Efficient Nondoped Blue Fluorescent Organic Light-Emitting Diodes (OLEDs) with a High External Quantum Efficiency of 9.4% @ 1000 cd m?2 Based on Phenanthroimidazole?Anthracene Derivative

Article abstract of DOI:10.1002/adfm.201705813

Organic light-emitting diodes (OLEDs) can promise flexible, light weight, energy conservation, and many other advantages for next-generation display and lighting applications. However, achieving efficient blue electroluminescence still remains a challenge

Full text of DOI:10.1002/adfm.201705813

FULL PAPER
OLEDs
www.afm-journal.de
Efficient Nondoped Blue Fluorescent Organic Light-
Emitting Diodes (OLEDs) with a High External
Quantum Efficiency of 9.4% @ 1000 cd m−2 Based
on Phenanthroimidazole−Anthracene Derivative
Xiangyang Tang, Qing Bai, Tong Shan, Jinyu Li, Yu Gao, Futong Liu, Hui Liu,
Qiming Peng, Bing Yang, Feng Li, and Ping Lu*
blue) primary colors but also act as host
for red and green dopants in full color dis-
play and white lightings.[2] In many cases,
the high efficiency is acquired by doping
strategy which demands appropriate host
material and fine-tuning doping concen-
tration.[3] This will complicate device fab-
rication and increase commercial cost.
In addition, reducing efficiency roll-off
is another crucial issue toward practical
application because high luminescence is
required for display and lightings. Hence,
it is of great significance to develop blue-
emitting materials capable of achieving
high efficiency at high luminescence for
nondoped OLEDs. Though phosphores-
cent materials, generally Ir- or Pt- based
complexes, can harvest all triplet excitons,
they should be dispersed into host mate-
rials with high triplet energy to reduce
the quenching effect of the long-lived tri-
plet excitons at high luminescence and
avoid back energy transfer from dopant
to host.[4] Furthermore, the scarcity of
Organic light-emitting diodes (OLEDs) can promise flexible, light weight,
energy conservation, and many other advantages for next-generation display
and lighting applications. However, achieving efficient blue electrolumines-
cence still remains a challenge. Though both phosphorescent and thermally
activated delayed fluorescence materials can realize high-efficiency via effec-
tive triplet utilization, they need to be doped into appropriate host materials
and often suffer from certain degree of efficiency roll-off. Therefore, devel-
oping efficient blue-emitting materials suitable for nondoped device with little
efficiency roll-off is of great significance in terms of practical applications.
Herein, a phenanthroimidazole−anthracene blue-emitting material is reported
that can attain high efficiency at high luminescence in nondoped OLEDs. The
maximum external quantum efficiency (EQE) of nondoped device is 9.44%
which is acquired at the luminescence of 1000 cd m−2. The EQE is still as
high as 8.09% even the luminescence reaches 10 000 cd m−2. The maximum
luminescence is ≈57 000 cd m−2. The electroluminescence (EL) spectrum
shows an emission peak of 470 nm and the Commission International de
L’Eclairage (CIE) coordinates is (0.14, 0.19) at the voltage of 7 V. To the best of
the knowledge, this is among the best results of nondoped blue EL devices.
these noble metal elements is not economical for commer-
cial production. Thermally activated delayed fluorescence
(TADF) materials can also utilize triplet excitons and obtain
nearly 100% internal quantum efficiency.[5] However, the slow
reverse intersystem crossing process from T1 (the lowest tri-
plet excited state) to S1 (the lowest singlet excited state) ren-
ders accumulation of triplet excitons at high luminescence. As
a result, like phosphorescent materials, most TADF materials
are also needed to be doped into proper host with high triplet
energy and suffer from a certain degree of efficiency roll-off.[6]
Triplet–triplet annihilation (TTA) can also effectively harness
triplet excitons in the form of generating one singlet exciton by
fusing two triplets.[7] In principle, TTA materials are applicable
for efficient nondoped OLEDs with reduced efficiency roll-off.
However, many TTA molecules still remain to be doped into
appropriate host materials to attain high efficiency.[8] There-
fore, developing efficient blue-emitting fluorescent materials
suitable for nondoped device which can achieve high efficiency
at high luminescence is a pressing issue toward practical appli-
cations, such as display and lightings.
1. Introduction
New-generation organic light-emitting diodes (OLEDs) mate-
rials with aspect of low cost like pure organic aromatic fluo-
rescent compounds while remaining high efficiency as iridium
(Ir) phosphorescent materials are being developed.[1] Especially,
efficient blue fluorescent materials are highly desired because
they not only can provide one of the RGB (red, green, and
Dr. X. Y. Tang, Dr. Q. Bai, Dr. T. Shan, Dr. J. Li,
Dr. Y. Gao, Dr. F. Liu, Dr. H. Liu, Dr. Q. Peng,[+]
Prof. B. Yang, Prof. F. Li, Prof. P. Lu
State Key Lab of Supramolecular Structure and Materials
Jilin University
2699 Qianjin Avenue, Changchun 130012, P. R. China
E-mail: lup@jlu.edu.cn
[+]Present address: Institute of Advanced Materials (IAM), Nanjing Tech
University (Nanjing Tech), 30 South Puzhu Road, Nanjing 211816, China
DOI: 10.1002/adfm.201705813
©
Adv. Funct. Mater. 2018, 1705813
1705813 (1 of 8)
2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.afm-journal.de
In this work, we report a highly efficient blue-emitting
material PIAnCN which is constructed by connecting phen-
anthroimidazole (PI) with cyano substituted anthracene.
PI has high photoluminescence quantum yield (PLQY)
because of its rigid planar π conjugation.[9] In addition, the
amine N atom in the imidazole ring together with the con-
jugation plane of whole molecule endows PI weak electron
donating ability and proper highest occupied molecular
orbital (HOMO) energy level to reduce hole injection barrier
in device.[10] Moreover, PI emits in the near ultraviolet region
because of its limited conjugation.[9] All these features make
PI a promising building block for blue-emitting molecules.
On the other hand, anthracene is also a highly emissive blue-
emitting moiety which is often adopted as functional group
to construct TTA materials.[7,11] By combining phenanthro-
imidazole with anthracene, it is anticipated that the obtained
molecule can possess efficient blue emitting as well as effec-
tive triplet fusion to utilize the nonemissive triplet excitons.
Cyano group is incorporated into the molecular structure. The
electron withdrawing property of cyano can lower the lowest
unoccupied molecular orbital (LUMO) energy level and
reduce electron injection barrier in device. Furthermore, the
presence of cyano group can enhance intermolecular interac-
tions.[12] For example, the existence of cyano group can pro-
mote tight intermolecular packing through enhanced supra-
molecular interactions in cyano substituent distyrylbenzene
(CN-DSB) derivatives.[12a] Considering TTA occurs between
two individual molecules, we assume that the introduction of
cyano group may facilitate TTA process and improve device
efficiency. We fabricated the doped device of PIAnCN which
shows deep blue emission with an electroluminescence (EL)
peak of 444 nm and CIE coordinates of (0.15, 0.07). Though
the maximum external quantum efficiency (EQE) of doped
device can achieve 6.77%, the efficiency roll-off is serious.
On the contrary, the nondoped device exhibits little efficiency
roll-off. The maximum EQE is 9.44% (13.16 cd A−1) which
is achieved at high luminescence of 1000 cd m−2. The EQE
can still remain as high as 8.09% (11.29 cd A−1) even when
the luminescence is increased to 10 000 cd m−2. The EL peak
locates at 470 nm with a CIE coordinates of (0.14, 0.19). The
maximum luminescence is ≈57 000 cd m−2. We assume that
TTA is the main reason for the high performance of non-
doped device. To the best of our knowledge, this is among the
best results of nondoped blue-emitting fluorescent OLEDs.[13]
2. Results and Discussion
2.1. Synthesis and Characterization
The synthetic route of PIAnCN is shown in Scheme 1. The
PPIBr is attained by one-pot reaction in high yield according
to our previous report.[9] Then PPIBr is treated with n-BuLi
and isopropoxyboronic acid pinacol ester at low temperature
to obtain 1-phenyl-2-(4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-
2-yl)phenyl)-1H-phenanthro[9,10-d]imidazole (PPIB). BrAnCN
is synthesized via Suzuki coupling reaction using 9,10-dibro-
moanthracene and 4-cyanophenylboronic acid as reactants.
Finally, PPIB is coupled with BrAnCN via Suzuki reaction to
afford the target compound PIAnCN. The purity and chemical
structure of PIAnCN are well characterized by NMR, mass
spectrum (MS), and elemental analysis. The synthetic details
are provided in the Supporting Information.
2.2. Thermal Properties
Thermal properties are examined by thermal gravimetric analysis
and differential scanning calorimetry (DSC). As displayed in
Figure S1a (Supporting Information), and Table 1 the decom-
position temperature (Td, corresponding to 5% weight loss) is
483 °C, indicating that PIAnCN can endure high temperature
and would not go through decomposition during device fabrica-
tion and operation. The DSC of the as-synthesized amorphous
sample is shown in Figure S1b (Supporting Information). Upon
heating procedure, no obvious exo- or endothermic signal can be
seen until 344 °C which can be ascribed to the melting point (Tm).
No glass transition (T ) temperature can be observed demon-
g
strating that PIAnCN can keep amorphous morphology in a wide
range of temperature which is beneficial for OLEDs application.
2.3. Photophysical Properties
To investigate the lowest singlet (S1) excited state of PIAnCN,
photoluminescence (PL) is measured in various solvents
with different polarities. As displayed in Figure 1, the solva-
tochromic effect is negligible as the PL does not change much
upon increasing solvent polarity. The emission peak shifts
from 434 nm in nonpolar slovent n-hexane to 446 nm in high
Scheme 1. Synthesis route of PIAnCN: a) CH3COONH4, acetic acid, 120 °C, N2 protection, overnight; b) n-BuLi, −78 °C, stirring for 3 h under N2
atmosphere, then isopropoxyboronic acid pinacol ester is added afterward. The mixture is stirred overnight at room temperature under N2 protection;
c) Suzuki coupling: Pd(PPh3)4, K2CO3 (2 m aq), toluene, 90 °C, 24 h under N2 protection.
©
Adv. Funct. Mater. 2018, 1705813
1705813 (2 of 8)
2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.afm-journal.de
Table 1. Photophysical properties of PIAnCN.
a)
c)
n)
b)
d)
Compound
PIAnCN
Td/Tm
[°C]
Eg
PLQYh)
(%)
HOMO/LUMOm) [eV]
ES1/ET1
[eV]
λmax, abs
[nm]
λmax, PL
[nm]
[eV]
483/344
394/374/360/315/260
2.97
464e)/444f)/442g)
50i)/80j)/82k)/62l)
2.86/1.74
−5.55/−2.78
a)Td: decomposition temperature; Tm: melting point; b)λmax, abs: absorption maximum in dilute THF (10−5); c)Eg: Optical gap calculated from the absorption onset in dilute
THF (10−5 m); d)λmax, PL: emission peak; e)Emission peak of nondoped film; f)Emission peak of doped film (15% in polymethyl methacrylate (PMMA); g)Emission peak in
dilute THF (10−5); h)PLQY: photoluminescence quantum yield measured by integrating sphere; i)PLQY of nondoped film; j)PLQY of doped film (15% in PMMA); k)PLQY of
doped film (15% in CzSi); l)PLQY in dilute THF (10−5); m)HOMO/LUMO energy levels estimated from cyclic voltammetry measurement; n)ES1/ET1: energy levels of S1 and
T1 states. ES1 is calculated from emission maximum in n-hexane and ET1 is calculated from the first vibronic peak of its phosphorescence at 77 K in THF.
polar solvent acetonitrile with a 12 nm variation, indicating the
localized excited property of S1 state. To clarify the origin of the
PL of PIAnCN, we further measured the PL of 1,2-diphenyl-
phenanthroimidazole (PPI) and 4-(10-phenylanthracen-
9-yl)benzonitrile (AnCN) in different solvents, respectively.
The synthesis of PPI and AnCN is displayed in Scheme S1
(Supporting Information). The results show that the PL of
PIAnCN is similar with that of AnCN while totally different
from that of PPI as can be seen in Figures S3 and S4 (Sup-
porting Information). Therefore, the S1 state of PIAnCN may
mostly localize on the AnCN part. The absorption of PIAnCN
in different solvents is nearly the same. The well-resolved
360, 374, and 394 nm absorption bands are characteristics
of the π–π* transition of anthracene moiety.[14] Compared
with the absorption of AnCN (Figure S4a, Supporting Infor-
mation), the newly emerged absorption band of PIAnCN
centered at 315 nm (Figure 1) can be assigned to the π–π*
transition of PPI moiety. The absorption at 260 nm arises
from the π–π* transition of the phenyl ring anchored at the
N atom in the imidazole ring.[15] The molar absorption coef-
ficient (εmax) at the longest absorption maximum (394 nm) is
determined to be about 2.25 × 104 L mol−1 cm−1 according to
Figure S6 and Equation S1 (Supporting Information). Then
the radiative decay constant (kr) of S1 state is estimated to be
2.25 × 108 s−1 according to Equation S2 (supporting informa-
tion). As displayed in Figure 2, the PIAnCN in tetrahydro-
furan (THF) shows a monoexponential decay with a lifetime
(τs) of 2.31 ns. No delayed component can be observed, indi-
cating the emission exclusively originates from prompt decay
of S1 state. The PLQY of PIAnCN in THF is measured to be
62%, in combination with τs, the radiative decay constant kr
can thus be calculated to be 2.68 × 108 s−1 according to Equa-
tion S3 listed in Supporting Information. This value is in
good agreement with the kr derived from εmax, demonstrating
the validity of our experimental data.
As shown in Figure S7 (Supporting Information), we are
not successful in obtaining the phosphorescence of PIAnCN
directly by normal phosphorescence measurement mode of PL
spectrometer in glassy THF at 77 K probably due to the inef-
ficiency of intersystem crossing (ISC) of S1→T1 as well as the
strong vibrational coupling between T1 and S0 as a result of
the narrow band gap of T1–S0 (energy gap law).[16] The key to
acquire phosphorescence of PIAnCN is forming large amounts
of T1 excitons of PIAnCN. PtOEP is an archetype of phospho-
rescent material featuring fast and efficient ISC from S1→T1.[17]
Therefore, the phosphorescence of PIAnCN is recorded at 77 K
in glassy THF using PtOEP as sensitizer. The S1 and T1 energy
of PtOEP is obtained according to literature.[18] As displayed in
Figure 3, excited by 550 nm ps laser source, PtOEP is pumped
to excited state, followed by fast triplet energy transfer to the T1
state of PIAnCN. Delayed by 10 ms, strong phosphorescence of
PIAnCN can be detected (Figure 3 and Figure S7, Supporting
Information). The first vibrational peak of phosphorescence is
713 nm, corresponding to a T1 energy of 1.74 eV. The phos-
phorescence originates from the anthracene moiety and is in
good accordance with literature.[19] Based on the PL peak of
PIAnCN in n-hexane (434 nm), the S1 energy is estimated to
be 2.86 eV. Obviously, TADF cannot occur considering the large
S1–T1 energy gap. However, TTA is energy favorable because
Figure 1. Absorption spectrum of PIAnCN in THF (10−5 m) and emission
spectra of PIAnCN in different solvents (10−5 m).
Figure 2. PL decay of PIAnCN in THF (10−5 m) at ambient condition.
©
Adv. Funct. Mater. 2018, 1705813
1705813 (3 of 8)
2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.afm-journal.de
Figure 3. Phosphorescence measurement diagram of PIAnCN by using PtOEP as sensitizer (10−4 m PIAnCN + 10−4 m PtOEP in THF, 77 K). The T1 energy
of PIAnCN is calculated from the first vibrational peak of its phosphorescence spectrum. The T1 and S1 energy of PtOEP is obtained from literature.[13]
2T1 > S1. Therefore, we assume that TTA may exert positive
Figure 4 and Figure S8 (Supporting Information), the T1 of
PIAnCN mainly distributes on the AnCN moiety and the cor-
responding energy is calculated to be 1.58 eV, and is close to
the value in the literature as well as our experimental value
1.74 eV. The good agreement between calculated values and
experimental data demonstrates that the theoretical method
we adopted is valid for the calculation of lowest excited states,
i.e., the S1 and T1 states. The calculated T2 state mainly locates
on PI moiety and the corresponding energy is 2.80 eV. The
calculated results of more high lying excited states are summa-
rized in Figure S8 (Supporting Information).
effect in boosting device efficiency.
2.4. Theoretical Calculations
To further examine the excited states properties of PIAnCN,
natural transition orbitals (NTOs) are performed using
TD-M062x/6-31G(d, p) method based on the optimized S1
geometry.[20] As shown in Figure 4, the NTOs of S1 state mainly
localize on the AnCN part, which is in accordance with the
result of solvatochromism measurement discussed above.
The calculated S1 state energy is 2.81 eV which agrees with the
emission energy of PIAnCN in n-hexane (434 nm, 2.86 eV).
According to literatures, the T1 energies of anthracene deriva-
tives are as low as ≈1.8 eV arising from the long conjugation
axis of anthracene.[21] As revealed by the NTOs diagram in
2.4. Electrochemical Properties
The HOMO/LUMO energy levels are derived from cyclic vol-
tammetry measurement (Figure S9, Supporting Information).
The oxidation and reduction onset against ferrocenium/ferro-
cene (Fc+/Fc) redox couple are 0.75 and −2.02 V, respectively.
Thus, the HOMO and LUMO energy levels are calculated to
be −5.55 and −2.78 eV according to Equations S4 and S5 (Sup-
porting Information). This HOMO/LUMO energy alignment is
beneficial for balanced charge injection and can avoid charge
accumulation at the interface between emission layer and
adjacent charge transporting layers of OLEDs device. These
advantages are in favor of improving device performance.[22]
2.5. Electroluminescence Properties
To examine the potential application of PIAnCN as solid state
emitter, we fabricated the nondoped OLEDs with the device
structure of indium tin oxide (ITO)/HATCN (6 nm)/TAPC
(25 nm)/TCTA (15 nm)/PIAnCN (20 nm)/TPBI (40 nm)/LiF
(0.5 nm)/Al (120 nm), where ITO (indium tin oxide) is the
anode, hexaazatriphenylenehexacabonitrile (HATCN) is the
hole injecting layer, TAPC (di-(4-(N,N-ditolyl-amino)-phenyl)
cyclohexan) is the hole transporting layer (HTL), TCTA (tris(4-
carbazoyl-9-ylphenyl)amine) is the buffer layer, PIAnCN is the
Figure 4. NTOs of PIAnCN on S1 geometry, S0: singlet ground state; S1: the
lowest singlet excited state; T1: the lowest triplet excited state; T2: the second
triplet excited state. The percentage indicates the possibility of the transition.
©
Adv. Funct. Mater. 2018, 1705813
1705813 (4 of 8)
2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.afm-journal.de
Supporting Information). The maximum luminescence is
≈16 000 cd A−1. The doped device suffers from serious effi-
ciency roll-off: the maximum EQE is 6.77% which is obtained
at initial luminescence (0.77 cd m−2); the EQE decreased to
5.97% at 100 cd m−2 and 4.77% at 1000 cd m−2 and 3.21%
at 10 000 cd m−2. By tuning the thickness of hole trans-
porting layers and selecting proper host materials, high effi-
ciency as well as deep blue emission can be obtained for the
doped device. We assume that cavity effect may be account-
able for this.[23] Besides, compared with the PL of 4,4′-Bis(N-
carbazolyl)-1,1′-biphenyl (CBP), the PL of CzSi exhibits
larger overlap with the absorption of PIAnCN (Figure S14,
Supporting Information), demonstrating more efficient
Förster energy transfer from CzSi to PIAnCN which can also
lead to the high efficiency. We notice that the nondoped and
doped devices display opposite trend in terms of EQE curves
(Figure 5). The EQE of doped device suffers from continuous
efficiency roll-off upon the device is turned on, while the
EQE of nondoped device goes through uphill process before
≈1000 cd m−2 then slight roll-off afterward. This implies that
the singlet excitons are severely quenched with increasing
current density for doped device. However, for nondoped
device, TTA process becomes more effective at higher cur-
rent density which can explain the roll-up trend of EQE curve
before ≈1000 cd m−2. With higher current density, triplet
quenching processes such as single-triplet quenching or triplet-
charge quenching may be dominant over TTA process, leading
to efficiency roll-off. But the TTA process can convert triplet
excitons into singlet ones to compensate the efficiency loss at
high current density, resulting in reduced efficiency roll-off at
high current density for nondoped device.
We measured the transient EL decay of both nondoped
and doped devices at different driving voltages. As displayed
in Figure S15 (Supporting Information), upon a fast electrical
excitation pulse, the transient EL decay of doped device shows
both a prompt EL decay at the timescale of submicroseconds
and a delayed EL response at the timescale of several tens of
microseconds. The prompt decay component can be assigned
to the fast fluorescence of PIAnCN. The delayed EL decay
shows little difference at different driving voltages which may
be assigned to the recombination of trapped charges since we
cannot give a short-time period reverse bias to eliminate these
trapped charges under our experimental condition. Likewise, as
shown in Figure 6, the transient EL decay of nondoped device
also consists of two components, a rapid EL decay which origi-
nates from the prompt fluorescence and a delayed EL response
on the order of several tens of microseconds. For the nondoped
Figure 5. Current efficiency (circle) and external quantum efficiency
(triangle) of nondoped and doped devices.
emitting layer (EML), TPBI (1,3,5-tris-(N-phenylbenzimidazol-
2-yl)benzene) is the electron transporting layer, LiF is the elec-
tron injecting layer and Al is the cathode. The turn-on voltage
is only 3.0 V indicating effective charge injection and recom-
bination in the EML. The EL spectra can coincide with the
nondoped PL spectrum and are very stable at various driving
voltages, demonstrating the EL is from the EML. The EL shows
an emission peak of ≈470 nm and CIE coordinates of (0.14,
0.19) at 7 V. The maximum power efficiency is 10.44 lm W−1
(Figure S11, Supporting Information). As displayed in Figure 5,
the maximum EQE is 9.44% which is obtained at high lumines-
cence of 1000 cd m−2. The efficiency roll-off is quite small. Even
at a luminescence of 10 000 cd m−2, the EQE can still remain
8.09%. To the best of our knowledge, this is among the best
results of nondoped fluorescent blue OLEDs.[13] A comparison
between our result and other nondoped blue-emitting OLEDs is
presented in Table S1 (Supporting Information). The high effi-
ciency at high luminescence makes PIAnCN a promising can-
didate for practical application.
We also optimized the doped devices by tuning the thick-
ness of hole transporting layers, changing the host materials,
and varying doping concentration. Detailed description of
device optimization is illustrated in Supporting Information.
The best result is obtained from the device of ITO/HATCN
(6 nm)/TAPC (25 nm)/TCTA (15 nm)/CzSi: PIAnCN 15%
(20 nm)/TPBI (40 nm)/LiF (0.5 nm)/Al (120 nm) where CzSi
(9-(4-tert-Butylphenyl)-3,6-bis(triphenylsilyl)-
9H-carbazole) is the host material. The
doped device shows deep blue emission
with EL peak of 444 nm and CIE coordinates
of (0.15, 0.07) at 7 V as shown in Figure
S13e (Supporting Information) and Table 2
or Table S2 (Supporting Information).
The turn-on voltage is increased to 3.6 V in
comparison with that of nondoped device.
The EL spectra are very stable at various
voltages. The maximum power efficiency
is 5.34 lm W−1 (Figure S12b and Table S2,
Table 2. EL performance of PIAnCN-based devices.
EQEd) [%]
c)
Device
Von
Lmax
CEmax
Max/1000/10
000
EL
[λmax
CIEf)
[nm] [x, y]
a)
b)
e)
[V]
3.0
3.6
[cd m−2
57787
16107
]
[cd A−1
13.16
5.95
]
]
Nondoped
Doped
9.44/9.44/8.09
6.77/4.77/3.21
470
444
0.14, 0.19
0.15, 0.07
a)
b)
V
on
: turn-on voltage at the luminescence of 1 cd m−2
;
L
max
: maximum luminescence; c)CEmax: maximum
current efficiency; d)EQEmax/1000/10 000: EQE of maximum/at 1000 cd m−2/10 000 cd m−2
sion peak of EL spectrum at 7 V; f)CIE coordinates at 7 V.
;
e)EL λmax: emis-
©
Adv. Funct. Mater. 2018, 1705813
1705813 (5 of 8)
2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.afm-journal.de
Figure 6. a) Transient EL decay of nondoped device at different voltages; b) comparison of transient EL decay between nondoped and doped devices
at the same current density; and c) amplified EL decay of the delayed component of PIAnCN nondoped device.
device, the ratio of delayed component gradually decreases
from 5 to 10 V. We assume the delayed response is not simply
caused by the recombination of trapped charges since higher
driving voltage would result in more trapped charges and
therefore an increased ratio of the delayed portion. The experi-
mental result in Figure 6a is exactly the other way around. On
the other hand, by comparing the transient EL decay profile
of both doped and nondoped devices at the same current den-
sity (Figure 6b), the ratio of delayed component of nondoped
device is obviously larger than that of doped device. This may
also indicate that the delayed component of nondoped device
is not solely from the recombination of trapped charges. Con-
sidering the large S1–T1 energy gap, TADF is not the reason for
the delayed component. To further exclude the TADF mecha-
nism, we also measured the transient EL decay of a TADF
TTA is very effective, resulting in a large proportion of delayed
component which can contribute to the enhanced device effi-
ciency. By increasing driving voltages, the triplet excitons may
be quenched by charges or go through other quenching pro-
cesses which would lead to the decrease of delayed component
and correspondingly a decrease of the EQE. The transient EL
response is inconsistent with the driving voltage-dependent
EQE curves as shown in Figure S10d (Supporting Informa-
tion). The EQE achieves its maximum at ≈5 V then slowly
decreases at higher voltages. To supply more straightforward
evidence of TTA, we measured the EL decay at 20 V under a
short time period pulse. The 20 V bias is to guarantee enough
triplet excitons, while the short time period pulse is to avoid the
device being burned at such high voltages. We amplified the
delayed component and take the logarithm of EL intensity as
Y and logarithm of time as X as shown in Figure 6c. According
to Monkman and co-workers,[24] if TTA is predominant in the
device, the delayed EL intensity should be proportional to t−2 (t
is the time). Therefore, by taking the logarithm of EL intensity
as Y and logarithm of time as X, the slope of obtained curve
should be −2 for TTA mechanism. Our result is consistent with
it as shown in Figure 6c which strongly supports TTA process
to be valid in PIAnCN nondoped device.
emitter
10,10′-(sulfonylbis(4,1-phenylene))bis(9,9-dimethyl-
9,10-dihydroacridine) (DMAC-DPS).[5a] As shown in Figure S16
(Supporting Information), the DMAC-DPS shows much more
prominent delayed component and much slower EL decay than
PIAnCN at the same current density. This demonstrates that
the delayed component of PIAnCN decays much faster than
that of the TADF emitter DMAC-DPS. Therefore, we can con-
clude that TADF does not account for the delayed component
of PIAnCN nondoped device. We assume that the delayed EL
results from TTA process. At the driving voltage of 5 V, the
Our previous investigation of CN-DSB derivatives discloses

that the CN group plays an important role in keeping tight
©
Adv. Funct. Mater. 2018, 1705813
1705813 (6 of 8)
2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.afm-journal.de
financial support from the National Basic Research Program of China
(Grant No. 2016YFB0401001), National Science Foundation of China
(Grant Nos. 21374038 and 21774047), and Jilin Provincial Science and
Technology Department (Grant No. 20160101302JC).
intermolecular packing through enhanced supramolecular
interactions.[12a] We infer that the introduction of CN group may

also enhance intermolecular interactions in PIAnCN neat film and
accordingly facilitates TTA process since TTA occurs between two

individual molecules. To clarify the functionality of CN group,

we synthesized the molecule without CN group PPIAn. The
Conflict of Interest
molecular structure and synthetic details are presented in Scheme
S1 (Supporting Information). We also fabricated the nondoped
device of PPIAn and measured the transient EL decay. The device
structure is ITO/HATCN (6 nm)/TAPC (25 nm)/TCTA (15 nm)/
PIAnCN (20 nm)/TPBI (40 nm)/LiF (0.5 nm)/Al (120 nm). As
displayed in Figure S17c and Table S3 (Supporting Information),
the maximum EQE is 4.20% which is inferior to that of PIAnCN
The authors declare no conflict of interest.
Keywords
blue electroluminescence, high quantum yield, low efficiency roll-off,
nondoped devices, organic light-emitting diodes (OLEDs)

nondoped device, implying that the CN group can improve
device efficiency to a great extent. As can be seen in Figure S18
(Supporting Information), the transient EL decay of PIAnCN non-
doped device shows larger delayed component than that of PPIAn
Received: October 7, 2017
Revised: November 20, 2017
Published online:

nondoped device, indicating that the CN containing molecule

PIAnCN possesses stronger TTA effect than PPIAn without CN

group. This evidences that the existence of CN group can indeed
[1] a) H. Uoyama, K. Goushi, K. Shizu, H. Nomura, C. Adachi,
Nature 2012, 492, 234; b) S. Reineke, Nat. Photonics 2014, 8, 269;
c) Q. Peng, A. Obolda, M. Zhang, F. Li, Angew. Chem., Int. Ed. 2015,
54, 7091; Angew. Chem. 2015, 127, 7197; d) H. Namai, H. Ikeda,
Y. Hoshi, N. Kato, Y. Morishita, K. Mizuno, J. Am. Chem. Soc. 2007,
129, 9032; e) L. Yao, B. Yang, Y. Ma, Sci. China: Chem. 2014, 57, 335.
[2] a) S. Krotkus, D. Kasemann, S. Lenk, K. Leo, S. Reineke, Light: Sci.
Appl. 2016, 5, e16121; b) Y. Sun, N. C. Giebink, H. Kanno, B. Ma,
M. E. Thompson, S. R. Forrest, Nature 2006, 440, 908; c) D. Zhang,
L. Duan, Y. Zhang, M. Cai, D. Zhang, Y. Qiu, Light: Sci. Appl. 2015,
4, e232; d) B. Q. Liu, L. Wang, D. Y. Gao, J. H. Zou, H. L. Ning,
J. B. Peng, Y. Cao, Light: Sci. Appl. 2016, 5, e16137; e) M. Zhu,
C. Yang, Chem. Soc. Rev. 2013, 42, 4963.
promote TTA process. In addition, balanced charge recombina-
tion is crucial for TTA devices since the triplet excitons are sub-
ject to quenching by redundant charges which will lead to severe
efficiency roll-off.[8c,22] The electron withdrawing ability of CN

group can lower electron injection barrier and is beneficial for
balanced charge recombination. As shown in Figures S19 and S9
(Supporting Information), the PIAnCN and PPIAn display similar
HOMO energy but different LUMO energy. The LUMO energy of
PIAnCN is ≈0.18 eV lower than that of PPIAn. Hence, the charge
recombination may be more balanced in the case of PIAnCN non-
doped device, leading to higher device efficiency.
[3] a) J. Shi, C. W. Tang, Appl. Phys. Lett. 1997, 70, 1665; b) C. W. Tang,
S. A. VanSlyke, C. H. Chen, J. Appl. Phys. 1989, 65, 3610;
c) J. Zhang, D. Ding, Y. Wei, F. Han, H. Xu, W. Huang, Adv. Mater.
2016, 28, 479; d) D. Zhang, J. Qiao, D. Zhang, L. Duan, Adv. Mater.
2017, 29, 1702847; e) M. A. Baldo, S. Lamansky, P. E. Burrows,
M. E. Thompson, S. R. Forrest, Appl. Phys. Lett. 1999, 75, 4;
f) H. Liu, G. Cheng, D. Hu, F. Shen, Y. Lv, G. Sun, B. Yang, P. Lu,
Y. Ma, Adv. Funct. Mater. 2012, 22, 2830.
[4] a) C. Adachi, R. C. Kwong, P. Djurovich, V. Adamovich,
M. A. Baldo, M. E. Thompson, S. R. Forrest, Appl. Phys. Lett. 2001,
79, 2082; b) R. J. Holmes, S. R. Forrest, Y. J. Tung, R. C. Kwong,
J. J. Brown, S. Garon, M. E. Thompson, Appl. Phys. Lett. 2003, 82,
2422; c) Y. Tao, C. Yang, J. Qin, Chem. Soc. Rev. 2011, 40, 2943.
d) M. A. Baldo, D. F. O’Brien, Y. You, A. Shoustikov, S. Sibley,
M. E. Thompson, S. R. Forrest, Nature 1998, 395, 151.
3. Conclusion
In conclusion, an efficient blue-emitting molecule PIAnCN,
which consists of PI and cyano substituted anthracene (AnCN),
is designed and synthesized. The doped device shows a max-
imum EQE of 6.77% and deep blue electroluminescence with
CIE coordinates of (0.15, 0.07). More importantly, the nondoped
OLEDs exhibit a maximum EQE of 9.44% which is acquired at
high luminescence of 1000 cd m−2. The efficiency roll-off is neg-
ligible. Even when the luminescence increases to 10 000 cd m−2,
the EQE is still as high as 8.09%. This result is among the best of
nondoped blue fluorescent OLEDs. We assume that TTA is the
reason for such high device efficiency. We believe that our result
is valuable for further practical applications, such as display and
lightings, which required to be operated at high luminescence.
[5] a) Q. Zhang, B. Li, S. Huang, H. Nomura, H. Tanaka, C. Adachi,
Nat. Photonics 2014, 8, 326; b) S. Hirata, Y. Sakai, K. Masui,
H. Tanaka, S. Y. Lee, H. Nomura, N. Nakamura, M. Yasumatsu,
H. Nakanotani, Q. Zhang, K. Shizu, H. Miyazaki, C. Adachi, Nat.
Mater. 2015, 14, 330.
[6] Q. Zhang, J. Li, K. Shizu, S. Huang, S. Hirata, H. Miyazaki,
C. Adachi, J. Am. Chem. Soc. 2012, 134, 14706.
Supporting Information
[7] C. J. Chiang, A. Kimyonok, M. K. Etherington, G. C. Griffiths, V. Jankus,
F. Turksoy, A. P. Monkman, Adv. Funct. Mater. 2013, 23, 739.
Supporting Information is available from the Wiley Online Library or
from the author.
[8] a) P. Y. Chou, H. H. Chou, Y. H. Chen, T. H. Su, C. Y. Liao, H. W. Lin,
W. C. Lin, H. Y. Yen, I. C. Chen, C. H. Cheng, Chem. Commun. 2014,
50, 6869; b) Y. H. Chen, C. C. Lin, M. J. Huang, K. Hung, Y. C. Wu,
W. C. Lin, R. W. C. Cheng, H. W. Lin, C. H. Cheng, Chem. Sci. 2016,
7, 4044; c) Y. J. Pu, G. Nakata, F. Satoh, H. Sasabe, D. Yokoyama,
J. Kido, Adv. Mater. 2012, 24, 1765; d) K. Okumoto, H. Kanno,
Acknowledgements
The authors appreciate the helpful discussions with Prof. Yuguang Ma
of South China University of Technology. The authors appreciate the
©
Adv. Funct. Mater. 2018, 1705813
1705813 (7 of 8)
2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.afm-journal.de
Y. Hamaa, H. Takahashi, K. Shibata, Appl. Phys. Lett. 2006, 89,
063504.
[9] Z. Wang, P. Lu, S. Chen, Z. Gao, F. Shen, W. Zhang, Y. Xu,
H. S. Kwok, Y. Ma, J. Mater. Chem. 2011, 21, 5451.
[10] X. Tang, Q. Bai, Q. Peng, Y. Gao, J. Li, Y. Liu, L. Yao, P. Lu, B. Yang,
Y. Ma, Chem. Mater. 2015, 27, 7050.
Y. Cao, Adv. Funct. Mater. 2015, 25, 5190; h) W.-C. Chen, Y. Yuan,
G.-F. Wu, H.-X. Wei, L. Tang, Q.-X. Tong, F.-L. Wong, C.-S. Lee, Adv.
Opt. Mater. 2014, 2, 626; i) J.-Y. Hu, Y.-J. Pu, F. Satoh, S. Kawata,
H. Katagiri, H. Sasabe, J. Kido, Adv. Funct. Mater. 2014, 24, 2064;
j) R. Kim, S. Lee, K.-H. Kim, Y.-J. Lee, S.-K. Kwon, J.-J. Kim,
Y.-H. Kim, Chem. Commun. 2013, 49, 4664.
[11] a) J.-Y. Hu, Y.-J. Pu, Y. Yamashita, F. Satoh, S. Kawata, H. Katagiri,
H. Sasabe, J. Kido, J. Mater. Chem. C 2013, 1, 3871; b) D. Zhang,
D. Zhang, L. Duan, ACS Appl. Mater. Interfaces 2016, 8, 23197.
[12] a) Y. Li, F. Li, H. Zhang, Z. Xie, W. Xie, H. Xu, B. Li, F. Shen, L. Ye,
M. Hanif, D. Ma, Y. Ma, Chem. Commun. 2007, 231; b) Y. Li,
F. Shen, H. Wang, F. He, Z. Xie, H. Zhang, Z. Wang, L. Liu, F. Li,
M. Hanif, L. Ye, Y. Ma, Chem. Mater. 2008, 20, 7312; c) C. Feng,
K. Wang, Y. Xu, L. Liu, B. Zou, P. Lu, Chem. Commun. 2016, 52,
3836; d) Y. Xu, H. Zhang, F. Li, F. Shen, H. Wang, X. Li, Y. Yu, Y. Ma,
J. Mater. Chem. 2012, 22, 1592.
[14] R. N. Jones, Chem. Rev. 1947, 41, 353.
[15] Z. M. Wang, X. H. Song, Z. Gao, D. W. Yu, X. J. Zhang, P. Lu,
F. Z. Shen, Y. Ma, RSC Adv. 2012, 2, 9635.
[16] B. Wardle, Principles and Applications of Photochemistry, John Wiley
& Sons Ltd, UK 2009.
[17] G. Ponterini, N. Serpone, M. A. Bergkamp, T. L. Netzel, J. Am.
Chem. Soc. 1983, 105, 4639.
[18] M. Aydemir, G. Haykır, A. Battal, V. Jankus, S. K. Sugunan, F. B. Dias,
H. A. Attar, F. Türksoy, M. Tavaslı, A. P. Monkman, Org. Electron.
2016, 30, 149.
[13] a) Q. Zhang, D. Tsang, H. Kuwabara, Y. Hatae, Bo Li, T. Takahashi,
S. Y. Lee, T. Yasuda, C. Adachi, Adv. Mater. 2015, 27, 2096;
b) C.-G. Zhen, Y.-F. Dai, W.-J. Zeng, Z. Ma, Z.-K. Chen, J. Kieffer,
Adv. Funct. Mater. 2011, 21, 699; c) B. Liu, Z. W. Yu, D. He,
Z. L. Zhu, J. Zheng, Y. D. Yu, W. F. Xie, Q. X. Tong, C. S. Lee,
J. Mater. Chem. C 2017, 5, 5402; d) S. Tang, W. Li, F. Shen, D. Liu,
B. Yang, Y. Ma, J. Mater. Chem. 2012, 22, 4401; e) S.-K. Kim,
B. Yang, Y. Ma, J.-H. Lee, J.-W. Park, J. Mater. Chem. 2008, 18, 3376;
f) Y. Wei, C.-T. Chen, J. Am. Chem. Soc. 2007, 129, 7478; g) M. Liu,
X. L. Li, D. C. Chen, Z. Xie, X. Cai, G. Xie, K. Liu, J. Tang, S. J. Su,
[19] L. Yao, Y. Pan, X. Tang, Q. Bai, F. Shen, F. Li, P. Lu, B. Yang, Y. Ma,
J. Phys. Chem. C 2015, 119, 17800.
[20] R. L. Martin, J. Chem. Phys. 2003, 118, 4775.
[21] S. Reineke, M. A. Baldo, Sci. Rep. 2014, 4, 3797.
[22] C. Ganzorig, M. Fujihira, Appl. Phys. Lett. 2002, 81, 3137.
[23] W. Brütting, J. Frischeisen, T. D. Schmidt, B. J. Scholz, C. Mayr,
Phys. Status Solidi A 2013, 21, 44.
[24] a) S. Sinha, A. P. Monkman, Appl. Phys. Lett. 2003, 82, 4651;
b) V. Jankus, C.-J. Chiang, F. Dias, A. P. Monkman, Adv. Mater.
2013, 25, 1455.
©
Adv. Funct. Mater. 2018, 1705813
1705813 (8 of 8)
2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim