Solution processable phosphorescent rhenium(i) dendrimers

Article abstract of DOI:10.1039/b707896j

A family of (1,10-phenanthroline)rhenium(i)(CO)₃Cl complex cored first generation dendrimers with one, two or three dendrons have been prepared. The first generation dendrons attached to the core complex are comprised of biphenyl units with 2-ethylhexyloxy surface groups at their distal ends. The number and position of attachment of the dendron to the core was found to have an effect on the properties of the dendrimers. When dendrons were attached to both the 2- and 9-positions of the 1,10-phenanthroline ligand, thus straddling the rhenium(i), the dendrimers became more electrochemically stable and less susceptible to solvatochromism. The dendrimers were generally found to have their emission blue-shifted and a higher photoluminescence quantum yield in the solid state than in the solution. The origin of this rigidochromism effect is discussed. A single layer device with a neat dendrimer film was found to have an external quantum efficiency of 0.4% (0.8 cd A^-1) and power efficiency of 0.2 lm W^-1 at 100 cd m^-2 and 12.8 V. The Royal Society of Chemistry.

Full text of DOI:10.1039/b707896j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Solution processable phosphorescent rhenium(I) dendrimers
Yong-Jin Pu,a Ruth E. Harding,c Stuart G. Stevenson,c Ebinazar B. Namdas,c Christine Tedeschi,a
Jonathan P. J. Markham,c Richard J. Rummings,a Paul L. Burn*ab and Ifor D. W. Samuel*c
Received 24th May 2007, Accepted 20th July 2007
First published as an Advance Article on the web 14th August 2007
DOI: 10.1039/b707896j
A family of (1,10-phenanthroline)rhenium(I)(CO)3Cl complex cored first generation dendrimers
with one, two or three dendrons have been prepared. The first generation dendrons attached to
the core complex are comprised of biphenyl units with 2-ethylhexyloxy surface groups at their
distal ends. The number and position of attachment of the dendron to the core was found to have
an effect on the properties of the dendrimers. When dendrons were attached to both the 2- and
9-positions of the 1,10-phenanthroline ligand, thus straddling the rhenium(I), the dendrimers
became more electrochemically stable and less susceptible to solvatochromism. The dendrimers were
generally found to have their emission blue-shifted and a higher photoluminescence quantum yield in
the solid state than in the solution. The origin of this rigidochromism effect is discussed. A single
layer device with a neat dendrimer film was found to have an external quantum efficiency of 0.4%
(0.8 cd A21) and power efficiency of 0.2 lm W21 at 100 cd m22 and 12.8 V.
ligand. In this paper we describe the synthesis of a series of
Introduction
(1,10-phenanthroline)rhenium(I)tricarbonyl complexes with
Phosphorescent emitters for organic light-emitting diodes
one, two, or three dendrons attached to the 1,10-phenanthro-
(OLEDs) are being increasingly investigated due to their high
line ligand (Fig. 1). The first generation dendron used was
device efficiencies that arise from the fact that both singlets
comprised of biphenyl units and contained 2-ethylhexyloxy
and triplets can be captured for emission. Most phosphorescent
surface groups. We discuss the effect of the number and
emitters are based on heavy metal complexes with the most
position of the dendrons on the photophysical and electro-
common being iridium(III) complexes.1–6 Less investigated
chemical properties of the materials. We also report the
are devices that utilize rhenium(I) complexes as the phos-
preliminary performance of a DLED incorporating one of the
phorescent emitter.7–15 While much of the work on phos-
materials.
phorescent emitters has been focusing on small molecules1–6
there has been an increasing focus on developing phos-
Results and discussion
phorescent light-emitting dendrimers to control the processing
and intermolecular interactions that govern device perfor-
mance.16–21 We have shown that by controlling the generation
and/or number of dendrons,17,18 and dendron type,22 it is
possible to control charge transport22,23 and light-emission24 in
iridium(III) complex cored dendrimers and form simple highly
efficient dendrimer light-emitting diodes (DLEDs).16 OLEDs
with solution processed layers comprised of polymers with
rhenium(I) units in the backbones or with rhenium(I) com-
plexes blended with a polymer host have generally given very
poor performance.8,10 The best rhenium(I) containing devices
have utilized evaporated layers with the devices including a
hole transport layer, blended emissive layer, and an electron
transport layer. We were therefore interested in applying the
dendrimer technology to rhenium(I) complexes.
Syntheses and physical properties
The synthetic routes to the dendronised rhenium(I) complexes
of Fig. 1 are shown in Scheme 1. Two different strategies were
used to attach the dendrons to the 1,10-phenanthroline. 1,10-
Phenanthroline is susceptible to nucleophilic attack at the 2(9)-
and 4(7)-positions and the dendrons were added to the 2- and
9-positions via an aryl anion. The dendrons could not be
attached to the 3(8)- and 5(6)-positions by the same method
and hence a Suzuki reaction was utilized. It should be noted
that apart from the 1,10-phenanthroline ligand with the
dendron attached to the 5-position (L2) it was not possible
to isolate the dendronised ligand in an analytically pure form
and hence they were reacted to form the complex which could
be purified. The key dendron intermediate for this work was
Most rhenium(I) complexes studied for light-emission have
had a bidentate heteroaromatic ligand such as 2,29-bipyridine
or 1,10-phenanthroline with three carbonyl and one chlorine
3,5-di[4-(2-ethylhexyloxy)phenyl]phenylbromide
(G1-Br).25
G1-Br was used directly for the formation of Re3, Re5,
and Re6 but converted to the corresponding boronic acid
[G1-B(OH2)] for the formation of Re2 and Re4. G1-B(OH2)
was formed by an analogous method as that reported for the
formation of an equivalent pinacolate boronate ester.25 G1-Br
was metallated with n-butyllithium at low temperature and
then reacted with trimethylborate followed by work-up with
dilute aqueous hydrochloric acid. The advantage of this route
aDepartment of Chemistry, University of Oxford, Chemistry Research
Laboratory, Mansfield Rd, Oxford, UK OX1 3TA
bCentre for Organic Photonics and Electronics, University of
Queensland, St Lucia, Queensland, 4072, Australia
cOrganic Semiconductor Centre, SUPA, School of Physics and
Astronomy, University of St Andrews, North Haugh, St Andrews, Fife,
UK KY16 9SS
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4255–4264 | 4255

View Article Online
Fig. 1 Chemical structures of the complexes Re1–6.
ligand with the dendron on the 5-position (L2) in a 54% yield.
