Synthesis and characterization of red light-emitting electrophosphorescent polymers with different triplet energy main chain

Article abstract of DOI:10.1016/j.polymer.2011.03.023

Red light-emitting electrophosphorescent polymers with poly(fluorene-alt- carbazole) (PFCz) and poly(3,6-carbazole) (PCz) as the main chain, and the quinoline-based iridium (Ir) complex as the side chain have been synthesized by Suzuki and modified Yamamoto polymerization, respectively. The triplet energy of the polymeric backbone is tuned from 2.1 eV of polyfluorene (PF) to 2.3 eV of PFCz and 2.6 eV of PCz. We find that, with the increasing triplet energy, the lifetime of the triplet excitons gradually increases, but the device efficiency becomes worse. The reason is that the alteration of the main chain structure leads to the increase of the highest occupied molecular orbital (HOMO) level of the polymeric host, which can facilitate the efficient hole injection and transportation. As a consequence of this enhancement of the hole current, charge balance in the emitting layer is destroyed, and correspondingly, the poorer device performance is achieved. Our results, we believe, indicate that besides the triplet energy, charge balance is a crucial determinant to develop high efficiency red light-emitting electrophosphorescent polymers.

Full text of DOI:10.1016/j.polymer.2011.03.023

Polymer 52 (2011) 2189e2197
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Synthesis and characterization of red light-emitting electrophosphorescent
polymers with different triplet energy main chain
,
b
,
,
a
a
a
,
a
Zhihua Ma a 1, Junqiao Ding a , Yanxiang Cheng , Zhiyuan Xie , Lixiang Wang , Xiabin Jing ,
*
*
Fosong Wang a
a State Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, PR China
b Graduate School of the Chinese Academy of Sciences, Beijing 100039, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Red light-emitting electrophosphorescent polymers with poly(fluorene-alt-carbazole) (PFCz) and poly
(3,6-carbazole) (PCz) as the main chain, and the quinoline-based iridium (Ir) complex as the side chain
have been synthesized by Suzuki and modified Yamamoto polymerization, respectively. The triplet
energy of the polymeric backbone is tuned from 2.1 eV of polyfluorene (PF) to 2.3 eV of PFCz and 2.6 eV of
PCz. We find that, with the increasing triplet energy, the lifetime of the triplet excitons gradually
increases, but the device efficiency becomes worse. The reason is that the alteration of the main chain
structure leads to the increase of the highest occupied molecular orbital (HOMO) level of the polymeric
host, which can facilitate the efficient hole injection and transportation. As a consequence of this
enhancement of the hole current, charge balance in the emitting layer is destroyed, and correspondingly,
the poorer device performance is achieved. Our results, we believe, indicate that besides the triplet
energy, charge balance is a crucial determinant to develop high efficiency red light-emitting electro-
phosphorescent polymers.
Received 19 January 2011
Received in revised form
6 March 2011
Accepted 15 March 2011
Available online 21 March 2011
Keywords:
Electrophosphorescent polymers
Light emitting diodes
Triplet energy
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
introduced as the side chain, the device showed a luminous effi-
ciency of 2.8 cd/A at 7.0 V and a luminance of 65 cd/m2 with the
Electrophosphorescent polymeric light-emitting devices
(PhPLEDs) have drawn great attention due to their ability to inte-
grate the low-cost solution processability of the polymers [1] and
the high efficiency of the electrophosphorescent transition metal
complexes [2]. Up to now, there are generally two strategies to
realize PhPLEDs. One is physical blending the phosphor with the
polymer host. However, these devices bear some problems, such as
potential aggregation and phase separation, which can result in
tripletetriplet annihilation, and thus lead to the fast decay of the
device efficiency at high current density [3e6].
peak emission at 610 nm [7]. By introducing isoquinoline-based Ir
complex as the side chain of poly(fluorene-alt-carbazole) (PFCz)
copolymers, Jiang et al. further improved the efficiency to 4.0 cd/A
[8]. Recently, we have realized an unexpected high efficiency based
on a simple polymeric chain owing to the charge balance. When
1 mol % of the quinoline-based Ir complex was incorporated into
the side chain of PF, a high efficiency of 5.0 cd/A was achieved for
the single layer device. Additionally, by inserting an alcohol-soluble
electron injection layer, the luminous efficiency was optimized to
8.3 cd/A [9].
Therefore, the other way, chemical attaching the phosphor to
the polymer host, has been developed. In early 2003, Chen and
coworkers reported electrophosphorescent polymers based on
poly0fluorene (PF) backbone with bis[2-(20-benzothienyl) pridinato-
N,C3 ](acetylacetonato) Ir complex [(btp)2Ir(acac)] as the pendant
attached to the 9-carbon position of fluorene. When carbazole was
Nonetheless, we note that the triplet energy (ET) of PF (2.1 eV) is
nearly the same as that of the quinoline-based Ir complex (2.1 eV)
[10]. It remains unknownwhether the back energy transfer from the
dopant to PF happens in this system, and to what extent it affects the
device performance [11]. To clarify this doubt, in this article, we
change the main chain from PF to PFCz and poly(3,6-carbazole)
(PCz), which have higher ET levels (PFCz: 2.3 eV; PCz: 2.6 eV) than
that of PF, and synthesize two series of the corresponding red light-
emitting phosphorescent polymers, PFCz05R0ePFCz5R0 and
PCz05R0ePCz5R0. In comparison to polymer PF1R0, the lifetimes of
PFCz1R0 and PCz1R0 gradually increase with the increasing ET of the
* Corresponding authors.
E-mail addresses: junqiaod@ciac.jl.cn (J. Ding), lixiang@ciac.jl.cn (L. Wang).
Fax: þ86 0431 8568 5653.
