High triplet energy polymer as host for electrophosphorescence with high efficiency

Article abstract of DOI:10.1021/ja060936t

We report the conjugated polymer P(tBu-CBP) as a host with high triplet energy (E_T 2.53 eV) and suitable HOMO (5.3 eV) and LUMO (2.04 eV) energy levels. Upon doping with green and red emission Ir-complexes, it gives devices with high luminous a

Full text of DOI:10.1021/ja060936t

Published on Web 06/14/2006
High Triplet Energy Polymer as Host for
Electrophosphorescence with High Efficiency
Yen-Chun Chen, Guo-Sheng Huang,† Chung-Chin Hsiao, and Show-An Chen*
Contribution from the Chemical Engineering Department, National Tsing-Hua UniVersity,
Hsinchu, 30041, Taiwan, R.O.C.
Received February 8, 2006; Revised Manuscript Received April 3, 2006; E-mail: sachen@che.nthu.edu.tw
Abstract: We report the conjugated polymer P(tBu-CBP) as a host with high triplet energy (ET 2.53 eV)
and suitable HOMO (5.3 eV) and LUMO (2.04 eV) energy levels. Upon doping with green and red emission
Ir-complexes, it gives devices with high luminous and external quantum efficiencies for green emission
(23.7 cd/A, 6.57%) and for red emission (5.1 cd/A, 4.23%), respectively, and low turn-on voltage (3 V). For
both devices, the efficiencies are higher than those of the corresponding devices with the same backbone
P(3,6-Cz) as a host by a factor of 4, even though the latter has an ET (2.6 eV) slightly higher than that of
the former. The results reflect that, in phosphorescent devices, the difference in ET between the host and
guest is not the only factor that determines the device efficiency, and the present side group modification
via the 9 position of carbazole also plays an important role, which allows a tuning of HOMO and LUMO
levels to provide more balance in electron and hole fluxes and provides prevention from formation of excimer.
Introduction
ability of high ET polymers (at least higher than the ET of the
green guest 2.4 eV for an Ir-complex).2b Recently a polycar-
Doping heavy metal complexes1 (e.g., Ir and Pt complexes)
into host materials in an organic light emitting diode for
obtaining high external quantum efficiency has attracted great
attention, owing to their efficient intersystem crossing from
singlet to triplet excited states followed by relaxing through
phosphorescence. Thus, an electrophosphorescent (EP) device,
allowing us to harvest both singlet and triplet excitons, is an
efficient approach over a purely fluorescent device where only
singlet exciton provides a radiative pathway. For an efficient
device, the host matrix should fulfill energy-level matching with
neighboring layers or electrodes for efficient charge injections
and with the guest for effective energy transfer of singlet
excitons and for efficient triplet confinement at the guest2
(requiring a higher triplet energy ET than that of the guest).
Studies on a high efficiency EP device with a conjugated
polymer as a host in red emission have been extensive,3 but in
green4a and blue emissions they are scarce due to the unavail-
bazole (P(3,6-Cz)) with an interconnection of carbazole units
in 3,6 positions was reported as a potential candidate for a host
from a quantum-chemical study4b-4c and found experimentally4a
to give a high efficiency green device (16 cd/A) when doped
with a green emitting Ir-complex. This polymer possesses high
ET (2.6 eV) and suitable highest occupied molecular orbital
(HOMO, 5.25 eV) and lowest unoccupied molecular orbital
(LUMO, 2.05 eV) levels allowing ease of charge injections
relative to the anode poly-(3,4-ethylenedioxythiophene):poly-
(styrenesulfonate) (denoted as PEDOT:PSS for simplicity) and
cathode (Ba/Al). However, it has drawbacks: imbalanced charge
fluxes4a and formation of a lower energy emission band, which
was claimed as an excimer emission.5 The formation of excimer
usually lowers the photoluminescence quantum efficiency
(PLQE) resulting in a lowering rate constant for the Fo¨rster
energy transfer.6
(3) (a) Chen, X.; Liao, J. L.; Liang, Y.; Ahmed, M. O.; Tseng, H. E.; Chen, S.
A. J. Am. Chem. Soc. 2003, 125, 636-637. (b) Jiang, C.; Yang, W.; Peng,
J.; Xiao, S.; Cao, Y. AdV. Mater. 2004, 16, 537-541. (c) Wu, F. I.; Shih,
P. I.; Tseng, Y. H.; Chen, G. Y.; Chien, C. H.; Shu, C. F.; Tung, Y. L.;
Chi, Y.; Jen, A. K. Y. J Phys. Chem. B 2005, 109, 14000-14005. (d)
Gong, X.; Ostrowski, J. C.; Bazan, G. C.; Moses, D.; Heeger, A. J.; Liu,
M. S.; Jen, A. K. AdV. Mater. 2003, 15, 45-49. (e) Sandee, A. J.; Williams,
C. K.; Evans, N.; Davies, J. E.; Boothby, C. E.; Kohler, A.; Friend, R. H.;
Holmes, A. B. J. Am. Chem. Soc. 2004, 126, 7041-7048. (f) Jiang, J.;
Jiang, C.; Yang, W.; Zhen, H.; Huang, F.; Cao, Y. Macromolecules 2005,
38, 4072-4080. (g) Zhen, H.; Luo, C.; Yang, W.; Song, W.; Du, B.; Jiang,
J.; Jiang, C.; Zhang, Y.; Cao, Y. Macromolecules 2006, 39, 1693-1700.
(4) (a) Dijken, A. v.; Bastiaansen, J. J. A. M.; Kiggen, N. M. M.; Langeveld,
B. M. W.; Rothe, C.; Monkman, A. P.; Bach, I.; Sto¨ssel, P.; Brunner, K.
J. Am. Chem. Soc. 2004, 126, 7718-7727. (b) Marsal, P.; Avilov, I.; da
Silva Filho, D. A.; Bre´das, J. L.; Beljonne, D. Chem. Phys. Lett. 2004,
392, 521-528. (c) Avilov, I.; Marsal, P.; Bre´das, J. L.; Beljonne, D. AdV.
Mater. 2004, 16, 1624-1629.
† Present address: National Laboratory of Applied Organic Chemistry,
Lanzhou University, Lanzhou 730000, China.
(1) (a) Baldo, M. A.; O’Brien, D. F.; Shoustikov, A.; Sibley, S.; Thompson,
M. E.; Forrest, S. R. Nature 1998, 395, 151-154. (b) Baldo, M. A.;
Thompson, M. E.; Forrest, S. R. Nature 2000, 403, 750-753. (c) Lamansky,
S.; Djurovich, P. I.; Murphy, D.; Abdel-Razzaq, F.; Kwong, R.; Tsyba, I.;
Bortz, M.; Mui, B.; Bau, R.; Thompson, M. E. Inorg. Chem. 2001, 40,
1704-1711. (d) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq,
F.; Lee, H.-E.; Adachi, C.; Burrows, P. E.; Forrest, S. R.; Thompson, M.
E. J. Am. Chem. Soc. 2001, 123, 4303-4312.
(2) (a) Sudhakar, M.; Djurovich, P. I.; Hogen-Esch, T. E.; Thompson, M. E.
J. Am. Chem. Soc. 2003, 125, 7796-7797. (b) Chen F.-C.; He, G.; Yang,
Y. Appl. Phys. Lett. 2003, 82, 1006-1008. (c) Adachi, C.; Kwong, R. C.;
Djurovich, P.; Adamovich, V.; Baldo, M. A.; Thompson, M. E.; Forrest,
S. R. Appl. Phys. Lett. 2001, 79, 2082-2084. (d) Holmes, R. J.; Forrest,
S. R.; Tung, Y.-J.; Kwong, R. C.; Brown, J. J.; Garon, S.; Thompson, M.
E. Appl. Phys. Lett. 2003, 82, 2422-2424. (e) Tokito, S.; Iijima, T.; Suzuri,
Y.; Kita, H.; Tsuzuki, T.; Sato, F. Appl. Phys. Lett. 2003, 83, 569-571. (f)
Holmes, R. J.; D’Andrade, B. W.; Forrest, S. R.; Ren, X.; Li, J.; Thompson,
M. E. Appl. Phys. Lett. 2003, 83, 3818-3820.
(5) (a) Siove, A.; Ade`s, D. Polymer 2004, 45, 4045-4049. (b) Zhang, Z. B.;
Fujiki, M.; Tang, H. Z.; Motonaga, M.; Torimitsu, K. Macromolecules 2002,
35, 1988-1990.
(6) Fo¨rster, T. Discuss. Faraday Soc. 1957, 27, 7-17.
9
10.1021/ja060936t CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 8549-8558
8549

A R T I C L E S
Chen et al.
Scheme 1. Synthetic Route for Intermediate 5
Scheme 2. Synthetic Route for Intermediates 7 and 8
Scheme 3. Synthetic Route for Monomer 9 and Compound 10
Here, we report the conjugated polymer P(tBu-CBP) (Scheme
4) for use as a host which possesses high ET (2.53 eV) and
suitable HOMO (5.3 eV) and LUMO (2.04 eV) energy levels.
Upon doping with green and red emission Ir-complexes, it gives
devices with high luminous and external quantum efficiencies
for green emission (ηLmax ) 23.7 cd/A, Qext ) 6.57%) and for
red emission (ηLmax ) 5.1 cd/A, Qext ) 4.23%), respectively,
and low turn-on voltage (3 V). For both devices, the efficiencies
are higher than those of the corresponding devices with P(3,6-
Cz) as the host by a factor of 4, even though the latter has an
ET (2.6 eV) slightly higher than that of the former. Evidently,
the device efficiency of the phosphorescent emission is not
solely dependent on the difference in ET between the host and
guest, and the side group in P(tBu-CBP) also plays an important
role. Incorporations of the comonomers with higher ET values,
dialkoxy substituted phenylene (compound 12) and dialkyl
substituted fluorene (compound 11), on the main chain to give
the alternating copolymers P(tBu-CBPP) and P(tBu-CBPF)
(Scheme 5), do not promote, in fact decrease, the levels of ET
and therefore lower the device efficiency due to back energy
transfer of ET.
Experimental Section
Synthesis of Monomers and Polymers. The synthetic routes for
the intermediates of monomers, monomers, and polymers used are
described in Schemes 1-5. The preparations of polymers P(3,6-Cz)
and P(tBu-CBP) were conducted using the Yamamoto coupling reaction.
P(tBu-CBPP) and P(tBu-CBPF) were prepared by the Suzuki coupling
reaction. The detailed synthetic procedures for the monomers and
polymers are described in the Supporting Information (SI).
Instrumentation. Ultraviolet-visible (UV-vis), photoluminescence
(PL), photoexcitation (PLE), electroluminescence (EL), phosphores-
cence, ultraviolet photoemission spectra (UPS), scanning probe mi-
croscopy, gel permeation chromatography (GPC), scanning probe
9
8550 J. AM. CHEM. SOC. VOL. 128, NO. 26, 2006

High Triplet Energy Polymer as Host
A R T I C L E S
Scheme 4. Synthetic Route for the Polymers P(tBu-CBP) and P(3,6-Cz)
Scheme 5. Synthetic Route for the Copolymers P(tBu-CBPF) and P(tBu-CBPP)
9
J. AM. CHEM. SOC. VOL. 128, NO. 26, 2006 8551

A R T I C L E S
Chen et al.
Figure 1. UV-vis (0) and prompt fluorescence (PF) (2) at room temperature, and delay luminescent (DL) with delay time of 5 ms after photoexcitation
([) spectra at 4 K of P(tBu-CBP), P(3,6-Cz), P(tBu-CBPP), and P(tBu-CBPF) thin solid films.
Table 1. Triplet Energy, HOMO and LUMO Levels, PLQE, and
microscopy (SPM), film thickness monitor and power supplies and
Hole Injection Barrier of the Polymers
luminance meter for measurements of device performance are also
hole
described in detail in the SI.
Device Fabrication and Characterization. An indium-tin oxide
injection
barriera
ET
HOMO
(eV)
LUMO
(eV)
PLQE
(%)
(ITO) glass was exposed on oxygen plasma at a power of 50 W and a
pressure of 200 mTorr for 5 min. A thin layer (15 nm) of poly(styrene
sulfonic acid)-doped poly(ethylenedioxythiophene) (Baytron P CH 8000
from Bayer, its conductivity is 10-5 S/cm) was spin-coated on the
treated ITO as a hole injection layer. Polymer solutions of P(3,6-Cz)
and P(tBu-CBP) in chlorobenzene (20 mg/mL) were filtered through a
5 µm filter and then spin-coated on top of the PEDOT:PSS layer. The
1,3,5-tris(2-henylbenzimidazolyl)benzene (TPBI) layer (30 nm), which
was used as a hole/exciton blocking layer,7 was grown by thermal
evaporation in a vacuum of 2 × 10-6 Torr. Finally, a thin layer of CsF
(about 2 nm) covered with a layer of aluminum for a bipolar device
was deposited in a vacuum thermal evaporator through a shadow mask
at a vacuum of 2 × 10-6 Torr. The active area of the device was about
10 mm2. The bipolar device structure was ITO/PEDOT:PSS (15 nm)/
Polymer or doped polymer (80 nm)/TPBI (30 nm)/CsF (2 nm)/Al. The
fabrication of a single carrier device was similar to that of the bipolar
device, but the device structures were different. The structure of the
hole dominating device was ITO/PEDOT:PSS (15 nm)/Polymer (120
nm)/Au and that of the electron dominating device was ITO/Ca (50
nm)/polymer (120 nm)/Ca (3 nm)/Al.
polymer
(eV)
(eV)
P(tBu-CBP)
P(tBu-CBPP)
P(tBu-CBPF)
P(3,6-Cz)
2.53
2.3
2.28
2.6
5.3
5.4
5.3
5.0
2.0
2.2
2.2
1.8
20
64
81
3
0.3
0.4
0.3
0.0
a The difference in HOMO levels between the polymer and PEDOT:
PSS (5.0 eV), which are measured by ultraviolet photoelectron spectroscopy.
of π-electrons along the polymer backbone. For the absorption
spectra of the alternating copolymers, P(tBu-CBPP) and P(tBu-
CBPF), the two higher energy peaks remain unchanged but the
lower energy peaks at 337 and 344 nm show red shifts by 3
and 9 nm, respectively, as compared to that of P(tBu-CBP).
For P(3,6-Cz), it shows a totally different shape of absorption
spectrum as compared to the CBP-based polymers; its absorption
peaks at 255 and 320 nm are the same as those with difference
alkyl substituents4a,5b,8 and similar to the characteristic absorption
peaks of carbazole dimer rather than those of a carbazole
monomer.8a,c-d
Results and Discussion
The emission peaks (PF in Figure 1) are observed at 416 nm
with a shoulder at 400 nm for P(tBu-CBP) and at 430 nm with
a shoulder at 400 nm and a low-energy band centered at about
525 nm for P(3,6-Cz). P(tBu-CBPP) exhibits a peak maximum
at 403 nm, and P(tBu-CBPF) exhibits maxima at 403 and 423
nm with a shoulder at 450 nm. The corresponding PLQE of
those polymers are shown in Table 1 and that of P(tBu-CBP)
(20%) is higher than that of P(3,6-Cz) (3%) due to the absence
of excimer emission5 in the former as evidenced by the lack of
A. Energy Levels Determinations (Singlet, Triplet, HOMO,
and LUMO). The chemical structures for the CBP-based
polymers, P(tBu-CBP), P(tBu-CBPP) and P(tBu-CBPF), and
P(3,6-Cz) are shown in Schemes 4 and 5, and their optical
spectra (UV-vis, prompt fluorescence (PF), and delay lumi-
nescence (DL) spectra) are shown in Figure 1. The homopoly-
mer P(tBu-CBP) exhibits absorption peaks at 238, 300, and 334
nm, which are similar to those of its repeat unit resemblance
(tBu-CBP) and monomer (tBu-CBP-Br) (Figure S2) (whose
chemical structures are referred to compounds 10 and 9 in
Scheme 3, respectively), but its absorption onset shows a red
shift by 16 nm as compared to the latter’s due to a delocalization
(8) (a) Romero, D. B.; Schaer, M.; Leclerc, M.; Ades, D.; Siove, A.; Zuppiroli,
L. Synth. Met. 1996, 80, 271-277. (b) Huang, J.; Niu, Y.; Yang, W.; Mo,
Y.; Yuan, M.; Cao, Y. Macromolecules 2002, 35, 6080-6082. (c) Paliulis,
O.; Ostrauskaite, J.; Gaidelis, V.; Jankauskas, V.; Strohriegl, P. Macromol.
Chem. Phys. 2003, 204, 1706-1712. (d) Ostrauskaite, J.; Strohriegl, P.
Macromol. Chem. Phys. 2003, 204, 1713-1718. (e) Iraqi, A.; Wataru, I.
J. Polym. Sci., Part A: Polym. Chem. 2004, 42, 6041-6051.
(7) Adamovich, V. I.; Cordero, S. R.; Djurovich, P. I.; Tamayo, A.; Thompson,
M. E.; D’Andrade, B. W.; Forrest, S. R. Org. Electron. 2003, 4, 77-87.
9
8552 J. AM. CHEM. SOC. VOL. 128, NO. 26, 2006

High Triplet Energy Polymer as Host
A R T I C L E S
lower-energy emission band (∼525 nm) as to be discussed later
in this section. Both the alternating copolymers, P(tBu-CBPP)
and P(tBu-CBPF), have a significant enhancement in PLQE of
64 and 81%, respectively. The delay luminescence (DL) (or
phosphorescence) at 4 K as shown in Figure 1 is utilized to
the neighboring carbazole unit (as illustrated in Figure 2a). This
explains why the ET remains constant from a carbazole dimer
2.75 eV to a carbazole trimer 2.74 eV. For the mixed carbazole/
fluorene compounds, for example, carbazole-fluorene-carbazole
linked via the 3,6 position of carbazole and the para position
of fluorene (as illustrated in Figure 2b), the conjugation length
extends to p-quaterphenyl by covering the inserted fluorene unit,
and thus ET decreases to 2.38 eV.
determine the ET of the polymers, in which the first vibronic
ν)0
transition (T1
f S0ν)0) of the phosphorescence is assigned
as ET.4a The ET of P(tBu-CBP) (2.53 eV, 490 nm) is only slightly
lower than that of P(3,6-Cz) (2.6 eV, 478 nm) due to the side
group modification. However, upon incorporation of the comono-
mers, dialkyl substituted fluorene (compound 11) and dialkoxy
substituted phenylene (compound 12), to yield the copolymers,
P(tBu-CBPF) and P(tBu-CBPP), the ET’s drop significantly to
2.28 eV (545 nm) and 2.3 eV (540 nm), respectively (Figure 1
and Table 1).
The same behavior of variation of ET with chemical structure
was held well in the poly(3,6-carbazole) derivatives.4a,10b Here,
the triplet exciton of P(tBu-CBP) can also be deduced from the
above discussions and expected to be predominantly delocalized
over the biphenyl unit across neighboring carbazole units as
illustrated in Figure 2c. For P(tBu-CBPP) and P(tBu-CBPF),
the triplet exciton can be expected to delocalize over the
p-terphenyl and p-quaterphenyl units as illustrated in Figure 2c
and b, respectively. Hence, an increased π-electron delocaliza-
tion in these copolymers, P(tBu-CBPP) and P(tBu-CBPF),
occurs as reflected in decreases of absorption and emission
(especially for phosphorescence) energies as compared to P(tBu-
CBP). Consequently, to maintain high ET of conjugated polymer,
the conjugation length of polymer should be carefully controlled.
Dijken et al.4a found that, for maintaining high ET (2.56-2.6
eV), incorporation of a comonomer (e.g., oxadiazole or fluorene)
into the carbazole main chain should only limit meta position
coupling. Here, the high ET of the conjugated polymer is
favorably obtained by the direct modification of a side group.
The HOMO levels of the present CBP-based polymers are
close to each other; in other words, the copolymerization hardly
affects the HOMO levels (Table 1). For P(tBu-CBP), the side
group acts as an inductive acceptor, leading to substantial
lowering of the HOMO (as well as LUMO) with respect to
P(3,6-Cz) by about 0.3 eV (0.2 eV). The reasons are as follows.
First, for small molecules, an aryl substituent at the 9 position
of carbazole was found to lower the HOMO level relative to
the alkyl substituent, but only to a slight extent by about 0.1
eV.10a Second, the additional carbazole group attached on the
side group of P(tBu-CBP) can also enhance the inductive effect
as in the case of tris(4-(9H-carbazol-9-yl)phenyl)amine (TCB),
in which the incorporation of Cz on the three phenyl ring leads
to a lowering in the HOMO and LUMO levels of triphenylamine
(TPA) by 0.14 and 0.6 eV, respectively.4b This ensures the
HOMO levels of CBP-based polymers are lower than or
comparable to those of most reported Ir-complexes,1c-d,2b,12
leading to the idea that Ir-complexes can act as a trap for hole
carriers to promote efficient charge trapping.
The HOMO levels of the polymers are determined from the
cyclic voltammetry measurements (experimental detail and
results are given in SI section 3) and listed in Table 1; no
reduction waves were observed down to about -1.2 V, and all
polymers showed irreversible oxidative behavior. The energy
level of LUMO was deduced from the onset of the UV-vis
spectrum and HOMO level (Table 1). The HOMO level of
P(tBu-CBP) 5.3 eV is lower than that of P(3,6-Cz) (5.0 eV).
However, the HOMO levels of P(tBu-CBPP) (5.4 eV) and
P(tBu-CBPF) (5.3 eV) are close to that of P(tBu-CBP). The
resulting hole injection barriers for these CBP- and Cz-based
polymers at the interfaces with PEDOT:PSS are significantly
improved relative to those of the mostly investigated polymer
hosts PFOs/PEDOT:PSS (0.6-0.8 eV)2b,3a-c and PVKs/PEDOT:
PSS (∼1 eV)9 and thus can lead to low operating voltage and
high device power efficiency.
B. Variations in Energy Levels of Singlet/Triplet and
HOMO/LUMO with Molecular Structures. The conjugation
lengths of carbazole derivatives4a,10 coupling via its 3,6 positions
have been investigated based on the studies for oligophen-
ylenes,11 in which the π-electron delocalizations are extended
along the longest molecular axis for para-linkage with ET
decreasing with the number of phenyls and interrupted at the
meta-linkage. For example, the ET of biphenyl (2.84 eV) is
higher than that of p-terphenyl (2.55 eV), and the ET of
m-terphenyl (2.81 eV) is close to that of biphenyl because its
conjugation is interrupted on the meta position for which the
triplet state is localized at every composing biphenyl structure.11
In this respect, Brunner et al.10a found that the ET of a carbazole
dimer, linked via the 3,6 position, reduces from a monomer
3.05 eV to a dimer 2.75 eV and is close to that of biphenyl
2.84 eV. This indicates not only that in a carbazole dimer the
triplet exciton is more delocalized in the carbazole dimer than
in the carbazole monomer but also that the triplet exciton is
predominantly delocalized over the biphenyl structure across
C. Absence of Excimer Emission in P(tBu-CBP) Contrary
to P(3,6-Cz). Figure 3 shows UV-vis, PL, and PLE spectra of
P(3,6-Cz) and P(tBu-CBP) in solutions and as thin solid films.
For P(3,6-Cz) thin solid film, the low-energy emission band
(LEEB) centered at about 525 nm (Figures 1b and 3c) can be
contributed from either an excimer or aggregates. An excimer
is an emitting species that forms from a dimer of two molecules
(9) (a) Yang, X.; Neher, D.; Hertel, D.; Daubler, T. K. AdV. Mater. 2004, 16,
161-166. (b) Kawamura, Y.; Yanagida, S.; Forrest, S. R. J. Appl. Phys.
2002, 92, 87-93. (c) Lamansky, S.; Djurovich, P. I.; Abdel-Razzaq, F.;
Garon, S.; Murphy, D. L.; Thompson, M. E. J. Appl. Phys. 2002, 92, 1570-
1575.
(12) (a) Lo, S. C.; Namdas, E. B.; Burn, P. L.; Samuel, I. D. W. Macromolecules,
2003, 36, 9721-9730. (b) Thomas, K. R. J.; Velusamy, M.; Lin, J. T.;
Chien, C. H.; Tao, Y. T.; Wen, Y. S.; Hu, Y. H.; Chou, P. T. Inorg. Chem.
2005, 44, 5677-5685. (c) Sajoto, T.; Djurovich, P. I.; Tamayo, A.;
Yousufuddin, M.; Bau, R.; Thompsom, M. E. Inor. Chem. 2005, 44, 7992-
8003. (d) Huang, Y. T.; Chuang, T. H.; Shu, Y. L.; Kuo, Y. C.; Wu, P. L.;
Yang, C. H.; Sun, I. W. Organometallics 2005, 24, 6230-6238. (e) Holmes.
R. J.; Forrest, S. R.; Sajoto, T.; Tamayo, A.; Djurovich, P. I.; Thompsom,
M. E.; Brooks, J.; Tung, Y. J.; D’Andrade, B. W.; Weaver, M. S. et al.
Appl. Phys. Lett. 2005, 87, 243507-243507-3.
(10) (a) Brunner, K.; Dijken, A. v.; Bo¨rner, H.; Bastiaansen, J. J. A. M.; Kiggen,
N. M. M.; Langeveld B. M. W. J. Am. Chem. Soc. 2004, 126, 6035-6042.
(b) Rothe, C.; Brunner, K.; Bach, I.; Heun, S.; Monkman, A. P. J. Chem.
Phys. 2005, 122, 84706-1∼84706-6.
(11) (a) Brike, J. B. Photophysics of Aromatic Compounds; John Wiley &
Sons: New York, 1970. (b) Higuchi, J.; Hayashi, K.; Yagi, M.; Kondo, H.
J. Phys. Chem. A 2002, 106, 8609-8618. (c) Higuchi, J.; Hayashi, K.;
Seki, K.; Yagi, M.; Ishizu, K.; Kohno, M.; Ibuki, E.; Tajima, K. J. Phys.
Chem. A 2001, 105, 6084-6091.
9
J. AM. CHEM. SOC. VOL. 128, NO. 26, 2006 8553

A R T I C L E S
Chen et al.
Figure 2. Chemical structures of conjugated units in the four polymers investigated in this study. The dashed line indicated the presence of oligo(para-
phenylene) in the polymer backbone.
of the same kind, one of which being in the excited state; it
exists only at the excited state but is dissociative in the ground
state. In contrast, an aggregate involves intermolecular interac-
tion between two or more lumophores in the ground state by
extending the delocalization of π-electrons over these conjugated
segments. According to the method of characterization for
aggregates in conjugated polymers used in our previous publica-
tions,13 the LEEBs of P(3,6-Cz) are contributed from excimer
rather than aggregates as to be revealed below.
by 3 nm (Figure 3b). Thus, there is no excimer or aggregate
emitting species in P(tBu-CBP) in solutions or as solid film.
For P(3,6-Cz), the PL spectra of its solutions do not vary with
concentration in the range 5 × 10-4 to 5 mg/mL, but the PL
spectrum of its spin-coated film shows a red shift in λmax by 6
nm and an additional LEEB centered at about 525 nm (Figure
3c). This indicates that new emission species (excimer or
aggregates) form in P(3,6-Cz) thin solid film.
If the new emission species is aggregates, a new “red shift”
and “structureless” aggregate emission should appear in the PL
spectrum relative to the emission from the solutions at an
excitation wavelength around the absorption edge, and a PLE
spectrum monitored at the LEEB (500-650 nm here) should
show a relatively elevated luminescence intensity in the longer
wavelength region (>430 nm) over that monitored at 430 nm
For P(tBu-CBP), the PL spectra of its solutions do not vary
with concentration in the range 5 × 10-4 to 5 mg/mL, and the
PL spectrum of its spin-coated film is similar to that at the
solution state except for a red shift in emission maximum (λmax
)
(13) (a) Peng, K. Y.; Chen S. A.; Fann, W. S. J. Am. Chem. Soc. 2001, 123,
11388-11397. (b) Peng, K. Y.; Chen, S. A.; Fann, W. S.; Chen, S. H.; Su,
A. C. J. Phys. Chem. B 2005, 109, 9368-9373.
9
8554 J. AM. CHEM. SOC. VOL. 128, NO. 26, 2006

High Triplet Energy Polymer as Host
A R T I C L E S
Figure 3. (a) UV-vis spectra of P(3,6-Cz) and P(tBu-CBP) solutions (5 × 10-3 mg/mL) in chlorobenzene and as thin solid film, (b) PL spectra of P(tBu-
CBP) solutions at various concentrations (5 × 10-4 to 5 mg/mL) and as thin solid film, (c) PL spectra of P(3,6-Cz) solutions at various concentrations (5
× 10-4 to 5 mg/mL) and as thin solid film, (d) PLE spectra of P(3,6-Cz) thin solid film monitored at various emission wavelengths (430-650 nm) and PL
spectra of P(3,6-Cz) thin solid film excited at various excitation wavelengths (320-380 nm). The arrows in the PL spectra (b and c) indicate the excitation
wavelength.
Figure 4. Electroluminescence spectra of P(3,6-Cz) (1) and P(tBu-CBP) doped with Ir-complexes: 0 wt % dopant (9), 5 wt % Ir-R (b), and 8 wt % Ir-G
(2). The bipolar device structure is ITO/PEDOT:PSS (15 nm)/polymer or polymer:dopant (80 nm)/TPBI (30 nm)/CsF (2 nm)/Al. The band diagram is
included on the right.
because of the existence of lumophore associates, which absorb
light at longer wavelengths than the monomeric lumophores
do.13 However, no such aggregate emissions appear as also
confirmed by the identical PLE spectra in P(3,6-Cz) thin solid
film (Figure 3d). Thus the LEEBs are not from aggregates but
from an excimer. Consequently, we can conclude that the LEEB
in P(3,6-Cz) is an excimer emission and the side group of P(tBu-
CBP) can significantly suppress the formation of excimer.
D. Green and Red Electrophosphorescence from the
Device with P(tBu-CBP) or P(3,6-Cz) as Host. P(tBu-CBP)
has an ET (2.53 eV) higher than those of the green and red
guests: bis(2-phenylpyridinato-N,C2′)iridium(acetylacetonate)
(denoted as Ir-G, ET ) 2.41 eV)1c and bis(1-phenyl-isoquino-
linato-N,C2′)iridium(acetylacetonate) (Ir-R, ET ) 2.00 eV);14
it is therefore a potential host for both guests. Figure 4 shows
electroluminescence spectra of P(3,6-Cz) and P(tBu-CBP) doped
with Ir-complexes and the band diagram of materials used in
the bipolar device structure (ITO/PEDOT:PSS (15 nm)/polymer
or doped polymer (80 nm)/TPBI (30 nm)/CsF (2 nm)/Al). The
energy levels of Ir-G,2b Ir-R,14 TPBI,15a and CsF15b are taken
from the literatures. The work functions of ITO and PEDOT:
PSS were measured by ultraviolet photoelectron spectroscopy
(14) Li, C. L.; Su, Y. J.; Tao, Y. T.; Chou, P. T.; Chien, C. H.; Cheng, C. C.;
Liu, R. S. AdV. Funct. Mater. 2005, 15, 387-395.
9
J. AM. CHEM. SOC. VOL. 128, NO. 26, 2006 8555

A R T I C L E S
Chen et al.
Figure 5. Brightness (top), current density (middle), and efficiency (bottom) versus electric field for Ir-R doped devices (left) and for Ir-G doped devices
(right). The Ir-R doping levels (by weight) are 5% (2), 10% (b), and 15% (1), and the Ir-G doping levels (by weight) are 5% (2), 8% (b), and 15% (1).
Current density versus electric field for P(tBu-CBP) (9) was put together with those for Ir-doped devices for a comparison. The unit for electric field (105
V/cm) is so taken so that, at the normal thickness of emitting layer 100 nm, the applied voltage is 1 V. The bipolar device structure is: ITO/PEDOT:PSS
(15 nm)/P(tBu-CBP) or doped P(tBu-CBP) (80 nm)/TPBI (30 nm)/CsF (2 nm)/Al.
Table 2. Device Performances of Ir-G or Ir-R as a Guest and
with the dopants above 5 wt % and identical to the PL spectra
of the dopants, respectively.
P(tBu-CBP), P(3,6-Cz), P(tBu-CBPP), or P(tBu-CBPF) as a Hosta
b,c
turn-on
(V)
Bmaxb (cd/m2)
at cd/A
ηLmaxb (cd/A) max Qext
Table 2 lists the performances (brightness, turn-on voltage,
luminous and external quantum efficiencies) of the devices with
the Ir-complexes doped with polymers at various doping levels,
and Figure 5 shows the brightness, current density, and
efficiency versus electric field for Ir-R and Ir-G doped P(tBu-
CBP) devices. For P(tBu-CBP), the best device performances
for Ir-R (ηLmax ) 5.1 cd/A, corresponding Qext ) 4.23%; Bmax
) 5590 cd/m2) and Ir-G (ηLmax ) 23.7 cd/A, corresponding
Qext ) 6.57%; Bmax ) 31 500 cd/m2) are observed at the optimal
doping levels 5 and 8 wt %, respectively, both exhibiting low
turn-on voltage (∼3 V) (Table 2 and Figure 5). The present
green emission device is as efficient as that with the reported
carbazole copolymer doped by an Ir-G with dendrimer-like
ligands16a (the most efficient green emission device with a
conjugated polymer, ηLmax ) 23 cd/A),4a even though the present
Ir-G seems to have a higher tendency toward phase separation,
as evidenced in Figure 6, than this Ir-G ligand, and might lead
to an inefficient use of guest molecules and a promoted triplet-
triplet annihilation.16 Figure 6 shows the SPM phase-contrast
images of both polymers, P(tBu-CBP) and P(3,6-Cz), doped
with the Ir-complexes and of their pristine polymer films. We
find that phase separation occurs in the Ir-complexes doped
polymers, as shown in Figure 6b, c, e, and f. But the extent of
host
guest
at V
(%)
P(tBu-CBP)
5 wt % Ir-R
10 wt % Ir-R
15 wt % Ir-R
5 wt % Ir-G
8 wt % Ir-G
10 wt % Ir-G
3.0
2.8
2.8
3.0
2.9
2.9
5590 (1.85)
6910 (1.64)
6450 (1.84)
26 560 (5.11) 21.0 (4.5)
31 500 (5.23) 23.7 (4.3)
28 030 (5.95) 19.3 (4.8)
5.1 (4)
4.8 (4)
4.2 (4.2)
4.23
3.98
3.48
5.82
6.57
5.35
P(3,6-Cz)
5 wt % Ir-R
8 wt % Ir-G
2.9
2.7
980 (0.28)
8160 (1.52)
1.2 (4.1)
5.4 (4.3)
1.0
1.5
P(tBu-CBPP)
P(tBu-CBPF)
5 wt % Ir-R
8 wt % Ir-G
6.0
6.8
1800 (1.3)
3400 (2.5)
1.4 (15.7)
2.7 (16.9)
1.17
0.75
5 wt % Ir-R
8 wt % Ir-G
5.4
5.0
6750 (2.0)
7800 (2.3)
2.0 (17)
2.6 (13.7)
1.67
0.72
a The device structure is ITO/PEDOT:PSS (15 nm)/Polymer:dopant (80 nm)/
TPBI (30 nm)/CsF (2 nm)/Al. (tBu-CBP) and P(3,6-Cz) were dissolved in
chlorobenzene (20 mg/mL) and P(tBu-CBPP) and P(tBu-CBPF) in tetrahydro-
furan (10 mg/mL). b The data for brightness (Bmax), luminous efficiency (ηLmax),
and external quantum efficiency (Qext) are the maximum values of the device.
The turn-on voltage was taken at 0.2 cd/m2. Qext was calculated from the total
c
photon number evaluated from luminance and emission spectra on the assumption
that the spatial emission pattern from the PLEDs was Lambertian.
(UPS). The emission spectrum of P(tBu-CBP) has a peak at
420 nm and no lower-energy emission (which appears in P(3,6-
Cz)s and was claimed from an electromer),4a indicating that the
bulky side group of tBu-CBP can suppress the stacking of
carbazole units in the main chain. Complete emissions from
Ir-G (519 nm) and Ir-R (618 nm) are observed in the devices
(16) (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, 975-979. (b)
Yang, X.; Neher, D.; Hertel, D.; Da¨ubler, T. K. AdV. Mater. 2004, 16,
161-166.
(15) (a) Anthopoulos, T. D.; Markham, J. P. J.; Namdas, E. B.; Samuel, I. D.
W.; Lo, S. C.; Burn, P. L. Appl. Phys. Lett. 2003, 82, 4842-4826. (b)
Michaelson, H. B. J. Appl. Phys. 1977, 48, 4729-4733.
9
8556 J. AM. CHEM. SOC. VOL. 128, NO. 26, 2006

High Triplet Energy Polymer as Host
A R T I C L E S
Figure 6. Phase-contrast image of the six polymer samples, (a) P(tBu-
CBP), (b) P(tBu-CBP) doped with 8 wt % Ir-G, (c) P(tBu-CBP) doped
with 5 wt % Ir-R, (d) P(3,6-Cz), (e) P(3,6-Cz) doped with 8 wt % Ir-G,
and (f) P(3,6-Cz) doped with 5 wt % Ir-R, as measured using a scanning
probe microscope (from Digital Instrument Nanoscope III).
phase separation for the case with P(tBu-CBP) is less than that
with P(3,6-Cz). Consequently, the doped devices show fast
decay in efficiency with the electric field as shown in Figures
5 and 7.
Figure 7. Brightness (b) and current density (0) versus electric field for
5 wt % Ir-R doped devices (left) and for 8 wt % Ir-G doped devices (right).
The inset is the efficiency versus electric field. The unit for electric field
(105 V/cm) is so taken so that, at the normal thickness of emitting layer
100 nm, the applied voltage is 1 V. The bipolar device structure is: ITO/
PEDOT:PSS (15 nm)/doped P(3,6-Cz) (80 nm)/TPBI (30 nm)/CsF (2 nm)/
Al.
For the polymer with the same backbone, P(3,6-Cz), at the
above optimal Ir-R and Ir-G doping levels for P(tBu-CBP), the
corresponding Bmax and ηLmax are both lower by a factor of 4
(Table 2 and Figures 6), even though it has an ET higher than
that of P(tBu-CBP) by 0.07 eV. Thus the higher efficiency in
doped P(tBu-CBP) devices is not only due to its high ET but
also its side group which may also play some role. As the two
copolymers P(tBu-CBPP) and P(tBu-CBPF) are used as hosts,
the devices with the dopants Ir-R and Ir-G at the doping levels
5 and 8 wt % give an efficiency lower by factors of about 3
(R) and 9 (G) relative to those with P(tBu-CBP) as the host
(Table 2), even though the HOMO and LUMO levels of the
three polymers are close (Table 1). Thus, such lowering in
efficiency must be due to an increased back transfer of triplet
energy from the dopant to the host.
E. Why P(tBu-CBP) Performs Better Than P(3,6-Cz)? To
understand why P(tBu-CBP) is a good host for an efficient
device, the charge transport properties of P(tBu-CBP) and P(3,6-
Cz) are investigated by single-carrier devices,17 and the results
are shown in Figure 8. P(tBu-CBP) shows balanced electron
and hole fluxes, while P(3,6-Cz) shows the current density of
the hole is higher than that of the electron by 2 orders of
magnitude at 4 × 105 V/cm. The results can be explained by
Figure 8. Current density versus electric field for single carrier devices
with P(3,6-Cz) (O for hole and b for electron) and P(tBu-CBP) (0 for
hole and 9 for electron) as the active polymer layer. The hole and electron
dominating device structures are ITO/PEDOT:PSS/Polymer/Au and ITO/
Ca/Polymer/CsF/Al, respectively.
the HOMO and LUMO energy levels of both polymers (Figure
4). For P(tBu-CBP), the injection barrier for the hole (0.3 eV)
is close to that for the electron (0.2 eV) (referred to as the anode
PEDOT:PSS and the cathode CsF/Al), whereas, for P(3,6-Cz),
(17) (a) Blom, P. W. M.; de Jong, M. J. M.; Vleggaar, J. J. M. Appl. Phys. Lett.
1996, 68, 3308-3310. (b) Yu, L. S.; Tseng, H. E.; Lu, H. H.; Chen, S. A.
Appl. Phys. Lett. 2002, 81, 2014-2016. (c) Yu, L. S.; Chen, S. A. AdV.
Mater. 2004, 16, 744-748. (d) Hsiao, C. C.; Chang, C. H.; Jen, T. H.;
Hung, M. C.; Chen, S. A. Appl. Phys. Lett. 2006, 88, 33512-33512-3.
9
J. AM. CHEM. SOC. VOL. 128, NO. 26, 2006 8557

A R T I C L E S
Chen et al.
the injection barrier for the hole (0 eV) is much smaller than
that for the electron (0.4 eV).
in ET between the host and guest is not the only factor that
determines the device efficiency, and the present side group
modification via the 9 position of carbazole also plays an
important role, which allows a tuning of HOMO and LUMO
levels for balancing electron and hole fluxes and provides a
depression of formation of excimer. Also, the tuning of HOMO/
LUMO levels allows a charge trapping of both electron and
hole in the dopants to occur. Thus, the present work provides
a new route for designing ambipolar transporting polymers with
energy level matching and high ET, simultaneously.
In addition, both Ir-G and Ir-R can act as traps for both
electron and hole carriers in P(tBu-CBP) as a host, but for P(3,6-
Cz), it only acts as an electron trap based on their corresponding
HOMO and LUMO levels (Figure 4). The other factor that leads
to the improvement could be due to the enhancement of PLQE,
which results in a more efficient Fo¨rster energy transfer from
the host to the guest.6,18
Conclusions and Significance
Acknowledgment. We thank the MOE for financial aid
through Project 91E-FA04-2-4A, the NSC for constant financial
support in the past, and Dr. Xiwen Chen for fruitful discussions.
In conclusion, we found that the high ET of conjugated
polymer P(tBu-CBP) exhibits balanced charge flux character-
istics. Upon doping with Ir-R and Ir-G, devices with high
efficiency and brightness for green and red emissions can be
obtained which are higher than the case with the polymer having
the same backbone and higher ET, P(3,6-Cz), as the host. The
results reflect that, in the phosphorescent device, the difference
Supporting Information Available: Instrumentation details,
detailed experimental procedures, and characterizations for all
the monomers and polymers. This material is available free of
charge via the Internet at http://pubs.acs.org.
(18) Shoustikov, A.; You, Y.; Burrows, P. E.; Thompson, M. E.; Forrest, S. R.
Synth. Met. 1997, 91, 217-221.
JA060936T
9
8558 J. AM. CHEM. SOC. VOL. 128, NO. 26, 2006