Copolyfluorenes containing pendant bipolar groups: Synthesis, optoelectronic properties and applications

Article abstract of DOI:10.1039/c0jm00707b

This article reports the synthesis of copolyfluorenes (P1-P3) containing pendant bipolar groups (2.1-8.2 mol%), directly linked to hole-transporting triphenylamine and electron-transporting aromatic 1,2,4-triazole, by the Suzuki coupling reaction and their application in improving PLED performance of conventional MEH-PPV. The bipolar groups not only suppress undesirable green emission of polyfluorene under thermal annealing, but also promote electron- and hole-affinity of the resulting copolyfluorenes. Blending bipolar model compound M0 with MEH-PPV results in enhancement of device performance [ITO/PEDOT:PSS/M0 + MEH-PPV/Ca(50 nm)/Al(100 nm)]. Maximum luminance and luminance efficiency were increased from 3160 cd/m² and 0.24 cd/A of MEH-PPV-only device to 6830 cd/m² and 0.50 cd/A. Moreover, blending the bipolar copolyfluorenes with MEH-PPV further improves the device efficiency, with the maximum luminance and luminance efficiency being significantly enhanced up to 11090 cd/m² and 0.56 cd/A (ca. 0.4 wt% of bipolar residue), respectively. Our results demonstrate the efficacy of the bipolar copolyfluorenes and model compound M0 in enhancing emission efficiency of MEH-PPV.

Full text of DOI:10.1039/c0jm00707b

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Copolyfluorenes containing pendant bipolar groups: Synthesis, optoelectronic
properties and applications†
*
Chia-Shing Wu and Yun Chen
Received 15th March 2010, Accepted 11th June 2010
DOI: 10.1039/c0jm00707b
This article reports the synthesis of copolyfluorenes (P1–P3) containing pendant bipolar groups (2.1–
8.2 mol%), directly linked to hole-transporting triphenylamine and electron-transporting aromatic
1,2,4-triazole, by the Suzuki coupling reaction and their application in improving PLED performance
of conventional MEH-PPV. The bipolar groups not only suppress undesirable green emission of
polyfluorene under thermal annealing, but also promote electron- and hole-affinity of the resulting
copolyfluorenes. Blending bipolar model compound M0 with MEH-PPV results in enhancement of
device performance [ITO/PEDOT:PSS/M0 + MEH-PPV/Ca(50 nm)/Al(100 nm)]. Maximum
luminance and luminance efficiency were increased from 3160 cd/m2 and 0.24 cd/A of MEH-PPV-only
device to 6830 cd/m2 and 0.50 cd/A. Moreover, blending the bipolar copolyfluorenes with MEH-PPV
further improves the device efficiency, with the maximum luminance and luminance efficiency being
significantly enhanced up to 11090 cd/m2 and 0.56 cd/A (ca. 0.4 wt% of bipolar residue), respectively.
Our results demonstrate the efficacy of the bipolar copolyfluorenes and model compound M0 in
enhancing emission efficiency of MEH-PPV.
crystallization-induced degradation and thermal breakdown
Introduction
during device operation might readily happen due to their low
glass transition temperature (Tg).13–15 Therefore, the develop-
ment of charge injection/transport materials with high Tg is
essential to improve PLEDs’ performance. Several research
groups prepared high-Tg polymers carrying electron- and/or
hole-transporting groups in an attempt to improve device
performance.16–22 Kim et al. synthesized high-Tg polyquinolines as
a hole-blocking/electron-transporting layer to improve emission
efficiency.18b Zheng et al. synthesized segmented copolymers con-
taining oxadiazole as an electron-transporting layer to enhance
performance of PPV-based device; the luminance and external
quantum efficiency were 2400 cd/m2 and 0.094% respectively.19c
Alam et al. synthesized polybenzobisazoles containing electron-
transporting groups with the same aim.20a Huang et al. used
poly[(9,9-bis(30-(N,N-dimethylamino)propyl)-2,7-fluorene)-alt-2,7-
(9,9-dioctyl fluorene)] as an electron-injection layer to improve
performance of MEH-PPV-based device.20b Kulkarni et al.
blended poly(9,9-dioctyl fluorene) with thermally stable poly(vinyl
diphenylquinoline) to overcome the poor spectral stability of blue-
emitting polyfluorenes.20d To our knowledge, few researches have
been focused on directly blending polymers containing bipolar
groups to improve emission efficiency of another light-emitting
polymer.
Since the first report on the electroluminescence of poly(p-phe-
nylenevinylene) (PPV) by Holmes et al. in 1990,1 polymeric light-
emitting diodes (PLEDs) have been studied extensively because
of their potential applications in large-area display and solid-
state lighting.2 Polymeric materials possess many advantages
over inorganic ones, such as good film-forming properties, facile
fabrication via spin-coating and ink-jet printing,3,4 and suscep-
tible to structural modification. Nevertheless, for most PPV
derivatives such as MEH-PPV, holes are more readily injected
and transported than electrons, leading to unbalanced carriers
injection when fabricated as PLEDs.5–9 Balanced and efficient
injection and transport for both carriers are required to attain
high device efficiency. Two methods have been attempted to
improve the device efficiency. One is to insert an additional
electron injection/transport layer between the emitter and the
cathode.10–12 Nevertheless, fabrication of the multilayer polymer
LEDs is usually a difficult task, since the emitting layer might be
re-dissolved during subsequent spin-coating of the electron
injection/transport layer. Single-layer devices are preferred from
the viewpoints of process simplicity and cost effectiveness.
Another method is to fabricate single-layer LEDs which blend
charge injection/transport molecules such as oxadiazole, 1,2,4-
triazole, triphenylamine or carbazole derivatives. However,
In this work, we synthesized a new bipolar monomer (M1) to
prepare copolyfluorenes (P1–P3) containing 2.1–8.2 mol%
pendant bipolar groups. The bipolar unit consists of electron-
transporting aromatic 1,2,4-triazole directly linked with hole-
transporting triphenylamine. The copolyfluorenes not only
enhance the injection of holes and electrons but also lessen
excimer formation under thermal annealing due to non-planar
structure of the bipolar units. Finally, the influence of bipolar
contents on device performance is investigated, using blends of
the copolyfluorenes and MEH-PPV as emitting layer. Both
Department of Chemical Engineering, National Cheng Kung University,
Tainan, Taiwan. E-mail: yunchen@mail.ncku.edu.tw
† Electronic supplementary information (ESI) available: Experimental
1
details and characterization of monomers (3 and 4), H NMR and 13C
NMR spectra of M1, TGA thermograms and DSC traces of PF and
P1–P3, Half-decay lifetime (t1/2) characteristics of PLEDs using blends
of MEH-PPV and P3 (0 and 5 wt%) as emitting layer, AFM images of
MEH-PPV-based and MEH-PPV/P3 (95/5)-based thin films before and
after thermal annealing. See DOI: 10.1039/c0jm00707b
7700 | J. Mater. Chem., 2010, 20, 7700–7709
This journal is ª The Royal Society of Chemistry 2010

View Article Online
maximum luminance and maximum luminance efficiency of the
blend devices are greatly enhanced up to 11090 cd/m2 and 0.56
cd/A, respectively. Consequently, copolyfluorenes with bipolar
moieties attached as pendant groups are applicable to enhance
emission efficiency of PPV-based polymers.