1
0032-3861/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2011.03.023

2190
Z. Ma et al. / Polymer 52 (2011) 2189e2197
Scheme 1. Synthetic route of the complex monomer.
main chain, indicative of the efficient confinement of the triplet
excitons on the phosphors. Notably, accompanied by the alteration
of the backbone’s structure, we find that the highest occupied
molecular orbital (HOMO) levels of the polymeric hosts have
increased to facilitate the hole injection as well as the transportation
due to the reduced hole injection barrier and hole-trap depth. As
a result, the balance between holes and electrons in the emitting
layer is disrupted, and thus the device performance decreases from
5.0 cd/A of PF1R0 to 1.7 cd/A of PFCz1R0 and 0.3 cd/A of PCz1R0,
respectively. Our results suggest that apart from the triplet energy of
the polymeric host, charge balance is a crucial determinant of high
efficiency PhPLEDs.
analyzed with a PerkineElmer-TGA 7 instrument under nitrogen at
a heating rate of 10 ꢀC minꢁ1. UVeVis absorption spectra were
recorded by a PerkineElmer Lambda 35 UVeVis spectrometer. PL
spectra were recorded using a PerkineElmer LS50B spectrofluo-
rometer. The PL decay curves were obtained from a Lecroy Wave
Runner 6100 digital oscilloscope (1 GHz) by using a tunable laser
(pulse width ¼ 4 ns, gate ¼ 50 ns) as excitation source (Continuum
Sunlite OPO).
2.2. Synthesis
2.2.1. 3,6-Dibromocarbazole (1)
6.8 mL bromine (21.1 g, 132 mmol) dissolved in 30 mL carbon
disulfide was added dropwise to a suspension of carbazole (10.0 g,
60 mmol) in 120 mL carbon disulfide at 50 ꢀC. After the addition,
the reaction was further stirred for 5 h. Then the reaction was
cooled to room temperature, the solid was filtered and washed
with carbon disulfide twice. The final product was recrystallized
2. Experimental
2.1. Materials and characterization
All reagents were purchased from Aldrich or Acros and used
without further purification. THF, DMF and toluene were purified
according to standard procedures. 2,4-diphenyl-quinoline (4) (PPQ)
[12], 3,6-dibromo-N-decylcarbazole (6) [13] and 2,7-bis-(trime-
thyleneborate)-9,9-dioctylfluorene (7) [7] were prepared according
to the literature procedure. 1H NMR spectra were obtained with
a Bruker Avance 300NMR spectrometer. The chemical shifts with
respect to tetramethylsilane as an internal reference are reported in
parts per million. Fourier transform infrared (FTIR) spectra were
recorded on a Bruker Vertex 70 spectrometer with KBr pellets.
Elemental analysis was performed using a Bio-Rad elemental
analysis system. Number- and weight-average molecular weights
(Mn and Mw) together with the polydispersity indexes of the
polymers were determined by gel permeation chromatography
(GPC) on a Waters 410 instrument with polystyrene as a standard
and THF as the eluent. Thermal properties of the polymers were
from ethanol (9.2 g, 47.0%). 1H NMR (300 MHz, CDCl3):
d 8.13 (d,
J ¼ 1.8 Hz, 2H), 8.08 (br, 1H), 7.52 (dd, J ¼ 8.7 Hz, 2H), 7.31 (d,
J ¼ 8.7 Hz, 2H).
2.2.2. 3,6-Dibromo-N-(10-bromodecyl)-carbazole (2)
To a stirred suspension of 3,6-dibromocarbazole 1 (3.0 g,
9.2 mmol) and KOH (0.9 g, 15.5 mmol) in 50 mL DMF, 1,10-dibro-
modecane (6.3 mL, 27.7 mmol) was added at room temperature.
After 24 h, the reaction mixture was poured into water, and
extracted with DCM (100 mL ꢂ3), then the combined organic phase
was washed with water for 3 times and dried over anhydrous
sodium sulfate. After removal of the solvent, the excess of 1,10-
dibromodecane was recovered from the mixture at 120 ꢀC under
reduced pressure. The crude product was purified by silica gel
column with petroleum: DCM ¼ 10:1 as eluent to give white solid

Z. Ma et al. / Polymer 52 (2011) 2189e2197
2191
Scheme 2. Synthetic routes of the polymers and the molecular structures of PFCz, PCz, PF as well as PF1R0.
(3.0 g, 58.8%). 1H NMR (300 MHz, CDCl3):
7.54 (dd, J ¼ 8.7 Hz, 2H), 7.25 (d, J ¼ 8.7 Hz, 2H), 4.22 (t, J ¼ 7.2 Hz,
2H), 3.39 (t, J ¼ 6.6 Hz, 2H), 1.87e1.77 (m, 4H), 1.40e1.23 (m, 12H).
d
8.13 (d, J ¼ 1.8 Hz, 2H),
a further 30 min at 0 ꢀC. N-butyllithium (1.6 M in hexane; 22.8 mL,
36.4 mmol) was added dropwise to the mixture at 0 ꢀC and the
resultant mixture was stirred for a further 30 min at 0 ꢀC. 3,6-
Dibromo-N-(10-bromodecyl)-carbazole 2 (6.6 g, 12.2 mmol) dis-
solved in THF (25 mL) was added to the stirred solution dropwisely
at 0 ꢀC. The reaction was kept at 0 ꢀC for 1 h and a further 1 h at
room temperature, after which it was quenched with aqueous
ammonium chloride (10 wt%, 40 mL). Concentrated hydrochloric
acid was added to the mixture until pH 1 was reached, and then the
2.2.3. N-(12,14-dioxo-quindecyl)-3,6-dibromocarbazole (3) [14]
Sodium hydride (50 wt% in mineral oil; 2.1 g, 42.5 mmol) was
washed with THF (20 mL each) three times. Acetylacetone (4.0 mL,
36.4 mmol) was added dropwise to a stirred suspension of sodium
hydride in THF (80 mL) at 0 ꢀC. The resultant mixture was stirred for
Table 1
Molecular Weights and Thermal Properties of Polymers.
