Aggregation-induced emission tetraphenylethene type derivatives for blue tandem organic light-emitting diodes

Article abstract of DOI:10.1016/j.orgel.2018.12.025

Tetraphenylethene (TPE) has been demonstrated to be an effective core molecule for generating excellent aggregation-induced emissions (AIE). Herein, we propose TPE moieties decorated with different biphenyl or anthracenyl fragments to avoid the crystalliz

Full text of DOI:10.1016/j.orgel.2018.12.025

Accepted Manuscript
Aggregation-induced emission tetraphenylethene type derivatives for blue tandem
organic light-emitting diodes
Chia-Wei Liao, Yu-Chien Hsu, Che-Chia Chu, Chih-Hao Chang, Gintare Krucaite,
Dmytro Volyniuk, Juozas V. Grazulevicius, Saulius Grigalevicius
PII:
S1566-1199(18)30667-0
DOI:
https://doi.org/10.1016/j.orgel.2018.12.025
ORGELE 5034
Reference:
To appear in: Organic Electronics
Received Date: 23 October 2018
Revised Date: 16 December 2018
Accepted Date: 16 December 2018
Please cite this article as: C.-W. Liao, Y.-C. Hsu, C.-C. Chu, C.-H. Chang, G. Krucaite, D. Volyniuk,
J.V. Grazulevicius, S. Grigalevicius, Aggregation-induced emission tetraphenylethene type derivatives
for blue tandem organic light-emitting diodes, Organic Electronics (2019), doi: https://doi.org/10.1016/
j.orgel.2018.12.025.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Aggregation-induced emission tetraphenylethene type derivatives
for blue tandem organic light-emitting diodes
Chia-Wei Liao a, Yu-Chien Hsu a, Che-Chia Chu a, Chih-Hao Chang a,*, Gintare Krucaite b,
Dmytro Volyniuk b, Juozas V. Grazulevicius b, Saulius Grigalevicius b,**
a Department of Electrical Engineering, Yuan Ze University, Chung-Li 32003, Taiwan
b
Department of Polymer Chemistry and Technology, Kaunas University of Technology,
Radvilenu plentas 19, Kaunas, LT50254, Lithuania
* Corresponding author. Department of Electrical Engineering, Yuan Ze University, Chung-Li
32003, Taiwan.
** Corresponding author. Department of Polymer Chemistry and Technology, Kaunas
University of Technology, Radvilenu plentas 19, Kaunas, LT50254, Lithuania.
E-mail addresses: chc@saturn.yzu.edu.tw (C.-H. Chang), saulius.grigalevicius@ktu.lt (S.
Grigalevicius).
-------------------------------------------------------------------------------------------------------------------
Abstract
Tetraphenylethene (TPE) has been demonstrated to be an effective core molecule for
generating excellent aggregation-induced emissions (AIE). Herein, we propose TPE moieties
decorated with different biphenyl or anthracenyl fragments to avoid the crystallization of neat
film and simultaneously harvest blue AIE. The two synthesized compounds,
2-(4-biphenyl)-1,1,2-triphenylethylene (2) and 2-(9-antracenyl)-1,1,2-triphenylethylene (3),
possess bipolar transport capabilities, high thermal stability, and strong sky-blue AIE. Both
compounds 2 and 3 were selected as emitters to investigate their electroluminescence

ACCEPTED MANUSCRIPT
applications. The optimized non-doped sky-blue organic light-emitting diodes (OLEDs) with 2
and 3 exhibited respective maximum efficiencies of 3.8% (8.8 cd/A) and 2.2% (4.0 cd/A).
These results illustrate the effective molecular design concepts of the TPE derivatives
containing biphenyl or anthracenyl fragments. Furthermore, a tandem device architecture was
used to improve device efficiency. Through adopting the specific designs of the intermediate
charge generation layer structure and the corresponding device architecture, respective peak
efficiencies of 8.2% (19.4 cd/A) and 4.6% (9.5 cd/A) were obtained in tandem devices with 2
and 3 emitters together with slight efficiency roll-off. The superior EL performance
demonstrated that the tandem structure possesses high potential for boosting the efficiency of
the AIE-based OLEDs.
Keywords: Tetraphenylethene (TPE), aggregation-induced emission (AIE), biphenyl,
anthracenyl, organic light-emitting diodes (OLEDs), tandem
-------------------------------------------------------------------------------------------------------------------

ACCEPTED MANUSCRIPT
1. Introduction
Organic light-emitting diodes (OLEDs) have been seen as one of the most important
solid-state sources, and are widely used as the primary display in cellphones and other portable
devices. Compared to other competing technologies, OLEDs stand out because of their superior
display quality and flexible properties allowing them to meet a variety of application needs.
However, for any given commercial application, OLEDs still have to pursue higher device
efficiency and operational stability. This has driven the aggressive development of luminogenic
compounds, including exciplex [1], thermally activated delayed fluorescence (TADF) [2], and
aggregation-induced luminescence (AIE) emitters [3]. The former two emitter types use a
co-doping technique to construct a host-guest system in the emitting layer (EML), thus
avoiding low efficiency induced by the concentration quenching. In contrast, the specific
molecular design of AIE-active chromophores such as the propeller-shaped structure can
prevent π-π interaction in the aggregates and thus significantly mitigates exciton quenching [3].
Therefore, AIE-based emitters could be used in OLEDs to construct a non-doped EML, which
could significantly reduce manufacturing complexity.
For OLED development, the fabrication of blue-emitting devices is a very arduous
process, which requires carrier transport materials or a host with an extremely wide energy
bandgap. Despite considerable research and development, blue-emitting OLEDs remains a
significant bottleneck and present substantial room for further improvement in terms of
electroluminescence (EL) efficiency and device lifetime. Many groups continue to seek to
improve OLED performance, incorporating blue-emitting AIE emitters to meet the stringent
requirements for practical applications. For example, in 2010, Tang’s group proposed a new
series of luminogens featuring TPA dimer (DTPA) cores with different tetraphenylethene (TPE)
peripheral moieties. The maximum efficiency of sky-blue OLEDs with 4TPEDTPA reached
3.7% (8.0 cd/A) [4]. In 2010, the same group developed a new AIE luminogen combining a
pyrene core with four TPE peripheries. The current and power efficiencies of such sky-blue

ACCEPTED MANUSCRIPT
OLEDs could respectively obtain 12.3 cd/A and 7.0 lm/W [5]. In 2014, Tang’s group modified
the NPB by integrating the covalent integration of TPE into its structure, subsequently named
(TPE-NPB). The maximum current, power efficiencies and external quantum efficiency (EQE)
of the TPE-NPB-based device respectively reached 13.1 cd/A, 7.8 lm/W and 4.2% [6]. In 2015,
they proposed a thermally stable luminogen, BTPE-PI, which possessed balanced carrier
injection properties and strong deep-blue AIE. The external quantum efficiency of the
non-doped device was up to 4.4% [7]. In 2016, Zhang’s group proposed an AIE-type blue
emitter by constructing a TPE group with two phenanthro[9,10-d]imidazole (PI) groups. The
fabricated device with mTPE-DPI as the non-doped EML exhibited a saturated blue emission
with a peak EQE of 3.69% [8].
Among the reported AIE fluorophores, tetraphenylethene (TPE) was a prototype
molecule because of its facile synthesis and excellent AIE effect; however, the TPE-based
derivatives usually demonstrate a strong tendency for crystallization [9]. If TPE is linked to
other big aromatic groups, the intrinsic extension of the π conjugation results in a longer
wavelength emission. Therefore, blue AIE-active chromophores are still very insufficient. In
this study, we have designed and synthesized new TPE type derivatives containing biphenyl or
anthracene fragments. The derivatives can form amorphous layers and could be used as
light-emitting materials for OLEDs due to bipolar charge transport in their layers and high
mobility values for organic low-molar-mass semiconductors. As a result, the sky-blue device
with the newly developed AIE emitter achieved a peak efficiency of up to 3.8%, demonstrating
the effectiveness of our specific molecular design. In addition, we not only investigated the EL
application of the synthesized AIE fluorophores but also further used the tandem device
structure to boost the EL efficiency of the AIE-based devices. In general, a qualified tandem
device should possess an external quantum efficiency about twice that of an ordinary device
with a single emission unit (SMU) [10-11]. As indicated, the peak EL efficiencies of the
tandem device achieved 8.2% (19.4 cd/A and 8.2 lm/W), which exceeded twice that of the

ACCEPTED MANUSCRIPT
device with a single emission unit. These outcomes indicate the high potential of the tandem
device structures for developing AIE-based OLEDs.
2. Experimental section
2.1 Instrumentation
Nuclear magnetic resonance (NMR) spectra were recorded using a Bruker Avance III
(400ꢀMHz) apparatus. Mass spectra were obtained on a Waters ZQ 2000 spectrometer.
Differential scanning calorimetry measurements were carried out using a Bruker Reflex II
thermosystem. Thermogravimetric analysis was performed on a Netzsch STA 409. TGA and
DSC curves were recorded in a nitrogen atmosphere at a heating rate of 10o C/min.
Photoelectron emission spectrometry was used to record the ionization potentials of studied
materials in solid state [12]. The samples for the experiments were fabricated by thermal
vacuum evaporation of the layers of the compounds onto fluorine doped tin oxide (FTO) coated
glass slides. Details of the electron photoemission experiments were previously published [13].
Hole-transporting properties of the layers of the synthesized compounds were tested by the
time-of-flight technique, which we have described earlier [14-15]. The thickness (d) of the
organic charge transporting layers was measured by the charge extraction using the linearly
increasing voltage (CELIV) technique [16].
2.2 Synthesis of complexes 1-3
2-Bromo-1,1,2-triphenylethylene (1), 4-biphenylboronic acid, 9-anthraceneboronic acid,
bis(triphenylphosphine)palladium(II) dichloride (Pd(PPh3)2Cl2) and potassium hydroxide were
purchased from Aldrich and used as received.
2-(4-Biphenyl)-1,1,2-triphenylethylene
(2).
0.56
g
(1.6
mmol)
of
2-bromo-1,1,2-triphenylethylene (1), 0.5 g (2.5 mmol) of 4-biphenyl boronic acid, 0.05 g (0.07
mmol) of PdCl2(PPh3)2 and 0.5 g (8.9 mmol) of powdered potassium hydroxide were stirred in