Experimental section
Measurements
Newly synthesized compounds were identified by 1H NMR
spectra, FT-IR spectra and elemental analysis (EA). 1H NMR and
13C NMR spectra were obtained on a Bruker AMX-400 MHz or
a AV500 MHz FT-NMR spectrometer with the chemical shifts
reported in ppm using tetramethylsilane (TMS) as an internal
standard. The FT-IR spectra were measured as KBr disk on
a Fourier transform infrared spectrometer, model 7850 from
Jasco. Elementalanalysiswas carriedoutonaHerausCHN-Rapid
elemental analyzer. The weight-average molecular weight (Mw)
andpolydispersity index(PDI)ofpolymers weremeasuredwithgel
permeation chromatography (GPC) using chloroform (CHCl3) as
eluent and monodisperse polystyrenes as calibration standards.
Thermogravimetric analysis (TGA) of polymers was performed
ꢀ
under nitrogen atmosphere at a heating rate of 20 C/min using
a PerkinElmer TGA-7 thermal analyzer. Thermal transition
properties of polymers were investigated using a differential
scanning calorimeter (DSC), PerkinElmer DSC 7, under nitrogen
atmosphere at a heating rate of 10 ꢀC/min. Absorption and pho-
toluminescence (PL) spectra were measured with a Jasco V-550
spectrophotometer and a Hitachi F-4500 fluorescence spectro-
photometer, respectively. Cyclic voltammograms were recorded
with a voltammetric analyzer (model CV-50W from Bioanalytical
Systems, Inc.) at room temperature under nitrogen atmosphere.
The measuring cell was made up of the concentration of low MW
compounds (TAZand TPA) was3 mg/1 mlCHCl3, while polymers
(PF, P1–P3) were coated on glassy carbon as working electrodes,
an Ag/AgCl electrode as the reference electrode, and a platinum
wire as the auxiliary electrode. The electrodes were immersed in
acetonitrile containing 0.1 M (n-Bu)4NClO4 as electrolyte. The
energy levels were calculated using ferrocene (FOC) as standard
(ꢁ4.8 eV with respect to vacuum level).23
Scheme 1 Synthesis of monomer M1 and structures of M0, TAZ and TPA
a comparative study, the model 3,4,5-triphenyl-4H-1,2,4-triazole
(TAZ) containing electron-transporting core was also prepared
from aniline and 1,2-bis(chloro(phenyl)methylene)hydrazine.
The synthesis of bipolar model compound M0 was reported
previously.22b
Synthesis of bipolar monomer (M1) (Scheme 1)
A mixture of 4 (0.87 g, 2 mmol) and 8 (0.52 g, 2 mmol) in 10 ml of
N,N-dimethylaniline was stirred at 135 ꢀC for 12 h under nitrogen
atmosphere. After adding aqueous solution of HCl (30 ml, 2 N),
the mixture was stirred for an additional 30 min. The precipitated
solid was collected by filtration, dried in vacuo, and recrystallized
from ethyl acetate to afford N-(4-(3-(2,5-dibromophenyl)-5-
phenyl-4H-1,2,4-triazol-4-yl)phenyl)-N-phenylbenzenamine (M1)
Materials
9,9-Dihexyl-2,7-dibromofluorene (Aldrich), 9,9-dihexylfluorene-
2,7-bis(trimethyleneborate) (Aldrich), Aliquat 336 (Alfa Aesar),
tetrakis(triphenylphosphine) palladium [Pd(PPh3)4] (Acros), methyl
2,5-dibromobenzoate (Acros), benzoyl chloride (Acros), phospho-
rous pentachloride (PCl5, Riedel-Dehaen), triphenylamine (TPA,
Acros) N,N-dimethylaniline (Acros), dimethyl sulfoxide
(DMSO, Tedia), toluene (Tedia), chloroform (CHCl3, Tedia)
and other solvents were of commercial sources and used without
purification. 1-((2,5-Dibromophenyl)chloromethylene)-2-(chlor-
o(phenyl)methylene) hydrazine (4) was prepared from 2,5-
dibromobenzohydrazide (2) by reacting consecutively with
benzoyl chloride and phosphorous pentachloride (PCl5) (Scheme
1). N0,N0-Diphenylbenzene-1,4-diamine (8) was synthesized via
direct reduction of N-(4-nitrophenyl)-N-phenylbenzenamine (7)
by hydrazine monohydrate, in which 7 was obtained from
1-fluoro-4-nitrobenzene (5) and diphenylamine (6). For
1
(58%). mp 178–180 ꢀC. H NMR (CDCl3, ppm): d 7.82–7.81 (d,
J ¼ 2.5 Hz, 1H, Ar-H), 7.67–7.62 (m, 2H, Ar-H), 7.49–7.42 (m,
5H, Ar-H), 7.33–7.30 (t, J ¼ 7.5 Hz, 4H, Ar-H), 7.17–7.15 (m, 2H,
Ar-H), 7.10–7.07 (t, J ¼ 7.5 Hz, 2H Ar-H), 6.98–6.97 (d, J ¼ 8.0
Hz, 4H, Ar-H), 6.86–6.83 (m, 2H, Ar-H). FTIR (KBr pellet,
cmꢁ1): n 3070, 2370, 1904, 1592, 1509 (C]N), 1485, 1462, 1333,
1264 (C–N), 1181, 1074 (C–Br), 860. Anal. Calcd. (%) for
C32H22Br2N4: C, 61.76; H, 3.56; N, 9.00. Found: C, 61.74; H,
3.58; N, 9.11.
Synthesis of bipolar copolyfluorenes (P1–P3) and poly(9,9-
dihexylfluorene) (PF) (Scheme 2)
Copolyfluorenes (P1–P3) containing pendant bipolar groups and
poly(9,9-dihexylfluorene) (PF) were prepared by the Suzuki coupling
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 7700–7709 | 7701

View Article Online
reaction using (PPh3)4Pd as the catalyst.24 For instance, stoichio-
metric amounts of 9,9-dihexylfluorene-2,7-bis(trimethyleneborate)
(9: 0.251 g, 0.50 mmol), 9,9-dihexyl-2,7-dibromofluorene (10: 0.222
g, 0.45 mmol), N-(4-(3-(2,5-dibromophenyl)-5-phenyl-4H-1,2,4-tri-
azol-4-yl)phenyl)-N-phenylbenzenamine (M1: 0.031 g, 0.05 mmol),
10 mg of (PPh3)4Pd and several drops of Aliquat 336 were added to
a mixture containing toluene (10 ml) and 2 M aqueous solution of
K2CO3 (8 ml). The mixture was refluxed for 48 h under nitrogen
atmosphere. Extra monomer 9 (16 mg, 0.03 mmol) was added to the
reaction mixture and stirred for additional 12 h. Finally, mono-
functional bromobenzene (10 mg, 0.07 mmol) was added and stirred
for 12 h to end-cap polymer chain with phenyl group. After cooling
to room temperature, the reaction mixture was precipitated from
methanol and distilled water (v/v ¼ 10/1). The precipitate was
collected by filtration, dissolved in chloroform and re-precipitated
from methanol twice. Then it was Soxhlet extracted with acetone for
48 h to remove trace oligomers and catalyst residues and then dried in
vacuo to give P2. 1H NMR (CDCl3, ppm): d 7.86–7.84 (m, 2H, Ar-
H), 7.71–7.67 (m, 4H, Ar-H), 7.59 (m, 4H, Ar-H), 7.52 (m, 4H, Ar-
H), 7.32 (m, 4H, Ar-H), 6.99 (m, 4H Ar-H), 2.13 (s, 4H, –CH2–), 1.26
(s, 4H, –CH2–), 1.14 (s, 12H, –CH2–), 0.80 (s, 6H, –CH3). FTIR (KBr
pellet, cmꢁ1): n 3032, 2926 (Ar-H), 2857, 1889, 1592, 1509 (C]N),
1455 (C]C), 1280 (C–N), 1112, 998, 891. Anal. Calcd. (%) for P2:C,
89.76; H, 9.41; N, 0.83. Found: C, 88.51; H, 9.26; N, 0.52.
hole-injection layer and dried at 150 ꢀC for 15 min. Emitting
layer was then deposited by spin-coating (1000 rpm) onto the
PEDOT:PSS layer from a polymer solution in 1,2-dichloroben-
zene (10 mg/ml). Finally, calcium and aluminum were consecu-
tively vacuum-deposited as cathode using a vacuum coater at
a pressure of 2 ꢂ 10ꢁ6 Torr. Luminance versus voltage and
current density versus voltage characteristics of the devices were
measured using a combination of a Keithley power source
(model 2400) and an Ocean Optics usb2000 fluorescence spec-
trophotometer. The optoelectronic measurements were con-
ducted in a glove-box filled with nitrogen.
Results and discussion
Characterization of bipolar monomer (M1) and copolyfluorenes
(P1–P3)
Scheme 1 illustrates the synthetic routes employed in the prep-
aration of bipolar monomer M1. N-(4-Nitrophenyl)-N-phenyl-
benzenamine (7)25 and N0, N0-diphenylbenzene-1,4-diamine (8)22b
were prepared according to the procedures reported previously.
The copolyfluorenes (P1–P3) were synthesized by the
Suzuki coupling reaction of 9,9-dihexylfluorene-2,7-bis-
(trimethyleneborate) (9) with functionalized dibromo aromatic
monomers (10 and M1) (Scheme 2), using Pd(PPh3)4 as the
reaction catalyst and Aliquat 336 as the phase-transfer catalyst.
The feed ratios of monomer M1 in the preparation of P1, P2 and
P3 were 3 mol%, 5 mol% and 10 mol% respectively. Extra
monomer 9 and monofunctional bromobenzene were added to
end-cap the polymer chain after the polymerization. Fig. S1
The synthetic procedures of PF, P1 and P3 were analogous to
those used in the preparation of P2, except with varied molar
feed ratios in 9, 10 and M1. PF: 9 (0.251 g, 0.50 mmol), 10 (0.246
g, 0.50 mmol). 1H NMR (CDCl3, ppm): d 7.85–7.83 (m, 2H, Ar-
H), 7.71–7.67 (m, 4H, Ar-H), 2.12 (s, 4H, –CH2–), 1.25 (s, 4H,
–CH2–), 1.13 (s, 12H, –CH2–), 0.84–0.78 (m, 6H, –CH3). FTIR
(KBr pellet, cmꢁ1): n 3032, 2926 (Ar-H), 2849, 2370, 1889, 1607,
1455 (C]C), 1409, 1371, 1250, 1089, 884. Anal. Calcd. (%) for
PF: C, 90.26; H, 9.74. Found: C, 88.37; H, 9.43. P1: 9 (0.251 g,
0.50 mmol), 10 (0.231 g, 0.47 mmol) and M1 (0.018 g, 0.03
1
(ESI†) shows the H NMR, 13C NMR, DEPT135, H–H COSY
and C–H HMQC spectra of M1. Assignments of each carbon
and proton were assisted by the two-dimensional NMR spectra
[Fig, S1 (d),(e), ESI†], which agree well with the proposed
molecular structure of M1. The actual molar percents of M1
residues in P1, P2 and P3 are 2.1, 3.3 and 8.2 mol%, respectively,
as estimated from the elemental analysis data. This indicates that
the bipolar monomer M1 is less reactive than 9,9-dihexyl-2,7-
dibromofluorene (10), which is probably due to steric hindrance
and electron-deficient characteristic of its bipolar moiety. The
1
mmol). H NMR (CDCl3, ppm): d 7.85–7.83 (m, 2H, Ar-H),
7.71–7.69 (m, 4H, Ar-H), 2.17–2.12 (m, 4H, –CH2–), 1.25 (s, 4H,
–CH2–), 1.13 (s, 12H, –CH2–), 0.83–0.78 (s, 6H, –CH3). FTIR
(KBr pellet, cmꢁ1): n 3032, 2926 (Ar-H), 2852, 1889, 1606, 1509
(C]N), 1455 (C]C), 1280 (C–N), 1114, 998, 891. Anal. Calcd.
(%) for P1: C, 89.96; H, 9.54; N, 0.50. Found: C, 88.90; H, 9.40;
N, 0.30. P3: 9 (0.251 g, 0.50 mmol), 10 (0.197 g, 0.40 mmol) and
1
M1 (0.062 g, 0.10 mmol). H NMR (CDCl3, ppm): d 7.85–7.84
(m, 2H, Ar-H), 7.71–7.67 (m, 5H, Ar-H), 7.59–7.49 (m, 7H, Ar-
H), 7.36–7.32 (m, 4H, Ar-H), 7.18–7.16 (m, 2H Ar-H), 7.13–7.11
(m, 2H, Ar-H), 2.12 (s, 4H, –CH2–), 1.25 (s, 4H, –CH2–), 1.13 (s,
12H, –CH2–), 0.79 (s, 6H, –CH3). FTIR (KBr pellet, cmꢁ1): n
3032, 2926 (Ar-H), 2857, 1889, 1592, 1509 (C]N), 1455 (C]C),
1280 (C–N), 1112, 998, 891. Anal. Calcd. (%) for P3: C, 89.28; H,
9.09; N, 1.63. Found: C, 88.16; H, 9.01; N, 1.14.
Fabrication of polymer light-emitting diodes
Double-layer polymer light-emitting diodes (ITO/PEDOT:PSS/
polymer/Ca/Al) were fabricated to investigate their optoelec-
tronic characteristics. Transparent indium tin oxide (ITO) glass
was successively cleaned with neutraler reiniger/deionized water
(1:3 ¼ v/v) mixture, deionized water, acetone and 2-propanol,
and then dried in vacuo overnight. Aqueous dispersion of
PEDOT:PSS was spin-coated on top of the cleaned ITO glass as
Scheme 2 Synthesis of bipolar copolyfluorenes (P1–P3) and poly-
fluorene (PF).
7702 | J. Mater. Chem., 2010, 20, 7700–7709
This journal is ª The Royal Society of Chemistry 2010

View Article Online
Table 1 Polymerization results and characterization of polymers
a
a
Yield Mn
Mw
(ꢂ104) (ꢂ104) PDIa Tg (ꢀC) Td (ꢀC) yc (%)
b
Polymer (%)
PF
P1
P2
P3
63
64
53
54
2.05
2.50
0.66
0.74
4.75
4.70
1.41
1.40
2.32
1.89
2.13
1.89
88
91
101
103
416
422
428
439
0
2.1
3.3
8.2
a
Mn, Mw, and PDI of the polymers were determined by gel permeation
chromatography using polystyrene standards. The temperature at 5
b
c
wt% loss under nitrogen atmosphere. The molar fractions of M1
residues were estimated from elemental analysis.
copolyfluorenes are readily soluble in common organic solvents
such as toluene, chloroform, chlorobenzene and 1,2-dichloro-
benzene. Their weight-average molecular weights (Mw) and
polydispersity indexes (PDI) are in the range 1.40–4.75 ꢂ 104 and
1.89–2.32 (Table 1) respectively, as determined by gel permeation
chromatography using monodisperse polystyrenes as calibration
standards.
Thermal decomposition temperatures (Td) (at 5 wt% loss) and
glass transition temperatures (Tg) of the copolyfluorenes were
evaluated with thermal gravimetric analysis (TGA) and differential
scanning calorimetry (DSC) respectively. The residual weights of
ꢀ
P1–P3 and PF (at 800 C) are above 50% under nitrogen atmo-
sphere (Fig. S2 in ESI†). As shown in Table 1, the thermal
decomposition temperatures (Td: 422–439 ꢀC) of the copoly-
fluorenes (P1–P3) are higher than that of PF (Td: 416 ꢀC).
Moreover, the copolꢀyfluorenes reveal higher glass-transition
Fig. 1 Absorption and PL spectra of PF and P1–P3: (a) in chloroform (1
ꢂ 10ꢁ5 M) (lex ¼ 389 nm for PF, 388 nm for P1, 384 nm for P2 and 383
nm for P3), (b) in film state (lex ¼ 391 nm for PF, 389 nm for P1, P2 and
387 nm for P3).
ꢀ
temperatures (91–103 C) than PF (88 C) and the Tgs increase
gradually increasing M1 residues in the main chain. This is
attributable to the rigid and non-planar structure of M1 residues
which not only raises chain rigidity but also restricts chain mobility
of the copolyfluorenes.22b,26 Accordingly, the copolyfluorenes
exhibit better thermal stability than PF; i.e., their thermal
decomposition (422–439 ꢀC) and glass-transition temperatures
(91–103 ꢀC) are higher than those of PF (416 ꢀC, 88 ꢀC).
Table 2 Optical properties of model M0 and polymers
UV-vis
lmax
solution
(nm)
UV-vis
lmax
film
PL
lmax
PL
a
lmax
filmb
(nm)
Molecule/
polymer
solutionb
(nm)
c
(nm)
FPL
M0
PF
P1
P2
P3
271,292
389
292,388
292,384
292,383
—
391
293,389
293,389
292,387
386
—
—
419,441 s
419,441 s
419,441 s
418,441 s
425,449 s
425,449 s
424,449 s
424,449 s
0.83
0.79
0.71
0.68
Photophysical properties of bipolar copolyfluorenes (P1–P3)
Fig. 1 illustrates the absorption and photoluminescence (PL)
spectra of PF and P1–P3 in CHCl3 and as films spin-coated from
CHCl3 solutions, with the characteristic optical data summarized
in Table 2. The absorption maximum and main emission peak of
the bipolar model M0 (Scheme 1) in CHCl3 appear at ca. 292 nm
(with a shoulder at 271 nm) and 386 nm, respectively.22b
However, the absorption maxima of P1–P3 in CHCl3 locate at
388, 384 and 383 nm, with only small shorter-wavelength
absorptions at ca. 292 nm. The major absorption of P1–P3 (383–
388 nm) can be attributed to the p–p* transitions of their
conjugated backbone, whereas the shorter-wavelength absorp-
tion (ca. 292 nm) is definitely originated from M1 residues. The
major absorption shifts slightly from 389 nm (PF) to 383 nm (P3)
because the conjugation of polyfluorene main chain is inter-
rupted by the bipolar moieties. This can be substantiated by the
existence of twist angles (72ꢀ and 16ꢀ) between adjacent benzene
rings of fluorene and bipolar M1 units (Fig. 2), simulated by
minimizing energy via semi-empirical MNDO calculations in the
gas states.27 Although the twist angles may be slightly different
a
b
c
In chloroform (1 ꢂ 10ꢁ5 M). s: Wavelength of shoulder. FPL
:
Determined in CHCl3, relative to quinine sulfate in 1N H2SO4(aq) at
a concentration of 10ꢁ5 M (FPL ¼ 0.55).
under the measuring conditions, they certainly reduce conjuga-
tion length of the copolyfluorenes. Solid state absorption
maxima of PF and P1–P3 are slightly red-shifted (1–5 nm)
relative to solution state ones due to aggregate formation via
intra- or inter-chain interaction. P1–P3 exhibit PL peaks at 418–
419 nm which are very close to 419 nm of PF. Moreover, the PL
spectra of P1–P3 were almost the same whether they were excited
with 292 nm (absorption maximum of M0) or 383 nm, suggesting
€
efficient Forster energy transfer from M1 residues to fluorene
segments under the photo-excitation. Partial overlap between
emission spectrum of M1 residues and absorption spectrum of
PF contributes greatly to the energy transfer.22b,26,28 It is note-
worthy that the emission spectrum of PF partially overlaps with
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 7700–7709 | 7703

View Article Online
Fig. 2 Optimized geometry of a bipolar unit connected with two fluo-
renes obtained from semi-empirical MNDO calculation.
the absorption spectrum of MEH-PPV, suggesting that energy
transfer from PF to MEH-PPV can be expected.22b Based on this
characteristic we fabricated efficient PLEDs using blends of
MEH-PPV and P1–P3 as emitting layers and will be discussed
later.
Fig. 3 PL spectra of PF and P1–P3 after annealing in vacuum at 150 ꢀC
for 24 h.
Relative PL quantum yields (FPL) of PF, P1, P2 and P3 in
chloroform (at a concentration of 10ꢁ5 M) are 0.83, 0.79, 0.71 and
0.68, respectively. The FPLs of the copolyfluorenes are slightly
lower than that of PF, probably caused by torsion-induced non-
radiative deactivation occurred at non-planar triphenylamine
groups.22b,29 Polyfluorenes usually show undesirable green emis-
sion after thermal annealing or under device operation, which has
been attributed to enhanced intermolecular interaction.30 To
examine color stability, PF and P1–P3 films were thermally
annealed at 150 ꢀC for 24 h to observe their PL spectra variations.
As shown in Fig. 3, the green emission appears at ca. 500–600 nm
in PF film after the thermal treatment; however, it is significantly
suppressed in P2 and P3 films. This result suggests that the non-
planar bipolar units are effective in suppressing the intermolecular
interaction in the copolyfluorenes.
Electrochemical investigations
Cyclic voltammetry (CV) was employed to investigate the elec-
trochemical properties of PF, P1–P3, 3,4,5-triphenyl-4H-1,2,4-
triazole (TAZ) and triphenylamine (TPA). Their HOMO and
LUMO energy levels were estimated by the equations: EHOMO
(eV) ¼ ꢁ(Eox,FOC + 4.8) and ELUMO (eV) ¼ ꢁ(Ered,FOC + 4.8),
where Eox,FOC and Ered,FOC are the onset oxidation and onset
reduction potentials, respectively, with respect to the ferrocene/
ferrocenium couple. The cyclic voltammograms are shown in
Fig. 4, with the representative electrochemical data summarized
in Table 3. Model M0 would emit green light corresponding to its
band-gap (2.53 eV), but its main emission peak (386 nm) deviates
from green emission significantly. In order to elucidate this
abnormal phenomena, both LUMO and HOMO levels of 3,4,5-
triphenyl-4H-1,2,4-triazole (TAZ) and triphenylamine (TPA)
were estimated from their cyclic voltammograms. The estimated
LUMO levels of M0, TAZ and TPA are ꢁ2.77, ꢁ2.71 and ꢁ2.12
eV, and the estimated HOMO levels are ꢁ5.3, ꢁ6.12 and ꢁ5.27
eV, respectively. The LUMO and HOMO levels of M0 are close
to LUMO level of TAZ and HOMO level of TPA, respectively.
This indicates that the LUMO level of M0 originates from
Fig. 4 Cyclic voltammograms of (a) TAZ and TPA at a concentration of
3 mg/1 mL CHCl3 (b) PF and P1–P3 coated on glassy carbon working
electrode (scan rate: 100 mV/s).
LUMO level of TAZ, whereas the HOMO level from TPA.
Under cyclic voltammetric analysis of bipolar M0, the oxidation
and reduction start from triphenylamine and aromatic triazole
groups, respectively. Consequently, the electrochemically-deter-
el
mined band-gap (Eg ) of M0 is roughly the energy difference
between LUMO level of TAZ and HOMO level of TPA. In this
7704 | J. Mater. Chem., 2010, 20, 7700–7709
This journal is ª The Royal Society of Chemistry 2010

View Article Online
Table 3 Electrochemical properties of model M0, TAZ, TPA and
polymers (PF, P1–P3)
Performance enhancement of MEH-PPV by model compound
M0 and copolyfluorenes
el
ꢀ
pt
Molecule/
polymer
E
oxvs.
E
red vs.
EHOMO ELUMO Eg
(eV)b
Eg
Polymer light-emitting devices using P1–P3 as emitting layer
[ITO/PEDOT:PSS/P1–P3/Ca(50 nm)/Al(100 nm)] were fabri-
cated to study their current density and luminance (versus bias)
characteristics. The turn-on voltages of P1–P3 devices increased
from 5.3 to 8.1 V with increasing M1 residues (2.1 / 8.2 mol%).
And the maximum luminance and maximum luminance effi-
ciency of P1–P3 devices decrease from 760 to 55 cd/m2 and from
0.05 to 0.01 cd/A respectively. The performance degradation at
high M1 contents is probably due to charges trapping in the
bipolar moieties that reduce charge recombination in poly-
fluorene segments.32,33 According to the energy band diagrams
depicted in Fig. 5, both LUMO and HOMO levels of model M0
are lower than those of MEH-PPV, suggesting that the M1
residues will promote electron transport and block hole transport
in MEH-PPV device. To elucidate this speculation, blends of
MEH-PPV with M0, TAZ or TPA were spin-coated as emitting
layers to fabricate double-layer light-emitting devices [ITO/
PEDOT:PSS/(MEH-PPV + M0, TAZ or TPA)/Ca(50 nm)/Al
(100 nm)] and their device performance investigated. Fig. 6
shows the current density and luminance versus bias character-
istics of the blend devices, with the representative performance
data summarized in Table 4. At 2 wt% addition the turn-on
FOC (V)a FOC (V)a (eV)b
(eV)c (eV) d
M0
TAZ
TPA
PF
P1
P2
0.5
ꢁ2.03
ꢁ2.09
ꢁ2.68
ꢁ2.21
ꢁ2.20
ꢁ2.18
ꢁ2.14
ꢁ5.3
ꢁ2.77
ꢁ2.71
ꢁ2.12
ꢁ2.59
ꢁ2.60
ꢁ2.62
ꢁ2.66
2.53
3.41
3.15
3.04
3.01
2.96
2.86
—
—
—
1.32
0.47
0.83
0.81
0.78
0.72
ꢁ6.12
ꢁ5.27
ꢁ5.63
ꢁ5.61
ꢁ5.58
ꢁ5.52
2.91
2.91
2.90
2.90
P3
a
b
EFOC ¼ 0.48 V vs. Ag/AgCl. EHOMO ¼ ꢁ(Eox,
+ 4.8) eV. Electrochemical band gap:
FOC
d
opt
+ 4.8) eV;
FOC
c
ELUMO
el
¼
ꢁ(Ered,
Eg ¼ LUMO – HOMO. Optical band gap: Eg ¼ hc/lonset
.
Fig. 5 Energy band diagrams of M0, TAZ, TPA, PF, P1–P3 and MEH-
PPV.
study, copolyfluorenes P1–P3 contain 2.1–8.2 mol% of bipolar
M1 residues derived from electron-transporting triazole and hole-
transporting triphenylamine. The estimated HOMO levels of PF,
P1–P3 are ꢁ5.63, ꢁ5.61, ꢁ5.58 and ꢁ5.52 eV, whereas the esti-
mated LUMO energy levels are ꢁ2.59, ꢁ2.60, ꢁ2.62 and ꢁ2.66
eV, respectively. The HOMO levels of P1–P3 are raised gradually
from ꢁ5.63 eV to ꢁ5.52 eV with an increase in M1 residues (2.1
mol%/8.2 mol%) (Fig. 5), suggesting that their hole affinity is in
the order of P3 > P2 > P1. On the contrary, the LUMO levels of
P1–P3 are lowered slightly (ꢁ2.59 eV/ꢁ2.66 eV) with increasing
M1 residues, meaning that their electron injection ability is
improved in the order of P3 > P2 > P1. Accordingly, both hole
and electron affinity of P1–P3 is gradually enhanced with
increasing M1 residues, resulting in improved carrier injection.
Both LUMO and HOMO levels of model M0 (ꢁ2.77 eV and ꢁ5.3
eV) are lower than those of MEH-PPV (ꢁ2.7 eV and ꢁ5.02 eV)
because of the electron-withdrawing aromatic 1,2,4-triazole
moiety.21b,22b,31 Therefore, blending P1–P3 with the conventional
MEH-PPV would increase its electron affinity and, moreover, at
the same time reduces its hole affinity. This might be an effective
way to improve balance in charges injection and transport for the
conventional MEH-PPV, whose holes are believed to be more
readily injected and transported.
Fig. 6 (a) Current density versus bias and (b) luminance versus bias
characteristics of PLEDs using blends of MEH-PPV and M0, TAZ or
TPA (2–4 wt%) as emitting layer. Device configuration: ITO/
PEDOT:PSS/MEH-PPV + M0, TAZ or TPA (90–110 nm)/Ca(50 nm)/
Al(100 nm).
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 7700–7709 | 7705

View Article Online
Table 4 Optoelectronic performance of polymer light-emitting diodesa
Maximum luminance
efficiency (cd/A)
Maximum
luminance (cd/m2)
Emitting-layer compositionb
Turn on voltage (V)c
CIE coordinates (x, y)d
MEH-PPV
3.9
3.6
4.5
3.8
4.1
3.7
3.9
3.3
3.1
3.1
3.3
3.4
3.4
3.4
0.24
0.50
0.42
0.25
0.29
0.40
0.30
0.40
0.50
0.56
0.56
0.39
0.38
0.39
3160
6830
4120
3240
3180
4600
3550
6370
8930
10550
11090
7240
6790
5840
(0.54, 0.46)
(0.54, 0.46)
(0.56, 0.44)
(0.54, 0.46)
(0.55, 0.45)
(0.54, 0.46)
(0.53, 0.46)
(0.53, 0.47)
(0.52, 0.47)
(0.52, 0.48)
(0.52, 0.48)
(0.52, 0.47)
(0.53, 0.47)
(0.54, 0.46)
MEH-PPV/M0 (98/2)
MEH-PPV/M0 (96/4)
MEH-PPV/TAZ (98/2)
MEH-PPV/TAZ (96/4)
MEH-PPV/TPA (98/2)
MEH-PPV/TPA (96/4)
MEH-PPV/P1 (90/10)
MEH-PPV/P2 (94/6)
MEH-PPV/P3 (97.5/2.5)
MEH-PPV/P3 (95/5)
MEH-PPV/P3 (92.5/7.5)
MEH-PPV/P3 (90/10)
MEH-PPV/P3 (87.5/12.5)
a
Device structure: ITO/PEDOT:PSS/emitting layer/Ca/Al. b The weight ratios are given in parentheses. c The voltage required for the luminance of 10
cd/m2. d The CIE coordinates at maximum luminance.
voltages are slightly reduced to 3.6 V–3.8 V, from 3.9 V of a neat
MEH-PPV device. Moreover, maximum luminance and
maximum luminance efficiency of the blend device containing 2
wt% M0 are enhanced to 6830 cd/m2 and 0.50 cd/A respectively,
which are superior to 3160 cd/m2 and 0.24 cd/A of the neat MEH-
PPV device. In addition, its performance is also better than the
other two blend devices. This is attributable to more balanced
charges injection and transport in the presence of bipolar M0
compound. However, the performance are deteriorated to 4120
cd/m2 and 0.42 cd/A at 4 wt% M0. This performance degradation
is probably caused by the hole-blocking effect of M0 because of
its low HOMO level (ꢁ5.3 eV). It is evident that a small amount
of the bipolar model M0 is effective in improving the perfor-
mance of conventional MEH-PPV.
because fluorene segment possesses lower HOMO (ꢁ5.63 eV)
and higher LUMO levels (ꢁ2.59 eV) than MEH-PPV (ꢁ5.02 eV,
ꢁ2.70 eV), which reduces injection and transport of both charges
in the blend and the reduction increases with increasing contents.
The device performance depends on contents of bipolar units
as well. Using P3-based blend device as an example, their current
MEH-PPV was blended with bipolar P1–P3 to solve its
common defect of unbalanced charges injection and transport.
Double-layer polymer light-emitting diodes [ITO/PEDOT:PSS/
(MEH-PPV + P1, P2 or P3)/Ca(50 nm)/Al (100 nm)] were
fabricated to investigate their device performances. The weight
ratios of MEH-PPV to the copolyfluorenes (MEH-PPV/P1,
MEH-PPV/P2 and MEH-PPV/P3) were controlled at 90/10, 94/6
and 97.5/2.5 wt%, respectively, to adjust the ratios of bipolar
units (ca. 0.2 wt%). Fig. 7 shows the current density and lumi-
nance versus bias characteristics of the blend devices, with the
representative data summarized in Table 4. The blend devices
exhibit lower turn-on voltages (3.1–3.3 V) than MEH-PPV
device (3.9 V). Moreover, the maximum luminance and
maximum luminance efficiency of the blend devices are signifi-
cantly enhanced to 6370–10550 cd/m2 and 0.40–0.55 cd/A,
respectively, when compared with 3160 cd/m2 and 0.24 cd/A of
the MEH-PPV device. Apparently the presence of bipolar M1
residues in emitting layer results in more balanced charges
injection under device operation. The device performance
depends mainly on the contents of fluorene segments (from
polyfluorene and copolyfluorenes). For instance, the P3-based
blend device reveals the best EL performance probably due to the
least content of fluorene segments (ca. 2.3 wt%). Increasing
percents of fluorene segments in P1- and P2-based devices (ca.
9.8 wt% and 5.8 wt%) leads to degraded performance. This is
Fig. 7 (a) Current density versus bias and (b) luminance versus bias
characteristics of PLEDs using blends of MEH-PPV and P1–P3 (2.5–10
wt%) as emitting layer. Device configuration: ITO/PEDOT:PSS/MEH-
PPV + P1, P2 or P3 (90–110 nm)/Ca(50 nm)/Al(100 nm).
7706 | J. Mater. Chem., 2010, 20, 7700–7709
This journal is ª The Royal Society of Chemistry 2010

View Article Online
density and luminance versus bias are shown in Fig. 8. The turn-
on voltages (3.1–3.4 V) are lower than that of MEH-PPV-based
device (3.9 V) due to improved electron injection and transport.
However, the turn-on voltages increase slightly from 3.1 V to 3.4
V with increasing P3 contents (2.5 wt%/12.5 wt%). The elec-
tron injection enhancement seems inferior to hole-blocking effect
so that extra bias is needed to turn on the devices.22b,34 To
elucidate the hole-blocking effect of P3, hole-only devices [ITO/
PEDOT:PSS/(MEH-PPV+ P3)/Au(20 nm)/Al (100 nm)] were
fabricated to investigate their current density versus bias char-
acteristics. As shown in Fig. 9, the curve shifts horizontally to
higher bias as P3’s ratio is increased from 0 to 5 wt%, indicating
diminished current density under the same bias. For instance, at
a bias of 8.0 V the current density decreases from 980 mA/cm2 to
570 mA/cm2, which is obviously due to increased hole-blocking
at high P3 contents. The maximum luminance and the maximum
luminance efficiency of the device are enhanced from 3160 cd/m2
and 0.24 cd/A to 11090 cd/m2 and 0.56 cd/A, respectively, as the
weight ratio of P3 is increased from 0 to 5%. This performance is
a little inferior to that obtained for our previous copolyfluorene
with 11.2 mol% bipolar groups in the main chain.22b The EL
performance difference is probably due to their different LUMO
(ꢁ2.66 eV versus ꢁ2.68 eV) and HOMO energy levels (ꢁ5.52 eV
versus ꢁ5.49 eV), resulting from different positions and contents
of the bipolar groups (pendant group versus main chain, 8.2
Fig. 9 Current density versus bias characteristics of the hole-only
devices using blends of MEH-PPV and P3 (0–5 wt%) as emitting layer.
Device structure: ITO/PEDOT:PSS/MEH-PPV + P3 (90–110 nm)/Au(20
nm)/Al(100 nm).
Fig. 10 SEM micrographs of blend films (ꢂ20000): (a) MEH-PPV/P3 ¼
95/5 (b) MEH-PPV/P3 ¼ 87.5/12.5.
mol% versus 11.2 mol%, respectively). Accordingly, the slightly
reduced EL device performance is mainly due to a less balanced
charge recombination than the previous copolyfluorene. In this
study, however, further increase in P3 contents leads to quick
degradation in maximum luminance and maximum current
efficiency; i.e., diminish to 5840 cd/m2 and 0.39 cd/A at MEH-
PPV/P3 ¼ 87.5/12.5. The results of the morphology investigation
can reasonably explain this performance reversion at higher P3
contents. As shown in Fig. 10, the SEM micrograph shows
a homogeneous surface morphology at low P3 weight ratio
(MEH-PPV/P3 ¼ 95/5). However, it becomes obviously micro-
phase-separated at high P3 ratio (MEH-PPV/P3 ¼ 87.5/12.5).
Therefore, the quick performance degradation of the blend
device is attributable to the phase separation formed during film
preparation. The operational stability of blend-based (MEH-
PPV/P3 ¼ 95/5) and MEH-PPV-based devices are shown in
Fig. S3 (ESI†). The half-decay lifetimes (t1/2) of MEH-PPV/P3
(95/5) and neat MEH-PPV device were about 93 min (L0 ¼ 668
cd/m2) and 95 min (L0 ¼ 855 cd/m2), respectively. For driving at
L0 ¼ 100 cd/m2, the estimated half-decay lifetimes (t1/2
)
35,36 of the
blend- and MEH-PPV-based devices were about 620 min and
810 min respectively, indicating that the blend device is slightly
inferior to MEH-PPV device in terms of operational stability.
The results of surface morphology investigation (Fig. S4: AFM
images†) after thermal annealing (90 ꢀC, 24 h) are also consistent
with the devices’ half-decay lifetimes; i.e., the blend film
Fig. 8 (a) Current density versus bias and (b) luminance versus bias
characteristics of PLEDs using blends of MEH-PPV and P3 (2.5–12.5
wt%) as emitting layer. Device structure: ITO/PEDOT:PSS/MEH-PPV +
P3 (90–110 nm)/Ca(50 nm)/Al(100 nm).
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 7700–7709 | 7707

View Article Online
maximum luminance efficiency: 0.56 cd/A). Current results
indicate that the copolyfluorenes containing bipolar moieties are
promising additives in greatly improving device performance of
MEH-PPV and other conjugated polymers.
Acknowledgements
We thank the National Science Council of the Republic of China
for financial aid through project NSC 98-2221-E-006-002-MY3.
References
1 J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. Marks,
K. Mackay, R. H. Friend, P. L. Burn and A. B. Holmes, Nature,
1990, 347, 539.
2 (a) G. Grem, G. Leditzky, B. Ullrich and G. Leising, Adv. Mater.,
1992, 4, 36; (b) Q. Pei and Y. Yang, J. Am. Chem. Soc., 1996, 118,
7416; (c) M. R. Andersson, O. Thomas, W. Mammo, M. Svensson,
Fig. 11 Emission spectra of PLEDs using blends of MEH-PPV and P1–
P3 as emitting layer. Device configuration: ITO/PEDOT:PSS/MEH-PPV
+ P1, P2 or P3 (90–110 nm)/Ca(50 nm)/Al(100 nm).
€
M. Theander and O. Inganas, J. Mater. Chem., 1999, 9, 1933.
3 (a) M. Aguiar, F. E. Karasz and L. Akcelrud, Macromolecules, 1995,
28, 4598; (b) E. S. Kolb, R. A. Gaudiana and P. G. Mehta,
Macromolecules, 1996, 29, 2359; (c) A. Kraft, A. C. Grimsdale and
A. B. Holmes, Angew. Chem., Int. Ed., 1998, 37, 402.
4 (a) R. H. Friend, R. W. Gymer, A. B. Holmes, J. H. Burroughes,
R. N. Marks, C. Taliani, D. D. C. Bradley, D. A. Dos Santos,
demonstrated higher surface roughness before and after the
thermal annealing (1.01 nm / 1.46 nm) than MEH-PPV (0.93
nm / 1.21 nm).
ꢀ
€
J. L. Bredas and M. Logdlund, Nature, 1999, 397, 121; (b)
M. T. Bernius, M. Inbasekaran, J. O’Brien and W. Wu, Adv.
Mater., 2000, 12, 1737; (c) L. Akcelrud, Prog. Polym. Sci., 2003, 28,
875.
However, at low P3 content (MEH-PPV/P3 ¼ 95/5) the device
performance is effectively enhanced. As shown in Fig. 11, the
electroluminescent emission of all devices is exclusively origi-
nated from MEH-PPV, with full width at half-maxima (fwhm)
being ca. 80 nm. The 1931 CIE coordinates (x, y) of the EL
emission only shift slightly from (0.54, 0.46) of MEH-PPV-based
device to (0.52, 0.48) of blend devices (MEH-PPV/P3 ¼ 97.5/2.5,
MEH-PPV/P3 ¼ 95/5). Furthermore, the excimer emission of
MEH-PPV at ca. 625 nm is also diminished due probably to
reduced aggregation after blending with the copolyfluorenes.
Therefore, slight amount of copolyfluorenes P1–P3 containing
bipolar groups (M1 residues) is effective in enhancing device
performance of the conventional MEH-PPV.
5 (a) T. Takiguchi, D. H. Park, H. Ueno, K. Yoshino and R. Sugimoto,
Synth. Met., 1987, 17, 657; (b) D. Braun and A. J. Heeger, Appl. Phys.
Lett., 1991, 58, 1982.
6 (a) A. Kraft, P. L. Burn, A. B. Holmes, D. D. C. Bradley, R. H. Friend
and J. H. F. Martens, Synth. Met., 1993, 57, 4163; (b) H. Antomiadis,
M. A. Abkowitz and B. R. Hsieh, Appl. Phys. Lett., 1994, 65, 2030.
€
7 (a) H. Meyer, D. Haarer, H. Naarmann and H. H. Horhold, Phys.
Rev. B: Condens. Matter, 1995, 52, 2587; (b) P. W. M. Blom,
M. J. M. de John and J. J. M. Vleggaar, Appl. Phys. Lett., 1996, 68,
3308.
8 (a) Y. Yang, Q. Pei and A. J. Heeger, J. Appl. Phys., 1996, 79, 934; (b)
Y. Z. Wang, D. D. Gebler, D. J. Spry, D. K. Fu, T. M. Swager,
A. G. MacDiarmid and A. J. Epstein, IEEE Trans. Electron
Devices, 1997, 44, 1263.
9 (a) X.-C. Li, A. C. Grimsdale, R. Cervini, A. B. Holmes,
€
S. C. Moratti, T. M. Yong, J. Gruner and R. H. Friend, ACS
Conclusions
Symp. Ser., 1997, 672, 322; (b) H. Meng, W. L. Yu and W. Huang,
Macromolecules, 1999, 32, 8841; (c) H. C. F. Martens,
J. N. Huiberts and P. W. M. Blom, Appl. Phys. Lett., 2000, 77, 1852.
10 C. Adachi, T. Tsutsui and S. Saito, Appl. Phys. Lett., 1989, 55, 1489.
11 (a) A. R. Brown, D. D. C. Bradley, J. H. Burroughes, R. H. Friend,
N. C. Greenham, P. L. Burn, A. B. Holmes and A. Kraft, Appl.
Phys. Lett., 1992, 61, 2793; (b) N. C. Greenham, S. C. Moratti,
D. D. C. Bradley, R. H. Friend and A. B. Holmes, Nature, 1993,
365, 628.
12 (a) I. D. Parker, Q. Pei and M. Marrocco, Appl. Phys. Lett., 1994, 65,
1272; (b) S. Son, A. Dodabalapur, A. J. Lovinger and M. E. Galvin,
Science, 1995, 269, 376.
13 M. Strukelj, F. Papadimitrakopoulos, T. M. Miller and
L. J. Rothberg, Science, 1995, 267, 1969.
We have successfully synthesized copolyfluorenes (P1–P3) con-
taining pendant bipolar groups (M1 residues: 2.1, 3.3 and 8.2
mol%). The copolyfluorenes were soluble in common organic
solvents and thermally stablꢀe (thermal decomposition tempera-
ture at 5 wt% loss: 422–439 C). The PL spectra of P1–P3 were
€
identical to that of PF, due to efficient Forster energy transfer.
Estimated LUMO and HOMO levels of model M0 were ꢁ5.3 eV
and ꢁ2.77 eV, which are mainly attributed to LUMO levels of
TAZ and HOMO levels of TPA, respectively. The HOMO levels
of PF and P1–P3 were raised gradually from ꢁ5.63 eV to ꢁ5.52
eV with increasing bipolar residues (from 0 to 8.2 mol%), while
their LUMO levels lowered slightly from ꢁ2.59 eV to ꢁ2.66 eV.
Blending bipolar M0 with MEH-PPV enhanced the device effi-
ciency (maximum luminance: 6830 cd/m2, maximum luminance
efficiency: 0.50 cd/A). Furthermore, blending the copoly-
fluorenes with MEH-PPV resulted in further significant perfor-
mance enhancement due to improved carriers injection and
transport. P3-based blend device (MEH-PPV/P3 ¼ 95/5) showed
the best performance (maximum luminance: 11090 cd/m2,
14 Z. Peng, Z. Bao and M. E. Galvin, Chem. Mater., 1998, 10, 2086.
15 X. Zhou, J. He, L. S. Liao, M. Lu, X. M. Ding, X. Y. Hou,
X. M. Zhang, X. Q. He and S. T. Lee, Adv. Mater., 2000, 12, 265.
16 E. S. Kolb, R. A. Gaudiana and P. G. Mehta, Macromolecules, 1996,
29, 2359.
€
17 M. Greczmiel, P. Strohriegl, M. Meier and W. Brutting,
Macromolecules, 1997, 30, 6042.
18 (a) S. Liu, X. Jiang, H. Ma, M. S. Liu and K.-Y. Jen, Macromolecules,
2000, 33, 3514; (b) J. L. Kim, J. K. Kim, H. N. Cho, D. Y. Kim,
C. Y. Kim and S. Hong, Macromolecules, 2000, 33, 5880; (c)
B.-J. Jung, J.-I. Lee, H. Y. Chu, L.-M. Do and H.-K. Shim,
Macromolecules, 2002, 35, 2282.
7708 | J. Mater. Chem., 2010, 20, 7700–7709
This journal is ª The Royal Society of Chemistry 2010

View Article Online
19 (a) E. Bellmann, S. E. Shaheen, R. H. Grubbs, S. R. Marder,
B. Kippelen and N. Peyghambarian, Chem. Mater., 1999, 11, 399;
(b) X. Jiang, R. A. Register, K. A. Killeen, M. E. Thompson,
F. Pschenitzka and J. C. Sturm, Chem. Mater., 2000, 12, 2542; (c)
26 (a) C.-F. Shu, R. Dodda, F.-I. Wu, M. S. Liu and A. K.-Y. Jen,
Macromolecules, 2003, 36, 6698; (b) F.-I. Wu, P.-I. Shih, C.-F. Shu,
Y.-L. Tung and Y. Chi, Macromolecules, 2005, 38, 9028.
27 MNDO semi-empirical calculations were performed with Gaussian R
98W and Gauss View 3.09 by Gaussian, Inc.
€
M. Zheng, L. Ding, E. E. Gurel, P. M. Lahti and F. E. Karasz,
Macromolecules, 2001, 34, 4124.
28 (a) V. P. Barberis and J. A. Mikroyannidis, J. Polym. Sci., Part A:
Polym. Chem., 2006, 44, 3556; (b) C.-H. Chien, P.-I. Shih, F.-I. Wu,
C.-F. Shu and Y. Chi, J. Polym. Sci., Part A: Polym. Chem., 2007,
45, 2073; (c) M.-C. Yuan, P.-I. Shih, C.-H. Chien and C.-F. Shu,
J. Polym. Sci., Part A: Polym. Chem., 2007, 45, 2925.
29 D. Oelkrug, A. Tompert, J. Gierschner, H.-J. Egelhaaf, M. Hanack,
M. Hohloch and E. Steinhuber, J. Phys. Chem. B, 1998, 102, 1902.
30 (a) J. Teetsov and M. A. Fox, J. Mater. Chem., 1999, 9, 2117; (b)
K. Becker, J. M. Lupton, J. Feldmann, B. S. Nehls, F. Galbrecht,
D. Q. Gao and U. Scherf, Adv. Funct. Mater., 2006, 16, 364.
31 S.-H. Chen and Y. Chen, Macromolecules, 2005, 38, 53.
32 X. Gong, J. C. Ostrowski, D. Moses, G. C. Bazan and A. J. Heeger,
Adv. Funct. Mater., 2003, 13, 439.
33 M. Uchida, C. Adachi, T. Koyama and Y. Taniguchi, J. Appl. Phys.,
1999, 86, 1680.
34 Y.-Z. Lee, X. Chen, S.-A. Chen, P.-K. Wei and W.-S. Fann, J. Am.
Chem. Soc., 2001, 123, 2296.
35 S. A. Van Slyke, C. H. Chen and C. W. Tang, Appl. Phys. Lett., 1996,
69, 2160.
36 M.-T. Lee, H.-H. Chen, C.-H. Liao, C.-H. Tsai and C. H. Chen, Appl.
Phys. Lett., 2004, 85, 3301.
20 (a) M. M. Alam and S. A. Jenekhe, Chem. Mater., 2002, 14, 4775; (b)
F. Huang, H. Wu, D. Wang, W. Yang and Y. Cao, Chem. Mater.,
2004, 16, 708; (c) M. M. Alam, C. J. Tonzola and S. A. Jenekhe,
Macromolecules, 2003, 36, 6577; (d) A. P. Kulkarni and
S. A. Jenekhe, Macromolecules, 2003, 36, 5285.
21 (a) K.-M. Yeh and Y. Chen, J. Polym. Sci., Part A: Polym. Chem.,
2006, 44, 5362; (b) K.-M. Yeh and Y. Chen, J. Polym. Sci., Part A:
Polym. Chem., 2007, 45, 2259.
22 (a) B. Ma, B. J. Kim, L. Deng, D. A. Poulsen, M. E. Thompson and
ꢀ
J. M. J. Frechet, Macromolecules, 2007, 40, 8156; (b) C.-S. Wu and
Y. Chen, Macromolecules, 2009, 42, 3729.
23 D. R. Coulson, L. C. Satek and S. O. Grim, Inorg. Synth., 1972, 13,
121.
24 M. Ranger, D. Rondeau and M. Leclerc, Macromolecules, 1997, 30,
7686.
25 (a) Y. Oishi, M. Ishida, M.-A. Kakimoto, Y. Imai and T. Kurosaki,
J. Polym. Sci., Part A: Polym. Chem., 1992, 30, 1027; (b)
K. H. Park, M.-A. Kakimoto and Y. Imai, J. Polym. Sci., Part A:
Polym. Chem., 1998, 36, 1987; (c) S.-H. Hsiao, C.-W. Chen and
G.-S. Liou, J. Polym. Sci., Part A: Polym. Chem., 2004, 42, 3302.
This journal is ª The Royal Society of Chemistry 2010
J. Mater. Chem., 2010, 20, 7700–7709 | 7709