Polymer
Mn
Mw
PDIa
(6 þ 7) or 6/BrCzR0 (molar ratio)d
a
a
b
Tg (ꢀC)
c
Td (ꢀC)
monomer in feed ratio
composition in polymere
PFCz05R0
PFCz1R0
PFCz2R0
PFCz5R0
PCz05R0
PCz1R0
18,000
17,000
19,000
20,000
32,000
45,000
64,000
78,000
40,000
36,000
39,000
52,000
48,000
89,000
129,000
168,000
2.21
2.14
2.54
2.56
1.51
1.95
2.01
2.16
101
102
105
106
128
128
135
137
418
416
418
358
437
443
441
350
99.5/0.5
99/1
98/2
95/5
99.5/0.5
99/1
e
e
99.4/0.6
96.9/3.1
e
e
PCz2R0
PCz5R0
98/2
95/5
99.7/0.3
98.6/1.4
a
Determined by GPC using THF as eluent with polystyrene as the standard.
b
c
Glass transition temperature measured by DSC under nitrogen at a heating rate of 10 ꢀC/min.
Temperature of 5% weight loss measured by TGA in nitrogen.
d
e
For PFCz-based polymers, the molar ratio is (6 þ 7)/BrCzR0; For PCz-based polymers, the molar ratio is 6/BrCzR0.
Calculated according to 1H NMR spectra.

2192
Z. Ma et al. / Polymer 52 (2011) 2189e2197
Fig. 1. IR spectra of the polymers. (a) PFCz-based polymers and PFCz; (b) PCz-based polymers and PCz.
aqueous phase was separated and extracted with diethyl ether. The
combined organic phases were washed with saturated sodium
bicarbonate, dried over anhydrous sodium sulfate. After removal of
the solvent, the crude product was purified by silica gel column
with petroleum: DCM ¼ 2:1 as eluent to give a white solid (4.8 g,
temperature. Then the resultant mixture was poured into water and
extracted with DCM. The organic phase was washed with brine and
dried with anhydrous sodium sulfate. After removal of the solvent,
the crude product was purified by alkaline alumina column with
petroleum: DCM ¼ 2:1 as eluent to give a red solid (0.33 g, 22.4%).1H
70.8%). 1H NMR (300 MHz, CDCl3):
d
15.48 (br, 0.8H), 8.13 (d,
NMR (300 MHz, CDCl3): d 8.59e8.52 (m, 2H), 8.15 (s, 2H), 8.03 (s,
J ¼ 1.8 Hz, 2H), 7.54 (dd, J ¼ 8.6 Hz, 2H), 7.26 (d, J ¼ 8.7 Hz, 2H), 5.48
(s, 0.8H), 4.23 (t, J ¼ 7.2 Hz, 2H), 3.56 (s, 0.4H), 2.48 (t, J ¼ 7.5 Hz,
0.4H), 2.25 (t, J ¼ 7.2 Hz, 1.6H), 2.23 (s, 0.4H), 2.05 (s, 1.6H),
1.88e1.77 (m, 2H), 1.62e1.53 (m, 2H), 1.28e1.21 (m, 14H).
1H), 7.97 (s, 1H), 7.87e7.78 (m, 4H), 7.70e7.50 (m, 12H), 7.73e7.40
(m, 4H), 7.26 (d, J ¼ 14.1 Hz, 2H), 6.96e6.88 (m, 2H), 6.70e6.55 (m,
4H), 4.68 (s,1H), 4.23 (t, J ¼ 6.9 Hz, 2H),1.83e1.64 (m, 4H),1.53e1.54
(m, 4H),1.26e0.80 (m,15H). IR (KBr, cmꢁ1): 3055 (w), 2922 (s), 2850
(s), 1594 (s), 1578 (s), 1536 (s), 1513 (s), 1494 (s), 1445 (s), 1406 (s),
1379 (s), 1288 (s), 1264 (s), 1047 (s), 762 (s), 701 (s).
0
2.2.4. 9-(iridium(Ⅲ)bis(2,4-diphenyl-quinoline-N,C2 )
(quindecanedionate-12,14))-3,6-dibromocarbazole (5)
Cyclometalated Ir(III)
m-chloro-bridged dimer was synthesized
2.2.5. General procedure of the Suzuki polymerization
according to the Nonoyama method [15]. Iridium chloride trihy-
drate (0.88 g, 2.5 mmol) and 2,4-diphenyl-quinoline 4 (1.48 g,
5.3 mmol) were added to a mixed solvent of ethoxyethanol (45 mL)
and water (15 mL). The mixture was refluxed for 24 h. After cooling
to room temperature, the precipitate was collected by filtration and
purified by alkaline alumina column with petroleum: DCM ¼ 2:1 to
DCM: methanol ¼ 50:1 as eluent. The resulting dimer (0.77 g,
0.56 mmol), N-(12,14-dioxo-quindecyl)-3,6-dibromocarbazole 3
(0.64 g, 1.15 mmol) and sodium carbonate (0.60 g, 5.6 mmol) were
added to a mixed solvent of acetonitrile (15 mL) and chloroform
(15 mL). The mixture was refluxed for 24 h before cooling to room
To a mixture of 3,6-dibromo-N-decylcarbazole 6, 2,7-bis-(trime-
thyleneborate)-9,9-dioctylfluorene
7 and 9-(iridium(Ⅲ)bis(2,4-
0
diphenyl-quinoline-N,C2 )(quindecanedionate-12,14))-3,6-dibro-
mocarbazole 5 with the corresponding feed ratios, aliquat 336
(20 mg) and Pd(PPh3)4 (1 mol %) were added. The mixture was
degassed for 30 min, and then 2 M degassed aqueous K2CO3 (3 mL)
and degassed toluene (8 mL) were added under argon. The mixture
was heated to 90 ꢀC and stirred for 24 h in the dark. Then the end-
capping reagents were added. First phenylboronic acid (100 mg)
stirring for 12 h, and then bromobenzene (0.5 mL) for another 12 h.
Afterworkup, themixturewaspouredinto methanol. Theprecipitate
Fig. 2. The absorption spectrum in DCM and PL spectrum in toluene of BrCzR0 as well as the PL spectra of PFCz and PCz in films (a) and the phosphorescent spectra of PF, PFCz and
PCz (b).

Z. Ma et al. / Polymer 52 (2011) 2189e2197
2193
2.2.5.3. PFCz2R0. 3,6-dibromo-N-decylcarbazole
6
(223.3 mg,
0.48 mmol), BrCzR0 (26.3 mg, 0.02 mmol), 2,7-bis-(trimethylene-
borate)-9,9-dioctylfluorene 7 (279.2 mg, 0.5 mmol) were used in
the polymerization. 1H NMR (300 MHz, CDCl3):
d 8.50 (s, 2H), 7.85
(d, J ¼ 7.5 Hz, 4H), 7.77e7.73 (m, 4H), 7.53 (d, J ¼ 8.4 Hz, 2H),
6.96e6.90 (m, ArH of Ir(PPQ)), 6.65e6.58 (m, ArH of Ir(PPQ)), 4.39
(s, 2H), 2.14 (br, 4H), 1.96 (br, 2H), 1.26e1.11 (m, 38H), 0.89e0.74 (m,
9H). Anal Calcd for (C29H40)50(C22H27N)48(C69H61IrN3O2)2: C, 87.23;
H, 9.46; N, 2.08. Found: C, 86.48; H, 9.21; N, 1.55. IR (KBr, cmꢁ1):
2954 (s), 2926 (s), 2854 (s), 1862 (w), 1629 (w), 1603 (m), 1580 (w),
1496 (m), 1463 (s), 1378 (m), 1268 (m), 879 (m), 825 (m), 802 (s),
761 (w), 722 (w).
Fig. 3. Energy level diagram containing the singlet ground state (S0), the lowest
excited singlet state (S1), and the lowest excited triplet state (T1) of the polymer hosts,
as well as the T1 level of the complex monomer BrCzR0.
2.2.5.4. PFCz5R0. 3,6-dibromo-N-decylcarbazole
6
(209.4 mg,
was collected by filtration and then dissolved in DCM. The solution
was then washed with water and then dried with anhydrous sodium
sulfate. The solution was concentrated to an appropriate volume and
then poured into methanol to get polymer fibers. The polymer was
purified by Soxhlet extraction in acetone for 48 h. The reprecipitation
procedure in DCM/methanol was then repeated for several times.
The final product was obtained after drying in vacuum with a yield of
50%e60%.
0.45 mmol), BrCzR0 (65.8 mg, 0.05 mmol), 2,7-bis-(trimethylenebo-
rate)-9,9-dioctylfluorene 7 (279.2 mg, 0.5 mmol) were used in the
polymerization. 1H NMR (300 MHz, CDCl3):
d 8.68e8.58 (m, ArH of Ir
(PPQ)), 8.50 (s, 2H), 7.85 (d, J ¼ 7.5 Hz, 4H), 7.77e7.73 (m, 4H), 7.53 (d,
J ¼ 8.4 Hz, 2H), 6.96e6.90 (m, ArH of Ir(PPQ)), 6.65e6.58 (m, ArH of Ir
(PPQ)), 4.39 (s, 2H), 2.14 (br, 4H), 1.96 (br, 2H), 1.26e1.11 (m, 38H),
0.89e0.74 (m, 9H). Anal Calcd for (C29H40)50(C22H27N)45
(C69H61IrN3O2)5: C, 85.86; H, 9.11; N, 2.16. Found: C, 84.95; H, 8.71; N,
2.85. IR (KBr, cmꢁ1): 2953 (s), 2926 (s), 2854 (s), 1862 (w), 1629 (w),
1603 (m), 1580 (m), 1538 (w), 1495 (m), 1463 (s), 1379 (m), 1268 (m),
879 (m), 825 (m), 802 (s), 762 (w), 723 (w).
2.2.5.1. PFCz05R0. 3,6-dibromo-N-decylcarbazole
6 (230.3 mg,
0.495 mmol), BrCzR0 (6.6 mg, 0.005 mmol), 2,7-bis-(trimethyle-
neborate)-9,9-dioctylfluorene 7 (279.2 mg, 0.5 mmol) were used in
the polymerization.1H NMR (300 MHz, CDCl3):
d 8.50 (s, 2H), 7.85 (d,
2.2.6. General procedure of the Yamamoto polymerization
J ¼ 7.5 Hz, 4H), 7.77e7.73 (m, 4H), 7.53 (d, J ¼ 8.4 Hz, 2H), 4.39 (s, 2H),
To a mixture of Ni(COD)2 (0.33 g, 1.2 mmol) and 2,20-bypyridyl
(0.19 g, 1.2 mmol), 0.12 mL COD and 10 mL anhydrous DMF were
added under argon. The resulting solution was kept stirring at 60 ꢀC
for 0.5 h, then slowly added to a solution of 3,6-dibromo-N-
2.14 (br, 4H), 1.96 (br, 2H), 1.26e1.11 (m, 38H), 0.89e0.74 (m, 9H).
Anal Calcd for (C29H40)50(C22H27N)49.5(C69H61IrN3O2)0.5: C, 87.99; H,
9.66; N, 2.03. Found: C, 86.65; H, 9.76; N, 1.68. IR (KBr, cmꢁ1): 2954
(s), 2926 (s), 2854 (s),1862 (w),1628 (w),1603 (m),1496 (s),1463 (s),
1378 (m), 1268 (m), 879 (m), 825 (m), 802 (s), 757 (w), 722 (w).
decylcarbazole
6
and 9-(iridium(Ⅲ)bis(2,4-diphenyl-quinoline-
0
N,C2 )(quindecanedionate-12,14))-3,6-dibromocarbazole
5
with
corresponding feed ratios in anhydrous DMF (2 mL). The reaction
was maintained at 60 ꢀC for 24 h in the dark. The resulting
precipitate was filtered, and dissolved in DCM, then washed
intensively with saturated EDTA-4Na solution. After dried with
anhydrous sodium sulfate, the solution was concentrated to an
appropriate volume and then poured into isopropanol to get
polymer fibers. The polymer was purified by Soxhlet extraction in
cyclohexane for 48 h. The reprecipitation procedure in DCM/iso-
propanol was then repeated for several times. The final product was
obtained after drying in vacuum with a yield of 30%e50%.
