Deep-blue thermally activated delayed fluorescence emitters containing diphenyl sulfone group for organic light emitting diodes

Article abstract of DOI:10.1166/jnn.2019.16702

Novel blue thermally activated delayed fluorescence (TADF) emitters, D1-DPS and D2-DPS, were designed and synthesized. Diphenyl sulfone (DPS) group functioned as a common acceptor, and it combined with each of two different spiro-acridine groups, D1 and D2. The calculated energy differences (ΔE_ST) of the singlet and triplet excited states of D1-DPS (0.062 eV) and D2-DPS (0.128 eV) had sufficiently small AE_ST values, which is favorable in the thermally activated reverse intersystem crossing (RISC) process from the T₁ state to the S₁ state. A device doped 10 wt% of D2-DPS with ADN host material, obtained 5.05% of external quantum efficiency with deep-blue emission having CIE_xy coordinates of (0.152, 0.065). The results showed that these molecules are promising host-free TADF deep-blue emitters by inhibiting concentration quenching.

Full text of DOI:10.1166/jnn.2019.16702

Article
Journal of
Nanoscience and Nanotechnology
Vol. 19, 4583–4589, 2019
Copyright © 2019 American Scientific Publishers
All rights reserved
www.aspbs.com/jnn
Printed in the United States of America
Deep-Blue Thermally Activated Delayed Fluorescence
Emitters Containing Diphenyl Sulfone Group for
Organic Light Emitting Diodes
In Hye Lee1, Ki Ju Kim1, Young Kwan Kim2ꢀ∗, Young Sik Kim2ꢀ∗, and Dong Myung Shin1ꢀ∗
1Department of Chemical Engineering, Hongik University, Seoul 04066, Korea
2Department of Information Display, Hongik University, Seoul 04066, Korea
Novel blue thermally activated delayed fluorescence (TADF) emitters, D1-DPS and D2-DPS, were
designed and synthesized. Diphenyl sulfone (DPS) group functioned as a common acceptor, and
it combined with each of two different spiro-acridine groups, D1 and D2. The calculated energy
differences (ꢀESTꢁ of the singlet and triplet excited states of D1-DPS (0.062 eV) and D2-DPS
(0.128 eV) had sufficiently small ꢀEST values, which is favorable in the thermally activated reverse
intersystem crossing (RISC) process from the T1 state to the S1 state. A device doped 10 wt%
of D2-DPS with ADN host material, obtained 5.05% of external quantum efficiency with deep-blue
emission having CIExy coordinates of (0.152, 0.065). The results showed that these molecules are
promising host-free TADF deep-blue emitters by inhibiting concentration quenching.
IP: 83.171.253.137 On: Fri, 10 May 2019 10:07:27
Keywords: Organic Light Emitting Diodes, Thermally Activated Delayed Fluorescence,
Copyright: American Scientific Publishers
Deep-Blue Emitter, Diphenyl Sulfone Derivative.
Delivered by Ingenta
1. INTRODUCTION
(ISC) by using sufficiently small energy difference (ꢃESTꢂ
between the lowest singlet (S1ꢂ and lowest triplet (T1ꢂ
excited states.4 This capability can achieve high external
EL quantum efficiencies (ꢁextꢂ, and realize 100% of ꢁint
when the rate of ISC from T1 into S1 is much higher
than the competing rate of the non-radiative decay for
T1.4–5 This property can be controlled by the optimization
of the chemical structural feature and the detailed under-
standing of the structure-property relationship.1ꢀ5 Recently,
versatile TADF molecular systems were reported, includ-
ing spiro-acridine,6 spirobifluorene, and diphenyl sulfone
(DPS) derivatives.1ꢀ6 A blue TADF material (DMAC-DPS)
containing DPS group showed an external quantum effi-
ciency of 19.5% with an emission peak at 450 nm2.
However, this material requires enhancement to solve the
problem of lower maximum brightness and efficiency roll-
off at high brightness.5
Organic light emitting diodes (OLED) are an attractive
technology because they can be used in a variety of sizes
and shapes, high-resolution flat panel displays, and power
consumption.1–4ꢀ6ꢀ13 The internal quantum efficiency (ꢁintꢂ,
which is the quantum statistical branching ratio of the
electron–hole pairs under electrical excitation, is limited
to 25% because of the forbidden spin of the radiative
decay of triplet excitons. However, phosphorescent OLEDs
(PHOLEDs) can obtain an ꢁint of 100%.2ꢀ4ꢀ6 To overcome
the disadvantages of PHOLEDs, such as lower electrolu-
minescence (EL) efficiency under high current density, and
low reliability in the blue region of the visible spectrum,
a new concept of thermally activated delayed fluorescence
(TADF) was considered in several works published.2ꢀ4ꢀ6
The third generation of OLED materials, TADF mate-
rials have attracted much interest because of their sta-
bility, high efficiency, and flexible molecular structure.1–4
TADF molecules have shown the thermally activated up-
conversion of non-emissive triplet (T1) excitons to emis-
sive singlet (S1ꢂ excitons through an intersystem crossing
In this paper, we describe the structural calculation and
synthesis as well as the photo-phycochemical characteris-
tics of two deep-blue emitter materials, which are types
of donor–acceptor (D–A) that exhibit TADF. In addition,
four EL properties, EQE, power efficiency (PE), CIExy
∗Authors to whom correspondence should be addressed.
J. Nanosci. Nanotechnol. 2019, Vol. 19, No. 8
1533-4880/2019/19/4583/007
doi:10.1166/jnn.2019.16702
4583

Deep-Blue Thermally Activated Delayed Fluorescence Emitters Containing Diphenyl Sulfone Group for OLEDs
Lee et al.
coordinates, and normalized EL emissions, are presented
to compare the two emitters.
2. EXPERIMENTAL DETAILS
2.1. Structural Design by DFT Calculation
In DPS, an oxygen unit has strong electronegativity, which
results in the electron-withdrawing nature of the sulfonyl
group. The sulfonyl group of DPS exhibits a tetrahedral
geometry, which limits the conjugation of compounds.
Consequently, DPS is the best-known typical electron
acceptor component in TADF molecules.16 The derivative
compounds based on DPS are expected to show a large
twisted conformation between the phenyl ring and the acri-
dine units by leading to a comparatively small ꢃEST and
the effective separation of molecular orbitals. However,
many TADF materials cannot achieve a deep-blue emis-
sion and show rather broad EL spectra. Therefore, in this
study, D1-DPS and D2-DPS that consisting of a common
acceptor group with two different spiro-acridine units were
designed and calculated.
Figure 1. Optimized geometries and calculated HOMO and LUMO
density maps for D1-DPS and D2-DPS.
These molecules showed sufficiently small ꢃEST between
the S1 and T1 of D1-DPS (0.062 eV) and D2-DPS
(0.128 eV). These small values indicate that the energy
barrier is sufficient for reverse intersystem crossing (RISC)
to exhibit TADF characteristics.
We investigated the factors responsible for the absorp-
tion energy and the electron population of molecular
orbitals by performing DFT and time-dependent den-
sity functional theory (TD-DFT) calculations in the
ground state by using the dependence on the amount
of charge transfer (CT) from the donor to acceptor for
2.2. Synthesis
2.2.1. General
Most reactants and solvents were purchased from commer-
cial sources and used without further purification. Tetrahy-
drofuran (THF) was purified using a solvent purification
1
system (Korea Kiyon Co., Ltd.). H-NMR was recorded
using a Bruker DRX400 spectrometer in DMSO or CDCl .
the optimal HF percentage in the TD-DFT exchange-
IP: 83.171.253.137 On: Fri, 10 May 2019 10:07:27
correlation. The geometries in the gas phase were opti-
mized by applying the DFT method and using the B3LYP
exchange-correlation function with the 6-31G∗ basis set
in the Gaussian 09 program package. We determined
an optimal HF% (OHF) by using the Multiwfn program
to analyze the optimal orbital composition for obtain-
ing a CT amount (q) from the donor to the acceptor;
the optimal HF% was proportional to the CT amount
according to the formula OHF = 42q. According to
the Franck-Condon principle, the crossing point between
absorption and emission corresponds to the E00 of the
CT transitions, as EVA ꢄS1ꢂ−E00ꢄS1ꢂ = E00ꢄS1ꢂ−EVEꢄS1ꢂ.
The value of E00ꢄS1ꢂ was calculated as (EVA ꢄS1, OHF)+
EVEꢄS1, OHF))/2.
Copyright: American Scientific Publishers
3
Delivered by Ingenta
UV-Vis spectra were obtained by a UV-Visible spec-
troscopy (Agilent 8453). PL spectra were recorded using
a fluorescence spectrometer (LS55, PerkinElmer precisely)
with hexane, dichloromethane (MC or DCM), toluene, and
dimethylformamide (DMF) as solvents to investigate the
solvent polarity in solution PL spectra. All the reaction
synthesis processes are described in Scheme 1.
2.2.2. Synthesis of 1-Bromo-4-(phenylsulfonyl)Benzene
(DPS)10
Benzenesulfonyl chloride (5 g, 28 mmol) and bromoben-
zene (4.84 g, 30.8 mmol) were placed in a round bottom
flask with N2 charging. Anhydrous aluminum chloride
The electron distributions of the highest occupied
molecular orbital (HOMO) and the lowest unoccupied
molecular orbital (LUMO) in these materials were calcu-
lated using the DFT method with the optimized ground-
state molecular geometry, as shown in Figure 1.
Table I. Calculated CT amount (q), OHF%, S1 and T1 state energies,
ꢃEST, and emission wavelengths [EVEꢄS1ꢂ] of D1-DPS and D2-DPS.
Parameter
D1-DPS
D2-DPS
The HOMO distribution was localized on the acridan
moieties, whereas the LUMO was localized on the DPS
moiety. The dihedral angles of D1-DPS and D2-DPS were
almost perpendicular. These larger dihedral angles led to
the separation between the HOMO and the LUMO, which
resulted in a small ꢃEST.
Table I shows the calculated CT amount (q), OHF%,
dihedral angle, S1 and T1 state energies, ꢃEST, and emis-
sion wavelengths [EVE(S1ꢂ] of D1-DPS and D2-DPS.
CT amount (q)
Optimal HF%
E0−0ꢄS1ꢂ
0ꢅ879
0ꢅ369
3ꢅ129
3ꢅ067
0ꢅ062
429ꢅ2
−5ꢅ130
−1ꢅ607
89.9ꢀ
0ꢅ877
0ꢅ368
3ꢅ161
3ꢅ033
0ꢅ128
424ꢅ5
−5ꢅ200
1ꢅ637
89.9ꢀ
E0−0ꢄT1ꢂ
ꢃEST
E
VEꢄS1ꢂ (nm)
HOMO
LUMO
Dihedral angle
Oscillator strength (f )
0ꢅ000
0ꢅ000
4584
J. Nanosci. Nanotechnol. 19, 4583–4589, 2019

Lee et al.
Deep-Blue Thermally Activated Delayed Fluorescence Emitters Containing Diphenyl Sulfone Group for OLEDs
IP: 83.171.253.137 On: Fri, 10 May 2019 10:07:27
Copyright: American Scientific Publishers
Delivered by Ingenta
Scheme 1. Reaction scheme for D1-DPS and D2-DPS.
(3.73 g, 28 mmol) was dissolved in a small amount of
upper reactant and added slowly to the flask with contin-
uous stirring. The mixture first changed to red and then to
dark reddish brown. The mixture was refluxed at 90 ꢀC for
7–8 h. The progress of the reaction was monitored by thin
layer chromatography (TLC) with a mixture of hexane and
ethyl acetate (5:1). After the reaction was completed, the
mixture was cooled and poured into a mixture of deion-
ized water and concentrated HCl. The obtained product
was recrystallized by ethanol.
49.4
mmol),
bis(dibenzylideneacetone)palladium(0)
(pd(dba)2, 0.2 g, 0.35 mmol), and [1,1-bis(diphenylphosphino)
ferrocene]dichloropalladium (pd(dppp)Cl2, 0.52 g,
0.71 mmol) were placed and stirred in a round bottom
flask with N2 charging. Then aniline (3.23 g, 35.3 mmol)
and 20 ml of toluene were injected. The mixture was
ꢀ
refluxed at 115 C for 2 h. The reaction was monitored
by TLC with a mixture of hexane and ethyl acetate
(10:1). After the reaction was completed, the mixture
was quenched by a saturated NaOH solution and then
worked up using water and ethyl acetate. The extracted
organic layer was dried by anhydrous MgSO4. It was then
filtered by celite and purified by column chromatography
with hexane and ethyl acetate. A slightly yellow oil was
obtained.
1
Yield: 74.4%; H-NMR (500 MHz, CDCl3ꢂ: ꢆ (ppm)
7.84–7.82 (d, 2H), 7.75 (d, 2H), 7.72–7.68 (d,
2H), 7.60–7.58 (d, 2H), 7.54–7.50 (t, 1H), 7.46–7.42
(t, 2H).
2.2.3. Synthesis of 2-Bromo-N-Phenylanniline (1)15
Intermediate compound 1 was synthesized by a Buchwald-
Hartwig reaction. 1-bromo-2-iodobenzene (10 g,
35.3 mmol), sodium-tert-butoxide (t-BuONa, 4.76 g,
1
Yield: 55.5%; H-NMR (500 MHz, CDCl3ꢂ: ꢆ (ppm)
7.51–7.48 (d, 2H), 7.32–7.28 (t, 2H), 7.24–7.22 (d, 1H),
7.14–7.12 (d, 3H), 7.04–7.00 (t, 1H), 6.73–6.69 (t, 1H),
6.07 (s, 1H).
J. Nanosci. Nanotechnol. 19, 4583–4589, 2019
4585

Deep-Blue Thermally Activated Delayed Fluorescence Emitters Containing Diphenyl Sulfone Group for OLEDs
Lee et al.
2.2.4. Synthesis of 10-(4-(phenylsulfonyl)phenyl)-10H-
Spiro[acridine-9,9ꢁ-fluorene] (D1-DPS)8, 11
2.2.5. Synthesis of 10-(4-(phenylsulfonyl)phenyl)-10H-
Spiro[acridine-9,9ꢁ-xanthene] (D2-DPS)8, 11
The other dopant material, D2-DPS, was synthesized
by the same procedure as described for D1-DPS in
Section 2.2.4. Only the reactant was changed in each step.
First, the –OH formation reaction to obtain the compound
1b and the cyclization reaction for compound 3 were per-
formed sequentially. Compound 1 (3 g, 12 mmol) was
dissolved in 50 ml of anhydrous THF under high purity
argon. In the n-butyllithuim reaction, a solution of xan-
thone (2.61 g, 13.3 mmol) in THF was used. The reaction
was ended after the confirmation of compound 1b by TLC
with a solution of hexane and ethyl acetate (10:1). Addi-
tional quenching was not required. The mixture was then
worked up using water and dichloromethane. The extracted
organic layer was dried by anhydrous MgSO4.
To obtain the D1-DPS, compound 1a of the –OH forma-
tion reaction, and compound 2 of the cyclization reaction
were synthesized. Compound 1 (3 g, 12 mmol) was dis-
solved in 50 ml of anhydrous tetrahydrofuran (THF) under
high purity argon in a one neck round bottom flask. It was
ꢀ
cooled to −78 C using solid carbon with acetone and
then n-butyllithuim (2.5 M in cyclohexane, 24.7 mmol)
was added dropwise slowly. Stirring was continued for
1 h before, cooling for 20–30 min. Then a solution of
9-fluorenone (2.4 g, 13.3 mmol) in THF was added at
−78 ꢀC under an argon atmosphere. The mixture was
stirred for 3 h in the same environment. The reaction was
completed after the confirmation of compound 1a. Two
side products and reactants appeared by TLC with a solu-
tion of hexane and ethyl acetate (10:1). Additional quench-
ing was not required. The mixture was then worked up
using water and dichloromethane. The extracted organic
layer was dried by anhydrous MgSO4.
For compound 3, compound 1b was placed in one-
neck round bottom flask, to which chloroform (10 ml)
and methane sulfonic acid (MSA, 5 ml) were added with
stirring at room temperature for 15 min. The progress of
the reaction was monitored by TLC with a solution of
hexane and ethyl acetate (5:1). The resulting mixture was
quenched with saturated aqueous NaHCO3 and extracted
with dichloromethane after the reaction was completed. In
addition, the solid product was recrystallized using chloro-
form and hexane. The obtained slightly yellow solid prod-
uct was purified by column chromatography with hexane
and ethyl acetate. A solid white product was obtained.
For compound 2, compound 1a was placed in a one-
neck round bottom flask, to which, chloroform (10 ml)
and methane sulfonic acid (MSA, 5 ml) were added while
stirring at room temperature for 15 min. The progress of
the reaction was monitored by TLC with a solution of
hexane and ethyl acetate (5:1). The resulting mixture was
quenched by saturated aqueous NaHCO and extracted
IP: 83.171.253.137 On: Fri, 10 May 2019 10:07:27
3
Copyright: American Scientific Publishers
with dichloromethane after the reaction completed. The
Two intermediates, DPS acceptor and compound 3,
Delivered by Ingenta
solid product was recrystallized using chloroform and hex-
were reacted for D2-DPS using Buchwald-Hartwig ami-
nation with the same procedure describe in Section 2.2.4.
A solid white product was obtained.
ane. A slightly yellow solid product was obtained, which
was then purified by column chromatography with hexane
and ethyl acetate.
1
Yield: 44.5%; H-NMR (500 MHz, CDCl3ꢂ: ꢆ (ppm)
Two intermediates, the DPS acceptor and the com-
pound 2 donor, were reacted for D1-DPS by the
Buchwald-Hartwig amination. Compound 2 (0.5 g,
1.38 mmol), DPS (0.48 g, 1.38 mmol), pd(dba)2 (0.008 g,
0.014 mmol), t-BuONa (0.27 g, 2.77 mmol), and triph-
enylphosphine (0.003 g, 0.011 mmol) showed high purity
Ar charging. Then toluene (20 ml) was added and refluxed
for 5 h at 115 ꢀC. Quenching was not required. The result-
ing mixture was worked up using ethyl acetate and water.
A small amount of water in the extracted organic layer
was removed by MgSO4 and filtered with celite to remove
the inorganic materials. Finally, column chromatog-
raphy was applied with hexane and dichloromethane.
The white solid product was recrystallized by
hexane.
8.28–8.26 (d, 2H), 8.11–8.09 (d, 2H), 7.70–7.67 (t, 2H),
7.64–7.61 (t, 4H), 7.26–7.14 (d, 4H), 7.10–7.08 (d, 2H),
6.96–6.94 (t, 2H), 6.87–6.84 (t, 4H), 6.71–6.68 (t, 2H),
6.15–6.12 (d, 2H). Elem, Anal. found for C37H25NO3S: C,
78.7617; H, 4.5699; N, 2.3221; O, 8.6813; S, 5.5692.
2.3. Device Fabrication and Measurement
We fixed the emitter concentration to a 10 wt% with
the ADN host material in the EML layer. The device
was fabricated by the deposition of the following mate-
rials and layers: indium-in-oxide (ITO) (150 nm)/TCTA
(50 nm)/10 wt% emitter: ADN (20 nm)/TPBi (30 nm)/Liq
(2 nm)/Al (100 nm). D1-DPS and D2-DPS emitters were
used separately, as shown in Figure 2. In this work, ITO
substrates with a sheet resistance of 10 ꢇ/sq were used,
and they were treated with oxygen plasma at 2×10−5 Torr
at 125 W for 2 min. The organic layers were deposited
by thermal evaporation through a shadow mask under a
high vacuum (8×10−5 Torr) condition. A water-getter was
used to absorb the residue moisture in the encapsulated
device. To measure the EL properties, Keithley SMU 238
and Minolta CS-100A were used, and 4 mm2 of OLEDs
1
Yield: 59.9%; H-NMR (500 MHz, CDCl3ꢂ: ꢆ (ppm)
8.30–8.28 (d, 2H), 8.11–8.10 (d, 2H), 7.81–7.79 (d, 2H),
7.64–7.61 (t, 4H), 7.39–7.36 (t, 4H), 7.26–7.14 (d, 4H),
7.10–7.08 (d, 2H), 6.91–6.85 (t, 2H), 6.60–6.56 (t, 2H),
6.41–6.36 (d, 2H), 6.26–6.20 (d, 2H). Elem, Anal. found
for C37H25NO2S: C, 75.1773; H, 4.5610; N, 2.1293;
O, 9.3124; S, 5.0175.
4586
J. Nanosci. Nanotechnol. 19, 4583–4589, 2019

Lee et al.
Deep-Blue Thermally Activated Delayed Fluorescence Emitters Containing Diphenyl Sulfone Group for OLEDs
showed deep-blue emission at 470∼480 nm. The result
indicates that two final products functioned as emitters bet-
ter than the host did. They were doped in the well-known
host material in each device. The emission peak of the
host and the absorption of the emitter should overlap to
produce an effective energy transfer from the host to the
dopant material in the EML. These materials overlapped
lightly in the range of 350–400 nm.
We also confirmed polarity dependence using the
Lippert-Mataga calculation, which indicated TADF char-
acteristics. The excited state of an emitter is stabilized in
more polar solvents, which is expected in an ICT property.
In this phenomenon, the excited electrons were moved
from the donor group to acceptor, and then the singlet was
Figure 2. Device structure using D1-DPS (Device 1). In D2-DPS
(Device 2), all components were exactly the same as in device 1, includ-
ing the doping concentration.
formed the triplet by ICT.
The negative slope in the Lippert-Mataga plot showed
the solvatochromism of emitters according to the polar-
ity of the solvents, and the emission characteristics were
confirmed by ICT. PL fluorescence emission peaks were
measured in four different solvents, hexane, toluene, MC,
and DMF. The emission peaked in DMF at about 490 nm,
which means a small amount of red shift was obtained
because of its molecular structures. Since chemical struc-
tures of D1-DPS and D2-DPS restrict the rotational, bend-
ing, and stretching modes because of their spiral structure.
These chemical structures interrupted the Frank-Condon
factor of the acridine moiety and led to the blue shift in
area were examined. The EL spectra and CIE coordinates
were identified and measured by a Minolta CS-2000 spec-
troradiometer. All measurements were carried out at room
temperature under ambient conditions.
3. RESULTS AND DISCUSSION
3.1. Material Property
We investigated the photochemical and photophysical
properties of two materials. As shown in Figure 3, UV-
Vis absorption and PL spectra wIePre: 8d3es.1cr7i1be.2d5u3s.i1n3g7bOotnh: Fri, 10 May 2019 10:07:27
the PL emission wavelength.4 This result indicates that the
Copyright: American Scientific Publishers
solution of D1-DPS and D2-DPS in MC. These molecules
electron donor group did not affect the molecule orbitals.
Delivered by Ingenta
showed three absorption peaks at about 250 nm, 275 nm,
As shown in Figure 4, D2-DPS demonstrated a stronger
emission intensity and a larger negative slope than D1-
DPS did. D2-DPS was expected to procedure more deep-
blue EL emissions than D1-DPS.
and 360 nm, respectively in UV-Vis spectra. The peak
at 360 nm appeared by charge transfer (CT) property. In
the solution PL spectra at room temperature, D2-DPS,
which introduced an oxygen atom into the acridine donor
unit, showed 5 nm of blue shift compared with D1-DPS.
The excitation wavelength was fixed at 275 nm for both
molecules during the PL measurements. Each material
As shown in Figure 5, mass transitions of the two emit-
ters were measured to investigate their thermal stability.
The emitters showed a similar tendency: 10% loss of mass
Figure 3. UV/PL spectra of D1-DPS and D2-DPS dissolved in MC
that has a cut off wavelength of 233 nm. Each solution was optimized at
10−5 M.
Figure 4. Lippert-Mataga plot using four different solvents (hexane,
toluene, MC, and DMF) of D1-DPS and D2-DPS.
J. Nanosci. Nanotechnol. 19, 4583–4589, 2019
4587

Deep-Blue Thermally Activated Delayed Fluorescence Emitters Containing Diphenyl Sulfone Group for OLEDs
Lee et al.
similar electro-emission peaks at 441 nm and 436 nm,
respectively. In the Figure 6(d), the CIExy coordinates
show that the emission of the fabricated devices was in
the deep-blue region, which means good color purity. The
CIE coordinates (0.153, 0.064) in Device 1 and (0.152,
0.065) in Device 2 were observed. D1-DPS and D2-DPS
exhibited the maximum external EL quantum efficiencies
(EQE) of 4.13% and 5.05%, respectively, as shown in
Figure 6(b). Both devices achieved slight efficiency roll-
off characteristics, so the decay time of the TADF mech-
anism was expected to be brief. These results could have
been caused by discordance between the absorption peak
of the two emitters at 360 nm and the emission peak of the
ADN at 400 nm. This discordance could have led to the
lower energy transfer between the luminescence materials.
Figure 6(c) shows the luminescence-PE spectra. Device 2
could not endure more than 1,300 cd/m2 of high lumi-
nescence. Device 2 showed a lower PE than Device 1. In
addition, the J (current density)–V (voltage) graph was
measured. Device 1 and Device 2 achieved 800 mA/cm2
and 85 mA/cm2 at 9 V, respectively. Because of their struc-
tural similarity, D1-DPS and D2-DPS showed similar per-
formance. However, D2-DPS demonstrated only one-third
of the luminescence that D1-DPS demonstrated. Thus, D2-
DPS functioned better than D1-DPS did in luminescence
efficiency. Fluorescent devices generally use low doping
concentration (1∼5 wt%) to obstruct the triplet–triplet
Figure 5. TGA curve showing mass transitions of D1-DPS and D2-
DPS in N2 atmosphere.
ꢀ
ꢀ
ꢀ
until 325 C. In the next step from 325 C to 410 C, the
emitters began to lose weight, which indicated that decom-
ꢀ
position also began at 410ꢀ C. Hence, both emitters were
thermally stable until 410 C. This stability resulted from
the twist structure that was designed to represent the TADF
property.
3.2. Device Performance
The normalized EL spectra of blue OLEDs at 9 V are
IP: 83.171.253.137 On: Fri, 10 May 2019 10:07:27
shown in Figure 6(a). Device 1 and Device 2 showed
annihilation (TTA) by suppressing the Dexter energy
Copyright: American Scientific Publishers
Delivered by Ingenta
Figure 6. Performance of Device 1 (D1-DPS) and Device 2 (D2-DPS). (a) Normalized EL characteristics, (b) luminescence-EQE characteristics,
(c) luminescence-PE, and (d) CIExy coordinates characteristics of blue OLEDs.
4588
J. Nanosci. Nanotechnol. 19, 4583–4589, 2019

Lee et al.
Deep-Blue Thermally Activated Delayed Fluorescence Emitters Containing Diphenyl Sulfone Group for OLEDs
transfer (DET). However, these two emitters inhibited con-
centration quenching at a concentration of 10 wt%.18 Con-
centration quenching was suppressed in the devices using
D1-DPS and D2-DPS emitters because the bulky and
largely twisted characteristics of the emitter molecules
obstruct ꢈ-stacking between them. Thus, these molecules
are promising host-free TADF deep-blue emitters because
they inhibit concentration quenching.
by the Ministry of Education (Nos. 201502080003,
2017R1D1A1B03028107 and 2015R1A6A1A03031833).
References and Notes
1. X. Wei, Y. Chen, R. Duan, J. Liu, R. Wang, Y. Liu, Z. Li, Y. Yi,
Y. Yamada-Takamura, P. Wang, and Y. Wang, J. Mater. Chem. C
5, 12077 (2017).
2. Q. Zhang, B. Li, S. Huang, H. Nomura, H. Tanaka, and C. Adachi,
Nature Photonics 8, 326 (2014).
3. B. Huang, Q. Qi, W. Jiang, J. Tang, Y. Liu, W. Fan, Z. Yin, F. Shi,
X. Ban, H. Xu, and Y. Sun, Dyes and Pigments 111, 135 (2014).
4. M. Numata, T. Yasuda, and C. Adachi, Chem. Commun. 51, 9443
(2015).
5. H. Wang, L. Xie, Q. Peng, L. Meng, Y. Wang, Y. Yi, and P. Wang,
Adv. Mater. 26, 5198 (2014).
4. CONCLUSION
In summary, we observed deep-blue OLEDs that exhib-
ited TADF characteristics, which we confirmed through
photochemical analysis by designing and synthesizing two
DPS derivative emitters, D1-DPS and D2-DPS. The ꢃEST
values of D1-DPS and D2-DPS were small enough to
demonstrate the TADF phenomenon through a small spa-
tial overlap between the HOMO and the LUMO; there
was a large dihedral angle between the donor and the
acceptor. Device 1 and Device 2, which were doped with
10 wt% of D1-DPS and D2-DPS, exhibited the maximum
EQE of 4.13% and 5.05%, respectively, with deep-blue
emissions having the CIE coordinates of (0.153, 0.064)
and (0.152, 0.065), respectively. Concentration quenching
was suppressed in the devices using D1-DPS and D2-
DPS emitters. These results showed that these molecules
are candidate materials for host-free TADF deep-blue
6. G. Mehes, H. Nomura, Q. Zhang, T. Nakagawa, and C. Adachi,
Angew. Chem. Int. Ed. 51, 11311 (2012).
7. S. Huang, Q. Zhang, T. Nakagawa, K. Kuwabara, K. Yoshizawa, and
C. Adachi, J. Chem. Theory Comput. 9, 3872 (2013).
8. S. O. Jeon and J. Y. Lee, Tetrahedron 66, 7295 (2010).
9. G. H. Lee, D. Y. Kwon, and Y. S. Kim, Mol. Cryst. Liq. Cryst. 621, 1
(2015).
10. A. Wani and S. Mahajan, Am. J. PharmTech. Res. 4, 1057 (2014).
11. A. H. Parham, P. Stoessel, T. Eberle, A. Jatsch, J. V. Kroe-
ber, T. Grossmann, and C. Pflumm, US patent US20170141327A1
(2014).
12. G. H. Lee and Y. S. Kim, Polym. Bull. 73, 2439 (2016).
13. F. Zinna, U. Giovanella, and L. D. Bari, Adv. Mater. 27, 1791 (2015).
14. Z. Yang, Z. Mao, Z. Xie, Y. Zhang, S. Liu, J. Zhao, J. Zhao, J. Xiu,
Z. Chi, and M. P. Aldred, Chem. Soc. Rev. 46, 915 (2017).
15. B. H. Yang and S. L. Buchwald, J. Organometallic Chem. 576, 125
IP: 83.171.253.137 On: Fri, 10(1M99a9y). 2019 10:07:27
emitters.
Copyright: American S1c6i.enMti.fiLciuP, uY.bSliesihnoe,rDs. Chen, S. Inomata, S. J. Su, H. Sasabe, and
Delivered by IngJ.eKnitdao, Chem. Soc. Rev. 51, 16353 (2015).
17. K. Wang, C. J. Zheng, W. Liu, K. Liang, Y. Z. Shi, S. L. Tao, C. S.
Acknowledgments: This work was supported by the
Basic Science Research Program through the National
Research Foundation of Korea (NRF) and, funded
Lee, X. M. Ou, and X. H. Zhang, Adv. Mater. 29, 1701476 (2017).
18. J. Y. Lee, N. Aizawa, M. Numata, C. Adachi, and T. Yasuda, Adv.
Mater. 29, 1604856 (2017).
Received: 23 June 2018. Accepted: 28 October 2018.
J. Nanosci. Nanotechnol. 19, 4583–4589, 2019
4589