Effect of the Number and Substitution Pattern of Carbazole Donors on the Singlet and Triplet State Energies in a Series of Carbazole-Oxadiazole Derivatives Exhibiting Thermally Activated Delayed Fluorescence

Article abstract of DOI:10.1021/acs.chemmater.8b02632

Substitution of a varying number of 9H-carbazole donors onto 2-(pentafluorophenyl)-5-phenyl-1,3,4-oxadiazole yields a series of donor-acceptor compounds with similar electrochemical band gaps, but dissimilar optical properties. Each substitution of a carbazole onto the 2- and 4-positions results in a 150-200 meV stabilization of the first singlet excited state, evident in the fluorescent emission. The first triplet excited state is less sensitive to these substitutions, leading to a reduced energy separation between the singlet and triplet states, δE_ST, with increasing carbazole content. OLED devices made using the donor-acceptor compounds exhibiting the lowest δE_ST were fabricated and exhibit sky blue emission with an external quantum efficiency of up to 24.4% at a luminance of 10 cd m^-2, and 9.2% at 1000 cd m^-2.

Full text of DOI:10.1021/acs.chemmater.8b02632

Article
pubs.acs.org/cm
Cite This: Chem. Mater. XXXX, XXX, XXX−XXX
Effect of the Number and Substitution Pattern of Carbazole Donors
on the Singlet and Triplet State Energies in a Series of Carbazole-
Oxadiazole Derivatives Exhibiting Thermally Activated Delayed
Fluorescence
Matthew W. Cooper,†,§ Xiaoqing Zhang,‡,§ Yadong Zhang,†,§ Soon Ok Jeon,∥ Hasup Lee,∥
Sunghan Kim,∥ Canek Fuentes-Hernandez,‡,§ Stephen Barlow,†,§ Bernard Kippelen,‡,§
and Seth R. Marder*,†,§
‡
§
†School of Chemistry and Biochemistry, School of Electrical and Computer Engineering and Center for Organic Photonics and
Electronics, Georgia Institute of Technology, Atlanta, Georgia 30332, United States
∥Samsung Advanced Institute of Technology, Samsung Electronics Co., Ltd., 130 Samsung-ro, Yeongtong-gu, Suwon-si,
Gyeonggi-do, Korea 16678
S
* Supporting Information
ABSTRACT: Substitution of a varying number of 9H-carbazole donors
onto 2-(pentafluorophenyl)-5-phenyl-1,3,4-oxadiazole yields a series of
donor−acceptor compounds with similar electrochemical band gaps,
but dissimilar optical properties. Each substitution of a carbazole onto
the 2- and 4-positions results in a 150−200 meV stabilization of the first
singlet excited state, evident in the fluorescent emission. The first triplet
excited state is less sensitive to these substitutions, leading to a reduced
energy separation between the singlet and triplet states, ΔEST, with
increasing carbazole content. OLED devices made using the donor−
acceptor compounds exhibiting the lowest ΔEST were fabricated and
exhibit sky blue emission with an external quantum efficiency of up to
24.4% at a luminance of 10 cd m−2, and 9.2% at 1000 cd m−2.
respectively, will decrease ΔEST.5 There is, however, a trade-off
INTRODUCTION
■
with retaining efficient fluorescence, since the radiative
Organic light-emitting diodes (OLEDs) have received
significant attention, both academically and in industry, since
transition probability will also decrease with increasing
the pioneering work of Tang and VanSlyke.1 Due to spin
statistics, devices made using conventional fluorescent emitters
are limited to roughly 25% internal quantum efficiency (IQE)
owing to the formation of ca. 75% of triplet excitons upon
charge recombination in the device.2 In emitters that exhibit
thermally activated delayed fluorescence (TADF), however,
the energy difference the lowest energy singlet excited state
(S1) and the lowest energy triplet state (T1), ΔEST, is
sufficiently low for a significant fraction of the triplets to be
thermally upconverted to singlets through reverse intersystem
crossing (RISC) before they decay to the ground state (S0);
this can then be followed, in some cases, by efficient
fluorescent emission from the S1 state.3,4 The approach most
often used to minimize ΔEST, and to increase RISC rates, is
predicated on the assumption that the S1 state for donor−
acceptor molecules can be well-described as a transition
between the highest occupied molecular orbital (HOMO) and
lowest unoccupied molecular orbital (LUMO). Increasing the
charge-transfer (CT) character of S1 by increasing spatial
separation of the HOMO and LUMO through increasing
localization of these orbitals on donor and an acceptor moiety,
HOMO−LUMO spatial separation. By utilizing TADF
emitters, OLEDs approaching 100% IQE and over 20%
external quantum efficiency (EQE) have been realized.6−8
2,5-Diphenyl-1,4,5-oxadiazole (DPO), which is well-known
for its electron transport properties, as well as high thermal
stability and photoluminescence quantum yield, and its
derivatives have been widely used as electron-transport
materials in OLEDs.9−14 The relatively high electron affinity
of this moiety has been utilized by incorporating DPO as the
acceptor into bipolar donor−acceptor (D-A) hosts for
phosphorescent emitters in OLED devices,15−17 including
both molecular compounds and polymers in which it is
combined with the ubiquitous 9-carbazolyl (Cz)
donor.15,16,18−20 Furthermore, CzDPO D-A compounds have
themselves been exploited as tunable emitters in conventional
fluorescent OLEDs.21−25 Use of DPO as the acceptor in TADF
emitters, however, has been limited. To date, DPO has been
Received: June 21, 2018
Revised: August 25, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.chemmater.8b02632
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
paired with relatively strong electron donors phenoxazine26,27
and 5-phenyl-5,10-dihydrophenazine,28 resulting in devices
with emission in the blue-green to green range and EQEmax
values of under 15%. Despite the common use of Cz as a donor
in TADF applications,29 examples of CzDPO emitters that
exhibit TADF have only very recently been demonstrated.30
The effect of an increasing number of donor substituents on
the photophysics of the resulting fluorophore has been
observed experimentally as a method for decreasing the
HOMO−LUMO gap, the S1 energy, and ΔEST.31 Several
recent computational studies have focused on elucidating the
mechanism for the purpose of developing advanced design
rules for TADF compounds. It has been calculated that a
model series of Cz-phthalonitrile-based TADF molecules show
a monotonic decrease in both S1- and T1-state energy with
increased number of Cz donors,32 with a larger decrease for the
S1 energy, leading to a parallel monotonic decrease in ΔEST.