2.2.5.2. PFCz1R0. 3,6-dibromo-N-decylcarbazole
6
(228.0 mg,
0.49 mmol), BrCzR0 (13.2 mg, 0.01 mmol), 2,7-bis-(trimethylene-
borate)-9,9-dioctylfluorene 7 (279.2 mg, 0.5 mmol) were used in the
polymerization. 1H NMR (300 MHz, CDCl3):
d 8.50 (s, 2H), 7.84 (d,
J ¼ 7.5 Hz, 4H), 7.77e7.73 (m, 4H), 7.53 (d, J ¼ 8.4 Hz, 2H), 4.39 (s, 2H),
2.14 (br, 4H), 1.96 (br, 2H), 1.26e1.11 (m, 38H), 0.89e0.74 (m, 9H).
Anal Calcd for (C29H40)50(C22H27N)49(C69H61IrN3O2)1: C, 87.73; H,
9.59; N, 2.05. Found: C, 87.26; H, 9.24; N, 2.36. IR (KBr, cmꢁ1): 2954
(s), 2926 (s), 2854 (s), 1862 (w), 1629 (w), 1603 (m), 1496 (m), 1463
(s), 1378 (m), 1268 (m), 879 (m), 825 (m), 802 (s), 757 (w), 722 (w).
2.2.6.1. PCz05R0. 3,6-dibromo-N-decylcarbazole 6 (463.0 mg, 0.995
mmol), BrCzR0 (6.6 mg, 0.005 mmol) were used in the polymeri-
zation. 1H NMR (300 MHz, CDCl3):
d 8.53 (br, 2H), 7.87 (br, 2H), 7.44
(br, 2H), 4.21 (br, 2H), 1.85 (br, 2H), 1.43e1.13 (m, 14H), 0.82 (t,
J ¼ 6 Hz, 3H). Anal Calcd for (C22H27N)99.5(C69H61IrN3O2)0.5: C, 86.23;
H, 8.84; N, 4.57. Found: C, 83.64; H, 8.64; N, 5.05. IR (KBr, cmꢁ1): 2952
(s), 2925(s), 2853 (s),1859 (w),1629 (w),1605 (m),1473(s),1379(m),
1348 (s),1302 (m),1268 (s),1241 (s),1135 (s), 874 (s), 797 (s), 722 (w).
2.2.6.2. PCz1R0. 3,6-dibromo-N-decylcarbazole 6 (460.6 mg, 0.99
mmol), BrCzR0 (13.2 mg, 0.01 mmol) were used in the polymeri-
zation. 1H NMR (300 MHz, CDCl3):
d 8.51 (br, 2H), 7.85 (br, 2H), 7.41
(br, 2H), 4.19 (br, 2H), 1.84 (br, 2H), 1.42e1.11 (m, 14H), 0.82 (t,
J ¼ 6.3 Hz, 3H). Anal Calcd for (C22H27N)99(C69H61IrN3O2)1: C, 85.96;
H, 8.78; N, 4.55. Found: C, 84.56; H, 9.11; N, 4.10. IR (KBr, cmꢁ1): 2952
(s), 2925 (s), 2853 (s), 1859 (w), 1629 (w), 1604 (m), 1472 (s), 1379
(m),1347 (s),1301 (m),1268 (m),1235 (s),1136 (m), 874 (m), 797 (s),
722 (w).
2.2.6.3. PCz2R0. 3,6-dibromo-N-decylcarbazole
0.98 mmol), BrCzR0 (26.3 mg, 0.02 mmol) were used in the
6
(456.0 mg,
Fig. 4. Decay of the PL intensity (excited at 355 nm) at room temperature from thin
films of PF1R0, PFCz1R0, PCz1R0.

2194
Z. Ma et al. / Polymer 52 (2011) 2189e2197
Fig. 5. The absorption and PL spectra of the polymers in films (a) PFCz-based polymers; (b) PCz-based polymers.
polymerization. 1H NMR (300 MHz, CDCl3):
d
8.53 (br, 2H), 7.86 (br,
another layer of 100 nm thick Al was deposited on the top as
a protecting layer for Ca. The active emissive area defined by the
two electrodes was 10 mm2. The hole-only devices were fabricated
with the configuration of ITO/PEDOT:PSS/polymer/Au. The EL
spectra, CIE coordinates, currentevoltage and brightnessevoltage
characteristics of the devices were measured with a Spectrascan
PR650 spectrophotometer in the forward direction and a computer-
controlled Keithley 2400 under ambient conditions.
2H), 7.42 (br, 2H), 6.96e6.90 (m, ArH of Ir(PPQ)), 6.61e6.58 (m, ArH
of Ir(PPQ)), 4.19 (br, 2H), 1.84 (br, 2H), 1.44e1.10 (m, 14H), 0.82 (t,
J ¼ 6 Hz, 3H). Anal Calcd for (C22H27N)98(C69H61IrN3O2)2: C, 85.44;
H, 8.65; N, 4.52. Found: C, 84.65; H, 8.54; N, 4.73. IR (KBr, cmꢁ1):
2952 (s), 2925 (s), 2853 (s), 1859 (w), 1629 (m), 1605 (s), 1578 (m),
1537 (w), 1471 (s), 1379 (s), 1348 (s), 1301 (s), 1268 (s), 1235 (s), 1136
(s), 874 (s), 797 (s), 722 (w).
