Efficient monochromatic red-light-emitting PLEDs based on a series of nonconjugated Eu-polymers containing a neutral terpyridyl ligand

Article abstract of DOI:10.1039/c3tc30681j

Three novel Eu-polymers containing hole transporting units and light-emitting units in the main chain have been designed and synthesized via radical copolymerization of N-vinylcarbazole and polymerizable Eu(TTA) ₃vinyl-tpy, where TTA is 2-theno

Full text of DOI:10.1039/c3tc30681j

Journal of
Materials Chemistry C
View Article Online
View Journal | View Issue
PAPER
Efficient monochromatic red-light-emitting PLEDs based
on a series of nonconjugated Eu-polymers containing a
neutral terpyridyl ligand†
Cite this: J. Mater. Chem. C, 2013, 1,
4885
Chaolong Yang,abc Jing Xu,ac Yunfei Zhang,ac Yinwen Li,ac Jian Zheng,ac Liyan Lianga
a
*
and Mangeng Lu
Three novel Eu-polymers containing hole transporting units and light-emitting units in the main chain have
been designed and synthesized via radical copolymerization of N-vinylcarbazole and polymerizable
Eu(TTA)3vinyl-tpy, where TTA is 2-thenoyltrifluoroacetonate. The chemical structure and composition of the
Eu-polymers were characterized by FTIR, UV-vis, 1H NMR, 13C NMR spectroscopy, GPC, ESI-MS, and
elemental analysis. The geometry of the polymerizable monomer Eu(TTA)3vinyl-tpy was predicted using
the Sparkle/PM6 model and suggested to be in a chemical environment with very low symmetry around
the Eu3+ ions (C1), in agreement with the fluorescence spectrum. All Eu-polymers exhibited good solubility,
as well as good thermal stability and high glass transition temperatures. The photoluminescence (PL)
properties of the copolymers in solution and in the solid state were investigated in detail. Intramolecular
energy transfer from the carbazole groups to the europium complexes occurred even in diluted solutions.
The efficiency of this process also depended on the composition of the Eu-polymers. In the solid state,
emission from the carbazole groups was suppressed and the absorbed excitation energy was transferred
effectively to the europium complexes in the Eu-polymers. Most importantly, the EL performances of eight
pure red-emission PLEDs based on P1, P2, and P3 as the emitting layer have been studied in detail. Bright
electroluminescence with a maximum luminance of 68.2 cd mꢀ2 from the double-layer devices of P1 was
demonstrated. Although the EL performance was only the third-best among those of the Eu-chelated
polymers reported so far, to the best of our knowledge, this is the first example of electroluminescent
devices of Eu-polymers based on tpy as a neutral ligand. These results illustrate the potential application of
polymerizable tpy ligands in high performance EL Eu-chelated polymers.
Received 12th April 2013
Accepted 16th June 2013
DOI: 10.1039/c3tc30681j
www.rsc.org/MaterialsC
performance displays, such as for high contrast ratios. In fact,
for some special applications, such as photosignal identica-
1 Introduction
Photoelectronic metal-containing polymers have attracted
enormous attention because they have extensive applications in
polymeric light-emitting diodes (PLEDs),1 memory devices,2
polymer solar cells,3 and so on. It is believed that electrolumi-
nescent (EL) polymers are superior in low-cost and large-area
display device fabrication because of their solution process-
ability.4 However, as uorescent emitters and mixtures with
polydisperse molecules, EL polymers have the common disad-
vantages of low EL efficiency and poor color purity, and these
poor properties remarkably restrain their applications in high
tion and processing, the color purity is as important as the
emission intensity. A number of reported results indicate that
small molecular electro-phosphorescent dyes, such as transi-
tion-metal complex units have been introduced into the main
chains or side chains of polymers to construct electro-phos-
phorescent polymers, which can efficiently improve the EL
efficiencies.5
However, the full widths at half maximum (FWHM) of these
polymers were still around 100 nm, which becomes a main
constraint for their application in high color purity display
devices. As is commonly known, lanthanide complexes, espe-
cially Eu- and Tb-complexes, have excellent color purity.6 To
date, a number of organic Eu-complex luminescent materials
have been extensively studied as one kind of the most important
red-emitting materials in organic light-emitting diodes (OLEDs)
because of their pure red monochromic characteristic emission
with FWHM of 5–10 nm, high photoluminescent (PL) efficiency,
chemical environment stability and theoretical internal
quantum efficiency approaching 100%.7
aKey Laboratory of Polymer Materials for Electronics, Guangzhou Institute of
Chemistry, Chinese Academy of Sciences, Guangzhou 510650, P.R. China. E-mail:
mglu@gic.ac.cn
bSchool of Materials Science and Engineering, Chongqing University of Technology,
Chongqing 400054, P.R. China. E-mail: yclzjun@163.com
cUniversity of Chinese Academy of Sciences, Beijing 100039, P.R. China. E-mail:
mglu@gic.ac.cn
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c3tc30681j
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4885–4901 | 4885

View Article Online
Journal of Materials Chemistry C
Paper
To further improve the EL efficiencies of Eu-complexes, remarkable solid emission amplication phenomenon was
many bright and efficient electroluminescent devices based on observed. Most importantly, the EL performances of eight pure
small molecular Eu-complexes are achieved through doping/ red-emission PLEDs based on P1, P2, and P3 as the emitting
blending, in which 4,40-N,N0-dicarbazolebiphenyl (CBP),8 layer have been researched in detail. Bright electrolumines-
2-biphenyl-4-yl-5-(4-tert-butyl-phenyl)-1,3,4-oxadiazole (PBD),9 cence with a maximum luminance of 68.2 cd mꢀ2 from the
and polyvinylcarbazole (PVK)10 with high triplet excited energy double-layer devices of P1 was demonstrated. Although the EL
levels are oen used as the hosts. Though such doped devices performance is only third-best among those of the Eu-chelated
have the advantage of allowing the fabrication of thin lms by polymers reported so far, to the best of our knowledge, this is
spin-coating, blending and doping of Eu-complexes in matrices, the rst example of electroluminescent devices (including small
they have a tendency to show phase separation over time, molecular Eu-complexes and Eu-polymer electroluminescent
leading to instability in the device's performance.11 Fortunately, devices) of Eu-polymers based on tpy as a neutral ligand.
the problems can be avoided by the new approach of incorpo-
rating the Eu-complexes into the polymer matrix through
covalent linkages. It is believed that through the covalent
linkage with the polymeric host matrices the Eu-complex
2 Experimental section
Materials and instruments
moieties can be dispersed more uniformly in the hosts, and All reagents used were of analytical grade. DMF and acetonitrile
more efficient energy transfer from the main chains to the Eu- were dried with CaH2 for about 24 h and distilled at reduced
complex moieties can be expected.12
pressure. FTIR spectra were recorded using a Tensor 27 (Bruker)
Recently, Eu-containing polymers based on three b-diketone Fourier Transform Infrared Spectrometer. Elemental analysis
ligands and polymerizable neutral ligands, including carboxylic data were obtained from a Vario EL elemental analyzer. The
acids, 2,20-bipyridine (bpy) and phenanthroline (phen) deriva- molecular weight of the Eu-polymers was determined by Waters
tives, have been focused on.13 Several results have shown that 1515-2414 GPC gel permeation chromatography, using THF as
the maximum brightness is more than 10 cd mꢀ2, but the an eluent and polystyrene as the standard. Differential scanning
practical requirements for application still cannot be achieved. calorimetry (DSC) was conducted on a Pyris Diamond TALAB
ꢀ1
ꢁ
The reason for the low brightness for these PLEDs may be system at a heating rate of 20 C min under nitrogen. Ther-
attributed to the neutral ligands being inferior in polymeriza- mogravimetric analysis was conducted with a NETZSCH TG
tion and oen inducing an uncontrollable proportion of Eu- 209F3 at a heating rate of 15 ꢁC minꢀ1 under an N2 atmosphere
complexes in the copolymers. Therefore, the use of stable and over a temperature range from 35 to 650 ꢁC. All the ESI-MS
polymerizable Eu-chelated monomers is still imperative for spectra were recorded in a LCQ DECA XP mass spectrometer.
efficient luminescent copolymers.
NMR spectra were recorded on a DRX-400 MHz (Bruker)
Terpyridine (tpy), a classic chelating tridentate ligand for superconducting-magnet NMR spectrometer with TMS as an
transition metal and rare earth ions, has played an important role internal standard. UV-vis absorption spectra were recorded on a
in the development of coordination chemistry and still continues Shimadzu spectrophotometer (UV 2550). Cyclic voltammetry
to be of considerable interest as a versatile starting material for (CV) measurements were made on a computer-controlled
organic, inorganic and supramolecular chemistry.14 tpy is a rigid, CHI600D electrochemical analyzer with a Pt working electrode,
planar, hydrophobic, electron-accepting heteroaromatic system, a Pt plate counter electrode, and an SCE reference electrode
whose three nitrogen atoms are beautifully placed to act coop- immersed in 0.1 M Bu4NClO4 in dry acetonitrile purged with
eratively in cation binding. These structural features determine dried argon. The scanning rate was 50 mV sꢀ1, and all electro-
its coordination ability toward metal ions. As is well-known, chemical potentials were calibrated with the ferrocene/ferroce-
many studies have conrmed that tpy is an excellent neutral nium (Fc/Fc+) standard. The photoluminescence (PL)
ligand for Eu-complexes; it can reduce the non-radiative decay of measurements in the solid state and THF solution were con-
the excited states of the europium ion, improve the stability of the ducted on a Hitachi F-4600 uorescence spectrophotometer.
europium complexes, and increase the energy transfer efficiency Fluorescent lifetimes were obtained with an FLS920 steady state
from the ligands to the Eu3+ ions.15 Surprisingly, although pho- spectrometer with a pulsed xenon lamp.
toluminescent property studies of many Eu-complexes based on
4-(4-Vinylbenzyloxy)benzaldehyde. Solid KOH (2.25 g, 40
tpy as a neutral ligand have been reported, we cannot nd any mmol) was added to a solution of 4-hydroxybenzaldehyde (4.89
one example of research into the electroluminescence of Eu- g, 40 mmol) and 4-vinylbenzyl chloride (7.16 g, 40 mmol) in
complexes based on tpy as a neutral ligand.