ACCEPTED MANUSCRIPT
10 ml of THF containing degassed water (1 ml) at 80 °C under nitrogen for 2 h. After thin layer
chromatography (TLC) to control the reaction, the mixture was cooled and quenched by the
addition of ice water. The product was extracted using ethyl acetate. The combined organic
extract was dried over anhydrous Na2SO4. The crude product was purified by silica gel column
chromatography using the mixture of toluene and hexane (vol. ratio 1:50) as an eluent. Yield: 0.5
g of white crystals. M.p.: 194 °C (DSC). MS (APCI+, 20 V): 408.1 ([M+H], 100 %). 1H NMR
(400 MHz, CDCl3, δ, ppm): 7.59-7.55 (m, 2H, Ar), 7.42-7.29 (m, 5H, Ar), 7.16-7.05 (m, 17H,
Ar). 13C NMR spectrum (CDCl3, δ, ppm): 143.8, 143.75, 142.78, 141.16, 140.67, 140.56,
138.88, 131.79, 131.44, 131.37, 128.70, 127.78, 127.71, 127.66, 127.17, 126.88, 126.50, 126.44,
126.22. Elemental analysis; found: C, 94.06; H, 5.93 %; molecular formula C32H24 requires C,
94.08; H, 5.92 %.
2-(9-Antracenyl)-1,1,2-triphenylethylene
(3).
0.5
g
(1.5
mmol)
of
2-bromo-1,1,2-triphenylethylene (1), 0.5 g (2.2 mmol) of 9-anthracene boronic acid, 0.04 g
(0.05 mmol) of PdCl2(PPh3)2 and 0.4 g (7.1 mmol) of powdered potassium hydroxide were
stirred in 7 ml of THF containing degassed water (1 ml) at 80 °C under nitrogen for 2 h. After
TLC control, the reaction mixture was cooled and quenched by the addition of ice water. The
product was extracted using ethyl acetate. The combined extract was dried over anhydrous
Na2SO4. The crude product was purified by silica gel column chromatography using a mixture
of toluene and hexane (vol. ratio 1:50) as an eluent. Yield: 0.24 g of yellowish crystals. M.p.:
1
233 °C (DSC). MS (APCI+, 20 V): 433.1 ([M+H], 100 %). H NMR (400 MHz, CDCl3, δ,
ppm): 8.41 (d, 2H, J=9.2Hz, Ar), 8.32 (s, 1H, Ar), 7.93 (d, 2H, J=8.4Hz, Ar), 7.47-7.38 (m, 4H,
Ar), 7.36-7.3 (m, 2H, Ar), 7.30-7.25 (m, 3H, Ar), 7.11-7.07 (m, 2H, Ar), 7.04-7.00 (m, 3H, Ar),
6.92-6.88 (m, 2H, Ar), 6.78-6.69 (m, 3H, Ar). 13C NMR spectrum (CDCl3, δ, ppm): 144.70,
143.30, 143.08, 142.66, 137.46, 136.27, 131.41, 131.37, 130.22, 130.12, 128.96, 128.64,
128.00, 127.92, 127.03, 126.03, 126.92, 126.73, 126.59, 126.50, 126.46, 125.81, 124.89.
Elemental analysis; found: C, 94.42; H, 5.57 %; molecular formula C40H28 requires C, 94.45;
H, 5.55 %.

ACCEPTED MANUSCRIPT
2.3 Characterization by X-ray diffractometer
The crystallographic nature of the powder materials 2 and 3 was determined using D8 Discover
X-ray diffractometer (Bruker AXS GmbH) with Cu Kα (λ= 1.54 Å) X-ray source. Parallel beam
geometry with 60 mm Göbel mirror (i.e. X-ray mirror on a high precision parabolic surface)
was used. This configuration enables transforming the divergent incident X-ray beam from a
line focus of the X-ray tube into a parallel beam that is free of Kβ radiation. The primary side
also had a Soller slit with an axial divergence of 2.5 º and a slit of 1 mm. The secondary side
had a LYNXEYE (1D mode) detector with an opening angle of 2.160 º and slit opening of 6.0
mm. X-ray generator voltage and the current was 40.0 kV and 40 mA, respectively. Coupled
2θ/θ scans were performed in the range of 3.0-130.0 º with a step size of 0.127 º, time per step
of 19.2 s and auto-repeat function enabled. Processing of the resultant diffractograms was
performed with DIFFRAC.EVA software.
2.4 Device fabrication
Low molar mass organic materials were purchased from Shine Materials Technology. All
organic compounds were subjected to temperature-gradient sublimation under high vacuum
before use. The indium-tin-oxide (ITO)-coated glass substrate was pretreated with UV-Ozone
for 5 minutes after a routine cleaning procedure. The organic and metal layers were deposited
onto ITO-coated glass by vacuum evaporation in a vacuum chamber with a base pressure of
10-6 Torr. The active area of the device was set to 2 x 2 mm2, as defined by the shadow mask
used for cathode deposition. Current density-voltage-luminance characterization was done
using two Keithley 2401 current source-measure units equipped with a calibrated
Si-photodiode. The electroluminescent spectra of the devices were recorded using an Ocean
Optics spectrometer.

ACCEPTED MANUSCRIPT
3. Results and discussion
3.1 Synthesis and structural characterization
The synthesis of the triphenylethylene type compounds with biphenyl or anthracenyl
fragments (2 and 3) was carried out by a one-step synthetic route as shown in Scheme 1.
2-(4-Biphenyl)-1,1,2-triphenylethylene (2) and 2-(9-antracenyl)-1,1,2-triphenylethylene (3)
were synthesized from commercially available 2-bromo-1,1,2-triphenylethylene (1) by Suzuki
reaction with an excess of 4-biphenyl boronic acid or 9-anthracene boronic acid,
correspondingly, in tetrahydrofurane.
The synthesized objective derivatives 2 and 3 were identified by mass spectrometry and
NMR spectroscopy. The data were found to be in good agreement with the proposed structures.
The materials were well soluble in common organic solvents. Transparent thin films of these
materials could be prepared by spin coating from solutions or by vacuum evaporation.
3.2 Thermal properties
The behavior under heating of compounds 2 and 3 was tested by thermogravimetric
analysis (TGA) and differential scanning calorimetry (DSC) under a nitrogen atmosphere. Both
the materials demonstrated acceptable thermal stability. The 5 % mass loss was established at
o
o
264 C for 2 and at 250 C for 3, as confirmed during the TGA analyses with a heating rate of
10o C/min. It is evident that the thermal stability of the materials depends slightly on
substituents connected to 1,1,2-triphenylethylene core. 9-Antracenyl-substituted derivative 2
demonstrates slightly higher thermal stability.
Both the TPE type derivatives 2 and 3 were obtained after synthesis as crystalline
materials demonstrating polymorphism; however, they could also form amorphous materials
with clear glass transition temperatures (Tg) as demonstrated by DSC experiment. The DSC
thermo-grams of material 2 are shown in Fig. 1 as an example. When the crystalline sample of
2 was heated, an endothermic peak due to transformation to other crystalline modification was

ACCEPTED MANUSCRIPT
observed at 182 oC and the new crystals melted at 194 oC. When the melted sample was cooled
down, it formed an amorphous material. During the second heating, the glass-transition
temperature was observed at 53 oC followed by crystallization at 112 oC, which formed crystals
o
of second crystalline modification with a melting temperature (Tm) of 194 C. The crystalline
sample of compound 3 demonstrated analogous behavior during the DSC test. It melted on first
o
o
heating at 233 C and formed glass (Tg = 78 C) upon cooling. During the second heating, the
glass-transition temperature was observed at 78 oC followed by crystallization at 145 oC, which
formed crystals with Tm of 233 oC.
X-ray diffraction (XRD) patterns of synthesized powders of the materials 2 and 3 were
also recorded (Figs. S5 and S6) and confirmed that the derivatives were obtained as crystalline
materials after synthesis. The XRD patterns displayed the di erent powder structures with the
crystallinity of 79.5 and 80.2 % for derivatives 2 and 3, respectively.
3.3 Photophysical properties
We first performed the photoluminescence (PL) measurement of both compounds to
assess their AIE properties. Both compounds were dissolved in different blending ratios of
THF/water solution, aiming to facilitate the molecular aggregation by varying the polarity of
the solvent. Figures 2(a) and (b) respectively show the PL spectra and the corresponding PL
intensity of compounds 2 and 3 in THF with different water fractions. As indicated, the PL
intensity of both compounds increased with the water fraction, indicating that the degree of AIE
acutely increased with the solvent polarity. In addition, the PL intensities of both solvent
samples abruptly increased when the water fractions exceeded 70%, which was a common
behavior similar to other AIE luminogenic compounds [17-19]. Figure 2(c) presents
photographs of compounds 2 and 3 in different THF/water fractions under UV light
illumination at room temperature. Clearly, an intense AIE was obtained when the water
fractions exceeded 70% in both solvent samples, and then the intensity continuously increased
to water fractions reaching 90%. In addition, we conducted PL quantum yield (PLQY)

ACCEPTED MANUSCRIPT
measurements of compounds 2 and 3 in the mixture of solvents (THF: water = 1:9), with
respective estimated PLQY values of 0.60 and 0.36.
Furthermore, we also performed PL measurements of the thin film samples. PL spectra of
the compound 2 and 3 neat films measured at room temperature and 77K are shown in Fig.
2(d). The PL spectra of the solvent samples with 90% water fractions were also added in the
figure for comparison. The solid samples with 2 or 3 exhibited spectral profiles similar to the
corresponding solvent samples at room temperature, illustrating that the radiation would result
from the same aggregation induced excited states. The slightly wider and red-shifted spectral
profiles observed in the solvent samples might be attributed to the solvation effect induced by
the strong polarity of water. On the other hand, PL spectra of both thin film samples with 2 and
3 were conducted at 77 K to investigate their phosphorescence. As indicated, the respective
phosphorescence peaks of 2 and 3 were recorded at 595 nm and 571 nm, with corresponding
triplet energy bandgaps of about 2.08 eV and 2.17 eV. In general, the PL spectral profiles of an
emitter will red-shift with the increased solution polarity. However, solvatochromism induced
red-shift is not pronounced in these molecules. Considering the difference of PL spectra of thin
film measured in RT and 77K, the large difference between S1 and T1 should result from the
considerable overlap of the LUMO/HOMO. Thus, no significant charge separation in the
excited state leads the traditional solvent effects that are not observed in this series of
molecules.
3.4 Ionization potentials and charge mobility
The ionization potentials (Ip) of layers of the compounds synthesized were measured by
the electron photoemission method. The photoemission spectra of the thin amorphous films of
compounds 2 and 3 are presented in Fig. 3(a). The established Ip values were 5.83 eV for the
layer of material 2 and 6.07 eV for the layer of 3. It is evident that the ionization potentials of
the materials are rather high and additional hole transporting layers would be necessary for the

ACCEPTED MANUSCRIPT
injection of holes from the ITO anode into the emitting layers of the AIE materials 2 and 3.
To check the potential of using 2 and 3 derivatives in electronic devices,
charge-transporting properties in the films of the materials were studied by the time-of-flight
(TOF) technique (Fig. 3(b)). At an electric field of 1.06×105 V/cm, the highest hole mobility
value of 2.0×10-2 cm2/V·s was obtained for compound 3; while a hole mobility value almost
one order lower of 1.9×10-3 cm2/V·s was obtained for similar compound 2 substituted by the
biphenyl group at an electric field of 1.06×105 V/cm. Very similar electron mobility values
(2.0×10-3 cm2/V·s at an electric field of 5.38×105 V/cm and 1.9×10-2 cm2/V·s at an electric field
of 1.06×105 V/cm) were obtained for compounds 2 and 3 in terms of their hole mobility. We
can predict that the compounds can be used as light-emitting materials for OLEDs due to their
bipolar charge transport and high mobility values as organic low-molar-mass semiconductors.
Hole and electron mobility values at the same electric fields for the derivatives are comparable
and a charge balance in a light-emitting layer of 2 or 3 based OLEDs could be achieved. The
thermal, electrochemical, and photophysical properties of compounds 2 and 3 are summarized
in Table 1.
Table 1. Selected thermal, electrical, and photophysical properties of compounds 2 and 3
Mobility (cm2/V·s)
Hole Electron
IP a
λmax.
λmax.
λmax.
ΦPL
b
c
d
b
Tg
Td
(oC)
(oC)
(eV)
(nm)
490
(nm)
478
(nm)
595
(%)
60
2
3
53
264
1.9×10-3 2.0×10-3
5.83
78
250
2.0×10-2 1.9×10-2
6.07
472
468
571
36
a. IP values were estimated from electron photoemission spectra. b. Experiments were conducted
in the mixture of THF and water with a ratio of 1:9. c. PL spectrum of thin film samples were
performed at room temperature. d. PL spectrum of thin film samples were executed at 77K.
3.5 Device performance
Since compounds 2 and 3 are recognized as efficient blue AIE-based emitters, both were

ACCEPTED MANUSCRIPT
selected for use as the non-doped emitting layer (EML) to examine their EL applications.
Herein,
1,1-bis[(di-4-tolylamino)phenyl]cyclohexane
(TAPC)
and
4,4’,4’’-tris(carbazol-9-yl)-triphenylamine (TCTA) were adopted as the hole transport layers
(HTLs) [20-21]. Considering both 2 and 3 possessed high ionization potentials (i.e. 6.07 and
5.83 eV), the high energy barriers between the ITO anode with the EML caused the device
architecture to inevitably adopt a specific structure design for facilitating hole injection.
Consequently, the respective HOMO levels of TAPC and TCTA were estimated to be 5.4 eV
and 5.7 eV, these two HTLs could be combined to form a stepwise HTL for advancing smooth
hole injection into EML. In addition, the hole mobility of TAPC was estimated to be 10-2
cm2/V·s whereas TCTA possesses a smaller hole mobility of 10-4 cm2/V·s. Because the carrier
recombination condition depends on the carrier transport capability of the AIE molecule, the
carrier balance in a non-doped EML is much harder to regulate than that in a doped EML in
general. Therefore, the selection of this combination of TAPC and TCTA could regulate the
hole and thus increase the possibility of carrier recombination. On the other hand,
1,3,5-tri[(3-pyridyl)-phen-3-yl]benzene (TmPyPB) was chosen as the electron transport layer
(ETL) because of its high carrier transport capability and high triplet energy gaps [22].
Furthermore, a wide triplet-energy-bandgap material, 4,4’-N,N’-dicarbazolylbiphenyl (CBP),
was inserted between the EML and ETL to serve as a thin exciton confining layer (ECL),
promoting exciton confinement as well as carrier adjustment [23]. Moreover, the thicknesses of
each organic layer were carefully adjusted to enhance radiation output and carrier balance.
Consequently, the sky-blue and blue OLEDs (i.e., devices A and B) were fabricated with the
following generalized architecture: ITO (120 nm)/TAPC (65 nm)/TCTA (5 nm)/Compounds 2
or 3 (20 nm)/CBP (2 nm)/TmPyPB (50 nm)/LiF (0.8 nm)/Al (150 nm), for which LiF and
aluminum were respectively used as the electron injection layer and reflective cathode. Figure 4
depicts the structures of the employed materials along with the schematic architecture of the
as-fabricated non-doped OLEDs.

ACCEPTED MANUSCRIPT
Figure 5 and Table 2 respectively depict the corresponding electroluminescent (EL)
characteristics and numeric data. Figure 5(a) shows the EL spectrum of devices measured at
1000 cd/m2, for which the profiles are analogous to the corresponding PL profiles of thin film
samples, confirming that these EL emissions indeed result from the AIE phenomenon of the
molecules 2 and 3. Furthermore, the pure AIE indicated that the exciton formation zone of both
devices A and B were generated and well confined within the EML [24]. In addition, the values
of full width at half maximum (FWHM) of the EL spectra were respectively calculated to be
96.5 nm and 80.0 nm for 2 and 3. Clearly, compound 3 presents a bluer AIE with a smaller
FWHM compared to compound 2, which gave a higher color purity to meet the display
application requirement. In contrast, compound 2 possesses a wider EL spectral profile,
suitable for use in the construction of a white OLED with high color rendering [25].
Figure 5(b) shows the current density-voltage-luminance (J-V-L) characteristics. As
indicated, although 3 possesses higher carrier mobilities and lower HOMO levels than 2, the
narrower energy band gap of 2 in device A contributed to a higher current density under higher
operating voltages, leading to a rather higher luminance. In addition, since the sensitivity of the
human eye is a function of wavelength, which is higher responsive to green emissions.
Therefore, the greenish-blue EL emission gave device A a higher luminance. [26]. The
respective recorded maximum luminance of devices A and B were 15561 cd/m2 and 8512 cd/m2.
Figures 5(c) and 5(d) respectively depict the efficiency curves of devices A and B. Moreover,
device A with a sky-blue emission achieved a peak efficiency of 3.8% (8.8 cd/A, 7.4 lm/W),
while the efficiencies remained high at 3.6% (8.4 cd/A, 5.5 lm/W) at a practical luminance of
100 cd/m2. The blue-emitting device B exhibited a maximum efficiency of 2.2% (4.0 cd/A, 3.0
lm/W) and slightly reduced efficiencies of 2.2% (3.9 cd/A, 3.0 lm/W) at 100 cd/m2. In addition,
the respective optical out-coupling factor and spin factor are estimated to be about 20~25% and
25%. Therefore, considering the PLQY values of the materials 2 and 3, both devices A and B
exhibited nearly 100% carrier balance factor, illustrating excellent device architecture designs.

ACCEPTED MANUSCRIPT
Notably, both devices A and B showed very small efficiency roll-off values, which could be
further verified by the fact that the EQE declined by half at a current density (J1/2) of 179.8 and
411.3 mA/cm2 for devices A and B, respectively [27]. These high values of J1/2 obtained in both
devices indicated insignificant exciton and polaron quenching. Nevertheless, devices A and B
still present relatively inferior performance compared to a common fluorescent OLED. Thus,
we attempted to modify the device architecture to construct a tandem device, aiming to double
the EL efficiency to fulfill the demands of display applications [10-11].
A tandem device consists of several individual EMUs connected by charge generation
layers (CGL). This device structure is proposed to obtain a qualified lighting source or a
display panel with an ample flux as well as high efficiencies, and thus has been a compelling
research topic in industry. Typically, if the device architecture is well-designed, the values of
efficiency, luminance, and operating voltage will be proportional to the number of connected
EMUs [28]. Furthermore, the CGL plays the most crucial role in the tandem OLED
performance. A typical CGL can be constructed through combining a p-doped organic layer and
an n-doped organic layer, which could generate abundant charges under the biases and then
separate the charges to facilitate the carrier injection into the EMUs [29]. In this study, an
intermediate CGL consists of an n-type doped layer and a 1,4,5,8,9,11-hexaazatriphenylene
hexacarbonitrile (HATCN) layer [30-31]. Herein, the electron transporting TmPyPB
co-doped with Li2CO3 was used as an n-type layer whereas HATCN was used to further
reinforce the electron injection due to its strong oxidizing property [32-33]. The charge
generation interface was expected to form between the HATCN layer and the adjacent
hole-transporting TAPC layer [31]. The schematic diagram of carrier generation and transport
in the intermediate CGL is depicted in Fig. 4(b). As a consequence, the optimal architectures of
the tested tandem AIE-based OLEDs (i.e., devices TA and TB) consisted of ITO (120
nm)/TAPC (65 nm)/TCTA (5 nm)/Compound 2 or 3 (20 nm)/CBP (2 nm)/TmPyPB (40
nm)/TmPyPB doped with 20 wt.% Li2CO3 (10 nm)/HATCN (6 nm)/TAPC (55 nm)/TCTA (5

ACCEPTED MANUSCRIPT
nm)/Compound 2 or 3 (20 nm)/CBP (2 nm)/TmPyPB (50 nm)/LiF (0.8 nm)/Al (150 nm). The
schematic device architectures and the corresponding EL performance of devices TA and TB
were also shown in Figs. 4 and 5.
The EL spectra of devices TA and TB measured at 1000 cd/m2 are also shown in Fig.
5(a). As indicated, the respective EL spectra of devices TA and TB were akin to those of
devices A and B. The slight variation observed in the spectral profiles between the
conventional device and the tandem devices might result from the effect of optical interference
[34]. In addition, no interfering emission from either exciton confining material or
carrier-transport materials was observed, indicating that the exciton generated from the
AIE-based molecules in both EMUs. The J-V-L curves showed that the current density of
device TB was lower than that of device TA, which was similar to the manifestation of devices
A and B (cf. Fig. 5(b)). As indicated, the turn-on voltage of device TA was double that of
device A (i.e. 6.2 V). However, device TB exhibited a rather higher turn-on voltage of 7.1 V,
which might be attributed to the higher energy bandgap of compound 3 or carrier imbalance in
the tandem structure. The maximum operating voltage of the source-measurement instrument is
limited to 20 V, and the maximum luminance of the tandem devices is collected in Table 2 for
reference.
As shown in Figs. 5(c) and 5(d), device TB exhibited a peak EL efficiency of 4.6% (9.5
cd/A and 2.6 lm/W), while device TA achieved a much greater maximum efficiency of 8.2%
(19.4 cd/A and 8.2 lm/W). Clearly, tandem devices TA and TB presented more than two-fold
maximum efficiency compared to those of the conventional devices A and B, indicating the
effectiveness of the CGL structure and the appropriate design for the device architecture. In
addition, it’s worth mentioning that the quantum efficiency and the luminance efficiency of
device TA reached the recorded high of the sky-blue or blue-emitting AIE-based OLEDs
[35-37]. On the other hand, device AT with a tandem structure exhibited improved power
efficiencies compared with that of the single-stack device A. The power efficiency of devices A

ACCEPTED MANUSCRIPT
and AT might differ in terms of changes in carrier balance conditions. Moreover, the
thicknesses of TmPyPB and TAPC layers in both emission units of device AT were thinner
than that of device A, thus device AT could be operated with lower operating voltages.
Furthermore, at a high luminance of 100 cd/m2 (1000 cd/m2), the EL efficiencies of devices TA
and TB respectively maintained efficiency levels of 8.1% (6.9%) and 4.6% (4.2%); hence, they
exhibited efficiency drops of only 1% (16%) and 0% (9%) from their respective peak values to
those recorded at 100 cd/m2 (1000 cd/m2). Similarly, the tandem devices also retained high
efficiency at high luminance levels and showed negligible efficiency roll-off.
Table 2. EL Characteristics of studied blue-emitting OLEDs with AIE emitters
Device
A
B
AT
2
BT
3
emitter
2
3
[a]
[b]
[c]
[a]
[b]
[c]
[a]
[b]
[c]
[d]
3.8
3.6
3.3
8.8
8.4
7.5
7.4
5.5
3.2
3.1
2.2
2.2
2.0
4.0
3.9
3.6
3.0
1.9
1.2
3.2
8.2
8.1
6.9
19.4
19.3
16.4
8.2
6.0
3.8
6.2
4.6
4.6
4.2
9.5
9.4
8.5
2.6
2.5
1.7
7.1
External quantum
efficiency (%)
Luminance efficiency
(cd/A)
Power efficiency
(lm/W)
Von (V)
15561
[14.2]
8512
8438
3923
Max. Luminance (cd/m2) [V]
[15.4]
[20.0]
[20.0]
λmax. (nm)
[c]
[c]
[b]
[c]
484
470
481
466
FWHM (nm)
96.5
80.0
105.0
83.0
(0.23, 0.38)
(0.22, 0.35)
(0.18, 0.22)
(0.18, 0.22)
(0.21, 0.36)
(0.21, 0.35)
(0.18, 0.27)
(0.18, 0.25)
CIE 1931 coordinates
[a] Maximum efficiency; [b] recorded at 102 cd/m2; [c] measured at 103 cd/m2; [d] turn-on
voltage measured at 1 cd/m2.

