changes, as in the
1
common bithiophene fragment. This implies that p-electron
delocalization is reduced compared to the two homologues,
and therefore intersystem crossing is more probably owed to
greater conformational flexibility.
Table 5 compares experimental emission energies and
fluorescence quantum yields with the theoretical S !S
1
0
transition energies and oscillator strengths from TD-DFT/
B3LYP/6-31G** calculations on 1–6, using RCIS S geome-
1
tries. For radiative decay processes from the first excited sin-
glet, oscillator strengths in the S geometry are proportional
1
to the decay radiative constant krad and consequently, for
molar extinction coefficients and S !S oscillator strengths
0
1
in absorption, theoretical and experimental comparisons can
be delineated for emission.
Table 5. Comparison for experimental and calculated data for the fluo-
rescence emission properties of molecules 1–6. Theoretically the S
transition is considered to be the relevant radiative transition. For 3,
1
!S
0
methyl groups rather than hexyl groups are considered in the model.
Figure 9. The CꢁC bond distance differences between the S
1
and S states
0
Compound
Experimental
Theoretical
for the compounds under study.
l
f
[nm]
DE [eV]
F
f
l
f
[nm] DE [eV]
f
1
2
3
4
5
6
455
465
455
2.72
2.67
2.72
0.38
0.24
0.42
0.46
0.20
0.01
476
509
478
566
484
351
2.60
2.44
2.59
2.19
2.56
3.54
1.68
1.99
1.79
2.09
1.27
0.44
Figure 9 presents RCIS/3-21G* ab initio optimized geo-
505*, 539 2.46
450*, 478 2.76
363
metries for the S states of 1–6. Note that the skeletal back-
1
bone is partially quinoidal and planar in the ground elec-
tronic state. Averaged skeletal deformations and hence ri-
3.42
5032
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 5023 – 5039

Organic Field-Effect Transistors
FULL PAPER
gidity in the 1–3 molecules are always midway between
those observed for 5 and 4. This rigidity likely justifies the
larger computed S !S oscillator strengths (1.68 in 1) com-
cates the presence of only a single phase characterized by a
st
th
d-spacing of ꢀ22 ꢃ (1 through 4 order reflections ob-
served). However, the XRD intensity of this film is far
weaker than that of 50 nm thick films of 4 and other oligo-
thiophenes. The diffraction pattern for 2 exhibits two
phases/orientations, one of which dominates. This high-in-
tensity phase is characterized by a d-spacing of 19.9 ꢃ,
versus a d-spacing of 21.7 ꢃ for the low-intensity phase.
However, the reflections of both phases are quite broad,
suggesting minimal microstructural order. From the comput-
ed 1 and 2 molecular lengths (Table 6) and the correspond-
1
0
pared to S !S (1.62 in 1), the small difference can be at-
0
1
tributed to the existing core stiffening in S by S···N interac-
0
tions. In this regard, the greater F in 1 vs. 2 can be also
f
argued to result from the stronger through-space S···N inter-
actions in the former (e.g., the larger computed rotational
barriers in Figure 8). This result is significant since it demon-
strates that for molecules with the same or similar numbers
of p-electrons, core rigidity enhances emissive properties.
Thin film characterization: Organic semiconductor film mi-
crostructure characterization is essential to understanding
charge transport in OFETs and other semiconductor-based
devices. For OFETs, critical factors are semiconductor film
crystallinity, molecular orientation with respect the dielectric
surface, crystal grain size and connectivity, and how these
properties are affected by the surface chemistry of the sub-
strate on which the semiconductor film is deposited. Films
of the new semiconductors (50 nm thick) were deposited by
thermal evaporation under high vacuum on glass substrates
and p -Si wafers havinga 300 nm thick thermal oxide coat-
ing. Semiconductor film crystallinity and molecular orienta-
tion on the dielectric surface were assessed by wide-angle x-
ray diffraction (WAXRD) measurements. Organic semicon-
ductor film morphology was studied by scanning electron
microscopy (SEM).
Table 6. Observed d spacing values and computed/experimental geomet-
ric parameters for films of compounds 1–6.
[a,b]
Compound
D spacing [ꢃ]
Molecular Length [ꢃ]
1
2
3
4
5
6
22.0
25.87
26.75
38.82
26.59 (26.14)
18.75 (18.17)
10.90 (10.61, 10.90)
19.9, 21.7
27.3, 18.6
23.3
15.43
7.82
[
a] Both lengths include van der Waals radii for carbon (1.70 ꢃ) and hy-
+
+
drogen atoms (1.20 ꢃ). [b] Numbers in parentheses indicate experimental
values from single crystal X-ray structures (see Ref. [68]).
ing d-spacings for the dominant phase, the average molecu-
lar tilt angles with respect the substrate normal are estimat-
ed to be ꢀ328 and ꢀ428, respectively. Note that these
angles are somewhat larger than those found for 4 films
WAXRD q–2q diffraction patterns of the thiophene-dia-
zine films deposited at the optimized temperature T =
D
1
108C indicate that all films are poorly textured (Figure 10).
A
T
N
T
E
U
G
Films deposited at lower temperatures (T =25, 70, 908C,
not shown) exhibit weaker/broader Bragg reflections indica-
tive of lower crystallinity. The diffraction pattern of 1 indi-
phases or orientations. The majority phase/orientation ex-
hibits a d-spacing of 27.3 ꢃ, whereas the other phase/orien-
tation is characterized by an unusually small d-spacing of
D
1
8.6 ꢃ. The corresponding molecular tilt angles for the two
phases are estimated to be ꢀ458 and ꢀ618, respectively.
SEM images of the present semiconductor films deposited
at 1108C show that all films are polycrystalline with the 3
films exhibiting the largest crystallites (Figure 11). Com-
pounds 1 and 2 film crystallites are ꢀ50 nm wide by 200–
3
00 and 50–150 nm in length, respectively. The surface of 3
reveals a smooth background and the presence of large
ribbon-like flakes ꢀ1 mm in size. Similar morphologies have
been observed for several oligothiophene films and are gen-
[77]
erally favorable for efficient carrier mobility.
Transistor characterization: The semiconductor performance
of the present new materials was evaluated in a top-contact
bottom-gate OFET geometry. All of the new materials ex-
hibit p-type transport. Compound 2 exhibits the highest mo-
ꢁ
3
2
ꢁ1 ꢁ1
bility of 4ꢄ10 cm V s , a threshold voltage V =ꢁ68 V,
T
7
and current on off ratio Ion:off =10 for films grown at T =
D
1
108C. In general, these saturation hole mobilities (m ) are
h
greater for films grown on hexamethyldisilazane (HMDS)-
Figure 10. WAXRD qꢁ2q diffraction patterns of 50 nm semiconductor
treated substrates (Table 7) than on bare SiO . Interestingly,
2
2
films vapor-deposited onto HMDS-treated Si/SiO substrates at 1108C.
electron transport is not observed for these devices, al-
though the diazine-thiophene semiconductors are clearly
The intensity of the first reflection is: 850 (1), 109400 (2), and 20300 cps
3).
(
Chem. Eur. J. 2009, 15, 5023 – 5039
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5033

J. T. Lꢂpez Navarrete, A. Facchetti, T. J. Marks et al.
this material, a pyrazine fragment joins two thiophenes
through para connections as opposed to pyrimidines in 1
and 3, which are meta functionalized. This meta functionali-
zation results in non-linear ring-ring connections, each of
which can have two conformations, which are computed to
ꢁ
1
lie within 1 kcalmol (Figure 8). Such energetically similar
conformations may create packing defects that would lower
the observed mh.
Charge transport vs. molecular structure: We conclude by
discussing OFET charge transport trends, analyzing thio-
phene-azine molecular electronic and structural parameters,
such as redox properties and intramolecular reorganization
energies (i.e., l ). The latter parameter considers the struc-
h
tural reorganization needed to accommodate charge as a
prerequisite for efficient transport. These l parameters,
h
[78]
computed as described in literature,
are compiled in
Figure 11. Scanning electron micrographs of films of the thiophene-dia-
Table 8 together with the experimental electrochemical oxi-
zine semiconductors A) 1, B) 2, and C) 3 vapor-deposited onto HMDS-
2
treated SiO /Si at 1108C.
Table 8. Reorganization energies, oxidation potentials, and maximum
hole mobilities measured for compounds 1–5.
2
ꢁ1 ꢁ1
Compound
H
lh(radical anion) [eV] lh(radical cation) [eV] Eox1 [V] m [cm V s ]
more electron-deficient than their p-type oligothiophene an-
alogues (see the Electrochemistry Section).
ꢁ
4
1
2
3
4
5
0.2751
0.2312
0.3009
–
0.2468
0.2353
0.2192
1.63
1.87
1.60
0.98
1.10
1ꢄ10
4ꢄ10
3ꢄ10
6ꢄ10
1ꢄ10
ꢁ3
ꢁ3
ꢁ2
ꢁ2
The alkyl-substituted pyrimidine 3 exhibits m comparable
h
[
a]
to that of 2 for most growth temperatures, with optimized
0.301
0.345
[
a]
ꢁ
3
2
ꢁ1 ꢁ1
–
parameters m =3ꢄ10 cm V s , V =ꢁ55 V, and I
=
on:off
h
T
7
2
ꢄ10 . Note that the m variations observed over a 858C
[a] Values are taken from reference [78].
h
growth range can be considered inconsequential for a mate-
rial having mobilities in this range. The 3 films have similar
mh values on both hydrophobic and hydrophilic substrates.
dation potentials Eox1 and OFET hole mobilities. With the
exception of 2, reorganization energies for anions which are
the molecular parameter relevant to electron transport in
these materials are larger than those predicted for cations.
The n-hexyl substitution of 1 in 3 decreases the hole reor-
ganization energy by 0.03 eV, in agreement with the ob-
served increase in p-type mobility. In the case of 4 and 5
versus the diazine molecules, the greater hole mobility ap-
pears to be aided by more facile hole formation (i.e., lower
Eox1) although the small reorganization energies in the dia-
zines suggest that mobilities should be similar. The FET mo-
bility of 6 has not been reported. The present semiconductor
description is in accord with the above discussion of the op-
These observed m variations may reflect small differences
h
in charge carrier trap densities and/or local film morphology
variations. The unsubstituted pyrimidine semiconductor 1
exhibits far lower m values and exhibits significant sensitivi-
h
ty to the dielectric surface energy. Thus, m on the bare hy-
h
3
drophilic SiO surface is typically 10 times lower than on
2
HMDS-treated substrates, and hole transport is not ob-
served for 258C growth on untreated SiO . Increased sub-
2
strate sensitivity and lower m has been observed previously
h
[36]
for unfunctionalized thiophenes such as 5 and 4.
The greater m observed in 2 is likely a result of a more
h
linear geometry, as shown by the TD-DFT calculations. In
2
ꢁ1 ꢁ1
Table 7. FET mobilities (m, cm V
s
), current on:off ratios (Ion:Ioff), and threshold voltages (V
T
, V) for semiconductor films of series 1–5, as a function
of deposition temperature.
2
I
58C 708C
908C 1108C
[
a]
Compound
Sbs.
m
on:Ioff
V
T
m
I
on:Ioff
V
T
m
I
on:Ioff
V
T
m
I
on:Ioff
V
T
ꢁ
7
2
5
6
7
5
6
4
5
ꢁ7
ꢁ4
ꢁ4
ꢁ4
ꢁ4
ꢁ3
ꢁ2
ꢁ3
3
4
5
5
4
7
4
4
ꢁ6
ꢁ5
ꢁ3
ꢁ3
ꢁ5
ꢁ4
3
4
5
7
4
5
S
H
S
H
S
H
H
H
NA
NA
NA
3ꢄ10
1ꢄ10
1ꢄ10
2ꢄ10
4ꢄ10
7ꢄ10
6ꢄ10
1ꢄ10
3ꢄ10
2ꢄ10
1ꢄ10
1ꢄ10
1ꢄ10
3ꢄ10
4ꢄ10
1ꢄ10
ꢁ70
ꢁ88
ꢁ87
ꢁ74
ꢁ66
ꢁ106
ꢁ12
ꢁ15
8ꢄ10
1ꢄ10
1ꢄ10
8ꢄ10
1ꢄ10
3ꢄ10
3ꢄ10
5ꢄ10
1ꢄ10
1ꢄ10
4ꢄ10
2ꢄ10
2ꢄ10
2ꢄ10
1ꢄ10
7ꢄ10
ꢁ124
ꢁ97
ꢁ107
ꢁ82
ꢁ85
ꢁ55
ꢁ10
ꢁ18
2ꢄ10
3ꢄ10
2ꢄ10
4ꢄ10
2ꢄ10
2ꢄ10
6ꢄ10
3ꢄ10
1ꢄ10
1ꢄ10
3ꢄ10
1ꢄ10
ꢁ85
ꢁ80
ꢁ64
ꢁ68
ꢁ103
ꢁ99
1
2
3
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
ꢁ
5
4
3
4
4
2
2
3
4
5
5
5
4
5
ꢁ4
ꢁ3
ꢁ3
ꢁ5
ꢁ4
ꢁ2
ꢁ2
2ꢄ10
3ꢄ10
2ꢄ10
3ꢄ10
2ꢄ10
4ꢄ10
1ꢄ10
5ꢄ10
1ꢄ10
2ꢄ10
1ꢄ10
2ꢄ10
2ꢄ10
1ꢄ10
ꢁ84
ꢁ67
ꢁ77
ꢁ90
ꢁ43
ꢁ14
ꢁ17
4
5
[
2
a] Substrate surface treatment; S: SiO , H: HMDS.
5034
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 5023 – 5039

Organic Field-Effect Transistors
FULL PAPER
tical spectroscopy and electrochemistry which indicate that
a significant portion of thiophene-diazine properties is domi-
nated by the central bithiophene fragment. The role of the
diazine units thus appears largely to extend the intrinsic
characteristics of the bithienyl units but, as a whole, not to
introduce unique electronic properties. In this sense, these
extended bithiophenes counter-intuitively behave as conven-
tional hole-semiconducting oligothiophenes with properties
similar to those of medium-sized oligothiophenes (i.e., 3–4
thiophene units).
the reaction flask. The crude solid product was removed by filtration,
and the filtrate was diluted with ether (100 mL), poured into an aqueous
+
ꢁ
solution of NH
After drying over MgSO
4
F
(3.0 g, 100 mL), and the organic phase separated.
and filtration, the solvent was evaporated af-
4
fording a white solid. The two solid portions were next combined and re-
crystallized from toluene to give the pure product as a colorless crystals
1
(
9
2.36 g, 9.66 mmol, 72.0% yield). M.p.=1468C; H NMR (CDCl
3
): d=
3
3
.07 (d, J=1.4 Hz, 1H), 7.86 (d, J=3.7 Hz, 1H), 7.83 (d, 1H), 7.57 (d,
3
J=5.1 Hz, 1H), 7.20 ppm (dd, 1H).
