Dumbbell-shaped dinuclear iridium complexes and their application to light-emitting electrochemical cells

Article abstract of DOI:10.1002/chem.201000600

A novel family of dumbbellshaped dinuclear complexes in which an oligophenyleneethynylene spacer is linked to two heteroleptic iridium(III) complexes is presented. The synthesis, as well as the electrochemical and photophysical characterization of the new complexes, is reported. The experimental results are interpreted with the help of density functional theory calculations. From these studies we conclude that the lowest triplet excited state corresponds to a ³π-π* state located on the conjugated spacer. The presence of this state below the ³MLCnT/ ³LLCT emitting states of the end-capping Ir^III complexes explains the low quantum yields observed for the dinuclear complexes (one order-of-magnitude less) with respect to the mononuclear complexes. The potential application of the novel dinuclear complexes in optoelectronic devices has been tested by using them as the primary active component in double-layer light-emitting electrochemical cells (LECs). Although the luminance levels are low, the external quantum efficiency suggests that a near-quantitative internal electron-tophoton conversion occurs in the device. This indicates that the emission inside the device is highly optimized and that the self-quenching associated with the high concentration of the complex in the active layer is minimized.

Full text of DOI:10.1002/chem.201000600

FULL PAPER
DOI: 10.1002/chem.201000600
Dumbbell-Shaped Dinuclear Iridium Complexes and Their Application to
Light-Emitting Electrochemical Cells
Rubꢀn D. Costa,[a] Gustavo Fernꢁndez,[b] Luis Sꢁnchez,[b] Nazario Martꢂn,*[b]
Enrique Ortꢂ,*[a] and Henk J. Bolink*[a]
Dedicated to Professor Josꢀ Barluenga on the occasion of his 70th birthday
Abstract: A novel family of dumbbell-
shaped dinuclear complexes in which
an oligophenyleneethynylene spacer is
this state below the 3MLCT/3LLCT
emitting states of the end-capping IrIII
complexes explains the low quantum
yields observed for the dinuclear com-
plexes (one order-of-magnitude less)
with respect to the mononuclear com-
plexes. The potential application of the
novel dinuclear complexes in optoelec-
tronic devices has been tested by using
them as the primary active component
in double-layer light-emitting electro-
chemical cells (LECs). Although the
luminance levels are low, the external
linked to two heteroleptic iridiumACTHNUTRGNEUG(N III)
complexes is presented. The synthesis,
as well as the electrochemical and pho-
tophysical characterization of the new
complexes, is reported. The experimen-
tal results are interpreted with the help
of density functional theory calcula-
tions. From these studies we conclude
that the lowest triplet excited state cor-
responds to a 3p–p* state located on
the conjugated spacer. The presence of
quantum efficiency suggests that a
near-quantitative internal electron-to-
photon conversion occurs in the device.
This indicates that the emission inside
the device is highly optimized and that
the self-quenching associated with the
high concentration of the complex in
the active layer is minimized.
Keywords: electroluminescence
iridium light-emitting electro-
chemical cells · quenching pathways
·
·
Introduction
(OLEDs).[2–9] A special type of building block for multifunc-
tional materials is organometallic complexes and, more spe-
Multifunctional materials are designed to accomplish multi-
ple performance objectives in a single system.[1] These objec-
tives are usually related with optical, electrochemical,
charge transport, and magnetic properties of the materials
and with their integration into devices, such as organic pho-
tovoltaics (PVs) or organic light-emitting diodes
cifically, those complexes based on an iridiumACTHNGUTRENNU(G III) metal
core. Iridium complexes have been widely utilized in optoe-
lectronics, owing to their relevant luminescent properties,
such as high emission quantum yields, stability, long excited-
state lifetimes, and easy tunability of the emission color.[10–14]
One of the most outstanding applications of ionic transi-
tion-metal complexes (iTMCs) is their use in the fabrication
of a new type of solid-state electroluminescent devices, the
so-called light-emitting electrochemical cells (LECs). LECs
mainly consist of one active layer of a RuII- or IrIII-based
iTMC,[15,16] in which the charge of the iTMC is compensated
with small mobile anions, such as hexafluorophosphate
[a] R. D. Costa, Prof. Dr. E. Ortꢀ, Dr. H. J. Bolink
Instituto de Ciencia Molecular (ICMol)
Universidad de Valencia
P.O. Box 22085, 46071 Valencia (Spain)
Fax : (+34)963543273
E-mail: enrique.orti@uv.es
ꢀ
(PF6 ). The presence of mobile ions leads to the formation
henk.bolink@uv.es
of ionic junctions when an external electric field is applied,
which lower the barriers for hole and electron injection and
make LECs independent of the work function of the elec-
trode material.[9,17–19] Therefore, air-stable electrodes, such as
gold, silver, or aluminum, can be used in LECs, which
makes its fabrication process cheaper and simpler compared
to OLEDs.
[b] Dr. G. Fernꢁndez, Dr. L. Sꢁnchez, Prof. Dr. N. Martꢀn
Departamento de Quꢀmica Orgꢁnica
Facultad de Ciencias Quꢀmicas
Universidad Complutense, 28040 Madrid (Spain)
Fax : (+34)913944103
E-mail: nazmar@quim.ucm.es
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000600.
Chem. Eur. J. 2010, 16, 9855 – 9863
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9855

LEC devices showing high stabilities[20–23] and a wide
range of emission colors, including white,[12,24–30] have been
reported. However, moderate efficiencies are generally ob-
served because of the high concentration of the iTMCs in
the active layer that leads to an efficient self-quenching.[31,32]
A useful strategy to enhance the efficiency of the device is
to increase the distance between the iTMCs by attaching
bulky groups in the periphery of the complexes.[25,32–35] This
strategy, however, has an important limitation, as the iTMCs
are responsible for all the processes, such as charge injec-
tion, charge transport by hopping, and emission. Therefore,
if the distances between the ionic complexes in the active
layer are greater than the hopping distances of the charge
carriers, the LEC device does not work properly.[25,34]
2,2’-bipyridine) and the p-conjugated bridge 6. The electro-
luminescence of LEC devices, built up using compounds 1a
and 1b as the active components, is also presented and is ra-
tionalized in terms of the photophysical properties of the
dumbbell-shaped dinuclear complexes.
Results and Discussion
Synthesis: The synthesis of the dinuclear complexes 1a and
1b was achieved by reacting the corresponding cyclometa-
lated iridium chloro-bridged dimer 7[36,37] with the p-conju-
gated spacer 6 in ethylene glycol under argon during 16 h.
Compound 6 was prepared in a two-step synthetic proce-
dure, starting from the nucleophilic substitution of the bro-
mine atoms of commercially available 1-(bromomethyl)-4-
iodobenzene (3) by the carbanion of dimethylbipyridine 2,
generated in situ upon addition of LDA at ꢀ788C. The sub-
sequent twofold Sonogashira cross-coupling reaction be-
tween bipyridine 4 and 1,4-di-
We will show a plausible strategy to control the separation
between the iTMCs by connecting two iridiumACTHNUTRGENUG(N III) com-
plexes by a p-conjugated spacer. Two novel dumbbell-
shaped dinuclear iridium complexes (compounds 1a and 1b
in Scheme 1) incorporating the oligophenyleneethynylene
ethynyl-2,5-bis-
AHCTUNGTRENNUNG
(hexyloxy)benzene (5)[38] yields
6 in 48%. All compounds were
fully characterized through a
variety of spectroscopic tech-
niques (see the Experimental
Section for details).
Electrochemical features: The
electrochemical behavior of the
dinuclear complexes 1a and 1b
has been studied by cyclic vol-
tammetry in dichloromethane
at room temperature (Figure 1
and Table 1). Table 1 also in-
cludes the redox potentials
measured for the bridge 6 and
the mononuclear complexes 8a
and 8b, which have been inves-
tigated as reference com-
pounds.
The mononuclear complexes
8a and 8b present dissimilar
oxidation behaviors. Although
Scheme 1. Synthesis of the dumbbell-shaped dinuclear complexes.
both complexes show a reversi-
ble first oxidation process, the
spacer endowed with two bipyridyl units (6) have been syn-
thesized. The terminal bipyridyl units are used to link two
heteroleptic iridiumACHTUNGTRENNUNG(III) moieties for which the coordination
sphere is completed with two 2-phenylpyridine (ppy) (com-
pound 1a) or two 2-(2’,4’-difluorophenyl)pyridine (F2ppy)
cyclometalating ligands (compound 1b). The electrochemi-
cal and photophysical properties measured for the novel di-
nuclear compounds are discussed with the help of density
functional theory (DFT) calculations and in comparison
with those obtained for the mononuclear IrIII complexes [Ir-
potential at which this process appears is highly influenced
by the presence of the fluorine atoms on the ppy ligands
(Table 1). The fluorine substituents make complex 8b
(+1.58 V) a remarkably weaker reducing agent than com-
plex 8a (+1.12 V), and 8a presents a second irreversible ox-
idation wave at +1.74 V that is not observed for 8b. The
electron-withdrawing effect of the fluorine atoms, therefore,
increases the electrochemical E1oxid–E1 gap from 2.70 V in
red
8a to 3.07 V in 8b. In contrast to 8a and 8b, the two oxida-
tion processes observed for the dinuclear complexes 1a and
1b remain mainly unaffected upon fluorine substitution
ACHTUNGTRENNUNG(ppy)2bpy]ACHTUNGTRENNUNG[PF6] (8a) and [IrACHTUNGETRG(NNUN F2ppy)2bpy]ACHTNUGTRENUN[NG PF6] (8b) (bpy=
9856
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9855 – 9863

Dumbbell-Shaped Dinuclear Iridium Complexes
FULL PAPER
Table 2. Selected bond lengths (in ꢃ) calculated for complexes 8 and 1.
Selected bonds
8a
8b
1a
1b
ꢀ
Ir Cppy
2.024
2.083
2.218
2.002
2.083
2.210
2.023
2.083
2.219
2.002
2.083
2.205
ꢀ
Ir Nppy
ꢀ
Ir Nbpy
from 1a to 1b and evidence the contraction effect that the
electron-withdrawing fluorine groups attached to the ppy li-
gands have on the coordination sphere of the metal. A simi-
lar contraction is computed for 8a and 8b. As expected, the
fluorine groups do not affect the structure of the wire in
compounds 1. The central part of the wire consists of a p-
conjugated structure with phenyl rings connected by ethyny-
ꢀ
ꢁ
lene groups (C C and C C bonds of 1.42 and 1.22 ꢃ, re-
spectively). The oligophenyleneethynylene spacer is linked
ꢀ
to the iTMC moieties by saturated ethylene bridges (C C
bonds of 1.55 ꢃ).
Figure 1. Cyclic voltammograms of the dinuclear complexes 1 measured
in CH2Cl2 at room temperature (scan rate: 100 mVsꢀ1).
Displayed in Figure 2 are the electronic density contours
calculated for the HOMOs and the LUMOs of 1a, together
with those obtained for the mononuclear complex 8a and
Table 1. Redox properties of the dinuclear complexes 1, the spacer 6,
and the mononuclear complexes 8.
[b]
Compound[a]
E1
E1
E2
oxid
1
1
2
=
red
=
oxid
2
1a
1b
6
8a
8b
ꢀ1.67
ꢀ1.58
ꢀ1.81[c]
ꢀ1.58
ꢀ1.49
+0.99
+1.06
+1.00
+1.12
+1.58
+1.73
+1.70
+1.50
+1.74
–
[a] V versus Ag/Ag+, CH2Cl2 as solvent, GCE as working electrode, Pt
as counter-electrode, Bu4NClO4 (0.1m) as supporting electrolyte; scan
rate 100 mVsꢀ1. [b] Very broad and irreversible wave. [c] Very weak and
irreversible wave.
(Table 1) and no appreciable increase of the electrochemical
gap is detected (1a: 2.66 V, 1b: 2.64 V). These results sug-
gest that oxidation in complexes 1a and 1b does not take
place on the end-capping iTMCs and should involve the p-
conjugated spacer. Indeed, oxidation of 6 (+1.00 V) oc-
curs at similar potentials than in 1a (+0.99 V) and 1b
(+1.06 V).
Ground-state molecular and electronic structures: The mo-
lecular and electronic structures of the dinuclear complexes
1, the p-conjugated spacer 6, and the mononuclear com-
plexes 8 were investigated by performing DFT calculations
at the B3LYP/(6-31G**+LANL2DZ) level. To simplify the
calculations the hexyloxy groups of the wire were removed
and all the molecules were fully optimized within C2 symme-
try constraints (the Computational Details are described in
the Experimental Section).
Figure 2. Schematic diagram showing the electronic density contours
(0.03 ebohrꢀ3) and energies (in eV) calculated for the frontier molecular
orbitals of the mononuclear complex 8a (left), the dinuclear complex 1a
(center), and the central bridge 6 (right). H and L denote HOMO and
LUMO, respectively. Molecular orbital energies correspond to those cal-
culated in acetonitrile solution.
the spacer 6. The HOMO of 8a is composed of a mixture of
iridium dp orbitals (t2g) and phenyl p orbitals distributed
equally among the two ppy ligands, whereas the LUMO re-
sides on the bpy ligand (see Figure 2, left). As the fluorine
groups in 8b are introduced on the phenyl rings of the ppy
ligands, their electron-withdrawing effect stabilizes the
HOMO (ꢀ5.96 eV) in a higher degree than the LUMO
(ꢀ2.44 eV) with respect to the frontier molecular orbitals of
The molecular structures calculated for compounds 1 are
almost identical to those obtained for the central bridge and
the mononuclear complexes 8. In Table 2 the values comput-
ed for selected bond lengths of the end-capping iTMCs of
compounds 1 are collected, they show a near octahedral co-
ordination of the IrIII metal. The Ir Cppy and Ir Nbpy bond
lengths shorten by 0.022 and 0.008 ꢃ, respectively, in passing
ꢀ
ꢀ
Chem. Eur. J. 2010, 16, 9855 – 9863
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9857

N. Martꢀn, E. Ortꢀ, H. J. Bolink et al.
complex 8a. The relative stabilizations of the HOMO
(0.41 eV) and the LUMO (0.11 eV) justify the higher oxida-
tion potential (+0.46 V) and the lower reduction potential
(+0.09 V) measured for 8b in comparison to those obtained
for 8a (see Table 1). The HOMO–LUMO energy gap in-
creases by 0.30 eV in passing from 8a to 8b in good agree-
ment with the increase experimentally measured for the
electrochemical E1oxid–E1red gap (0.37 V).
Calculations show that the LUMO and LUMO+1 in
compounds 1 reside on the bpy ligands and correspond to
the in-phase and out-of-phase combination of the LUMOs
of the mononuclear complexes (Figure 2, centre). They are
degenerate and appear at slightly lower energies for 1b
(ꢀ2.42 eV) than for 1a (ꢀ2.35 eV), as is found for 8b and
8a and in agreement with the lower reduction potential
measured for 1b. The LUMO of the wire appears as the
LUMO+2 and is calculated 0.46 (1a) and 0.51 eV (1b)
higher in energy. In contrast to the LUMO, the HOMO in
compounds 1 is found to be located on the conjugated part
of the wire and corresponds to the p-HOMO of 6. This ex-
plains the similar oxidation po-
Figure 3. Absorption (open symbols) and photoluminescence (full sym-
bols) spectra of IrIII complexes 1a (diamonds), 8a (stars), 1b (squares),
and 8b (circles) in de-aerated acetonitrile (ꢂ10ꢀ6 m) and of the central
spacer 6 (black triangles) in o-dichlorobenzene (ꢂ10ꢀ6 m).
tentials recorded for com-
Table 3. Photophysical properties of the dinuclear complexes 1, the p-conjugated spacer 6, and the mononu-
clear complexes 8.
pounds 1 (Table 1), as fluorine
substitution has no significant
effect on the wire and the
energy of the HOMO remains
Compound
Absorption
Emission
Decay Dynamics[c]
t [ms][a] kr [sꢀ1
lmax [nm][a]
lmax [nm][a]
fr[a]
]
knr [sꢀ1
]
1a
1b
6
8a
8b
217, 270, 372
245, 305, 363
294, 367[b]
222, 270
588
9ꢄ10ꢀ3
7ꢄ10ꢀ3
0.84[b]
0.14
0.4
1.2
2.3ꢄ104
0.6ꢄ104
2.8ꢄ108
2.3ꢄ105
1.1ꢄ105
2.8ꢄ106
8.5ꢄ105
5.3ꢄ107
1.4ꢄ106
5.0ꢄ105
mostly
unaffected
(1a:
521
443[b]
589
3.0ꢄ10ꢀ3[b]
ꢀ5.55 eV, 1b: ꢀ5.58 eV). The
calculations support that reduc-
tion takes place on the bpy li-
gands of the end-capping
iTMCs, whereas the oxidation
implies the conjugated spacer.
This determines that, in con-
0.6
1.7
250, 296
512
0.18
[a] Measured in nitrogen-saturated acetonitrile (298 K, ꢂ10ꢀ6 m). [b] Measured in nitrogen-saturated o-DCB
(298 K, ꢂ10ꢀ6 m). [c] lexc =310 nm, fr and t are ꢃ10%.
trast with what is found for complexes 8a and 8b, the
HOMO–LUMO gap in compounds 1 remains mainly con-
stant (3.2 eV), in agreement with the electrochemical meas-
urements. The electronic properties calculated for the dinu-
clear complexes 1 suggest that the electronic communication
between the end-capping iTMCs and the central conjugated
structure of the spacer is rather weak.
the phenylpyridine and the bipyridine ligands.[11,12] Some
contribution of the oligophenyleneethynylene spacer is also
observed in this region for compound 1b. The low-intensity
bands that complexes 8 present in the l=350–400 nm
region, which are associated with the metal-to-ligand charge
transfer (1MLCT) transitions, are reinforced for the dinu-
clear complexes by the p–p* absorptions of the conjugated
spacer and a better-defined band emerges at lꢂ370 nm for
1a and 1b.
The photoluminescence of 6 is characterized by a poor-re-
solved structured band with maximum emission at l=
443 nm (Figure 3). It corresponds to an efficient fluores-
cence process with a high quantum yield (fr =0.84) and a
short excited-state lifetime (t=3 ns). In contrast, the photo-
luminescence of compounds 1 is described as a broad struc-
tureless band centered at l=588 nm (2.11 eV) and l=
521 nm (2.38 eV) for 1a and 1b, respectively. The shape of
the bands and the emission maxima recorded for com-
pounds 1 agree perfectly with those obtained for the respec-
tive mononuclear complexes 8a and 8b (see Figure 3 and
Table 3). The emission maximum blue-shifts in going from
8a to 8b by 77 nm as it is also observed for 1a and 1b.
Therefore, the nature of the emitting excited state should be
Photophysical properties and excited electronic states: Dis-
played in Figure 3 are the absorption and photolumines-
cence spectra recorded for the dinuclear complexes 1 to-
gether with those registered for the mononuclear complexes
8 and the p-conjugated spacer 6. Summarized in Table 3 are
the absorption and emission properties. Compound 6 exhib-
its two intense absorption bands at l=294 and 367 nm that
are assigned to the p–p* excitations of the central oligophe-
nyleneethynylene spacer on the basis of time-dependent
DFT (TD-DFT) calculations performed for 6 by using the
ground-state optimized structure. The absorption bands ob-
served for compound 1 in the high-energy region (l=250–
300 nm) correlate with the bands recorded for the mononu-
clear complexes 8. These bands are attributed to singlet
spin-allowed ligand-centered (1LC) transitions that imply
9858
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9855 – 9863

Dumbbell-Shaped Dinuclear Iridium Complexes
FULL PAPER
the same and the emission of compounds 1 can be clearly
ascribed to the phosphorescence of the end-capping iTMCs.
The mononuclear complexes 8 show a strong phosphores-
cence emission with moderate quantum yields (8a: 0.14, 8b:
0.18) and long excited-state lifetimes in the ms scale. In con-
trast, compounds 1 present very low quantum yields (1a: 9ꢄ
10ꢀ3, 1b: 7ꢄ10ꢀ3), whereas the excited-state lifetimes have
similar values (see Table 3). The photophysical parameters
obtained for compounds 1 yield radiative-rate constants one
order-of-magnitude lower than those determined for the
mononuclear complexes and non-radiative-rate constants
that approximately double those calculated for 8. This pho-
tophysical behavior suggests that compounds 1 should pres-
ent an additional non-radiative deactivation pathway com-
pared with complexes 8. To investigate the origin of the de-
activation pathway, transient absorption experiments in the
ms time domain were performed in combination with DFT
calculations.
The transient absorption spectra recorded for 1a and 8a
are compared in Figure 4 (Figure S1 in the Supporting Infor-
mation shows those recorded for 1b and 8b). Instantaneous
formation of broad stimulated emission features are ob-
served at l=600 and 520 nm for the mononuclear com-
plexes 8a and 8b, respectively. Their decay kinetics are ade-
quately fitted to mono-exponential decays with lifetimes of
0.6 (8a) and 1.7 ms (8b) and are assigned to the decay of the
lowest triplet state. Similar transient absorption spectra have
been observed for other ionic ruthenium and iridium com-
plexes.[36,37] DFT calculations including full-geometry optimi-
zation show that the lowest-energy triplet state of com-
pounds 8a and 8b result from the monoelectronic
Figure 4. Transition difference spectra obtained by means of laser flash-
photolysis experiments at l=310 nm in degassed acetonitrile solutions at
room temperature with low concentration (10ꢀ6 m) of complex 8a (top)
and compound 1a (bottom).
HOMO!LUMO
excitation
and implies a mixing of metal-
to-ligand and ligand-to-ligand
charge
transfer
(3MLCT/3LLCT). This mixed
nature is confirmed by the spin-
density distribution calculated
for 8a (Figure 5) that perfectly
matches the electronic distribu-
tion of the HOMO and the
LUMO.
The transient absorption
spectrum of 1a presents new
features if compared with that
of 8a (Figure 4). Photoexcita-
tion of compound 1a leads to
ms transient signals that include
a broad stimulated emission at
l=600 nm with mono-exponen-
tial decay of 0.4 ms and absorp-
tion bands centered at l=445
Figure 5. Schematic energy diagram showing the adiabatic energy difference between the T1 and T2 states
(DE) calculated for compounds 1a and 1b. The electronic nature of the excited states is illustrated by the spin
density contours (0.05 ebohrꢀ3) calculated for the T1 and T2 states of 1a. The spin density of the lowest triplet
of the free ligand 6 (top, left) and the mononuclear complex 8a (top, right) are included for comparison pur-
poses
and 520 nm. In compound 1b, the stimulated emission ap-
pears in the same region that the positive absorption band
at l=520 nm and only a low-intensity absorption band is
observed (see Figure S1 in the Supporting Information). For
compound 1a, the position of the stimulated emission (l=
600 nm) and the fitted decay lifetime (0.4 ms) confirm that
the emitting triplet state is the same than in 8a (l=600 nm,
0.6 ms). The absorption bands are fitted to a long mono-ex-
ponential decay of around 19 ms for both 1a and 1b. These
absorption bands are not observed for the mononuclear
Chem. Eur. J. 2010, 16, 9855 – 9863
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9859

N. Martꢀn, E. Ortꢀ, H. J. Bolink et al.
complexes and they should be ascribed to the excitation of a
low-energy triplet state that involves the spacer.
To describe the triplet excited-state manifold of com-
pounds 1, time-dependent DFT (TD-DFT) calculations
were performed for 1a and 1b on the ground-state fully-op-
timized geometries. For both compounds, the lowest-energy
triplet state (T1, 3B) results from the monoelectronic
HOMO!LUMO+2 excitation that only implies the wire
3
and corresponds to the p–p* HOMO!LUMO transition
of the oligophenyleneethynylene spacer (see Figure 2). The
next two triplets, T2 (3A) and T3 (3B), are degenerate and
originate in the transitions from the HOMOꢀ1 and
HOMOꢀ2 to the LUMO and LUMO+1. These states cor-
3
respond to the lowest-energy MLCT/3LLCT triplet of the
mononuclear complexes. T1 is calculated at 2.26 eV above
the ground state and is separated from T2 and T3 by 0.20
(1a) and 0.56 eV (1b). TD-DFT calculations therefore sug-
gest that the lowest-energy triplet of the dinuclear com-
Figure 6. Photoluminescence (full symbols) in acetonitrile and electrolu-
minescence (open symbols) spectra of compounds 1a (squares) and 1b
(triangles). The electroluminescence spectra are recorded at a constant
voltage of 3 V.
3
plexes 1 is the p–p* state of the wire.
To further investigate the nature of the lowest-energy
triplet state, the geometries of the T1 and T2 states of com-
pounds 1 were optimized under C2 symmetry restrictions.
The EL maxima are red-shifted by 10–20 nm, with respect
to the photoluminescence maxima, which is often observed
in LECs.[15,16]
3
After full-geometry relaxation, the p–p* triplet of the wire
continue to be the most stable and is calculated to lie 0.47
LEC devices based on compounds 1 show low emission
levels with luminance maxima below 10 cdmꢀ2. The low lu-
minance values result from the low emission quantum yield
observed for compounds 1, owing to the quenching of the
3
(1a) and 0.61 eV (1b) below the MLCT/3LLCT triplets of
the end-capping iTMCs compounds. The nature of T1 and T2
is confirmed by the spin-density distributions shown in
Figure 5, which perfectly correlate with the spin densities
calculated for the lowest triplet state of the p-conjugated
central bridge and of the mononuclear complex, respective-
ly. Theoretical calculations, therefore, suggest that the ab-
sorption bands observed in the transient absorption spectra
of compounds 1 result from the excitation of the lowest trip-
3
phosphorescence of IrIII complexes by the p–p* state of the
conjugated spacer. Although the luminance level is low, the
external quantum efficiency (EQE) values obtained for
LECs based on 1a (0.16%) and 1b (0.13%) are rather close
to the maximum EQE values (1a: 0.20%, 1b: 0.16%) esti-
mated from the photoluminescence quantum yields.[39,40]
This indicates that emission inside the device is highly opti-
mized, as it is nearly as efficient as we can expect from the
photophysical properties of 1a and 1b, and suggests that the
self-quenching due to the high iTMC concentration in the
device has been minimized. To obtain higher luminances,
3
let state (T1) centered on the wire. The presence of the p–
p* triplet below the emitting states centered on the iTMCs
depopulates these states and explains the lower photolumi-
nescence quantum yield observed for compounds 1 com-
pared with complexes 8.
3
the low p–p* state of the p-conjugated bridge linking the
Electroluminescence properties: Simple two-layer LECs
were built by using the dinuclear complexes 1a and 1b. The
details concerning the device fabrication are described in
the Experimental Section. Upon applying a bias of 3 V, the
build up of the luminance is synchronous with that of the
current density (see Figure S2 in the Supporting Informa-
tion). The observed time-delayed response is a typical fea-
ture of LECs and reflects the mechanism of device opera-
IrIII complexes should be shifted to energies above the emit-
3
ting MLCT/3LLCT iTMC triplets. This will suppress the un-
desired non-radiative deactivation pathway through the wire
and LEC devices with high luminances can be expected.
Conclusion
ꢀ
tion for which the PF6 counter-ions have to reach the inter-
The synthesis and the electrochemical, photophysical, and
theoretical characterization of a novel family of dumbbell-
shaped dinuclear complexes (1), in which an oligophenyle-
neethynylene spacer is linked with two heteroleptic iridium-
face of the electrodes which enhances the charge-injection
process.[15,16]
The electroluminescence (EL) spectra of compounds 1
consist of a broad structureless band with a shape similar to
the photoluminescence spectra recorded in solution
(Figure 6). As expected, compound 1a emits in the yellow
region (CIE coordinates: x=0.4527; y=0.4164) with a maxi-
mum at l=596 nm, whereas 1b emits in the green region
(x=0.3696; y=0.4625) with a maximum at l=540 nm.[38]
AHCTUNGTREG(NNNU III) complexes, have been reported. From these studies we
conclude that the lowest triplet excited state in the new
3
complexes corresponds to a p–p* state located on the con-
jugated spacer. The appearance of this state below the
3MLCT/3LLCT emitting states of the end-capping IrIII com-
plexes is an efficient quenching pathway, which explains the
9860
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9855 – 9863

Dumbbell-Shaped Dinuclear Iridium Complexes
FULL PAPER
Compound 6: Compounds 4 (1.34 g, 3.37 mmol) and 5 (0.5 g, 1.53 mmol)
were dissolved in dry THF (25 mL) under argon atmosphere and diiso-
propylamine (1.3 mL), tetrakis(triphenylphosphine)palladium(0) (0.17 g,
0.15 mmol) and copper(I) iodide (0.03 g, 0.15 mmol) were added. The re-
action mixture was heated at 508C and stirred overnight. After evapora-
tion of the solvent under reduced pressure, the residue was washed with
NH4Cl, extracted with methylene chloride and dried over MgSO4. After
evaporation of the solvent, the residue was purified by column chroma-
tography (silica gel, chloroform/ethanol (200:1) yielding 6 as a yellowish
solid (0.50 g, 48%). 1H NMR (CDCl3, 300 MHz): d=8.55 (d, J=1.5 Hz,
4H; Hg +Hg’), 8.30 (d, J=1.5 Hz, 4H; Hf +Hf’), 7.47 (d, J=9 Hz, 4H;
Ha), 7.15 (d, J=9 Hz, 4H; Hb), 7.14 (d, J=1.5 Hz, 4H; He), 7.08 (d, J=
1.5 Hz, 4H; He’), 7.01 (s, 2H; Hj), 4.03 (t, J=9 Hz, 4H; Hk), 3.02 (m, 8H;
Hc +Hd), 2.45 (s, 2H; Hi), 1.85 (m, 4H), 1.54 (m, 4H), 1.38 (m, 4H),
0.92 ppm (t, J=9 Hz, 6H); 13C NMR (CDCl3, 75MHz): d=159.5, 133.8,
133.5, 124.6, 115.5, 115.1, 90.8, 87.2, 72.3, 71.2, 71.1, 71.0, 70.0, 67.9,
59.5 ppm; FTIR (neat): n˜ =3050, 3014, 2927, 2958, 1927, 1624, 1594, 1554,
1482, 1456, 1376, 1247, 1195, 1107, 1062, 900, 822, 765, 707 cmꢀ1; ESI-MS:
m/z: [M+]=870.
low quantum yields observed for the dinuclear complexes in
comparison with the mononuclear IrIII complexes. Although
complexes 1 present poor photophysical properties, their ap-
plication in optoelectronic devices has been tested by using
them as the primary active component in double-layer light-
emitting electrochemical cells (LECs). Electroluminescence
emission in the green and yellow regions is obtained under
applying an external voltage of 3 V. The EQE values ob-
tained for the devices are almost the same that the maxi-
mum EQE values estimated from the photoluminescence
quantum yields of the complexes. This suggests that a near-
quantitative internal electron-to-photon conversion occurs
and that emission inside the device is highly optimized. The
introduction of a molecular spacer to control the separation
of the iTMCs in the active layer is, therefore, a useful strat-
egy to minimize the self-quenching in the operating device.
The design of new dinuclear IrIII complexes connected by p-
conjugated spacers that do not act as quenchers of the phos-
phorescent emission is of high interest for production of
highly-efficient LECs.
General procedure for the heteroleptic iridium
plexes 1 were synthesized by mixing of starting iridium complexes 7 (7a:
0.0564 g, 0.05 mmol; 7b: 0.0608 g, 0.05 mmol) and spacer (0.04 g,
ACHTUNGERTN(NUNG III) complexes (1): Com-
6
0.055 mmol) in an ethyleneglycol solution heated to reflux during 16 h.
Upon cooling to room temperature, the resulting yellow mixture was
transferred to a separation funnel with water (40 mL) and washed with
diethyl ether (3ꢄ30 mL). A concentrated solution of ammonium hexa-
fluorophosphate (1 g) in water (10 mL) was slowly added, at 08C, to the
reaction mixture, yielding a colored suspension. The compound was
transferred to the fridge for 16 h and the precipitate was collected by fil-
tration. The yellow product was recrystallized from acetonitrile/diethyl
ether.
Experimental Section
Materials and solvents: All solvents were dried according to standard
procedures. Reagents were used as purchased. All air-sensitive reactions
were carried out under an argon atmosphere.
1
Complex 1a: Yield: 37%. H NMR (CD2Cl2, 700 MHz): d=8.49–8.43 (m,
4H); 7.8–7.82 (m, 4H); 7.71 (d, J=5.1 Hz, 6H); 7.54 (m, 4H); 7.47–7.39
(m, 6H); 7.17–7.09 (m, 9H); 6.98–6.96 (m, 6H); 6.85–6.82 (m, 4H); 6.08
(s, 4H); 3.12–3.04 (m, 8H); 2.60 (s, 6H; Mebpy); 2.13 (s, 12H;
Meppy);1.56–0.88 ppm (m, 26H); 13C NMR (CDCl3, 175 MHz): d=
167.86, 160.38, 155.30, 154.97, 150.53, 150.01, 148.27, 141.00, 137.75,
132.33, 129.31, 128.90, 128.72, 128.54, 128.31, 128.17, 128.14, 126.60,
124.77, 124.48, 123.57, 123.50, 122.54, 121.38, 119.37, 119.30, 119.06, 77.53,
41.08, 36.74, 35.75, 31.56, 29.69, 29.36, 29.28, 25.69, 22.64, 21.31,
13.82 ppm; ESI-MS: m/z: [M+2]=966.
Complex 1b: Yield, 63%.1H NMR (CDCl3, 700 MHz): d=8.37 (m, 4H;
z), 7.88 (m, 4H), 7.52 (d, J=7.5 Hz, 4H), 7.48 (d, J=7.5 Hz, 4H), 7.36 (s,
4H), 7.34 (d, J=7.5 Hz, 4H), 7.29 (d, J=7.5 Hz, 4H), 7.24 (d, J=7.5 Hz,
4H), 7.10 (s, br, 4H), 7.06 (s, 4H), 6.65 (t, J=11.9 Hz, 4H), 5.78 (dd, J1 =
2 Hz, 4H), 4.07 (t, J=9 Hz, 4H; Hk), 3.25 (t, J=7.7 Hz, 4H), 3.11 (t, J=
7.7 Hz, 4H), 2.66 (s, 6H; Hi), 1.87 (m, 4H), 1.54 (m, 4H), 1.41 (m, 4H),
1.38 (m, 4H), 0.92 ppm (t, J=7 Hz, 6H); 13C NMR (CDCl3, 175 MHz):
d=164.44, 164.34, 164.30, 162.31, 160.82, 155.57, 155.32, 155.02, 153.93,
153.56, 153.03, 149.95, 149.76, 148.55, 140.49, 139.12, 131.65, 129.33,
128.72, 128.55, 127.66, 125.78, 125.11, 123.88, 123.76, 123.56, 121.45,
116.71, 114.02, 113.78, 99.98, 69.59, 36.74, 35.75, 31.56, 29.69, 29.28, 25.69,
22.64, 21.31, 13.82 ppm; ESI-MS: m/z: [M+2]=1008.
Synthesis and characterization: Flash chromatography was performed by
using silica gel (Merck, Kieselgel 60, 230–240 mesh or Scharlau 60, 230–
240 mesh). Analytical thin layer chromatography (TLC) was performed
by using aluminum coated Merck Kieselgel 60 F254 plates. The NMR
spectra were recorded by using a Bruker Avance 300 (1H: 300 MHz; 13C:
75 MHz) and a Bruker Avance III 700 MHz spectrometer at 298 K using
partially deuterated solvents as internal standards. Coupling constants (J)
are denoted in Hz and chemical shifts (d) in ppm. Multiplicities are de-
noted as follows: s=singlet, d=doublet, t=triplet, m=multiplet, br=
broad.
Compound 5,[41] the cyclometalated iridium chloro-bridged dimer 7,[42,43]
and compounds 8[11] were prepared according to previously reported syn-
thetic procedures and showed identical spectroscopic properties to those
reported therein.
Compound 4: 4-Methyl-2-(4-methylpyridin-2-yl)pyridine (2, 4.03 g,
21.9 mmol) in dry THF was added dropwise to a freshly prepared lithium
diisopropylamide (LDA, 2.3 g, 23.07 mmol) at ꢀ788C and under an
argon atmosphere. The resulting solution was stirred in these conditions
for one hour. After that, this solution was added through a cannula to a
solution of 1-(bromomethyl)-4-iodobenzene (3, 5 g, 16.84 mmol) in dry
THF (50 mL). The solution was stirred in these conditions overnight al-
lowing it to reach room temperature. After that, methanol (20 mL) was
added. After evaporation of the solvent under reduced pressure, the resi-
due was washed with water, extracted with methylene chloride and dried
over MgSO4. After evaporation of the solvent, the residue was purified
by column chromatography (silica gel, chloroform/ethanol (100/1)) yield-
Electrochemical characterization: Cyclic voltammetry was performed by
using an Autolab PGStat 30 equipment. These measurements were made
in a low-volume BAS cell. A glassy carbon working electrode (BAS MF-
2012) was used after being polished with alumina (0.3 m) for 1 min, and
platinum wire was used as counter electrode. A Ag/AgNO3 electrode was
used as a reference. Tetrabutylammonium perchlorate (0.1m) was used as
the supporting electrolyte and acetonitrile as solvent. The samples were
purged with argon prior to measurement.
ing compound
4 as a
white solid. (4.5 g, 66%). 1H NMR (CDCl3,
300 MHz): d=8.55 (d, J=1.5 Hz, 2H; Hg +Hg’), 8.25 (d, J=1.5 Hz, 2H;
Hf +Hf’), 7.60 (d, J=9 Hz, 2H; Ha), 7.15 (d, J=1.5 Hz, 1H; He), 7.13 (d,
J=1.5 Hz, 1H; He’), 6.92 (d, J=9 Hz, 2H; Hb), 2.97 (m, 4H; Hc +Hd),
2.45 ppm (s, 3H; Hi); 13C NMR (CDCl3, 75 MHz): d=156.8, 156.4, 151.4,
149.5, 149.3, 148.4, 140.9, 137.9, 130.8, 125.1, 124.2, 12.4, 121.5, 91.6, 37.4,
36.5, 21.5 ppm; FTIR (neat): n˜ =3052, 2928, 2860, 2208, 1727, 1595, 1554,
1515, 1461, 1414, 1380, 1276, 1212, 1116, 1019, 941, 895, 827, 761, 723,
694 cmꢀ1; ESI-MS: m/z: [M+]=400.
Photophysical characterization: UV/Vis spectra were recorded in a 1 cm
path-length quartz cell by using a 845x UV/Vis Agilent spectrophotome-
ter, operating with UV/Vis ChemStation Software. Steady-state lumines-
cence spectra were measured by using a Photon Technology spectrofluor-
ometer, equipped with a lamp power supply (LPS-220B), working at
room temperature. The excited-state lifetimes and the transient absorp-
tion spectra were measured from fresh solutions, which were degassed by
Chem. Eur. J. 2010, 16, 9855 – 9863
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9861

N. Martꢀn, E. Ortꢀ, H. J. Bolink et al.
nitrogen bubbling for 30 min. The excited-state lifetimes of the molecular
wire 6 and compound 1b were measured in the same setup using a nitro-
gen-laser/dye-laser/frequency-doubler source (PTI). Transient absorption
spectra for complexes 8 and compounds 1 were measured by means of a
laser flash-photolysis system based on a pulsed Nd:YAG laser, using l=
310 nm as the exciting wavelength. The single pulses were approximately
10 ns duration and the energy was approximately 15 mJ per pulse. A
Lo255 Oriel xenon lamp was employed as the detecting light source. The
laser flash-photolysis apparatus consisted of the pulsed laser, the Xe
time-dependent DFT (TD-DFT) calculations[54–56] were performed in ace-
tonitrile solution to determine the energies and the electronic nature of
the lowest excited triplet states at the ground-state geometry. Vertical
electronic excitation energies were determined for the lowest 10 triplet
states.
lamp,
a 77200 Oriel monochromator, and an Oriel photomultiplier
(PMT) system made up of a 77348 PMT power supply. The oscilloscope
was a TDS-640 A Tektronix. The excited-state lifetimes were deduced
from this technique. The quantum yields of 1a and 1b compounds were
determined in de-aerated acetonitrile solutions at an excitation wave-
length of l=310 nm using Equation (1),
Acknowledgements
This work was supported by the ESF (Eurocores-05SONS-FP-021 and
FP7-212942-2, FUNMOLS), the Spanish Ministry of Science and Innova-
tion (MICINN) (MAT2006-28185-E, MAT2007-61584, CSD2007-00010,
CTQ2006-14987-C02-02, CTQ2009-08790, and CTQ2008-00795), the
Generalitat Valenciana(ACOMP/2009/269), and European FEDER
funds. R.D.C. and G.F. acknowledge the support of a FPU grant of the
MICINN.
ðD=Aexc
Þ
ðD=AexcÞx
ꢀx ¼ ꢀrKopt
ð1Þ
r
for which the subscript x denotes the 1a and 1b compounds for which
the quantum yield is to be determined (10ꢀ6 m in CH3CN), subscript r
refers to the reference substance bisulfate quinoline (10ꢀ6 m in CH3CN)
for which the luminescence quantum yield is assumed to be 0.98, D is the
integrated area under the emission spectrum, and Aexc is the absorbance
at the exciting wavelength. The quantum yields for complexes 8 and the
molecular wire 6 were determined in de-aerated acetonitrile solution by
using an integrated sphere (Hamamatsu model C9920–0). The system is
made up of an excitation light source, consisting of a xenon lamp linked
to a monochromator, an integration sphere, and a multi-channel spec-
trometer.
[1] G. Saito, F. Wudl, R. C. Haddon, K. Tanigaki, T. Enoki, H. E. Katz,
M. Maesato, Multifunctional Conducting Molecular Materials, RSC,
Cambridge 2007.
[2] C. J. Brabec, N. S. Sariciftci, J. C. Hummelen, Adv. Funct. Mater.
2001, 11, 15.
[3] K. M. Coakley, M. D. McGehee, Chem. Mater. 2004, 16, 4533.
[4] B. C. Thompson, J. M. J. Frꢆchet, Angew. Chem. 2008, 120, 62;
Angew. Chem. Int. Ed. 2008, 47, 58.
Device preparation: The pre-patterned ITO glass plates were extensively
cleaned, by using chemical and UV-Ozone methods, just before the depo-
sition of the organic layers. Compounds 1 were used as the single active
component in preparing double-layer LECs. A first layer of poly(3,4-eth-
ylenedioxythiophene)/polystyrenesulfonate (PEDOT:PSS) (100 nm) was
deposited on top of a patterned ITO substrate to increase the device
preparation yield. Compounds 1 were then spin-coated from an acetoni-
trile solution (20 mgmLꢀ1) incorporating the ionic liquid (IL) 1-butyl-3-
methylimidazolium hexafluorophosphate in a 1:1 iTMC/IL molar ratio to
yield an 80 nm thick layer. The thickness of the films was determined by
using an Ambios XP1 profilometer. The IL helps to enhance the perfor-
mance level of the LEC device.[44, 45] On top of the active layer, the metal
cathode (aluminum) was thermally evaporated under high vacuum by
using an evaporator integrated into an inert atmosphere glovebox
(<0.1 ppm O2 and H2O). The base pressure for evaporation was <1ꢄ
10ꢀ6 mbar. Current density and luminance versus voltage were measured
using a Keithley 2400 source meter and a photodiode coupled to a Keith-
ley 6485 picoampmeter by using a Minolta LS100 to calibrate the photo-
current. An Avantes luminance spectrometer was used to measure the
EL spectrum. Devices were characterized in inert atmosphere.
[5] J. G. C. Veinot, T. J. Marks, Acc. Chem. Res. 2005, 38, 632.
[6] Z. H. Kafafi, Organic Electroluminescence, CRC Press, Boca Raton,
2005.
[7] K. Mꢇllen, U. Scherf, Organic Light Emitting-Devices: Synthesis
Properties and Applications, Wiley-VCH, Weinheim, 2006.
[8] Z. Li, H. Meng, Organic Light-Emitting Materials and Devices,
CRC, Boca Raton, 2007.
[9] Q. Pei, G. Yu, C. Zhang, Y. Yang, A. J. Heeger, Science 1995, 269,
1086.
[10] V. Balzani, S. Campagna, Photochemistry and Photophysics of Coor-
dination Compounds I, Vol. 280, Springer, Berlin, 2007.
[11] M. S. Lowry, W. R. Hudson, R. A. Pascal, S. Bernhard, J. Am.
Chem. Soc. 2004, 126, 14129.
[12] A. B. Tamayo, S. Garon, T. Sajoto, P. I. Djurovich, I. M. Tsyba, R.
Bau, M. E. Thompson, Inorg. Chem. 2005, 44, 8723.
[13] P. T. Chou, Y. Chi, Chem. Eur. J. 2007, 13, 380.
[14] A. B. Tamayo, B. D. Alleyne, P. I. Djurovich, S. Lamansky, I. Tsyba,
N. N. Ho, R. Bau, M. E. Thompson, J. Am. Chem. Soc. 2003, 125,
7377.
[15] J. D. Slinker, J. Rivnay, J. S. Moskowitz, J. B. Parker, S. Bernhard,
H. D. AbruÇa, G. G. Malliaras, J. Mater. Chem. 2007, 17, 2976.
[16] J. Slinker, D. Bernards, P. L. Houston, H. D. AbruÇa, S. Bernhard,
G. G. Malliaras, Chem. Commun. 2003, 2392.
[17] J. C. deMello, N. Tessler, S. C. Graham, R. H. Friend, Phys. Rev. B
1998, 57, 12951.
[18] J. D. Slinker, J. A. DeFranco, M. J. Jaquith, W. R. Silveira, Y. Zhong,
J. M. Moran-Mirabal, H. G. Graighead, H. D. AbruÇa, J. A. Marohn,
G. G. Malliaras, Nat. Mater. 2007, 6, 894.
Computational details: Density functional calculations (DFT) were com-
pleted with the D.02 revision of the Gaussian 03 program package,[46]
using Beckeꢅs three-parameter B3LYP exchange-correlation function-
al[47–49] together with the 6-31G** basis set for C, H, F, and N atoms[50]
and the “double-z” quality LANL2DZ basis set for the Ir element.[51] An
effective core potential (ECP) replaces the inner core electrons of iridi-
um leaving the outer core (5s)2(5p)6 electrons and the (5d)6 valence elec-
trons of IrIII. The geometries of the singlet ground state (S0) and of the
lowest triplet excited states were fully optimized by imposing C2 symme-
try restrictions. The hexyloxy groups of the wire and the methyl groups
in the bpy and ppy ligands of the end-capping iTMCs were removed.
Triplet states were calculated at the spin-unrestricted UB3LYP level with
a spin multiplicity of 3. Geometry optimization of the lowest triplet
states T1 (3B) and T2 (3A) of compounds 1 was achieved by imposing
symmetry restrictions to the wavefunction. Solvent effects were consid-
ered within the SCRF (self-consistent reaction field) theory using the po-
larized continuum model (PCM) approach to model the interaction with
the solvent.[52,53] Molecular orbitals were calculated in acetonitrile using
gas-phase optimized geometries. Starting from the closed-shell S0 state,
[19] P. Matyba, K. Maturova, M. Kemerink, N. D. Robinson, L. Edman,
Nat. Mater. 2009, 8, 672.
[20] H. J. Bolink, E. Coronado, R. D. Costa, E. Ortꢀ, M. Sessolo, S.
Graber, K. Doyle, M. Neuburger, C. E. Housecroft, E. C. Constable,
Adv. Mater. 2008, 20, 3910.
[21] S. Graber, K. Doyle, M. Neuburger, C. E. Housecroft, E. C. Consta-
ble, R. D. Costa, E. Ortꢀ, D. Repetto, H. J. Bolink, J. Am. Chem.
Soc. 2008, 130, 14944.
[22] R. D. Costa, E. Ortꢀ, H. J. Bolink, S. Graber, C. E. Housecroft, M.
Neuburger, S. Schaffner, E. C. Constable, Chem. Commun. 2009,
2029.
9862
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 9855 – 9863

Dumbbell-Shaped Dinuclear Iridium Complexes
FULL PAPER
[23] R. D. Costa, E. Ortꢀ, H. J. Bolink, S. Graber, S. Schaffner, M. Neu-
burger, C. E. Housecroft, E. C. Constable, Adv. Funct. Mater. 2009,
19, 3456.
[24] H. C. Su, H. F. Chen, F. C. Fang, C. C. Liu, C. C. Wu, K. T. Wong,
Y. H. Liu, S. M. Peng, J. Am. Chem. Soc. 2008, 130, 3413.
[25] H. C. Su, F. C. Fang, T. Y. Hwu, H. H. Hsieh, H. Chen, G. Lee, S.
Peng, K. T. Wong, C. C. Wu, Adv. Funct. Mater. 2007, 17, 1019.
[26] L. He, L. Duan, J. Qiao, R. Wang, P. Wei, L. Wang, Y. Qiu, Adv.
Funct. Mater. 2008, 18, 2123.
[40] J. C. Scott, G. G. Malliaras in Conjugated Polymers, Vol. 13, Wiley-
VCH, New York 1999.
[41] C. Atienza, B. Insuasty, C. Seoane, N. Martꢀn, J. Ramey, G. M. A.
Rahmand, D. M. Guldi, J. Mater. Chem. 2005, 15, 124.
[42] M. Nonoyama, Bull. Chem. Soc. Jpn. 1974, 47, 767.
[43] S. Sprouse, K. A. King, P. J. Spellane, R. J. Watts, J. Am. Chem. Soc.
1984, 106, 6647.
[44] S. T. Parker, J. Slinker, M. S. Lowry, M. P. Cox, S. Bernhard, G. G.
Malliaras, Chem. Mater. 2005, 17, 3187.
[45] J. D. Slinker, C. Y. Koh, G. G. Malliaras, M. S. Lowry, S. Bernhard,
Appl. Phys. Lett. 2005, 86, 173506.
[27] L. He, J. Qiao, L. Duan, G. Dong, D. Zhang, L. Wang, Y. Qiu, Adv.
Funct. Mater. 2009, 19, 2950.
[46] Gaussian 03, Revision C.02, 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., Wall-
ingford CT, 2004.
[28] R. D. Costa, F. J. Cꢆspedes-Guirao, E. Ortꢀ, H. J. Bolink, J. Giersch-
ner, F. Fernꢁndez-Lꢁzaro, A. Sastre-Santos, Chem. Commun. 2009,
3886.
[29] J. L. Rodrꢀguez-Redondo, R. D. Costa, E. Ortꢀ, A. Sastre-Santos,
H. J. Bolink, F. Fernꢁndez-Lꢁzaro, Dalton Trans. 2009, 9787.
[30] H. J. Bolink, L. Cappelli, S. Cheylan, E. Coronado, R. D. Costa, N.
Lardiꢆs, M. K. Nazeeruddin, E. Ortꢀ, J. Mater. Chem. 2007, 17, 5032.
[31] H. Rudmann, M. F. Rubner, J. Appl. Phys. 2001, 90, 4338.
[32] H. Rudmann, S. Shimada, M. F. Rubner, J. Am. Chem. Soc. 2002,
124, 4918.
[33] J. D. Slinker, A. A. Gorodetsky, M. S. Lowry, J. Wang, S. Parker, R.
Rohl, S. Bernhard, G. G. Malliaras, J. Am. Chem. Soc. 2004, 126,
2763.
[34] C. Rothe, C.-J. Chiang, V. Jankus, K. Abdullah, X. Zeng, R. Jitchati,
A. S. Batsanov, M. R. Bryce, A. P. Monkman, Adv. Funct. Mater.
2009, 19, 2038.
[35] E. S. Handy, A. J. Pal, M. F. Rubner, J. Am. Chem. Soc. 1999, 121,
3525.
[36] D. W. Thompson, J. F. Wishart, B. S. Brunschwig, N. Sutin, J. Phys.
Chem. A 2001, 105, 8117.
[37] R. D. Costa, F. J. Cꢆspedes-Guirao, H. J. Bolink, F. Fernꢁndez-
Lꢁzaro, A. Sastre-Santos, E. Ortꢀ, J. Gierschner, J. Phys. Chem. C
2009, 113, 19292.
[47] A. D. Becke, J. Chem. Phys. 1988, 88, 2547.
[48] C. T. Lee, W. T. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785.
[49] A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
[50] 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.
[51] P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299.
[52] C. S. Cramer, D. G. Truhlar, Solvent Effects and Chemical Reactivity,
Kluwer, Dordrecht 1996.
[53] J. Tomasi, M. Persico, Chem. Rev. 1994, 94, 2027.
[54] C. Jamorski, M. E. Casida, D. R. Salahub, J. Chem. Phys. 1996, 104,
5134.
[55] M. Petersilka, U. J. Gossmann, E. K. U. Gross, Phys. Rev. Lett. 1996,
76, 1212.
[56] M. E. Casida, C. Jamorski, K. C. Casida, D. R. Salahub, J. Chem.
Phys. 1998, 108, 4439.
[38] CIE, 1933.
[39] The EQE can be calculated by the equation h=bf/2n2, in which b
is the recombination efficiency, f is the fraction of excitons that
decay radiatively, and n is the refractive index of the glass substrate
(the factor 1/2n2 accounts for the coupling of light out of the
device). The b factor depends on the injection characteristics and is
equal to unity when both contacts are ohmic. In first approximation,
the parameter f should be equal to the quantum yield of these com-
pounds since the EL emission involves the same emitting states than
the photoluminescence emission in these devices. For the glass sub-
strate, the n parameter is approximated to 1.5.
Received: March 8, 2010
Published online: June 25, 2010
Chem. Eur. J. 2010, 16, 9855 – 9863
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9863