ACCEPTED MANUSCRIPT
Overall, these AIE-based devices with single or double EMUs presented great EL
performance, which raises the feasibility of using the AIE emitters in commercial applications.
Furthermore, the AIE-based emitter affords the possibility for an efficient device with a
non-doped EML, which could reduce the complexity of manufacturing in tandem devices and
thus increase production yields.
4. Conclusions
In summary, molecules consisting of a TPE moiety decorated with biphenyl or
anthracenyl fragments were designed to generate blue-emitting AIE. The strong AIE
characteristics of compounds 2 and 3 were confirmed by the PL measurements in highly polar
THF/water mixed solutions and their respective PLQY values were estimated to be 0.60 and
0.36. From the DSC results, 9-antracenyl-substituted derivative 3 presented a higher glass
transition temperature than that of 2 due to the slightly larger molecular configuration of 3. In
addition, the adequate decomposition temperatures observed in both compounds indicated that
they could be effectively deposited to form thin films by vacuum evaporation. The TOF
measurements indicated that both compounds 2 and 3 possessed high hole and electron
mobilities. Their bipolar carrier transportability benefited the realization of non-doped OLEDs
with carrier balance. As a result, the peak efficiencies of both sky-blue and blue AIE OLEDs
with compounds 2 and 3 (i.e. devices A and B) respectively reached 3.8% (8.8 cd/A, 7.4 lm/W)
and 2.2% (4.0 cd/A, 3.0 lm/W). Furthermore, the carefully optimized device architecture led
both to achieve nearly 100% carrier balance factor. Moreover, both devices A and B showed
very small efficiency roll-off compared to the other reported TPE-based blue-emitting OLEDs,
demonstrating the effective design concepts of a TPE moiety decorated with biphenyl and
anthracenyl fragments. In addition, a specific CGL consisting of TmPyPB: Li2CO3 and
HATCN layers was adopted to construct tandem AIE OLEDs (i.e. devices TA and TB), aiming
to significantly improve efficiency at practical luminance. As indicated, devices TA and TB

ACCEPTED MANUSCRIPT
exhibited respective peak EL efficiencies of 4.6% (9.5 cd/A and 2.6 lm/W) and 8.2% (19.4
cd/A and 8.2 lm/W), which were more than double those of the conventional devices A and B.
These excellent results illustrate the high potential of the AIE-based luminogenic compounds
for use in tandem devices.
Acknowledgments
The authors gratefully acknowledge financial support from the Ministry of Science and
Technology of Taiwan (MOST 107-2221-E-155-028-) and from the Research Council of
Lithuania (S-MIP-17-64). We are thankful to Dr. Algirdas Lazauskas for X-ray diffraction
measurements.
Appendix A. Supplementary data
Supplementary data associated with this article, TOF current transients in films of the materials
2 and 3, XRD patterns of synthesized powders of the materials 2 and 3, and DSC curves of 3
can be found in the online version.
SCHEME CAPTIONS
Scheme 1. Synthetic route of TPE type derivatives 2 and 3.
FIGURE CAPTIONS
Fig. 1. DSC curves of the objective material 2. Heating rate: 10 oC/min.
Fig. 2. (a) PL spectra of compounds 2 and 3 in THF with different water fractions; (b) the
intensity of the PL spectra; (c) compounds 2 and 3 in THF with different water fractions
under UV light illumination at room temperature; (d) PL spectra of compounds 2 and 3
in solutions and neat films measured at room temperature and 77K.
Fig. 3. (a) Photoemission spectra of thin amorphous films of compounds 2 and 3; (b)

ACCEPTED MANUSCRIPT
visualization of the experimental TOF charge mobilities in the layers of compounds 2
and 3.
Fig. 4. (a) Structural drawings of the materials; (b) schematic structures of devices A, B, AT,
and BT.
Fig. 5. (a) Normalized EL spectra at a luminance of 103 cd/m2 for devices A, B, AT, and BT; (b)
current density-luminance-voltage (J-V-L) characteristics, (c) external quantum
efficiency versus luminance, and (d) luminance efficiency/power efficiency versus
luminance for devices A, B, AT, and BT.

ACCEPTED MANUSCRIPT
Reference
[1] W.-Y. Hung, T.-C. Wang, P.-Y. Chiang, B.-J. Peng, K.-T. Wong, ACS Appl. Mater.
Interfaces 9 (2017) 7355.
[2] Y.-J. Lien, T.-C. Lin, C.-C. Yang, Y.-C. Chiang, C.-H. Chang, S.-H. Liu, Y.-T. Chen, G.-H.
Lee, P.-T. Chou, C.-W. Lu, Y. Chi, ACS Appl. Mater. Interfaces 9 (2017) 27090.
[3] Z. Zhao, J.W.Y. Lam, B.-Z. Tang, J. Mater. Chem. 22 (2012) 23726.
[4] W.-Z. Yuan, P. Lu, S. Chen, J.W.Y. Lam, Z. Wang, Y. Liu, H.S. Kwok, Y. Ma, B.-Z. Tang,
Adv. Mater. 22 (2010) 2159.
[5] Z. Zhao, S. Chen, J.W.Y. Lam, P. Lu, Y. Zhong, K.S. Wong, H.S. Kwok, B.-Z. Tang, Chem.
Commun. 46 (2010) 2221.
[6] W. Qin, J. Liu, S. Chen, J.W.Y. Lam, M. Arseneault, Z. Yang, Q. Zhao, H.S. Kwok, B.-Z.
Tang, J. Mater. Chem. C 2 (2014) 3756.
[7] W. Qin, Z. Yang, Y. Jiang, J.W.Y. Lam, G. Liang, H.S. Kwok, B.-Z. Tang, Chem. Mater. 27
(2015) 3892.
[8] C. Li, J. Wei, J. Han, Z. Li, X. Song, Z. Zhang, J. Zhang, Y. Wang, J. Mater. Chem. C 4
(2016) 10120.
[9] Y. Dong, J.W.Y. Lam, A. Qin, J. Liu, Z. Li, B.-Z. Tang, Appl. Phys. Lett. 91 (2007) 011111.
[10] M.-Y. Chan, S.-L. Lai, K.-M. Lau, M.-K. Fung, C.-S. Lee, S.-T. Lee, Adv. Funct. Mater. 17
(2007) 2509.
[11] C.-H. Chang, Z.-J. Wu, Y.-H. Liang, Y.-S. Chang, C.-H. Chiu, C.-W. Tai, H.-H. Chang,
Thin Solid Films 548 (2013) 389.
[12] E. Miyamoto, Y. Yamaguchi, M. Yokoyama, Electrophotography 28 (1989) 364.
[13] N.A. Kukhta, D. Volyniuk, L. Peciulyte, J. Ostrauskaite, G. Juska, J.V. Grazulevicius,
Dyes. Pigm. 117 (2015) 122.
[14] V. Mimaite, J. V. Grazulevicius, R. Laurinaviciute, D. Volyniuk, V. Jankauskas, G. Sini, J.
Mater. Chem. C 3 (2015) 11660.
[15] D. Lo, C.-H. Chang, G. Krucaite, D. Volyniuk, J. V. Grazulevicius, S. Grigalevicius, J.
Mater. Chem. C 5 (2017) 6054.