Synthesis of 5,5’-bis-(6-chloropyrimid-4-yl)-2,2’-dithiophene (7): A mix-
ture of 5,5’-bis(tri-n-butylstannyl)-2,2’-dithiophene (7.58 g, 10.18 mmol),
4
,6-dichloropyrimidine (6.00 g, 40.27 mmol) and tetrakis(triphenylphos-
phine)palladium(0) (0.26 g, 0.22 mmol) in dry toluene (70 mL) was deaer-
ated twice with nitrogen. The reaction mixture was next refluxed for 6 h
and, after cooling, the resulting precipitate was collected by filtration.
The crude solid product was washed several times with hexane and then
Conclusions
with methanol, to afford the essentially pure product as a yellow powder
Three new diazine-functionalized oligothiophenes have been
synthesized and their structural, optical, vibrational, electro-
chemical, and semiconductor properties studied in compari-
son with those of the corresponding oligothiophenes, with
the goal of enhancing transport properties of value in organ-
ic electronics. The molecular properties of these systems are
dominated by the central bithiophene fragment, and hence
their properties correspond largely to p-electron extended
bithiophenes. This is surprising in the case of charge mobili-
ty since, a priori, these materials would appear to be best
suited for electron transport (i.e., azine units behave as elec-
tron acceptors). These diazine-functionalized oligothio-
phenes are reasonably efficient hole transporters, an innate
characteristic of readily oxidized oligothiophenes. To probe
these issues further, a broad set of interconnected physico-
chemical data (optical, electrochemical, vibrational, confor-
mational, energetic, etc.) are acquired and analyzed. Re-
garding organic electronics, the present diazine-oligothio-
phene OFET mobilities are moderate. Further work will
focus on modification of azine synthons, for example with
electron acceptors, which could effect majority charge carri-
er sign inversion, or even more interesting, afford ambipolar
semiconductors able to transport in a similar regime, holes
1
(
2.80 g, 7.16 mmol, 70.3% yield). M.p.=2618C (sublimation); H NMR
3
3
(CDCl
2
3
): d=8.91 (d, J=1.3 Hz, 2H), 7.73 (d, J=4.0 Hz, 2H), 7.60 (d,
(70 eV): m/z (%): 390.9(100%) 392.9 (75%);
: C 49.11 ,H 2.06, N 14.32;
H), 7.37 ppm (d, 2H); MS
A
H
U
G
R
N
U
G
ACHTUNGTRENNUNG
elemental analysis calcd (%) for C16
8 2 4 2
H Cl N S
found: C 49.21, H 2.19, N 14.16.
Synthesis of 5,5’-bis(6-(thien-2-yl)pyrimid-4-yl)-2,2’-dithiophene (1): A
mixture of 5,5’-bis(6-chloropyrimid-4-yl)-2,2’-dithiophene (1.40 g,
3.58 mmol), tri-n-butylstannylthiophene (3.00 g, 8.04 mmol), tetrakis(tri-
phenylphosphine)palladium(0) (0.20 g, 0.17 mmol), and few crystals of
2
,6-di-tert-butyl-4-methylphenol in dry toluene (120 mL) was deaerated
twice with nitrogen. The reaction mixture was then refluxed for 10 h and,
after cooling, the precipitate was collected by filtration (2.00 g). The solid
residue was washed several times with hexane and then recrystallized
from pyridine (170 mL) to afford the pure product as an orange solid
(0.88 g, 1.81 mmol, 50.5% yield). Extremely pure samples can be ob-
1
tained by gradient sublimation. mp=3068C; H NMR (CDCl
3
): d=9.09
3
3
3
(
7
d, J=1.2 Hz, 2H), 7.94 (d, J=3.7 Hz, 2H), 7.85 (d, J=4.0 Hz, 2H),
3
.83 (d, 2H), 7.61 (d, J=4.9 Hz, 2H), 7.41 (d, 2H), 7.24 (dd, 2H); MS-
(70 eV): m/z (%): 487.0 (100); elemental analysis calcd (%) for
C H N S : C 59.23, H 2.91, N 11.52; found: C 59.17, H 2.97, N 11.37.
AHCTUNGTRENNUNG
2
4
14
4 4
Synthesis of 5,5’-bis(6-(5-hexylthien-2-yl)pyrimid-4-yl)-2,2’-dithiophene
3): A mixture of 5,5’-bis(6-chloropyrimid-4-yl)-2,2’-dithiophene (1.05 g,
.68 mmol), 2-(tri-n-butylstannyl)-5-hexylthiophene (2.60 g, 5.68 mmol),
(
2
tetrakis(triphenylphosphine)palladium(0) (0.15 g, 0.13 mmol), and few
crystals of 2,6-di-tert-butyl-4-methylphenol in dry toluene (90 mL) was
deaerated twice with nitrogen. The reaction mixture was then refluxed
for 12 h and, after cooling, the precipitate was collected by centrifugation
(
1.66 g). The solid crude product was washed once with hexane and then
(
owing to the bithiophene portion) and electrons (owing to
dissolved in hot chloroform (150 mL). The warm solution was filtered
and the solvent evaporated to give the pure product as a brown solid
the azine units). Such charge transport characteristics, if
properly combined with emissive properties, may provide
new OLET materials combining electron-hole transport
with efficient luminescence.
(
1.21 g, 1.85 mmol, 71.1% yield). An analytically pure sample was ob-
1
tained by recrystallization from toluene. M.p.=2308C; H NMR
3
3
3
(CDCl ): d=9.04 (d, J=1.1 Hz, 2H), 7.78 (d, J=4.0 Hz, 2H), 7.72 (d,
3
3
J=3.8 Hz, 2H), 7.83 (d, 2H), 7.37 (d, J=4.2 Hz, 2H), 6.90 (d, 2H), 2.89
3
3
(
t, J=7.5 Hz, 4H), 1.76 (m, 4H), 2.00–1.40 (m, 12H), 0.91 ppm (t, J=
7
5
.3 Hz, 6H); elemental analysis calcd (%) for C36
.86, N 8.56; found: C 65.88, H 5.57, N 8.57.
38 4 4
H N S : C 66.01, H
Experimental Section
[
82]
Synthesis of 3-(thien-2-yl)-6-chloropyridazine (8): A mixture of tri-n-
butylstannylthiophene (6.26 g, 16.78 mmol), 3,6-dichloropyridazine
Materials and methods: The reagents 5,5’-bis(tri-n-butylstannyl)-2,2’-bi-
thiophene and 2-(tri-n-butylstannyl)thiophene were synthesized accord-
(
0
(
5.00 g, 33.56 mmol), tetrakis(triphenylphosphine)palladium(0) (0.20 g,
.17 mmol), and few crystals of 2,6-di-tert-butyl-4-methylphenol in DMF
50 mL) was deaerated twice with N . The reaction mixture was then
[
79]
ing to known procedures.
thiophene was synthesized as reported earlier.
spectra were measured in CDCl or CD Cl on a Varian Mercury 400
room temperature) or a Varian Inova 400 (high temperature) instru-
ment.
The reagent 2-(tri-n-butylstannyl)-5-hexyl-
[
80] 1
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
H NMR (400 MHz)
2
3
2
2
heated at 808C for 6 h and, after cooling, poured into water (100 mL).
The resulting white precipitate was collected, washed several times with
water, and dried under vacuum. This solid was next taken up in ether
(25 mL) and filtered to afford nearly pure solid product (3.01 g, 36.7%
yield) after evaporation of the ether. Finally, this solid was recrystallized
(
[
81]
Synthesis of 4,6-dithien-2-ylpyrimidine (9): A mixture of 2-tri-n-butyl-
stannylthiophene (10.52 g, 28.19 mmol), 4,6-dichloropyrimidine (2.00 g,
1
0
3.42 mmol),
tetrakis(triphenylphosphine)palladium(0)
(0.345 g,
from MeOH-H O to give the pure target compound as white crystals
2
1
.30 mmol), and a few crystals of 2,6-di-tert-butyl-4-methylphenol in dry
3
(1.56 g, 7.90 mmol, 47.3% yield). M.p.=1558C; H NMR (CDCl ): d=
3
3
3
toluene (20 mL) was deaerated twice with nitrogen. The reaction mixture
was then refluxed for 6 h and, after cooling, a white precipitate formed in
7.75 (d, J=11.0, 1H), 7.68 (d, J=3.6, 1H), 7.54 (d, J=5.0, 1H), 7.51
(d, 1H), 7.18 ppm (dd, 1H).
Chem. Eur. J. 2009, 15, 5023 – 5039
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5035

J. T. Lꢂpez Navarrete, A. Facchetti, T. J. Marks et al.
Synthesis of 5,5’-bis(6-(thien-2-yl)pyridazin-3-yl))-2,2’-dithiophene (2): A
mixture of 3-(thien-2-yl)-6-chloropyridazine (1.50 g, 7.63 mmol), 5,5’-
bis(tri-n-butylstannyl)-2,2’-dithiophene (2.84 g, 3.81 mmol), and tetrakis-
over, DFT force fields calculated using the B3LYP functional yield infra-
[
86]
red spectra in very good agreement with experiment. The standard 6-
[
87]
31G** basis set was used. Optimized molecular geometries were deter-
mined on isolated entities. Vertical one-electron excitations were comput-
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(triphenylphosphine)palladium(0) (0.09 g, 0.08 mmol) in dry DMF
[
88]
(
30 mL) was deaerated twice with nitrogen. The reaction mixture was
ed using time-dependent DFT (TDDFT) methods.
energy electronic excited states were computed for 1 and 2. The geome-
try optimization of the first excited S state was carried out using ab
initio methods with restricted configuration interaction singles (RCIS) in-
The 20 lowest-
then heated at 708C overnight. After cooling, the resulting precipitate
was collected and washed several times with hexane, MeOH, and ether.
After drying, the pure product was obtained as a light-orange solid
1
[
89]
(
1.60 g, 3.29, 86.3% yield). M.p.>3508C; elemental analysis calcd (%)
corporated in Gaussian 03, and the 3–21G* basis was chosen for all
molecules. TD-DFT/B3LYP/6-31G* calculations were used to calculate
for C24 : C 59.23, H 2.91, N 11.52; found: C 58.85, H 3.11, N 11.57.
14 4 4
H N S
NMR experiment was not recorded, owing to the low solubility of this
compound. However the elemental analysis and mass allowed us to
assess the compounds identity and purity.
S
1
!S
0 1
electronic transition frequency from the optimized (relaxed) S
state. These calculations were performed over the geometries optimized
by RCIS method.
Raman and optical spectroscopic measurements: FT-Raman scattering
spectra were collected on
a Bruker FRA106/S instrument with a
Nd:YAG laser source (lexc =1064 nm), in a back-scattering configuration.
The operating power for the exciting radiation was limited to 100 mW in
all the experiments. Samples were analyzed as pure solids, averaging
Acknowledgements
ꢁ
1
1
000 scans with 4 cm spectral resolution. Raman spectra of thin films
We thank ONR (N00014–02–0909), the NSF-MRSEC program through
the Northwestern Materials Research Center (DMR-0520513), and Poly-
era Corporation for support of this research at Northwestern. Research
at the University of Malaga was supported by the Ministerio de Educa-
cion y Ciencia (MEC) of Spain through Project CTQ2006-14987-C02-01.
We are also indebted to the Junta de Andalucia for funding our FQM-
0159 scientific group and for the Project P06-FQM-01678. R. P. O. thanks
the MEC for a personal postdoctoral grant.
were recorded using a RENISHAW Microscope Invia Reflex Raman
working with the 514.5 nm laser excitation wavelength. UV/Vis absorp-
tion spectra were collected on an Agilent 8453 instrument equipped with
a diode array for the fast acquisition of all absorptions in the 190–
1
100 nm spectral range. Fluorescence spectra were recorded with a Jasco
FP-750 spectrometer. Solutions were prepared with an absorbance be-
tween 0.1 and 0.2 at the wavelength region of experimental interest.
Fluorescence quantum yields were determined by comparison with 0.1m
quinine sulfate in 0.05m sulfuric acid as reference and corrected for the
refractive index of the solvent.
[1] a) Handbook of Oligo- and Polythiophenes (Ed.: D. Fichou), Wiley-
VCH, Weinheim, 1999; b) Polythiophenes - Electrically Conductive
Polymers (Eds.: G. Schopf, G. Kossmehl), Springer, Berlin, 1997.
[2] a) P. Frꢅre, J.-M. Raimundo, P. Blanchard, J. Delaunay, P. Ri-
chomme, J.-L. Sauvajol, J. Orduna, J. Garin, J. Roncali, J. Org.
Chem. 2003, 68, 5357; b) J. Z. Wang, Z. H. Zheng, H. W. Li, W. T. S.
Huck, H. Sirringhaus, Nat. Mater. 2004, 3, 171; c) A. Afzali, T. L.
Breen, C. R. Kagan, Chem. Mater. 2002, 14, 1742.
3] a) P. Bꢆuerle, Electronic Materials: The Oligomer Approach (Eds.:
K. Mꢇllen, G. Egner), Wiley-VCH, Weinheim, 1998, p. 105; b) L.
Groenendaal, E. W. Meijer, J. A. J. M. Vekemans, Electronic Materi-
als: The Oligomer Approach (Eds.: K. Mꢇllen, G. Egner), Wiley-
VCH, Weinheim, 1998, p. 235; c) J. Gruber, R. W. Chia Li, I. A.
Hꢇmmelgen, Handbook of Advanced Electronic and Photonic Mate-
rials and Devices, Vol. 8 (Ed.: H. S. Nalwa), Academic Press, San
Diego, 2000, pp. 163–184; d) H. Meng, J. Zheng, A. J. Lovinger, B.-
C. Wang, P. G. Van Patten, Z. Bao, Chem. Mater. 2003, 15, 1778;
e) J. L. Reddinger, J. R. Reynolds, Adv. Polym. Sci. 1999, 145, 57.
4] a) S. C. Moratti, Handbook of Conducting Polymers; 2nd ed. (Eds.:
T. A. Skotheim, R. L. Elsenbaumer, J. R. Reynolds), Marcel
Dekker, New York, 1998, pp. 343; b) R. D. McCullough, Adv. Mater.
Electrochemical measurements: Cyclic voltammetry experiments were
performed in 0.1m tetrabutyl ammonium hexafluorophosphate (TBAPF )
6
solutions in dry, oxygen-free THF. Glassy carbon was used as the working
electrode, platinum gauze as the auxiliary electrode, and Ag/AgCl as the
+
reference electrode, which was checked against the Fc/Fc couple after
each measurement.
Thermal measurements: Differential scanning calorimetry (DSC) and
thermogravimetric analysis (TGA) were performed by using TA DSC
[
2
920 and Mettler–Toledo TGA instruments, respectively.
Thin film growth, characterization, and device fabrication: Prime grade
p-doped silicon wafers (100) of 300 nm thermally grown oxide (Process
Specialties Inc. and Montco Silicon Technologies Inc.) were used as
device substrates. They were first rinsed with water, methanol, and ace-
tone before film deposition. Trimethylsilation of the Si/SiO
carried out by exposing the silicon wafers to HMDS vapor at room tem-
perature in a closed container under N overnight. Organic semiconduc-
tors were deposited by vacuum evaporation (pressure <10 Torr) at se-
2
surface was
2
ꢁ
5
[
[
ꢁ
1
lected substrate temperatures and at a growth rate of 0.2–0.3 ꢃs
.
Evaporated films were 500 ꢃ thick (as determined by a calibrated in situ
quartz crystal monitor). For FET device fabrication, top-contact electro-
1
998, 10, 93.
ꢁ
5
des (500 ꢃ) were deposited by evaporating gold (pressure <10 Torr);
channel dimensions were 50/100 mm (L) by 5.0 mm (W). The capacitance
5] a) H. Meng, J. Zheng, A. J. Lovinger, B.-C. Wang, P. G. Van Patten,
Z. Bao, Chem. Mater. 2003, 15, 1778; b) M. Mushrush, A. Facchetti,
M. Lefenfeld, H. E. Katz, T. J. Marks, J. Am. Chem. Soc. 2003, 125,
ꢁ
8
ꢁ2
of the insulator is 2ꢄ10 Fcm for 300 nm SiO
2
. TFT device measure-
ments were carried out in a customized high-vacuum probe station
9
414; c) A. Facchetti, M. Mushrush, H. E. Katz, T. J. Marks, Adv.
ꢁ
6
pumped down to 8ꢄ10 Torr before being backfilled with argon or air.
Coaxial and/or triaxial shielding was incorporated into Signatone probe
stations to minimize the noise level. TFT characterization was performed
with a Keithly 6430 sub-femtoamp meter and a Keithly 2400 source
meter, operated by a locally written Labview program and GPIB commu-
nication. Thin films were analyzed by using wide-angle X-ray film diffrac-
tometry (WAXRD) on a Rikagu ATX-G using standard q-2q techniques,
with monochromated CuKa radiation. All q-2q scans were calibrated in
situ using the Si (100) substrate reflections.
Mater. 2003, 15, 33; d) D. M. Delongchamp, S. Sambasivan, E. K. L.
Fischer, P. Chang, A. R. Murphy, J. M. J. Frꢈchet, V. Subramanian,
Adv. Mater. 2005, 17, 2340.
6] a) M. Pasini, S. Destri, W. Porzio, C. Botta, U. Giovanella, J. Mater.
Chem. 2003, 13, 807; b) M. Suzuki, M. Fukuyama, Y. Hori, S. Hotta,
J. Appl. Phys. 2002, 91, 5706.
[
[
7] D. Pisignano, M. Anni, G. Gigli, R. Cingolani, M. Zavelani-Rossi,
G. Lanzani, G. Barbarella, L. Favaretto, Appl. Phys. Lett. 2002, 81,
3
534.
Computational methods: Density functional theory (DFT) calculations
were carried with the Gaussian 03 program running on an SGI Origin
[8] a) L. Torsi, A. J. Lovinger, B. Crone, T. Someya, A. Dodabalapur,
H. E. Katz, A. Gelperin, J. Phys. Chem. B 2002, 106, 12563; b) J.
Roncali, J. Mater. Chem. 1999, 9, 1875.
[9] K. Hara, M. Kurashige, Y. Dan-Oh, C. Kasada, A. Shinpo, S. Suga,
K. Sayama, H. Arakawa, New J. Chem. 2003, 27, 783.
[83]
2
000 supercomputer. Becke’s three-parameter exchange functional com-
[
84]
bined with the LYP correlation functional (B3LYP) was used.
It is
known that the B3LYP functional yields similar geometries for medium-
[
85]
sized molecules as do MP2 calculations with the same basis sets. More-
[10] G. Barbarella, Chem. Eur. J. 2002, 8, 5072.
5036
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 5023 – 5039

Organic Field-Effect Transistors
FULL PAPER
[
[
11] C. Xia, X. Fan, J. Loklin, R. C. Advincula, Org. Lett. 2002, 4, 2067.
12] B. Jousselme, P. Blanchard, E. Levillain, J. Delaunay, M. Allain, P.
Richomme, D. Rondeau, N. Gallego-Planas, J. Roncali, J. Am.
Chem. Soc. 2003, 125, 1363.
Lucas, M. Heeney, I. McCulloch, J. E. Northrup, Adv. Mater. 2007,
19, 833.
[24] a) A. Facchetti, M.-H. Yoon, T. J. Marks, Adv. Mater. 2005, 17, 1705;
b) J. Veres, S. Ogier, G. Lloyd, D. de Leeuw, Chem. Mater. 2004, 16,
4543; c) D. M. Delongchamp, B. M. Vogel, Y. Jung, M. C. Gurau,
C. A. Richter, O. A. Kirillov, J. Obrzut, D. A. Fischer, S. Sambasi-
van, L. J. Richter, E. K. Lin, Chem. Mater. 2005, 17, 5610.
[25] a) H. Fukumoto, A. Kumagai, Y. Fujiwara, H. Koinuma, T. Yama-
moto, Heterocycles 2006, 68, 1349; b) T. Yasuda, Y. Sakai, S. Arama-
ki, T. Yamamoto, Chem. Mater. 2005, 17, 6060; c) G. Sonmez,
C. K. F. Shen, Y. Rubin, F. Wudl, Adv. Mater. 2005, 17, 897; d) L. M.
Campos, A. Tontcheva, S. Gunes, G. Sonmez, H. Neugebauer, N. S.
Sariciftci, F. Wudl, Chem. Mater. 2005, 17, 4031; e) R. D. Champion,
K.-F. Cheng, C.-L. Pai, W.-C. Chen, S. A. Jenekhe, Macromol. Rapid
Commun. 2005, 26, 1835; f) A. K. Agrawal, S. A. Jenekhe, Macro-
molecules 1993, 26, 895; g) Y. Cui, X. Zhang, S. A. Jenekhe, Macro-
molecules 1999, 32, 3824; h) M. Svensson, F. Zhang, S. C. Veenstra,
W. J. H. Verhees, J. C. Hummelen, J. M. Kroon, O. Inganꢆs, M. R.
Andersson, Adv. Mater. 2003, 15, 988.
[
13] a) S. R. Forrest, Nature 2004, 428, 911; b) M. Muccini, Nat. Mater.
2
006, 5, 605.
[
14] For example, see: a) A. Kraft, A. C. Grimsdale, A. B. Holmes,
Angew. Chem. 1998, 110, 416; Angew. Chem. Int. Ed. 1998, 37, 42;
b) U. Mitschke, P. Bauerle, J. Mater. Chem. 2000, 10, 1471; c) A. J.
Heeger, Solid State Commun. 1998, 107, 673; d) A. P. Kulkarmi, C. J.
Tonzola, A. Babel, S. A. Jenekhe, Chem. Mater. 2004, 16, 4556;
e) M. M. Richter, Chem. Rev. 2004, 104, 3003; f) N. R. Armstrong,
R. M. Wightman, E. M. Gross, Annu. Rev. Phys. Chem. 2001, 52,
3
91; g) F. Fungo, K.-T. Wong, S.-Y. Ku, Y.-Y. Hung, A. J. Bard, J.
Phys. Chem. B 2005, 109, 3984; h) H. Yan, P. Lee, N. R. Armstrong,
A. Graham, G. A. Evmenenko, P. Dutta, T. J. Marks, J. Am. Chem.
Soc. 2005, 127, 3172; i) I. F. Perepichka, D. F. Perepichka, H. Meng,
F. Wudl, Adv. Mater. 2005, 17, 2281.
[
15] a) G. Barbarella, L. Zanelli, A. Gigli, M. Mazzeo, M. Anni, A. Bon-
gini, Adv. Funct. Mater. 2005, 15, 664; b) D. Wasserberg, P. Marsat,
S. C. J. Meskers, R. A. J. Janseen, D. Beljonne, J. Phys. Chem. B
[
26] Handbook of Organic Conductive Molecules and Polymers, Vol. 2–4
Ed.: H. S. Nalva), Wiley, New York, 1997.
(
2005, 109, 4410; c) M. Rubio, M. Merchꢀn, E. Ortꢉ, ChemPhysChem
2005, 6, 1357; d) A. Kçhler, D. Beljonne, Adv. Funct. Mater. 2004,
14, 11; e) F. Garnier, Acc. Chem. Res. 1999, 32, 209; f) J. Cornil,
[
27] a) J. Krajcovic, G. Cik, D. Vꢈgh, F. Sersen, Synth. Met. 1999, 105, 79;
b) G. Cꢉk, J. Krajcovic, P. Veis, D. Vꢈgh, F. Sersen, Synth. Met. 2001,
1
18, 111.
D. A. dos Santos, X. Crispin, R. Silbey, J. L. Brꢈdas, J. Am. Chem.
Soc. 1998, 120, 1289; g) R. S. Becker, J. Seixas de Melo, A. L. Maca-
nita, F. Elisei, J. Phys. Chem. 1996, 100, 18683; h) D. Grebner, H.
Helbig, S. Rentsch, J. Phys. Chem. 1995, 99, 16991; i) A. J. Janssen,
L. Smilowitz, S. N. Sariciftci, D. Moses, J. Chem. Phys. 1994, 101,
[
[
28] A. I. Khodair, P. Bertrand, Tetrahedron 1988, 44, 4859.
29] E. E. Schweizer, A. L. Rheingold, M. Bruch, J. Org. Chem. 1993, 58,
4
339.
30] a) R. Centore, C. Garzillo, J. Chem. Soc. Perkin Trans. 1 1997, 22,
9; b) P. Espinet, J. Etxebarria, M. Marcos, J. Pꢈres, A. Remꢂn, J. L.
[
7
1
787; j) S. Hotta, K. Waragai, J. Phys. Chem. 1993, 97, 7427; k) P.
Serrano, Angew. Chem. 1989, 101, 1076; Angew. Chem. Int. Ed.
Engl. 1989, 28, 1065.
Garcia, J. M. Pernaut, P. Happiot, V. Wintgens, P. Valat, F. Garnier,
D. Delabouglise, J. Phys. Chem. 1993, 97, 513; l) D. Birnbaum, D.
Fichou, B. Kohler, J. Chem. Phys. 1992, 96, 165.
[
[
31] H. S. Nalwa, A. Kakatu, A. Mukoh, J. Appl. Phys. 1993, 73, 4743.
32] M. X. Hong, H. E. Katz, A. J. Lovinger, B.-C. Wang, K. Raghava-
chari, Chem. Mater. 2001, 13, 4686.
[
[
16] M. Ichikawa, R. Hibino, M. Inoue, T. Haritani, S. Hotta, K.-I. Araki,
T. Koyama, Y. Taniguchi, Adv. Mater. 2005, 17, 2073.
17] a) M. Ichikawa, R. Hibino, M. Inoue, T. Haritani, S. Hotta, T.
Koyama, Y. Taniguchi, Adv. Mater. 2003, 15, 213; b) M. Nagawa, R.
Hibino, S. Hotta, H. Yanagi, M. Ichikawa, T. Koyama, Y. Taniguchi,
Appl. Phys. Lett. 2002, 80, 544.
18] a) H. Meng, J. Zheng, A. J. Lovinger, B.-C. Wang, G. P. Van Patten,
Z. Bao, Chem. Mater. 2003, 15, 1778; b) K. T. Wong, C.-F. Wang,
C. H. Chou, Y. O. Su, G.-H. Lee, S. M. Peng, Org. Lett. 2002, 4,
[
33] a) Y. Zhu, R. D. Champion, S. A. Jenekhe, Macromolecules 2006, 39,
8
712; b) R. D. Champion, K.-F. Cheng, C.-L. Pai, W.-C. Chen, S. A.
Jenekhe, Macromol. Rapid Commun. 2005, 26, 1835; c) T. Yamamo-
to, T. Yasuda, Y. Sakai, S. Aramaki, Macromol. Rapid Commun.
2
005, 26, 1214.
[
[
[
[
34] a) A. Abbotto, S. Bradamante, A. Facchetti, G. A Pagani, J. Org.
Chem. 2002, 67, 5753; b) K. B. Wiberg, T. P. Lewis, J. Am. Chem.
Soc. 1970, 92, 7154.
35] a) S. Sharma, D. Lacey, P. Wilson, Liq. Cryst. 2003, 30, 461; b) P.
Liu, H. Nakano, Y. Shirota, Liq. Cryst. 2001, 28, 581; c) R. Azumi,
G. Gçtz, P. Bꢆuerle, Synth. Met. 1999, 101, 544.
36] a) A. Facchetti, M.-H. Yoon, C. L. Stern, G. R. Hutchison, M. A.
Ratner, T. J. Marks, J. Am. Chem. Soc. 2004, 126, 13480; b) A. Zen,
M. Saphiannikova, D. Neher, J. Grenzer, S. Grigorian, U. Pietsch, U.
Asawapirom, S. Janietz, U. Scherf, I. Lieberwirth, G. Wegner, Mac-
romolecules 2006, 39, 2162.
4
429.
[
19] a) T. Noda, I. Imae, N. Noma, Y. Shirota, Adv. Mater. 1997, 9, 239;
b) T. Noda, H. Ogawa, N. Noma, Y. Shirota, Adv. Mater. 1997, 9,
7
1
20; c) T. Noda, H. Ogawa, N. Noma, Y. Shirota, Appl. Phys. Lett.
997, 70, 699; d) T. Noda, H. Ogawa, N. Noma, Y. Shirota, J. Mater.
Chem. 1999, 9, 2177.
[
20] a) Y. Shirota, M. Kinoshita, T. Noda, K. Okumoto, O. Takahiro, J.
Am. Chem. Soc. 2000, 122, 11021; b) H. Doi, M. Kinoshita, K.
Kumoto, Y. Shirota, Chem. Mater. 2003, 15, 1080.
21] a) H. Akimichi, K. Waragai, S. Hotta, H. Kano, H. Sakaki, Appl.
Phys. Lett. 1991, 58, 1500; b) F. Garnier, A. Yassar, R. Hajlaoui, G.
Horowitz, F. Deloffre, B. Servet, S. Ries, P. Alnot, J. Am. Chem.
Soc. 1993, 115, 8716; c) H. E. Katz, L. Torsi, A. Dodabalapur, Chem.
Mater. 1995, 7, 2235; d) H. E. Katz, J. Mater. Chem. 1997, 7, 369.
22] a) A. Tsumara, H. Koezuka, T. Ando, Appl. Phys. Lett. 1986, 49,
[
[
37] J. Jagur-Grodzinski, Polym. Adv. Technol. 2002, 13, 615.
38] a) W. Domagala, M. Laokowski, S. Guillerez, G. Bidan, Electrochim.
Acta 2003, 48, 2379; b) K. Meerholz, J. Heinze, Electrochim. Acta
[
1
996, 41, 1839.
[
[
39] J. Jagur-Grodzinski, Polym. Adv. Technol. 2002, 13, 615.
40] a) W. Domagala, M. Laokowski, S. Guillerez, G. Bidan, Electrochim.
Acta 2003, 48, 2379; b) K. Meerholz, J. Heinze, Electrochim. Acta
1996, 41, 1839.
[
[
1
210; b) A. Assadi, C. Svensson, M. Willander, O. Inganꢆs, Appl.
Phys. Lett. 1988, 53, 195.
23] a) C. D. Dimitrakopoulos, B. K. Furman, T. Graham, S. Hedge, S.
Purushothaman, Synth. Met. 1998, 92, 47; b) H. E. Katz, A. J. Lo-
vinger, J. G. Laquindanum, Chem. Mater. 1998, 10, 457; c) H. E.
Katz, J. G. Laquindanum, A. J. Lovinger, Chem. Mater. 1998, 10,
[41] M.-H. Yoon, C. Kim, A. Facchetti, T. J. Marks, J. Am. Chem. Soc.
2006, 128, 12851.
[42] A. Abbotto, S. Bradamante, A. Facchetti, G. A. Pagani, J. Org.
Chem. 2002, 67, 5753.
6
33; d) M. Halik, H. Klauk, U. Zschieschang, G. Schmid, S. Pono-
[43] a) J. Roncali, Chem. Rev. 1997, 97, 173; b) J. Casado, L. L. Miller,
K. R. Mann, T. M. Pappenfus, V. Hernꢀndez, J. T. Lꢂpez Navarrete,
J. Phys. Chem. B 2002, 106, 3597.
[44] a) J. Casado, R. Ponce Ortiz, M. C. Ruiz Delgado, V. Hernꢀndez,
J. T. Lꢂpez Navarrete, J.-M. Raimundo, P. Blanchard, M. Allain, J.
Phys. Chem. B 2005, 109, 16616; b) M. C. Ruiz Delgado, V. Hernꢀn-
marenko, S. Kirchmeyer, W. Weber, Adv. Mater. 2003, 15, 917;
e) D. M. Delongchamp, B. M. Vogel, Y. Jung, M. C. Gurau, C. A.
Richter, O. A. Kirillov, J. Obrzut, D. A. Fischer, S. Sambasivan, L. J.
Richter, E. K. Lin, Chem. Mater. 2005, 17, 5610; f) P. M. Delong-
champ, R. J. Kline, E. K. Lin, D. A. Fischer, L. J. Richter, L. A.
Chem. Eur. J. 2009, 15, 5023 – 5039
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5037

J. T. Lꢂpez Navarrete, A. Facchetti, T. J. Marks et al.
dez, J. Casado, J. T. Lꢂpez Navarrete, J. M. Raimundo, P. Blanchard,
J. Roncali, Chem. Eur. J. 2003, 9, 3670.
Garnier, Adv. Mater. 1998, 10, 382; d) T. Siegrist, C. Kloc, R. A.
Laudise, H. E. Katz, R. C. Haddon, Adv. Mater. 1998, 10, 379; a4:
e) G. Horowitz, B. Bachet, A. Yassar, P. Lang, F. Demanze, J.-L.
Fave, F. Garnier, Chem. Mater. 1995, 7, 1337.
[
[
[
[
45] S. Hamai, F. Hirayama, J. Phys. Chem. 1983, 87, 83.
46] B. Xu, S. Holdcroft, Macromolecules 1993, 26, 4457.
47] M. Zander, Z. Naturforsch. A: Phys. Sci. 1989, 44, 1116.
48] J. R. Larowicz, Principles of fluorescence spectroscopy; Kluwer Aca-
demic Press, New York, 1999.
49] M. Pope, C. E. Swenberg, Electronic Processes in Organic Crystals,
Oxford University Press, New York, 1982.
50] a) J. Casado, V. Hernꢀndez, S. Hotta, J. T. Lꢂpez Navarrete, Synth.
Met. 2001, 119, 305; b) C. Moreno Castro, M. C. Ruiz Delgado, V.
Hernꢀndez, S. Hotta, J. Casado, J. T. Lꢂpez Navarrete, J. Chem.
Phys. 2002, 116, 10419.
[69] Handbook of Chemistry and Physics, CRC Press, Boca Raton, 1995.
[70] The energy level of the normal hydrogen electrode (NHE) is
ꢁ4.6 eV below the vacuum level (see A. J. Bard, L. R. Faulkner,
Electrochemical Methods —Fundamentals and Applications, Wiley,
New York, 1984). The oxidation potential of ferrocene is 0.67 V
versus the NHE (see J.-F. Wang, Y. Kawabe, S. E. Shaheen, M. M.
Morrell, G. E. Jabbour, P. A. Lee, J. Anderson, N. R. Armstrong, B.
Kippelen, E. A. Mash, N. Peyghambarian, Adv. Mater. 1998, 10,
230). The HOMO/LUMO energies were determined using the ferro-
cene calibration against the vacuum energy level. The terms HOMO
and LUMO are a convenient short hand since the actual energies
correspond to states, not levels.
[
[
[
51] A. Yassar, G. Horowitz, P. Valat, V. Wintgens, M. Hmyene, F. De-
loffre, P. Srivastava, P. Lang, F. Garnier, J. Phys. Chem. 1995, 99,
9
155.
[
[
52] F. Garnier, Acc. Chem. Res. 1999, 32, 209.
53] M. Muccini, E. Lunedei, C. Taliani, D. Beljonne, J. Cornil, J. L.
Brꢈdas, J. Chem. Phys. 1998, 109, 10513.
[71] a) R. W. Godby, M. Schluter, L. J. Sham, Phys. Rev. B 1988, 37,
10159; b) R. Stowasser, R. Hoffmann, J. Am. Chem. Soc. 1999, 121,
3414.
[
[
54] H. Meng, J. Zheng, A. J. Lovinger, B.-C. Wang, P. G. Van Patten, Z.
Bao, Chem. Mater. 2003, 15, 1778.
55] a) V. Hernandez, J. Casado, F. J. Ramirez, G. Zotti, S. Hotta, J. T.
Lopez Navarrete, J. Chem. Phys. 1996, 104, 9271; b) J. Casado, V.
Hernandez, S. Hotta, J. T. Lopez Navarrete, J. Chem. Phys. 1998,
[72] M. J. Kamlet, J. L. M. Abboud, M. H. Abraham, R. W. Taft, J. Org.
Chem. 1983, 48, 2877.
[73] M. Jonsson, A. Houmam, G. Jocys, D. D. M. Wayner, J. Chem. Soc.
Perkin Trans. 2 1999, 425.
[74] V. Hernꢀndez, J. T. Lꢂpez Navarrete, J. Chem. Phys. 1994, 101, 1369.
[75] a) R. S. Becker, J. Seixas de Melo, A. L. Macanita, F. Elisei, Pure
Appl. Chem. 1995, 67, 9; b) R. S. Becker, J. Seixas de Melo, A. L.
Macanita, F. Elisei, J. Phys. Chem. 1996, 100, 18683; c) D. Beljonne,
J. Cornil, R. H. Friend, R. A. J. Janssen, J. L. Brꢈdas, J. Am. Chem.
Soc. 1996, 118, 6453; d) J. Seixas de Melo, L. M. Silva, L. G. Arnaut,
R. S. Becker, J. Chem. Phys. 1999, 111, 5427.
[76] K. Schmidt, S. Brovelli, V. Coropceanu, D. Beljonne, J. Cornil, C.
Bazzini, T. Caronna, R. Tubino, F. Meinardi, Z. Shuai, J. L. Brꢈdas,
J. Phys. Chem. A 2007, 111, 10490.
[77] A. Facchetti, M. Mushrush, M.-H. Yoon, G. R. Hutchison, M. A.
Ratner, T. J. Marks, J. Am. Chem. Soc. 2004, 126, 13874.
[78] M. Malagoli, J. L. Bredas, Chem. Phys. Lett. 2000, 327, 13.
[79] Y. Wei, Y. Yang, J.-M. Yeh, Chem. Mater. 1996, 8, 2659.
[80] A. Facchetti, M.-H. Yoon, C. L. Stern, G. R. Hutchison, M. A.
Ratner, T. J. Marks, J. Am. Chem. Soc. 2004, 126, 13480.
[81] S. Dufresne, G. S. Hanan, W. G. Skene, J. Phys. Chem. B 2007, 111,
11407.
1
09, 10419; c) C. Moreno Castro, M. C. Ruiz Delgado, V. Hernan-
dez, S. Hotta, J. Casado, J. T. Lopez Navarrete, J. Chem. Phys. 2002,
16, 10419; d) C. Moreno Castro, M. C. Ruiz Delgado, V. Hernan-
dez, Y. Shirota, J. Casado, J. T. Lopez Navarrete, J. Phys. Chem. B
002, 106, 7163.
1
2
[
56] T. Kojima, J-I. Nishida, S. Tokito, H. Tada, Y. Yamashita, Chem.
Commun. 2007, 1430.
[
57] a) T. Kojima, J-I. Nishida, S. Tokito, H. Tada, Y. Yamashita, Chem.
Commun. 2007, 1430; b) T. Yasuda, Y. Sakai, S. Aramaki, T. Yama-
moto, Chem. Mater. 2005, 17, 6060; c) B. Ackers, A. J. Blake, S. J.
Hill, P. Hubberstey, Acta Crystallogr. Sect. C 2002, 58, o640; d) F.
Iwasaki, K.-Y. Akiba, Acta Crystallogr. Sect. C 1987, 43, 2338; e) R.
Ghosh, S. H. Simonsen, Acta Crystallogr. Sect. C 1993, 49, 1031.
58] a) J. Ma, S. Li, Y. Jiang, Macromolecules 2002, 35, 1109; b) M.
Springborg, K. Schmidt, H. Meider, L. De Maria, Organic Electron-
ic Materials: Conjugated Polymers and Low Molecular Weight Or-
ganic Solids, Vol. 41 (Ed.: G. Grosso), Springer, Berlin, 2001, p. 39;
c) J.-L. Brꢈdas, J. Cornil, D. Beljonne, D. A. Dos Santos, Z. Shuai,
Acc. Chem. Res. 1999, 32, 267; d) S. Stafstrçm, J.-L. Bredas, Theo-
chem 1989, 183–203, 393; e) G. R. Hutchison, M. A. Ratner, T. J.
Marks, J. Phys. Chem. A 2002, 106, 10596.
[
[82] a) A. J. Goodman, S. P. Stanforth, B. Tarbit, Tetrahedron 1999, 55,
15067.
[83] Gaussian 03, Revision B.04, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Pitts-
burgh PA, 2003
[
59] W. J. Hehre, A Guide to Molecular Mechanics and Quantum Chemi-
cal Calculations, Wavefunction, Irvine, 2003.
[
60] a) F. Grein, THEOCHEM 2003, 624, 23; b) A. A. Mohamed, Int. J.
Quantum Chem. 2000, 79, 367; c) F. Grein, J. Phys. Chem. A 2002,
1
06, 3823.
[
[
[
[
[
[
[
61] a) O. V. Prezhdo, Adv. Mater. 2002, 14, 597; b) J. Wang, B. G. John-
son, R. J. Boyd, L. A. Eriksson, J. Phys. Chem. 1996, 100, 6317.
62] J. C. Rienstra-Kiracofe, G. S. Tschumper, H. F. Schaefer, S. Nandi,
G. B. Ellison, Chem. Rev. 2002, 102, 231.
63] C. G. Zhan, J. A. Nichols, D. A. Dixon, J. Phys. Chem. A 2003, 107,
4
184.
64] C. G. Zhan, J. A. Nichols, D. A. Dixon, J. Phys. Chem. A 2003, 107,
184.
65] J. Muscat, A. Wander, N. M. Harrison, Chem. Phys. Lett. 2001, 342,
97.
4
3
66] G. R. Hutchison, A. Facchetti, M. A. Ratner, T. J. Marks, unpublish-
ed results.
67] a) D. Fichou, J. Mater. Chem. 2000, 10, 571; b) T. Siegrist, C. Kloc,
R. A. Laudise, H. E. Katz, R. C. Haddon, Adv. Mater. 1998, 10, 379;
c) L. Antolini, G. Horowitz, F. Kouki, F. Garnier, Adv. Mater. 1998,
[84] A. D. Becke, J. Chem. Phys. 1993, 98, 1372.
[85] a) P. J. Stephens, F. J. Devlin, F. C. F. Chabalowski, M. J. Frisch, J.
Phys. Chem. 1994, 98, 11623; b) J. J. Novoa, C. Sosa, J. Phys. Chem.
1995, 99, 15837.
1
0, 382.
[86] a) A. P. Scott, L. Radom, J. Phys. Chem. 1996, 100, 16502; b) G.
Rauhut, P. Pulay, J. Phys. Chem. 1995, 99, 3093.
[87] M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S. Gordon,
D. J. Defrees, J. A. Pople, J. Chem. Phys. 1982, 77, 3654.
[
68] a6: a) G. J. Visser, G. J. Heeres, J. Wolters, A. Vos, Acta Crystallogr.,
Sect. B 1968, 24, 467; b) M. Pelletier, F. Brisse, Acta Crystallogr.
Sect. A 1994, 50, 1942; a5: c) L. Antolini, G. Horowitz, F. Kouki, F.
5038
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 5023 – 5039

Organic Field-Effect Transistors
FULL PAPER
[
88] a) E. Runge, E. K. U. Gross, Phys. Rev. Lett. 1984, 52, 997;
b) E. K. U. Gross, W. Kohn, Adv .Quantum Chem. 1990, 21, 255;
c) Density Functional Theory (Eds.: E. K. U. Gross, R. M. Driezler),
Plenum Press, New York, 1995, p. 149; d) M. E. Casida, Recent Ad-
vances in Density Functional Methods, Part I (Ed.: D. P. Chong),
World Scientific, Singapore, 1995, p. 115.
[89] J. B. Foresman, M. Head-Gordon, J. A. Pople, M. J. Frish, J. Phys.
Chem. 1992, 96, 135.
Received: November 20, 2008
Published online: February 27, 2009
Chem. Eur. J. 2009, 15, 5023 – 5039
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5039