Synthesis and electroluminescence properties of fluorene-co-diketopyrrolopyrrole-co-phenothiazine polymers

Article abstract of DOI:10.1016/j.polymer.2009.12.044

A series of fluorene-co-diketopyrrolopyrrole(DPP)-co-phenothiazine polymers, named as Flu-DPP-Phen, were synthesized by a palladium-catalyzed Suzuki coupling reaction with different feed ratios among 9,9-dihexylfluorene-2,7-bis(trimethylene boronate), 2,5-dioctyl-3,6-bis(4-bromophenyl)pyrrolo[3,4-c]pyrrole-1,4-dione, and 3,7-dibromo-10-octylphenothiazine. Their chemical structures and compositions were confirmed by ¹H NMR and elemental analysis. These terpolymers were found to be thermally stable and readily soluble in common organic solvents. Absorption and photoluminescence (PL) properties of Flu-DPP-Phen exhibit regular change with increasing of DPP contents in the terpolymers. Electroluminescence (EL) properties of all the terpolymers were characterized with the device configurations of ITO/PEDOT/terpolymer/Ba/Al and ITO/PEDOT/PVK/terpolymer/Ba/Al. Owing to exciton confinement on the narrow band gap DPP unit, the emission of fluorene segments is quenched completely with very low content of DPP (0.2 mol%). The EL spectra of all the terpolymers show exclusive long-wavelength emission originating from DPP units, which implied that the energy transfer in the terpolymer is very efficient. EL colors of the terpolymers vary from orange to red, the maximum emission is gradually red-shifted from 582 nm to 600 nm. The best EL performance was achieved by Flu-DPP-Phen(50:30:20) with maximum external quantum efficiency (EQE) of 0.25% and maximum brightness of 259 cd/m² in the device configuration of ITO/PEDOT/PVK/terpolymer/Ba/Al. The preliminary EL results proved that DPP units could effectively improve the electron affinity, and phenothiazine units could significantly enhance the hole injection ability, which resulted in the remarkable improvement of EQE.

Full text of DOI:10.1016/j.polymer.2009.12.044

Polymer 51 (2010) 1016–1023
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Synthesis and electroluminescence properties of
fluorene-co-diketopyrrolopyrrole-co-phenothiazine polymers
,
Zhi Qiao a, Junbiao Peng b, Yi Jin a, Qilin Liu c, Jiena Weng a, Zhicai He b, Shaohu Han b, Derong Cao a
*
a School of Chemistry and Chemical Engineering, South China University of Technology, Guangzhou 510640, China
b Institute of Polymer Optoelectronic Materials and Devices, South China University of Technology, Guangzhou 510640, China
c Guangzhou Institute of Chemistry, Chinese Academy of Sciences, Guangzhou 510650, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of fluorene-co-diketopyrrolopyrrole(DPP)-co-phenothiazine polymers, named as Flu-DPP-Phen,
were synthesized by a palladium-catalyzed Suzuki coupling reaction with different feed ratios among
9,9-dihexylfluorene-2,7-bis(trimethylene boronate), 2,5-dioctyl-3,6-bis(4-bromophenyl)pyrrolo[3,4-
c]pyrrole-1,4-dione, and 3,7-dibromo-10-octylphenothiazine. Their chemical structures and composi-
tions were confirmed by 1H NMR and elemental analysis. These terpolymers were found to be thermally
stable and readily soluble in common organic solvents. Absorption and photoluminescence (PL) prop-
erties of Flu-DPP-Phen exhibit regular change with increasing of DPP contents in the terpolymers.
Electroluminescence (EL) properties of all the terpolymers were characterized with the device config-
urations of ITO/PEDOT/terpolymer/Ba/Al and ITO/PEDOT/PVK/terpolymer/Ba/Al. Owing to exciton
confinement on the narrow band gap DPP unit, the emission of fluorene segments is quenched
completely with very low content of DPP (0.2 mol%). The EL spectra of all the terpolymers show exclusive
long-wavelength emission originating from DPP units, which implied that the energy transfer in the
terpolymer is very efficient. EL colors of the terpolymers vary from orange to red, the maximum emission
is gradually red-shifted from 582 nm to 600 nm. The best EL performance was achieved by Flu-DPP-
Phen(50:30:20) with maximum external quantum efficiency (EQE) of 0.25% and maximum brightness of
259 cd/m2 in the device configuration of ITO/PEDOT/PVK/terpolymer/Ba/Al. The preliminary EL results
proved that DPP units could effectively improve the electron affinity, and phenothiazine units could
significantly enhance the hole injection ability, which resulted in the remarkable improvement of EQE.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 23 September 2009
Received in revised form
6 December 2009
Accepted 30 December 2009
Available online 14 January 2010
Keywords:
Fluorene
Diketopyrrolopyrrole
Phenothiazine
1. Introduction
and their derivatives (PFs) have been extensively studied during the
past few years [9–12]. Moreover, the emission color of PFs can be
Since the initial polymeric light-emitting diodes (PLEDs) based
on poly(p-phenylenevinylene) were reported by the Cambridge
group [1], these new apparatus have received considerable atten-
tion due to their potential application in flat-panel displays and
lighting applications [2–4]. PLEDs are promising devices for use in
full-color displays because they have a number of advantages over
conventional devices such as a low driving voltage, wide view
angle, self-emitting property, full-color capability, low weight,
potentially large area color displays and flexibility [5–7]. PLEDs are
expected to become dominant displays in the future [8].