ACCEPTED MANUSCRIPT
[16] G. Juska, K. Arlauskas, M. Viliunas, K. Genevicius, Phys. Rev. B 62 (2000) 16235.
[17] S. Odabas, E. Tekin, F. Turksoy, C. Tanyeli, J. Mater. Chem. C 1 (2013) 7081.
[18] J. Huang, R. Tang, T. Zhang, Q. Li, G. Yu, S. Xie, Y. Liu, S. Ye, J. Qin, Z. Li, Chem.-Eur. J.
20 (2014) 5317.
[19] F. Tang, J. Peng, R. Liu, C. Yao, X. Xu, L. Li, RSC Adv. 5 (2015) 71419.
[20] K. Goushi, R. Kwong, J. J. Brown, H. Sasabe, C. Adachi, J. Appl. Phys. 95 (2004) 7798.
[21] W. Song, H.L. Lee, J.Y. Lee, J. Mater. Chem. C 5 (2017) 5923.
[22] S.-J. Su, T. Chiba, T. Takeda, J. Kido, Adv. Mater. 20 (2008) 2125.
[23] H. Matsushima, S. Naka, H. Okada, H. Onnagawa, Curr. Appl. Phys. 5 (2005) 305.
[24] C.-H. Chang, Z.-J. Wu, C.-H. Chiu, Y.-H. Liang, Y.-S. Tsai, J.-L. Liao, Y. Chi, H.-Y. Hsieh,
T.-Y. Kuo, G.-H. Lee, H.-A. Pan, P.-T. Chou, J.-S. Lin, M.-R. Tseng, ACS Appl. Mater.
Interfaces 5 (2013) 7341.
[25] C.-H. Chang, C.-C. Chen, C.-C. Wu, S.-Y. Chang, J.-Y. Hung, Y. Chi, Org. Electron. 11
(2010) 266.
[26] D.-G. Chen, T.-C. Lin, C.-L. Chen, Y.-T. Chen, Y.-A. Chen, G.-H. Lee, P.-T. Chou, C.-W.
Liao, P.-C. Chiu, C.-H. Chang, Y.-J. Lien, Y. Chi, ACS Appl. Mater. Interfaces 10 (2018)
12886.
[27] H. Sasabe, J.-I. Takamatsu, T. Motoyama, S. Watanabe, G. Wagenblast, N. Langer, O.
Molt, E. Fuchs, C. Lennartz, J. Kido, Adv. Mater. 22 (2010) 5003.
[28] H.W. Bae, Y.H. Son, B.Y. Kang, J.M. Lee, H. Nam, J.H. Kwon, Opt. Express 23 (2015)
30701.
[29] H. Sun, Q, Guo, D. Yang, Y. Chen, J. Chen, D. Ma, ACS Photonics 2 (2015) 271.
[30] X. Zhang, M. Zhang, M. Liu, Y. Chen, J. Wang, X. Zhang, J. Zhang, W.-Y. Lai, W. Huanga,
Org. Electron. 53 (2018) 353.
[31] L.S. Liao, K.P. Klubek, Appl. Phys. Lett. 92 (2008) 223311.
[32] C.-H. Chang, M.-K. Hsu, S.-W. Wu, M.-H. Chen, H.-H. Lin, C.-S. Li, T.-W. Pi, H.-H.
Chang, N.-P. Chen, Phys. Chem. Chem. Phys. 17 (2015) 13123.
[33] S.M. Park, Y.H. Kim, Y. Yi, H.-Y. Oh, J.W. Kim, Appl. Phys. Lett. 97 (2010) 063308.
[34] C.-L. Lin, H.-W. Lin, C.-C. Wu, Appl. Phys. Lett. 87 (2005) 021101.

ACCEPTED MANUSCRIPT
[35] Y. Hong, J.W.Y. Lam, B.-Z. Tang, Chem. Soc. Rev. 40 (2011) 5361.
[36] C.Y. K. Chan, J.W.Y. Lam, Z. Zhao, S. Chen, P. Lu, H.H. Y. Sung, H.S. Kwok, Y. Ma, I.D.
Williams, B.-Z. Tang, J. Mater. Chem. C 2 (2014) 4320.
[37] M. Bian, Z. Zhao, Y. Li, Q. Li, Z. Chen, D. Zhang, S. Wang, Z. Bian, Z. Liu, L. Duan, L.
Xiao, J. Mater. Chem. C 6 (2018) 745.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT