A rapid route to carbazole containing dendrons and phosphorescent dendrimers

Article abstract of DOI:10.1039/b717750j

A convergent strategy for the synthesis of three generations of dendrons comprised of carbazole moieties is described. The procedure to build the dendrons involves an iterative palladium catalysed amination-debenzylation sequence using N-benzyl-3,6-dibromocarbazole. The three carbazolyl focussed dendrons are then attached to a reactive fac-tris[2-phenylpyridyl]iridium(iii) core by a palladium catalysed amination to give the dendrimers. The three generations of dendrons have one, three, and seven carbazole units leading to dendrimers with fac-tris[2-phenylpyridyl]iridium(iii) cores and three, nine and twenty one carbazole units. The use of 9,9′-dialkylfluorenyl surface groups gave the dendrimers excellent solubility. The attachment of the carbazolyl-based dendrons did not change the emission colour significantly with the dendrimers emitting green phosphorescence. The dendrimers were highly luminescent with solution photoluminescence quantum yields of the order of 70%. Ground state molecular orbital calculations showed that while the LUMO was concentrated on the core iridium(iii) complex the HOMO was delocalised across the core and each of the dendrons. This was reflected in the oxidation properties of the dendrimers whereby the increased carbazolyl character of the HOMO resulted in the first oxidation being moved to more positive potentials. The Royal Society of Chemistry 2008.

Full text of DOI:10.1039/b717750j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
A rapid route to carbazole containing dendrons and phosphorescent
dendrimers†
Kevin A. Knights,a Stuart G. Stevenson,c Christopher P. Shipley,a Shih-Chun Lo,a Seth Olsen,b
c
Ruth E. Harding, Salvatore Gambino, Paul L. Burn and Ifor D. W. Samuel
c
b
c
*
*
Received 16th November 2007, Accepted 21st February 2008
First published as an Advance Article on the web 14th March 2008
DOI: 10.1039/b717750j
A convergent strategy for the synthesis of three generations of dendrons comprised of carbazole
moieties is described. The procedure to build the dendrons involves an iterative palladium catalysed
amination–debenzylation sequence using N-benzyl-3,6-dibromocarbazole. The three carbazolyl
focussed dendrons are then attached to a reactive fac-tris[2-phenylpyridyl]iridium(III) core by
a palladium catalysed amination to give the dendrimers. The three generations of dendrons have one,
three, and seven carbazole units leading to dendrimers with fac-tris[2-phenylpyridyl]iridium(III) cores
and three, nine and twenty one carbazole units. The use of 9,90-dialkylfluorenyl surface groups gave the
dendrimers excellent solubility. The attachment of the carbazolyl-based dendrons did not change the
emission colour significantly with the dendrimers emitting green phosphorescence. The dendrimers
were highly luminescent with solution photoluminescence quantum yields of the order of 70%. Ground
state molecular orbital calculations showed that while the ‘‘LUMO’’ was concentrated on the core
iridium(III) complex the ‘‘HOMO’’ was delocalised across the core and each of the dendrons. This was
reflected in the oxidation properties of the dendrimers whereby the increased carbazolyl character of
the ‘‘HOMO’’ resulted in the first oxidation being moved to more positive potentials.
in small molecule hosts incorporated into the dendrons to form
Introduction
a covalently linked guest–host system where the number of
carbazolyl to emissive units is precisely known and the
intermolecular interactions that govern the luminescence proper-
ties are controlled at a molecular level by the dendrimer
generation.20–25 This has been in part due to the lack of efficient
routes to higher generation carbazolyl containing dendrons. In
addition, previous routes to phosphorescent dendrimers have
generally involved the preparation of the dendronised ligand,
which is subsequently reacted with an iridium(III) salt to give
the dendrimer.23,24,26 These latter reactions can sometimes be
low yielding.
Phosphorescent complexes based on iridium(III) are likely to play
an important role in bringing organic light-emitting diodes to
market. In many cases the most efficient devices based on iridiu-
m(III) complexes have the complex blended with a host that is
comprised of carbazole units.1–7 Phosphorescent complexes are
sensitive to concentration quenching of the luminescence and
hence one of the critical roles the host plays is to keep the emis-
sive chromophores separated. However, while the guest–host
composition may be well defined it is much more difficult to
determine what the distribution and arrangement of the two
components are in the deposited film. There is now a significant
effort towards controlling the intermolecular interactions of
iridium(III) complexes in the solid state by using dendritic archi-
tectures.8–13 Until recently the most efficient dendrimer based
light-emitting diodes (DLEDs) had the light-emitting dendrimers
blended with carbazole containing hosts14–17 although hostless
dendrimer devices have now been reported,18 and the efficiencies
of the devices have been found to be strongly dependent on the
processing regime.19 What have been explored in less detail are
dendritic materials that have the carbazolyl units normally found
In this paper we describe a simple iterative convergent proce-
dure that rapidly leads to first, second, and third generation
dendrons. We show that phosphorescent dendrimers with
a fac-tris[2-phenylpyridyl]iridium(III) (Irppy3) core can then be
formed in a single step where the dendrons are reacted with an
activated preformed complex. The third generation dendrimer
has six mole percent (mol%) of (Irppy3) core to carbazole units,
which is similar to the most efficient small molecule devices
with (Irppy3) : 4,40-bis(N,N-carbazolyl)biphenyl (CBP) blends
(4.5 mol%). In addition, we report the optoelectronic properties
of the dendrimers and relate these to their molecular orbital
distribution.
aDepartment of Chemistry, University of Oxford, Chemistry Research
Laboratory, Mansfield Rd, Oxford, UK OX1 3TA
bCentre for Organic Photonics & Electronics, School of Molecular &
Microbial Sciences, Chemistry Building, University of Queensland, St
Lucia, Queensland, 4072, Australia
cOrganic Semiconductor Centre, SUPA, School of Physics and Astronomy,
University of St Andrews, St Andrews, Fife, UK KY16 9SS
† Electronic supplementary information (ESI) available: Cartesian
coordinates of first, second and third generation dendrimer models. See
DOI: 10.1039/b717750j
Results and discussion
Synthesis
When developing the strategies for the preparation of
dendrimers we have focussed on minimising the reaction types
and steps for the formation of the different generations of
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 2121–2130 | 2121

View Article Online
materials. The synthesis of first and second generation carbazole
containing dendrons with tert-butyl surface groups and subse-
quent metal complex formation via a dendronised ligand has
been previously reported.23,27 The procedure for building the
dendrons required the reaction of 3,6-di-tert-butylcarbazole
with 3,6-diiodo-N-tosyl carbazole using the relatively harsh
Ullmann conditions. Although we found this methodology could
be used with the new 9,9-di-n-propylfluorenyl surface groups
used in this work the purification of the second and third gener-
ation dendrons was found to be difficult and the yield of the third
generation dendron was low with significant detosylation occur-
ring during the Ullmann coupling. We therefore developed
a much simpler convergent procedure using Buchwald type
aminations with N-benzyl protected 3,6-dibromocarbazole as
the basic building block. Once the fluorenyl surface group was
attached the third generation dendron could be produced in
four simple steps. Scheme 1 shows the strategy for the formation
of the three generations of dendrons and subsequent attachment
to the core to form the dendrimers. The first step in the synthesis
was the reaction of 9,9-di-n-propylfluorenyl-2-boronic acid28
with 3,6-dibromocarbazole.29 Under the Suzuki conditions
used there was no need to N-protect the carbazole unit and the
first generation dendron 3 was formed in an excellent 76% yield.
Two equivalents of the first generation dendron were then
reacted with N-benzyl-3,6-dibromocarbazole 4 using tris(diben-
zylideneacetone)dipalladium(0) as the catalyst. We found that
Scheme 1 Reagents and conditions: i) PhMe, EtOH, Na2CO3(aq), Pd(PPh3)4, 100 ꢀC. ii) NaOtBu, xylenes or toluene, HPtBu3$BF4, Pd2(dba)3, heat
(for the dendrimer formation the reaction also contained fac-tris[2-(3-bromophenyl)pyridyl]iridium(III)). iii) AlCl3, MeOPh, heat.
2122 | J. Mater. Chem., 2008, 18, 2121–2130
This journal is ª The Royal Society of Chemistry 2008

View Article Online
using tri-tert-butylphosphonium tetrafluoroborate gave more
reliable yields than the simpler tri-tert-butylphosphine and that
the benzyl-protected second generation dendron 5 could be
formed in an 80% yield. Deprotection of 5 was achieved by
treatment with aluminium chloride in anisole at 80 ꢀC and under
these conditions 6 was isolated in 51% yield. The cycle could be
repeated to give the third generation benzyl-protected dendron
7 and deprotected material 8 in yields of 72% and 63%
respectively. In the past we have normally taken the dendron
and attached it to the ligand before complexation to the
iridium(III). However, the formation of iridium complexes can
be capricious so in this synthesis we have changed strategy.
The final step in the synthesis of the three generations of
dendrimers was the tris-dendronisation of the fac-tris[2-(3-bro-
mophenyl)pyridyl]iridium(III) complex. The palladium catalysed
aminations were carried out under the same conditions as for the
dendron formation and the first (9), second (10), and third (11)
generation materials were isolated in the respectable yields of
65%, 32% and 34%.
The 1H NMR spectra of the dendrimers were found to be more
complex than those of the dendrimers with biphenyl dendrons.
In particular the aromatic region was complex with some of
the signals, even for the first generation dendrimer, being broad,
which may indicate hindered rotation due to steric encumbrance.
The dendrimers were thermally stable to decomposition, for
example, the first generation only showed significant weight
loss above 385 ꢀC. The hydrodynamic radii of the dendrimers
held further apart by the larger dendrons in the second generation
dendrimer.31 In contrast, it was found that when the branching
moiety was changed from a phenyl group to a carbazolyl moiety
the calculations showed that the HOMO was not only found on
the Irppy3 core but also on the carbazolyl units.24 We were
therefore interested to see how the molecular orbital distribution
and energies correlated with the optical and electronic properties
of the higher generation carbazole dendrimers.
The first part of the study was to calculate the molecular
orbital distribution and energies (see Experimental section for
details of the method). Given the size of the dendrimers and
the concomitant complexity of the calculations we locked the
angles between different aromatic units in the dendrons and their
attachment points to the core at 30ꢀ to take into account the
interactions between neighbouring aromatic units. Of course
steric interactions could differ between different generations
causing deviations from this structural arrangement, and hence
only general trends should be drawn from the results. For
example, if steric interactions caused larger twisting in the
dendrons than allowed for in the calculations then the latter
might give a slightly lower energy gap for the orbitals than is
observed experimentally. Optimisation of a first generation
model indicated a predilection for C3 symmetry—calculations
for higher generation models were performed within this
symmetry group. We found that the HOMO and LUMO
(belonging to the totally symmetric irreducible representation
A) were close in energy to the HOMO ꢁ 1 and HOMO ꢁ 2,
and LUMO + 1 and LUMO + 2 respectively (Fig. 1 illustrates
the results for the third generation dendrimer 11). The HOMO
ꢁ 1 and HOMO ꢁ 2 belong to the E irreducible representation,
as do the LUMO + 1 and LUMO + 2. Therefore, each orbital
pair is degenerate within itself. The energy gap preceding the
next lower and higher sets of orbitals (of E and A symmetry) is
larger by about an order of magnitude; their influence is corre-
spondingly less and it is reasonable not to consider them. The
fact the first three HOMO and the first three LUMO orbitals
are similar in energy means that the orbital distribution can be
considered to be an average of these. Any asymmetric perturba-
tion (for example, deviations from symmetric geometry or the
effects of amorphous environments) will mix each triad within
itself at lowest order. As can be seen in Fig. 1 the molecular
orbital distribution of the ‘‘LUMO’’ (comprised of LUMO,
LUMO + 1, and LUMO + 2) is strongly localised on the core
complex, whereas the ‘‘HOMO’’ (comprised of HOMO,
HOMO ꢁ 1, and HOMO ꢁ 2) has contributions from both
the dendrons and the core complex. Similar results are observed
for the first and second generation dendrimers.
ꢀ
were calculated from the Mvs determined by gel permeation
chromatography using polystyrene standards in combination
with the Hester–Mitchell equation and Mark–Houwink relation-
ship.30 The first generation dendrimer had a hydrodynamic
˚
radius of 10.3 A, which is similar to that of the first generation
biphenyl dendrimer 12.26 It is interesting to note that in spite
of the fact that the aromatic units in the carbazolyl dendrons
of 9 are significantly larger than in the biphenyl dendrons of
12 the hydrodynamic radii are similar. This indicates that for
the biphenyl-based dendrons the 2-ethylhexyloxy groups play
an important role in controlling the intermolecular interactions
of the emissive core. The second generation and third generation
˚
dendrimers, 10 and 11, had hydrodynamic radii of 14.0 A and
˚
18 A respectively; that is, the third generation has a hydrody-
namic radius of almost twice that of the first. The three
dendrimers were found to be monodisperse.
Molecular orbital calculations
Molecular orbital calculations in conjunction with experimental
results can assist in understanding the optical and electronic
properties of new materials. For example, in a previous study of
a first generation Irppy3-cored dendrimer with biphenyl based
dendrons we calculated that the HOMO and LUMO of the
dendrimer were strongly localised on the Irppy3 core.24 The fact
that the orbital density was concentrated on the core complex
was reflected in the difference in the photoluminescence and
charge transporting properties of the first and second generation
dendrimers. It was found that the second generation dendrimer
was more photoluminescent in the solid state and that the hole
mobility was lower. These two properties were consistent with
the optically and electronically active cores being on average
We have analysed the molecular orbital distributions in more
detail and the results for the first generation biphenyl 12 and
carbazolyl dendronised dendrimers 9–11 are summarised in
Table 1. In addition, Fig. 2 uses target diagrams to illustrate
the distribution for dendrimers 9 and 12 to facilitate the explana-
tion. For the first generation dendrimer with biphenyl-based
dendrons 12 the LUMO density resides almost entirely on the
2-phenylpyridyl ligand (97%) with only around 2% on the
iridium, and the remainder on the dendron. The LUMO distri-
bution for the carbazolyl-dendronised dendrimers is essentially
the same indicating that the carbazole units do not strongly
affect the LUMO. As stated earlier the big difference between
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 2121–2130 | 2123

View Article Online
2-phenylpyridyl ligands. In contrast to 12, where there was little
HOMO density on the dendrons, for 9 there was 40% of the
HOMO density on the first layer of carbazolyl units of the
dendrons. For the second generation carbazolyl dendrimer 10
the HOMO distribution had even less on the iridium (10%)
and ligand (21%) with most of the density now being found on
the first (41%) and second (27%) layers of carbazole units. The
dilution of the HOMO density on the iridium(III) complex
continued with the third generation with only 9% found on the
core complex with the remainder on the first (22%), second
(38%), and third (31%) carbazole layers.
The fact that the HOMO is distributed over the whole of the
carbazolyl dendronised dendrimers as opposed to being localised
on the core complex suggests that the hole transport in the
materials should be significantly different to the biphenyl
dendronised dendrimers in which the HOMOs on adjacent cores
are effectively isolated from one another. We have measured the
hole mobility of 9 by the time-of-flight technique and obtain
a value of 6 ꢂ 10ꢁ5 cm2 Vꢁ1 sꢁ1 at a field of 0.2 MV cmꢁ1. This
contrasts with the hole mobility of 4 ꢂ 10ꢁ6 cm2 Vꢁ1 sꢁ1 measured
at the same field for the first generation dendrimer 12.31 The
second and third generation carbazole dendrimers 10 and 11
have even higher mobilities than 9. Therefore, the calculations
give important insight into how the distribution of the molecular
orbitals can lead to an understanding of the improved hole
mobilities in films of the carbazole dendrimers, and how the
carbazole units can enhance hole transport. We were also inter-
ested in seeing whether we could gain some insight into the
nature of the optical properties of the dendrimers and hence
carried out time dependent density functional theory (TD-
DFT) calculations to determine the nature of the lowest energy
excitations. For these calculations we used the same functional
and basis sets as those used for determining the molecular orbital
distributions. From these calculations it was found that the three
lowest energy transitions for all three dendrimers, 9–11, were
triplet in nature. For the first generation dendrimer 9 the three
lowest energies were 2.38, 2.41, and 2.44 eV, while for 10 and
11 they were 2.26, 2.28 and 2.29 eV, and 2.27, 2.29 and
2.30 eV respectively. The first point to note from these calcula-
tions is that the excitation energies for the three dendrimers are
very similar. There is a slight red-shift in moving from the first
to second generation but the second and third generation have
similar energies. The TD-DFT calculations also show that the
three lowest energy transitions are dominated by excitation
combinations between the HOMO, HOMO ꢁ 1 and HOMO
ꢁ 2, and LUMO, LUMO + 1 and LUMO + 2. That is, the
Fig. 1 Frontier orbitals in A and E symmetry for the third generation
dendrimer 11. The HOMO ꢁ 1 and HOMO ꢁ 2, and LUMO + 1 and
LUMO + 2 are degenerate in energy.
the dendrimers with the biphenyl-dendronised dendrimers such
as 12 and the carbazole containing materials is in the HOMO
distribution. For 12 the HOMO distribution is similar to that
for the simple complex with 53% of the density on the iridium,
and 41% and 6% on the ligand and dendron respectively. That
is, nearly all of the HOMO orbital density is on the central
core metal complex. Moving to the first generation carbazole-
dendronised dendrimer 9 the amount of HOMO density on the
iridium is found to decrease to 25% with 34% on the
Table 1 Mulliken populations of the frontier molecular orbitals of the first generation biphenyl based dendrimer and the three generations of
carbazolyl dendronised dendrimers. Ir ¼ iridium(III), ppy ¼ 2-phenylpyridyl ligand, Layer 1 ¼ first level of carbazolyl or phenyl branching units, Layer
2 ¼ second level of carbazolyl branching units, and Layer 3 ¼ final level of carbazolyl units
9
10
11
12
Orbital
HOMO (%)
LUMO (%)
HOMO (%)
LUMO (%)
HOMO (%)
LUMO (%)
HOMO (%)
LUMO (%)
Ir
ppy
Layer 1
Layer 2
Layer 3
24.8
34.3
40.4
1.5
97.8
0.3
9.7
21.5
41.5
27.3
0.5
99.1
0.4
2.1
6.7
22.3
37.8
31.1
0.5
99.1
0.4
0.01
0.0
52.6
41.2
6.2
2.0
97.4
0.3
0.0
2124 | J. Mater. Chem., 2008, 18, 2121–2130
This journal is ª The Royal Society of Chemistry 2008

View Article Online
Fig. 2 ‘Targets’ showing the divisions of the Mulliken populations of the frontier molecular orbitals for the first generation biphenyl 12 (R ¼ 2-ethyl-
hexyl) and carbazolyl 9 dendrimers used in the generation of Table 1. The percentages shown in Table 1 arise from the orbital populations in the area
between a given circle and the next inward circle. The fluorenyl surface groups for the carbazolyl dendrimers have been omitted to enable a simpler
calculation.
excitations are not strictly metal–ligand charge transfer transi-
tions as the ‘‘HOMO’’ has iridium metal, ligand, and carbazole
dendron character. However, care needs to be taken in not
over-interpreting these energies with regard to the excitation
and light emission properties of the dendrimers as they are based
on their ground state geometries.
to-ligand charge transfer states. In comparison to 12 the carba-
zole containing dendrimers have strong p–p* absorptions at
longer wavelengths and this is due to the fact that the carbazolyl
units have a smaller HOMO–LUMO energy gap than phenyl
rings. As expected the absorption at 320 nm, which is due
primarily to the carbazolyl units, increased with generation,
that is, with number of carbazole units. Due to the large oscil-
lator strength and number of chromophores with p–p* absorp-
tions the absorptions involving the iridium(III) are swamped for
the carbazolyl containing dendrimers (the extinction coefficients
are summarised in the Experimental section). It should be noted
that the similarity in the lowest excitation energy determined
from the TD-DFT calculations is reflected in the solution
absorption spectra which all have shoulders at around 480 nm.
The next stage in understanding the photophysical properties
of the dendrimers was the determination of the effect of the car-
bazolyl dendrons on the photoluminescence properties of the
dendrimers. The solution photoluminescence (PL) spectra of
the dendrimers are shown in Fig. 4. It can be seen in Fig. 4 there
are only slight differences in the PL maxima with 12 having
a maximum at 514 nm26 with 9–11 having maxima in the range
510–524 nm, that is, the dendrimers emit green light. The fact
that there is not a red-shift in the emission spectrum in moving
from the first to the second and third generation dendrimers
seems at first to contradict the calculations of the energies of
the lowest transitions. However, it should be noted that to
simplify the calculations we set the angles between the carbazolyl
units at 30ꢀ, and if the angles were larger then the delocalisation
would be reduced with the emission occurring at shorter wave-
lengths. Alternatively the PL spectra provides some evidence
that calculation of the excitation energies based on a ground
state geometry may not be a good method for understanding
the emissive properties of these materials. That is, the excited
state geometry may be significantly different from the ground
state particularly if the symmetry is broken. The view that the
ground and excited state geometries are different is strengthened
Photophysical properties
The solution UV-visible spectra of the iridium(III) complex-cored
dendrimers of 9, 10, 11, and 12 are shown in Fig. 3. The UV-
visible spectra of such dendrimers are comprised of two
components.24,26 The short wavelength component is due to the
p–p* absorptions of the dendrons and ligand while the longer
wavelength weak absorptions are due to the nominally metal-
Fig. 3 Solution UV-visible spectra of 9, 10, 11, and 12. 9, 10, and
11 were measured in dichloromethane while 12 was a solution in
toluene.
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 2121–2130 | 2125

View Article Online
Fig. 5 Cyclic voltammograms of 9, 10, 11, and 5. The voltammograms
have been offset for clarity. All the oxidation potentials are quoted
against the ferricenium/ferrocene couple.
Fig. 4 Solution photoluminescence spectra of 9, 10, 11, and 12.
by time resolved PL measurements, which showed that the
radiative decay rates of the carbazole dendronised dendrimers
were around 5.6 ꢂ 105 sꢁ1 32
.
If the emissive state had more
generation dendrimer appears to have only one oxidation
process with an Epa of 0.65 V and an Epc of 0.45 V.
dendron character then this would correspond to greater
p-orbital triplet character and hence slower radiative rate than
the simple complex should be observed. This is not the case as
The first point to note from the electrochemistry is that there
appears to be a trend for the first oxidation to move to more
positive potentials: 0.29 V for 9 and 0.36 V for 10. Close exam-
ination of the onset to the oxidation process(es) for 11 shows
that it is less steep for 11 than 9 and 10 hinting that there is
another process occurring that is not resolved under the experi-
mental conditions. DPV was only partially successful in
resolving the first oxidation from the latter processes but
a shoulder on the main oxidation peak could be seen clearly.
From the DPV it was estimated that the first oxidation of 11
was at z0.42 V, a further shift to more positive potentials. An
explanation for the shift to more positive potentials for the first
oxidative process with increasing generation is that while for 12
the oxidation is predominantly an Irppy3 complex based process
for 9–11 the first oxidation has carbazolyl character. The molec-
ular orbital calculations show that the carbazolyl character
increases from the first to the third generation dendrimer and so
the oxidation potential which is partly comprised of the core
complex should move closer to those of the carbazolyl units.
The oxidation cyclic voltammogram of the second generation
benzyl protected dendron 5 is also shown in Fig. 5. It can be
seen that its oxidations occur at similar potentials to that of the
second oxidations of 9 and 10, and the oxidation of 11. Given
that the only redox active moieties in 5 are the carbazole units
then the latter oxidations in the dendrimers could be significantly
dendron based. This view is strengthened by the electrochemical
results for the second and third generation dendrimers where there
are a series of overlapping oxidations rather than just a single
oxidative process. Evidence for this comes from the fact that the
difference is between the Epa and Epc is larger for the higher
generation dendrimers (150–200 mV) than for the first generation
dendrimer (z95 mV) even though the scan rates are the same.
This makes sense as the higher generation dendron 5 and
dendrimers 10 and 11 all have carbazole units that are in slightly
different environments and hence different oxidation potentials.
Irppy3 has a very similar radiative decay rate of 6.7 ꢂ 105 sꢁ1 33
.
The computation of the excited state geometries is a challenging
task with the current methodologies. The final step in the photo-
physical study was the measurement of the solution PL quantum
yields (PLQY) to determine whether the addition of the carba-
zolyl dendrons was detrimental to the luminosity of the
materials. 12 has been reported to have a PLQY of 70%33 and
9, 10, and 11 have PLQYs of 75%, 68%, and 66% respectively.
Clearly these results show that the carbazolyl dendrons do not
appreciably quench the luminescence of the core.
Electrochemistry
The oxidation properties of the dendrimers were studied by
cyclic voltammetry at a scan rate of 50 mV sꢁ1. The Irppy3
complex that is at the core of the dendrimers undergoes one
oxidative process with an E1/2 of 0.26 V being reported.34
Dendrimer 12 with biphenyl dendrons attached to the Irppy3
core also has a single oxidative process with the E1/2 being
0.24 V.26 This indicates that the oxidation of dendrimer 12 occurs
essentially on the iridium(III) complex at the core and is consis-
tent with the calculations of the molecular orbital calculations
that shows the HOMO of 12 is essentially on the complex. In
contrast the first generation carbazolyl dendrimer 9 has three
clearly defined chemically reversible oxidations at 0.29 V, 0.57,
and 0.73 V, and an ill-defined process at more positive potentials
(Fig. 5). The fourth oxidation can be resolved using differential
pulse voltammetry (DPV). On moving to the second generation
dendrimer 10 the first oxidation process is seen to move to
a more positive potential with an E1/2 of 0.36 V with the second
oxidation having an apparent E1/2 of 0.56 V. There is also a third
oxidation process that occurs at a similar position to the first
generation carbazole dendrimer 9. At first glance the third
2126 | J. Mater. Chem., 2008, 18, 2121–2130
This journal is ª The Royal Society of Chemistry 2008

View Article Online
platinum working, Ag/AgCl/NaCl reference, and platinum
counter electrodes, and a 50 mV sꢁ1 scan rate. All measurements
were made at room temperature on samples at a 1 mM concen-
tration in dichloromethane, with 0.1 M tetra-n-butylammonium
tetrafluoroborate as the electrolyte. HPLC grade dichlorome-
thane was used for the oxidation studies. The solutions were
deoxygenated with argon and the ferrocenium/ferrocene couple
was used as standard.35 In all cases several scans were carried
out to confirm the chemical reversibility of the redox processes.
Solution PLQYs were measured by a relative method using
quinine sulfate in 0.5 M sulfuric acid, which has a PLQY of
54.6%, as the standard.36 The dendrimers were dissolved in tetra-
hydrofuran and freeze–thaw degassed. Photoluminescence
spectra were recorded in a JY Horiba Fluoromax 2 fluorimeter,
with the dendrimer solutions excited at 360 nm. The optical
densities of the standard and sample were similar and small
(less than/equal to 0.1). The accuracy of these measurements is
estimated to be ꢃ10% of the stated value. The time-resolved
luminescence measurements were performed using the time-
correlated single photon counting technique, with excitation at
390 nm from a pulsed gallium nitride laser. The hole mobility
was measured using the charge generation layer time-of-flight
method. Dendrimer films 200–300 nm thick were prepared by
spin-coating, and a 10 nm lumogen red charge injection layer
evaporated on top of them. The sample was excited by a nitrogen
pumped dye laser at 580 nm.
Conclusion
We report a new simple strategy for preparing carbazolyl-based
dendrons of different generations. The key to the strategy is that
there is a simple iterative two-step procedure, namely a palladium
catalysed amination followed by a benzyl deprotection. The
dendrimers containing
a fac-tris[2-phenylpyridyl]iridium(III)
complex core were also prepared by a new method involving
the attachment of the dendrons to a preformed tris-brominated
complex again using palladium catalysis. This meant that the
third generation dendrimer could be prepared in six simple steps.
While it was found that the attachment of the carbazolyl
dendrons did not affect the photoluminescence properties of
the dendrimers with respect to those with the same core but
biphenyl based dendrons, molecular orbital calculations on the
carbazolyl containing dendrimers indicated that the ‘‘HOMO’’
was not just located on the core but also over the carbazolyl units
in the dendrons. The delocalized HOMO has the effect of
increasing the hole-transport properties of the dendrimers and
this will be reported in detail elsewhere.
Experimental
Measurements
NMR spectra were recorded on Bruker DPX400, DQX400, or
AVII500 spectrometers: CarbH ¼ carbazole H, LigH ¼ ligand
H, and FlH ¼ fluorenyl H. All J values are in Hertz and are
rounded to the nearest 0.5 Hz. UV-visible spectra were recorded
on a Perkin-Elmer UV-visible Lambda 25 spectrometer and were
recorded as solutions in spectroscopic grade dichloromethane.
Infrared spectra were recorded on a Perkin Elmer Spectrum
Density functional theory (DFT) calculations were performed
in C3 symmetry using the Gaussian 03 suite of programs37 with
a LanL2DZ basis set38–41 and B3LYP level of theory42 Geome-
tries were optimised structures with the pendant phenyl and
carbazole units held at 30ꢀ to the ligand phenyl or ring bearing
them. Compositions of molecular orbitals were calculated using
the AOMix Program.43,44 Linear response TD-DFT calculations
of the 6 lowest states of singlet and triplet character were
performed within the random phase approximation using the
same functional and basis set. The Cartesian co-ordinates of
1000 FT-IR or a Bruker Tensor 27 FT-IR spectrometer. Mass
¨
spectra were recorded on an Applied Biosystems Voyager
matrix-assisted laser desorption/ionisation time-of-flight
(MALDI-TOF) from 2-[(2E)-3-(4-tert-butylphenyl)-2-methyl-
prop-2-enylidene]malononitrile (DCTB) in positive reflectron
mode, or positive linear mode for 11, at the EPSRC National
Mass Spectrometry Centre, Swansea, UK, or a VG AutoSpec
for CI. Melting points were recorded on a Gallenkamp or a Leica
Galen III melting point apparatus and are uncorrected.
Microanalyses were carried out in the Inorganic Chemistry
Laboratory, Oxford or at London Metropolitan University,
UK. Gel permeation chromatography (GPC) was carried out
using PLgel Mixed-A columns (600 mm + 300 mm lengths,
7.5 mm diameter) from Polymer Laboratories calibrated with
˚
the models (in A) are provided in the ESI.† Although the theo-
retical foundations which underlie the use of Kohn–Sham
orbitals in chemical analyses are less rigorous than those for,
e.g., Hartree–Fock canonical orbitals, they have been found
empirically to give useful orbital energies45 and singlet–triplet
excitation splittings.46
3,6-(Bis[9,90-di-n-propylfluoren-2-yl])carbazole 3
6
3,6-Dibromocarbazole 2 (9.03 g, 27.8 mmol) and 9,90-di-n-pro-
pylfluorenyl-2-boronic acid 1 (19.9 g, 67.8 mmol) were dissolved
in toluene (75 cm3) and ethanol (59 cm3). Aqueous sodium
carbonate (2 M, 75 cm3) was added, and the mixture was
deoxygenated with argon for 5 min. The mixture was heated to
100 ꢀC and tetrakis(triphenylphosphino)palladium(0) (1.77 ꢀg,
1.53 g) was added and then the reaction was heated at 100 C
for a further 23 h. The mixture was allowed to cool to room
temperature and water (200 cm3) was added. The organic layer
was separated and the aqueous layer was extracted with ether
(2 ꢂ 200 cm3). The combined organic layers were washed with
brine (200 cm3), dried over anhydrous magnesium sulfate,
filtered, and the solvent was removed. The residue was purified
in three stages: first, it was passed through a silica plug using
ꢀ
polystyrene narrow standards (Mp ¼ 580 to 3.2 ꢂ 10 ) in tetra-
hydrofuran with toluene as flow marker. The tetrahydrofuran
was degassed with helium and pumped with a rate of 1 cm3
minꢁ1 at 30.0 ꢀC. Thermal gravimetric analysis was performed
on a Perkin Elmer Thermogravimetric Analyzer TGA 7. Light
petroleum refers to the fraction of boiling point 40–60 ꢀC.
When solvent mixtures are used for chromatography over silica
the proportions are given by volume. All solvents used were of
HPLC or Analar grade. Tetrahydrofuran was dried by distilla-
tion from sodium benzophenone ketal or by passage through
activated alumina. Xylene and toluene were dried by distillation
from calcium hydride.
Electrochemistry was performed using an EG&G Princeton
Applied Research potentiostat/galvanostat model 263A using
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 2121–2130 | 2127

View Article Online
a dichloromethane : light petroleum mixture (1 : 1) as eluent;
second, it was purified by column chromatography over silica
using a dichloromethane : light petroleum mixture (1 : 4 to 3 : 7)
as eluent; and finally, it was recrystallised from a dichlorome-
thane : methanol mixture to give 3 (14.05 g, 76%): mp 229–230
ꢀC (Found: C, 90.3%; H, 7.5%; N, 2.1%. C50H49N requires C,
90.45%; H, 7.4%; N, 2.1%); nmax(KBr) 3396 cmꢁ1 (NH);
lmax(CH2Cl2) 269 nm (log 3/dm3 molꢁ1 cmꢁ1 4.49), 319 (4.89),
399sh (4.77); dH(400 MHz, CDCl3) 0.70–0.85 (20 H, m,
CH2CH3 and CH3), 2.00–2.10 (8 H, m, CH2CH2CH3),
7.30–7.42 (6 H, m, CarbH and/or FlH), 7.57 (2 H, d, J ¼ 8,
CarbH), 7.70–7.85 (10 H, m, CarbH and/or FlH), 8.17 (1 H,
bs, NH), 8.48 (2 H, s, CarbH); m/z (CI, NH3) 664 [MH]+ (100%).
chromatography over silica using a dichloromethane : light
petroleum mixture (1 : 3 to 2 : 5) as eluent to give 6 as a white
powder (2.06 g, 51%): mp 238–240 ꢀC (Found: C, 89.8%; H,
7.2%; N, 2.7%. C112H103N3 requires C, 90.2%; H, 7.0%; N,
2.8%); nmax(KBr) 3408 cmꢁ1 (NH); lmax(CH2Cl2) 272 nm (log
3/dm3molꢁ1cmꢁ1 4.85), 302sh (5.09), 321 (5.22), 342sh (5.08);
dH(400 MHz, CDCl3) 0.67–0.83 (40 H, m, CH2CH3 and
CH2CH3), 2.00–2.15 (16 H, m, CH2CH2CH3), 7.31–7.45 (12 H,
m, CarbH and/or FlH), 7.56 (4 H, d, J ¼ 8, CarbH), 7.70–7.85
(24 H, m, CarbH and/or FlH), 8.36 (2 H, s, CarbH), 8.53
(1 H, s, NH), 8.58 (4 H, s, CarbH); m/z [MALDI-TOF] Anal.
Calcd for C112H103N3: 1489.8 (79%), 1490.8 (100%), 1491.8
(64%), 1492.8 (26%), 1493.8 (7%), 1494.8 (2%). Found: 1489.8
(86%), 1490.8 (100%), 1491.8 (55%), 1492.9 (33%), 1493.9
(9%), 1494.9 (2%) (M+).
N-Benzyl-3,6-bis[3,6-bis(9,90-di-n-propylfluoren-
2-yl)carbazolyl]carbazole 5
N-Benzyl-3,6-bis[3,6-bis(3,6-bis{9,90-di-n-propylfluoren-
2-yl}carbazolyl)carbazolyl]carbazole 7
N-Benzyl-3,6-dibromocarbazole 4 (1.97 g, 4.75 mmol), 3 (7.07 g,
10.6 mmol), sodium tert-butoxide (1.92 g, 20.0 mmol) and
tri-tert-butylphosphonium tetrafluoroborate (506 mg, 1.74
mmol) were placed into a pressure tube fitted with a Young’s
valve and deoxygenated by placing under vacuum and back
filling with nitrogen. Mixed xylenes (30 cm3) were added and
the resultant solution deoxygenated three times. Tris(dibenzyli-
deneacetone)dipalladium(0) (200 mg, 218 mmol) was added, the
solution was deoxygenated three times and the tube was sealed
and heated at 135 ꢀC for 17 h. The mixture was cooled and
filtered through a silica plug using dichloromethane as eluent.
The filtrate was collected and the solvent removed. The crude
product was purified by flash chromatography over silica using
a dichloromethane : light petroleum mixture (1 : 4 to 7 : 13) to
give a mixture of 5 and 3. The mixture was adsorbed onto silica
and 3 was removed by elution with a 2-propanol : light petrol
mixture (1 : 9). After removal of 3, elution with dichloromethane
gave 5 as a white solid (6.00 g, 80%): mp 263–265 ꢀC (Found: C,
90.4%; H, 6.9%; N, 2.6%. C119H109N3 requires C, 90.4%; H,
6.95%; N, 2.7%); lmax(CH2Cl2) 272 nm (log 3/dm3 molꢁ1 cmꢁ1
4.87), 302sh (5.09), 321 (5.22), 342sh (5.06); dH(400 MHz,
CD2Cl2) 0.70–0.85 (40 H, m, CH2CH3 and CH3), 2.02–2.18
(16 H, m, CH2CH2CH3), 5.84 (2 H, s, CH2Ph), 7.33–7.50
(17 H, m, CarbH and/or FlH, and PhH), 7.60 (4 H, d, J ¼ 8.5,
CarbH), 7.76–7.88 (24 H, m, CarbH and/or FlH), 8.46 (2 H, s,
CarbH), 8.52 (4 H, s, CarbH); m/z (MALDI-TOF) Anal. Calcd
for C183H165IrN6: 2637.3 (21%), 2638.3 (41%), 2639.3 (78%),
2640.3 (100%), 2641.3 (89%), 2642.3 (55%), 2643.3 (27%),
2644.3 (12%), 2645.3 (6%), 2646.3 (2%). Found: 2637.3 (32%),
2638.3 (56%), 2639.3 (89%), 2640.3 (100%), 2641.3 (89%),
2642.3 (64%), 2643.3 (30%), 2644.3 (11%), 2645.3 (5%), 2646.3
(2%) (M+).
4 (415 mg, 1.00 mmol), 6 (3.50 g, 2.35 mmol), sodium tert-butox-
ide (497 mg, 5.18 mmol) and tri-tert-butylphosphonium tetra-
fluoroborate (300 mg, 1.03 mmol) were placed into a pressure
tube fitted with a Young’s valve and deoxygenated by placing
under vacuum and then back filling with nitrogen. Mixed xylenes
(25 cm3) were added and the resultant solution deoxygenated
five times. Tris(dibenzylideneacetone)dipalladium(0) (120 mg,
130 mmol) was added, the solution was deoxygenated three times
and the tube was sealed and heated at 130 ꢀC for 14 h. The
mixture was cooled and filtered through a silica plug using
dichloromethane as eluent. The filtrate was collected and the
solvent removed. The residue was purified by flash chromatog-
raphy over silica using a dichloromethane : light petroleum
mixture (3 : 7 to 2 : 5) as eluent to give 7 as a white solid
(2.35 g, 73%): mp >280 ꢀC (Found: C, 90.0%; H, 6.9%; N,
2.85%. C243H217N7 requires C, 90.2%; H, 6.8%; N, 3.0%);
lmax(CH2Cl2) 272 nm (log 3/dm3 molꢁ1 cmꢁ1 5.19), 304sh
(5.41), 320 (5.52), 343sh (5.35); dH(500 MHz, CDCl3) 0.69–0.88
(80 H, m, CH2CH3 and CH3), 2.00–2.16 (32 H, m,
CH2CH2CH3), 5.90 (2 H, s, CH2Ph), 7.31–7.43 (24 H, m, CarbH
and/or FlH), 7.46–7.56 (5 H, m, PhH), 7.60 (8 H, d, J ¼ 8.5,
CarbH), 7.73–7.87 (48 H, m, CarbH and/or FlH), 7.89–8.01
(4 H, m, CarbH and/or FlH), 8.49 (4 H, s, CarbH), 8.59 (8 H,
s, CarbH), 8.69 (2 H, s, CarbH); m/z [MALDI-TOF] Anal. Calcd
for C243H217N7: 3232.7 (27%), 3233.7 (73%), 3234.7 (100%),
3235.7 (91%), 3236.7 (60%), 3237.7 (34%), 3238.7 (16%),
3239.7 (5%), 3240.7 (3%). Found: 3232.7 (43%), 3233.7 (90%),
3234.7 (98%), 3235.7 (100%), 3236.7 (68%), 3237.7 (33%),
3238.7 (17%), 3239.7 (5%), 3240.7 (3%) (M+).
3,6-Bis[3,6-bis(3,6-bis{9,90-di-n-propylfluoren-
2-yl}carbazolyl)carbazolyl]carbazole 8
3,6-Bis[3,6-bis(9,90-di-n-propylfluoren-2-yl)carbazolyl]
carbazole 6
7 (1.01 g, 312 mmol) and aluminium chloride (5.00 g, 38 mmol)
were dissolved in anisole and heated at 90 ꢀC for 2 h. After
cooling, the reaction was quenched with water (50 cm3). Ether
(50 cm3) and dilute hydrochloric acid (3 M, 50 cm3) were added
and the layers were separated. The organic layer was washed
with water (50 cm3) and brine (50 cm3), dried over magnesium
sulfate and filtered before the solvent was removed. The crude
5 (4.26 g, 2.69 mmol) was dissolved in anisole (250 cm3).
Aluminium chloride (17.00 g, 128 mmol) was added, and the
mixture heated at 90 ꢀC overnight. After cooling, methanol
(20 cm3) was added and the mixture was passed though a plug
of silica using dichloromethane as eluent. The solvent was
removed and the residue was purified by column
2128 | J. Mater. Chem., 2008, 18, 2121–2130
This journal is ª The Royal Society of Chemistry 2008

View Article Online
material was purified by column chromatography over silica
using a dichloromethane : light petroleum mixture (3 : 7 to
1 : 1) as eluent to give 8 as a white solid (0.61 g, 62%): mp
>280 ꢀC (Found: C, 89.8%; H, 6.5%; N, 2.95%. C236H210N7
requires C, 90.1%; H, 6.8%; N, 3.1%); nmax(KBr) 3413 cmꢁ1
(NH); lmax(CH2Cl2) 272 nm (log 3/dm3 molꢁ1 cmꢁ1 5.17),
304sh (5.40), 320 (5.51), 342sh (5.36); dH(500 MHz, CDCl3)
0.68–0.87 (80 H, m, CH2CH3 and CH3), 2.00–2.14 (32 H, m,
CH2CH2CH3), 7.33–7.55 (24 H, m, CarbH and/or FlH), 7.60
(8 H, d, J ¼ 8.5, CarbH), 7.72–7.85 (48 H, m, CarbH and/or
FlH), 7.95 (4 H, m, CarbH and FlH), 8.47 (4 H, s, CarbH),
8.57 (8 H, s, CarbH), 8.61 (2 H, s, CarbH), 8.71 (1 H, s, NH);
m/z [MALDI-TOF] Anal. Calcd for C236H211N7: 3142.7 (27%),
3143.7 (75%), 3144.7 (100%), 3145.7 (88%), 3146.7 (57%),
3147.7 (28%), 3148.7 (14%), 3149.7 (5%), 3150.7 (2%). Found:
3142.7 (34%), 3143.7 (84%), 3144.7 (100%), 3145.7 (86%),
3146.7 (53%), 3147.7 (26%), 3148.7 (13%), 3149.7 (5%), 3150.7
(2%) (M+).
fac-Tris[2-(3-{3,6-bis[3,6-bis(9,90-di-n-propylfluoren-2-
yl)carbazolyl]carbazolyl}phenyl)pyridyl]iridium(III) 10
A
mixture of fac-tris[2-(3-bromophenyl)pyridyl]iridium(III)
(150 mg, 168 mmol), 6 (0.95 g, 0.64 mmol), sodium tert-butoxide
(140 mg, 1.46 mmol) and tri-tert-butylphosphonium tetrafluoro-
borate (176 mg, 0.61 mmol) was placed into a pressure tube fitted
with a Young’s valve and degassed. Mixed xylenes (10 cm3) were
added and the resultant solution was deoxygenated by placing
under vacuum and backfilling with nitrogen three times. Tris
(dibenzylideneacetone)dipalladium(0) (70 mg, 76 mmol) was
added, the solution was deoxygenated by placing under vacuum
and backfilling with nitrogen a further three times, and then
heated at reflux for 3 days. The mixture was cooled, filtered
through a plug of silica using dichloromethane as eluent, and
then the filtrate was collected and the solution removed. The
residue was purified by column chromatography over silica in
two steps: first using a dichloromethane : light petroleum mixture
(2 : 3) as eluent and then by using a benzene : light petroleum
mixture (3 : 2) as eluent to give 10 as a yellow powder
(274 mg, 32%): mp >280 ꢀC (Found: C, 86.4%; H, 6.5%; N,
3.2%. C369H327IrN12 requires C, 86.5%; H, 6.4%; N, 3.3%);
lmax(CH2Cl2) 273 nm (log 3/dm3 molꢁ1 cmꢁ1 5.43), 306sh
(5.63), 320 (5.71), 400sh (4.10), 440sh, (3.66) 480sh (3.22);
dH(400 MHz, C6D6) 0.74 (72 H, t, J ¼ 7.5, CH3), 0.90–1.10 (48
H, m, CH2CH3), 2.00–2.17 (48 H, m, CH2CH2CH3), 6.44 (3
H, dd, J ¼ 8, J ¼ 8, LigH), 7.04 (3 H, dd, J ¼ 8.5, J ¼ 8.5,
LigH), 7.31–8.32 (141 H, m, LigH, CarbH, FlH), 8.82 (12 H, s,
CarbH); m/z [MALDI-TOF] Anal. Calcd for C369H327IrN12:
5116.6 (4%), 5117.6 (13%), 5118.6 (33%), 5119.6 (62%), 5120.6
(87%), 5121.6 (100%), 5122.6 (88%), 5123.6 (71%), 5124.6
(43%), 5125.6 (28%), 5126.6 (12%), 5127.6 (6%), 5128.6 (3%).
Found: 5116.5 (5%), 5117.5 (17%), 5118.5 (44%), 5119.5 (69%),
5120.5 (93%), 5121.5 (100%), 5122.5 (90%), 5123.5 (66%),
fac-Tris[2-(3-{3,6-bis(9,90-di-n-propylfluoren-2-yl)carbazoyl}
phenyl)pyridyl]iridium(III) 9
A
mixture of fac-tris[2-(3-bromophenyl)pyridyl]iridium(III)47
(383 mg, 439 mmol), 3 (1.00 g, 1.51 mmol), sodium tert-butoxide
(192 mg, 2.00 mmol) and tri-tert-butylphosphonium tetrafluoro-
borate (100 mg, 345 mmol) was placed into a pressure tube fitted
with a Young’s valve and deoxygenated by placing under
vacuum and backfilling with nitrogen. Toluene (10 cm3) was
added and the resultant solution deoxygenated by placing under
vacuum and filling with nitrogen three times. Tris(dibenzylide-
neacetone)dipalladium(0) (40 mg, 44 mmol) was added, and the
solution deoxygenated by placing under vacuum and filling
with nitrogen three times. The tube was sealed and the reaction
ꢀ
mixture was heated at 110 C for 17 h. The mixture was cooled
5124.5 (43%), 5125.5 (23%), 5126.5 (8%), 5127.5 (5%), 5128.5
+
and then filtered through a plug of silica using dichloromethane
as eluent. To this solution was added silica and then the solvent
was removed. The preloaded silica was added onto a silica
chromatography column packed with light petroleum and a light
petroleum : 2-propanol solvent mixture (4 : 1) was used to elute
the majority of the free carbazole and then dichloromethane was
used to elute crude 9. Crude 9 was further purified by column
chromatography over silica using a dichloromethane : light
petroleum mixture (3 : 7 to 7 : 13) as eluent to give 9 as a yellow
powder (759 mg, 65%): mp >280 ꢀC (Found: C, 83.2%; H, 6.3%;
N, 3.2%. C183H165IrN6 requires C, 83.2%; H, 6.3%; N, 3.2%);
lmax(CH2Cl2) 271 nm (log 3/dm3molꢁ1cmꢁ1 5.10), 307sh (5.31),
320 (5.39), 336sh (5.29), 401sh (4.29), 438sh (3.88), 480sh
(3.43); dH(400 MHz, C6D6) 0.77 (36 H, t, J ¼ 7, CH3),
0.89–1.11 (24 H, m, CH2CH3), 2.00–2.19 (24 H, m,
CH2CH2CH3), 6.36 (3 H, dd, J ¼ 6.5, J ¼ 6.5, LigH), 6.94
(3 H, dd, J ¼ 8, J ¼ 8, LigH), 7.11–8.11 (69 H, m, LigH, CarbH,
FlH), 8.79 (2 H, s, CarbH); m/z [MALDI-TOF] Anal. Calcd
for C183H165IrN6: 2637.3 (21%), 2638.3 (41%), 2639.3 (78%),
2640.3 (100%), 2641.3 (89%), 2642.3 (55%), 2643.3 (27%),
2644.3 (12%), 2645.3 (6%), 2646.3 (2%). Found: 2637.3 (32%),
2638.3 (56%), 2639.3 (89%), 2640.3 (100%), 2641.3 (89%),
ꢀ
ꢀ
(3%) (M ). GPC: Mw ¼ 3995, Mv ¼ 3921, PD ¼ 1.15
fac-Tris[2-(3-{3,6-bis[3,6-bis(3,6-bis{9,90-di-n-propylfluoren-2-
yl}carbazolyl)carbazolyl]carbazolyl}phenyl)pyridyl]iridium(III) 11
A
mixture of fac-tris[2-(3-bromophenyl)pyridyl]iridium(III)
(36 mg, 40 mmol), 8 (578 mg, 184 mmol), sodium tert-butoxide
(75 mg, 0.78 mmol) and tri-tert-butylphosphonium tetrafluoro-
borate (64 mg, 0.22 mmol) was placed into a pressure tube fitted
with a Young’s valve and deoxygenated by placing under
vacuum and backfilling with nitrogen. Toluene (10 cm3) was
added and the resultant solution deoxygenated by placing under
vacuum and backfilling with nitrogen five times. Tris(dibenzyli-
deneacetone)dipalladium(0) (29 mg, 30 mmol) was added, the
solution was deoxygenated by placing under vacuum and back-
filling with nitrogen five times before being heated at 110 ꢀC for
40 h. The mixture was cooled, filtered through a plug of silica
using dichloromethane as eluent, and the solvent removed. The
residue was purified by column chromatography over silica in
two steps: first, using a dichloromethane : light petroleum
mixture (2 : 3 to 1 : 1) as eluent, and second, using a benzene :
light petroleum mixture (1 : 1 to 13 : 7) as eluent to give 11 as
a yellow powder (136 mg, 34%): mp >280 ꢀC (Found: C,
88.4%; H, 6.5%; N, 3.3%. C741H651IrN24 requires C, 88.25%;
2642.3 (64%), 2643.3 (30%), 2644.3 (11%), 2645.3 (5%), 2646.3
+
ꢀ
ꢀ
(2%) (M ); GPC: Mw ¼ 2334, Mv ¼ 2294, Polydispersity
(PD) ¼ 1.1.
This journal is ª The Royal Society of Chemistry 2008
J. Mater. Chem., 2008, 18, 2121–2130 | 2129

View Article Online
H, 6.5%; N, 3.3%); lmax(CH2Cl2) 273 nm (log 3/dm3 molꢁ1 cmꢁ1
5.67), 310sh (5.88), 319 (5.93), 400sh (3.85), 443sh (3.39), 480sh
(2.98); dH(400 MHz, CD2Cl2) 0.52–0.77 (240 H, bm, CH3 and
CH2CH3), 1.87–2.07 (96, m, CH2CH2CH3), and 7.21–7.36,
7.52–7.57, 7.60–7.79, 8.27–8.37, 8.42–8.57, 8.71 (315 H, LigH,
CarbH, FlH); m/z [MALDI: DCTB] Anal. Calcd for
21 Y. Q. Li, A. Rizzo, M. Salerno, M. Mazzeo, C. Huo, Y. Wang,
K. C. Li, R. Cingolani and G. Gigli, Appl. Phys. Lett., 2006, 89,
061125.
22 T.-H. Kwon, M.-K. Kim, J. Kwon, D.-Y. Shin, S.-J. Park, C.-L. Lee,
J.-J. Kim and J.-I. Hong, Chem. Mater., 2007, 19, 3673.
23 J. Q. Ding, J. Gao, Y. X. Cheng, Z. Y. Xie, L. X. Wang, D. G. Ma,
X. B. Jing and F. S. Wang, Adv. Funct. Mater., 2006, 16, 575.
24 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.
+
ꢀ
C
741H651IrN24: 10077. Found: 10090 (broad) (M ). GPC: Mw
ꢀ
¼ 6175, Mv ¼ 6065, PD ¼1.1.
25 N. Iguchi, Y.-J. Pu, K. Nakayama and J. Kido, J. Photopolym. Sci.
Technol., 2007, 20, 73.
Acknowledgements
26 S.-C. Lo, E. B. Namdas, P. L. Burn and I. D. W. Samuel,
Macromolecules, 2003, 36, 9721.
We thank the EPSRC for financial support. We also thank the
EPSRC National Service for Computational Chemistry
Software (URL: http://www.nsccs.ac.uk) and the APAC
National Facility (Canberra, Australia) for computer time, and
the EPSRC National Spectroscopy Centre, Swansea, UK.
Professor Paul L. Burn is the recipient of an Australian Research
Council Federation Fellowship (project number FF0668728).
27 K. R. J. Thomas, J. T. Lin, Y.-T. Tao and C.-W. Ko, J. Am. Chem.
Soc., 2001, 123, 9404.
28 I. D. W. Samuel, P. L. Burn and S.-C. Lo, Pat. Appl. GB0222268.5,
2002.
29 M. E. Wright and M.-J. Jin, J. Org. Chem., 1989, 54, 965.
30 P. L. Burn, R. Beavington, M. J. Frampton, J. N. G. Pillow,
M. Halim, J. M. Lupton and I. D. W. Samuel, Mater. Sci. Eng., B,
2001, 85, 190.
31 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.
32 J. C. Ribierre, A. Ruseckas, K. Knights, S. V. Staton, N. Cumpstey,
P. L. Burn and I. D. W. Samuel, Phys. Rev. Lett., 2008, 100, 017402.
33 E. B. Namdas, A. Ruseckas, I. D. W. Samuel, S.-C. Lo and
P. L. Burn, J. Phys. Chem. B, 2004, 108, 1570.
References
1 Y. You, C.-G. An, J.-J. Kim and S.-Y. Park, J. Org. Chem., 2007, 72,
6241.
34 V. V. Grushin, N. Herron, D. D. LeCloux, W. J. Marshall and
V. A. Petrov, Chem. Commun., 2001, 1494.
2 Z. L. Lu, K. Q. Xing, C. P. Luo, Y. Liu, Y. P. Yang, Q. Gan,
M. X. Zhu, C. Y. Jiang, Y. Cao and W. G. Zhu, Chem. Lett., 2006,
35, 538.
35 G. Gritzner and J. Kuta, Electrochim. Acta, 1984, 29, 869.
36 J. N. Demas and G. A. J. Crosby, J. Phys. Chem., 1971, 75, 991.
37 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr, T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Lyengar, 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, 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. Akrzewski, 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. Cioslowski, B. B. Stefanov,
G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, 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 and J. A. Pople, Gaussian 03,
revision C.01 or E.01, Gaussian, Inc., Wallingford, CT, 2004.
38 T. H. Dunning Jr and P. J. Hay, in Modern Theoretical Chemistry,
ed. H. F. Schaefer, III, Plenum, New York, 1976, vol. 3, p. 1.
39 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270.
40 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 284.
41 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299.
42 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
3 M. A. Baldo, S. Lamansky, P. E. Burrows, M. E. Thompson and
S. R. Forrest, Appl. Phys. Lett., 1999, 75, 4.
4 M. Ikai, S. Tokito, Y. Sakamato, T. Suzuki and Y. Taga, Appl. Phys.
Lett., 2001, 79, 156.
5 T. Watanabe, E. Nakamura, S. Kawami, Y. Fukuda, T. Tsuji,
T. Wakimoto, S. Miyaguchi, M. Yahiro, M. J. Yang and
T. Tsutsui, Synth. Met., 2001, 122, 203.
6 X. Zhou, D. S. Qin, M. Pfeiffer, J. Blochwitz-Nimoth, A. Werner,
J. Dreschel, B. Maennig, K. Leo, M. Bold, P. Erk and
H. Hartmann, Appl. Phys. Lett., 2002, 81, 4070.
7 X. H. Yang and D. Neher, Appl. Phys. Lett., 2004, 84, 2476.
8 M. Velusamy, K. R. J. Thomas, C.-H. Chen, J. T. Lin, Y. S. Wen,
W.-T. Hsieh, C.-H. Lai and P.-T. Chou, Dalton Trans., 2007, 3025.
9 X. H. Li, Z. Chen, Q. Zhao, L. Shen, F. Y. Li, T. Yi, Y. Cao and
C. H. Huang, Inorg. Chem., 2007, 46, 5518.
10 R. N. Bera, N. Cumpstey, P. L. Burn and I. D. W. Samuel, Adv.
Funct. Mater., 2007, 17, 1149.
11 N. Cumpstey, R. N. Bera, P. L. Burn and I. D. W. Samuel,
Macromolecules, 2005, 38, 9564.
12 P. L. Burn, S.-C. Lo and I. D. W. Samuel, Adv. Mater., 2007, 19,
1675.
13 S.-C. Lo and P. L. Burn, Chem. Rev., 2007, 107, 1097.
14 S.-C. Lo, N. A. H. Male, J. P. J Markham, S. W. Magennis,
P. L. Burn, O. V. Salata and I. D. W. Samuel, Adv. Mater., 2002,
14, 975.
15 T. D. Anthopoulos, M. J. Frampton, E. B. Namdas, P. L. Burn and
I. D. W. Samuel, Adv. Mater., 2004, 16, 557.
43 S. I. Gorlesky, AOMix: Program for Molecular Orbital Analysis,
York University, Toronto, Canada, 1997. Available from: http://
www.sg-chem.net.
44 S. I. Gorlesky and A. B. P Lever, J. Organomet. Chem., 2001, 635,
187.
45 R. Stowasser and R. Hoffmann, J. Am. Chem. Soc., 1999, 121, 3414.
46 A. Vazquez-Mayagoitia, R. Vargas, J. A. Nichols, P. Fuentealba and
J. Garza, Chem. Phys. Lett., 2006, 419, 207.
47 W. Yang, H. Y. Zhen, C. Y. Jiang, L. J. Su, J. X. Jiang, H. H. Shi and
Y. Cao, Synth. Met., 2005, 153, 189.
16 S.-C. Lo, G. J. Richards, J. P. J. Markham, E. B. Namdas, S. Sharma,
P. L. Burn and I. D. W. Samuel, Adv. Funct. Mater., 2005, 15, 1451.
17 E. B. Namdas, T. D. Anthopoulos, I. D. W. Samuel, M. J. Frampton,
S.-C. Lo and P. L. Burn, Appl. Phys. Lett., 2005, 86, 161104.
18 S.-C. Lo, T. D. Anthopoulos, E. B. Namdas, P. L. Burn and
I. D. W Samuel, Adv. Mater., 2005, 17, 1945.
19 J. C. Ribierre, S. G. Stevenson, I. D. W. Samuel, S. V. Staton and
P. L. Burn, J. Display Technol., 2007, 3, 233.
20 B.-L. Li, L. Wu, Y.-M. He and Q.-H. Fan, Dalton Trans., 2007, 2048.
2130 | J. Mater. Chem., 2008, 18, 2121–2130
This journal is ª The Royal Society of Chemistry 2008