High-triplet-energy dendrons: Enhancing the luminescence of deep blue phosphorescent iridium(III) complexes

Article abstract of DOI:10.1021/ja903157e

Solution-processable blue phosphorescent emitters with high luminescence efficiency are highly desirable for large-area displays and lighting applications. This report shows that when a fac-tris[1-methyl-5-(4-fluorophenyl) -3-n-propyl-1H-[1,2,4]triazolyl]

Full text of DOI:10.1021/ja903157e

Published on Web 07/10/2009
High-Triplet-Energy Dendrons: Enhancing the Luminescence
of Deep Blue Phosphorescent Iridium(III) Complexes
Shih-Chun Lo,† Ruth E. Harding,‡ Christopher P. Shipley,§ Stuart G. Stevenson,‡
Paul L. Burn,*,† and Ifor D. W. Samuel*,‡
Centre for Organic Photonics & Electronics, The UniVersity of Queensland, School of Chemistry
& Molecular Biosciences, QLD 4072, Australia, Organic Semiconductor Centre, SUPA, School
of Physics & Astronomy, UniVersity of St Andrews, North Haugh, Fife, KY16 9SS, U.K., and
Chemistry Research Laboratory, Department of Chemistry, UniVersity of Oxford,
Mansfield Road, Oxford, OX1 3TA, U.K.
Received April 20, 2009; E-mail: p.burn2@uq.edu.au; idws@st-andrews.ac.uk
Abstract: Solution-processable blue phosphorescent emitters with high luminescence efficiency are highly
desirable for large-area displays and lighting applications. This report shows that when a fac-tris[1-methyl-
5-(4-fluorophenyl)-3-n-propyl-1H-[1,2,4]triazolyl]iridium(III) complex core is encapsulated by rigid high-triplet-
energy dendrons, both the physical and photophysical properties can be optimized. The high-triplet-energy
and rigid dendrons were composed of twisted biphenyl dendrons with the twisting arising from the use of
tetrasubstituted branching phenyl rings. The blue phosphorescent dendrimer was synthesized using a
convergent approach and was found to be solution-processable and to possess a high glass transition
temperature of 148 °C. The dendrimer had an exceptionally high solution photoluminescence quantum
yield (PLQY) of 94%, which was more than three times that of the simple parent core complex (27%). The
rigid and high-triplet-energy dendrons were also found to control the intermolecular interactions that lead
to the quenching of the luminescence in the solid state, and the film PLQY was found to be 60% with the
emission having Commission Internationale de l’Eclairage coordinates of (0.16, 0.16). The results
demonstrate that dendronization of simple chromophores can enhance their properties. Single layer neat
dendrimer organic light-emitting diodes (OLEDs) had an external quantum efficiency (EQE) of 0.4% at 100
cd/m2. Bilayer devices with an electron transport layer gave improved EQEs of up to 3.9%. Time-resolved
luminescence measurements suggest that quenching of triplets by the electron transport layer used in the
bilayer OLEDs limits performance.
radiatively thus leading to higher efficiencies.2 For example,
Introduction
OLEDs based on the green emissive fac-tris(2-phenylpyridyl)-
Organic light-emitting diodes (OLEDs)1 based on luminescent
heavy metal complexes are playing a key role in next generation
flat panel displays and solid-state lighting. Phosphorescence from
osmium(II), platinum(II), and in particular iridium(III) gives an
advantage over fluorescent materials in that both the singlet and
triplet excitons that form in a device can be captured and decay
iridium(III) (Irppy3) have shown a nearly 100% internal quantum
efficiency.3 While OLEDs with red and green phosphorescent
iridium(III) complexes have proved fairly straightforward to
produce, the same cannot be said for saturated blue emissive
materials. Highly luminescent deep blue phosphorescent materi-
als with Commission Internationale de l’Eclairage (CIE) coor-
dinates less than (0.20, 0.20) have proved a challenge. For full
color displays and solid-state lighting based on OLED technol-
ogy, it would be advantageous to have deep blue phosphorescent
materials with high photoluminescence quantum yields. Early
work on materials that emitted blue phosphorescence involved
attaching fluorine substituents on the basic Irppy3 structure4 and
then moved to heteroleptic complexes with high triplet energy5
and/or strong ligand field ancillary ligands.6 More recently
homoleptic small molecule iridium(III) complexes based on
carbene7 or phenyl triazole ligands8 with saturated blue emission
have been reported. In the former case, judicious choice of host
† The University of Queensland.
‡ University of St Andrews.
§ University of Oxford.
(1) (a) Tang, C. W.; van Slyke, S. A. Appl. Phys. Lett. 1987, 51, 913. (b)
Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.;
Mackay, K.; Friend, R. H.; Burn, P. L.; Holmes, A. B. Nature 1990,
347, 539. (c) Forrest, S. R. Nature 2004, 428, 911. (d) Holder, E.;
Langeveld, B. M. W.; Schubert, U. S. AdV. Mater. 2005, 17, 1109.
(e) Grimsdale, A. C.; Chan, K. L.; Martin, R. E.; Jokisz, P. G.; Holmes,
A. B. Chem. ReV. 2009, 109, 897.
(2) (a) Baldo, M. A.; O’Brien, D. F.; You, Y.; Shoustikov, A.; Sibley, S.;
Thompson, M. E.; Forrest, S. R. Nature 1998, 395, 151. (b) Baldo,
M. A.; Thompson, M. E.; Forrest, S. R. Nature 2000, 403, 750. (c)
Bernhard, S.; Gao, X. C.; Malliaras, G. G.; Abrun˜a, H. D. AdV. Mater.
2002, 14, 433. (d) Brooks, J.; Babaya, T.; Lamansky, S.; Djurovich,
P. I.; Tsyba, I.; Bau, R.; Thompson, M. E. Inorg. Chem. 2002, 41,
3055. (e) Ikai, M.; Tokito, S.; Sakamoto, Y.; Suzuki, T.; Taga, Y.
Appl. Phys. Lett. 2001, 79, 156. (f) Gong, X.; Ma, W. L.; Ostrowski,
J. C.; Bazan, G. C.; Moses, D.; Heeger, A. J. AdV. Mater. 2004, 16,
615.
(3) Adachi, C.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. J. Appl.
Phys. 2001, 90, 5048.
(4) Adachi, C.; Kwong, R. C.; Djurovich, P.; Adamovich, V.; Baldo,
M. A.; Thompson, M. E.; Forrest, S. R. Appl. Phys. Lett. 2001, 79,
2082.
9
10.1021/ja903157e CCC: $40.75  2009 American Chemical Society
J. AM. CHEM. SOC. 2009, 131, 16681–16688 16681

A R T I C L E S
Lo et al.
has led to relatively efficient OLEDs with good CIE coordinates.
A key aspect of these studies has been the focus on small
molecules that require evaporation together with a host. This
places strict requirements on the properties of the host including
the ability to be evaporated, transport charge, and have a triplet
energy sufficiently high so as to not quench the luminescence.9
There is therefore a need for blue phosphorescent materials that
can be processed from solution for use in displays and lighting
applications.
of the dendron is above that of the core chromophore. However,
none of the dendrons reported thus far have been suitable for
devices containing saturated blue emissive iridium(III) com-
plexes either having too low a triplet energy leading to the
quenching of the luminescence, or giving dendrimers with
thermal properties not suitable for device fabrication.14,15
In this work we report the development of a new rigid high-
triplet-energy dendron that can be used to give a solution-
processable saturated blue phosphorescent dendrimer with a high
photoluminescence quantum yield. We show that when twisted
biphenyl-based dendrons are attached to the saturated blue
emissive core, fac-tris[1-methyl-5-(4-fluorophenyl)-3-n-propyl-
1H-[1,2,4]triazolyl]iridium(III) (1 in Figure 1), the properties
of the simple complex are significantly enhanced. We compare
the properties of the dendrimer 1 with the core complex 2, and
dendrimers with simple biphenyl 3 or diphenylethylene 4
dendrons and the same surface groups (Figure 1), and show
that not only are the photophysical properties improved but the
physical properties are now suitable for device fabrication.
Finally, we report the performance of single and bilayer devices
and describe the next challenges in materials development
required for efficient large-area devices containing solution-
processed deep blue phosphorescent materials.
One successful approach to solution processable phospho-
rescent materials has been the development of light-emitting
dendrimers. Phosphorescent light-emitting dendrimers generally
consist of a light-emitting core, dendrons (branches), and surface
groups, with the last playing the dominant role in the process-
ability of the materials. The dendrons that have been developed
for phosphorescent light-emitting dendrimers have fallen into
two categories: those that are electrically insulating10 and merely
play a structural role in controlling the intermolecular interac-
tions of the core, and those that are electroactive.11 In both cases
the precise control over the intercore distances means that the
emissive cores can be photophysically independent thus avoiding
interactions that lead to the quenching of the luminescence.12
However, the cores can still be sufficiently close to allow the
necessary charge transport in a device. This has made it possible
to use dendrimers to make highly efficient solution-processed
OLEDs with a neat light-emitting dendrimer layer, avoiding the
need for a host.13 The lack of a host material is attractive, both
from the point of view of simplifying device fabrication and
because finding suitable hosts for deep blue emission is
challenging. The development of deep blue phosphorescent
dendrimers places additional requirements on the properties of
the dendrons. In particular it is important that the triplet energy
Results and Discussion
To achieve the desired processing, thermal and photophysical
properties dendrimer 1 (Scheme 1) incorporates several impor-
tant design features: first, the fac-tris[1-methyl-5-(4-fluorophe-
nyl)-3-n-propyl-1H-[1,2,4]triazolyl]iridium(III) complex used as
the core emits blue phosphorescence with CIE coordinates in
solution of (0.16, 0.13). The photoluminescence (PL) peaks are
at 428 and 456 nm,8 which corresponds to energies of 2.9 and
2.7 eV, respectively. Second, the dendrons are composed of
tetrasubstituted phenyl rings, ensuring that adjacent phenyl rings
are twisted out of plane thus breaking the π-electron delocal-
ization. Molecular orbital calculations show that the chro-
mophores within the tetrasubstituted phenyl-based dendrons
have triplet energies of 3.4 eV, which is 0.5 eV higher in energy
than the emission of the core. The high triplet energy of the
chromophores within the dendrons should ensure that the PL
of the core is not quenched by the dendrons.14 Third, the
presence of the second generation and first generation dendrons
attached to the triazolyl and phenyl rings of the ligands
respectively, ensures that the emissive core is encapsulated by
the dendrons thus protecting it from intermolecular interactions
that can lead to quenching of the PL. Fourth, the rigid nature
of the dendrons will provide for a high glass transition
temperature (Tg), and finally, the 2-ethylhexyloxy surface groups
ensure solubility and that good-quality spin-coated films can
be formed.
(5) (a) Holmes, R. J.; D’Andrade, B. W.; Forrest, S. R.; Ren, X.; Li, J.;
Thompson, M. E. Appl. Phys. Lett. 2003, 83, 3818. (b) Li, J.;
Djurovich, P. I.; Alleyne, B. D.; Yousufuddin, M.; Ho, N. N.; Thomas,
J. C.; Peters, J. C.; Bau, R.; Thompson, M. E. Inorg. Chem. 2005, 44,
1713. (c) Coppo, P.; Plummer, E. A.; De Cola, L. Chem. Commun.
2004, 1774. (d) Yang, C.-H.; Cheng, Y.-M.; Chi, Y.; Hsu, C.-J.; Fang,
F.-C.; Wong, K.-T.; Chou, P.-T.; Chang, C.-H.; Tsai, M.-H.; Wu, C.-
C. Angew. Chem., Int. Ed. 2007, 46, 2418. (e) Chang, C.-F.; Cheng,
Y.-M.; Chi, Y.; Chiu, Y.-C.; Lin, C.-C.; Lee, G.-H.; Chou, P.-T.; Chen,
C.-C.; Chang, C.-H.; Wu, C.-C. Angew. Chem., Int. Ed. 2008, 47,
4542.
(6) (a) Nazeeruddin, M. K.; Humphry-Baker, R.; Berner, D.; Rivier, S.;
Zuppiroli, L.; Gra¨tzel, M. J. Am. Chem. Soc. 2003, 125, 8790. (b)
Dedeian, K.; Shi, J.; Forsythe, E.; Morton, D. C.; Zavalij, P. Y. Inorg.
Chem. 2007, 46, 1603. (c) Di Censo, D.; Fantacci, S.; De Angelis, F.;
Klein, C.; Evans, N.; Kalyanasundaram, K.; Bolink, H. J.; Gra¨tzel,
M.; Nazzeeruddin, M. K. Inorg. Chem. 2008, 44, 980.
(7) (a) Sajoto, T.; Djurovich, P. I.; Tamayo, A.; Yousufuddin, M.; Bau,
R.; Thompson, M. E.; Holmes, R. J.; Forrest, S. R. Inorg. Chem. 2005,
44, 7992. (b) Holmes, R. J.; Forrest, S. R.; Saioto, T.; Tamayo, A.;
Djurovich, P. I.; Thompson, M. E. Appl. Phys. Lett. 2005, 87, 243507.
(8) Lo, S.-C.; Shipley, C. P.; Bera, R. N.; Harding, R. E.; Cowley, A. R.;
Burn, P. L.; Samuel, I. D. W. Chem. Mater. 2006, 18, 5119.
(9) (a) Tokito, S.; Iijima, T.; Suzuri, Y.; Kita, H.; Tsuzuki, T.; Sato, F.
Appl. Phys. Lett. 2003, 83, 569. (b) Ren, X.; Li, J.; Holmes, R. J.;
Djurovich, P. I.; Forrest, S. R.; Thompson, M. E. Chem. Mater. 2004,
16, 4743.
Synthesis. Our synthetic route to the dendrons and dendrimer
is shown in Scheme 1. The strategy involved the formation of
the elaborated [1-methyl-5-(4-fluorophenyl)-3-n-propyl-1H-
[1,2,4]triazolyl] ligand with three bromine groups (19) available
for attachment to a first generation dendron 18 with a boronic
acid at its focus.
We will first discuss the synthesis of the tribromo triazole ligand
19 and then the first generation boronic acid focused dendron 18
and finally the dendrimer formation. The key step in the synthe-
(10) Lo, S.-C.; Namdas, E. B.; Burn, P. L.; Samuel, I. D. W. Macromol-
ecules 2003, 36, 9721.
(11) (a) Lo, S.-C.; Namdas, E. B.; Shipley, C. P.; Markham, J. P. J.;
Anthopolous, T. D.; Burn, P. L.; Samuel, I. D. W. Org. Electron.
2006, 7, 85. (b) Knights, K. A.; Stevenson, S. G.; Shipley, C. P.; Lo,
S.-C.; Olsen, S.; Harding, R. E.; Gambino, S.; Burn, P. L.; Samuel,
I. D. W. J. Mater. Chem. 2008, 18, 2121.
(12) (a) Adronov, A.; Fre´chet, J. M. J. Chem. Commun. 2000, 1701. (b)
Lo, S.-C.; Burn, P. L. Chem. ReV. 2007, 107, 1097. (c) Burn, P. L.;
Lo, S.-C.; Samuel, I. D. W. AdV. Mater. 2007, 19, 1675.
(13) Lo, S.-C.; Anthopoulos, T. D.; Namdas, E. B.; Burn, P. L.; Samuel,
I. D. W. AdV. Mater. 2005, 17, 1945.
(14) Lo, S.-C.; Bera, R. N.; Harding, R. E.; Burn, P. L.; Samuel, I. D. W.
AdV. Funct. Mater. 2008, 18, 3080.
(15) Lo, S.-C.; Harding, R. E.; Brightman, E.; Burn, P. L.; Samuel, I. D. W.
J. Mater. Chem. 2009, 19, 3213.
9
16682 J. AM. CHEM. SOC. VOL. 131, NO. 46, 2009

High-Triplet-Energy Dendrons
A R T I C L E S
Figure 1. Chemical structures of the parent iridium(III) complex 2, and dendrimers 1, 3, and 4.
sis of 19 was the reaction of the benzylhydrazine 8 with the
butanimidic acid ethyl ester 13 to form the triazole ring. The 3,5-
dibromo-4-methylbenzylhydrazine 8 was formed in three steps from
3,5-dibromo-4-methylbenzoic acid 5. In the first step 5 was
transformed by reduction with borane-tetrahydrofuran complex
into the benzyl alcohol 6 in an excellent yield of 94%. The
benzyl alcohol 6 was converted into benzyl bromide 7 by
treatment with phosphorus tribromide at 90 °C in a 96% yield.
The benzyl bromide 7 was stored as the advanced intermediate
with 8 being formed immediately prior to the reaction to form
the triazole ring. For the phenyl part of the eventual ligand, we
chose methyl 4-fluoro-3-iodobenzoate 9 as the starting materi-
al.15 9 was reacted with 4-bromo-2,5-dimethylphenylboronic
acid at 66 °C under Suzuki reaction conditions to give 10 in an
isolated yield of 87%. Ester 10 was then saponified to benzoic
acid 11 in a 97% yield by treatment with aqueous lithium
hydroxide in a methanol/tetrahydrofuran mixture heated at
reflux. Acid chloride 12 was obtained by heating 11 with excess
thionyl chloride, and this was directly converted into the
butanimidic acid ethyl ester 13 by reacting with ethyl butyr-
imidate hydrochloride in the presence of triethylamine at room
temperature. 13 was then used immediately in the cyclization
reaction to form the triazole ring. Benzylhydrazine 8 was formed
from 7 using hydrazine monohydrate, and then cyclization of
benzylhydrazine 8 with the preformed acid ester 13 gave the
corresponding tribrominated triazole ligand 19 in an isolated
yield of 73% for the three steps with respect to 11.
To prepare the first generation boronic acid focused twisted
biphenyl dendron 18, we used 1,3-dibromo-2-methyl-5-ni-
trobenzene 14 as the starting material. 14 was initially reacted
with ∼2.7 equiv of 4-(2-ethylhexyloxyl)phenylboronic acid at
reflux for 60 h under Suzuki reaction conditions. The coupling
was found to be sluggish, and under these conditions, a mixture
of 15 and some of the monosubstituted derivative were formed.
The reaction could be driven to completion by using a further
∼0.46 equiv of 4-(2-ethylhexyloxyl)phenylboronic acid under
the same Suzuki coupling conditions to give the desired first
generation nitro-focused dendron 15 in an overall yield of 96%.
The nitro group of 15 was then reduced to the corresponding
amine 16 using palladium on carbon and hydrogen in a 93%
yield. The amine group at the focus of 16 could then be
smoothly converted into the iodo derivative 17 in a 75% yield
via diazotization and reaction with potassium iodide in the
presence of urea.16
With the two key precursor materials in hand, the final steps
to prepare the doubly dendronized ligand 20 involved conversion
of 17 into the boronic acid 18 and subsequent Suzuki reactions
with 19. Boronic acid 18 was formed by metalation of 17 with
tert-butyllithium, followed by reaction with tri(n-butyl)borate
at -78 °C, and subsequent hydrolysis with aqueous hydrochloric
9
J. AM. CHEM. SOC. VOL. 131, NO. 46, 2009 16683

A R T I C L E S
Lo et al.
Scheme 1a
a Conditions and reagents: a. Borane-tetrahydrofuran complex, THF, 0-2 °C to r.t., Ar(g), and then ice, 94%. b. Phosphorus tribromide, heat, Ar(g), 96%.
c. Hydrazine monohydrate, EtOH, heat, Ar(g). d. 4-Bromo-2,5-dimethylphenylboronic acid, Pd(PPh3)4, 2 M Na2CO3(aq), MeOH, toluene, heat, Ar(g), 87%. e.
Lithium hydroxide, water, MeOH, THF, heat, Ar(g), 97%. f. Thionyl chloride, heat, Ar(g). g. Ethyl butyrimidate hydrochloride, NEt3, CH2Cl2, r.t., Ar(g). h. 8
and 13, CHCl3, r.t., Ar(g), 73% with respect to 11. i. 4-(2-Ethylhexyloxy)phenylboronic acid, Pd(PPh3)4, 2 M Na2CO3(aq), EtOH, toluene, heat, Ar(g), 96%. j.
10% Palladium on carbon, ethyl acetate and methanol, r.t., H2(g), 93%. k. Sodium nitrite, water, CH2Cl2, H2SO4, water, 0-2 °C, then urea, and then KI,
water, 0-2 °C to r.t., 75%. l. tert-Butyllithium, THF, -78 °C, Ar(g), then tri(n-butyl)borate -78 °C and then 3 M HCl(aq), -78 °C to r.t. m. 18, Pd(PPh3)4,
2 M Na2CO3(aq), EtOH, toluene, heat, Ar(g), 98% with respect to 19. n. Iridium(III) chloride trihydrate, water, 2-(n-butoxy)ethanol, heat, Ar(g), and then silver
trifluoromethanesulfonate, 20, diglyme, heat, Ar(g), 64% (R ) 2-ethylhexyl).
acid. Excess 18 was then reacted with 19 under standard Suzuki
conditions to give 20 in an excellent yield of 98% based on 19.
The final step in the synthesis of dendrimer 1 was the
complexation of the ‘doubly dendronized’ ligand 20 with
‘iridium(III)’. This was achieved by following the standard two-
step procedure.10 First, 20 was reacted with iridium(III) chloride
trihydrate in a water/2-(n-butoxy)ethanol mixture heated at
reflux, and then the intermediate chloro-bridged dimer was
reacted with an excess of ligand 20 in the presence of silver
trifluoromethane sulfonate in diglyme at high temperature. Under
these conditions, dendrimer 1 was formed in a good yield of
64% for the two steps after purification by column chromatog-
raphy. As shown in Figure 2, the isotopic pattern observed in
the MALDI-TOF mass spectrum was consistent with the
calculated molecular weight distribution for 1. The H NMR
1
spectrum of 1 displayed one simple set of ligand signals
indicating that only one isomer of the core complex had formed.
The iridium(III) complex at the core was assigned as the facial
isomer by analogy to the core complex 2 for which a crystal
structure has been obtained.8 Further evidence for there being
only a single isomer came from the 19F NMR spectrum in which
only one 19F signal was observed.
The dispersity and size of the doubly dendronized ligand 20
and dendrimer 1 were investigated using gel permeation
chromatography (GPC) against polystyrene standards. Both the
dendronized ligand and dendrimer were found to be monodis-
(16) Endres, A.; Maas, G. Tetrahedron 2002, 58, 3999.
9
16684 J. AM. CHEM. SOC. VOL. 131, NO. 46, 2009

High-Triplet-Energy Dendrons
A R T I C L E S
Figure 2. Parent peak of the MALDI-TOF mass spectrum of dendrimer 1 (a. calculated isotopic distribution; b. observed distribution).
Table 1. Summary of the Properties of the Iridium(III) Complex 2 and Dendrimers 1, 3, and 4
solutiona
filmb
peaks
(nm)
CIEs
(x,y)
kr
knr
Peaks
(nm)
CIEs
(x,y)
τ
kr
knr
Tg
E1/2
PLQY (%) τ (µs) (×105 s-1
)
(×105 s-1
)
PLQY (%) (µs) (×105 s-1
)
(×105 s-1
)
(°C) (ox)c (V)
28
428, 456 (0.16, 0.13)
27
59
45
94
1.25
22
1.65
3.6
2.2
0.27
2.7
2.6
5.8
0.19
3.3
460
(0.17, 0.19)
18
17
49
60
-
-
2.1
2.8
-
-
2.3
2.1
-
-
2.4
1.4
-
75
–3
0.50
0.53
0.45
0.61
314 441, 468 (0.15, 0.16)
443, 467 (0.19, 0.22)
441, 469 (0.16, 0.19)
438, 466 (0.16, 0.16)
415 441, 470 (0.16, 0.18)
1
435, 465 (0.16, 0.16)
0.17
148
a Measured in toluene. b Spin-coated from dichloromethane solution. c Quoted against the ferricenium/ferrocene couple.
perse. The hydrodynamic radii of the dendronized ligand 20
and dendrimer 1 were determined from the Mvs given by GPC
device fabrication. These results clearly indicate that for
designing light-emitting dendrimers, it is important to control
the ratio of flexible to nonflexible units. Dendrimer 1 also
showed excellent solubility in common organic solvents and
good-quality thin films could be prepared by spin-coating.
j
in combination with the Hester-Mitchell equation and Mark-
Houwink relationship.17 Ligand 20 and iridium(III) dendrimer
j
1 had Mvs of 2542 and 4528 with corresponding hydrodynamic
radii of 10.9 and 15.2 Å, respectively. The hydrodynamic radius
of 1 is larger than those for dendrimers 3 (biphenyl)14 and 4
(diphenylethylene)15 (Figure 1), which have hydrodynamic radii
of 12.8 and 13.8 Å, respectively. The larger size of 1 is due to
the fact that it contains one first-generation (attached via a xylyl
group) and one second-generation dendron while 3 and 4 only
have two first generation dendrons per ligand.
Thermogravimetric analysis (TGA) and differential scanning
calorimetry were used to determine the thermal properties of
dendrimer 1. TGA indicated that dendrimer 1 had good thermal
stability with a decomposition temperature, corresponding to a
5% weight loss, at 417 °C. Importantly, dendrimer 1 was found
to have a glass transition (Tg) at 148 °C using a scan-rate of
300 °C/min. In contrast dendrimers 3 and 4 had the much lower
Tgs of 75 °C14 and -3 °C,15 respectively. Dendrimer 4 has a
low Tg due to the surface groups and flexible 1,2-diphenyleth-
ylene dendrons. By moving to the more rigid dendrimer 3 with
biphenyl dendrons, the Tg is increased. In the case of 1 the
twisted biphenyl dendrons are more rigid (less rotation around
the biphenyl bonds), and this coupled with the use of a second-
generation dendron10 leads to the high Tg, which is suitable for
Photophysical and Electronic Properties. We first investigated
the effect of the twisted biphenyl dendrons on the electronic
properties of the core complex. The electronic properties of 1
were investigated with cyclic voltammetry at a scan rate of 40
mV s-1. As for the parent complex 2,8 we were able to observe
a chemically reversible one-electron oxidation for 1 (Table 1).
The key point to note from the electrochemical studies is that
the E1/2 for the oxidation of 1 (0.61 V) was similar to that of
the core, indicating that the electroactive component of the
dendrimer is the emissive core. The oxidation of 1 occurs at a
slightly more positive potential than that of 4,15 reflecting the
different electronic effects of the alkyl and xylyl moieties.
The UV-visible absorption spectra of dendrimers 1, 3 (with
simple biphenyl dendrons), and 4 (with phenylethylene den-
drons), and their parent small molecule iridium(III) complex 2,
are shown in Figure 3. For iridium(III) complexes, the absorption
spectra can be divided into two regions: the longer wavelength
(>280 nm) absorptions are associated with a mixture of singlet
and triplet ‘metal-to-ligand’-like transitions while the absorptions
in the shorter wavelength region are primarily due to the singlet
π-π* transitions of the ligands (see spectrum of 2 in Figure
3). Dendrimer 1 has a peak with a high molar extinction
coefficient at around 256 nm that is mainly due to the π-π*
transitions of the chromophores within the dendrons. The peak
(17) Burn, P. L.; Beavington, R.; Frampton, M. J.; Pillow, J. N. G.; Halim,
M.; Lupton, J. M.; Samuel, I. D. W. Mater. Sci. Eng., B 2001, 85,
190.
9
J. AM. CHEM. SOC. VOL. 131, NO. 46, 2009 16685

A R T I C L E S
Lo et al.
Figure 3. Solution UV-visible spectra for 1-4 in dichloromethane.
Figure 4. Solution and film photoluminescence spectra for 1, 3, and 4.
The spectra have been normalized and offset for comparison (the excitation
wavelength is 360 nm for solution and 325 nm for film).
absorption for the biphenyl dendrons of 3 is at 271 nm, which
is red-shifted compared to 1 due to biphenyl dendrons being
able to adopt a more planar arrangement. The dendron absorp-
tion of 4 at around 270 nm is much weaker due to the phenyl
rings being isolated chromophores.
able feature of the emissive properties of 4 was that in the solid-
state the PLQY remained a creditable 49% although the CIE
coordinates were less blue.
Dendrimer 1 has a solution PLQY of 94%, which is over
three times that of the core complex. The PL lifetime of 3.6 µs
indicates that the emission is phosphorescent in nature and due
to the core. This shows that the twisting of the biphenyl units
in the dendrons raises their singlet and triplet energies such that
the excited state is localized on the core complex. The radiative
decay rates of 1 and 2 are the same, and hence the improvement
in PLQY is primarily due to the dramatic reduction in the
nonradiative decay rate for 1 (Table 1). We attribute the reduced
non-radiative decay rate to the rigidifying effect of the dendrons
which reduces the geometry change in the excited state, and
hence also the amount of vibrational quenching. The strong
weighting of the (0,0) transition in the solution PL and the lack
of red tail in the film PL give rise to the good blue emission
from 1. Importantly, the attachment of the ‘twisted’ biphenyl-
based dendrons has not changed the color of the emission
appreciably with the CIE coordinates of 1 being (0.16, 0.16)
versus (0.16, 0.13) for 2. Interestingly, in forming films of 2
(the unadorned complex) the color of the emission changes
notably with CIE coordinates of (0.17, 0.19). In contrast, the
emission color of 1 remains essentially unchanged with the
peaks of the emission within a few nanometers and the CIE
coordinates being (0.16, 0.16). The fact that the color has not
changed suggests that the intermolecular interactions that lead
to aggregate or excimer emission have been controlled. This
can be understood by considering the arrangement of the
dendrons around the core complex. The core iridium(III)
complex is the facial isomer, and hence having dendrons on
each of the aromatic rings of all the ligands means that the core
is encapsulated by the dendrons. This is reflected in the
hydrodynamic radii of 1, which is the largest of all three
dendrimers shown here. The hydrodynamic radius of 4 is larger
than that of 3 due to the extension of the dendrons caused by
the linking 1,2-diphenylethylene units. For 1, the radius (15.2
Å) is larger still, due to the fact that the dendron on the triazolyl
The next step in the study was to see whether the rigid
dendrons had a positive effect on the emissive properties of
the phosphorescent core complex. The solution and film PL
spectra of 1, 3, and 4 are shown in Figure 4 and the emissive
properties summarized in Table 1. The parent complex core 2
has a solution photoluminescence quantum yield (PLQY) of
27% at room temperature, CIE coordinates of (0.16, 0.13), and
a photoluminescence (PL) lifetime of 1.25 µs. A PL lifetime of
1.25 µs is consistent with phosphorescence associated with a
ligand-to-metal charge transfer type emission. It is important
to note that the relatively low PLQY of 2 is due to a relatively
large nonradiative decay rate, which has been attributed to strong
vibrational coupling.18 We have observed that attachment of
biphenyl-based dendrons can increase the PLQY of simple
complexes, and this may be in part due to the large group
decreasing the vibrational modes that lead to the quenching of
the luminescence. For example, attachment of biphenyl-based
dendrons to 2 to give 3 was found to increase the solution PLQY
from 27% to 59% with little change in the emission color.
However, the addition did cause a large change in the PL
lifetime, with it increasing from 1.25 µs for 2 to 22 µs for 3.
The increase in PL lifetime is due to the core and the dendrons
having similar triplet energies allowing the triplet excitons to
spend time on the dendrons although emission still occurs from
the complex at the core. The fact that the triplet exciton could
reside on the dendrons of 3 meant that there was a dramatic
drop in the solid-state PLQY, which fell to 17%.14 By replacing
the biphenyl dendrons of 3 with the 1,2-diphenylethylene
dendrons of 4 it proved possible to confine the triplet exciton
on the core of the dendrimer.15 Dendrimer 4 had a respectable
solution PLQY of 45% and a PL lifetime of 1.65 µs with the
latter indicating emission from the core complex. The remark-
(18) Harding, R. E.; Lo, S.-C.; Shipley, C. P.; Burn, P. L.; Samuel, I. D. W.
Org. Electron. 2008, 9, 377.
9
16686 J. AM. CHEM. SOC. VOL. 131, NO. 46, 2009

High-Triplet-Energy Dendrons
A R T I C L E S
Figure 6. Bilayer device structure and energy level diagram for the
materials showing HOMO, LUMO, and triplet energies (T1).
Figure 5. Film PL spectrum for dendrimer 1 and EL spectra of the single
(ITO/dendrimer 1/Ca:Al) and bilayer (ITO/dendrimer 1/PBD/LiF:Al)
devices. The inset is a photo of the single layer device.
ring is second-generation and the first-generation dendron is
linked to the ligand phenyl ring via another phenyl moiety. The
benefit of encapsulating the emissive core well can be seen in
the solid-state PLQY of 1, which at 60% is very high for a
material that emits saturated blue phosphorescence. The lower
solid-state PLQY is primarily due to a larger nonradiative decay
rate. Large nonradiative decay rates in this family of iridium(III)
complexes has been attributed to enhanced coupling of the
excited state to vibrational modes,18 which is exhibited by an
increase in the strength of the (0,1) to the (0,0) transition. It
can be seen that for 1 there is a change in the ratio of the (0,1)
to (0,0) transition in going from solution to the solid-state with
the (0,1) transition being larger in the film (Figure 4). This
suggests that there is stronger coupling of the excited state and
vibrational modes for 1 in the solid-state than in solution, which
could arise from the packing of the dendrimer in the film causing
a slightly different conformation.
Organic Light-Emitting Diodes. Given that 1 can be spin-
coated to form good quality thin films that are morphologically
stable, simple OLEDs were fabricated. The device performance
is summarized here with the full characteristics shown in the
Supporting Information. We first fabricated a single layer device
with an indium tin oxide (ITO) anode and calcium cathode [ITO/
dendrimer 1 (100 nm)/Ca/Al]. The color of the emission was
blue (see inset of Figure 5) with CIE coordinates of (0.17, 0.20).
Although the blue color is very good, the small reduction in
color saturation for the EL spectrum versus the film PL is due
to a difference in weighting of the (0,0), (0,1), and (0,2) tra-
nsitions and an increase in the emission at longer wavelengths.
At 100 cd/m2 the device had an external quantum efficiency of
0.41% at 26.7 V, giving power and luminous efficiencies of
0.08 lm/W and 0.65 cd/A, respectively. The high voltage and
increasing efficiency at increasing bias (see Supporting Informa-
tion) indicate that large barriers to charge injection limited the
efficiency. The work function of oxygen plasma-treated ITO is
around 4.8 eV19 depending on its pretreatment while calcium
Figure 7. Time-resolved PL of films of dendrimer 1 in PMMA, TPBI, or
PBD. Excitation ) 393 nm.
has a work function of 2.9 eV. To gain an understanding of the
barriers to charge injection (summarized in Figure 6), we
calculated the Highest Occupied Molecular Orbital (HOMO)
energy level to be 5.4 eV from the E1/2 for the oxidation (0.61
V against the ferrocenium/ferrocene couple) given that the
ionization potential of ferrocence is 4.8 eV,20 and the Lowest
Unoccupied Molecular Orbital (LUMO) energy of 2.2 eV was
then estimated by adding the singlet energy of 3.2 eV of the
core complex8 to the HOMO energy level.
In previous work on OLEDs containing phosphorescent
dendrimers the device efficiency was found to be high when
an electron transporting hole blocking layer was used.13,14,21
Therefore, in order to improve electron injection into the devices
we prepared bilayer OLEDs with a neat film of dendrimer 1
and an electron transport layer (ETL) of either 1,3,5-tris(2-N-
phenylbenzimidazolyl)benzene (TPBI) or 2-(4-biphenyl)-5-(4-
(20) D’Andrade, B. W.; Datta, S.; Forrest, S. R.; Djurovich, P.; Polikarpov,
E.; Thompson, M. E. Org. Electron. 2005, 6, 11.
(21) (a) Lo, S.-C.; Male, N. A. H.; Markham, J. P. J.; Magennis, S. W.;
Burn, P. L.; Salata, O. V.; Samuel, I. D. W. AdV. Mater. 2002, 14,
975. (b) Anthopoulos, T. D.; Frampton, M. J.; Namdas, E. B.; Burn,
P. L.; Samuel, I. D. W. AdV. Mater. 2004, 16, 557.
(19) Kim, K.-P.; Hussain, A. M.; Hwang, D.-K.; Woo, S.-H.; Lyu, H.-K.;
Baek, S.-H.; Jang, Y.; Kim, J.-H. Jpn. J. Appl. Phys. 2009, 48, 021601.
9
J. AM. CHEM. SOC. VOL. 131, NO. 46, 2009 16687

A R T I C L E S
Lo et al.
tert-butylphenyl)-1,3,4-oxadiazole (PBD) (ITO/dendrimer 1/ETL/
LiF/Al). For the devices with TPBI, the quantum efficiency
increased with increasing voltage, and at 18 V a brightness of
100 cd/m2 was obtained with an EQE of 2.2% (0.46 lm/W, 2.6
cd/A). For PBD, the EQE was highest at low brightness and
decreased as the voltage increased. The maximum external
quantum efficiency (EQE) was 3.9% (5.0 V, 3.4 lm/W, 5.4 cd/
A) and at a brightness of 100 cd/m2 the EQE was 2.1% (15.0
V, 0.61 lm/W, 2.9 cd/A). The EL spectrum (Figure 5) shows
deep blue emission was improved relative to the single layer
device and the CIE coordinates of (0.16, 0.17) are very close
to the PL coordinates of (0.16, 0.16). The EQE of 2.1% at 100
cd/m2 is very good for such a simple device structure and not
too far from the efficiencies at comparable brightnesses (≈8%
at 100 cd/m2) reported for much more complex state-of-the-art
deep blue phosphorescent devices.22 These latter devices are
prepared by evaporation and contain seven layers and six
compounds in contrast to the two layers and two materials of
this work. The improved performance of the bilayer devices
arises from better charge balance due to the electron transporting
and hole blocking action of the ETLs. LiF-Al has a work
function of 2.6 eV,23 TPBI has a reduction potential of 2.7 V
versus the ferrocenium/ferrocene couple, which corresponds to
a LUMO energy of approximately 2.1 eV (based on ferrocene
having an ionization potential of 4.8 eV20), and PBD has a
LUMO of energy of 2.4 eV24 (see Figure 6). This means that
there is still a substantial barrier to electron injection in the TPBI
device, which explains its higher operating voltage than the PBD
device. The relatively high operating voltage of both devices is
also likely to be due to the remaining substantial barrier to hole
injection.
dendrimer 1. To probe whether this is indeed a factor, we
compared the PL lifetime of emission from dendrimer 1 in
blends with poly(methylmethacrylate) (PMMA, used as an inert
host), TPBI, and PBD (Figure 7). It can be readily seen that
the PL decays much faster in the blends with TPBI and PBD.
The time for the emission to fall to 1/e of its initial value is
2.18 µs in PMMA, 0.23 µs in TPBI, and 0.024 µs in PBD. The
shortened lifetime indicates that the ETLs quench the lumines-
cence of the dendrimer.
Clearly both the barrier to hole injection and quenching of
the EL by the ETL are limiting device performance. Therefore,
to improve the efficiency of the devices, a hole-injection layer
(which is insoluble in the dendrimer processing solvent) needs
to be included. In addition, the hole and electron transport layers
will have to have triplet energies higher than that of the emissive
layer to enable solution processed deep blue phosphorescent
dendrimers to give optimum performance in an OLED.
Conclusion
In conclusion, we have shown that dendrimers provide a route
to solution-processable saturated blue phosphorescent materials
with a high solid-state photoluminescence quantum yield and
promise for OLEDs. This advance has been achieved by the
use of new high triplet energy ‘twisted’ dendrons that can be
used to confine the excited state on a blue emissive phospho-
rescent core. This new family of dendrons was found to
significantly reduce the nonradiative decay rate while leaving
the radiative decay rate essentially unchanged, leading to a blue
phosphorescent dendrimer with the exceptionally high solution
PLQY of 94% at room temperature. The film PLQY of 60%
with CIE coordinates of (0.16, 0.16) means that the dendrimer
is an excellent candidate for OLEDs. Preliminary devices
indicate that good efficiencies can be achieved, and the next
steps are the design of hole and electron transport materials with
suitable triplet energies.
The external quantum efficiency of the devices is less than
might be expected from a film with a PLQY of 60%. One
potential reason for the lower EQEs in the bilayer devices is
that excitons will be formed near the dendrimer:ETL interface
and could then be quenched by the ETL. It has been reported
that the triplet energies for TPBI and PBD are 2.6 eV25 and 2.5
eV,26 respectively, which is lower than the emission energy of
Acknowledgment. This work was supported by CDT Oxford
Ltd and the EPSRC. Professor Paul L. Burn is recipient of an
Australian Research Council Federation Fellowship (project number
FF0668728). Professor Ifor Samuel holds an EPSRC Senior
Research Fellowship (grant number EP/C542398).
(22) Chiu, Y.-C.; Hung, J.-Y.; Chi, Y.; Chen, C.-C.; Chang, C.-H.; Wu,
C.-C.; Cheng, Y.-M.; Yu, Y.-C.; Lee, G.-H.; Chou, P.-T. AdV. Mater.
2009, 21, 2221.
Supporting Information Available: Experimental procedures,
spectral data for compounds, and device characteristics. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(23) Hung, L. S.; Tang, C. W.; Mason, M. G.; Raychaudhuri, P.; Madathil,
J. Appl. Phys. Lett. 2001, 78, 544.
(24) Janietz, S.; Wedel, A. AdV. Mater. 1997, 9, 403.
(25) Fukagawa, H.; Watanabe, K.; Tokito, S. Org. Electron. 2009, 10, 798.
(26) Suzuki, M.; Tokito, S.; Sato, F.; Igarashi, T.; Kondo, K.; Koyama,
T.; Yamaguchi, T. Appl. Phys. Lett. 2005, 86, 103507.
JA903157E
9
16688 J. AM. CHEM. SOC. VOL. 131, NO. 46, 2009