easily tuned in the entire visible region by introducing narrow band
gap comonomers into the polyfluorene backbone [13]. Neverthe-
less, there are some drawbacks that continue to hamper their
potential applicability. Firstly, undesired green emission that
appears upon thermal annealing or device operation. This low-
energy emission band has been attributed to the formation of
excimers or aggregates in the solid state [12]. Secondly, there is
high ionization potential (5.8 eV) for PFs in PLEDs, i.e., a high energy
barrier exists when holes travel into PFs from the indium tin oxide
(ITO)/poly (3,4-ethylenedioxythiophene) (PEDOT) layer (5.0–
5.2 eV), which results in higher driving voltages [14]. Phenothiazine
is a well-known heterocyclic compound with electron-rich sulfur
and nitrogen heteroatoms. Because of unique electrooptical prop-
erties, polymers containing phenothiazine and its derivatives were
potential candidate materials for PLEDs [15]. Obviously, phenothi-
azine and its derivatives which were incorporated into PFs should
Because of their high photoluminescence and electrolumines-
cence efficiencies as well as good thermal properties, polyfluorene
* Corresponding author. Tel./fax: þ86 20 87110245.
E-mail address: drcao@scut.edu.cn (D. Cao).
0032-3861/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2009.12.044

Z. Qiao et al. / Polymer 51 (2010) 1016–1023
1017
improve the injection and transport of holes. And what’s more, the
nonplanar structure of the phenothiazine ring should impede
stacking aggregation and intermolecular excimer formation in the
polymer main chain [16].
under a heating rate of 10 ꢀC/min and a N2 flow rate of 100 ml/min.
Elemental analysis was performed using a Vario EL III instrument.
UV–vis absorption spectra were recorded on an HP 4803 Instru-
ment. PL and EL spectra were recorded on an Instaspec IV CCD
spectrophotometer. The absolute PL quantum yields were deter-
mined in an Integrating sphere IS080 (Labsphere) with 325 nm
excitation of He–Cd laser (Mells Griod), as the percent of the total
output photons in all directions vs the total input photons. The
luminance-voltage curves (L–V) were measured using a Keithley
236 source measurement unit and a calibrated silicon photodiode.
The luminance was calibrated using a PR-705 SpectraScan spec-
trophotometer (Photo research). The external quantum efficiency
(EQE) was collected by measuring the total light output in all
directions in an integrating sphere. It was determined as the
percent of the total output photons vs the total input electrons from
each electrode.
Another notable problem for PFs in PLEDs applications is that
most of the fluorene-containing copolymers synthesized to date
have low electron affinity and are typically p-type or hole-trans-
port-dominated polymers [17]. Because electroluminescence in
a PLED arises from the recombination of holes and electrons in an
active polymer layer, any unipolar characteristics in the polymer
layer may result in an unbalance in hole and electron injection and/
or transport, leading to lower efficiency of the devices [12,18,19]. It
is important to develop efficient PFs with high electron affinity to
achieve a more balanced injection and transport of carriers (holes
and electrons).
Despite many attempts to tune the color of blue-emitting PFs,
appropriate red-emitting materials that meet the requirements of
display applications have yet to be obtained [16]. If two chromo-
phores with different energy band gaps are mixed together, then
the emission will predominately or exclusively come from the
chromophore with the narrower energy band gap because of
energy transfer [15,16,20]. Hence, comonomers with narrow
energy band gap can be introduced into PFs for color tuning. To
date, those aromatic heterocycles such as thiophene, bithiophene,
benzothiazole, and benzodithiazole derivatives are the most widely
used as narrow energy band gap comonomers [12,16]. However,
only a limited number of fluorene-based copolymers are known for
red electroluminescence [21]. The exploration of novel classes is
still needed for further improvements.
2.2. Materials
2,5-Dioctyl-3,6-bis(4-bromophenyl)pyrrolo[3,4-c]pyrrole-1,4-
dione (1) was synthesized as described before [27]. 9,9-Dihexyl-
fluorene-2,7-bis(trimethylene boronate) (2) and tetrakis
(triphenylphosphine) palladium (Pd(PPh3)4) were purchased from
Sigma Aldrich and Alfa Aesar, respectively, and were used without
any further purification. 3,7-Dibromo-10-octylphenothiazine (3)
was synthesized according to the procedures in the literature [31].
2-Bromo-9,9-di-n-octyl-fluorene was synthesized from fluorene as
a starting material according to the literature [32].
In recent years, DPP derivatives with the pyrrolo[3,4-c]pyrrole-
1,4-dione unit as the new electroluminescent materials have drawn
broad attention due to their unique structure and properties
[22,23]. DPP derivatives possess brilliant red, high photo-
luminescence quantum yields and excellent chemical and thermal
stability as well as photostability. They are potential materials
applied in luminescent field as a kind of new electroluminescent
materials [24–27]. Additionally, DPP derivatives are convenient
obtained from commercial resources or simple synthetic proce-
dures and easy optimization of the structure through variations of
aryls on 3- and 6-positions. It will be very attractive to incorporate
DPP derivatives into PFs back bone for red emission. According to
our former study [28,29], DPP derivatives exhibit high electron
affinity because of their two lactam units.
We have briefly reported that a novel polyfluorene derivative
which contained DPP and phenothiazine units had remarkable
improvement of EQE [30]. In order to get a full understanding of
their role in the conjugated PFs, we report the systematically study
on the synthesis of a series of three component copolymers with
different ratios of fluorene, phenothiazine, diketopyrrolopyrrole
units and the structure–property relationships of DPP-fluorene-
phenothiazine polymers here. PLEDs based on the terpolymers are
expected to emit a red light, and exhibit better hole-injection/
transport in polymer electronic devices.
2.3. General procedure of polymerization
Monomer 1, K2CO3 (17 equiv), monomer 2, Pd (PPh3)4 (2 mol%),
and monomer 3 were dissolved in a mixture of THF and water (50/
50, v/v) under N2. The mixture was heated to 80 ꢀC and stirred
vigorously for 48 h. Additional 2 (6 mol%) was added, and the
mixture was refluxed under N2 for 6 h; then 2-bromo-9,9-di-n-
octyl-fluorene (12 mol%) was added, and the mixture was refluxed
for another 6 h. The reaction mixture was cooled, and the solid was
precipitated in methanol. The solid was filtered off, washed with
dilute HCl, dissolved in chloroform, and precipitated in acetone.
After being extracted with acetone in a Soxhlet apparatus for 24 h,
terpolymers were obtained as pale brown to dark red solid with
yields of 73–81%.
2.4. EL devices fabrication and characterization
Flu-DPP-Phen were dissolved in toluene or p-xylene and filtered
through a 0.45 mm filter. Patterned indium tin oxide (ITO)-coated
glass substrates were cleaned with acetone, detergent, distilled
water, and 2-propanol, subsequently in an ultrasonic bath. After
treatment with oxygen plasma, 50–60 nm of poly(3,4-ethyl-
enedioxythiophene) (PEDOT) doped with poly(styrenesulfonic
acid) (PSS) (Batron-P 4083, Bayer AG) was spin-coated onto the ITO
2. Experimental section
substrate followed by drying in a vacuum oven. Poly(-
vinylcarbazole) (PVK) from 1,1,2,2-tetrachloroethane solution was
coated on top of a dried PEDOT:PSS layer subsequently. A thin film
of Flu-DPP-Phen was coated onto the anode by spin-casting inside
a drybox. The film thickness of the active layer was 70–80 nm, as
measured with an Alfa Step 500 surface profiler (Tencor). For EL
devices with Ba as cathodes, a thin layer of Ba (4–5 nm) and
subsequently 150 nm layers of Al were vacuum-evaporated
subsequently on the top of an EL polymer layer under a vacuum of
1 ꢁ10ꢂ4 Pa.
2.1. Measurement and characterization
1H NMR spectra were collected on a Bruker DRX 400 spec-
trometer in CDCl3 with tetramethylsilane as inner reference.
Number-average (Mn) and weight-average (Mw) molecular weights
were determined by a Waters GPC 515-410 in tetrahydrofuran
(THF) using a calibration curve of polystyrene standards. Ther-
mogravimetric analysis (TGA) was conducted on a Pyris 1 TGA

1018
Z. Qiao et al. / Polymer 51 (2010) 1016–1023
3. Results and discussion
because boron and bromine units could quench emission and
contribute to excimer formation in PLED applications [33]. All the
terpolymers were found to be soluble in common organic
solvents such as CH2Cl2, CHCl3, and THF. Molecular weights and
polydispersity indices (PDI) are listed in Table 1. Mn and Mw
ranged from 5605 to 30,454 (Mn) and from 10,320 to 70,487 (Mw),
respectively, with PDI ranging from 1.84 to 2.79. The actual
contents of monomers 1 and 3 in the terpolymer were calculated
according to elemental analysis, and the results were very close to
the feed compositions. The chemical structures of the terpolymers
were confirmed by 1H NMR, as shown in Fig. 1. As expected, the
intensity of signals corresponding to protons of DPP moieties
(protons a, b, c, and d), relative to protons of phenothiazine
moieties (protons f and g), increase gradually along with DPP
contents increasing in the terpolymers. The thermal properties of
the terpolymers are shown in Table 1. TGA revealed that onset
decomposition temperatures (based on 5% weight loss, Td) were
in range of 198–357 ꢀC. Flu-DPP-Phen(50:0.2:49.8) and Flu-DPP-
Phen(50:30:20) exhibited very good thermal stability, losing less
than 5% weight on heating to w350 ꢀC in TGA runs under
nitrogen.
3.1. Synthesis and characterizations
The conjugated terpolymers derived from monomers 1, 2 and
3
were prepared by palladium(0)-catalyzed Suzuki coupling
reactions with an equivalent molar ratio of the diboronic ester
monomer to the dibromo monomers (Scheme 1). The molar ratio
of 2 to 1 and 3 was always kept as 2: (1 þ3) ¼ 1:1, the molar feed
ratios of 1–3 were 0.2:49.8, 2.0:48.0, 10.0:40.0, 20.0:30.0,
30.0:20.0, and the corresponding terpolymers were named as
Flu-DPP-Phen(50:0.2:49.8), Flu-DPP-Phen(50:2:48), Flu-DPP-
Phen(50:10:40),
Flu-DPP-Phen(50:20:30),
Flu-DPP-Phen(50:
30:20), respectively. Their chemical structures and compositions
were verified by 1H NMR and elemental analysis. For these
terpolymers, n-octyl substituents in the 2,5-positions of DPP and
10-position of phenothiazine were employed to improve the
solubility of the resulting terpolymers. At the end of polymeri-
zation, 9,9-dihexylfluorene-2,7-bis(trimethylene boronate) was
added to remove bromine end groups and 2-bromo-9,9-di-n-
octyl-fluorene was added to remove boronic ester end groups,
C8H17
N
Η
O
O
CN
Br
N
O
O
O
Br
O
Br
C8H17Br/NMP
t-C5H11OK
Br
O
Br
O
N
N
Η
C8H17
1
C8H17
N
C6H13
C6H13
O
O
O
B
B
Br
S
O
Br
2
3
1, K2CO3, Pd(PPh3)4
THF/H2O
C6H13
C8H17
N
C6H13
C6H13
C6H13
Br
S
O
O
x
y
B
N
n
O
O
N
C8H17
C8H17
1) 2
2) 2-bromo-9,9-dioctyl-fluorene
C8H17
C8H17
C6H13
C6H13
C8H17
C8H17
C6H13
C6H13
C8H17
C6H13
C6H13
N
O
S
y
x
O
N
N
C8H17
C8H17
Flu-DPP-Phen
Flu-DPP-Phen(50:0.2:49.8) (x:y=0.2:49.8)
Flu-DPP-Phen(50:2:48) (x:y=2:48)
Flu-DPP-Phen(50:10:40) (x:y=10:40)
Flu-DPP-Phen(50:20:30) (x:y=20:30)
Flu-DPP-Phen(50:30:20) (x:y=30:20)
Scheme 1. Synthetic route of the terpolymers.

Z. Qiao et al. / Polymer 51 (2010) 1016–1023
1019
Table 1
Yields, molecular weights, and thermal properties of the terpolymers.
1.2
Flu-DPP-Phen(50:0.2:49.8)
Flu-DPP-Phen(50:2:48)
Flu-DPP-Phen(50:10:40)
Flu-DPP-Phen(50:20:30)
Flu-DPP-Phen(50:30:20)
Polymers
Yield Mw
(%)
Mn
PDI
1
3
Td
(mol%)a (mol%)a (ꢀC)b
1.0
0.8
0.6
0.4
0.2
0.0
Flu-DPP-Phen
(50:0.2:49.8)
Flu-DPP-Phen
(50:2:48)
Flu-DPP-Phen
(50:10:40)
Flu-DPP-Phen
(50:20:30)
Flu-DPP-Phen
(50:30:20)
73
76
71
79
81
45,316 16,232 2.79
25,842 13,137 1.97
342
2.1
47.9
38.8
28.9
21.8
220
307
198
357
10,320
5605 1.84 11.2
36,299 18,305 1.98 21.1
70,487 30,454 2.32 28.2
a
Calculated from the results of elemental analysis.
Based on 5% weight loss.
b
3.2. Optical and photoluminescence properties
300
350
400
450
500
550
600
650
The UV–vis absorption spectra of the terpolymers in thin films
are shown in Fig. 2. Flu-DPP-Phen(50:0.2:49.8) and Flu-DPP-
Phen(50:2:48) in thin films possess very similar absorption spectra
Wavelength(nm)
Fig. 2. UV–vis absorption spectra of the terpolymers in thin films.
with maximum peaks at 331 nm, attributed to
p–p* transition of
the terpolymers. In addition, a weak absorption peak at 382 nm is
attributed to the fluorene unit. However, Flu-DPP-Phen(50:10:40),
Flu-DPP-Phen(50:20:30) and Flu-DPP-Phen (50:30:20) show
different absorption spectra. The additional long-wavelength
absorption peak at 513 nm is attributed to the DPP segments.
Moreover, the absorption intensity at 513 nm increased gradually
with DPP content increasing in the terpolymer backbone. In the
case of Flu-DPP-Phen(50:0.2:49.8) and Flu-DPP-Phen(50:2:48)
containing 0.2% and 2% of DPP in the terpolymers, respectively, the
absorption peak of the DPP unit is not apparent in the solid film. In
one word, the two distinguishable absorption features of all
terpolymers demonstrate that the electronic configurations of
three components (fluorene, DPP and phenothiazine) in the
terpolymers are not well mixed.
at a concentration of 1 ꢁ10ꢂ5 M also change regularly with the
variation of DPP contents. The PL emission spectra of dilute solu-
tions of Flu-DPP-Phen(50:0.2:49.8) and Flu-DPP-Phen(50:2:48)
reveal that they are nearly identical with maximum emission peaks
at 486 nm. No typical emission from fluorene (at about 450 nm)
was observed, which implied that the emission of fluorene was
quenched. There is no signature of DPP emission. Increasing the
DPP contents to 10% for Flu-DPP-Phen(50:10:40), the DPP emission
emerges with maximum at 560 nm, but the intensity is weak
relatively. Once the DPP units increase to 20%, the intensity of the
DPP emission peak at 574 nm of Flu-DPP-Phen(50:20:30) is
enhanced dramatically and become stronger than the short-
wavelength emission (482 nm). Moreover, the emission in short-
wavelength region nearly disappears with DPP content up to 30%
for Flu-DPP-Phen(50:30:20). This can be attributed to strong
intramolecular energy transfer. Compared with the maximum PL
emission peak (578 nm) of alternating fluorene-DPP copolymer P1
in the previous work [28], all the maximum PL emission peaks
(486–574 nm) of the terpolymers in solutions are regularly blue-
shifted; and what’s more, the emission peak in short-wavelength
Photoluminescence spectra in CHCl3 solutions and in thin solid
films are shown in Figs. 3 and 4, respectively. All characteristic data
of absorption and photoluminescence are summarized in Table 2.
As shown in Fig. 3, PL spectra of the terpolymers in CHCl3 solutions
d
g
c
S
N
e
O
b
a
Flu-DPP-Phen(50:0.2:49.8)
y
n
N
f
x
1.0
O
N
Flu-DPP-Phen(50:2:48)
Flu-DPP-Phen(50:10:40)
Flu-DPP-Phen(50:20:30)
Flu-DPP-Phen(50:0.2:49.8)
Flu-DPP-Phen(50:2:48)
0.8
Flu-DPP-Phen(50:30:20)
0.6
0.4
0.2
0.0
Flu-DPP-Phen(50:10:40)
Flu-DPP-Phen(50:20:30)
Flu-DPP-Phen(50:30:20)
a
b
e
d
g
c
f
400
450
500
550
600
650
700
750
800
850
Wavelength(nm)
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
Fig. 3. Photoluminescent spectra of the terpolymers in CHCl3 solutions (1 ꢁ10ꢂ5 M,
Fig. 1. 1H NMR spectra of the terpolymers in CDCl3.
excitation wavelength 325 nm).

1020
Z. Qiao et al. / Polymer 51 (2010) 1016–1023
0.00006
Flu-DPP-Phen(50:0.2:49.8)
1.0
0.8
0.6
0.4
0.2
0.0
0.00004
0.00002
0.00000
-0.00002
-0.00004
-0.00006
-0.00008
-0.00010
-0.00012
-0.00014
Flu-DPP-Phen(50:2:48)
Flu-DPP-Phen(50:10:40)
Flu-DPP-Phen(50:20:30)
Flu-DPP-Phen(50:30:20)
Flu-DPP-Phen(50:0.2:49.8)
Flu-DPP-Phen(50:2:48)
Flu-DPP-Phen(50:10:40)
Flu-DPP-Phen(50:20:30)
Flu-DPP-Phen(50:30:20)
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
300
400
500
600
700
800
900
1000
1100
1200
E(V)
wavelength(nm)
Fig. 5. Cyclic voltammograms of the terpolymers on a Pt electrode at a scanning rate of
Fig. 4. Photoluminescent spectra of the terpolymers in thin films (excitation wave-
50 mV/s.
length 325 nm).
the fluorescent quenching effect caused by carbonyl groups. The
most interesting insight from these photophysics results is that the
phenothiazine ring is an excellent building block for impeding
region around 480 nm disappears completely when DPP content
increases to 50 mol% regard to the alternating fluorene-DPP
copolymer P1.
p-stacking aggregation and intermolecular excimer formation,
PL spectra of the terpolymers in thin films closely resemble
spectra in chloroform solution, as shown in Fig. 4. However, some
differences should be noticed: first, the threshold of DPP content for
complete quenching of emission from fluorene and phenothiazine
segments is lowered to 10%; second, the maximum emission of PL is
some red-shifted for the terpolymers, compared with those in
chloroform solution. These facts can be attributed to more intra-
molecular or intermolecular interactions in the film than in the
solution [34].
Interestingly, as the fraction of DPP increased, the PL emission
maximum for all terpolymers in thin films in the longer wavelength
region gradually is red-shifted (566–604 nm), and this phenom-
enon is identical with the fluorene-DPP alternating copolymers
[29]. Two interpretations are possible to explain the phenomenon:
because phenothiazine ring is nonplanar [16].
It should be noted that all the PL efficiencies of these terpoly-
mers are lower than the fluorene-DPP alternating copolymers
reported previously [29]. The low PL efficiencies of the terpolymers
might be due to the presence of the phenothiazine unit, which is
more effective as a quencher than the carbonyl group of DPP unit. In
addition, the fluorene content in the terpolymer (50 mol%) is less
than the fluorene content (50–99 mol%) in the DPP-fluorene
alternating copolymers in ref. 29, which might result lower PL
efficiencies than the DPP-fluorene alternating copolymer because
fluorene segments usually exhibit high photoluminescence effi-
ciencies. Another possible reason may be attributed to the disor-
dered main chain structure of these random terpolymers.
increase in the p-conjugation length with increase in the fraction of
3.3. Electrochemical properties of the terpolymers
DPP, and aggregate formation between the moieties of DPP.
The absolute PL efficiencies of the terpolymers in thin films were
measured in the integrating sphere using a HeCd laser line of
325 nm as the excitation source, and the results are listed in Table 2.
The PL efficiencies decrease quickly with increasing DPP content in
the terpolymers from 10.95% for Flu-DPP-Phen(50:0.2:49.8) to
4.54% for Flu-DPP-Phen(50:10:40), which is probably due to
quenching effect of the carbonyl groups in DPP unit. Carbonyl
groups which facilitate intersystem crossing of singlet to triplet
The electrochemical properties of the terpolymers were inves-
tigated by cyclic voltammetry (CV). The oxidation and reduction
potentials are closely related to the HOMO and LUMO levels of the
copolymers. The terpolymer films deposited on a platinum wire
electrode were scanned both positively and negatively (scan rate of
50 mV/s) in 0.10 M tetrabutylammonium hexafluorophosphate
(Bu4NPF6) in anhydrous acetonitrile by using calomel as reference
electrode. As shown in Fig. 5, in the anodic scans, the onset
oxidation potentials of the terpolymers ranged from 0.99 to 1.16 V.
HOMO energy levels of the terpolymers were between ꢂ5.39 eV
and ꢂ5.56 eV calculated according to the empirical formula EHO-
with their n–p* transition, and possible intramolecular charge
transfer is responsible for the loss of PL efficiency. However, the PL
efficiencies do not change dramatically when DPP content
increases to 10% or more. The PL quantum yield of Flu-DPP-
Phen(50:0.2:49.8) is the highest among the five terpolymers, which
implies that nonplanar phenothiazine ring can effectively restrain
¼ ꢂ(Eox þ 4.4) eV. Compared with the alternating fluorene-DPP
MO
copolymer P1 [28], HOMO energy levels of the terpolymers were
higher than the energy level of P1 (ꢂ5.6 eV), implying that the
energy barriers between the work function of PEDOT/PSS (ꢂ5.0 eV)
Table 2
Photophysical properties of the terpolymers.
Table 3
Polymers
UV, lmax (nm) PL, lmax (nm)
Thin film Solution Thin film Thin film
Flu-DPP-Phen(50:0.2:49.8) 331,382
hPL (%)
Electrochemical properties of the terpolymers.
Polymers
Eox (V) Ered (V) HOMO(eV) LUMO(eV) Eg(eV)
486 483,566
486 476,579
10.95
6.93
4.51
4.24
4.72
Flu-DPP-Phen(50:0.2:49.8) 1.16
ꢂ1.07
ꢂ1.03
ꢂ1.02
ꢂ1.01
ꢂ1.02
ꢂ5.56
ꢂ5.55
ꢂ5.55
ꢂ5.54
ꢂ5.39
ꢂ3.33
ꢂ3.37
ꢂ3.38
ꢂ3.39
ꢂ3.38
2.23
2.18
2.17
2.15
2.01
Flu-DPP-Phen(50:2:48)
Flu-DPP-Phen(50:10:40)
Flu-DPP-Phen(50:20:30)
Flu-DPP-Phen(50:30:20)
331,382
346,513
335,513
289,513
Flu-DPP-Phen(50:2:48)
Flu-DPP-Phen(50:10:40)
Flu-DPP-Phen(50:20:30)
Flu-DPP-Phen(50:30:20)
1.15
1.15
1.14
0.99
486,560
482,574
482,574
587
587
604

Z. Qiao et al. / Polymer 51 (2010) 1016–1023
1021
emission maxima for the terpolymers are gradually red-shifted
from 582 nm for Flu-DPP-Phen(50:0.2:49.8) to 600 nm for Flu-DPP-
Phen(50:30:20), and this change pattern coincides with the trend
of the red shift of PL emission maximum. Compared with the EL
emission spectrum of alternating fluorene-DPP copolymer P1 [28],
all the maximum EL emission peaks (582–600 nm) of the terpoly-
mers are regularly blue-shifted. As summarized in Table 4, CIE
coordinates change from (0.51, 0.48) for Flu-DPP-Phen(50:0.2:49.8)
to (0.62, 0.38) for Flu-DPP-Phen(50:30:20), and the emission color
of the devices changes from orange to red, which indicate that the
more DPP contents in the terpolymers, the more saturated red
emission can be achieved. In particular, the chromaticity value of
Flu-DPP-Phen(50:30:20) is (0.62, 0.38) which is very close to the
standard red emission (0.66, 0.34) demanded by NTSC [17].
Flu-DPP-Phen(50:0.2:49.8)
Flu-DPP-Phen(50:2:48)
Flu-DPP-Phen(50:10:40)
Flu-DPP-Phen(50:20:30)
Flu-DPP-Phen(50:30:20)
1.0
0.8
0.6
0.4
0.2
0.0
Fig. 7 shows brightness-voltage curves of the terpolymers in the
devices with the configuration of ITO/PEDOT/terpolymer/Ba/Al.
The maximum EQE, maximum brightness, and turn-on voltages of
the devices are also listed in Table 4. The turn-on voltages of the
devices range from about 5.1 to 10.2 V. Furthermore, the PLED device
based on Flu-DPP-Phen(50:30:20) exhibits the lowest turn-on
voltage among all the devices. It is worthy to note that the turn-on
voltage (5.1 V) of Flu-DPP-Phen(50:30:20) is lower obviously than
the turn-on voltage (8 V) of the alternating fluorene-DPP copolymer
P1. This fact can also be attributed to easier hole injection and
enhanced hole transport, which is mainly contributed by the intro-
duction of phenothiazine segments into copolymers. The maximum
brightness of all the devices are in the range of 37–195 cd/m2, and
the maximum EQEs change from 0.01% to 0.11%. With increasing
the DPP content, maximum brightness and maximum EQEs of light-
emitting devices first decrease, and then gradually increase. The
best device performance of the terpolymers is observed for Flu-
DPP-Phen(50:30:20) with the maximum brightness of 195 cd/m2
and maximum EQE of 0.11%. Furthermore, the maximum brightness
(195 cd/m2) of Flu-DPP-Phen(50:30:20) is increased by 27%
compared with the maximum brightness (153 cd/m2) of the alter-
nating fluorene-DPP copolymer P1. Terpolymers exhibit good EL
performances with low DPP contents (0.2%) and high DPP contents
(30%), while EL performances are very poor of terpolymers with
medium DPP contents (2–20%). These results are consistent with
fluorene-DPP copolymers [29]. Even for Flu-DPP-Phen(50:10:40)
with DPP content of 10% in the feed, both the brightness–voltage
plot and the EQE–voltage plot of the light-emitting device (LED)
could not be detected because of too low luminance and efficiency.
Flu-DPP-Phen(50:0.2:49.8) might be superior to other terpolymers
in consideration of conjugation that is responsible for carrier
mobility, which is critical for efficient EL. In addition, the pheno-
400
500
600
700
800
900
wavelength(nm)
Fig. 6. Electroluminescent spectra of the terpolymers in the devices with the config-
uration of ITO/PEDOT/terpolymer/Ba/Al.
and HOMO energy levels of the terpolymers reduced due to the
incorporation of phenothiazine units into polymer main chains.
During the cathodic scans, the onset reduction potentials of the
terpolymers were between ꢂ1.01 V and ꢂ1.07 V, and LUMO energy
levels were estimated to be ranged from ꢂ3.33 to ꢂ3.39 eV
according to the empirical formula ELUMO ¼ ꢂ(Ered þ 4.4) eV. The
electrochemical band gaps were calculated from HOMO and LUMO
levels, and the results are summarized in Table 3. It can be seen that
the energy band gaps of the terpolymers gradually decrease from
2.23 to 2.01 eV with increasing DPP contents in the terpolymers.
3.4. Electroluminescence properties
A double-layer device with a configuration of ITO/PEDOT/
terpolymer/Ba/Al was fabricated, where PEDOT doped with poly(-
styrenesulfonic acid) (PSS) functions as a buffer layer and a hole-
injection/transport layer. The EL spectra of the terpolymers are
shown in Fig. 6. The emission maxima and chromaticity coordinates
for the terpolymers are summarized in Table 4. In contrast to PL
spectra, EL emission consists exclusively of orange or red emission
peaked around 582–600 nm originating from DPP units for the
terpolymers. The emission from fluorene and phenothiazine
segments is completely quenched. The differences in the PL and EL
spectra at low doping concentrations were attributed to the
differences in the recombination zone for photo- and electric
excitations. It is important to note that the short-wavelength
emission is completely quenched even at 0.2% DPP loading in the
feed, indicating intramolecular and intermolecular energy transfer
is a very quick and efficient process. As a result, the generated
excitons are efficiently confined to DPP sites. The DPP moiety even
at the low content can serve as an efficient trap for all of generated
excitons, and the terpolymer emits exclusively from the narrow
energy band gap component. With increasing DPP contents, the EL
thiazine units can impede p-stacking aggregation and intermolec-
ular excimer formation due to nonplanar ring. However, when DPP
content increases to 30% in the feed for Flu-DPP-Phen(50:30:20), it
exhibits best EL performance, which partially attributes to more
efficient energy transfer from wide band gap fluorene segments to
narrow band gap DPP segments. Moreover, more DPP units will
largely improve electron affinity and achieve more balanceable
mobility of carriers (holes and electrons).
Table 4
EL performances of the devices with the configuration of ITO/PEDOT/terpolymer/Ba/Al.
Polymers
EL Emission lmax(nm)
CIE Coordinate (x,y)
Turn-on Voltage (V)a
Maximum Brightness (cd/m2)
Maximum EQE (%)
Flu-DPP-Phen(50:0.2:49.8)
Flu-DPP-Phen(50:2:48)
Flu-DPP-Phen(50:10:40)
Flu-DPP-Phen(50:20:30)
Flu-DPP-Phen(50:30:20)
582
589
590
592
600
(0.51,0.48)
(0.56,0.42)
(0.57,0.42)
(0.58,0.47)
(0.62,0.38)
7.6
10.2
178
80
0.049
0.021
6.4
5.1
37
195
0.018
0.110
a
Turn-on voltage corresponds to luminance of about 1.5 cd/m2.

1022
Z. Qiao et al. / Polymer 51 (2010) 1016–1023
220
200
180
160
140
120
100
80
results, we can conclude that the holes injection and transport
Flu-DPP-Phen(50:0.2:49.8)
Flu-DPP-Phen(50:2:48)
Flu-DPP-Phen(50:20:30)
Flu-DPP-Phen(50:30:20)
ability are improved with the phenothiazine units incorporated
into the backbone of copolymers. Similar results have been repor-
ted in the literature [36]. Moreover, these results also support the
other fact that incorporation of DPP units into polymer main chain
can effectively improve electron affinity.
4. Conclusions
In conclusion, a series of soluble conjugated random terpoly-
mers Flu-DPP-Phen contained fluorene, DPP and phenothiazine
units were successfully synthesized by palladium(0)-catalyzed
Suzuki coupling reactions with various molar ratios of the electron-
rich phenothiazine and the narrow energy band gap DPP contents.
1H NMR and elemental analysis show that these polymers possess
well-defined chemical structures. The actual contents of DPP units
and phenothiazine units are close to feed ratios. TGA measure-
ments display excellent thermal stabilities of most terpolymers.
Absorption spectra and PL spectra of the terpolymers vary regularly
with increase of DPP contents. Only very few DPP units (0.2%) are
needed to completely quench emission from fluorene segments in
the terpolymers, implying that energy transfer is very efficient due
to exciton trapping on the narrow band gap DPP sites. Emission
colors of EL devices change from orange to red, and the EL peaks are
gradually red-shifted from 582 nm to 600 nm with increasing DPP
content in polymer chains, corresponding CIE coordinates from
(0.51, 0.48) to (0.62, 0.38). Parameters of the devices with structures
of ITO/PEDOT/PVK/terpolymer/Ba/Al support the conclusion that
incorporation of DPP units into polymer main chain can effectively
improve electron affinity, and incorporation of phenothiazine units
can obviously enhance the holes injection and transport ability,
which would result in balance in carriers injection and transport,
and therefore can significantly improve the EL performances.
60
40
20
0
-20
4
6
8
10
12
14
16
18
20
22
Bias(V)
Fig. 7. Brightness-voltage curves of the terpolymers in the devices with the configu-
ration of ITO/PEDOT/terpolymer/Ba/Al.
To further investigate the expected enhancement of hole
injection arising from the electron-rich sulfur and nitrogen
heteroatoms of phenothiazine units, we fabricated another kind of
device with a configuration of ITO/PEDOT(50 nm)/PVK(30 nm)/
terpolymer (80 nm)/Ba/Al, and PVK was introduced as the hole
injection anode [35]. Responding EL spectra (not shown) are almost
identical to those of the EL devices without a PVK layer. Other
parameters are summarized in Table 5. EL properties of the
terpolymers with medium DPP contents are still poor. We found
that turn-on voltages obviously increased when PVK was intro-
duced. Compared with those of the EL devices without a PVK layer,
maximum EQEs of the devices based on the terpolymers with more
phenothiazine units or low DPP units, Flu-DPP-Phen(50:0.2:49.8)
and Flu-DPP-Phen (50:2:48), dramatically decreased, indicating
that hole mobility was higher than electron mobility for Flu-DPP-
Phen(50:0.2:49.8) and Flu-DPP-Phen(50:2:48) due to more elec-
tron-rich phenothiazine units were incorporated into the main
chain of the terpolymers, which resulted in unbalance in hole and
electron injection and transporting. Therefore superfluous holes led
fluorescent quenching and thereby led significant falling of EQE.
With phenothiazine contents decreasing or DPP contents
increasing, the maximum EQEs of the devices using Flu-DPP-
Phen(50:10:40), Flu-DPP-Phen(50:20:30) and Flu-DPP-Phen
(50:30:20) as active layers were evidently improved compared
with the corresponding EL devices without PVK layer. This might be
attributed to enhancement of electron affinity by DPP units. It is
worthy to note that the maximum brightness of the devices with
the configuration of ITO/PEDOT/PVK/terpolymer/Ba/Al exhibit the
nearly variational trend compared with the devices with the
configuration of ITO/PEDOT/terpolymer/Ba/Al. The EL device with
Flu-DPP-Phen(50:30:20) as the emissive layer shows the highest
EQE (0.25%) and the maximal brightness (259 cd/m2). From these
Acknowledgments
This work was supported by the National Natural Science
Foundation of China (20872038), Research Fund for the Doctoral
Program of Higher Education of China (20060561024), Science and
Technology Planning Project of Guangdong Province, China
(2007A010500011).
References
[1] Burroughes JH, Bradley DDC, Brown AR, Marks RN, Mackay K, Friend RH, et al.
Nature 1990;347:539.
[2] Wu GL, Yang Y, He C, Chen XM, Li YF. Eur Polym J 2008;44:4047.
[3] Jung MS, Lee TW, Lee JH, Sohn BH, Jung IS. Polymer 2006;47:2670.
[4] Lee CC, Yeh KM, Chen Y. Polymer 2008;49:4211.
[5] Jin Y, Song S, Park SH, Park JA, Kim J, Woo HY, et al. Polymer 2008;49:4559.
[6] Peng Q, Lu ZY, Huang Y, Xie MG, Han SH, Peng JB, et al. Macromolecules
2004;37:260.
[7] Park MJ, Lee J, Jung IH, Park JH, Hwang DH, Shim HK. Macromolecules
2008;41:9643.
[8] Geffroy B, Roy PL, Prat C. Polym Int 2006;55:572.
[9] Lee JF, Hsu SLC. Polymer 2009;50:2558.
[10] Park JW, Park SJ, Kim YH, Shin DC, You H, Kwon SK. Polymer 2009;50:102.
[11] Cho NS, Huang DH, Jung BJ, Lim E, Lee J, Shim HK. Macromolecules
2004;37:5265.
[12] Akcelrud L. Prog Polym Sci 2003;28:875.
[13] Wang F, Luo J, Yang KX, Chen JW, Huang F, Cao Y. Macromolecules
2005;38:2253.
[14] Wu FI, Shih PI, Shu CF. Macromolecules 2005;38:9028.
[15] Park JH, Cho NS, Jung YK, Cho HJ, Shim HK, Kim H, et al. Org Electron
2007;8:272.
[16] Cho NS, Park JH, Lee SK, Lee J, Shim HK. Macromolecules 2006;39:177.
[17] Wu WC, Liu CL, Chen WC. Polymer 2006;47:527.
[18] Ku CH, Kuo CH, Leung MK, Hsieh KH. Eur Polym J 2009;45:1545.
[19] Manjunatha MG, Adhikari AV, Hegde PK. Eur Polym J 2009;45:763.
[20] Cho NS, Park JH, Shim HK. Curr Appl Phys 2006;6:686.
[21] Chan LH, Lee YD, Chen CT. Macromolecules 2006;39:3262.
Table 5
EL performances of the devices with the configuration of ITO/PEDOT/PVK/
terpolymer/Ba/Al.
Polymers
Turn-on voltage Maximum brightness Maximum
(V)a
(cd/m2)
EQE (%)
Flu-DPP-Phen(50:0.2:49.8) 12.6
159
90
10
15
259
0.029
0.017
0.003
0.026
0.250
Flu-DPP-Phen(50:2:48)
Flu-DPP-Phen(50:10:40)
Flu-DPP-Phen(50:20:30)
Flu-DPP-Phen(50:30:20)
15.8
5.6
9.8
9.4
a
Turn-on voltage corresponds to the luminance of about 1.5 cd/m2.

Z. Qiao et al. / Polymer 51 (2010) 1016–1023
1023
[22] Beyerlein T, Tieke B. Macromol Rapid Commun 2000;21:182.
[23] Rabindranath AR, Zhu Y, Heim I, Tieke B. Macromolecules 2006;39:8250.
[24] Rabindranath AR, Zhu Y, Zhang K, Tieke B. Polymer 2009;50:1637.
[25] Zhu Y, Heim I, Tieke B. Macromol Chem Phys 2006;207:2206.
[26] Zhu Y, Rabindranath AR, Beyerlein T, Tieke B. Macromolecules 2007;40:6981.
[27] Zhang K, Tieke B. Macromolecules 2008;41:7287.
[30] Qiao Z, Cao DR, Liu QL, Weng J, He Z, Han S, et al. Sci China Ser B Chem
2009;52:2038.
[31] Kong X, Kulkarni AP, Jenekhe SA. Macromolecules 2003;36:8992.
[32] Lee JI, Klaerner G, Miller RD. Chem Mater 1999;11:1083.
[33] Yang XH, Yang W, Yuan M, Hou Q, Huang J, Zeng XR, et al. Synth Met
2003;135:189.
[28] Cao DR, Liu QL, Zeng WJ, Han SH, Peng JB, Liu SP. J Polym Sci Part A Polym
Chem 2006;44:2395.
[29] Cao DR, Liu QL, Zeng WJ, Han SH, Peng JB, Liu SP. Macromolecules
2006;39:8347.
[34] Kim JH, You NH, Lee H. J Polym Sci Part A Polym Chem 2006;44:3729.
[35] Ullrich Mitschke, Bauerle P. J Mater Chem 2000;10:1471.
[36] Simokaitiene J, Danilevicius A, Grigalevicius S, Grigalevicius JV, Getautis V,
Jankauskas V. Synth Met 2006;156:926.