DMF (100 mL), and the mixture was stirred for 12 h at room
In this work, a novel polymerizable vinyl-tpy ligand 40-(4-((4- temperature. Water (800 mL) was added to precipitate the
vinylbenzyl)oxy)phenyl)-2,20:60,200-terpyridine and the corre- product, which was isolated, washed with water (3 ꢂ 50 mL),
ꢁ
sponding Eu-complex were designed and synthesized. Three and dried in a vacuum at 60 C. Excess 4-vinylbenzyl chloride
copolymers, P1, P2 and P3 were prepared by using the Eu- was removed by triturating the product in hexane (50 mL). The
complex and N-vinylcarbazole as the monomers with composi- product was then washed with additional hexane (3 ꢂ 20 mL)
tions of 1 : 100, 1 : 50 and 1 : 30. To enhance the ligand-medi- and dried under a positive air ow. Yield: 8.11 g (85%). 1H NMR
ated energy transfer, PVK with good hole transporting (400 MHz CDCl3): 5.14 (s, 2H); 5.27–5.29 (d, 1H, JHH ¼ 11); 5.75–
properties was chosen as the host segment. Dramatically high 5.80 (d, 1H, JHH ¼ 18); 6.69–6.76 (dd, 1H, JHH ¼ 18, JHH ¼ 12);
PL efficiencies of around 40% in the solid state were realized. A 7.06–7.09 (d, 2H, JHH ¼ 9); 7.38–7.46 (dd, 4H, JHH ¼ 8, JHH ¼ 8);
4886 | J. Mater. Chem. C, 2013, 1, 4885–4901
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
7.83–7.85 (d, 2H, JHH ¼ 8); 9.89 (s, 1H). ESI-MS: m/z 242.4 (M + d 162.51 (carbon of Ar), 156.43 (carbon of tpy), 155.89 (carbon of
H)+. FTIR (KBr pellet, cmꢀ1): 2935, 1706 (C]O stretching), 1601, tpy), 149.12 (carbon of tpy), 139.96 (carbon of carbazole), 137.25
1512, 1380, 1261, 1160, 1008, 832.
(carbon of carbazole), 125.06 (carbon of carbazole), 123.66 (carbon
40-(4-((4-Vinylbenzyl)oxy)phenyl)-2,20:60,200-terpyridine (vinyl- of carbazole), 121.86 (carbon of carbazole), 120.10 (carbon of
tpy). 2-Acetylpyridine (2.42 g, 20 mmol) was added into a solu- carbazole), 118.95 (carbon of carbazole), 110.18 (carbon of
tion of 4-(4-vinylbenzyloxy)benzaldehyde (2.38 g, 10 mmol) in carbazole), 107.96 (carbon of carbazole), 102.25, 69.81 (carbon
EtOH (100 mL). KOH pellets (1.55 g, 85%, 20 mmol) and of ethyl), 65.05 (carbon of TTA). FTIR (KBr pellet, cmꢀ1): 1601,
aqueous NH3 (30 mL, 29.3%, 25 mmol) were then added to the 1536 (C]C stretching in TTA), 1492, 1455, 1329, 1221, 1158, 1124,
solution. The solution was stirred at room temperature for 12 h. 747, 718. Elemental analysis calcd (%): C 83.94, H 6.43, N 6.94.
The milk-white solid was collected by ltration and washed with Found: C 84.12, H 6.27, N 7.04. Molecular weight: Mn ¼ 10 053,
EtOH (3 ꢂ 20 mL). Yield: 2.73 g (62%). 1H NMR (400 MHz Mw ¼ 18 438, PDI ¼ 1.83. Content of Eu3+ ¼ 0.73 wt%. Polymer
CDCl3): 8.65–8.71 (m, 6H, tpy), 7.83–7.88 (t, 4H, ArH), 7.40–7.45 composition: n : m ¼ 1 : 100, determined according to the results
(m, 4H, ArH), 7.26–7.33 (t, 2H, tpy), 7.07–7.09 (d, 2H, ArH), 6.70– of elemental analysis and Eu3+ titration.
6.77 (m, 1H, CH]CH2), 5.75–5.80 (d, 1H, CH]CH2), 5.25–5.28
P2: with yield of 0.38 g (65%) as milk-white powder. 1H NMR
(d, 1H, CH]CH2), 5.08 (s, 2H, –CH2–). 13C NMR (100 MHz (400 MHz, CDCl3): d 8.05, 7.68, 7.43, 6.97, 6.89, 6.50, 6.37, 4.77,
CDCl3): 159.5, 156.2, 155.7, 149.5, 148.9, 137.1, 136.8, 136.3, 3.50, 3.16, 2.48, 1.40–1.92, 0.94–1.24 ppm. 13C NMR (100 MHz,
130.9, 128.4, 127.7, 126.4, 123.7, 121.3, 118.2, 115.1, 114.1, 69.7. CDCl3): d 162.50, 156.43, 155.90, 149.11, 139.93, 137.22, 125.06,
ESI-MS: m/z 442.3 (M + H)+. FTIR (KBr pellet, cmꢀ1): 1603, 1584, 123.56, 121.87, 120.13, 118.95, 110.16, 107.96, 102.25, 69.83,
1566, 1513, 1463, 1386, 1252, 1183, 1038, 992, 830, 793, 734.
65.04 (carbon of TTA). FTIR (KBr pellet, cmꢀ1): 1600, 1536 (C]C
Preparation of the europium monomer Eu(TTA)3vinyl-tpy stretching in TTA), 1484, 1451, 1336, 1227, 1159, 1124, 743, 719.
(EuVTPY). The complex was prepared according to the well- Elemental analysis calcd (%): C 82.09, H 6.23, N 6.73. Found: C
established method.16 Vinyl-tpy (0.44 g, 1 mmol) was dissolved 82.15, H 6.06, N 6.85. Molecular weight: Mn ¼ 8228, Mw ¼ 15367,
in THF (50 mL), to which Eu(TTA)3$2H2O (0.85 g, 1 mmol) was PDI ¼ 1.87. Content of Eu3+ ¼ 1.38 wt%. Polymer composition:
added. The whole mixture was reuxed for 3 h and cooled to n : m ¼ 1 : 50, determined according to the results of elemental
room temperature. The resulting precipitate was collected and analysis and Eu3+ titration.
1
washed twice with water to give the target complex (0.93 g, 74%)
P3: with yield of 0.32 g (51%) as milk-white powder. H NMR
as a light-yellow powder. (Found: C, 50.08; H, 2.94; N, 3.19. (400 MHz, CDCl3): d 8.74, 8.09, 7.99, 7.51, 7.68, 6.98, 6.44, 6.50,
EuC54H35F9N3O7S3 [EuVTPY] requires C, 51.60; H, 2.81; N, 5.14, 4.89, 3.48, 3.16, 2.72, 0.86–1.75 ppm. 13C NMR (100 MHz,
1
3.35%). H NMR (400 MHz CDCl3): 14.34 (s, 2H, tpy), 12.87 (s, CDCl3): d 162.48, 156.43, 155.81, 149.15, 139.98, 137.21, 125.06,
2H, tpy), 11.20 (s, 2H, tpy), 10.76 (s, 2H, tpy), 8.19 (s, 2H, ArH), 123.69, 121.86, 120.05, 118.95, 110.15, 107.96, 102.18, 69.81, 65.01.
7.24–7.68 (m, 4H, ArH), 6.95 (s, 2H, ArH), 6.82–6.85 (m, 3H, Th– FTIR (KBr pellet, cmꢀ1): 1601, 1536 (C]C stretching in TTA), 1484,
H), 6.72–6.79 (m, 1H, CH]CH2), 6.48 (m, 3H, Th–H), 6.28 (s, 1333, 1222, 1160, 747, 726. Elemental analysis calcd (%): C 79.92,
2H, Th–H), 5.61–5.68 (m, 4H, CH]CH2), 5.23–5.27 (m, CH] H 5.99, N 6.49. Found: C 80.01, H 5.87, N 6.55. Molecular weight:
CH2), 5.14 (s, 2H, CH2). ESI-MS m/z: [Eu(TTA)2vinyl-tpy]+ calcu- Mn ¼ 6226, Mw ¼ 10 396, PDI ¼ 1.67. Content of Eu3+ ¼ 3.69 wt%.
lated for EuC46H31F6N3O5S2, 1036.08; found, 1036.0. FTIR (KBr Polymer composition: n : m ¼ 1 : 30, determined according to the
pellet, cmꢀ1): 1617 (C]O stretching in TTA), 1538 (C]C results of elemental analysis and Eu3+ titration.
stretching in TTA), 1417, 1312, 1188, 1139, 1058, 780.
Calculation of quantum efficiency (h) in solid state. The
Preparation of the Eu-chelated copolymers (Eu-polymer P1, intrinsic uorescent quantum efficiency (h) of the 5D0 emission
P2 and P3). A mixture of N-vinyl carbazole (0.51 g, 2.6 mmol), level in EuVTPY, P1, P2, and P3 at room temperature was
EuVTPY (33 mg, 0.027 mmol for P1, 68 mg, 0.054 mmol for P2, obtained based on the luminescence data (emission spectra and
and 113 mg, 0.090 mmol for P3), and AIBN (2,20-azobisisobutyr- emission decay curves of these Eu-complexes). Eqn (1) is a
onitrile) initiator (6 mg, about 1 wt% of the total monomers) was means to determine the h values from experimental spectro-
dissolved in dry DMF (3 mL) in a glass polymerization tube. The scopic data:25
homogeneous solution was purged with argon for 5 min and
Arad
sealed under a reduced argon atmosphere. The mixture was
h ¼
(1)
ꢁ
Arad þ Anrad
heated to 65 C with continuous stirring for 24 h. The reaction
mixture remained clear throughout the copolymerization
process. The viscous mixture was diluted with DMF (3 mL) and
precipitated into methanol (50 mL) under vigorous stirring. The
reprecipitation procedure was repeated three times. The result-
ing solid material was collected by ltration. The copolymer was
further puried by Soxhlet extraction with boiling acetone for 48
h and nally dried in a vacuum oven at 60 ꢁC for 24 h. The feed
ratios in the polymerization experiments are shown in Table S1.†
P1: with yield of 0.45 g (83%) as milk-white powder. 1H NMR
(400 MHz, CDCl3): d 7.68, 7.39, 6.89, 6.44, 4.88, 3.41, 3.21, 2.56,
2.19, 1.66–1.73, 0.87–1.53 ppm. 13C NMR (100 MHz, CDCl3):
where, Arad and Anrad are radiative and nonradiative transition
rates, respectively. The denominator in eqn (1) is calculated
from the lifetime of the emitting level (1/s ¼ Arad + Anrad). In the
case of europium uorescence the value of Arad can be estimated
by spectral analysis with the help of eqn (2):
4
X
A0ꢀ1hu0ꢀ1
S0ꢀJ
Arad
¼
(2)
S0ꢀ1
J¼0 hu0ꢀJ
7
where, J represents the nal F0–6 levels, S is the integrated
intensity of the particular emission lines and hu stands for the
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4885–4901 | 4887

View Article Online
Journal of Materials Chemistry C
Paper
corresponding transition energies. A0–1 is the Einstein coeffi- efficiencies were determined by a Si photodiode with calibration
cient of spontaneous emission between the 5D0 and the 7F1 in an integrating sphere (IS080, Labsphere).
Stark levels. The branching ratios for the 5D0 / 7F5,6 transitions
must be neglected as they are too weak to be observed experi-
3 Results and discussion
3.1 Geometry optimization, and simulation of UV and IR
mentally. Therefore, their inuence can be ignored in the
depopulation of the 5D0 excited state. The 5D0 / 7F1 transition
spectra for EuTPY
is independent of the local ligand eld seen by the europium
ions and, thus, may be used as a reference for the whole spec- In the eld of coordination compounds, the semiempirical
trum, in vacuo A0–1 ¼ 14.65 sꢀ1
.