2.2.6.4. PCz5R0. 3,6-dibromo-N-decylcarbazole
0.95 mmol), BrCzR0 (65.8 mg, 0.05 mmol) were used in the poly-
merization. 1H NMR (300 MHz, CDCl3):
8.52 (br, 2H), 7.85 (br, 2H),
6
(442.0 mg,
3. Results and discussion
d
3.1. Synthesis and characterization
7.44 (br, 2H), 6.96e6.91 (m, ArH of Ir(PPQ)), 6.64e6.59 (m, ArH of Ir
(PPQ)), 4.21 (br, 2H), 1.84 (br, 2H), 1.44e1.11 (m, 14H), 0.82 (t,
J ¼ 6.3 Hz, 3H). Anal Calcd for (C22H27N)95(C69H61IrN3O2)5: C, 84.04;
H, 8.31; N, 4.43. Found: C, 83.38; H, 8.34; N, 5.18. IR (KBr, cmꢁ1):
2952 (s), 2925 (s), 2853 (s), 1859 (w), 1629 (w), 1604 (m), 1579 (m),
1537 (w), 1473 (s), 1379 (m), 1348 (s), 1301 (m), 1268 (s), 1235 (s),
1135 (m), 874 (s), 797 (s), 722 (w).
Scheme 1 illustrates the synthetic route of the Ir complex
monomer 5 (BrCzR0). This key phosphorescent monomer 5 was
obtained by reacting 3,6-dibromocarbazole with a diketone-ended
alkyl chain attached to its N-position (3) with 2,4-diphenylquino-
line (PPQ)-based Ir chloride-bridged dimer [15]. And the synthesis
of the corresponding electrophosphorescent polymers is shown in
Scheme 2. The polymers with PFCz as the main chain were
2.3. Device fabrication
produced by
a
typical palladium-catalyzed Suzuki poly-
condensation, and after the completion of the polymerization, the
polymers were sequentially end-capped by phenylboronic acid and
bromobenzene. On the other hand, a modified Yamamoto coupling
method was used to prepare the polymers with PCz as the main
chain, which utilized a reverse order of the addition of the nickel
reagent into the monomer solution, and in this way, PCz-based
polymers with high molecular weights were obtained [16]. As
determined by GPC (Table 1), the Mn values are around 20,000 with
polydispersity indexes (PDIs) of 2.21e2.56 and 32,000e78,000
with PDIs of 1.51e2.16 for the PFCz- and PCz-based polymers,
respectively. And their thermal properties were also characterized
by DSC and TGA in a nitrogen atmosphere. As collected in Table 1,
The indium-tin oxide (ITO) glass plates were cleaned in an
ultrasonic solvent bath and then baked in a heating chamber at
120 ꢀC, and treated with O2 plasma for 25 min before use. A 50 nm
thick ITO-modifying layer of poly(ethylenedioxythiophene):poly
(styrene sulfonic acid) (PEDOT:PSS) was spin-coated on top of the
ITO and then baked for 40 min at 120 ꢀC. Thin films of the polymers
were spin-coated from their solution in chlorobenzene
(15 mg mLꢁ1). The substrate was then thermally annealed in
vacuum at 100 ꢀC for 30 min. After being cooled, the substrate was
transferred to a vacuum thermal evaporator. A 10 nm thick layer of
Ca was deposited at a pressure of 2 ꢂ 10ꢁ3 Pa through a mask, and
Table 2
Device Performance of the Polymers.
Polymer
lmax (nm)
Voltagea (V)
Luminanceb (cd/m2)
EQEb (%)
LEb (cd/A)
PEb (lm/w)
CIE (x,y)
PFCz05R0
PFCz1R0
PFCz2R0
PFCz5R0
PCz05R0
PCz1R0
608
608
612
620
608
612
616
620
4.2
4.2
5.3
5.3
4.0
4.0
5.4
6.0
3968
3745
2057
638
88
425
263
33
1.5
1.2
0.5
0.3
0.1
0.2
0.1
0.04
2.3
1.7
0.7
0.3
0.1
0.3
0.1
0.04
1.1
0.7
0.2
0.2
0.1
0.1
0.04
0.02
(0.60,0.35)
(0.62,0.36)
(0.63,0.36)
(0.65,0.35)
(0.56,0.35)
(0.61,0.36)
(0.63,0.35)
(0.63,0.34)
PCz2R0
PCz5R0
a
The voltage at a luminance of 1 cd/m2.
The maximum values for the luminance, external quantum efficiency (EQE), luminous efficiency (LE) and power efficiency (PE).
b

Z. Ma et al. / Polymer 52 (2011) 2189e2197
2195
Fig. 6. The current density dependence of the luminous efficiency of the polymers. (a) PFCz-based polymers; (b) PCz-based polymers.
the PFCz-based polymers, PFCz05R0ePFCz5R0, have similar glass
transition temperatures (Tg: 101e106 ꢀC), corresponding to that of
the PFCz backbone (100 ꢀC), due to the attachment of the low
content of the Ir complex in the final polymers. Additionally,
PFCz05R0ePFCz2R0 show close decomposition temperatures (Td:
416e418 ꢀC), indicating their good thermal stability. However, the
Td of PFCz5R0 decreases to 358 ꢀC when higher content of (PPQ)2Ir
(acac) is incorporated to the final polymers. As for the PCz-based
polymers, PCz05R0- PCz5R0, the same trend was observed.
Fig. 1(a) and (b) give the IR spectra of the resulting polymers in
comparison with PFCz and PCz, respectively. As can be seen clearly,
the characteristic bands at 1579 and 1538 cmꢁ1 assigned to the
acetylacetone group of (PPQ)2Ir(acac) (R0) become more intense as
the feed ratio increases, suggesting the increasing content of this Ir
segment incorporated to the final polymers [17]. To quantify the
actual attached number of (PPQ)2Ir(acac), 1H NMR spectra were
characterized. However, no meaningful information can be found
when the feed ratio is below 2 mol %. Only when the feed ratio
reaches to 2 mol %, the visible separated signals appear at
6.97e6.87 and 6.72e6.55 ppm, which can be attributed to the
protons of PPQ ligand in (PPQ)2Ir(acac). Therefore, the relative ratio
of the (PPQ)2Ir(acac) in the PFCz-based polymers can be calculated
by comparing the integration of the 1H NMR signals at
6.72e6.55 ppm with the sum of the first CH2 groups of the alkyl
chains to the 9-position of fluorene (d 2.15) together with that to the
Fig. 7. The comparison of PL and EL spectra (at a driving voltage of 10 V) of PF1R0
Fig. 8. The comparison of the luminous efficiency and triplet energy among PF1R0,
(a), PFCz1R0 (b) and PCz1R0 (c).
PFCz1R0 and PCz1R0.

2196
Z. Ma et al. / Polymer 52 (2011) 2189e2197
estimated the ET levels of PF, PFCz and PCz to be 2.1 eV, 2.3 eV and
2.6 eV, respectively. Fig. 3 displays all energy levels of the hosts and
guest. As can be clearly seen, the triplet energies of both PFCz and
PCz were higher than that of BrCzR0. Compared with PF, this triplet
energy enhancement would further suppress the backward energy
transfer from the phosphor to the polymeric host, and avoid the loss
of the triplet excitons [11,19]. To confirm our hypothesis, the tran-
sient PL spectra of PF1R0, PFCz1R0 and PCz1R0 were measured, and
their lifetimes were obtained by single exponential fit of emission
decay curves (Fig. 4). With the increasing triplet energy, the lifetime
gradually increases from 1.15
ms of PF1R0 to 1.23 ms of PFCz1R0 and
1.46 s of PCz1R0. This observation further demonstrates that the
m
triplet excitons can be effectively confined on the phosphor for the
electrophosphorescent polymer with higher triplet energy main
chain.
Fig. 5(a) and (b) depict the UVeVis as well as PL spectra of the
PFCz- and PCz-based electrophosphorescent polymers in films,
respectively. For both of these two series of polymers, the
absorption spectra resemble to those of their pristine polymeric
backbone, and no distinct absorptions belonging to the Ir complex
are observed because of its low content loading. However, their PL
spectra show obvious red light emission from the Ir complex even
though the feed ratio is as low as 0.5 mol %. Moreover, the
intensity of the emission from the polymer backbone decreases
with the increasing feed ratio. Especially for PFCz5R0, PCz2R0 and
PCz5R0, the emission from PFCz or PCz is almost quenched
completely. These results indicate the efficient energy transfer
from PFCz or PCz to the Ir complex. In addition, with the
increasing Ir content in the final polymers, the emission shows
a small red shift, which means that the aggregation exists in these
polymers to some degree.
Fig. 9. The proposed energy-level diagram for the devices.
N-position of carbazole (d 4.21). For PCz-based ones, the results are
obtained in a similar way. The calculations and the feed ratios for
comparison are all listed in Table 1. Similar to previously reported
results, the actual Ir complex content in the final polymer is found
to be substantially lower than the feed ratio of the monomer [9].
3.2. Photophysical properties
Fig. 2(a) shows UVeVis absorption and photoluminescent (PL)
spectra of the monomer BrCzR0 in solutions, and the PL spectra of
PFCz and PCz in films. According tothe literature [12], the absorption
bands around 275 and 350 nm can be attributed to the ligand-
centered (LC) transitions of the Ir complex, and the
pep
* absorp-
3.3. Device performance
tions of the carbazole moiety locate at 240 and 305 nm. The weak
absorptions below 400 nm are assigned to the metal-to-ligand
charge-transfer (MLCT) absorption of the Ir complex. As indicated in
Fig. 2(a), the PL spectra of both PFCz and PCz have a good overlap
with the MLCT absorption of the phosphorescent guest BrCzR0,
which means that the Förster energy transfer from PFCz and PCz to
BrCzR0 is efficient [18]. Simultaneously, in order to determine the ET
level of the polymer backbone, we measured the phosphorescent
spectra of the pristine polymers at 77 K, and the results are displayed
in Fig. 2(b). According to the maximum of the first vibronic mode
Single layer devices with the configuration of ITO/PEDOT:PSS/
Polymer/Ca/Al were fabricated and the results are summarized in
Table 2. The current density dependence of luminous efficiency of
the two series of polymers is displayed in Fig. 6. The best results
were obtained from PFCz05R0-based device, and the maximum
luminous efficiency reached to 2.3 cd/A with the Commission
Internationale de L’Eclairage (CIE) coordinate of (0.60, 0.35). When
the current density increased to 100 mA/cm2, the luminous effi-
ciency still remained as high as 1.7 cd/A, indicative of a gentle
efficiency roll-off for the chemical attaching system in contrast to
the physical blending system [4e6]. As for the PCz-based polymers,
the device performance was relatively poor, and only 0.3 cd/A was
achieved for PCz1R0.
n
n
(S0 ¼0 ) T1 ¼0) of the corresponding phosphorescence spectra, we
Fig. 7 gives the PL and electroluminescence (EL) spectra of
PF1R0, PFCz1R0 and PCz1R0. Similar to PF1R0, the EL spectra of
PFCz1R0 and PCz1R0 are obviously different from their PL coun-
terparts. In the EL spectra, the emission is mainly from the Ir
complex guest, and the emission derived from the polymeric host is
completely quenched, which demonstrates that direct charge
trapping on the Ir complex is the main mechanism in the EL process
[20]. However, it is interesting to note that the efficiency of PFCz1R0
and PCz1R0 is much lower than that of PF1R0 (Fig. 8), although
their triplet energies of the main chain are improved. This result is
out of our expectation, but we can reasonably interpret it from the
standpoint of charge balance.
As can be clearly seen in Fig. 9, the HOMO levels of the polymeric
backbones change at the same time while tuning the triplet energy.
For example, the HOMO level increases from ꢁ5.80 eV of PF
to ꢁ5.25 eV of PCz [21e23]. This improvement, on one hand, can
decrease the hole injection barrier, and facilitate the efficient hole
injection. On the other hand, the hole-trap depth between the
Fig. 10. Comparison of the hole current density of PF1R0, PFCz1R0 and PCz1R0.

Z. Ma et al. / Polymer 52 (2011) 2189e2197
2197
polymeric host and the Ir guest has decreased from 0.62 eV for PF to
Acknowledgements
0.07 eV for PCz, which will weaken the hole-trap capability and
enhance the hole transportation in contrary. Taking these two
factors into account, the consequence is that the hole current will
increase, and result in the charge imbalance in the emitting layer
accompanied by the poor device performance. This can be further
verified by the comparison of the hole current measured from the
hole-only devices. Ongoing from PF1R0 to PFCz1R0 and PCz1R0, the
hole current is found to increase in turn, as shown in Fig. 10.
Consequently, the calculated hole mobility using space-charge-
limited current (SCLC) method [24] improves from 1.9 ꢂ 10ꢁ12 cm2/
(Vs) of PF1R0 to 1.9 ꢂ 10ꢁ8 and 9.2 ꢂ 10ꢁ6 cm2/(Vs) of PFCz1R0 and
PCz1R0, respectively.
Based on the above-mentioned observations, we can speculate
that the triplet energy of the polymeric host is not the only factor
that affects the device performance of the PhPLEDs. As for the red
light-emitting electrophosphorescent polymers, where the main
chain has the comparable triplet energy with the Ir complex
dopant, charge balance in the emissive layer plays a paramount
role in determining the performance of the resulting materials.
The authors are grateful to the National Natural Science Foun-
dation of China (No. 20923003), Science Fund for Creative Research
Groups (No. 20921061), and 973 Project (2009CB623601) for
financial support of this research.
References
[1] Kraft A, Grimsdale AC, Holmes AB. Angew Chem Int Ed Engl 1998;37:402e28.
[2] Baldo MA, O’Brien DF, You Y, Shoustikov A, Sibley S, Thompson ME, et al.
Nature 1998;395:151e4.
[3] Baldo MA, Adachi C, Forrest SR. Phys Rev B 2000;62:10967e77.
[4] Lee CL, Lee KB, Kim JJ. Appl Phys Lett 2000;77:2280e2.
[5] Noh YY, Lee CL, Kim JJ, Yase K. J Chem Phys 2003;118:2853e64.
[6] Gong X, Ostrowski JC, Bazan GC, Moses D, Heeger AJ, Liu MS, et al. Adv Mater
2003;15:45e9.
[7] Chen X-W, Liao J-L, Liang Y-M, Ahmed MO, Tseng H-E, Chen S-A. J Am Chem
Soc 2003;125:636e7.
[8] Jiang JX, Jiang CY, Yang W, Zhen HG, Huang F, Cao Y. Macromolecules 2005;
38:4072e80.
[9] Ma ZH, Ding JQ, Zhang BH, Mei CY, Cheng YX, Xie ZY, et al. Adv Funct Mater
2010;20:138e46.
[10] Ding JQ, Lv JH, Cheng YX, Xie ZY, Wang LX, Jing XB, et al. Adv Funct Mater
2008;18:2754e62.
[11] Sudhakar M, Djurovich PI, Hogen-Esch TE, Thompson ME. J Am Chem Soc
2003;125:7796e7.
4. Conclusions
[12] Ding JQ, Gao J, Fu Q, Cheng YX, Ma DG, Wang LX. Synth Met 2005;155:
539e48.
[13] You Y, Kim SH, Jung HK, Park SY. Macromolecules 2006;39:349e56.
[14] Evans NR, Devi LS, Mak CSK, Watkins SE, Pascu SI, Kohler A, et al. J Am Chem
Soc 2006;128:6647e56.
[15] Nonoyama M. Bull Chem Soc Jpn 1974;47:767e8.
[16] Zhang Z-B, Fujiki M, Tang H-Z, Motonaga M, Torimitsu K. Macromolecules
2002;35:1988e90.
[17] Mei CY, Ding JQ, Yao B, Cheng YX, Xie ZY, Geng YH, et al. J Polym Sci Part A:
Polym Chem 2007;45:1746e57.
[18] Förster T. Discuss Faraday Soc; 1959:7e17.
[19] Cleave V, Yahioglu G, Le Barny P, Hwang DH, Holmes AB, Friend RH, et al. Adv
Mater 2001;13:44e7.
[20] Gong X, Ostrowski JC, Moses D, Bazan GC, Heeger AJ. Adv Funct Mater 2003;
13:439e44.
[21] Yuan M-C, Shih P-I, Chien C-H, Shu C-F. J Polym Sci Part A: Polym Chem 2007;
45:2925e37.
We have synthesized red light-emitting electrophosphorescent
polymers with PFCz and PCz as the main chain. Compared with PF,
the triplet energy of the backbone among these polymers is
improved, and the triplet excitons can be effectively confined on
the Ir complex. However, the higher triplet energy of the polymer
main chain doesn’t mean higher luminous efficiency of the corre-
sponding polymers. As for PFCz1R0 and PCz1R0, the maximum
luminous efficiencies are 1.7 and 0.3 cd/A, respectively, which are
substantially lower than that of PF1R0 (5.0 cd/A). Based on the
comparison of the hole current obtained from the hole-only
devices, we consider that, while tuning the triplet energy of the
polymeric host, the HOMO level alters simultaneously, which leads
to charge imbalance in the emitting layer, and thus poor device
performance. It is believed that, from these results, charge balance
must be firstly considered to develop high efficiency red light-
emitting electrophosphorescent polymers.
[22] Li YN, Ding JF, Day M, Tao Y, Lu JP, D’Iorio M. Chem Mater 2004;16:2165e73.
[23] Huang J, Niu YH, Yang W, Mo YQ, Yuan M, Cao Y. Macromolecules 2002;35:
6080e2.
[24] Chu TY, Song OK. Appl Phys Lett 2007;90:203512e4.