Synthesis of new iridium complexes and their electrophosphorescent properties in polymer light-emitting diodes

Article abstract of DOI:10.1039/b205310a

Several new iridium complexes with p-substituted 2-phenylpyridine (R-PPy) ligands have been synthesized and characterized. The complexes were incorporated into phosphorescent polymer light-emitting devices using soluble poly[1,4-bis(6′-cyano-6′-methylhept

Full text of DOI:10.1039/b205310a

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis of new iridium complexes and their electrophosphorescent
properties in polymer light-emitting diodes
Weiguo Zhu, Chuanzhen Liu, Lijun Su, Wei Yang, Ming Yuan and Yong Cao*
Institute of Polymer Optoelectronics Materials and Devices, South China University of
Technology, Guangzhou 510640, P. R. China. E-mail: poycao@scut.edu.cn
Received 31st May 2002, Accepted 3rd October 2002
First published as an Advance Article on the web 6th November 2002
Several new iridium complexes with p-substituted 2-phenylpyridine (R-PPy) ligands have been synthesized and
characterized. The complexes were incorporated into phosphorescent polymer light-emitting devices using
soluble poly[1,4-bis(6’-cyano-6’-methylheptyloxy)phenylene] (CNPPP) as the host and the resultant materials
compared with Ir(PPy)3-doped devices. Green electrophosphorescence was observed, with peak emission at
about 495 and 515 nm. Among the devices fabricated, highly efficient polymer light-emitting diodes were
obtained with CNPPP doped with fac-tris[2-(4-tert-butylphenyl)pyridinato]iridium. An external quantum
efficiency of 4.4% (photoluminescence/electroluminescence) and a luminous efficiency of 10 cd A21 were
obtained at 120 cd m22. These values remain at 4.2% and 10 cd A21, respectively, at 2500 cd m22. The
improvement is attributed to improved interaction between the guest and host, and to better and more
complete energy transfer from the host singlet to the guest triplet state. These results demonstrate that efficient
electrophosphorescence is not limited to small molecule organic light-emitting diodes, it can also be achieved in
devices made with polymer hosts.
PtOEP-doped polyfluorene device, respectively.11 The quan-
Introduction
tum efficiencies for phosphorescent dye-doped polymer light-
emitting diodes (PLED) prepared so far are much lower than
those of dye-doped OLEDs using small molecules as host.
It is of interest to use polymers instead of organic small
molecules as host materials, since PLEDs have the potential to
be used for large area displays which can be made using simpler
processes and at a lower cost. In order to improve performance
of phosphorescence-based PLEDs and decrease phase segrega-
tion between host and guest, it is very important to design
highly efficient phosphorescent guest materials.12 In this paper,
we report the synthesis of three new iridium complexes
(Scheme 1) which show high phophorescence efficiencies, and
report their optical and electroluminescent (EL) properties. We
found that altering the length of the of alkyl substitution on the
phenylpyridine ligands could lead to significant differences in
EL performance.
Recently, high efficiency organic light-emitting diodes (OLEDs)
have been fabricated using electrophosphorescent molecules
such as 2,3,7,8,12,13,17,18-octaethyl-21H,23H-porphyrinato-
platinum
(PtOEP),1,2
fac-tris(2-phenylpyridinato)iridium
[Ir(PPy)3]3–5 and bis{2-[2’-benzo(4,5-a)thienyl]pyridinato-N,C3}-
iridium acetylacetonate [(Btp)2Ir(acac)].6,7 This has heralded
a breakthrough in improving external quantum efficiency
and luminous efficiency in OLEDs. Unlike fluorescence-based
light-emitting diodes (LEDs), the quantum efficiency of phos-
phorescence-based LEDs are not limited and can theoretically
reach 100% by using both singlet and triplet excitons. As a
result, they have attracted attention worldwide, and are
considered to be promising candidates for realizing commercial
full color LED displays.
Much effort has been focused on developing highly efficient
phosphorescent guest materials, selecting appropriate host
materials and designing efficient device structures. At present,
most of the phosphorescent guest materials investigated are
complexes of heavy metals, as strong spin–orbit coupling leads
to singlet–triplet state mixing, thus resulting in high efficiency
electrophosphorescence in LEDs. Ortho-metalated complexes
with d6 and d8 metal ions such as Pt(II),2 Os(II),8 and Ir(III)3–7,9
have been extensively investigated for electrophosphorescent
device applications. An energy efficiency of 40 lm W21 and an
external quantum efficiency of 15% were reported in OLEDs
using an iridium complex as the phosphorescent emitter.3
Research into host materials has mainly focused on small
molecules. Typically, small molecule blue emitters such as
tris(8-hydroxyquinolinato)aluminium (Alq3),1 4,4’-dicarbazole-
biphenyl (CBP),3,4 2,9-dimethyl-4,7-diphenyl-1,10-phenanthro-
line (BCP),3,4 1,3-bis(tert-butylphenyl)-1,3,4-oxadiazole (OXD7),7
3-phenyl-4-(1’-naphthyl)-5-phenyl-1,2,4-triazole (TAZ)3,7 have
been used in electrophosphorescent devices. In contrast, there
are only a few reports on phosphorescence-based LEDs
using a polymer as the host material. Peak quantum efficiencies
of 1.9 and 3.5% have been reported
Experimental details
Reagents
p-tert-Butylbromobenzene, 2-bromopyridine, n-butyllithium,
4-bromophenol, and anhydrous zinc chloride were pur-
chased from Aldrich. Tetrakis(triphenylphosphine)palladium
was obtained from TCI Co. Bromobenzene and anhydrous
aluminium chloride were used as received. Anhydrous
tetrahydrofuran was distilled over sodium–benzophenone
under nitrogen prior to use, CS2 was distilled over calcium
chloride. 4-(3’,7’-Dimethyloctyloxy)bromobenzene, decanoyl
chloride, 1-bromo-4-decylbenzene, and 2,5-dichloro-1,4-bis-
(6’-cyano-6-’methylheptyloxy)benzene were prepared following
published procedures.13–16
Syntheses
1-Bromo-4-decanoylbenzene (1). In a 250 ml three-necked
round-bottom flask, 39.0 g (0.25 mol) of bromobenzene and
36.5 g (0.27 mol) of anhydrous aluminium chloride were added
to 100 mL of dry CS2, and the mixture was cooled to 5–10 uC.
for
a
Ir(PPy)3-doped polyvinylcarbazole device10 and
a
50
J. Mater. Chem., 2003, 13, 50–55
DOI: 10.1039/b205310a
This journal is # The Royal Society of Chemistry 2003

View Article Online
Scheme 1 Synthetic routes to and chemical structures of the materials used herein.
45.0 g (0.24 mol) of decanoyl chloride were added dropwise
with vigorous stirring over 2 h. The resulting deep brown
mixture was refluxed for 2 h. The mixture was cooled to room
temperature and cautiously poured into a mixture of 250 g of
crushed ice and 150 mL of concentrated hydrochloric acid in a
beaker. The organic layer was extracted with ether. The
resulting ether layer was washed successively with dilute
sodium hydroxide (1%, w/w) and water, and dried over
magnesium sulfate. The ether was evaporated to give a orange
oil, which yielded a colorless liquid product (b.p. 141–143 uC/
0.1 mmHg ) after vacuum distillation.
25.0 mmol) and tetrakis(triphenylphosphine)palladium (0.90 g,
0.078 mmol) in THF (15 mL) was added, followed by stirring
for 1.5 h, and a yellow solution was obtained. The reaction
mixture was neutralized to pH 8 with solution of aqueous
sodium carbonate (10%, w/w), then the oil was removed and
the aqueous layer was extracted with diethyl ether (2 6 25 mL).
The ether and the oil layer were mixed together, washed with
water, dried over magnesium sulfate, and evaporated to give
an orange–red crude product. The crude product was purified
by column chromatography (silica gel, n-hexane–ethyl ace-
tate 3 : 1) to provide 3.10 g of a yellow liquid in 58.7% yield.
GC-MS, m/e: 211 (M1, 23.1), 196 ([M 2 CH3]1, 100), 154
([M 2 C4H9]1, 4.5%).
1-Bromo-4-decylbenzene (2). 31.1 g (0.1 mol) of 1-bromo-4-
decanoylbenzene, 15 mL (0.3 mol) of hydrazine monohydrate,
and 18.5 g (0.33 mol) of potassium hydroxide were dissolved
in 80 mL of 2-hydroxyethyl ether. The mixture was heated to
130 uC and stirred for 2 h, then the excess hydrazine hydrate
was removed. The resulting mixture was reacted for 2 h at
200 uC, cooled to room temperature, and slowly poured into
100 mL of 18% (w/w) aqueous hydrochloric acid to give an oil
and an aqueous layer. The oil was removed and the aqueous
layer was extracted with diethyl ether (2 6 20 mL). The ether
and the oil layer were mixed together, washed with water, dried
over magnesium sulfate, and the ether evaporated. The residue
was distilled under 0.1 mmHg of vacuum and the colorless
liquid product collected at 134–136 uC. GC-MS, m/e: 297 (M1,
2.6), 298 ([M 1 1]1, 25), 296 ([M 2 1]1, 25), 171 ([M 1 1 2
C9H19]1, 100), 169 ([M 2 1 2 C9H19]1, 100%).
2-[4’-(3@,7@-Dimethyloctyloxy)phenyl]pyridine (4). This com-
pound was synthesized according to the procedure used for
the preparation of compound 3. A yellow liquid product
was obtained in 54.2% yield. GC-MS, m/e: 311 (M1, 7.9), 171
([M 1 1 2 C10H21]1, 100), 154 ( M 2 OC10H21]1, 6.6%).
2-(4’-Decylphenyl)pyridine (5). This compound was synthe-
sized according to the procedure used for the preparation of
compound 3. A yellow liquid product was obtained in 51.1%
yield. GC-MS, m/e: 295 (M1, 21.1), 182 ([M 2 C8H7]1, 67.1),
167 ([M 2 C9H19]1, 100%).
fac-Tris[2-(4’-tert-butylphenyl)pyridinato]iridium, Ir(BuPPy)3
(6). 200 mg (0.4 mmol) of tris(acetylacetonato)iridium
[Ir(acac)3] and 646 mg (3.06 mmol) of 2-(4-tert-butylphenyl)-
pyridine (BuPPy) were added to 20 mL of degassed glycerol in a
three-necked flask, and the mixture refluxed under a nitrogen
atmosphere for 12 h. The reaction mixture was cooled to room
temperature and then poured into 120 mL of dilute aqueous
hydrochloric acid (0.1 M) to give a red–brown precipitate. The
precipitate was filtered off, washed with water and methanol,
and purified by column chromatography (silica gel, dichloro-
methane) to give 0.205 g of an orange product in 62.3% yield.
Elemental analysis found: C, 65.22; H, 6.00; N, 4.93; calcd: C,
65.66; H, 5.88; N, 5.11%.
2-(4-tert-Butylphenyl)pyridine (3). 1-Bromo-4-tert-butylben-
zene (5.33 g, 25.0 mmol) was dissolved in 50 mL of THF,
and the solution was cooled to between 270 and 280 uC
under nitrogen. n-Butyllithium solution (1.6 M in n-hexane,
15.6 mL, 25.0 mmol) was added dropwise over 10 min and
the mixture was stirred for 15 min. Then, a solution of zinc
chloride (3.40 g, 25.0 mmol) in THF (50 mL) was added
dropwise at between 245 and 250 uC for 15 min and the
mixture stirred for 30 min. After warming the reaction mixture
to room temperature, a solution of 2-bromopyridine (3.95 g,
J. Mater. Chem., 2003, 13, 50–55
51

View Article Online
fac-Tris{2-[4’-(3@,7@-dimethyloctyloxy)phenyl]pyridinato}iri-
dium, Ir(DecOPPy)3 (7). This compound was synthesized
according to the procedure used for the preparation of
compound 6. An orange–yellow product was obtained in
50% yield. Elemental analysis found: C, 67.09; H, 7.56; N, 3.77;
calcd: C, 67.35; H, 7.50; N, 3.71%.
vacuum vapor deposition at a pressure below 3 6 1024 Pa.
Immediately after deposition of Ba, a 200 nm capping layer of
aluminium was deposited on top of the Ba metal layer. The
cathode area defines the active area of the device, which is
typically 0.15 mm2 for the devices in this study. The deposition
speed and the thickness of the barium and aluminium layers
were monitored with a Sycon Instruments Model STM-100
thickness/rate meter. All steps except the PVK-coating were
performed in N2-filled dry boxes with oxygen and water
contents of less than 1 ppm. I–V Characteristics were measured
with a computerized Keithley 236 source measuring unit.
The luminance of the devices was measured with a calibrated
photodiode.
fac-Tris[2-(4’-decylphenyl)pyridinato]iridium, Ir(DecPPy)3
(8). This compound was synthesized according to the proce-
dure used for the preparation of compound 6. A green–yellow
product was obtained in 50% yield. Elemental analysis found:
C, 70.08; H,7.99; N, 3.83; calcd: C, 70.35; H, 7.87; N, 3.92%.
Poly[1,4-bis(6’-cyano-6’methylheptyloxy)phenylene], CNPPP
(9).16 Zinc powder (2.88 g, 44.2 mmol), nickel chloride (0.08 g,
0.63 mmol), bipyridine (0.095 g, 0.6 mmol), triphenylphosphine
(1.90 g, 7.3mmol) and 2,5-dichloro-1,4-bis(6’-cyano-6-’methyl-
heptyloxy)benzene (4.03 g, 13.5 mmol) were added to a three-
necked flask. Under a nitrogen atmosphere, DMF (40 ml) was
added to the flask via syringe. The mixture was stirred at 80 uC
for 48 h, cooled to room temperature, then poured into the
solution of 75 mL methanol and 10 ml concentrated HCl. The
resulting mixture was stirred overnight to yield a sticky
mixture, which was filtered to give precipitate. The precipitate
was dissolved in THF to form a pale yellow solution. A white
precipitate was obtained by adding the solution to water. The
product was purified by reprecipitaion from THF with acetone.
Results and discussion
Synthesis of iridium complex
The synthesis of the ligands and iridium complexes is depicted
in Scheme 1. The 2-arylpyridine ligands were synthesized via
Pd-catalyzed cross-coupling using the modified method
reported previously.18 The use of zinc (intermediate electro-
negativity) rather than lithium (low electronegativity) leads to
favorable results in Pd-catalyzed coupling between aryl and
pyridyl. The reaction was carried out in one step with a yield
of over 50%. Iridium complexes were prepared according to
the procedure reported previously.19 The reaction between
arylpyridines and tris(acetylacetonato)iridium in refluxing
glycerol can afford cycloiridium complexes. The yields of
about 55% were obtained by varying the ratio of 2-arylpyridine
to tris(acetylacetonato)iridium, using a higher ratio than that
reported for fac-tris(2-phenylpyridinato)iridium in the litera-
ture.19 The elemental analysis results for the complexes are
consistent with the structures shown in Scheme 1.
Instrumentation
All GC-MS data were obtained using a Finnigan Trace GC-
MS-2000 Series system. All NMR spectra were acquired with a
Bruker Dex-400NMR instrument, using CDCl3 as a solvent.
Elemental analysis was performed on a Harrios elemental
analyzer. The UV-Visible absorption spectra of films con-
taining iridium complexes were recorded using an HP-8453
UV-Visible system. Photoluminescence (PL) and electrolumi-
nescence (EL) spectra were obtained with an Oriel InstaSpec IV
CCD system. The PL quantum efficiencies of the blends as
solid thin films on quartz substrates were measured with a
LabSphere IS80 integrating sphere, together with a UDT S370
digital photometer, according to the method described by
Greenham et al.17 Excitation for the measurement of PL
spectra and PL efficiency was achieved using the 325 nm line
from a He–Cd laser (Melles Griot).
Optical and photoluminescent properties
Fig. 1 shows the UV-Visible absorption spectra of three
complexes in solid films on quartz substrates. All three iri-
dium complexes have almost identical UV-Visible absorption
spectra, similar to that of Ir(PPy)3, at a wavelength from 200
to 800 nm. An intense absorption band is observed in the
ultraviolet parts of the spectra from 200 to 400 nm. The intense
absorption band around 290 nm can be assigned to a spin-
allowed 1p–p* transition on the cyclometalated ligands, and the
broad absorption band at lower energy (380 nm) is typical for
spin-allowed metal to ligand charge-transfer (1MLCT) transi-
tions, as has been discussed in the literature.19 The 380 nm peak
for Ir(DecOPPy)3 is slightly blue-shifted compared with those
due to Ir(DecPPy)3 and Ir(BuPPy)3. This indicates that
modification of the PPy molecule by adding alkyl or alkoxy
Device fabrication and characterization
The polymer light-emitting diodes were fabricated on com-
mercial indium tin oxide (ITO) substrates with a sheet
21
resistance of 15 V % (Nanbo, Shengzhen, China). Before
spin coating, the ITO substrates were cleaned using acetone,
detergent, deionized water, and 2-propanal, then treated with
oxygen plasma for 10 min. The LED structure is illustrated in
Scheme 2. A 40 nm layer of polyvinylcarbazole (PVK) was spin
cast on top of the ITO substrate. The PVK layer functions as a
hole-injection layer. A 70 nm phosphorescent dye-doped
emitting layer was spin coated on top of the PVK layer. The
thickness of the PVK and emitting layers were measured using
a Tencor Alpha-step 500 surface profiler. Devices were made
with a thin layer of barium as the cathode, deposited using
Scheme 2 Schematic representation of the PLED structure.
Fig. 1 UV-Visible absorption spectra of films of the iridium complexes.
52
J. Mater. Chem., 2003, 13, 50–55

View Article Online
Table 1 PL efficiencies of pure Ir complexes and CNPPP, and blends
thereof, measured in the integrating sphere
Dopant
concentration/wt%
PL
efficiency (%)
Material
Ir(PPy)3–CNPPP
4
4
4
4
32.1
41.7
28.4
12.3
Ir(BuPPy)3–CNPPP
Ir(DecPPy)3–CNPPP
Ir(DecOPPy)3–CNPPP
CNPPP
Ir(PPy)3
Ir(BuPPy)3
0
—
—
26.3
1.1
1.0
Device performance
Polymer LEDs were fabricated from the Ir complex-doped
CNPPP films. A schematic representation of the device struc-
ture is shown in Scheme 2. Since the highest occupied
molecular orbital (HOMO) level of alkoxy-substituted PPPs
is 5.7 eV,15 a 40 nm polyvinylcarbazole film was placed between
the indium tin oxide and the emitting layer to facilitate hole
injection. A thin layer of Ba with a 200 nm Al capping layer
was used as the cathode. The thickness of the Ir complex–
polymer blend was 70–80 nm in all cases. Fig. 2(b) shows the
corresponding EL spectra obtained from the devices with
different Ir(BuPPy)3 dopant concentrations. In contrast to the
PL spectra [Fig. 2(a)], which show a significant contribution
from the CNPPP host at low dopant concentrations, no
difference can be seen between the EL emission profiles for
devices with dopant concentrations between 1 and 8 wt%. The
spectra of the devices contain only emissions from the Ir
complexes, even at dopant concentrations as low as 1 wt%. This
result clearly indicates that Forster transfer is not the only
mechanism responsible for EL performance. O’Brien et al.11
suggested that there is a significant contribution from carrier
trapping. Recently, Lane et al. investigated Pt complex-doped
polyfluorene and concluded that Forster transfer of singlet
excitons was weak and the dominant emission mechanism in
their doped polymer LEDs was charge trapping followed by
recombination on the Pt complex.22 McGhee et al. reported a
large difference between the PL and EL spectra of Eu complex–
CNPPP devices at low dopant concentrations.23 They attrib-
uted this behavior to the difference in recombination zone for
photo- and electrical excitation.
Fig. 2 (a) PL spectra of Ir(BuPPy)3-doped CNPPP films on quartz
substrates. (b) EL spectra of Ir(BuPPy)3-doped CNPPP devices.
groups to the phenyl ring has little influence on the absorption
properties of the corresponding iridium complexes.
Fig. 2(a) shows the normalized photoluminescence (PL)
spectra of Ir(BuPPy)3-doped CNPPP devices with dopant
concentrations of 0, 1, 2, 4, and 8 wt%. The photoluminescence
profile contains two peaks: one is centered at 430 nm, resulting
from the emission of the CNPPP host and the other, observed
at 515 nm, is due to Ir(BuPPy)3 triplet emission.20 The emission
at 430 nm decreases significantly as the dopant concentration is
increased. At dopant concentrations of 4 wt% or more, the PL
profiles are dominated by the 515 nm triplet emission. The
observed concentration dependence of the PL profile resembles
that observed for Ir(PPy)3-doped CNPPP films.12 The results
indicate that Forster transfer from CNPPP singlets takes place.
In order gain a greater understanding of the energy transfer in
such blend films, the absolute PL quantum efficiencies of thin
(100 nm) blend films were measured in the integrating sphere,
according to the method proposed by Greenham et al.17
Table 1 lists the PL efficiencies of blend films containing the
four Ir complexes in 4 wt% dopant concentrations (the point at
which the emission of the Ir complex becomes dominant over
that of CNPPP). For the purposes of comparison, the PL
efficiencies of undoped CNPPP film and pure Ir complexes
were also measured (Table 1). As expected, the PL efficiencies
of the Ir complexes are only ca. 1%, due to concentration
quenching. This value is consistent with the results reported for
Ir complexes with other ligands.21 As can be seen from Table 1,
the PL efficiency varies with alkyl chain length; Ir(BuPPy)3-
doped CNPPP shows the highest PL quantum efficiency among
the four blends, which is consistent with the electrolumines-
cence (EL) results (see below). There are several possible
explanations for the dependence of the PL properties on the
alkyl chain length, and further photophysics studies are
necessary to obtain a clearer picture.
Fig. 3. compares the EL spectra of devices prepared with
CNPPP doped with different Ir complexes at 4 wt% dopant
concentration. No emission peak at 420 nm corresponding to
Fig. 3 EL spectra of devices containing CNPPP doped with various
iridium complexes.
J. Mater. Chem., 2003, 13, 50–55
53

View Article Online
Table 2 Device performance of different Ir complexes blended with
CNPPP
external QE and EL efficiency were 4.4% and 10 cd A21
,
respectively, at 120 cd m22. The efficiencies remain at 4.2%
and 10 cd A21, respectively, at 2500 cd m22. A maximum
external QE of 5.1% and an EL efficiency of 12 cd A21
were observed at 800 cd cm22 and at a current density of
6.8 mA cm22. According to Greenham et al.,24 the internal
quantum efficiency can be estimated from the optical power
emitted in the forward direction. Assuming the refractive index
of the substituted polyphenylene as 1.6, as observed for most
conjugated polymers,25 the maximal internal quantum effi-
ciency is 22% for the Ir(BuPPy)3-doped CNPPP devices. Better
blending and more homogeneous distribution of the guest
molecules in the host polymer matrix may account for the
improved energy transfer and the suppression of triplet–triplet
annihilation.
Device performance
Luminance/
V/V I/mA cd m22
EL efficiency/
QE (%) cd A21
Guest
Ir(PPy)3
Ir(BuPPy)3
Ir(DecOPPy)3 28
Ir(DecPPy)3 26
30
30
4.7
3.8
7.5
2.7
1668
2501
1717
1022
2.27
4.22
1.46
1.39
5.4
10.0
3.5
3.3
the host is observed, indicating that efficient energy transfer
from CNPPP to the iridium complexes occurs at a dopant
concentration of 4 wt% for all these devices. The Ir(BuPPy)3
device emits green light centered at 515 nm, with exactly the
same peak position and emission spectrum as for the Ir(PPy)3
device. It is interesting to note that the attaching decyl and
decyloxy groups at the p-position of the phenyl ring of
2-phenylpyridine shifts the emission peak significantly into the
blue region. Both the Ir(DecOPPy)3 and Ir(DecPPy)3 devices
emit blue–green light centered at 495 nm.
Overall, these are very encouraging results. Although the
external quantum efficiency for phosphorescent dye-doped
PLEDs is still much lower than the corresponding OLEDs,3,26
significant improvement in QE reduction rate with increasing
current density breaks the barrier to phosphorescent PLEDs
being used for passive displays operated at low duty cycles.
These results also demonstrate that efficient electrophosphor-
escence is not specific to small molecule OLEDs, it can also be
achieved in devices made with a polymer host. At this stage of
our experiments we don’t know why the longer 2-phenylpyr-
idine chain substitution in Ir(DecOPPy)3 and Ir(DecPPy)3)
leads to lower QEs than both Ir(PPy)3 and Ir(BuPPy)3.
However, this fact clearly indicates that there are many factors
which affect the energy transfer process in such systems. An
overlap of the emission spectrum of the host and the absorption
spectrum of the guest is considered to be an important
condition to obtain high efficiency in phosphorescent dye-
doped LEDs. In this study, all of these iridium complexes have
the same overlap between the emission spectrum of the CNPPP
host and their absorption spectra, similar to that for Ir(PPy)3.
Nevertheless, the device performance is very different in each
case. These results suggest that the energy transfer from the
blue light-emitting host to the phosphorescent dye is a
complicated process. Overlap of the emission of the host and
the absorption of the guest (Fig. 1) is a necessary condition, but
not a sufficient one. Other processes need to be considered. One
such is the matching of the LUMO and HOMO levels between
the host and guest in the singlet and triplet states. Another is
phase segregation between the crystalline guest molecules and
the amorphous host polymer. We have designed a series of
experiments to identify the underlying mechanism and these are
currently in progress.
Table 2 shows the device performance of the three new
complexes along with Ir(PPy)3 in 4 wt% dopant concentration.
Attaching a butyl group at the p-position of the phenyl ring
of 2-phenylpyridine improves the device performance. The
Ir(BuPPy)3-doped CNPPP device shows a generally higher
quantum efficiency (QE) than the Ir(PPy)3-doped device at
each dopant concentration. For the 4 wt% Ir(BuPPy)3 device,
an external quantum efficiency of 4.2% (PL/EL) and a lumi-
nous efficiency of 10 cd A21 were obtained at 2500 cd m22
,
while for the Ir(PPy)3-doped device fabricated under the
same experimental conditions, the values are only 2.3% and
5.4 cd A21, respectively, at 1770 cd m22 (Table 2). Devices
prepared using Ir(DecOPPy)3 and Ir(DecPPy)3 show much
lower efficiencies than the Ir(PPy)3 device, as can be seen from
Table 2. It is important to note that the reduction rate of the
QE under higher bias voltages (and currents) for the devices
with different substitutions of the phenyl ring in 2-phenylpyr-
idine differ remarkably. It is well known that the external QE in
Ir(PPy)3–TAZ devices3 reaches its maximum (15%) at very low
current densities (v1022 mA cm22) and decays very quickly
with increasing current density due to triplet–triplet annihila-
tion.3 Fig. 4 compares the luminous efficiencies of devices
prepared from Ir(PPy)3-, Ir(BuPPy)3-, and Ir(DecOPPy)3-
doped CNPPP at a dopant concentration of 2 wt%. The
Ir(BuPPy)3 device again shows the best performance at high
current densities. For one of the best Ir(BuPPy)3-doped
CNPPP devices with a dopant concentration of 4 wt%, the
Conclusions
We have synthesized and characterized three new iridium
complexes with different substituted PPy ligands. The EL
performance of these three complexes was compared with
that of Ir(PPy)3 in the same device configuration and using
the same fabrication conditions. We found that device per-
formance is quite sensitive to the substitution on the 2-phenyl-
pyridine ligand. Highly efficient PLEDs were obtained with
Ir(BuPPy)3-doped CNPPP. An external QE of 4.4% PL/
EL and a luminous efficiency of 10 cd A21 were obtained
at 120 cd m22. These values remain at 4.2% and 10 cd A21
,
respectively, at 2500 cd m22. We attribute these improvements
over other similar devices to improved interaction between the
guest and host, and to better and more complete energy
transfer from the host singlet to the guest triplet state. These
results demonstrate that efficient electrophosphorescence is not
only a feature of small molecule OLEDs, it can also be achieved
in devices made with a polymer host. This approach allows
high efficiency electrophosphorescent devices fabricated via
Fig. 4 Luminous efficiency vs. bias for devices prepared with blend
films with 2 wt% dopant concentration.
54
J. Mater. Chem., 2003, 13, 50–55

View Article Online
10 C. L. Lee, K. B. Lee and J. J. Kim, Appl. Phys. Lett., 2000, 77(15),
2280.
11 D. E. O’Brien, C. Giebler, R. B. Flecher, J. Cadlby, L. C. Palilis,
D. G. Lidzey, P. A. Lane, D. D. C. Bradley and N. Blan, Synth.
Met., 2001, 116, 379.
simpler and more cost-efficient processes, such as ink-jet print-
ing and screen printing.
Acknowledgements
12 W. Zhu, Y. Mo, M. Yuan, W. Yang and Y. Cao, Appl. Phys. Lett.,
2002, 80, 2045.
This work is supported by the National Natural Science
Foundation of China (Project No. 29992530-6) and the
Postdoctoral Science Foundation of China (Project No.
2001-5).
13 H. Spreitzer, W. Krreuder, H. Becker, E. Kluge, H. Schoo and
R. Demandt (PCT), WO Pat. Appl., 98/27136, 1996.
14 G. J. Solan and W. R. Vaughan, J. Org. Chem., 1957, 22, 750.
15 G. W. Gray, M. Hird, D. Lacey and K. J. Toyne, J. Chem. Soc.,
Perkin Trans. 2, 1989, 2041.
16 Y. Yang, Q. Pei and A. J. Heeger, J. Appl. Phys., 1996, 79, 934.
17 N. C. Greenham, I. D. W. Samuel, G. R. Hayes, R. T. Phillips,
Y. A. R. R. Kessener, S. C. Moratti, A. B. Holmes and
R. H. Friend, Chem. Phys. Lett., 1995, 241, 89.
18 E. Negishi, F. T. Luo, R. Frisbee and H. Matsushita, Heterocycles,
1982, 18, 117.
19 Y. Zhang, C. D. Baer, C. C. Neto, P. O’Brien and D. A. Sweigart,
Inorg. Chem., 1991, 30, 1685.
20 M. G. Colombo, T. C. Brunold, T. Riedener and H. U. Gudel,
Inorg. Chem., 1994, 33, 545.
21 X. Gong, M. R. Robinson, J. C. Ostrowski, D. Moses, G. C. Bazan
and A. J. Heeger, Adv. Mater., 2002, 14, 581.
22 P. A. Lane, L. C. Palilis, D. F. O’Brien, C. Giebler, A. J. Cadby,
D. G. Lidzey, A. J. Cambell, W. Blau and D. D. C. Bradley, Phys.
Rev. B, 2002, 63, 2 235 206.
23 M. D. McGhee, T. T. Bergstedt, C. Zhang, A. P. Saab,
M. B. O’Regan, G. C. Bazan, V. I. Srdanov and A. J. Heeger,
Adv. Mater., 1999, 11, 1349.
24 N. C. Greenham, R. H. Friend and D. D. C. Bradley, Adv. Mater.,
1994, 6, 491.
References
1
2
3
4
5
6
D. F. O’Brien, Y. You, A. Shoustikov, S. Sibley, M. E. Thompson
and S. R. Forrest, Nature., 1998, 395, 151.
D. E. O’Brien, M. A. Baldo, M. E. Thompson and S. R. Forrest,
Appl. Phys. Lett., 1999, 74, 442.
C. Adachi, M. A. Baldo, S. R. Forrest and M. E. Thompson, Appl.
Phys. Lett., 2000, 77(6), 904.
M. A. Baldo, S. Lamansky, P. E. Burrows, M. E. Thompson and
S. R. Forrest, Appl. Phys. Lett., 1999, 75(1), 4.
Ot. Watarabe, K. Nakamura, S. Kawami and Y. Fukuda, Synth.
Met., 2001, 122, 203.
C. Adachi, M. A. Baldo, S. R. Forrest, S. Lamansky,
M. E. Thompson and R. C. Kwong, Appl. Phys. Lett., 2001,
78(11), 1622.
7
8
M. E. Thompson, S. Lamansky, P. Djurovich, D. Murphy and
F. A. Razzaq, SID Proc., 2000, 337.
S. Lamansky, P. Djurovich, D. Murphy, F. A. Razzaq, H. E. Lee,
C. Adachi, P. Burrough, S. R. Forrest and M. E. Thompson,
J. Am. Chem. Soc., 2001, 123, 4304.
25 I. D. Parker and Q. Pei, Appl. Phys. Lett., 1994, 65, 1272.
26 M. A. Baldo, M. E. Thompson and S. R. Forrest, Nature, 2000,
403, 750.
9
Y. Ma, H. Y. Zhang, J. C. Shen and C. Che, Synth. Met., 1998, 94,
245.
J. Mater. Chem., 2003, 13, 50–55
55