An average refractive index Sparkle/PM6 model has proven to be effective in lanthanide
26
equal to 1.5 was considered, leading to A0–1 z 50 sꢀ1
.
chemistry because it allows the prediction of the coordination
Calculation of quantum yield (F) in THF solution. Solution geometry for both small lanthanide complexes and more
uorescent yields of EuVTPY, P1, P2, and P3 were determined sophisticated structures in a relatively short time and with a low
using quinine sulfate (dissolved in 0.5 M H2SO4 with a computational demand.17 Recent results have shown good
concentration of 10ꢀ6 M, assuming FPL of 0.55) as a standard.27 prediction for the ground state geometry, when compared to
The quantum yield was calculated according the following single crystal data.18 Here the ground state geometry shown in
equation:
Fig. S1† was calculated using Sparkle/PM6 implemented in the
Mopac2012 package. The Sparkle model replaces the lantha-
nide ion by a core with +3e charge, so that only the electrostatic
interaction between the ion and the ligand is considered. The
Eu3+ ion in this compound is nine-coordinate and the coordi-
nation polyhedron can be approximately described as a tricap-
ped trigonal prism, and the europium complex belongs to the
C1 point-group. The structural parameters of the coordination
polyhedron, shown in Table S2,† are given by the distance
between the europium ion and ligand (R), the angle between the
z axis and the ligand atom (q) and the angle of the projection of
vector r in the xy plane and the x axis (4). The average distance
Ar S n2
F ¼ Fr
(3)
2
A Sr nr
where F is the uorescence quantum yield, S represents the
area of the corrected emission uorescence spectrum, A is the
absorbance of the solution at the excitation wavelength, and n is
the refractive index of the solvent used. The subscript r denotes
the reference substance whose uorescence quantum yield is
already known.
Fabrication and testing of PLEDs. Single-layer and double-
layer PLEDs were fabricated by spin coating with three cong-
urations of ITO/PEDOT (40 nm)/Eu-polymer (80 nm)/Ba (4 nm)/
Al (120 nm), ITO/PEDOT (40 nm)/PVK (40 nm)/Eu-polymer (80
nm)/Ba (4 nm)/Al (120 nm) and ITO/PEDOT (40 nm)/Eu-poly-
mer:OXD-7 (30%, 40%, 80 nm)/Ba (4 nm)/Al (120 nm), where
PEDOT is poly(3,4-ethylenedioxythiophene) as the hole injec-
tion material, OXD-7 is 1,3-bis[(4-tert-butylphenyl)-1,3,4-oxi-
diazolyl]phenylene as the electron-transporting material, and
ITO and Ba/Al were used as the anode and cathode, respectively.
PLED was fabricated on patterned indium tin oxide (ITO) with a
sheet resistance of 15–20 U per square. The substrate was
ultrasonically cleaned with acetone, detergent, deionized water,
and 2-propanol. Oxygen plasma treatment was applied for 4
min as the nal step just before lm coating. Onto the ITO glass
was spin-coated a layer of PEDOT lm with a thickness of 40 nm
from its aqueous dispersion. The PEDOT lm was dried at 80 ꢁC
for 3 h in a vacuum oven. The solution of the polymer was
prepared under nitrogen atmosphere and spin-coated onto the
PEDOT layer. The typical thickness of the emitting layer was 80
nm. Then a thin layer of barium as an electron injection
cathode and the subsequent 140 nm thick aluminum protection
layers were thermally deposited by vacuum evaporation through
a mask at a base pressure below 2 ꢂ 10ꢀ4 Pa. The cathode area
denes the active area of the device. The typical active area of
the devices in this study is 0.15 cm2. The EL layer spin coating
process and the device performance tests were carried out
within a glovebox under a nitrogen atmosphere. Current–
voltage (I–V) characteristics were recorded with a Keithley 236
source meter. EL spectra were obtained by an Oriel Instaspec IV
CCD spectrograph. Luminance was measured by a PR 705
photometer (Photo Research). The external quantum
˚
for Eu–O (TTA) of 2.403 and Eu–N (tpy) of 2.551 A are close to the
distances estimated using X-ray diffraction data for the corre-
sponding single crystal Eu-complex based on TTA and terpyridyl
derivatives.19
As is well known, knowledge of the location of orbitals in
luminescent lanthanide research is important. Therefore we
calculated the molecular orbitals (MOs) of EuVTPY using
Gabedit soware using the optimized data by Mopac2012.
Fig. 1(a) and (b) show the HOMO and LUMO of EuVTPY; the
calculated results indicate that the HOMO is mostly localized on
the vinylbenzene moiety, and the LUMO is mostly localized on
the tpy group. Molecular magnets are a very active research area.
Recently, there has been interest in the development of
molecular magnets with paramagnetic lanthanide ions.20
Fig. 1(c) shows the spin density distribution along the EuVTPY
molecule, the blue balloons are considered to be regions where
the spin density is positive and the red balloons are the regions
where the spin density is negative. It is seen that the spin
density of the Eu-complex is mainly located at the vinylbenzene
moieties, and is mostly negative; meanwhile, little positive spin
density in the vicinity of the para-C–C of the benzene group can
be found. The electron density distribution of EuVTPY is dis-
played in Fig. 1(d);the distribution of the electron density in this
complex is very interesting, it is mostly located on the vinyl-
benzene and TTA moieties, however, electron density in the tpy
moiety almost cannot be observed. Obtaining the electron
density distribution may help understand the energy transfer
progress, and will be studied in detail in future work.
The experimental and simulated FTIR spectra of the mono-
mer EuVTPY are shown in Fig. S2.† The calculation was made
4888 | J. Mater. Chem. C, 2013, 1, 4885–4901
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
so we predict that the monomer EuVTPY and the corresponding
Eu-polymers P1, P2 and P3 will have excellent PL and EL
properties. The synthesis procedures of EuVTPY and P1, P2 and
P3 are outlined in Scheme 1. In the rst step, 4-(4-vinyl-
benzyloxy)benzaldehyde is obtained from 1-(chloromethyl)-4-
vinylbenzene and 4-hydroxybenzaldehyde reacting in DMF in
the presence of KOH at room temperature in a high yield (85%).
40-(4-((4-Vinylbenzyl)oxy)phenyl)-2,20:60,200-terpyridine
was
formed by the condensation of 2-acetylpyridine and 4-(4-vinyl-
benzyloxy)-benzaldehyde, according to a slightly modied tpy
derivate preparation method.21 The chromophore monomer
EuVTPY was synthesized according to the well-established
procedure with a high yield of 74%. The Eu-polymers P1, P2 and
P3 were prepared by the classic free radical copolymerization of
EuVTPY and N-vinylcarbazole (NVK), using AIBN as the initi-
ator, in dry DMF. All of the compounds were characterized by
NMR, ESI-MS, FTIR, and elemental analysis.
Compared with the ligand vinyl-tpy, in the FTIR spectrum of
polymerizable EuVTPY, three new absorbance peaks were
observed at 1613, 1538 and 1413 cmꢀ1 (Fig. 2). They are
attributed to the C]O and C]C stretching vibrations of the
coordinated TTA ligand and the C]N stretching vibration of
the coordinated vinyl-tpy ligand in the Eu-complex, respectively.
These results indicate that the neutral ligand vinyl-tpy and the
ionic ligands TTA had successfully coordinated with the Eu3+
ion emission centers. The FTIR spectra of the Eu-polymers are
shown in Fig. 2, and are very similar to the FTIR spectrum of
pure PVK. Two sharp and strong absorption peaks at 750 and
722 cmꢀ1 are attributed to the characteristic absorptions of the
carbazole moieties. Because of the formation of polymer and
the low content of EuVTPY in the polymers, the strong and
widely-dispersed absorption peaks of the PVK segments almost
cover all of the characteristic absorption peaks of EuVTPY, such
as the TTA absorption band at 1613 cmꢀ1. Fortunately, covalent
attachment of the Eu-complex unit to the polymer backbone is
strongly supported by the C]C stretching vibrations of the
coordinated TTA ligand around 1536 cmꢀ1, which is a strong
absorption peak in the spectrum of EuVTPY and a much weaker
peak in those of the Eu-polymers.
Fig. 1 (a) HOMO and (b) LUMO of the monomer EuVTPY. The red and blue lobes
denote the positive and negative phases of the coefficients of the molecular
orbitals. The size of each lobe is proportional to the MO coefficient. (c) Calculated
spin density distribution at the EuVTPY surface (the blue balloons indicate regions
where the spin density is positive and the red balloons the regions where the spin
density is negative). (d) Calculated electronic density distribution at the EuVTPY
surface. (Light green, black, blue, pink, blue-green, yellow, and red balls corre-
spond to hydrogen, carbon, nitrogen, oxygen, fluorine, sulfur, and europium
atoms, respectively.)
for a free molecule in vacuum, while the experiment was per-
formed for the solid state; furthermore, the anharmonicity in
the real system was neglected for the vibration calculations.
Therefore, they are disagreements between the calculated and
experimental vibrational wavenumbers. Nevertheless, the
characteristic stretching peaks of EuVTPY can be realized, these
absorbance peaks located at 1725, 1640 and 1453 cmꢀ1 are
attributed to the C]O and C]C stretching vibrations of the
coordinated TTA ligand and C]N of the coordinated vinyl-tpy
ligand in the Eu-complex, respectively. A comparison of the
calculated and experimental UV spectra for the monomer
EuVTPY is presented in Fig. S3.† As can be seen, the agreement
between the simulated and experimental spectra is very good.
Whether in the calculated or in the experimental spectrum, two
main absorption bands of the monomer EuVTPY are easily
observed, which are attributed to the singlet–singlet p / p enol
absorption of TTA and tpy moieties in the monomer Eu-
complex. It is noteworthy that the last absorption band (located
at near 344 nm) in the simulation is very similar to that in the
experiment. However, an obvious red shi of the rst absorp-
tion band in the calculated spectrum can be observed, which
might be attributed to the solvent effects neglected in the
calculation or any errors inherent in the method itself.
GPC analysis (using poly-styrene as the standard for cali-
bration) showed that the number-average molecular weight
(Mn) and polydispersity indexes (PDI) of the Eu-polymers are
10 053, 1.83 for P1, 8228, 1.87 for P2, and 6226, 1.67 for P3. The
appropriate molecular weight and relative narrow PDI are
benecial to form uniform thin lms and enhance the
mechanical capacities of the lms. According to the results of
the elemental analysis and titration experiments, the compo-
3.2 Design, synthesis, and characterization
With the aim of developing the rst PLEDs of Eu-polymers sitions of the copolymers (n : m molar ratio) were about 1 : 100,
based on tpy as a neutral ligand, a novel tridentate ligand vinyl- 1 : 50 and 1 : 30. The feed ratios of Eu-polymer are very close to
tpy and a corresponding polymerizable monomer EuVTPY have its actual composition, which means that the reactivity of
been designed and synthesized. To increase the energy transfer EuVTPY to NVK is similar with the reactivity between NVKs.
efficiency from the ligand to the emission center of the Eu3+ However, it is noticed that the molecular weight and yield of the
ions, blue-emitting PVK (which possesses good hole trans- copolymers remarkably decrease along with an increase of the
porting properties) was chosen as the host segment in these Eu- EuVTPY : NVK feed ratio. In addition, the differences between
polymers. Most importantly, due to the neutral ligand vinyl-tpy the copolymer compositions and the feed ratios increase from
has a strong coordination ability, structure adjustment ability, P1 to P3. The main reason might be that along with the increase
protuberant coordinate sites and inertia towards modication, of the molar quantity of EuVTPY, the possibility of collision
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4885–4901 | 4889

View Article Online
Journal of Materials Chemistry C
Paper
Scheme 1 Synthetic procedures of the monomer EuVTPY and Eu-polymers P1, P2 and P3.
molecular weight of P2 and P3. The resulting copolymers are
readily soluble in common organic solvents, such as THF,
dichloromethane, chloroform and toluene, and can be easily
cast into transparent and uniform thin lms.
The 1H NMR spectrum of the Eu-polymer P2 obtained at
400 MHz in chloroform solution is shown in Fig. 3. The spec-
trum consists of several broad peaks with chemical shis in the
ranges of 4.5–8.0 and 0.50–4.0 ppm, which are associated with
the aromatic and alkyl protons, respectively. It has been
reported that the two aromatic rings of the carbazole group are
magnetically non-equivalent, despite the fact that the carbazole
moiety belongs to the C2v symmetry group.22,32 This phenom-
enon is due to the fact that, in the polymer chain, the neigh-
boring carbazole units may be partially or totally overlapped in
different stereochemical sequences (meso or racemic confor-
mations, as shown in structure (a) of Fig. 3); similar phenomena
have been observed in poly(c-glutamic acid) ester systems.33 The
non-equivalence of the two carbazole rings is reected in the 1H
NMR spectrum. For example, the more severely shielded proton
Fig. 2 The FTIR spectra of vinyl-tpy, monomer EuVTPY and Eu-polymers.
between EuVTPY units also increases. Although the tridentate H1 appears in the high eld region (4.87 ppm), while the proton
tpy ligand has a suitable structure, the polymerization between H8 of the less severely shielded ring appears at 6.35 ppm. The
bulky EuVTPY molecules is much more difficult than that ring current effect is the main contribution to this difference in
between EuVTPY and NVK. Thus, these collisions might induce chemical shis. The other peaks in the specied low eld region
chain termination or chain transfer, which reduce the are reasonably assigned to the aromatic protons. The
4890 | J. Mater. Chem. C, 2013, 1, 4885–4901
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
Fig. 3 1H NMR (400 MHz) and 13C NMR (100 MHz) spectra of P2 in CDCl3. (a) The meso (m) and racemic (r) conformations of the neighboring carbazole (Cz) units; (b)
the head-to-tail (h–t), tail-to-tail (t–t) and head-to-head (h–h) sequences in the polymer chain.
resonances of methenyl protons (H9, H11) and methylene heterotactic (mr) sequences existing in the polymer chain. The
protons (H10, H12) in the polymer backbone appear with multi- other aliphatic carbon atoms (C10–C12) resonate at a higher
peaks in the regions of 2.4–3.5 ppm and 1.1–1.7 ppm, respec- eld (33–40 ppm), compared to that of carbon C9. Again, the
tively. This observation is attributed to the existence of different resonance signals of the carbon atoms in the ligands TTA and
stereo-chemical sequences in the polymer chain (structures (a) vinyl-tpy, which are located in the resonance region of the
and (b)). Due to a low content of EuVTPY in the Eu-polymers, carbazole carbons, are difficult to recognize. However, the ethyl
the chemical shis of the protons in the TTA and tpy groups are carbon (C17) and phenyl carbon (C18) of the vinyl-tpy are
weak and buried by those of the carbazole group.
bonded to an electron-withdrawing atom (O) and the resonance
The chemical shis in the 13C NMR spectrum of Eu-polymer is, therefore, shied to lower eld, compared to the normal
P2 are expanded in several regions (Fig. 3). The non-equivalence ethyl carbon and phenyl carbon. Owing to the effects of the
of the two carbazole rings is also reected in the carbon oxygen atoms bonded to the two neighboring carbon atoms, the
chemical shis. Chemical shis for all the different carbon aliphatic carbon (C22) of the TTA group is also detected at a
species in the carbazole units are clearly shown in the 12 groups lower eld (at about 63 ppm), compared to the other aliphatic
of broad peaks.23,32 The resonance of carbon C9, which is carbon atoms in the backbone. In addition, due to the effect of
bonded to a nitrogen atom, appears clearly as three peaks electron-withdrawing atoms (N), the carbons C19, C20 and C21
located at 48.2, 49.1 and 50.1 ppm. These three chemical shis of the tpy groups are observed at low eld, and are located at
correspond to the isotactic (mm), syndiotactic (rr) and about 156.4, 155.9 and 149.2 ppm, respectively. Based on the
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4885–4901 | 4891

View Article Online
Journal of Materials Chemistry C
Paper
above, the NMR results further conrm the successful forma-
tion of the designed Eu-polymers.
3.3 Thermal properties
The thermal stability of the Eu-complexes is very important
because decomposition leads to decreased EL performance. The
differential scanning calorimetry (DSC) analysis of the Eu-
polymer was measured to investigate their phase stability
(Fig. 4). The glass transition temperature (Tg) of P1 is 171.7 ꢁC,
which is lower than that of the PVK homopolymer (Tg ¼ 200 ꢁC).
It is noted that the Tg values of P2 and P3 are reduced to 163.5 ꢁC
and 143.1 ꢁC, respectively, which means that the rigidity of the
Eu-polymers was reduced with increasing Eu-chelated content.
The Tg values of P2 and P3 are lower than that of P1, and this
can be ascribed to the increase of the content of Eu-chelated
moieties accompanied with the steric effect of the related
macromolecular monomer EuVTPY and the disruption of the
ordered structures of the entire polymers. Nevertheless, the
phase stability of P1, P2 and P3 is still favorable among
the soluble light-emitting polymers. The thermal stability of Eu-
polymers was investigated by thermogravimetric analysis (TGA)
(Fig. 5). The temperatures of the thermal decomposition (Td, 5%
weight loss temperature) of P1, P2 and P3 are 294.8, 276.9 and
261.1 ꢁC, respectively, which are lower than that of PVK (439 ꢁC).
Obviously, since the coordinate bond is much weaker, the initial
decomposition process mainly involves the rupture of the O–Eu
bond in EuVTPY and decomposition of Eu(TTA)3. Nevertheless,
P1, P2 and P3 still have good thermal stability with 5% weight
Fig. 5 TGA curves of P1, P2 and P3.
in the monomer Eu-complex. Due to the formation of larger
conjugated chelate rings in EuVTPY, those main absorption
bands shi to longer wavelengths compared with correspond-
ing ligands. Because the content of Eu3+ chelated segments in
the polymers is rather small, in the UV-vis absorption spectra of
P1, P2 and P3 in dilute THF solution, all of the absorption bands
from the monomer EuVTPY are completely buried by the strong
bands at 250, 260, 294, 330, and 344 nm corresponding to the
transitions of the carbazole groups.
The excitation spectra of EuVTPY, P1, P2 and P3 in a solid
and in THF solution were obtained by monitoring the 5D0 / 7F2
transition of the europium ion, which is shown in Fig. 6 and 7.
For solution excitation spectra, one broad excitation band at
288–404 nm can be easily observed for EuVTPY, P1, P2 and P3.
In the solid state excitation spectra, it can be seen clearly that
the intense broad band between 200 nm and 423 nm dominates
large portions of the excitation spectra of all the Eu-complexes,
which is attributed to the p / p* transitions of TTA, tpy, and
ꢁ
loss occurring at greater than 260 C.
3.4 Optical properties
The UV-vis absorption spectra for vinyl-tpy, EuVTPY, P1, P2 and
P3 in THF solution (1 ꢂ 10ꢀ5 mol Lꢀ1) are shown in Fig. S4†.
Two main absorption bands of the monomer EuVTPY are easily
observed, at 286 nm and 344 nm, which are attributed to the
singlet–singlet p / p enol absorption of TTA and tpy moieties
Fig. 6 The PL spectra of EuVTPY and P1, P2, and P3 in the solid state. Left:
excitation spectra (lem ¼ 615 nm), right: emission spectra (lex ¼ 360 nm) (+: false
peak).
Fig. 4 DSC curves of P1, P2 and P3.
4892 | J. Mater. Chem. C, 2013, 1, 4885–4901
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
the carbazole moieties from the comparison of the UV-vis
spectra in Fig. S4.† In particular, in the excitation spectrum of
the monomer EuVTPY in the solid state, other than the intense
broad bands, a weak excitation peak was also observed at
465 nm, which results from the 5D0
/
7F2 transition of the
europium ion. In comparison with the excitation broad bands
of the ligands, the direct excitation peaks of the Eu3+ ions were
much weaker. This result suggests that uorescence sensitiza-
tion by excitation of the ligands was much more efficient than
direct excitation of the Eu3+ ion absorption levels.
The PL spectra of the monomer EuVTPY, P1, P2 and P3 in
THF solution (1 ꢂ 10ꢀ5 mol Lꢀ1) are shown in Fig. 7. As with
most Eu-chelated PVK copolymers, the PL spectra of P1, P2 and
P3 in THF solution consist of two groups of emission peaks. The
rst group from about 350 to 500 nm contains the emission
bands originating from p* / p transitions of the carbazole
moieties. The second group comprises the characteristic emis-
Fig. 8 Left: CIE coordinate diagram of EuVTPY, P1, P2, and P3 in the solid state
and in THF solution. Right: luminescent image of Eu-complexes excited at 365 nm
sions of Eu-chelated compounds at 580, 594, 615, 653 and in the solid state and in THF solution.
697 nm, which originate from the Eu3+ ion corresponding to 5D0
7
/ Fj (j ¼ 0–4) transitions. For the Eu-polymer P1, due to the
content of EuVTPY units being rather low, the emission of solid enhances the intermolecular interactions, which facili-
¨
carbazole is much stronger than that in P2 and P3, and the color tates the inter- and intrachain Forster energy transfer. More-
as observed by the naked eye in THF solution appears pink over, the shorter intermolecular distance and the more effective
under UV 365 nm excitation (Fig. 8), and the CIE chromaticity inter- and intrachain winding and wrapping of Eu-chelated
coordinates from the emission spectra are (0.31, 0.12). For P2 moieties increase the possibility of electron overlap between the
and P3, with the content of EuVTPY units increasing, the PVK sequences and the EuVTPY segments, which consequently
emission intensity of carbazole gradually decreases, and the improves the Dexter-type energy transfer. Thus, in the solid
color changes from pink (P1) to pure red (P3), and the CIE state the strong intermolecular interactions remarkably facili-
coordinates are (0.58, 0.28) and (0.60, 0.29) for P2 and P3, tate the energy transfer from the blue-emitting PVK sequences
respectively. Fig. 6 shows the PL spectra of the monomer to the red-emitting EuVTPY moieties. Because the copolymer is
EuVTPY, P1, P2 and P3 in the solid state. In contrast to the PL designed to have integration between the host and guest, the
spectra of the Eu-polymer P1, P2 and P3 in solution, the PL high efficiency energy transfer between the different moieties is
spectra of the Eu-polymer is dominated by red emission from the basis of the high performance of the PL and EL. Although
the EuVTPY moieties. For the Eu-polymers P1, P2 and P3 in the the Eu-complex content is low (ca. 0.73% for P1, 1.38 for P2, and
solid state, the blue emission attributed to the PVK sequences is 3.69 for P3) in these Eu-polymers, the energy transfer is also
too weak to be recognized. The molecular aggregation in the highly efficient. Therefore, nding the balance point between
efficient host–guest energy transfer and limited concentration
quenching is most likely. In addition, the monomer EuVTPY
and the Eu-polymers (P1, P2 and P3) exhibit characteristic red
emission of europium ions under UV 365 nm (Fig. 8) excitation,
and it is suggested that these complexes can be potential red
uorescent materials. The CIE chromaticity coordinates of
EuVTPY, P1, P2, and P3 from the emission spectra are (0.67,
0.33), (0.61, 0.30), (0.66, 0.32) and (0.66, 0.33) in the solid state,
respectively, which indicates pure red emission. The results are
important because they indicate an advantage for rare-earth
complexes for preparing PLEDs.
Whether in the solid state or in THF solution, the intensity
ratios (I2/I1) of these Eu-complexes were very high. In the solid
state, the intensity ratios (I2/I1) of EuVTPY, P1, P2 and P3 were
11.35, 12.10, 12.54, and 14.21, respectively, and in THF solution
they were 16.49, 12.47, 13.86, and 14.08, respectively. This ratio
was only possible when the Eu3+ ion did not occupy a site with
inversion symmetry. It was clear that strong coordination
interactions took place between the ligands and Eu3+ ion.
Fig. 7 The PL spectra of EuVTPY and P1, P2, and P3 in THF solution (1 ꢂ 10ꢀ5
mol Lꢀ1). Left: excitation spectra (lem ¼ 615 nm), right: emission spectra (lex
¼
Furthermore, the emission spectra of the complexes showed
359 nm).
only one line for the 5D0 7F0 transition, indicating the
/
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4885–4901 | 4893

View Article Online
Journal of Materials Chemistry C
Paper
presence of a single chemical environment around the euro- Judd–Ofelt intensity parameters U2 and U4 can be calculated
pium ions.24
from the emission spectra as in ref. 28. In particular, U2 is more
To better understand the uorescent properties of EuVTPY, sensitive to the symmetry and sequence of ligand elds. To
P1, P2 and P3 in the solid state and in THF solution, the room obtain faster europium radiation rates, antisymmetrical euro-
5
temperature (RT) uorescence decay curves of the D0 excited pium complexes with large U2 parameters need to be designed.
state were measured by monitoring the most intense emission The spontaneous emission probability A0–l (l ¼ 2, 4) of the
7
lines (5D0 / F2) of the europium ion center at 616 nm, and transitions are related to its dipole strength according to the
under excitation by a 360 nm Xenon lamp. As shown in Fig. S5,† equation:
whether in a solid or in THF solution, the decay curves of the
A0–l ¼ (64p4n3)/|3h(2J + 1)|h|n(n2 + 2)/9|S(ED) + S(MD)
i
(4)
complexes exhibited monoexponential behavior, indicative of
the presence of a single chemical environment around the Eu3+
ion in these Eu-complexes, which was in agreement with the
Here, n is the average transition energy in cmꢀ1, h is the
Planck constant, and 2J + 1 is the degeneracy of the initial state.
S(ED) and S(MD) are the electric dipole strength and magnetic
dipole strengths, respectively. Among all these transitions, the
5
results of only one D0 / 7F0 line in the emission spectra and
the calculated results of Sparkle/PM6. The uorescent lifetime
values (s) of EuVTPY, P1, P2, and P3 are 0.63, 0.60, 0.63 and
0.74 ms in the solid, and 0.48, 0.49, 0.50 and 0.56 ms in THF
solution, respectively.
7
5D0 / F0,3,5 transitions are forbidden both in magnetic and
electric dipole schemes (S(ED) and S(MD) are zero). In addition,
5
7
the D0 / F1 transition is the isolated magnetic dipole tran-
sition and has no electric dipole contribution, which is practi-
cally independent of the lanthanide ion chemical environment
and can be used as a reference. The Judd–Ofelt parameters U2
and U4 can be calculated according to the equation:
The uorescent quantum efficiency (h) in the solid and
quantum yield (F) in THF solution of EuVTPY, P1, P2, and P3 at
room temperature were calculated and are shown in Table 1.
In the solid state, the monomer EuVTPY, P1, P2 and P3
exhibited high uorescent quantum efficiency; the values are
37.92, 38.21, 42.49, and 51.07%, respectively. These results
show that, although the content of Eu3+ units in copolymer P1,
P2 and P3 is very low (0.73, 1.38, and 3.69%, respectively), the
quantum efficiency is higher than that of monomer EuVTPY.
This is attributed to the fact that the carbazole is an efficient
hole transport unit. Aer inducing the carbazole units into the
Eu-polymers, the energy transfer can be more efficient than that
in EuVTPY. In particular, the h of P3 is higher by about 13% that
of EuVTPY, and this further conrms that the energy transfer
from PVK sequences to the Eu3+ ion emission centers was very
efficient in Eu-polymers P1, P2 and P3. In THF solution, the
quantum yield (F) of monomer EuVTPY is relatively high
(8.22%), however, the F of P1 is only 1.31, which is far lower
than that of EuVTPY. It can be interpreted that the interactions
between the THF molecules and Eu-polymer P1 increase the
distance between the PVK moieties and the Eu3+ ion emission
3hc3A0ꢀl
4e2w3c 5D0kUhlik FJ
Ul ¼
(5)
D
E
2
7
where e is the electronic charge and w is the angular frequency
2
of the transition. c ¼ n0(n0 + 2)2/9 is a Lorenz local eld
correction. The square reduced matrix elements are
5
2
5
2
h D0kU(2)kF2i ¼ 0.0032 and h D0kU(4)kF4i ¼ 0.0023, and an
average index of refraction equal to 1.5 was used. The U2 and U4
intensity parameters for all Eu-complexes are shown in Table 1.
A point to be noted in the results is the relatively high U2
parameter for all Eu-complexes. In particular, it is evident that
the Eu-polymers P1, P2 and P3 had higher U2 values than their
corresponding monomer EuVTPY, suggesting an increase of the
covalence degree in the rst coordination shell of Eu3+ ions and
an enhancement of the 5D0
/
7F2 hypersensitive transition.
¨
center, disrupting the inter- and intrachain Forster energy
This may also be due to the change of the chemical environ-
ment surrounding the Eu3+ ions, which was induced by the
intra- and inter-molecular interactions between the EuVTPY
unit and the neighboring chain consisting of PVK moieties. The
higher values of U4 for the Eu-polymers as compared with that
of their corresponding monomer EuVTPY indicated a pertur-
bation on the coordination effect of the ligands vinyl-tpy and
TTA by the steric factors from the surrounding PVK units.
transfer. In addition, the THF molecule vibration can reduce the
quantum yield in THF solution. With the content of EuVTPY
moieties in the Eu-polymer increasing, the
F gradually
increases; when the content of EuVTPY moieties is up to 3.69%,
the F of P3 is higher than that of the monomer EuVTPY.
To investigate the possible structural changes around the
Eu3+ ion emission centers among these Eu-complexes, the
Table 1 Solid state and solution luminescence data of Eu-complexes
In solid state
In THF solution
U2
(10ꢀ20 cm2)
U4
(10ꢀ20 cm2)
Compound
s (ms)
Arad (sꢀ1
)
I2/I1
h (%)
CIE
I2/I1
s (ms)
F (%)
CIE
EuVTPY
P1
P2
0.63
0.60
0.63
0.74
595.28
634.71
673.02
692.18
11.35
12.1
12.54
14.21
37.92
38.21
42.49
51.07
15.04
16.21
17.35
17.82
1.71
1.77
1.74
1.89
0.67, 0.33
0.61, 0.30
0.66, 0.32
0.66, 0.33
16.49
12.47
13.86
14.08
0.48
0.49
0.50
0.56
8.22
1.31
4.11
8.59
0.64, 0.31
0.31, 0.12
0.58, 0.28
0.60, 0.29
P3
4894 | J. Mater. Chem. C, 2013, 1, 4885–4901
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
3.5 Electrochemical properties
The redox behavior of the monomer EuVTPY, P1, P2 and P3 is very
important, as it can give information on the frontier orbitals of
these compounds. In particular, by comparing with the monomer
EuVTPY, the investigation is helpful to clearly understand the
carrier migration between the PVK segments and EuVTPY moie-
ties during the electroluminescent process. The electrochemical
properties of EuVTPY, P1, P2 and P3 were investigated at 298 K in
acetonitrile solutions versus the saturated calomel electrode (SCE)
(Table 2 and Fig. 9). For monomer EuVTPY, one irreversible
oxidation wave at 1.78 V was easily observed, which is attributed
to the oxidation of terpyridyl moieties in vinyl-tpy. The corre-
oxd
onset
sponding half-wave potential (Vo1/x2d) and onset potential (V
)
are 1.60 and 1.23 V, respectively. According to the equation
red
reported by de Leeuw et al.,29 ELUMO ¼ ꢀ(V
+ 4.4 eV) and
onset
oxd
onset
EHOMO ¼ ꢀ(V
+ 4.4 eV), in which ELUMO is the energy level of
Fig. 9 CV cures of monomer EuVTPY and Eu-polymers measured in acetonitrile
the LUMO and EHOMO is the energy level of the HOMO; EHOMO of
EuVTPY is about ꢀ5.63 eV. Meanwhile, for EuVTPY, two irre-
versible reduction waves at ꢀ1.24 and ꢀ1.57 V were observed. The
former corresponds to the reduction of vinyl moieties in vinyl-tpy,
and the latter should originate from its tpy moieties. Considering
that aer polymerization the vinyl moieties react to produce
saturated ethenyl moieties, ELUMO of polymerized EuVTPY should
be ꢀ3.01 eV corresponding to the second reduction peak.
solution, containing 0.1 M Bu4NClO4 at 298 K. Scans rate 50 mV sꢀ1
.
electronegative Eu-complex segments to form charge-transfer
(CT) excitons.
3.6 Electroluminescence performance
In order to investigate the electroluminescence (EL) properties
of the designed Eu-polymers, three types of spin-coated devices
were fabricated to investigate the EL performance of P1, P2 and
P3. Devices A, B and C were single-layer PLEDs, and they were
based on P1, P2 and P3 with the conguration of ITO/PEDOT
(40 nm)/Eu-polymer (80 nm)/Ba (4 nm)/Al (120 nm). Devices D, E
and F were double-layer PLEDs, and they were based on P1, P2
and P3 with the conguration of ITO/PEDOT (40 nm)/PVK
(40 nm)/copolymer (80 nm)/Ba (3 nm)/Al (120 nm). Devices G
and H were based on P1 with the conguration of ITO/PEDOT
(40 nm)/PVK (40 nm)/P1:OXD-7 (40% and 30%, 80 nm)/Ba
(3 nm)/Al (120 nm).
EL spectra of the eight devices consist of the characteristic
Eu3+ ion emission corresponding to 5D0 / 7Fj (j ¼ 0–4) (Fig. 10).
Moreover, the peaks at 580, 594, 653, and 697 nm are much
weaker than those in the PL spectra in the solid state. Therefore,
the main peak at 615 nm becomes more dominant as pure red
emission, which exhibits extremely high color purity (CIE coor-
dinates were around 0.66, 0.33) with a full width at half-
maximum of 10 nm. In particular, even at the highest voltage, EL
spectra of devices A–H were still very stable. No distinct short-
wavelength emission from PVK moieties was observed, which
indicated that the efficient energy transfer from PVK segments to
Eu(TTA)3 moieties at the high current densities occurred.
For polymers P1, P2, and P3, two irreversible oxidations
could be observed. The rst irreversible oxidation wave was
located 1.14, 1.12, and 1.16 V, and the second oxidation wave
was located at 1.61, 1.65, and 1.59. The rst irreversible oxida-
tion wave is attributed to the oxidation of the tpy moieties, and
the second is attributed to the carbazole moieties. In addition,
due to the very low content of EuVTPY in these polymers, no
distinct reduction peak is observed for P1, P2, and P3. Accord-
ing to their onset oxidation potentials, EHOMO of P1, P2 and P3 is
about ꢀ5.31 (5.67) eV. The energy gap between the LUMO and
HOMO energy levels of P1, P2 and P3 is calculated to be 3.49 eV
by reference to their absorption edge (carbazole moieties) of
355 nm. Thus, ELUMO of P1, P2 and P3 is about ꢀ2.19 eV. P1, P2
and P3 exhibit much stronger hole-injection ability but much
weaker electron-injection ability than EuVTPY. Nevertheless, as
the majority in the copolymers, PVK segments are dominant in
carrier injection. The wrapping and embedment of Eu-complex
segments in the nonconjugated systems would further enhance
this situation. Therefore, this implies that under the electric
eld, rstly carriers are injected in PVK segments, then elec-
trons can be captured by Eu-complex segments through intra-
and inter-chain carrier migration and nally holes hopping to
the surrounding PVK segments can be attracted by
Table 2 Electrochemical properties of monomer EuVTPY and Eu-polymers
red
1/2
red
onset
oxd
1/2
oxd
Compound
V
(V)
V
(V)
V
(V)
V (V)
onset
EHOMO (eV)
ELUMO (eV)
EuVTPY
P1
P2
ꢀ1.06, ꢀ1.53
ꢀ0.91, ꢀ1.39
1.60
1.23
ꢀ5.63
ꢀ3.49, ꢀ3.01
ꢀ1.83, ꢀ2.19
ꢀ1.80, ꢀ2.17
ꢀ1.82, ꢀ2.22
—
—
—
—
—
—
1.07, 1.42
1.06, 1.49
1.09, 1.51
0.92, 1.28
0.89, 1.26
0.91, 1.31
ꢀ5.32, ꢀ5.68
ꢀ5.29, ꢀ5.66
ꢀ5.31, ꢀ5.71
P3
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4885–4901 | 4895

View Article Online
Journal of Materials Chemistry C
Paper
For single-layer PLEDs, the energy band diagram is shown excellent hole transporting material, and upon increasing the
schematically in Fig. 10(b). In the presence of ideal intra- content of PVK segments in the Eu-polymers, the hole-injection
molecular energy transfer, electrons are preferentially injected ability also increased, and more balanced carrier injection and
from the Ba/Al cathode into the LUMO of the Eu-polymers P1, transport can be obtained, so the properties of device A based
P2 and P3. Holes are injected from the ITO anode into the on P1 as the emitting layer are superior to those of B and C. As
HOMO of the PEDOT, and then transported to the HOMO of P1, mentioned above, one important problem can be raised. If we
P2 and P3. With the subsequent recombination of holes and continued to increase the content of PVK in single-layer PLEDs,
electrons in the Eu-polymers P1, P2 and P3, excitons are formed, could the electroluminescence performances of corresponding
and EL emission occurs. As show in Fig. 11, the turn-on voltage PLEDs be further improved? To better conrm this idea, on the
of A is 24 V at 1 cd mꢀ2, and its maximum brightness is 5.96 cd basis of single-layer devices A, B and C, we added one additional
mꢀ2 at 25.5 V with the current density of 178 mA cmꢀ2. Due to PVK layer (40 nm) and fabricated corresponding double-layer
the maximum brightness of devices B and C being lower than devices D, E and F, and the corresponding B–J–V and CE–J
1 cd mꢀ2, we cannot dene their turn-on voltage. The maximum curves are shown in Fig. 12. Aer adding the PVK layer, the turn-
brightnesses of B and C are 0.82 and 0.1 cd mꢀ2 at 25.5 and on voltage of D is increased to 27 V, but its maximum brightness
15.8 V with the current densities of 25.8 and 5.5 mA cmꢀ2
,
improved to 18.9 cd mꢀ2 at 35.7 V with a current density of 81
respectively. It is shown that the turn-on voltage of A is relatively mA cmꢀ2. Compared with the corresponding single-layer device
high, and the maximum brightness of A is far higher than that A, although the turn-on voltage increased 3 V, the maximum
of B and C. Moreover, the current density of A is much higher brightness increased 3 times. Most importantly, due to the
than those of B and C at the same voltages, and these results addition of the PVK layer, devices E and F had turn-on voltages
indicate that P1 exhibited much stronger carrier injection and of 25.7 and 38 V, respectively. The maximum brightness was
transport ability than P2 and P3.30 The main reason might be 4.05 and 2.37 cd mꢀ2 at 33.0 and 38.5 V with current density of
that the majority of the Eu-polymers P1, P2 and P3 are the PVK 18.9 and 25.8 mA cmꢀ2, respectively. Compared with the cor-
segments, as is well known, which is an efficient hole trans- responding single-layer devices B and E, the maximum bright-
porting material, so the major carrier in devices A, B and C is ness increased nearly 5 and 24 times, respectively. As shown in
hole carrier. The hole-injection ability of vinyl-tpy is much Fig. 10(c), compared with single-layer PLEDs, although the
weaker than the PVK segments, thus, the hole-injection and channel of electron injection is not changed, the passage of
transporting ability of the main chain is weakened upon holes from the ITO to emission layer Eu-polymer P1, P2 or P3
increasing the content of vinyl-tpy. Although the electron will be easier due to the existence of the hole transport layer
injection ability of EuVTPY is much stronger than that of the PVK. Therefore the carrier-injection in D, E and F can be more
PVK segments, because of the wrapping and embedment of the balanced than in A, B and C, and more excitons can be formed
pendent EuVTPY moieties by the PVK segments, the improve- with recombination in the emission layer. This is the reason
ment of electron injection and transporting upon increasing the why the electroluminescence performances of devices D, E and
content of EuVTPY from 0.73% to 3.69% is limited. Therefore, F are much better than those of A, B and C.
the carrier-injection and transporting in A is superior to that in
B and C.
In particular, although the brightness–current density (B–J)
curves of A, B, D and E show similar tendencies at the same
For single-layer PLEDs, we nd the content of PVK segments current densities, the brightnesses of B and E are higher than
is higher, and the maximum brightness and current density are those of A and D (Fig. S6 and S7†). This is very different from the
much higher. This may be attributed to the fact that PVK is an doping devices, in which the brightness remarkably decreased
Fig. 10 EL spectra of the devices at their highest voltages and the schematic energy level diagram of the devices.
4896 | J. Mater. Chem. C, 2013, 1, 4885–4901
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
Fig. 11 Left: brightness–current density–voltage (B–J–V) curves of devices A–C. Right: CE–current density (CE–J) curves of devices A–C.
along with an increase of the doping concentration. According those of corresponding single-layer A, B and C, the maximum
to the EL spectra of A, B, D and E, the absence of the short- brightness and efficiency of the PLEDS are relatively low. In
wavelength emission from PVK segments suggests the efficient addition, the above mentioned results show that the difficulty of
¨
exciton migration and Forster energy transfer in both Eu-poly- electron injection and transporting is the main reason for the
mers P1 and P2. If the EL process of A, B, D and E mainly low brightness of A to F. In order to improve the electron
involves these two channels, due to the Eu-chelated units in injection and preserve the advantage of the good hole injection,
device B (E) being higher than that of A (D), B (E) should have an excellent electron transport material OXD-7 was blended in
much higher brightness than that of A (D) at the same current the emitting layer. Due to the EL performances of devices A and
density. However, the brightness of A (D) is higher than that of B D based on P1 as the emitting layer being the best among the
(E) because of the worse concentration quenching of P2. devices, we next chose P1 as the emitting layer, OXD-7 (electron
According to electrochemical analysis, the LUMO energy levels transport material) as blend material, and PVK as hole transport
of the PVK segments and Eu-chelated segments are ꢀ2.0 and material to fabricate two novel PLEDs. The conguration of the
ꢀ3.5 eV, respectively. Thus, the Eu-chelated repeating units devices was ITO/PEDOT (40 nm)/PVK (40 nm)/P1: OXD-7 (40%,
introduce electron traps with a depth of as much as 1.5 eV, and 30%, 80 nm)/Ba (3 nm)/Al (120 nm). The LUMO level of
which makes the electron capture efficient. Thus, aer electrons OXD-7 is ꢀ2.8 eV, which is close to the vacuum level of Ba/Al of
are trapped in Eu-chelated segments, the surrounding holes ꢀ2.7 eV (Fig. 10(d)). It is shown that the turn-on voltage of G is
would be attracted by the coulomb force to form charge trans- remarkably reduced from 27 V to 17 V, and its maximum
form (CT) excitons. Obviously, P2 with the higher Eu-chelated brightness of 68.2 cd mꢀ2 is achieved at 23.5 V with the current
content could form more excitons than P1, and nally realize density of 88.7 mA cmꢀ2. H had a turn-on voltage of 20 V. The
the higher EL brightness at the same current density. This maximum brightness of H was 35.4 cd mꢀ2 at 28.0 V with the
implies that for devices A, B, D and E, the carrier trapping might current density of 69.8 mA cmꢀ2 (Fig. 13). Compared with
be the main channel involved in the EL process.31
the corresponding undoped device D, the turn-on voltages of G
Aer adding an additional PVK layer, although the EL and H were much lower because of their improved electron
performances of double-layer D, E and F are much better than injection and transporting. The maximum brightnesses of G
Fig. 12 Left: brightness–current density–voltage (B–J–V) curves of devices D–F. Right: CE–current density (CE–J) curves of devices D–F.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4885–4901 | 4897

View Article Online
Journal of Materials Chemistry C
Paper
and H were improved 3.6 and 1.9 times, respectively. It is Aꢀ1 at 6.8 mA cmꢀ2, corresponding to 0.018 lm Wꢀ1, and 0.24%.
noteworthy that the turn-on voltage decreased from 20 to 17 V Obviously, the more balanced carrier injection and transporting
with increasing doping content of OXD-7 from 30 wt% for in G and H was effective in improving the device efficiencies.
device H to 40 wt% for device G. This result indicates that the Compared with single-layer device A, for device G, the CE, PE
doping devices G and H mainly operate by the carrier-trapping and EQE increased 53, 18, and 55 times, respectively, and for
mechanism. The reason for the better EL performance of device device H, the CE, PE and EQE increased 35, 10.5, and 36 times,
G or H can be attributed to the fact that the electron can be respectively. It is noteworthy that the EL efficiencies of the
easily injected in the emitting layers aer blending the electron devices do not drop remarkably when the current density is
transporting material OXD-7. The increasing concentration of increased (Fig. 13), and when the current density reaches its
electrons greatly facilitates the formation of excitons. In addi- maximum, the devices still maintain a relatively high value.
tion, with the content of OXD-7 increasing, although parts of This is very different to the devices based on pure small mole-
the excitons were still formed by carrier-trapping, most of the cule Eu-complexes or Eu-containing doping systems, whose
excitons were formed at PVK segments, then, the energy trans- efficiencies greatly decrease along with an increase of the
fers to Eu-chelated segments through exciton migration and current density. Therefore, this indicates that in P1, P2 and P3
Forster energy transfer. For these reasons, the EL properties of the uniform dispersion of the emissive Eu3+ chelate moieties
¨
device G with a high content of OXD-7 (40 wt%) are better than in the host matrix through covalent bonds can efficiently
those of device H with a low content of OXD-7 (30 wt%).
restrain the concentration quenching and T–T annihilation
The EL efficiencies of A–H, except B and C, were also calcu- even at the highest exciton concentrations.32 It is also noted that
lated. Their current efficiency (CE)–J curves are shown in although P2 and P3 have higher PL quantum efficiency in solid
Fig. 11–13. The maximum CE of A was 0.0036 cd Aꢀ1 at 167.99 than P1, the devices based on P1 exhibit much higher efficien-
mA cmꢀ2, corresponding to the power efficiency (PE) of 0.0017 cies than those based on P2 and P3. This should be attributed to
lm Wꢀ1 and EQE (external quantum efficiency) of 0.0067%. the different emission mechanisms of PL and EL. The multi-
Aer adding one PVK layer, the maximum efficiency of D was particle quenching effect is much more dominant and effective
greatly improved to 0.023 cd Aꢀ1, 0.021 lm Wꢀ1, and 0.009% at in the EL process than in the PL process. Therefore, the higher
57 mA cmꢀ2. Meanwhile, devices E and F had maximum CE of Eu3+ content may facilitate the interaction between the
0.031 and 0.0091 cd Aꢀ1 at 8.67 and 25.8 mA cmꢀ2, corre- segments and consequently reduce the EL efficiencies by those
sponding to 0.003 and 7.5 ꢂ 10ꢀ4 lm Wꢀ1, and 0.012 and quenching effects.
0.004%, respectively (Fig. 12). Compared with corresponding
The EL performances from several recent representative
single-layer device A (P1 as emitting layer), the CE, PE and EQE reports about Eu-polymers are listed in Table 3. The EL
of the double-layer device D increased 6, 12 and 1.3 times, and performance of device G with PVK as a hole transport material,
the results indicated that introducing the additional PVK layer and OXD-7 (40 wt%) as the electron transport material
into single-layer Eu-polymer PLEDs is very efficient for compares favorably with these results. The maximum bright-
improving the EL performances. To the best of our knowledge, ness, PE and CE of G is the third-best reported, and the EQE is
this is the rst report in which the EL properties of these PLEDs the second-best reported so far. These excellent performances
could be modied by adding an additional PVK layer into Eu- originate from the good comprehensive properties of P1 in
polymers. Aer blending the electron transport material OXD-7 intra- and interchain energy transfer, high PL efficiencies,
with P1, the EL efficiency of G (OXD-7 content: 40%) was greatly stable structure and efficient mitigation of concentration
improved to 0.18 cd Aꢀ1, 0.031 lm Wꢀ1, and 0.37% at 9.5 mA quenching and T–T annihilation. Although the EL perfor-
cmꢀ2 (Fig. 13). Meanwhile, H had the maximum CE of 0.12 cd mances of
G are not the highest among the reported
Fig. 13 Left: brightness–current density–voltage (B–J–V) curves of devices G–H. Right: CE–current density (CE–J) curves of devices G–H.
4898 | J. Mater. Chem. C, 2013, 1, 4885–4901
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
Eu-containing copolymers, to the best of our knowledge, this is the frontier orbital levels of the tpy ligands can be conveniently
the rst example of electroluminescent diodes based on Eu- tuned, so that lower turn-on voltage, higher brightness and
complexes as emitting layers with tpy as a neutral ligand. These higher efficiency can be obtained.
results illustrate the potential application of polymerizable tpy
ligands in high performance EL Eu-chelated polymers.
However, the EL performance based on P1, P2 and P3 as the
Acknowledgements
The authors appreciate the Guangdong-Hongkong Technology
Cooperation Funding (Project no. 2009A091300012) and
National Natural Science Foundation of China (Project no.
20974121).
emission layer is still inferior to those of small molecular Eu-
complexes as the emission layer, which may be attributed to the
simple conguration of devices fabricated through spin
coating, the possibility of reversible energy transfer between Eu-
complexed moieties and PVK segments, and the different
emission mechanism of PL and EL. Through the device opti-
mization, the EL performance of P1, P2 and P3 would be further
improved.
Notes and references
1 (a) M. J. Park, J. H. Lee, J. H. Kwak, I. H. Jung, J. H. Park,
H. Y. Kong, C. H. Lee, D. H. Hwang and H. K. Shim,
Macromolecules, 2009, 42, 5551–5557; (b) G. R. Whittell,
M. D. Hager, U. S. Schubert and I. Manners, Nat. Mater.,
2011, 10, 176–188; (c) J. X. Jiang, C. Y. Jiang, W. Yang,
H. Y. Zhen, F. Huang and Y. Cao, Macromolecules, 2005, 38,
4072–4080; (d) 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, 539–541;
(e) S. Basak, Y. S. L. V. Narayana, M. Baumgarten,
4 Conclusions
In this work, we have designed and synthesized three nearly
monochromatic red electroluminescent chelating polymers
containing carbazole segments and tpy moieties which serve as
neutral ligands to coordinate with the Eu(TTA)3 complex. The
resulting copolymers exhibited good solubility, as well as good
thermal stability and high glass transition temperatures. Our
investigations showed that the functional tpy ligand and the
corresponding Eu-chelate segments signicantly improve the
properties of the Eu-complex nonconjugated copolymers. It is
shown that the strong coordination ability of tridentate tpy
ligands facilitates the stability of the complex monomer during
polymerization. P1, P2 and P3 exhibit enhanced emission from
Eu3+ ions and high PL quantum efficiency in solid of 40%,
which implies the efficient intra- and inter-chain energy trans-
fer in P1, P2 and P3. The pure-red emission from the devices of
P1, P2 and P3 was demonstrated. Favorable EL performance
including a relatively low turn-on voltage of 17 V and maximum
brightness of 68.2 cd mꢀ2 were realized. To the best of knowl-
edge, this is the rst report of the EL performance of Eu-
complexes based on tpy as a neutral ligand. Our results
demonstrate the potential application of polymerizable tpy
ligands in high performance EL europium-complex polymers.
The further purposeful chemical modication of tpy ligands is
ongoing in our group. It has been shown that the suitable
excited energy levels of tpy ligands have a strong effect on the
intra-chain energy transfer and charge trapping is one of the
most important channels in the EL process. Through intro-
ducing different functional groups, the excited energy levels and
¨
K. Mullen and R. Chandrasekar, Macromolecules, 2013, 46,
362–369.
2 (a) A. Bandyopadhyay, S. Sahu and M. Higuchi, J. Am. Chem.
Soc., 2011, 133, 1168–1171; (b) Q. D. Ling, Y. Song, S. J. Ding,
C. X. Zhu, D. S. H. Chan, D. L. Kwong, E. T. Kang and
K. G. Neoh, Adv. Mater., 2005, 17, 455–459; (c) B. Korybut-
Daszkiewicz, R. Bilewicz and K. Wozniak, Coord. Chem.
Rev., 2010, 254, 1637–1660; (d) M. J. Cho, D. H. Choi,
A. Sullivan, A. J. P. Akelaitis and L. R. Dalton, Prog. Polym.
Sci., 2008, 33, 1013–1058; (e) Q. D. Ling, D. J. Liaw,
C. X. Zhu, D. S. H. Chan, E. T. Kang and K. G. Neoh, Prog.
Polym. Sci., 2008, 33, 917–978; (f) X. Y. Chen, X. P. Yang
and B. J. Holliday, J. Am. Chem. Soc., 2008, 130, 1546–1547;
(g) Z. M. Liu, A. A. Yasseri, J. S. Lindsey and D. F. Bocian,
Science, 2003, 302, 1543–1545.
3 (a) I. Radivojevic, A. Varotto, C. Farley and C. M. Drain,
Energy Environ. Sci., 2010, 3, 1897–1909; (b) Y. Li,
Y. Z. Bian, M. Yan, P. S. Thapaliya, D. Johns, X. Z. Yan,
D. Galipeau and J. Z. Jiang, J. Mater. Chem., 2011, 21,
11131–11141; (c) M. Jurow, A. E. Schuckman, J. D. Batteas
and C. M. Drain, Coord. Chem. Rev., 2010, 254, 2297–2310;
(d) C. M. Yan, Y. P. Fan and D. H. Zhao, Macromolecules,
Table 3 EL performances of the Eu-containing polymers
Max. EL efficiency
Turn-on voltage
(V)
Max. brightness
Reference
(cd mꢀ2
)
CE (cd Aꢀ1
)
PE (lm Wꢀ1
)
EQE (%)
30(a)
30(b)
31(a)
31(b)
17
30
13
12.5
8
16.4
46
36.9
149.1
126
—
—
0.247 (6.29 cd mꢀ2
)
—
—
0.035 (9.8 cd mꢀ2
)
0.035 (21.5 cd mꢀ2
)
0.0073 (21.5 cd mꢀ2
)
0.037 (21.5 cd mꢀ2
)
0.41 (71.6 cd mꢀ2
0.52 (1 cd mꢀ2
0.18 (9.5 cd mꢀ2
)
0.089 (71.6 cd mꢀ2
)
0.43 (71.6 cd mꢀ2
)
32
)
Device G of this work
17
68.2
)
0.031 (15.7 cd mꢀ2
)
0.37 (9.5 cd mꢀ2
)
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4885–4901 | 4899

View Article Online
Journal of Materials Chemistry C
Paper
2012, 45, 133–141; (e) P. T. Wu, T. Bull, F. S. Kim,
C. K. Luscombe and S. A. Jenekhe, Macromolecules, 2009,
42, 671–681.
Chem., 2007, 45, 388–394; (d) Z. G. Zhang, J. B. Yuan,
H. J. Tang, H. Tang, L. N. Wang and K. L. Zhang, J. Polym.
Sci., Part A: Polym. Chem., 2009, 47, 210–221.
4 X. H. Yang, D. C. Muller, D. Neher and K. Meerholz, Adv. 14 (a) W. Andreas, W. Andreas, F. Schlutterand and
Mater., 2006, 18, 948–954.
U. S. Schubert, Chem. Soc. Rev., 2011, 40, 1459–1511; (b)
P. P. Kumar, G. Premaladha and B. G. Maiya, Chem.
Commun., 2005, 3823–3825; (c) M. Licini and
J. A. G. Williams, Chem. Commun., 1999, 1943–1944; (d)
P. P. Laine, F. Loiseau, S. Campagna, I. Cioni and
C. Adamo, Inorg. Chem., 2006, 45, 5538–5551; (e)
P. P. Laine, S. Campagna and F. Loiseau, Coord. Chem.
Rev., 2008, 252, 2552–2571; (f) H. Hofmeier and
U. S. Schubert, Chem. Soc. Rev., 2004, 33, 373–399.
5 (a) M. J. Park, J. H. Lee, J. H. Kwak, I. H. Jung, J. H. Park and
H. Y. Kong, Macromolecules, 2009, 42, 5551–5557; (b)
A. J. Sandee, C. K. Williams, N. R. Evans, J. E. Davies,
¨
C. E. Boothby and A. Kohler, J. Am. Chem. Soc., 2004, 126,
7041–7048; (c) P. I. Lee and S. L. C. Hsu, J. Polym. Sci., Part
A: Polym. Chem., 2007, 45, 1492–1498; (d) X. Chen,
J. L. Liao, Y. Liang, M. O. Ahmed, H. E. Tseng and
S. A. Chen, J. Am. Chem. Soc., 2003, 125, 636–637.
6 (a) A. C. Grimsdale, K. L. Chan, R. E. Martin, P. G. Jokisz and 15 (a) O. Kotova, R. Daly, C. M. G. dos Santos, M. Boese,
A. B. Holmes, Chem. Rev., 2009, 10897–10915; (b)
C. Coluccini, A. K. Sharma, D. Merli, D. V. Griend,
B. Mannucci and D. Pasini, Dalton Trans., 2011, 40, 11719–
11725; (c) C. Coluccini, P. Metrangolo, M. Parachini,
D. Pasini, G. Resnati and P. Righetti, J. Polym. Sci., Part A:
Polym. Chem., 2008, 46, 5202–5213; (d) S. F. Li,
G. Y. Zhong, W. H. Zhu, F. Y. Li, J. F. Pan, W. Huang and
H. Tian, J. Mater. Chem., 2005, 15, 3221–3228; (e)
G. Zucchi, T. Jeon, D. Tondelier, D. Aldakov, P. Thuery,
M. Ephritikhine and B. Geffroy, J. Mater. Chem., 2010, 20,
2114–2120; (f) P. Lenaerts, A. Storms, J. Mullens, J. D'Haen,
C. Gorller-Walrand, K. Binnemans and K. Driesen, Chem.
P. E. Kruger, J. J. Boland and T. Gunnlaugsson, Angew.
Chem., Int. Ed., 2012, 51, 7208–7212; (b) W. S. Lo,
W. M. Kwok, G. L. Law, C. T. Yeung, C. T. L. Chan,
H. L. Yeung, H. K. Kong, C. H. Chen, M. B. Murphy,
K. L. Wong and W. T. Wong, Inorg. Chem., 2011, 50, 5309–
5311; (c) R. Shunmugam and G. N. Tew, J. Am. Chem. Soc.,
2005, 127, 13567–13572; (d) B. Song, G. L. Wang,
M. Q. Tan and J. L. Yuan, J. Am. Chem. Soc., 2006, 128,
13442–13450; (e) J. Costa, R. Ruloff, L. Burai, L. Helm and
A. E. Merbach, J. Am. Chem. Soc., 2005, 127, 5147–5157; (f)
N. Chandrasekhar and R. Chandrasekar, J. Org. Chem.,
2010, 75, 4852–4855.
Mater., 2005, 17, 5194–5201; (g) D. Pasini, P. P. Righetti 16 L. Matthews and E. T. Knobbe, Chem. Mater., 1993, 5, 1697–
and V. Rossi, Org. Lett., 2002, 4, 23–26; (h) C. Coluccini, 1700.
A. K. Sharma, M. Caricato, A. Sironi, E. Cariati, S. Righetto, 17 J. J. P. Stewart, MOPAC 2012 Manual, Stewart Computational
E. Tordin, C. Botta, A. Forni and D. Pasini, Phys. Chem.
Chem. Phys., 2013, 15, 1666–1674.
7 (a) J. Kido and Y. Okamoto, Chem. Rev., 2002, 102, 2357–
2368; (b) S. V. Eliseeva and J. C. G. Bunzli, Chem. Soc. Rev.,
2010, 39, 189–227.
Chemistry, Colorado Springs, 2012.
18 (a) A. P. Souza, F. A. A. Paz, R. O. Freire, L. D. Carlos,
O. L. Malta, S. A. Junior and G. F. de Sa, J. Phys. Chem. B,
2007, 111, 9228–9238; (b) O. Freire, G. B. Rocha and
A. M. Simas, Inorg. Chem., 2005, 44, 3299–3310.
8 C. Adachi, M. A. Baldo and S. R. Forrest, J. Appl. Phys., 2000, 19 (a) D. P. Li, C. H. Li, J. Wang, L. C. Kang, T. Wu, Y. Z. Li and
87, 8049–8055.
X. Z. You, Eur. J. Inorg. Chem., 2009, 4844–4849; (b) X. L. Li,
F. R. Dai, L. Y. Zhang, Y. M. Zhu, Q. Peng and Z. N. Chen,
Organometallics, 2007, 26, 4483–4490.
9 J. Kido, H. Hayase, K. Hongawa, K. Nagai and K. Okuyama,
Appl. Phys. Lett., 1994, 65, 2124–2126.
10 T. X. Li, H. Fukuyama, Y. Yamagata, H. L. Lan and J. Kido, 20 B. W. Wang, S. D. Jiang, X. T. Wang and S. Gao, Sci. China,
Polym. Adv. Technol., 2004, 15, 302–305. Ser. B: Chem., 2009, 52, 1739–1758.
11 H. Xu, K. Yin and W. Huang, Chem.–Eur. J., 2007, 13, 10281– 21 J. H. Wang and G. S. Hanan, Synlett, 2005, 8, 1251–1254.
10293.
22 A. Karali, P. Dais, E. Mikros and F. Heatley, Macromolecules,
2001, 34, 5547–5554.
12 (a) J. Pei, L. X. Liu, L. W. Yu, Y. H. Lai, Y. H. Niu and Y. Cao,
Macromolecules, 2002, 35, 7274–7280; (b) Q. D. Ling, 23 A. Karali, G. E. Froudakis, P. Dais and F. Heatley,
E. T. Kang, K. G. Neoh and W. Huang, Macromolecules, Macromolecules, 2000, 33, 3180–3183.
2003, 36, 6995–7003; (c) P. Lenaerts, K. Driesen, D. R. Van 24 (a) R. Feng, F. L. Jiang, M. Y. Wu, L. Chen, C. F. Yan and
and K. Binnemans, Chem. Mater., 2005, 17, 2148–2154; (d)
Q. D. Ling, W. Wang, Y. Song, C. X. Zhu, D. S. H. Chan
and E. T. Kang, J. Phys. Chem.B, 2006, 110, 23995–24001;
(e) A. Balamurugan, M. L. P. Reddy and M. Jayakannan, J.
Phys. Chem. B, 2009, 113, 14128–14138.
M. C. Hong, Cryst. Growth Des., 2010, 10, 2306–2313; (b)
Y. J. Li and B. Yan, Dalton Trans., 2010, 39, 2554–2562; (c)
C. L. Yang, J. X. Luo, J. Y. Ma, M. G. Lu, L. Y. Liang and
B. H. Tong, Dyes Pigm., 2011, 92, 696–704.
25 (a) A. Beeby, I. M. Clarkson, R. S. Dickins, S. Faulkner,
D. Parker, L. Royle, A. S. Sousa, J. A. G. Williams and
M. Woods, J. Chem. Soc., Perkin Trans. 2, 1999, 493–504; (b)
C. L. Yang, J. Xu, J. Y. Ma, D. Y. Zhu, Y. F. Zhang,
L. Y. Liang and M. G. Lu, Photochem. Photobiol. Sci., 2013,
12, 330–338.
13 (a) M. J. Yang, Q. D. Ling, M. Hiller, X. Z. Fun, X. Liu and
L. H. Wang, J. Polym. Sci., Part A: Polym. Chem., 2000, 38,
3405–3411; (b) Y. X. Cheng, X. W. Zou, D. Zhu, T. S. Zhu,
Y. Liu and S. W. Zhang, J. Polym. Sci., Part A: Polym. Chem.,
2007, 45, 650–660; (c) G. A. Wen, X. R. Zhu, L. H. Wang,
J. C. Feng, R. Zhu and W. Wei, J. Polym. Sci., Part A: Polym. 26 W. H. Melhuish, J. Phys. Chem., 1961, 65, 229–235.
4900 | J. Mater. Chem. C, 2013, 1, 4885–4901
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry C
27 (a) R. Ferreira, P. Pires, B. D. Castro, R. A. SaFerreira, L. D. Carlos
and U. Pischel, New J. Chem., 2004, 28, 1506–1513; (b) C. L. Yang,
J. Xu, R. Zhang, Y. F. Zhang, Z. X. Li, Y. W. Li, L. Y. Liang and
M. G. Lu, Sens. Actuators, B, 2013, 177, 437–444.
28 (a) M. H. V. Werts, R. T. F. Jukes and J. W. Verhoeven, Phys.
Chem. Chem. Phys., 2002, 4, 1542–1548; (b) H. Liang and
F. Xie, Spectrochim. Acta, Part A, 2010, 75, 1191–1914.
29 D. M. de Leeuw, M. M. J. Simenon, A. R. Brown and
R. E. F. Einerhand, Synth. Met., 1997, 87, 53–59.
Chem., 2000, 38, 3405–3411; (b) Z. G. Zhang, J. B. Yuan,
H. J. Tang, H. Tang, L. N. Wang and K. L. Zhang, J. Polym.
Sci., Part A: Polym. Chem., 2009, 47, 210–221.
31 (a) H. Xu, R. Zhu, P. Zhao, L. H. Xie and W. Huang, Polymer,
2011, 52, 804–813; (b) H. Xu, R. Zhu, P. Zhao and W. Huang,
J. Phys. Chem. C, 2011, 115, 15627–15638.
32 Q. D. Ling, Q. J. Cai, E. T. Kang, K. G. Neoh, F. R. Zhu and
W. Huang, J. Mater. Chem., 2004, 14, 2741–2748.
33 A. Pacini, M. Caricato, S. Ferrari, D. Capsoni, A. M. de
Ilarduya, S. Munoz-Guerra and D. Pasini, J. Polym. Sci., Part
A: Polym. Chem., 2012, 50, 4790–4799.
30 (a) M. J. Yang, Q. D. Ling, M. Hiller, X. Z. Fun, X. Liu,
L. H. Wang and W. G. Zhang, J. Polym. Sci., Part A: Polym.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. C, 2013, 1, 4885–4901 | 4901