Efficient green OLEDs achieved by a terbium(III) complex with photoluminescent quantum yield close to 100%

Article abstract of DOI:10.1007/s11426-021-1036-0

Electroluminescence of f-f transition lanthanide complex is a traditional topic for display over decades. Here we report highly efficient organic light-emitting diodes based on a new terbium(III) complex with novel ligand CPMIP (1-(4-cyanophenyl)-3-methyl-4-isobutyryl-pyrazoline-5-one). The high triplet energy level of CPMIP (3.0 eV) and inhibited quenching effects in the solid-state lead to a nearly 100% photoluminescent quantum efficiency of Tb(CPMIP)₃. The best Tb(CPMIP)₃ device exhibited maximum external quantum efficiency up to 19.7%, setting a new record of OLEDs based on f-f transition lanthanide complexes. [Figure not available: see fulltext.].

Full text of DOI:10.1007/s11426-021-1036-0

SCIENCE CHINA
Chemistry
•ARTICLES•
https://doi.org/10.1007/s11426-021-1036-0
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Efficient green OLEDs achieved by a terbium(III) complex with
photoluminescent quantum yield close to 100%
Zifeng Zhao1, Mengying Bian2, Chenjian Lin1, Xuzheng Fu1, Gang Yu1*, Huibo Wei3*,
Zhiwei Liu1, Zuqiang Bian1 & Chunhui Huang1
1Beijing National Laboratory for Molecular Sciences (BNLMS), State Key Laboratory of Rare Earth Materials Chemistry and Applications,
College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, China;
2State Key Laboratory for Mesoscopic Physics, Department of Physics, Peking University, Beijing 100871, China;
3Jiangsu JITRI Molecular Engineering Institute Co., Ltd., Changshu 215500, China
Received April 29, 2021; accepted May 28, 2021; published online August 20, 2021
Electroluminescence of f-f transition lanthanide complex is a traditional topic for display over decades. Here we report highly
efficient organic light-emitting diodes based on a new terbium(III) complex with novel ligand CPMIP (1-(4-cyanophenyl)-3-
methyl-4-isobutyryl-pyrazoline-5-one). The high triplet energy level of CPMIP (3.0 eV) and inhibited quenching effects in the
solid-state lead to a nearly 100% photoluminescent quantum efficiency of Tb(CPMIP)3. The best Tb(CPMIP)3 device exhibited
maximum external quantum efficiency up to 19.7%, setting a new record of OLEDs based on f-f transition lanthanide complexes.
luminescence, terbium, lanthanide complex, organic light-emitting diodes
Citation:
Zhao Z, Bian M, Lin C, Fu X, Yu G, Wei H, Liu Z, Bian Z, Huang C. Efficient green OLEDs achieved by a terbium(III) complex with
photoluminescent quantum yield close to 100%. Sci China Chem, 2021, 64, https://doi.org/10.1007/s11426-021-1036-0
1 Introduction
f transition of the central ions have extremely sharp emission
bands (in general, <10 nm) [14,15], and their potential in
display technology is never off the stage.
Organic light-emitting diodes (OLEDs) with narrowband
emission are urgent for next-generation display [1–3] of
large color gamut, high contrast ratio and device-flexibility.
High efficiency and favorable color purity are both critical
for OLED applications [4]. Over decades of efforts, external
quantum efficiency (EQE) exceeding 20% in OLED has
been demonstrated by phosphorescence [5], thermally acti-
vated delayed fluorescence (TADF) [6], and organic radical
[7] emitters. Recently, d-f transition lanthanide complex is
verified to utilize 100% excitons [8–10] and 20% EQE was
achieved in OLEDs [11]. Most of these materials have wide
emission spectral bandwidth, thus require special molecular
design [12] or incorporated with other dyes [13] to achieve
OLEDs with high color-purity. Lanthanide complexes with f-
Since the efforts of Kido et al. [16,17], OLEDs based on f-f
transition lanthanide complexes have been studied for over
30 years. Theoretically, The f-f transition lanthanide com-
plexes also have potential to reach 20% EQE in OLEDs [18].
However, till now, the most efficient case is a Tb(PMIP)3-
based device with EQE of 15% optimized by better con-
finement of triplet excitons [19]. The efficiency of f-f tran-
sition lanthanide complex device still has the potential to be
greatly improved.
High photoluminescent quantum yield (PLQY) of emitter
is crucial for achieving highly efficient OLEDs. The suitable
triplet energy level and suppressed non-radiative relaxation
of ligand are critical for constructing highly efficient lumi-
nescent lanthanide complexes [20]. In this work, we present
a new ligand CPMIP with a high triplet energy level up to
*Corresponding authors (email: yugang@pku.edu.cn; weihuibo@163.com)
© Science China Press and Springer-Verlag GmbH Germany, part of Springer Nature 2021
chem.scichina.com link.springer.com

2
Zhao et al. Sci China Chem
3.0 eV, which could sensitize Tb3+ efficiently both in dilute
solution and in solid state. Intriguingly, the PLQY of the
crystal and doped film is nearly 100%, which is about three
times higher than that in dilute solution. Cyano substitute
enhanced the thermal stability of Tb(CPMIP)3 by coordina-
tion to the adjacent terbium ion without changing the vola-
tility, which is important for fabricating devices by thermal
evaporation method. The optimized device based on
Tb(CPMIP)3 exhibited a maximum EQE of 19.7%, setting a
new record of f-f transition lanthanide complexes based
OLEDs.
obtain a clear solution. This solution was heated to 70 °C,
and TbCl3·6H2O (0.462 g, 1.24 mmol) in methanol was ad-
ded dropwise. Then 10 mL of deionized water was added and
white solids precipitated. Maintain the temperature for an-
other 10 min, then cool to room temperature and filter. The
solid was washed with a mixed solvent of ethanol/water (1:1)
and then dried under vacuum at 70 °C. The sample was then
sublimed at 330 °C and 10–5 Pa to obtain 0.768 g of white
crystalline solid. The yield was 68%. Anal. calcd for
C45H42N9O6Tb (found): N: 13.25 (13.08), C: 56.07 (56.08),
H: 4.52 (4.39).
2.4 General characterization
2 Experimental
Elemental analyses were performed on a VARIO EL analy-
zer (GmbH, Hanau, Germany). The high-resolution mass
spectrum was collected on Bruker Solarix XR FTMS (Ger-
many) by electrospray ionization (ESI) method. UV-visible
(UV-Vis) absorption spectra were recorded on a Shimadzu
UV-3100 spectrometer (Japan). Fluorescence and transient
PL decay spectra were measured on an Edinburgh Analytical
Instruments FLS980 spectrophotometer (UK). PLQYs were
measured on the C9920-02 absolute quantum yield mea-
surement system from Hamamatsu Company (Japan).
Thermogravimetric analysis (TGA) and differential scanning
calorimetry (DSC) were undertaken with a Q600SDT and
Q100DSC instrument (USA), respectively. Cyclic voltam-
metry was measured in nitrogen-purged CH2Cl2 solution at
room temperature using a CHI600C voltammetric analyzer
and tetrabutylammonium hexafluorophosphate (TBAPF6)
(0.1 M) as the supporting electrolyte. The conventional
three-electrode configuration consists of a platinum working
electrode, a platinum wire auxiliary electrode, and an Ag/
AgCl wire pseudoreference electrode.
2.1 Synthesis of CPMP
4-Cyanophenylhydrazine hydrochloride (2.86 g, 16.8 mmol)
was stirred and suspended in ethanol (50 mL), and then 2 M
NaOH solution was added dropwise to obtain a yellow solu-
tion. Ethyl acetoacetate (2.21 g, 17.0 mmol) was slowly added
to the solution obtained in the previous step. 2 M hydrochloric
acid was added to the solution to adjust the pH to 4–5, and the
mixture was stirred at room temperature for 40 min, then
heated to 70 °C and stirred overnight. The solution was then
concentrated to 30 mL by reduced pressure distillation, and a
large amount of a yellow solid precipitated from the solution.
The mixture was filtered and recrystallized from ethanol to
give a pale-yellow solid. Finally, 2.27 g 1-(4-cyanophenyl)-3-
methyl-pyrazoline-5-one (CPMP) was obtained after vacuum
drying at 70 °C. The yield was 67.5%. MS (m/z, ESI): calcd
199.1, found 200.1 (M+H+).
2.2 Synthesis of CPMIPH
Under N2 protection, CPMP (2.23 g, 11.2 mmol) and dry
Ca(OH)2 (0.88 g, 11.9 mmol) powder were added to 100 mL
of dry 1,4-dioxane and stirred. Then, isobutyryl chloride
(1.34 g, 12.6 mmol) was added dropwise to the suspension,
and then the solution was refluxed for one hour and then
cooled to room temperature. After this mixture was poured
into 200 mL of water and the pH was adjusted to neutral with
hydrochloric acid, the solids will precipitate directly. The
product was filtered and vacuum dried at 70 °C. 1.33 g white
solid CPMIPH was obtained by sublimation purification at
120 °C and 10–5 Pa. The yield was 44%. MS (m/z, ESI):
2.5 OLEDs fabrication and measurement
MoO3, mCP, DPEPO, TPBi, and LiF were purchased from
commercial sources without further purification. DCPPO
was synthesized and purified using the method reported in
the literature.
The indium tin oxide (ITO) glass substrates were pur-
chased from commercial source. The thickness of the ITO is
80 ± 5 nm, and the sheet resistance is 17–18 Ω per square.
The active area of the device is 2.0 mm×2.0 mm. The ITO
substrates were ultrasonically cleaned by detergent solution,
deionized water, acetone, and ethanol, and dried with N2 gun
before transferred to vacuum chambers. The organic and
metal layers were deposited in different vacuum chambers
with a deposition rate of around 1 and 1–2 Å s−1, respectively
under a base pressure better than 1 × 10−4 Pa, while the de-
position rates of MoO3 and LiF were roughly 0.1–0.2 Å s−1.
All electric testing and optical measurements were per-
1
calcd 269.1, found 270.1 (M+H+). H NMR (400 MHz,
DMSO-d6) δ 8.01–7.90 (m, 4H), 3.70–3.55 (m, 1H), 2.44 (s,
3H), 2.09 (s, 1H), 1.03 (d, J = 6.8 Hz, 6H).
2.3 Synthesis of Tb(CPMIP)3
CPMIPH (1.00 g, 3.72 mmol) and NaOH (0.143 g,
3.58 mmol) were added to10 mL of methanol, and stirred to

3
Zhao et al. Sci China Chem
formed under ambient conditions. The EL spectra, J-V-L, and
EQE characteristics were measured by computer-controlled
Keithley 2400 source meter and absolute EQE measurement
system (C9920-12) with a photonic multichannel analyzer.
3 Results and discussion
CPMIPH could be effectively synthesized from 4-cyano-
phenylhydrazine hydrochloride and ethyl acetoacetate, fol-
lowed by reaction with isobutyryl chloride. Ln(CPMIP)3
could be obtained by CPMIPH and lanthanide chloride hy-
drate salt (LnCl3·6H2O). The synthetic method for CPMIPH
and Ln(CPMIP)3 is shown in Scheme 1.
Scheme 1 Syntheses of CPMIPH and Ln(CPMIP)3 complexes. Reaction
conditions: (a) ethanol, pH 4–5, 70 °C, over night; (b) Ca(OH)2, dry 1,4-
dioxane, reflux, 1 h, under nitrogen; (c) LnCl3·6H2O, NaOH, ethanol/water,
reflux.
Crystals of Ln(CPMIP)3(EtOH)(H2O) were obtained by
slowly evaporating the ethanol/water solution of complexes
at room temperature. Due to the large ion radius, lanthanide
trivalent ions usually require a high coordination number of 8
or 9 to stabilize the coordination environment. Therefore,
additional neutral ligands are needed to satisfy saturated
coordination and 3:1 (monoanionic ligands vs. Ln3+ ions)
electrical neutral complexes when bidentate monoanionic
ligands were used to construct Ln3+ complexes. In this case,
an ethanol molecule and a water molecule coordinate to a
Ln3+ ion as neutral ligands (Figure S1, Supporting In-
formation online). Solvent-free Ln(CPMIP)3 crystals for X-
ray analysis were obtained by sublimation of Ln(CPMIP)3-
(EtOH)(H2O) at 300 °C and 10–4 Pa. The neutral ligands
(EtOH and H2O) would leave during this process. The
structure of Tb(CPMIP)3 was deduced from X-ray diffrac-
tion data. As shown in Figure 1, the terbium ions of the
complex are seven-coordinated by six oxygen atoms from
three bidentate β-diketone ligands and a nitrogen atom from
–CN group, forming a binuclear structure. The details of
bond length and crystal data were exhibited in Tables S1 and
S2 (Supporting Information online).
Figure 1 Oak Ridge Thermal Ellipsoid Plot view of the crystal structure
of binuclear dimer [Tb(CPMIP)3]2 at 30% probability level. All hydrogen
atoms have been omitted for clarity (color online).
As demonstrated in Figure 2, the UV-Vis absorption
spectrum exhibits a strong absorption band corresponding to
the π-π* transition with maxima at 286 nm for Tb(CPMIP)3
(ε = 9.18 × 104 M−1 cm−1, where ε is the molar absorption
coefficient). The bandgap (Eg), which can be calculated from
the onset wavelength of the absorption spectrum, is specu-
lated to be around 3.54 eV (28,600 cm−1). To estimate triplet
energy (ET) value of CPMIP ligand, we obtained the phos-
phorescent emission spectrum of the Gd(CPMIP)3 with
25 ms delay at 77 K. The ET was estimated as 3.00 eV
(24,150 cm−1), which could be calculated from the first
phosphorescent emission peak at 414 nm. The energy gap
between ET (24150 cm−1) and 5D4 state (20500 cm−1) of Tb3+
ion lies at 3650 cm−1, which is benefit for energy transfer
from ligand to Tb3+ ion and inhibiting the reverse energy
transfer process. The room temperature emission spectrum of
Tb(CPMIP)3 exhibits four strong and sharp emission bands
Figure 2 The absorption and emission spectra of Tb(CPMIP)3 at room
temperature, and the phosphorescence spectrum of Gd(CPMIP)3 at 77 K.
The spectra were recorded in ethanol solution at 10–5 M (color online).
corresponding to the 5D4→7FJ (J = 6, 5, 4, 3) transitions with
the main peak at 545 nm. The full width at half maxima

4
Zhao et al. Sci China Chem
(FWHM) of the main peak is only 10 nm, which should be
ascribe to the shielded f-f transition of lanthanide ions.
Tb(CPMIP)3 shows strong emission in both solution and
solid-state. Furthermore, the absolute PL quantum effi-
ciencies of Tb(CPMIP)3 were investigated in different states.
The data were shown in Table 1. Notably, the PLQY of
sublimated crystals could reach 97.1%, while it dramatically
decreased to 32.6% when the crystals were dissolved in
CH3OH. It is widely recognized that the O–H vibration
would quench the emission of lanthanide ions. Hence deut-
erated methanol was introduced to replace CH3OH. The
PLQY raised to 48.6% and the emission showed a longer
lifetime. It should be attributed to the lower frequency of O–
D bond vibration, reducing the quenching effect for lantha-
nide emission. But the quantum yield in deuterated solution
is still only half that of crystals. That implies there is another
non-radiative pathway in the complex molecules. The pho-
tophysical properties are listed in Table 1.
The thermal properties of the complex Tb(CPMIP)3 were
measured by TGA and DSC. The decomposition tempera-
tures (Td) and the glass transition temperature (Tg) of
Tb(CPMIP)3 were observed to be 382 and 156 °C, respec-
tively (Figure S2), which are both higher than that of Tb
(PMIP)3. The enhanced thermal stability could be attributed
to the increase in molecular weight and the binuclear struc-
ture of Tb(CPMIP)3. A high glass transition temperature is
desirable for materials in OLEDs because it prevents mor-
phological changes and suppresses the formation of ag-
gregates upon heating.
Cyclic voltammetry was used to evaluate the highest oc-
cupied molecular orbital (HOMO) energy level of the com-
plex. The test was conducted in dichloromethane solution
using ferrocene (Fc/Fc+) as an internal reference. Two oxi-
dation peaks were detected during positive scanning (Figure
S3). The onset value was adopted for the potential of the
irreversible redox peak according to the reference. The
HOMO energy level was estimated using the method de-
scribed by Forrest et al. [21]. The lowest unoccupied mole-
cular orbital (LUMO) energy level can obtained from the
HOMO value and the Eg which comes from the UV-Vis
absorption spectrum. According to the above methods, the
HOMO and LUMO energy levels of Tb(CPMIP)3 were es-
timated to be −5.7 and −2.2 eV, which could fit most carrier
transport and host materials for OLEDs fabrications.
Green OLEDs were fabricated using the terbium complex
as the dopant of emitting layers. The base device structure of
the OLEDs is ITO/MoO3 (2 nm)/m-bis(N-carbazolyl)ben-
zene (mCP):MoO3 (25 nm, 20 wt%)/mCP (15 nm)/dicarba-
zolylphenylphosphine oxide (DCPPO):Tb(CPMIP)3 (20 nm,
15 wt%)/bis(2-(diphenylphosphino)phenyl)ether oxide
(DPEPO,
10 nm)/1,3,5-tris(N-phenylbenzimidazol-2-yl)
benzene (TPBi, 30 nm)/LiF (1 nm)/Al (100 nm). The fron-
tier molecular orbital energy levels and triplet energy levels
of the materials were shown in Figure 3a. DCPPO was
chosen as the host material of the devices because DCPPO
possesses bipolar carrier mobility and high triplet energy
level up to 3.0 eV [22,23], which is suitable as the host of
Tb(CPMIP)3. The film of DCPPO:Tb(CPMIP)3 (15 wt%)
shows pure emission from Tb3+ ion with 98.9% PLQY, in-
dicating the energy transfer from host to Tb(CPMIP)3 is
highly efficient. mCP (ET = 2.9 eV) and TPBi (ET = 2.6 eV)
serve as the hole and electron transport materials, respec-
tively. The triplet energy level of mCP is slightly lower than
that of the host and dopant, and that of TPBi is inadequate for
these OLEDs. Neither of them can completely confine triplet
excitons in the emitting layer. To solve this problem, a 10 nm
thick DPEPO is introduced considering its high triplet energy
level (3.1 eV) and relative weaker electron-transporting
ability (compared with TPBi). Additionally, a 20 nm MoO3
doped mCP was introduced between the MoO3 and mCP
layers to enhance the hole injection and transportation. These
methods, on the one hand, make the carrier recombination
zone away from the hole transport layer to suppress the
diffusion of triplet excitons from emission layer (EML) to
mCP. On the other hand, the leakage of triplet excitons from
EML to TPBi is inhibited by DPEPO. Moreover, MoO3 is
used as the hole injection layer which modifies the work
function of ITO, and LiF is the electron injection layer. The
organic molecular structures of these used materials are re-
presented in Figure S4.
For further optimization, the thicknesses of EML and
electron transporting layer (ETL) were modified in devices
A, B and C. The parameters and performances of the devices
are listed in Table 2.
All of the devices exhibit typical Tb3+ emission from
Tb(CPMIP)3 complex. As an example, Figure 3b depicts the
electroluminescence spectra of device C at different vol-
tages, indicating the favorable carrier confinement in EML
Table 1 Photophysical, electrochemical and thermal analysis data of Tb(CPMIP)3
PLQY (%)/τobs (μs)
a)
b)
HOMO/LUMO c)
(eV)
Eg (eV)
ET (eV)
Td (°C)
Tg (°C)
d)
In CH3OH
32.6/942
In CD3OD
Crystals
97.1/561
Doped film
98.9/824
48.6/1289,
1837
Tb(CPMIP)3
3.54
3.00
−5.7/−2.2
382
156
a) Calculated from the absorption edge; b) calculated from the 0-0 transition peak of the phosphorescent spectrum; c) HOMO was obtained by the cyclic
voltammetry method, and the LUMO was speculated from the Eg and HOMO energy level; d) doped in DCPPO with 15 wt% concentration.

5
Zhao et al. Sci China Chem
Figure 3 (a) HOMOs, LUMOs and ET levels of materials in fabricated OLEDs; (b) electroluminescence spectra of device C at different voltages; (c) current
density-voltage-luminance (J-V-L) characteristics of the devices; (d) power efficiency-luminance-external quantum efficiency (PE-L-EQE) of the devices
(color online).
Table 2 The thickness parameters and performances of devices
Thickness (nm)
EQE (%)
@100 cd m−2
Devices
Von (V)
Lmax (cd m−2)
λmax (nm) CIE (x, y) @100 cd m−2
EML
ETL
30
max
10.8
15.0
19.7
A
B
C
20
20
40
4.1
4.1
4.7
5.6
7.6
774
971
543
543
543
0.32, 0.62
0.32, 0.62
0.32, 0.62
50
50
10.9
1383
and excellent energy transfer from host to emitter. The cur-
rent density-voltage-luminance (J-V-L) curves are shown in
Figure 3c. The current density and luminance of device A
rise fastest among the three devices, indicating device A
owns smallest resistance. The maximum EQE and luminance
could reach 10.8% and 774 cd m−2, respectively. As the
thickness of TPBi increases to 50 nm (Device B), the max-
imum EQE and luminance increase to 15.0% and
971 cd m−2, respectively, which should be ascribe to the shift
of recombination zone away from the mCP layer to reduce
the triplet excitons quenching. The relative smaller current
density and luminance at the same voltage exhibit larger
resistance of device B than that of device A. Considering the
long lifetime of lanthanide complexes (millisecond level),
the probability of triplet-triplet annihilation and polaron-tri-
plet annihilation is much larger than that of common phos-
phorescent materials. To further improve device efficiency,
the thickness of the emission layer was optimized from 20 to
40 nm in device C. Six devices were fabricated with an
average EQE maxima of 18.6% and a highest one of 19.7%
(Table S3). The high EQE were achieved by extended re-
combination zone for reducing triplet exciton annihilation in
the emission layer. The maximum luminance of device C is
1383 cd m−2, and the EQE at 100 cd m−2 could reach 10.9%,
which is the best performance among OLEDs based on f-f
transition lanthanide complexes.
Compared with current state-of-the-art performance for
green OLEDs with different types of emitter [24–26], our
devices show evident disadvantages such as low maximum
luminance and serious efficiency roll-off due to the intrinsic
millisecond radiation lifetime of f-f transition. The multiple
emission bands of Tb3+ reduce the color purity of the cor-
responding OLEDs. These properties are unfavorable for
commercial application. However, d-f transition lanthanide
complex can circumvent these issues and open a new stage
for future application in OLEDs.

6
Zhao et al. Sci China Chem
ME, Forrest SR. Nature, 1998, 395: 151–154
4 Conclusion
6
7
8
Uoyama H, Goushi K, Shizu K, Nomura H, Adachi C. Nature, 2012,
492: 234–238
In conclusion, nearly 100% PLQY was achieved in a new f-f
transition lanthanide complex Tb(CPMIP)3 by rational li-
gand design. Suitable HOMO and LUMO energy levels and
excellent thermal stabilities endow Tb(CPMIP)3 with huge
potential in electroluminescence. The EQE of corresponding
OLEDs could approach the theoretical limit of 20% by op-
timizing the device structure carefully. Just like phosphor-
escence and TADF materials, the f-f transition lanthanide
complex was demonstrated as an efficient emitter with 100%
exciton utilization efficiency.
Ai X, Evans EW, Dong S, Gillett AJ, Guo H, Chen Y, Hele TJH,
Friend RH, Li F. Nature, 2018, 563: 536–540
Wang L, Zhao Z, Zhan G, Fang H, Yang H, Huang T, Zhang Y, Jiang
N, Duan L, Liu Z, Bian Z, Lu Z, Huang C. Light Sci Appl, 2020, 9:
157
9
Li J, Wang L, Zhao Z, Sun B, Zhan G, Liu H, Bian Z, Liu Z. Nat
Commun, 2020, 11: 5218
10 Zhan G, Wang L, Zhao Z, Fang P, Bian Z, Liu Z. Angew Chem Int Ed,
2020, 59: 19011–19015
11 Zhao Z, Wang L, Zhan G, Liu Z, Bian Z, Huang C. Natl Sci Rev, 2021,
8: nwaa193
12 Kondo Y, Yoshiura K, Kitera S, Nishi H, Oda S, Gotoh H, Sasada Y,
Yanai M, Hatakeyama T. Nat Photonics, 2019, 13: 678–682
13 Shahalizad A, Malinge A, Hu L, Laflamme G, Haeberlé L, Myers
DM, Mao J, Skene WG, Kéna-Cohen S. Adv Funct Mater, 2021, 31:
2007119
Acknowledgements
This work was supported by the National Basic
Research Program of China (2017YFA0205100), and the Key Project of
Science and Technology Plan of Beijing Education Commission
(KZ201910028038). Zifeng Zhao acknowledges financial support from the
China Postdoctoral Science Foundation (2018M641065, 2021T140009) and
BMS-Junior Fellow.
14 Wei H, Zhao Z, Wei C, Yu G, Liu Z, Zhang B, Bian J, Bian Z, Huang
C. Adv Funct Mater, 2016, 26: 2085–2096
15 Bünzli JCG, Piguet C. Chem Soc Rev, 2005, 34: 1048–1077
16 Kido J, Nagai K, Ohashi Y. Chem Lett, 1990, 19: 657–660
17 Kido J, Nagai K, Okamoto Y, Skotheim T. Chem Lett, 1991, 20:
1267–1270
Conflict of interest The authors declare no conflict of interest.
18 Wang L, Zhao Z, Wei C, Wei H, Liu Z, Bian Z, Huang C. Adv Opt
Mater, 2019, 7: 1801256
19 Yu G, Ding F, Wei H, Zhao Z, Liu Z, Bian Z, Xiao L, Huang C. J
Mater Chem C, 2016, 4: 121–125
20 Wei C, Sun B, Cai Z, Zhao Z, Tan Y, Yan W, Wei H, Liu Z, Bian Z,
Huang C. Inorg Chem, 2018, 57: 7512–7515
21 Dandrade B, Datta S, Forrest S, Djurovich P, Polikarpov E, Thompson
M. Org Electron, 2005, 6: 11–20
Supporting information The supporting information is available online at
http://chem.scichina.com and http://link.springer.com/journal/11426. The
supporting materials are published as submitted, without typesetting or
editing. The responsibility for scientific accuracy and content remains en-
tirely with the authors.
22 Zhao Z, Yu G, Chang Q, Liu X, Liu Y, Wang L, Liu Z, Bian Z, Liu W,
Huang C. J Mater Chem C, 2017, 5: 7344–7351
23 Tao Y, Xiao J, Zheng C, Zhang Z, Yan M, Chen R, Zhou X, Li H, An
Z, Wang Z, Xu H, Huang W. Angew Chem, 2013, 125: 10685–10689
24 Ma X, Liang J, Bai F, Ye K, Xu J, Zhu D, Bryce MR. Eur J Inorg
Chem, 2018, 2018(42): 4614–4621
25 Im Y, Lee JY. J Inf Display, 2017, 18: 101–117
26 Lee DR, Kim BS, Lee CW, Im Y, Yook KS, Hwang SH, Lee JY. ACS
Appl Mater Interfaces, 2015, 7: 9625–9629
1
2
3
4
5
Zhang Q, Li B, Huang S, Nomura H, Tanaka H, Adachi C. Nat
Photon, 2014, 8: 326–332
Tao Y, Yuan K, Chen T, Xu P, Li H, Chen R, Zheng C, Zhang L,
Huang W. Adv Mater, 2014, 26: 7931–7958
Li X, Zhang J, Zhao Z, Wang L, Yang H, Chang Q, Jiang N, Liu Z,
Bian Z, Liu W, Lu Z, Huang C. Adv Mater, 2018, 30: 1705005
Hatakeyama T, Shiren K, Nakajima K, Nomura S, Nakatsuka S, Ki-
noshita K, Ni J, Ono Y, Ikuta T. Adv Mater, 2016, 28: 2777–2781
Baldo MA, O'Brien DF, You Y, Shoustikov A, Sibley S, Thompson