L2 was then reacted with pentacarbonylchlororhenium(I) in
toluene heated at reflux to give Re2 in a 52% yield. Re4 with
dendrons in the 3- and 8-positions on the ligand was formed
in a similar manner with G1-B(OH2) being reacted with 3,8-
dibromo-1,10-phenanthroline under Suzuki conditions to give
L4. L4 was then reacted with pentacarbonylchlororhenium(I)
in refluxing toluene to give Re4 in a 24% yield for the two
steps. While in principle the dendronised ligand L3 could be
formed from the Suzuki reaction of G1-B(OH2) with 4,7-
dibromo-1,10-phenanthroline, we found that the L3 formed
was very difficult to purify. However, by converting the G1-Br
into the corresponding zinc reagent (G1-ZnCl) by metallation
and reaction with zinc chloride it was possible to carry out a
Negishi coupling to give L3. L3 was then reacted with
pentacarbonylchlororhenium(I) giving Re3 in a 14% yield for
the two steps. Re5 with the dendrons in the 2- and 9-positions
was prepared by reaction of the lithium salt of G1-Br, G1-Li,
with 1,10-phenanthroline followed by oxidation with manga-
nese dioxide and subsequent complex formation in an overall
43% yield. Re6 with dendrons in the 2-, 5-, and 9-positions
was made in a 17% yield by an analogous method to Re5
except that L2 was used as the starting phenanthroline. All
the dendritic complexes, Re2–Re6, showed good solubility in
common organic solvents such as tetrahydrofuran, dichloro-
methane, and toluene.
Scheme 1 Synthetic route of the complexes Re2–6. Reagents and
conditions: (a) Pd(PPh3)4, Na2CO3aq., EtOH, toluene, 100 uC, 24 h; (b)
Re(CO)5Cl, toluene, reflux; (c) (i) n-BuLi, THF, 278 uC, 1 h, (ii)
ZnCl2, THF, 278 uC; (d) Pd(PPh3)4, THF, 80 uC, 17 h; (e) n-BuLi,
diethyl ether, 1 h; (f) (i) r.t., (ii) MnO2, dichloromethane, 1 h.
Infrared analysis of the complexes in the carbonyl region
(1880–2025 cm21, Table 1) revealed that there were three
stretches, which was consistent with the complexes only having
the facial configuration.26 In the facial configuration there are
only two carbonyl environments and two of the stretches
correspond to the symmetric and asymmetric stretches of the
two carbonyls in the same plane, with the third stretch due to
the third carbonyl which is opposite the chlorine atom. If the
complexes were the mer isomers only two stretches would
be infrared active. The 13C-NMR confirmed that there was
only one isomer present. Interestingly Re3, Re4, and Re5 all
had two peaks due to the carbonyl carbons in the region of
over forming the pinacolate boronate ester is that is cheaper
to form on a large scale. However, the difficulty is that the
product is not a single compound with dimers and cyclic
trimers also formed although these can also be used in the
Suzuki reaction.
The 5-substituted complex (Re2) was formed from 5-bromo-
1,10-phenanthroline in two steps. In the first step it was
coupled with G1-B(OH2) under Suzuki conditions to give the
4256 | J. Mater. Chem., 2007, 17, 4255–4264
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Table 1 13C-NMR chemical shift and IR streching frequencies of
the carbonyl groups, and decomposition (Td) temperatures of the
complexes
d/ppma
n/cm21 b
Td/uCc
Re1
Re2
Re3
Re4
Re5
Re6
a
2017, 1930, 1897
2020, 1916, 1893
2020, 1920, 1888
2023, 1923, 1883
2019, 1921, 1884
2022, 1926, 1885
c
348, 358
306, 339
344, 393
276, 307
320, 357
365, 392
189.4, 196.98, 197.03
189.7, 197.2
189.8, 197.1
192.1, 192.6
192.2, 192.7
b
125 MHz, CDCl3. Film on NaCl. 5 and 10% weight loss.
189–199 ppm. In contrast the 13C-NMR spectrum of Re2
exhibited three carbonyl signals because of the asymmetric
nature of its structure, caused by the 1,10-phenanthroline
ligand being substituted by the dendron in the 5-position. The
1H-NMR spectra were also consistent with the structures of
the complexes. There was an important difference between the
1H-NMR spectra of Re5 and Re6 and the other complexes.
The 1H-NMR spectra showed that the symmetry of the
protons in the dendrons had been broken (Fig. 2 illustrates the
aromatic region of Re5). This means that when the dendrons
are attached to the 2- and 9-positions their rotation is hindered
by the other ligands on the metal. That is, the dendrons are
encapsulating the rhenium(I) at the centre. As will be discussed
later this imparts important properties to the rhenium(I)
complex cores.
Fig. 3 UV-visible spectra in dichloromethane of Re1–Re6.
(heating rate = 100 uC min21) respectively. Re4 had a more
crystalline appearance and had a melting point at 260–263 uC
while the thermal properties of Re5 and Re6 were much more
complex with overlapping transitions observed.
Photophysical properties
The thermal properties of the dendrimers were studied
by Thermal Gravimetric Analysis (TGA) and Differential
Scanning Calorimetry (DSC). TGA of all the complexes
showed that they had high thermal stability with 5% weight
loss occurring at temperatures greater than 275 uC (Table 1).
Both Re2 and Re3 had clearly observable glass transition
temperatures at 114 uC (heating rate = 30 uC min21) and 121 uC
The first step in the photophysical study of the dendrimers was
to understand their UV-visible absorption spectra, which are
shown in Fig. 3. The parent complex (Re1) has an absorption
peak at 267 nm and a broad absorption between 320 nm and
440 nm. The absorption at short wavelengths is primarily
due to the p–p* transition of the 1,10-phenanthroline in the
Fig. 2 1H-NMR and molecular model of Re5.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4255–4264 | 4257

View Article Online
ligand with the longer wavelength absorption being assigned
to metal-to-ligand charge transfer transitions.27 Each of the
dendrimers have a similar absorption pattern although there
are some important differences. At short wavelengths
(#270 nm) the absorption of the dendrimers is stronger than
that of Re1 and that is due to the biphenyl units in the
dendrons that also absorb at that wavelength. The strength of
the absorption is generally in line with the number of dendrons
(biphenyl units) attached to the core. Re2 has a single dendron
and hence has a lower molar absorption coefficient at
#270 nm than Re3, Re4, and Re5 that have two dendrons.
Re6 with three dendrons attached to the core has the largest
absorption in this part of the UV-visible spectrum. It might be
thought that the attachment of the conjugated dendrons would
have a large effect on the MLCT transitions. However, it is
important to remember that the branching in the dendrons
means that while they are fully conjugated the p-electrons are
not fully delocalised. In fact the MLCT transitions of Re2,
Re3, Re5, and Re6 occur at similar wavelengths and have
similar molar extinction coefficients to Re1. This can be easily
understood for Re3, Re5, and Re6 where in principle the first
phenyl ring of the dendron could be conjugation with the 1,10-
phenanthroline ring. For these dendrimers though, the plane
of the dendrons is orthogonal to that of the 1,10-phenanthro-
line due to steric interactions and hence there is less overlap of
the dendron and ligand p-orbitals. The UV-visible spectrum
of Re4 was different from the other dendrimers at longer
wavelengths. For Re4 there is a second absorption peak at
346 nm and this is most probably due to the fact that when the
dendrons are attached to the 3- and 8-positions there are less
steric interactions meaning that the first phenyl rings of the
two dendrons increases the conjugation length of the ligand.
The fact that the molar absorption coefficient is of order of
33 000 M21cm21 strongly suggests that it has significant p–p*
transition character although it is undoubtedly overlapping
MLCT transitions.
Fig. 4 Comparison of the UV-visible spectra of Re2 (solid line) and
Re5 (dotted line) in dichloromethane (open circle) and toluene–
pentane (4 : 96 v/v) (filled circle).
emission at 421 nm is probably due to a small amount of free
ligand that remained after purification and illustrates the
power of luminescence spectroscopy to identify different
emissive species when a weakly emitting chromophore, in this
case the rhenium(I) complex, is present. Indeed this analytical
technique allowed us to compare the purity of different
batches. Fascinatingly, the solvatochromatism observed in
absorption spectra does not occur in the PL spectra, which is
almost independent of solvent. The extent of solvatochromism
effects on the luminescence appear to be dependent on the
bidentate nitrogen containing ligand used.26,27
One of the features of rhenium(I) complexes is that they can
show solvatochromism, with non-polar solvents causing a red
shift in the MLCT transitions in the UV-visible absorption
spectrum.27 It has been reported that this may be due to
interactions of the solvent with the bidentate ligand orbitals or
a ‘charge separated excited state’.26 We therefore investigated
whether the dendritic structure could create an environment
that gives control over the interactions that lead to solvato-
chromism. The UV-absorption spectra of the dendrimers in
dichloromethane and a toluene–heptane mixture (4 : 96 v/v)
were compared. For the dendrimers with dendrons further
away from the rhenium (Re2, Re3, and Re4) there was a red
shift in the absorption spectra in the less polar toluene–
heptane mixture (the spectra of Re2 are shown in Fig. 4). In
contrast there was little change in the wavelengths of the
absorption spectra for Re5 (Fig. 4) and Re6 indicating that
having the shielding dendrons decrease the solvatochromism
observed in the absorption spectra.
The solution photoluminescence (PL) spectra of the
dendrimers are shown in Fig. 5. The spectra of the dendrimers
are all broad and similar to that of the parent complex Re1,
except that of Re6, which shows a second emission peak at
the shorter wavelength of 421 nm. We believe that the small
Fig. 5 Solution PL spectra of Re2–6 (the spectra have been
normalized).
4258 | J. Mater. Chem., 2007, 17, 4255–4264
This journal is ß The Royal Society of Chemistry 2007

View Article Online
We propose that the differences in spectrum and PLQYs
between the solution and the solid state are due to differences
in molecular geometry and vibronic coupling. In solution the
geometry of the molecules is less constrained than in the solid
state, and so they can reach a more fully relaxed geometry,
resulting in the red shift of solution PL spectra with respect to
the film PL spectra. Non-radiative decay can arise from
overlap of ground and excited state vibronic wavefunctions
with the same energy. Since relaxation energy is proportional
to vibronic coupling,29 this non-radiative decay process is
stronger in the solution than in the solid state. This explains
the lower PLQYs in solution and accounts for the effect of
rigidochromism which has not been fully understood.
Normally the dendrons would be expected to control the
intermolecular interactions in the solid state that lead to
quenching of the luminescence with the best case being where
the solution and solid state PLQYs are the same.16 However,
with these rhenium(I) complex cored dendrimers it is more
difficult to elucidate the role of the dendrons as the PLQY is
larger in the solid state when compared to solution. For
example, when Re2 was blended with m-bis(carbazolyl)ben-
zene the film PLQY increased to 32%. However, this was
associated with a larger blue shift than the neat film and hence
the increased PLQY could be due to smaller relaxation and
vibronic coupling as well as decreasing the intermolecular
interactions that quench the luminescence.
Fig. 6 Film PL spectra of Re2–6 (the spectra have been normalized).
The film PL spectra are shown in Fig. 6 and the key
difference is that the film PL spectra are blue shifted relative to
the solution spectra by approximately 20 nm. A blue shift in
the solid state relative to the solution PL spectra of rhenium(I)
complexes has been reported previously for a complex in a
frozen solvent or in a polymer blend and has been termed
‘rigidochromism’.27,28 Here, we are observing a similar effect,
but in a neat film of the material. The second difference
between the solution and solid state PL spectra occurs for Re6,
which only shows emission from the core complex. This can be
explained by the fact that intermolecular energy transfer
between any free ligand and the dendrimer core can occur
easily in the solid state.
Electrochemical properties
The redox properties of the dendrimers were studied by cyclic
voltammetry (CV) (Fig. 7 and Table 3). The redox potentials
are quoted against the ferrocenium/ferrocene couple.30 All the
complexes showed a chemically reversible reduction at a scan
rate of 100 mV s21 in N,N-dimethylformamide. For rhenium(I)
complexes the LUMO is predominantly on the heterocyclic
ligand, in this case the 1,10-phenanthroline ligand. In spite
of the different substitution patterns of the dendrons of the
dendrimer there was not a large difference in the reduction
potentials of the materials with the reduction E1/2s falling
in the range of 21.63 V to 21.76 V. Many neutral rhenium(I)
complexes have less chemically reversible oxidations at the
usual scan rates used in cyclic voltammetry,31 which is in
contrast to iridium(III) complexes whose oxidations are
generally stable. The first oxidation rhenium(I) complexes
undergo is normally associated with the rhenium(I)/
rhenium(II) redox couple. For example, Re1 has been reported
to have an essentially chemically irreversible oxidation at scan
rates of 1 V s21 but at a scan rate of 50 V s21 the redox process
becomes more chemically reversible.31 We found that dichloro-
methane was the best solvent for the oxidation studies. We also
observed that Re1 had a chemically irreversible oxidation with
Epa (potential of the peak on the anodic scan) of 0.93 V when a
scan rate of 100 mV s21 was used. Attachment of a dendron to
the 5-, 3- and 8-, and 4- and 7-positions was found to have no
effect on the chemical reversibility of the oxidation processes
and no cathodic process was observed for the oxidized species
for dendrimers, Re2, Re3, and Re4. In contrast, for Re5 and
Re6 a cathodic wave was observed for the oxidation with a
scan rate of 100 mV s21 (Fig. 7). However, even under the best
We further probed the photophysical properties of the
dendrimers by measuring their PL quantum yields (PLQYs) in
tetrahydrofuran with the results summarized in Table 2. The
solution PLQYs of the dendrimers ranged between 0.5 and 3%
which is similar to that for the parent complex Re1 (2%) and
typical of such complexes. It should be noted that the solution
PLQY of Re6 also contains a component from the ligand
fluorescence. In neat film, the PLQYs of all the dendrimers
were found to increase relative to their solution values and
were in the range from 7% for Re6 to 19% for Re2 (Table 2).
This is very unusual behaviour; in most complexes, inter-
molecular interactions in neat films results in extra quenching
with respect to the solution.
Table 2 Photoluminescence data
PLQY PLQY
PL (THF)/nm PL (film)/nm EL/nm (THF) (%) (film) (%)
Re1 629a
Re2 632a,b
Re3 637b
Re4 639b
Re5 632
Re6 451, 645a
a
—
—
2
2
3
0.5
1
—
19
13
10
11
7
598
616
602
603
608
612
613
596
601
612
0.5
Excitation wavelength = 360 nm. b Excitation wavelength = 300 nm.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4255–4264 | 4259

View Article Online
Table 3 Electrochemical data
E1/2 (oxidation)
E1/2 (reduction)
E1/2 Eonset Epa
Eonset Epa
Epc
Epc
E1/2
Re1 0.82
Re2 0.81
0.93
0.94
1.09
0.92
1.16
1.15
21.68 21.70 21.79 21.74
21.66 21.66 21.74 21.70
Re3 0.82
21.60 21.63 21.71 21.67
21.56 21.60 21.67 21.63
Re4 0.93
Re5 0.73
Re6 0.76
0.86 0.76 0.81 21.70 21.72 21.80 21.76
0.88 0.75 0.82 21.68 21.69 21.78 21.74
this is an important step forward as it shows that the bulky G1
dendrons on the 2- and 9-positions of the 1,10-phenanthroline
enhance the stability of the core complex. The E1/2s of Re5 and
Re6 were just over 0.8 V (Table 3).
Electroluminescent properties
We prepared simple single layer devices comprised of indium
tin oxide/neat dendrimer film/calcium/aluminium. Of the six
dendrimers Re5 gave the best device performance with an
external quantum efficiency of 0.4% (0.8 cd A21) and power
efficiency of 0.2 lm W21 at 100 cd m22 and 12.8 V. With a
solid state PLQY of 11% the maximum external quantum
efficiency of the device could be 2.5% taking into account
an outcoupling of light of 20% from the device. We can
understand the lack of optimum performance from the device
by considering the HOMO and LUMO energies of Re5 relative
to Irppy3. The oxidation of Re5 is around 0.5 V more positive
than that of Irppy3 meaning that its HOMO energy is around
6.13 eV. In contrast the reduction potential of Re5 is
approximately 1.1 V less negative than Irppy3 meaning that
the LUMO has an energy of 3.6 eV. The electrochemical
results therefore suggest that a calcium cathode should have at
least an ohmic contact with the dendrimer while there will be
a large barrier to hole injection. That is, there will be an
imbalance of charge injection with electrons being injected
more readily than holes. The performance of the devices could
be improved by the use of an insoluble hole transport layer and
by blending the dendrimers with a host material.
Conclusion
We have demonstrated that the light-emitting dendrimer
technology can be used in conjunction with rhenium(I)
complex emitters. By attaching dendrons strategically to the
ligands of rhenium(I) it is possible to prepare solution
processable materials that have enhanced electrochemical
stability and control the intermolecular interactions that cause
solvatochromism in solution and PL quenching in the solid
state. Finally, we have shown that in spite of modest solid state
PLQYs and unbalanced charge injection reasonably efficient
devices can be prepared.
Fig. 7 Cyclic voltammograms of Re1–6 (the currents for the
oxidations and reductions for each material are shown on the left
and right axes respectively).
Experimental
conditions the current decreased on consecutive scans indicat-
ing that the redox process was still not fully chemically
reversible under the experimental conditions. Nevertheless,
Unless otherwise noted, all chemicals were obtained from
commercial suppliers and used as received. Melting points
4260 | J. Mater. Chem., 2007, 17, 4255–4264
This journal is ß The Royal Society of Chemistry 2007

View Article Online
were measured on a Leica Galen III hotstage and are
1
uncorrected. The H-NMR spectra were measured in deuter-
of 300 or 360 nm. Spectra were corrected after measurement
using the emission calibration obtained from measuring a
calibrated lamp spectrum. PLQYs were measured by a relative
method using quinine sulfate in 0.5 M sulfuric acid as a
standard.32 The error in this method is estimated to be
approximately 10% of the measured value. Films for the
solid state PLQYs were spin-coated from a chloroform or
dichloromethane solution with a dendrimer concentration of
10 mg ml21 at 1000 rpm for 1 min to give a thickness of about
150 nm. Their PLQYs were measured using an integrating
sphere in accordance with Greenham et al. with a Helium
Cadmium laser (Kimmon) as the excitation source.33 The
excitation intensity was 0.2 mW at 325 nm and the sphere was
purged with nitrogen during the measurements.
Devices were made from solutions spin-coated at 20 mg ml21
concentration using chloroform or dichloromethane onto
cleaned and plasma etched ITO substrates. The samples were
then transferred to a vacuum evaporator capable of base
pressures of ,2 6 1026. A metal cathode of 20 nm of calcium
followed by a capping layer of 100 nm of aluminium was
deposited.
ated solvents with either Bruker DPX 400 MHz, AVB
400 MHz, or AVC 500 MHz spectrometers. All J values are
quoted in Hertz to the nearest 0.5 Hz. The phenyl rings in the
dendron are denoted sp-H (surface phenyl H) and bp-H
(branch phenyl H), and the ligand protons are described as
L-H. Microanalyses were carried out in the Inorganic
Chemistry Laboratory, Oxford University, UK or by the
Elemental Analysis Service, London Metropolitan University,
UK. The UV-visible absorption spectra were recorded as
solutions in HPLC grade dichloromethane with a Perkin-
Elmer UV-vis Lambda 25 spectrometer. IR spectra were
recordered on a Perkin Elmer Paragon 1000 Infrared spectro-
meter as film on a sodium chloride plate. Mass spectra were
recorded on an Applied Biosystems Voyager matrix-assisted
laser desorption/ionisation-time-of-flight (MALDI-TOF)
from trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]
malononitrile (DCTB) in positive reflectron mode or a
Finnigan MAT 95XP at the EPSRC National Mass
Spectrometry Centre, Swansea, UK, or a Bruker FT-MS
Apex III (EI). Thermal gravimetric analysis was performed
L2. A mixture of G1 boronic acid (3.72 g, 7.01 mmol),
5-bromo-1,10-phenanthroline34 (1.24 g, 4.79 mmol), tetrakis-
(triphenylphosphine)palladium(0) (162 mg, 0.140 mmol),
aqueous sodium carbonate (2.1 M, 2.5 cm3), ethanol
(2.5 cm3), and toluene (13 cm3) was degassed and placed
under nitrogen, and then heated at 100 uC for 24 h. The
mixture was allowed to cool to room temperature and then
dichloromethane (100 cm3) was added. The organic layer was
washed with water (3 6 100 cm3), dried over magnesium
sulfate, and filtered. The solvent was removed and the residue
was purified by column chromatography over silica using a
methanol–ethyl acetate mixture (1 : 10) as eluent to give L2
(1.70 g, 54%). Found: C, 83.15; H, 8.0; N, 4.7. C46H52N2O2
requires C, 83.1; H, 7.9; N, 4.2%; lmax(CH2Cl2)/nm 271
[loge/dm3mol21cm21 (5.1)]; dH(500 MHz, CDCl3) 0.87–1.03
(12 H, m, CH3), 1.32–1.59 (16 H, m, CH2), 1.77 (2 H, m, CH),
3.92 (4 H, m, OCH2), 7.02 (4 H, 1/2AABB9, sp-H), 7.61 (1 H,
dd, J = 8, J = 4, L-H), 7.62–7.65 (6 H, m, sp-H and bp-H), 7.68
(1 H, dd, J = 8, J = 4, L-H), 7.86 (1 H, s, L-H), 7.87 (1 H, dd,
J = 1.5, J = 1.5, bp-H), 8.28 (1 H, dd, J = 8.5, J = 1.5 Hz),
8.43 (1 H, dd, J = 8.5, J = 1.5, L-H), and 9.23 (2H, m, L-H);
dC(125 MHz, CDCl3) 11.1, 14.1, 23.0, 23.9, 29.1, 30.5, 39.4,
70.6, 114.9, 122.9, 123.4, 124.8, 126.49, 126.52, 128.0, 128.1,
128.2, 132.8, 134.7, 136.0, 139.0, 139.7, 141.8, 145.6, 146.3,
150.1, 150.2, 159.3. m/z (EI: M+) Found: 664.4 (24%).
C46H52N2O2 requires 664.9.
on
a Perkin-Elmer thermogravimetric analyzer TGA7.
Differential Scanning Calorimetry (DSC) was performed on
a Perkin Elmer Pyris 1. The samples for DSC analysis were
prepared by transferring a toluene solution of the dendrimer
into the aluminium DSC pan and then the toluene was
removed by heating at 100 uC. The sample was then dried
under vacuum at room temperature for 1 h before the mea-
surement. Light petroleum refers to the fraction of boiling
point 40–60 uC. When solvent mixtures are used for chromato-
graphy over silica, the proportions are given by volume. The
alumina used for chromatography was basic Brockmann grade
I. Electrochemistry was performed using an EG&G Princeton
Applied Research potentiostat/galvanostat model 263A. All
measurements were made at room temperature on samples
dissolved in dichloromethane or N,N-dimethylformamide with
0.1 M tetra-n-butylammonium tetrafluoroborate as the elec-
trolyte. The sample concentration was 1 mM, and a platinum
working electrode, platinum counter electrode in 0.1 M tetra-
n-butylammonium tetrafluoroborate in the same solvent as
used for the sample, and a Ag/0.1 M AgNO3 in acetonitrile
reference electrode were used. The scan rate was 100 mV s21
.
The electrolyte was purified by recrystallization from a mixture
of ethylacetate and diethylether. The solutions were deoxy-
genated with argon. The ferricenium/ferrocene couple was
used as the standard30 and the ferrocene was purified by sub-
limation. All potentials are quoted relative to the ferricenium/
ferrocene couple. In all cases several scans were carried out to
confirm the chemical reversibility of the redox processes. For
room temperature photoluminescence measurements, samples
were dissolved in spectroscopic grade tetrahydrofuran in
quartz degassing cuvettes, degassed by three freeze–pump–
thaw cycles, sealed under vacuum, and warmed to nominal
room temperature in a bath of tepid water. The optical density
(OD) of the samples and standard were similar and small
(¡0.1 at ¢360 nm). Photoluminescence spectra in solution
were recorded using a Jobin Yvon Fluoromax 2 fluorimeter, at
the highest spectral resolution, using an excitation wavelength
Re2.
A solution of L2 (127 mg, 0.191 mmol) and
pentacarbonylchlororhenium(I) (77.5 mg, 0.214 mmol) in
toluene (10 cm3) was heated at reflux under nitrogen for
2.0 h. The solution was allowed to cool to room temperature
and the solvent was removed. The residue was purified by
column chromatography over silica using a light petroleum–
dichloromethane mixture (1 : 4) to give a yellow solid of
Re2 (96 mg, 52%). Found: C, 60.5; H, 5.5; N, 2.8. C49H52-
ClN2O5Re requires C, 60.6; H, 5.4; N, 2.9%; nmax(thin film)/
cm21 1893 (CO), 1916 (CO), and 2020 (CO); lmax(CH2Cl2)/nm
273 [loge/dm3mol21cm21 (4.85)], 329sh (3.93), and 383sh
(3.72); dH(400 MHz, CDCl3) 0.90–0.99 (12 H, m, CH3),
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4255–4264 | 4261

View Article Online
1.31–1.60 (16 H, m, CH2), 1.78 (2 H, m, CH), 3.92 (4 H, m,
OCH2), 7.03 (4 H, 1/2AA9BB9, sp-H), 7.59–7.63 (6 H, m, sp-H
and bp-H), 7.81 (1 H, dd, J = 8.5, J = 5, L-H), 7.89–7.92 (2 H,
m, bp-H and L-H), 8.06 (1 H, s), 8.56 (1 H, dd, J = 8.5, J = 1.5,
L-H), 8.69 (1 H, dd, J = 8.5, J = 1.5, L-H), and 9.43 (2 H, m,
L-H); dC(125 MHz, CDCl3) 11.1, 14.1, 23.0, 23.9, 29.1, 30.5,
39.4, 70.7, 115.1, 125.6, 125.7, 126.1, 126.3, 127.1, 128.2, 130.2,
130.6, 132.3, 137.2, 137.4, 137.9, 141.0, 142.4, 146.5, 147.4,
152.8, 152.9, 159.6, 189.4 (CO), 196.98 (CO), and 197.03 (CO);
m/z (FAB: M–Cl+) Found: 933.5 (64%), 934.5 (36%),
935.5 (100%), 936.5 (52%), 937.5 (17%), and 938.5 (3%).
C49H52N2O5Re requires 933.3 (54%), 934.3 (31%), 935.3
(100%), 936.3 (53%), 937.3 (15%), and 938.4 (3%); TGA5%
306 uC (2 uC min21).
1419.4 (100%), 1420.5 (76%), 1421.5 (36%), and 1422.5 (13%).
C83H96N2O7Re requires 1417.7 (47%), 1418.7 (44%), 1419.7
(100%), 1420.7 (82%), 1421.7 (38%), and 1422.7 (12%); TGA5%
344 uC (5 uC min21).
Re4. A mixture of G1 boronic acid (3.31 g, 6.24 mmol), 3,8-
dibromo-1,10-phenanthroline 336 (0.846 g, 2.50 mmol), tetra-
kis(triphenylphosphine)palladium(0) (0.223 g, 0.193 mmol),
aqueous sodium carbonate (2.0 M, 3.0 cm3), ethanol (3.0 cm3)
and toluene (9 cm3) was degassed and placed under nitrogen.
The reaction mixture was heated at 100 uC under nitrogen
for 23 h. After the mixture was cooled to room temperature
dichloromethane (100 cm3) was added. The organic layer was
washed with water (3 6 100 cm3), dried over anhydrous
magnesium sulfate, and filtered. The solvent was removed and
the residue was purified by column chromatography over
alumina with dichloromethane as eluent to give a white solid
of L4 (0.380 g, 13%). dH(400 MHz, CDCl3) 0.91–0.98 (24 H,
m, CH3), 1.30–1.60 (32 H, m, CH2), 1.78 (4 H, m, CH), 3.94
(8 H, m, OCH2), 7.05 and 7.67 (16 H, AA9BB9, sp-H), 7.83
(2 H, dd, J = 1.5, J = 1.5, bp-H), 7.87 (4 H, d, J = 1.5, bp-H),
7.98 (2H, s, L-H), 8.58 (2 H, s, L-H), and 9.60 (2 H, s, L-H). L4
was used for the complexation reaction without further
purification. L4 (0.419 g, 0.364 mmol) and pentacarbonyl-
chlororhenium(I) (0.133 g, 0.368 mmol) in toluene (18 cm3)
was heated at reflux for 2.0 h. After the solution was allowed
to cool to room temperature the solvent was removed, and the
residue was purified by column chromatography over silica
using a light petroleum–dichloromethane mixture (1 : 3) as
eluent to give a yellow solid of Re4 (0.126 g, 24%). Mp 260–
263 uC (found: C, 68.6; H, 7.1; N, 2.1. C83H96ClN2O7Re
requires C, 68.5; H, 6.65; N, 1.9%); nmax(thin film)/cm21 1883
(CO), 1923 (CO), and 2023 (CO); lmax(CH2Cl2)/nm 277
[loge/dm3mol21cm21 (5.11)], 346 (4.53), and 420sh (3.45);
dH(400 MHz, CDCl3) 0.90–1.00 (24 H, m, CH3), 1.33–1.61
(32 H, m, CH2), 1.78 (4 H, m, CH), 3.91 (8 H, m, OCH2), 7.05
and 7.63 (16 H, AA9BB9, sp-H), 7.74 (4 H, d, J = 1.5, bp-H),
7.84 (2 H, bm, bp-H), 7.97 (2H, s, L-H), 8.59 (2 H, d, J = 1.5,
L-H), 9.62 (2 H, d, J = 1.5, L-H); dC(125 MHz, CDCl3) 11.2,
14.2, 23.2, 23.9, 29.2, 30.6, 39.5, 70.7. 115.1, 124.0, 126.5,
127.9, 128.4, 130.3, 132.2, 135.0, 135.6, 138.9, 143.0, 145.3,
151.8, 159.7, 189.8 (CO), and 197.1 (CO). m/z (FAB: M–Cl+)
Found: 1417.5 (72%), 1418.5 (61%), 1419.4 (100%), 1420.5
(74%), 1421.4 (41%), and 1422.4 (19%). C83H96N2O7Re
requires 1417.7 (47%), 1418.7 (44%), 1419.7 (100%), 1420.7
(82%), 1421.7 (38%), and 1422.7 (12%); TGA5% 276 uC
(2 uC min21).
Re3. n-Butyllithium (1.6 M in hexane, 2.4 cm3, 3.84 mmol)
was added to a solution of G1-Br25 (2.01 g, 3.55 mmol) in
tetrahydrofuran (74 cm3) cooled in an acetone–dry ice bath
under nitrogen. The solution was stirred with acetone–dry ice
bath cooling for 1 h. Zinc chloride (0.50 g, 3.67 mmol) in
tetrahydrofuran (19 cm3) was added to the solution cooled in
an acetone–dry ice bath, and then the solution was allowed
to warm to room temperature and stirred for 30 min.
4,7-Dibromo-1,10-phenanthroline 235 (0.403 g, 1.19 mmol)
and tetrakis(triphenylphosphine)palladium(0) (0.026 g,
0.0225 mmol) were added to the solution and then the solution
was heated at 80 uC for 17 h. The solvent was removed and the
residue was dissolved in dichloromethane (300 cm3). The
organic layer was washed with saturated aqueous ethylenedi-
amine tetraacetic acid (3 6 200 cm3) and water (2 6 200 cm3),
dried over anhydrous magnesium sulfate, and filtered. The
solvent was removed and the residue was partially purified by
column chromatography over silica using ethylacetate and
then a methanol–dichloromethane mixture (8 : 92) as the
eluent to give L3 as pale brownish solid (#0.181 g, #13%).
dH(400 MHz, CDCl3) 0.88–0.97 (24 H, m, CH3), 1.30–1.60
(32 H, m, CH2), 1.75 (4, H, m, CH), 3.88 (8 H, m, OCH2), 6.99
(8 H, 1/2AA9BB9, sp-H), 7.59–7.63 (12 H, m, sp-H and bp-H),
7.73 (2 H, d, J = 4, L-H), 7.85 (2 H, dd, J = 1, J = 1, bp-H),
8.02 (2H, s, L-H), and 9.32 (2 H, d, J = 4, L-H). L3 was used in
the complexation reaction without further purification. A
solution of L3 (0.181 g, 0.157 mmol) and pentacarbonyl-
chlororhenium(I) (0.061 g, 0.169 mmol) in toluene (1.0 cm3)
was heated at reflux for 1.5 h. The solution was allowed to cool
to room temperature and the solvent removed. The residue
was purified by column chromatography over silica using a
light petroleum–dichloromethane mixture (1 : 2) as the eluent
gave a yellow solid of Re3 (33 mg, 14%). Found: C, 68.5; H,
6.7; N, 2.0. C83H96ClN2O7Re requires C, 68.5; H, 6.65; N,
1.9%; nmax(thin film)/cm21 1888 (CO), 1920 (CO), and 2020
Re5. n-Butyllithium (1.6 M in hexane, 1.9 cm3, 3.04 mmol)
was added to a solution of G1-Br25 (1.51 g, 2.67 mmol) in
diethyl ether (2.7 cm3) cooled in an acetone–dry ice bath under
nitrogen, and then the solution was allowed to warm to room
temperature and stirred for 1.0 h. The mixture was then cooled
in an acetone–dry ice bath and a solution of 1,10-phenanthro-
line 4 (0.12 g, 0.67 mmol) in toluene (9 cm3) was added slowly.
The reaction mixture was allowed to warm to room
temperature and then stirred for 18 h. A small amount of
water and dichloromethane (100 cm3) were added, and then
the mixture was washed with water (150 cm3), dried over
anhydrous magnesium sulfate, and filtered. The solvent was
removed and the residue was dissolved in dichloromethane
(CO);
l
max(CH2Cl2)/nm 277 [loge/dm3mol21cm21 (5.05)],
331sh (4.23), and 385sh (3.98); dH(400 MHz, CDCl3) 0.90–
0.99 (24 H, m, CH3), 1.30–1.60 (32 H, m, CH2), 1.76 (4 H, m,
CH), 3.90 (8 H, m, OCH2), 7.01 (8 H, 1/2AA9BB9, sp-H), 7.58–
7.62 (12 H, m, sp-H and bp-H), 7.90–7.93 (4 H, m, bp-H and
L-H), 8.19 (2 H, s, L-H), and 9.48 (2 H, d, J = 5, L-H);
dC(125 MHz, CDCl3) 11.1, 14.1, 23.0, 23.8, 29.1, 30.5, 39.4,
70.6. 115.1, 125.8, 125.9, 126.1, 126.4, 128.2, 129.2, 132.1,
136.5, 142.5, 147.7. 151.4, 152.5, 159.6, 189.7 (CO), and 197.2
(CO); m/z (FAB: M–Cl+) Found: 1417.5 (62%), 1418.5 (55%),
4262 | J. Mater. Chem., 2007, 17, 4255–4264
This journal is ß The Royal Society of Chemistry 2007

View Article Online
(12 cm3) and activated manganese dioxide (3.06 g) was added.
The mixture was stirred at room temperature for 1.0 h and
then the reaction mixture was filtered through celite. The
filtrate was collected and the solvent removed. The residue was
purified by column chromatography over alumina using a light
petroleum–dichloromethane mixture (3 : 1) as the eluent to
give a brownish solid of L5 (0.432 g, 56%). dH(400 MHz,
CDCl3) 0.90–1.00 (24 H, m, CH3), 1.33–1.61 (32 H, m, CH2),
1.78 (4 H, m, CH), 3.85 (8 H, m, OCH2), 6.89 and 7.71 (8 H,
AA9BB9, sp-H), 7.81 (2 H, s, L-H), 7.88 (2H, bs, bp-H), 8.25
(2 H, d, J = 8.5, L-H), 8.34 (2 H, d, J = 8.5, L-H), and 8.58
(4 H, bs, bp-H). L5 was used for the complexation reaction
without further purification. L5 (0.432 g, 0.376 mmol) and
pentacarbonylchlororhenium(I) (0.136 g, 0.376 mmol) in
toluene (19 cm3) was heated at reflux for 2.0 h. The solution
was allowed to cool and the solvent was removed. The residue
was purified by column chromatography over silica using a
light petroleum–dichloromethane mixture (1 : 5) as the eluent
to give a yellow solid of Re5 (0.235 g, 43%). Found: C, 68.5; H,
6.65; N, 1.9. C83H96ClN2O7Re requires C, 68.5; H, 6.65; N,
1.9%; nmax(thin film)/cm21 1884 (CO), 1921 (CO), and 2019
1634.3. L6 was used for the complexation reaction without
further purification. L6 (0.204 g, 0.125 mmol) and penta-
carbonylchlororhenium(I) (0.050 g, 0.14 mmol) in toluene
(1.0 cm3) was heated at reflux for 1.0 h. The solution was
cooled to room temperature and the solvent removed. The
residue was purified by column chromatography over silica
using a light petroleum–dichloromethane mixture (1 : 2) as the
eluent to give a yellow solid of Re6 (0.042 g, 17%). Found: C,
72.5; H, 7.2; N, 1.5. C117H140ClN2O9Re requires C, 72.4; H,
7.3; N, 1.4%; nmax(thin film)/cm21 1885 (CO), 1926 (CO), and
2022 (CO); lmax(CH2Cl2)/nm 272 [loge/dm3mol21cm21 (5.25)],
338sh (4.40), and 400sh (3.73); dH(400 MHz, CDCl3) d 0.90–
1.00 (36 H, m, CH3), 1.27–1.61 (48 H, m, CH2), 1.76 (6H, m,
CH), 3.90 (12 H, m, OCH2), 6.97–7.10 [12 H, m (3 6
1/2AA9BB9), sp-H], 7.63–7.75 (14 H, m, sp-H and bp-H), 7.77–
7.81 (3 H, m, bp-H), 7.83 (1 H, bs, bp-H), 7.90–7.94 (4 H, m,
bp-H and L-H), 8.01 (1 H, d, J = 8.5, L-H,), 8.12 (1 H, s, L-H),
8.57 (1 H, d, J = 8.5, L-H), and 8.73 (1 H, d, J = 8.5, L-H); m/z
(MALDI: M+) Found: 1937.0 (38%), 1938.0 (48%), 1939.0
(97%), 1940.0 (100%), 1941.0 (74%), 1942.0 (45%), 1943.0
(24%), and 1944.0 (10%). C117H140ClN2O9Re requires 1937.0
(33%), 1938.0 (44%), 1939.0 (95%), 1940.0 (100%), 1941.0
(80%), 1942.0 (50%), 1943.0 (25%), and 1944.0 (9%); TGA5%
356 uC (5 uC min21).
(CO);
l
max(CH2Cl2)/nm 273 [loge/dm3mol21cm21 (5.05)],
340sh (4.18), and 395sh (3.64); dH(500 MHz, CDCl3) 0.90–
1.00 (24 H, m, CH3), 1.34–1.60 (32 H, m, CH2), 1.78 (4 H, m,
CH), 3.94 (8 H, m, OCH2), 6.97–7.02 [8 H, m (2 6
1/2AA9BB9), sp-H], 7.66 (4 H, 1/2AA9BB9, sp-H), 7.71 (4 H,
1/2AA9BB9, sp-H), 7.77 (2H, bm, bp-H), 7.81 (2 H, bm, bp-H),
7.91 (2 H, dd, J = 1.5, J = 1.5 bp-H), 7.97 (2 H, d, J = 8.5,
L-H), 8.06 (2 H, s, L-H), and 8.55 (2 H, d, J = 8.5 L-H);
dC(125 MHz, CDCl3) 11.1, 14.1, 23.0, 23.8, 29.1, 30.5, 39.3,
70.5, 114.9, 125.7, 126.0, 126.88, 126.92, 127.2, 128.4, 128.6,
130.0, 132.61, 132.65, 137.9, 141.5, 142.3, 142.8, 148.2, 159.2,
159.3, 164.8, 192.1, and 192.6; m/z (FAB: M+) Found: 1452.4
(58%), 1453.4 (50%), 1454.4 (100%), 1455.4 (79%), 1456.4
(58%), 1457.4 (42%), and 1458.4 (26%). C83H96ClN2O7Re
requires 1452.6 (41%), 1453.7 (39%), 1454.6 (100%), 1455.7
(83%), 1456.6 (60%), 1457.7 (33%), and 1458.6 (13%); TGA5%
320 uC (2 uC min21).
Acknowledgements
Y.-J. Pu thanks Japan Society for the Promotion of Science.
References
1 M. A. Baldo, D. F. O’Brien, Y. You, A. Shoustikov, S. Sibley,
M. E. Thompson and S. R. Forrest, Nature, 1998, 395, 151.
2 C. Adachi, M. A. Baldo, S. R. Forrest and M. E. Thompson, Appl.
Phys. Lett., 2000, 77, 904.
3 W.-Y. Wong, C.-L. Ho, Z.-Q. Gao, B.-X. Mi, C.-H. Chen,
K.-W. Cheah and Z. Lin, Angew. Chem., Int. Ed., 2007, 46,
1558.
4 S.-Y. Takizawa, Y. Sasaki, M. Akhtaruzzaman, H. Echizen,
J.-I. Nishida, T. Iwata, S. Tokito and Y. Yamashita, J. Mater.
Chem., 2007, 17, 841.
Re6. n-Butyllithium (2.8 cm3, 1.6 M in hexane, 4.48 mmol)
was added to a solution of G1-Br25 (2.15 g, 3.80 mmol) in
diethyl ether (3.8 cm3) cooled in an acetone–dry ice bath under
nitrogen, and then the solution was allowed to warm to room
temperature and stirred for 1.0 h. The reaction mixture was
then cooled in an acetone–dry ice bath and a solution of L2
(0.63 g, 0.95 mmol) in toluene (12 cm3) was added slowly.
After addition the reaction mixture was allowed to warm to
room temperature and was stirred for 21 h. A small amount of
water and dichloromethane (100 cm3) were added and the
mixture was washed with water (150 cm3). The organic layer
was collected, dried over anhydrous magnesium sulfate, and
filtered. The solvent was removed and the residue was
dissolved in dichloromethane (17 cm3) and activated manga-
nese dioxide (8.0 g) was added. The mixture was stirred at
room temperature for 1.0 h before being filtered through celite.
The filtrate was collected and the solvent removed. The residue
was purified by column chromatography over basic alumina
using a light petroleum–dichloromethane mixture (5 : 1) as
the eluent to give a yellowish solid of L6 (0.710 g, 46%). m/z
(MALDI: M+) Found: 1634.1 (100%). C114H140N2O requires
5 S.-Y. Kim, J.-H. Kim, Y. Ha, S.-H. Lee, J.-H. Seo and Y.-K. Kim,
Curr. Appl. Phys., 2007, 7, 380.
6 S.-C. Lo, C. P. Shipley, R. N. Bera, R. E. Harding, A. R. Cowley,
P. L. Burn and I. D. W. Samuel, Chem. Mater., 2006, 18, 5119.
7 Y.-Y. Lu, C.-C. Ju, D. Guo, Z.-B. Deng and K.-Z. Wang, J. Phys.
Chem. C, 2007, 111, 5211.
8 N. J. Lundin, P. J. Walsh, S. L. Howell, J. J. McGarvey,
A. G. Blackman and K. C. Gordon, Inorg. Chem., 2005, 44,
3551.
9 Z. Ming, L. Ping, X. M. Wang, H. Lin, X. Hong, Z. Wu, Y. Bing,
L. L. Liu, Y. Li, Y. Min, Y. G. Ma, J. K. Feng, D. J. Wang and
N. Tamai, J. Phys. Chem. B, 2004, 108, 13185.
10 B. Li, M. T. Li, Z. R. Hong, W. L. Li, T. Z. Yu and H. Z. Wei,
Appl. Phys. Lett., 2004, 85, 4786.
11 F. Li, M. Zhang, G. Cheng, J. Feng, Y. Zhao, Y. G. Ma, S. Y. Liu
and J. C. Shen, Appl. Phys. Lett., 2004, 84, 148–150.
12 S. Ranjan, S. Y. Lin, K. C. Hwang, Y. Chi, W. L. Ching, C. S. Liu,
Y. T. Tao, C. H. Chien, S. M. Peng and G. H. Lee, Inorg. Chem.,
2003, 42, 1248.
13 Y. Q. Li, Y. Liu, J. H. Guo, F. Wu, W. J. Tian, B. F. Li and
Y. Wang, Synth. Met., 2001, 118, 175.
14 W. K. Chan, P. K. Ng, X. Gong and S. J. Hou, Appl. Phys. Lett.,
1999, 75, 3920.
15 X. Gong, P. K. Ng and W. K. Chan, Adv. Mater., 1998, 10,
1337.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 4255–4264 | 4263

View Article Online
16 J. P. J. Markham, S.-C. Lo, S. W. Magennis, P. L. Burn and
I. D. W. Samuel, Appl. Phys. Lett., 2002, 80, 2645.
17 T. D. Anthopoulos, J. P. J. Markham, E. B. Namdas,
J. R. Lawrence, I. D. W. Samuel, S.-C. Lo and P. L. Burn, Org.
Electron., 2003, 4, 71.
26 P. J. Giordano and M. S. Wrighton, J. Am. Chem. Soc., 1979, 101,
2888.
27 M. Wrighton and D. L. Morse, J. Am. Chem. Soc., 1974, 96, 998.
28 T. G. Kotch, A. J. Lees, S. J. Fuerniss, K. I. Papathomas and
R. W. Snyder, Inorg. Chem., 1993, 32, 2570.
18 S.-C. Lo, T. D. Anthopoulos, E. B. Namdas, P. L. Burn and
I. D. W. Samuel, Adv. Mater., 2005, 17, 1945.
19 G. J. Zhuo, W.-Y. Wong, B. Yao, Z. Y. Xie and L. X. Wang,
Angew. Chem., Int. Ed., 2007, 46, 1149.
20 T. Tsuzuki, N. Shirasawa, T. Suzuki and S. Tokito, Jpn. J. Appl.
Phys., 2005, 44, 4151.
29 M. Pope and C. E. Swenberg, Electronic Processes in Organic
Crystals and Polymers, Oxford University Press, New York,
2nd edn, 1999.
30 G. Gritzner and J. Kuta, Pure Appl. Chem., 1984, 56, 461.
31 J. C. Luong, L. Nadjo and M. S. Wrighton, J. Am. Chem. Soc.,
1978, 100, 5790.
21 J. Ding, J. Gao, Y. Cheng, Z. Xie, L. Wang, D. Ma, X. Jing and
F. Wang, Adv. Funct. Mater., 2006, 16, 575.
22 S.-C. Lo, E. B. Namdas, C. P. Shipley, J. P. J. Markham,
T. D. Anthopolous, P. L. Burn and I. D. W. Samuel, Org.
Electron., 2006, 7, 85.
32 J. N. Demas and G. A. Crosby, J. Phys. Chem., 1971, 75, 991.
33 N. C. Greenham, I. D. W. Samuel, G. R. Hayes, R. T. Phillips,
Y. Kessener, S. C. Moratti, A. B. Holmes and R. H. Friend, Chem.
Phys. Lett., 1995, 241, 89.
34 M. Hissler, W. B. Connick, D. K. Geiger, J. E. McGarrah,
D. Lipa, R. J. Lachicotte and R. Eisenberg, Inorg. Chem., 2000, 39,
447.
23 J. P. J. Markham, I. D. W. Samuel, S. C. Lo, P. L. Burn, M. Weiter
and H. Ba¨ssler, J. Appl. Phys., 2004, 95, 438.
24 E. B. Namdas, A. Ruseckas, I. D. W. Samuel, S.-C. Lo and
P. L. Burn, J. Phys. Chem. B, 2004, 108, 1570.
25 S.-C. Lo, E. B. Namdas, P. L. Burn and I. D. W. Samuel,
Macromolecules, 2003, 36, 9721.
35 G. I. Graf, D. Hastreiter, L. E. da Silva, R. A. Rebelo,
A. G. Montalban and A. McKillop, Tetrahedron, 2002, 58, 9095.
36 Y. Saitoh, T. Koizumi, K. Osakada and T. Yamamoto, Can. J.
Chem., 1997, 75, 1336–1339.
4264 | J. Mater. Chem., 2007, 17, 4255–4264
This journal is ß The Royal Society of Chemistry 2007