The effect on the CT-like S1 state has been explained through
stabilization of the acceptor LUMO via inductive electron
withdrawal from the acceptor by the Cz substituents, with each
additional carbazole increasing this effect.33−35 The smaller
effect on the T1-state energy, conversely, is due to the T1 state,
at least for the less substituted compounds, being well
described by a localized excitation (LE) rather than an
intramolecular CT HOMO−LUMO transition.32,36 Addition-
ally, for efficient RISC, not only low ΔEST but also high spin−
orbit coupling between the triplet and singlet manifolds is
desirable,37,38 and it has also been shown that increased
substitution by donors can increase spin−orbit coupling when
the S1 and T1 states differ in character, whether they are CT vs
LE states, respectively, or have different CT character from one
another.36,39
The specific topology of the D-A TADF compound has only
very recently come under systematic investigation. Kim
described calculations on Cz-phthalonitrile TADF compounds,
suggesting that attachment of the donor at chemically
inequivalent positions significantly affects not the LUMO of
the acceptor, but the nature and energy of the donor HOMO
instead. The effect on the triplet state was, again, less
pronounced and exhibited diminishing significance as the
number of donors increased and the additive inductive effects
began to dominate over mesomeric effects.40 Geng et al. found
very little change in the energy of the S1 state by varying the
connectivity in xanthone-phenylene-carbazole compounds, but
were able to raise the T1 energy by moving from para- to meta-
to ortho-substitution patterns.41 To contribute to the under-
standing of the role of topology and substitution in the design
of TADF molecules, we have synthesized a series of
CznF(5−n)DPO compounds in which the effect of the number
of Cz groups on the acceptor has been partially decoupled
from the inductive stabilization of the LUMO on the acceptor.
This allows us to examine the mesomeric effects on the S1 and
T1 states more directly. OLEDs made utilizing the CzDPO
compounds as the emitter yielded EQEmax values ranging from
6% to over 24%, and include the most efficient OLEDs
containing DPO-based emitters reported to date.
shifts are listed in parts per million (ppm) and were referenced using
the residual nondeuterated solvent 1H signal or the solvent 13C
resonance, or using an internal standard (trifluoroacetic acid, −76.55
ppm) for 19F NMR. Column chromatography was carried out using
silica gel (60 Å, 40−63 μm, Sorbent) as the stationary phase. Mass
spectra were measured on an Applied Biosystems 4700 Proteomics
Analyzer using MALDI or a VG Instruments 70-SE using electron
impact (EI) mode. UV−vis spectra were measured using a Cary 5000
UV−vis−NIR spectrophotometer. Fluorescence, phosphorescence,
and emission transients were collected on a Jobin Yvon Fluorolog-3
equipped with a pulsed xenon lamp, photomultiplier tube (Horiba
R928), and pulsed LED excitation source (NanoLED). Fluorescent
lifetimes were fit using DAS6 decay analysis software. Solution and
solid-state absolute photoluminescence quantum yields were
determined on a Quantaurus-QY C11347 spectrometer. Electro-
chemical data were acquired using cyclic voltammetry in 0.1 M
−
nBu4N+PF6 in dry tetrahydrofuran (for reductive scans) or
dichloromethane (oxidative scans) under nitrogen, using a CH
Instruments 620D potentiostat, a glassy carbon working electrode, a
platinum wire auxiliary electrode, and, as a pseudo-reference
electrode, a silver wire anodized in 1 M aqueous potassium chloride
solution. The scan rate was 50 mV s−1, and ferrocene was added as an
internal reference. Thermogravimetric analysis was carried out using a
PerkinElmer Pyris 1 TGA with a heating rate of 10 °C min−1.
Computational Methodology. The molecular geometries in the
ground (S0) state were optimized using the B3LYP42 functional and
the 6-31G(d,p) basis set using the Gaussian 09 program.43 Excited-
state geometries and energies were derived by time-dependent density
functional theory (TD-DFT) at the MPW1B95/6-31G(d,p)44 level of
theory. Electronic properties were calculated by TD-DFT.
2-(Pentafluorophenyl)-5-phenyl-1,3,4-oxadiazole (2a). To a
solution of benzohydrazide (5.9 g, 43 mmol) in dry THF (100 mL)
was slowly added 2,3,4,5,6-pentafluorobenzoyl chloride (10 g, 43
mmol) at room temperature under nitrogen. During addition a white
solid appeared. After the addition was complete, the reaction was
warmed to room temperature. The reaction mixture was stirred for 18
h at room temperature. Pyridine (10 mL) was then added, the mixture
was stirred for an additional 1 h, and water (500 mL) was added. The
white solid was collected by filtration and dried under vacuum. The
product (N’-benzoyl-2,3,4,5,6-pentafluorobenzohydrazide) was ob-
tained as a white powder (14 g) and was used for the next step
without any future purification.
N′-Benzoyl-2,3,4,5,6-pentafluorobenzohydrazide (14.0 g, 42.4
mmol) from the previous step was suspended in POCl3 (50 mL),
and the reaction was stirred in an oil bath at 120 °C. During heating,
the white solid starting materials dissolved, and a solution was
obtained. The reaction was monitored by thin-layer chromatography
(dichloromethane as eluent). After 4 h, the reaction mixture was
brought to room temperature and was carefully dropped into ice
water (700 mL). A white solid precipitated, which was collected by
filtration, and was washed with water. After drying, the crude product
was purified by column chromatography, eluting with 5% ethyl acetate
in dichloromethane. After evaporating solvents, the white solid was
recrystallized from acetone/water. The white solid was filtered and
then dried under vacuum (11.5 g, 87%, two steps).
1H NMR (400 MHz, CDCl3): δ 8.14 (dt, J = 8, 1.6 Hz, 2H), 7.63−
7.54 (m, 3H). 19F NMR (376.5 MHz, CDCl3, CF3COOH): δ
−135.38 (m, 2F), −147.14 (m, 1F), −159.46 (m, 2F). The NMR
data are similar to those previously reported for the same compound
synthesized by oxidative coupling of 2-phenyl-1,3,4-oxadiazole and
pentafluorobenzene.45
2-(4-(tert-Butyl)phenyl)-5-(pentafluorophenyl)-1,3,4-oxa-
diazole (2b). To a solution of 4-(tert-butyl)benzohydrazide (5.0 g,
26 mmol) in dry THF (100 mL) was slowly added 2,3,4,5,6-
pentafluorobenzoyl chloride (6.0 g, 26 mmol) at room temperature
under nitrogen, during which a white solid appeared. After the
addition was complete, the reaction mixture was stirred at room
temperature for 2 h. Pyridine (8.0 mL) was added and stirred for
additional 1 h. Water (500 mL) was added to the reaction mixture,
and a brown solid was collected by filtration and dried under vacuum.
EXPERIMENTAL SECTION
■
All reagents were purchased from commercial sources and were used
without further purification. 1H, 13C, and 19F NMR spectra were
recorded using a Bruker Avance IIIHD 500 instrument operating at
500, 125, and 470.4 MHz, respectively, or on a Varian Mercury Vx
400 operating at 400, 100, and 376.5 MHz, respectively. Chemical
B
DOI: 10.1021/acs.chemmater.8b02632
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
The product (N′-(4-(tert-butyl)benzoyl)-2,3,4,5,6-pentafluorobenzo-
hydrazide) was obtained as a brown powder (9.8 g) and was used for
the next step without any further purification.
phenyl)-5-phenyl-1,3,4-oxadiazole (0.50 g, 1.60 mmol), carbazole
(0.803 g, 4.80 mmol), and potassium carbonate (2.20 g, 16.0 mmol).
The flask was evacuated and purged with nitrogen and DMSO (20
mL) was added. The reaction mixture was stirred overnight at room
temperature. Upon completion, the reaction mixture was precipitated
with water and the crude product was separated via filtration. Column
chromatography (silica gel, 10% hexanes/dichloromethane as eluent)
was performed to yield the higher Rf product 2-phenyl-5-(2,4,6-
tri(9H-carbazol-9-yl)-3,5-difluorophenyl)-1,3,4-oxadiazole as a yellow
powder, and the lower Rf product 2-phenyl-5-(2,4,5-tri(9H-carbazol-
9-yl)-3,6-difluorophenyl)-1,3,4-oxadiazole as a white crystalline
powder.
N′-(4-(tert-Butyl)benzoyl)-2,3,4,5,6-pentafluorobenzohydrazide
(9.0 g, 27 mmol) was suspended in POCl3 (50 mL). The reaction was
stirred at 100 °C (oil bath). During heating, the white solid starting
materials dissolved and a clear solution was obtained. The reaction
was monitored by thin-layer chromatography (dichloromethane as
eluting solution). After 6 h, the reaction mixture was brought to room
temperature and was carefully dropped into ice water (700 mL). A
white solid precipitated, which was collected by filtration and washed
with water. After drying, the crude product was purified by column
chromatography, eluting with dichloromethane. After evaporating
solvents, the product was recrystallized from acetone/water; a white
solid was collected by filtration and dried under vacuum (8.5 g, 84%,
two steps).
2,4,6CzDPO. Yield: 0.226 g (19%). 1H NMR (500 MHz, CDCl3):
δ 8.19 (d, J = 4 Hz, 2H), 8.11 (d, J = 8 Hz, 4H), 7.58 (t, J = 8 Hz,
2H), 7.51 (t, J = 8 Hz, 4H), 7.46−7.37 (m, 8H), 7.33 (t, J = 8 Hz,
4H), 7.17 (t, J = 8 Hz, 2H), 6.89 (d, J = 8 Hz, 2H). 13C{1H} NMR
(125 MHz, CDCl3): δ 164.59, 156.65 (t, JCF = 2.5 Hz), 156.41 (dd,
1JCF = 268 Hz, JCF = 5 Hz), 140.90, 139.49, 131.64, 128.46, 126.67,
1H NMR (400 MHz, CDCl3): δ 8.06 (d, J = 8 Hz, 2H), 7.57 (d, J =
8 Hz, 2H), 1.37 (s, 9H). 13C{1H} NMR (125 MHz, CDCl3): δ
165.83, 156.22 154.40 (m), 145.17 (dm, 1JCF = 256 Hz), 143.21 (dtt,
126.62, 126.54, 126.35, 124.39, 124.04, 123.66 (dd, JCF = 14 Hz, JCF
=
1
1JCF = 259 Hz, JCF = 13, 5 Hz), 138.13 (dm, JCF = 252 Hz), 127.08,
5 Hz), 122.18, 121.60, 120.69, 119.84 (t, JCF = 15 Hz), 109.89,
108.96. 19F NMR (470 MHz, CDCl3): δ −113.00 (s). HRMS
(MALDI, m/z): Calcd for C50H29N5OF2 (M+), 754.2413; Found,
754.2410. Anal. Calcd for C50H29N5OF2: C, 79.67; H, 3.88; N, 9.29.
Found: C, 79.75; H, 3.73; N, 9.45.
126.22, 101.00 (td, JCF = 15, 5 Hz), 35.16, 31.05. 19F NMR (376.5
MHz, CDCl3, CF3COOH): δ −135.45 (m, 2F), −147.04 (m, 1F),
−159.56 (m, 2F). HRMS (EI, m/z): Calcd for C18H13N2OF5 (M+),
368.0948; Found, 368.0944.
2,4,5CzDPO. Yield: 0.273 g (23%). 1H NMR (500 MHz, CDCl3):
δ 8.12 (d, J = 4 Hz, 2H), 7.80−7.78 (m, 2H), 7.76−7.74)m, 2H), 7.52
(t, J = 8 Hz, 2H), 7.44 (d, J = 8 Hz, 2H), 7.41 (d, J = 8 Hz, 1H), 7.35
(t, J = 8 Hz, 2H), 7.28 (t, J = 8 Hz, 2H), 7.24−7.21 (m, 4H), 7.15−
7.08 (m 10H). 13C{1H} NMR (125 MHz, CDCl3): δ 165. 05, 156.11
2-(4-(9H-Carbazol-9-yl)-2,3,5,6-tetrafluorophenyl)-5-phen-
yl-1,3,4-oxadiazole (CzDPO). In a round-bottom flask, 2-(penta-
fluorophenyl)-5-phenyl-1,3,4-oxadiazole (1.03 g, 3.30 mmol), carba-
zole (0.570 g, 3.41 mmol), and potassium carbonate (4.42 g, 32.0
mmol) were placed under vacuum and then under nitrogen. DMSO
(20 mL) was added and the reaction mixture was stirred overnight at
room temperature. The reaction progress was monitored via thin-layer
chromatography. Upon completion, the mixture was precipitated with
water and the crude product was collected via filtration. Column
chromatography (9:1 ethyl acetate/hexanes as eluent) yielded the
1
(q, JCF = 2.5 Hz), 154.22 (dd, JCF = 263 Hz, JCF = 3.8 Hz), 153.27
1
(dd, JCF = 255 Hz, JCF = 3.8 Hz), 140.76, 138.58, 138.43, 131.91,
128.71, 128.12 (dd, JCF = 14 Hz, JCF = 3.8 Hz), 128.76, 126.75,
125.73, 125.71, 125.10 (dd, JCF = 15 Hz, JCF = 3.8 Hz), 124.14,
124.13, 124.05, 122.45, 121.26, 121.15, 121.03, 120.62, 120.04,
120.02, 114.01 (d, JCF = 14 Hz), 109.91, 109.83, 109.35. 19F NMR
(470 MHz, CDCl3): δ −111.64 (d, J = 14 Hz, 1F), −120.54 (d, J = 14
Hz, 1F). HRMS (MALDI, m/z): Calcd for C50H29N5OF2 (M+),
754.2413; Found, 753.2347. Anal. Calcd for C50H29N5OF2: C, 79.67;
H, 3.88; N, 9.29. Found: C, 79.60; H, 4.00; N, 9.32.
1
higher Rf fluorescent product as a white powder (0.819 g, 54%). H
NMR (500 MHz, CDCl3): δ 8.21 (d, J = 6.5 Hz, 2H), 8.16 (d, J = 7.5
Hz, 2H), 7.63−7.57 (m, 3H), 7.50 (t, J = 7.5 Hz, 2H), 7.39 (t, J = 8
Hz, 2H), 7.22 (d, J = 8 Hz, 2H). 13C{1H} (125 MHz, CDCl3): δ
165.84, 155.16, 145.33 (ddt, 1JCF = 260 Hz, JCF = 14 Hz, JCF = 4 Hz),
144.14 (ddt, 1JCF = 255 Hz, JCF = 14 Hz, JCF = 4 Hz), 139.40, 132.50,
129.27, 127.30, 126.66, 124.35, 123.01, 121.65, 120.61, 119.83 (t, JCF
= 14 Hz), 109.94, 104.51 (t, JCF = 14 Hz). 19F NMR (470 MHz,
CDCl3): δ −134.94 (m, 2F), −140.38 (m, 2F). HRMS (EI, m/z):
Calcd for C26H13N3OF4 (M+), 459.0995; Found, 459.0995.
2-Phenyl-5-(2,3,4,6-tetra(9H-carbazol-9-yl)-5-fluorophenyl)-
1,3,4-oxadiazole (2,3,4,6CzDPO). A solution of 2-(pentafluoro-
phenyl)-5-phenyl-1,3,4-oxadiazole (1.0 g, 3.2 mmol), carbazole (2.13
g, 12.7 mmol), and potassium carbonate (7.0 g, 50. mmol) in DMSO
(20 mL) was stirred at room temperature for 27 h under nitrogen.
Water (100 mL) was added and a pale yellow solid was obtained. The
solid product was collected by filtration, and washed with water. After
drying, the crude product was purified by column chromatography
using dichloromethane/hexane (8:2) as eluent. After removal of
solvent, the yellow solid was dissolved in dichloromethane, and
methanol (100 mL) was added. Dichloromethane was removed and a
pale-yellow solid was recrystallized from methanol solution. The solid
product was collected by filtration. After drying under vacuum, the
2-(2,4-Di(9H-carbazol-9-yl)-3,5,6-trifluorophenyl)-5-phenyl-
1,3,4-oxadiazole (2,4CzDPO). In a round-bottom flask, 2-(penta-
fluorophenyl)-5-phenyl-1,3,4-oxadiazole (0.482 g, 1.54 mmol),
carbazole (0.408 g, 2.44 mmol), and potassium carbonate (3.5 g, 25
mmol) were placed under vacuum and then under nitrogen. DMSO
(10 mL) was added and the reaction mixture was stirred overnight at
room temperature. Upon completion, the mixture was precipitated
with water and the crude product was separated via filtration. Column
chromatography (10% ethyl acetate in hexanes as eluent) yielded the
1
product was obtained (1.6 g, 55%). H NMR (500 MHz, CDCl3): δ
1
8.09 (d, J = 8 Hz, 2H), 7.74−7.73 (m, 2H), 7.62 (d, J = 8 Hz, 2H),
7.49−7.44 (m, 4H), 7.39 (d, J = 8 Hz, 2H), 7.32−7.27 (m, 3H),
7.23−7.20 (m, 4H), 7.15−7.06 (m, 8H), 7.03 (t, J = 8 Hz, 2H),
6.85−6.80 (m, 4H), 6.73 (t, J = 8 Hz, 2H). 13C{1H} NMR (125
MHz, CDCl3): δ 164.30, 157.14 (d, J = 4 Hz), 156.39 (d, 1JCF = 261
Hz), 140.95, 139.71, 138.72, 138.12, 136.64 (d, J = 2.5 Hz), 134.13
(d, J = 4 Hz), 131.47, 129.35 (d, J = 14 Hz), 128.37, 127.16 (d, J = 15
Hz), 126.95 (d, J = 1 Hz), 126.55, 126.48, 125.61, 125.55, 124.68,
124.14, 124.01, 123.72, 122.26, 121.04, 120.97, 120.70, 120.56,
120.47, 120.04, 119.54, 110.05, 109.30, 109.28. 19F NMR (470 MHz,
CDCl3): δ −113.83 (s). HRMS (MALDI, m/z): Calcd for
C67H37N6OF (M+) 900.3013; Found 900.3005. Anal. Calcd for
C67H37N6OF: C, 82.65; H, 4.14; N, 9.33. Found: C, 82.43; H, 4.08;
N, 9.32.
product as a white powder (0.177 g, 19%). H NMR (500 MHz,
CDCl3): δ 8.16 (d, J = 5 Hz, 2H), 8.06 (d, J = 10 Hz, 2H), 7.52 (t, J =
10 Hz, 2H), 7.44−7.38 (m, 5H), 7.31−7.28 (m, 4H), 7.25 (t, J = 10
Hz, 2H), 7.20 (d, J = 10 Hz, 2H), 7.13 (d, J = 10 Hz, 2H). 13C{1H}
1
NMR (125 MHz, CDCl3): δ 165.09, 155.69, 152.77 (d, JCF = 259
Hz), 147.97 (dd, 1JCF = 262 Hz, JCF = 16 Hz), 146.93 (dd, 1JCF = 265
Hz, JCF = 16 Hz) 140.83, 139.46, 131.99, 128.71, 126.76, 126.75,
126.69, 124.38, 123.95, 122.30, 121.64, 121.35, 121.14, 120.66,
120.58, 119.84 (t, JCF = 15 Hz), 114.47 (d, JCF = 11 Hz), 114.42,
109.93, 109.07. 19F NMR (470 MHz, CDCl3): δ −120.34 (d, J = 14
Hz, 1F), −131.78 (d, J = 24 Hz, 1F), −131.89 (dd, J = 24 Hz, 14 Hz,
1F). HRMS (ESI, m/z): Calcd for C38H22N4OF3 (M+), 607.174;
Found, 607.1723.
2-Phenyl-5-(2,4,6-tri(9H-carbazol-9-yl)-3,5-difluorophenyl)-
1,3,4-oxadiazole (2,4,6CzDPO) and 2-Phenyl-5-(2,4,5-tri(9H-
carbazol-9-yl)-3,6-difluorophenyl)-1,3,4-oxadiazole (2,4,
5CzDPO). To a round-bottom flask were added 2-(pentafluoro-
2-(4-(tert-Butyl)phenyl)-5-(2,3,4,5,6-penta(9H-carbazol-9-
yl)phenyl)-1,3,4-oxadiazole (2,3,4,5,6CzDPO). A solution of 2-
(4-(tert-butyl)phenyl)-5-(pentafluorophenyl)-1,3,4-oxadiazole (0.20
C
DOI: 10.1021/acs.chemmater.8b02632
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
g, 0.54 mmol), carbazole (0.50 g, 3.0 mmol), and potassium
carbonate (2.0 g, 14 mmol) in DMSO (10 mL) was stirred at 100
°C for 10 h under nitrogen. After cooling to room temperature, water
(75 mL) was added and a pale yellow solid was obtained. The solid
product was collected by filtration and washed with water. After
drying, the crude product was purified by column chromatography
using dichloromethane/hexane (7:3) as eluent. After removal of
solvents, the solid was dissolved in dichloromethane, methanol (30
mL) was added, dichloromethane was removed, and the resulting
yellow solid was collected by filtration and dried under vacuum (0.55
g, 92%). 1H NMR (500 MHz, CDCl3): δ 7.64−7.61 (m, 4H), 7.36 (d,
J = 8 Hz, 4H), 7.30 (d, J = 4 Hz, 2H), 7.24−7.20 (m, 10H), 7.08 (d, J
= 8.5 Hz, 2H), 6.97−6.95 (m, 8H), 6.79−6.73 (m, 6H), 6.68−6.65
(m, 6H), 6.61 (t, J = 7.5 Hz, 2H), 1.22 (s, 9H). 13C{1H} NMR (125
MHz, CDCl3): δ 164.12, 157.34, 154.83, 139.73, 139.57, 138.34,
137.95, 137.04, 127.30, 126.30, 125.49, 125.24, 124.37, 124.23,
123.80, 123.60, 120.49, 120.31, 120.21, 120.02, 119.57, 119.41,
119.31, 110.75, 110.51, 109.38, 34.86, 30.98. HRMS (MALDI, m/z):
Calcd for C78H53N7O 1103.4312 (M+); Found 1103.4295. Anal.
Calcd for C78H53N7O: C, 84.84; H, 4.84; N, 8.88. Found: C, 84.65;
H, 4.72; N, 8.91.
Chart 1. Compounds Studied
OLED Fabrication. The performance of three of the emitters was
evaluated in OLEDs with the structure: glass/ITO/MoO3 (15 nm)/
Poly-TriCZ (80 nm)/DPEPO:emitter (25 nm, 25 wt %)/TP3PO (4
nm)/TPBi (50 nm)/LiF (1 nm)/Al (50 nm)/Ag (100 nm) as shown
in Figure 4a (see Chart S1 in the Supporting Information for
molecular structures of the materials used). The thin layer of
molybdenum trioxide was evaporated as a hole-injection layer. The 80
nm thick Poly-TriCZ46 hole-transporting layer was spin-coated (500
rpm for 60 s) from a chlorobenzene solution (10 mg mL−1) and
thermally annealed (110 °C for 2 min) in a nitrogen-filled glovebox.
The emissive layer (EML) was deposited via a co-evaporation of
DPEPO:emitter with a dopant concentration of 25 wt %. Sequentially
deposited TP3PO and TPBi layers were used as electron transport
layers. Lithium fluoride was deposited to improve the electron
injection between a 50 nm thick aluminum cathode and the organic
layers. A silver layer was deposited atop the aluminum to assist with
the contact between the devices and the test setup. The chemical
structures of all organic materials used in the devices are shown in the
Supporting Information. The typical OLED active area was 3.0 mm ×
3.0 mm. All thermal depositions were performed at pressures below
1.0 × 10−7 Torr.
phere; decomposition temperatures (Td), defined here as the
temperature at which 5% mass loss in the TGA curve was
observed (Figure S5 in the Supporting Information), are
summarized in Table 1. All six compounds show decom-
position temperatures in excess of 300 °C, and Td increases
with increasing replacement of fluorine with carbazole.
Oxidation and reduction potentials were measured using
cyclic voltammetry in dichloromethane and tetrahydrofuran,
+
respectively, and all values were referenced against an FeCp2 /
FeCp2 internal standard. In the case of nonreversible oxidation
processes, potentials were estimated from the onset at a scan
rate of 50 mV s−1. The reduction waves were sufficiently
reversible processes that E1/2 values could be estimated. The
compounds became slightly more difficult to reduce as
additional carbazole groups were added, with reduction
potentials shifting from −2.00 V for CzDPO to −2.19 V for
2,3,4,5,6CzDPO, suggesting fluoro has a greater stabilizing
effect on the LUMO than carbazolyl, consistent with it acting
as a slightly stronger electron-withdrawing group. All
compounds showed irreversible oxidation behavior during
the cathodic sweep, consistent with electrochemical oxidation
occurring on the Cz donor.47 The compounds are increasingly
easily oxidized with additional Cz substitution, again consistent
with the slightly greater electron-withdrawing effects of fluoro
substituents. Values ranged from +1.17 V for CzDPO down to
+1.00 V for 2,3,4,5,6CzDPO. Taken together, the electro-
chemical gap (Egec = e(Eox − Ered)) changed minimally within
the series, only varying between 3.12 and 3.19 eV.
RESULTS AND DISCUSSION
■
The carbazole donor/2,5-diphenyl-1,3,4-oxadiazole acceptor
compounds in this study (Chart 1) were synthesized through
nucleophilic aromatic substitution with differing stoichiome-
tries of 9H-carbazole on the pentafluorophenyl DPO
precursors 2-(pentafluorophenyl)-5-phenyl-1,3,4-oxadiazole or
2-(4-(tert-butyl)phenyl)-5-(pentafluorophenyl)-1,3,4-oxadia-
zole. Very recently, a close analogue of 2,3,4,5,6CzDPO,
differing only in lacking the tert-butyl group, has been obtained
through similar chemistry.30 Compounds 2,4,6CzDPO and
2,4,5CzDPO were obtained from the same reaction in a 0.8:1
ratio (based on isolated yields). Detailed synthetic procedures
and characterization data can be found in the Experimental
1
Section. The CzDPO compounds were characterized by H,
13C{1H}, and, when appropriate, 19F NMR, and high-
resolution mass spectrometry (HRMS), and purity was
confirmed through elemental analysis (EA). In particular, the
regiochemistry of 2,4CzDPO, 2,4,6CzDPO, 2,4,5CzDPO,
and 2,3,4,6CzDPO was confirmed through 19F NMR and
13C{19F} NMR spectroscopy. Additional purification by
vacuum sublimation was performed on compounds before
incorporation into OLEDs.
The B3LYP/6-31G(d,p)-calculated energies of the HOMO
and LUMO (illustrated in Figure 2) are included in Table 1 for
comparison. Although the energies do not precisely track with
the observed electrochemical potentials, HOMO and LUMO
energies are in good agreement and, similarly, increase slightly
relative to CzDPO with increased Cz substitution. The
resulting difference in energy of the HOMO and LUMO,
The thermal stability of the compounds was examined using
thermogravimetric analysis (TGA) under a nitrogen atmos-
D
DOI: 10.1021/acs.chemmater.8b02632
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
+
Table 1. Thermal and Electrochemical Potentials of Compounds vs FeCp2 /FeCp2 and DFT-Calculated Frontier-Orbital
Energies
electrochemical potential
calculated orbital energies
LUMO/eV
a
b
c
compound
CzDPO
2,4CzDPO
2,4,6CzDPO
2,4,5CzDPO
2,3,4,6CzDPO
2,3,4,5,6CzDPO
Td/°C
Eox/V
Ered/V
Eg/eV
HOMO/eV
ΔEHL/eV
308
373
421
409
426
478
1.17
1.15
1.11
1.08
1.02
1.00
−2.00
−1.99
−2.04
−2.05
−2.10
3.17
3.14
3.15
3.13
3.12
3.19
−5.78
−5.54
−5.54
−5.61
−5.52
−5.42
−2.14
−2.08
−2.03
−2.10
−1.97
−1.89
3.64
3.46
3.52
3.51
3.55
3.53
−2.19
b
a
c
From 5% weight loss temperature by TGA at 10 °C min−1. Eg = e(Eox − Ered). ΔEHL = ELUMO − EHOMO
.
Table 2. Solution Photophysical Properties in Dilute Toluene Solution
a
fluorescence
phosphorescence
b
c
compound
CzDPO
2,4CzDPO
2,4,6CzDPO
2,4,5CzDPO
2,3,4,6CzDPO
2,3,4,5,6CzDPO
λabs/nm
288, 344
288, 315, 329, 347
288, 316, 329, 345
289, 314, 329, 361
290, 319, 331
λmax/nm
onset/eV
λmax/nm
onset/eV
ΔEST /eV
ϕFl
τFl/ns
444
449
490
459
483
482
3.24
3.10
2.85
3.06
2.86
2.93
508
516
521
484
471
459
2.75
2.72
2.73
2.71
2.79
2.81
0.49
0.38
0.12
0.35
0.07
0.12
0.21/0.27
0.06/0.08
0.04/0.06
0.12/0.20
0.06/0.10
6.7
4.8
3.9
7.7
3.2
2.9
291, 333
0.03/0.13
a
b
c
Recorded in toluene matrix at 77 K. Estimated from the difference of onset of fluorescence and onset of phosphorescence. Photoluminescence
quantum yield in aerated/nitrogen sparged solution.
ΔEHL, ranges from a maximum 3.64 eV in CzDPO to a
minimum of 3.46 eV, a difference of 0.17 eV.
Solution state photophysical data were collected in dry
toluene unless otherwise stated and are shown in Table 2. All
compounds exhibit UV absorptions from ca. 288 to 297 nm
and weaker absorptions from 314 to 333 nm, attributable to
π−π* absorptions by electronic states localized on the Cz and
DPO moieties. The broad, featureless absorptions at wave-
lengths above ca. 340 nm are assigned to intramolecular
charge-transfer (ICT) transitions from the Cz donor to DPO
acceptors. As seen in Figure 1a, fluorescent emission of all
molecules displayed the broad, structureless line shape
expected from an emission from a CT S1 state. Furthermore,
the emission for all compounds is solvent-dependent, while the
absorption spectra were largely insensitive to solvent (see the
Supporting Information). The emission maximum for CzDPO,
for example, in dichloromethane (475 nm) is red-shifted by
350 meV versus that in cyclohexane (419 nm). If solvation is
similar for all the compounds studied, electron density from Cz
moieties contributing to the S1 state and aligned partially
against the overall direction of the excited-state dipole,
specifically those in the 2- and 6-positions, would be expected
to decrease the difference in dipole between the ground and
excited states, and subsequently the emission dependence on
solvent polarity.48 Thus, as expected, more highly substituted
compounds show less solvatochromic fluorescence. The
pentasubstituted 2,3,4,5,6CzDPO, for example, exhibits a
lower red shift of 240 meV between cyclohexane (461 nm) and
dichloromethane (506 nm).
The energy of the S1 state, estimated using the onset of
fluorescence, differs more significantly over the series (0.39 eV
variation) than the electrochemical gap (0.07 eV variation).
Compound 4CzDPO exhibits a fluorescence onset at 3.24 eV.
Substituting one fluorine with carbazole yields 2,4CzDPO and
lowers the S1-state energy to 3.10 eV. Trisubstituted
compounds 2,4,6CzDPO and 2,4,5CzDPO are at 2.85 and
3.06 eV, respectively. Addition of a fourth Cz donor on
Figure 1. (a) Steady-state fluorescence spectra collected at room
temperature in dilute toluene solution. (b) Photoluminescence
spectra collected in toluene matrix at 77 K with a 500 μs delay
before data collection and attributed to phosphorescence.
2,3,4,6CzDPO resulted in onset at 2.86 eV, and pentasub-
stituted 2,3,4,5,6CzDPO is at 2.93 eV. Fluorescence maxima
follow the same trend as the onsets.
E
DOI: 10.1021/acs.chemmater.8b02632
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 2. HOMO, LUMO, and singlet and triplet NTOs for compounds studied. NTO hole (blue) and particle (red) wave functions are shown
superimposed.
When interpreting the structural effects on the singlet state
energies, two observations stand out. First, the range of S1-state
energies (ca. 400 meV) is significantly higher than in the
electrochemical gaps, (ca. 70 meV). Similarly, the TD-DFT
calculated singlet energies vary by up to 330 meV, while the
calculated HOMO−LUMO gap varies by only 170 meV.
Second, substitution of Cz into the 2- and 6-positions provides
strong stabilization of S1, while substitution at the 3- and 5-
positions has relatively little effect. The molecules can be
classified into three groups based on their S1 energies, and on
how many donors are substituted in the critical 2- and 6-
positions: compounds with 2,6-donors exhibit the most red-
shifted fluorescence (red shifts of 0.39, 0.38, and 0.31 eV for
the onsets of 2,4,6CzDPO, 2,3,4,6CzDPO, and 2,3,4,5,6Cz-
DPO vs that of CzDPO); compounds substituted in only the
2-position exhibit intermediate S1-state energies (red shifts of
0.14 and 0.18 eV for 2,4CzDPO and 2,4,5CzDPO vs
CzDPO), while CzDPO has the highest S1 energy.
Delayed photoluminescence spectra (500 μs delay before
detection) were collected at 77 K and are attributable to
phosphorescence. The T1 energy was estimated using the onset
of phosphorescence, and ΔEST was calculated as the difference
in estimated S1- and T1-state energies, as is common in the
literature. In contrast to the S1 energy, which, as noted above,
varied by ca. 400 meV in the series, the T1-state energy was
relatively insensitive to donor substitution, varying by ca. 100
meV. However, the line shapes of the phosphorescence spectra
do fall into two distinct classes. The compounds that retain the
fluoro substituent in both 3- and 5-positions (Figure 1b, lower
left) displayed a line shape resembling the phosphorescent
emission of an isolated donor or acceptor; specifically, a Cz-
like line shape7 is seen in 4CzDPO and a DPO-like emission15
is seen for 2,4CzDPO and 2,3,5CzDPO. Previously reported
quantum-chemical calculations have found the T1 state on
similar CzDPO compounds to be an LE state located on the
DPO moiety.18 The compounds that contain Cz in the 3- and/
or 5-positions show an emission line shape distinct from the
previous compounds, with a well-resolved 0,0-peaked high-
frequency (ca. 1300 cm−1) vibronic structure (Figure 1b, lower
right). This may indicate a different T1-state character. In
particular, the line shape indicates that the T1 and S0
geometries are primarily related through high-frequency
modes; this could possibly be due to suppression of low-
frequency torsional relaxations by the steric crowding of
adjacent carbazole groups in the 3- and/or 5-Cz-substituted
species. Noda et al. have also recently described donor
substitution in the 3- and 5-positions of benzonitrile as
modifying the energies of the triplet states independent of
those of the CT state.49
The highest T1 energies of the series were found for the
more highly substituted 2,3,4,6CzDPO (2.79 eV) and
2,3,4,5,6CzDPO (2.81 eV). Due to the relatively small
variation in T1 energies, the estimated ΔEST values roughly
followed the trend for the S1 energy and are shown in Table 2.
Compounds 2,4,6CzDPO, 2,3,4,6CzDPO, and 2,3,4,5,6Cz-
DPOthe compounds with the lowest energy S1 states
F
DOI: 10.1021/acs.chemmater.8b02632
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
exhibit the smallest ΔEST values, suggesting they are the most
promising candidates for TADF emission.
DFT calculations (Figure 2) indicate that the HOMOs are,
as expected, primarily located on one or more of the electron-
rich carbazoles, with small coefficients on the bridging phenyl.
The LUMOs are more broadly spread across the entire 2,5-
diphenyl-1,3,4-oxadiazole moiety, with higher coefficients on
the bridging phenyl that is substituted with the inductively
electron-withdrawing carbazolyl and fluoro substituents than
t
on the terminal C6H5 or C6H4 Bu group.
Natural transition orbitals (NTOs), based on TD-DFT
calculations at optimized S1 and T1 geometries, are shown
alongside the HOMOs and LUMOs. The NTOs indicate a
singlet emission generally well represented by a LUMO-to-
HOMO CT-type transition, although the partitioning of the
hole wave function between the different carbazole moieties
differs a little in some cases from that seen for the HOMO
wave function. For all compounds besides CzDPO, significant
density of the hole wave function is located on carbazoles in
the 2- and/or 6-positions. This is consistent with the
experimentally observed decrease in the sensitivity of the
Stokes shift to solvent polarity in the series, and with the key
role of donor substitution in the 2- and 4-positions to
stabilizing the fluorescent emission, perhaps due to electro-
static stabilization of more closely adjacent hole and electron.
In contrast to the NTOs of the lowest singlet emission, the
T1 NTOs show mixed CT and LE character, with that of
CzDPO showing particularly predominant DPO LE character.
As in the S1 NTOs, the particle wave functions are well
confined to the DPO moiety. The hole wave functions in the
T1 states, while displaying significant density on the Cz donors,
are not as strongly Cz-localized as those in the S1 states and
also have significant density on the bridging phenyl groups and
the oxadiazoles. It has been calculated that, for a given ΔEST,
compounds containing S1 and T1 states with different orbital
nature exhibit increased RISC rates relative to those for which
S1 and T1 have similar orbital character due to increased spin−
orbit coupling between the singlet and triplet manifolds.
To probe the potential application of these compounds as
TADF emitters, fluorescence transients were collected for the
more promising compounds in toluene solution that had been
sparged by bubbling nitrogen for ca. 15 min; these are shown
in Figure 3. Compound 2,3,4,5,6CzDPO displayed a
biexponential fluorescent decay composed of a prompt
component with a lifetime of 3.2 ns and a strong delayed
luminescence component, attributed to TADF, with a lifetime
of 62.8 μs. Compounds 2,4,6CzDPO and 2,3,4,6CzDPO also
displayed biexponential fluorescent decay (prompt fluores-
cence τfl = 3.9 and 3.2 ns, respectively) with minor delayed
fluorescence components that, while distinguishable from the
instrument response function (IRF), could not be reliably
fitted to determine the delayed lifetime. The long-lived
luminescence components were quenched after exposure to
air, consistent with TADF, where the triplet state can be
quenched by oxygen before undergoing RISC. It is worth
noting that, while 2,4,6CzDPO exhibited detectable TADF,
regioisomer 2,4,5CzDPO, which exhibits a larger ΔEST value,
does not, illustrating the importance of D-A topology in the
design of TADF compounds.
Figure 3. Transient photoluminescent decay curves at 300 K. (a)
Data in toluene solution at room temperature after sparging with
nitrogen. The instrument response function (IRF) is shown in black;
although its width and the weakness of delayed fluorescence preclude
extraction of τDF values for 2,4,6CzDPO and 2,3,4,6CzDPO, the data
clearly show a delayed component for these compounds that is much
weaker than that for 2,3,4,5,6CzDPO. (b) Data for evaporated thin
films in which fluorophores (25 wt %) are doped into DPEPO
(transients collected under nitrogen).
maxima of the films were 472, 484, and 479 nm for
2,4,6CzDPO, 2,3,4,6CzDPO, and 2,3,4,5,6CzDPO, respec-
tively. Each compound displayed fluorescent decay with
contribution from fast and slow components when measured
under a nitrogen stream. When fit with a biexponential decay
function, delayed fluorescent lifetimes (τDF) of 308.7, 166.2,
and 75.0 μs, respectively, were determined. Absolute photo-
luminescent quantum yields were significantly higher than
when determined in toluene solution, presumably due to
decreased nonradiative relaxation rates in thin film vs in
solution. Under a stream of nitrogen, the PLQYs were 0.32,
0.54 and 0.61, respectively.
The performance of the three emitters exhibiting TADF
discussed above was evaluated in OLEDs with the structure
shown in Figure 4a (see Chart S1 in the Supporting
Information for chemical structures of the materials). Figure
4b displays the normalized electroluminescence (EL) spectra
of the three devices. The EL spectra of the devices exhibit
more structure than the corresponding solution fluorescence
spectra; in addition to the peak at ca. 490 nm, each device also
exhibits a subsidiary peak or shoulder at ca. 470 nm, and longer
wavelength shoulders. However, the positions of the strongest
Thin films of the three compounds displaying TADF
properties were deposited by co-evaporation of the fluo-
rophore (25 wt %) with a bis[2-(diphenylphosphino)phenyl]-
ether oxide (DPEPO) host on a quartz substrate. The emission
G
DOI: 10.1021/acs.chemmater.8b02632
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Figure 4. (a) Device structure, (b) electroluminescence (EL) spectra, (c) current−voltage−luminance (J−V−L) characteristics, and (d) EQE of
three devices containing 2,4,6CzDPO, 2,3,4,6CzDPO, and 2,3,4,5,6CzDPO as emitters, respectively.
a
Table 3. Device Performance of OLEDs Made Using CzDPO Emitters
emitter
Von/V
EQEmax/%
CEmax/cd A−1
PEmax/lm W−1
CIE/(x, y)
2,4,6CzDPO
2,3,4,6CzDPO
2,3,4,5,6CzDPO
5.1
5.5
4.1
6.1
17.8
24.4
12.3
40.6
46.3
7.7
22.8
34.6
(0.17, 0.30)
(0.18, 0.36)
(0.16, 0.29)
a
Turn-on voltage (Von), defined as that required to achieve 10 cd m−2, external quantum efficiency (EQE), current efficacy (CE), power efficacy
(PE), and CIE coordinates of OLEDs (from EL spectra under a bias of 5 V).
maximum and the overall emission widths are consistent with
those seen in the solution fluorescence spectra. Similar
differences have been seen between the fluorescence and EL
spectra of other emitters; e.g., the EL of a benzofurocarbazole-
triazine TADF emitter has recently been found to exhibit
concentration-dependent structure,50 while the corresponding
fluorescence spectra are structureless.51 Although the EL
spectra are similar for all three emitters, as are the
corresponding fluorescence spectra, the relative intensity of
the component at 470 nm is significantly smaller for the device
containing the 2,3,4,6CzDPO emitter, leading to a greener
light emission of the device.
devices ranged from 6.1% in 2,4,6CzDPO, to 17.8% in
2,3,4,6CzDPO, and up to 24.4% at a reference luminance of
10 cd/m2 in 2,3,4,5,6CzDPO.52 Note that, under higher
injection conditions and at a luminance of 1000 cd/m2, more
relevant for applications, devices with 2,3,4,5,6CzDPO yield
EQE values of 9.2% in nonoptimized devices. The highest
EQEmax values are large despite moderately large τDF values;
however, these lifetimes likely contribute, via processes such as
triplet-triplet or triplet-polaron annihilation, to the substantial
roll-off in EQE seen with luminance (Figure 4c). Moreover,
values of EQEmax increase with decreased thin-film τDF values.
CONCLUSIONS
Current−voltage−luminance characteristics and EQE as a
■
function of luminance are shown in Figure 4c,d, respectively,
and device data are summarized in Table 3. Despite having
similar estimated ΔEST in solution, the compounds in which
there are donors in the 3- and 5-positions displayed stronger
delayed fluorescence with shorter delayed fluorescent lifetimes.
This trend of spectroscopic characteristics favoring TADF was
continued in the device efficiencies. EQEmax values in the
In summary, we have synthesized a series of DPO compounds
with a varying number and connectivity of fluorine atoms and
carbazole groups. The compounds exhibit very similar
electrochemical gaps but display S1-state energies that have a
strong dependence on D-A topology. In previously studied
systems, additional donors attached through electronegative
heteroatoms can change the S1 energy due to inductive effects
H
DOI: 10.1021/acs.chemmater.8b02632
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
stabilizing the LUMO regardless of position. In this system,
where the substitution of carbazole replaces an electronegative
fluorine instead of hydrogen, the inductive effect on the
LUMO is largely negated. Substitution in the 2- and/or 6-
positions is shown to lead to stabilization of the S1 state
relative to that of 4-CzDPO. However, substitution in the 3-
and/or 5-positions does not affect the S1 state significantly.
Triplet state energies were much less sensitive to the
substitution of Cz versus fluorine atoms; however, there is
some evidence that the substitution pattern plays a more
important role in the nature of the triplet state and, as a result,
reverse intersystem crossing.
In the compounds that displayed TADF, delayed fluorescent
lifetimes decreased and fluorescence quantum yield under a
stream of nitrogen increased as carbazole substitution in the 3-
and 5-positions increased. OLEDs were fabricated using the
three compounds with the lowest estimated ΔEST values as the
emitters. Despite similar estimates of ΔEST, controlled largely
by stabilization of the S1 states, devices exhibited a range of
EQEmax from 6.1% to 24.4% at a luminance of 10 cd/m2,
increasing with increased donor substitution in the 3- and 5-
positions of the emitters. At a luminance of 1000 cd/m2, an
EQE of 9.2% was achieved with the best compound. This work
helps elucidate the importance of the regiochemistry of the
donor and acceptor in TADF compounds, beyond the
previously postulated ability of the donor to simply inductively
stabilize the LUMO on the acceptor.
REFERENCES
■
(1) Tang, C. W.; VanSlyke, S. A. Organic electroluminescent diodes.
Appl. Phys. Lett. 1987, 51, 913−915.
(2) Baldo, M. A.; O’Brien, D. F.; You, Y.; Shoustikov, A.; Sibley, S.;
Thompson, M. E.; Forrest, S. R. Highly efficient phosphorescent
emission from organic electroluminescent devices. Nature 1998, 395,
151−154.
(3) Fister, J. C.; Rank, D.; Harris, J. M. Delayed Fluorescence
Optical Thermometry. Anal. Chem. 1995, 67, 4269−4275.
(4) Salazar, F. A.; Fedorov, A.; Berberan-Santos, M. N. A study of
thermally activated delayed fluorescence in C60. Chem. Phys. Lett.
1997, 271, 361−366.
(5) Yang, Z.; Mao, Z.; Xie, Z.; Zhang, Y.; Liu, S.; Zhao, J.; Xu, J.;
Chi, Z.; Aldred, M. P. Recent advances in organic thermally activated
delayed fluorescence materials. Chem. Soc. Rev. 2017, 46, 915−1016.
(6) Hirata, S.; Sakai, Y.; Masui, K.; Tanaka, H.; Lee, S. Y.; Nomura,
H.; Nakamura, N.; Yasumatsu, M.; Nakanotani, H.; Zhang, Q.; Shizu,
K.; Miyazaki, H.; Adachi, C. Highly efficient blue electroluminescence
based on thermally activated delayed fluorescence. Nat. Mater. 2015,
14, 330−336.
(7) Zhang, Q.; Li, J.; Shizu, K.; Huang, S.; Hirata, S.; Miyazaki, H.;
Adachi, C. Design of Efficient Thermally Activated Delayed
Fluorescence Materials for Pure Blue Organic Light Emitting Diodes.
J. Am. Chem. Soc. 2012, 134, 14706−14709.
(8) Uoyama, H.; Goushi, K.; Shizu, K.; Nomura, H.; Adachi, C.
Highly efficient organic light-emitting diodes from delayed
fluorescence. Nature 2012, 492, 234−238.
(9) Adachi, C.; Tsutsui, T.; Saito, S. Confinement of charge carriers
and molecular excitons within 5-nm-thick emitter layer in organic
electroluminescent devices with a double heterostructure. Appl. Phys.
Lett. 1990, 57, 531−533.
(10) Adachi, C.; Tsutsui, T.; Saito, S. Blue light-emitting organic
electroluminescent devices. Appl. Phys. Lett. 1990, 56, 799−801.
(11) Kulkarni, A. P.; Tonzola, C. J.; Babel, A.; Jenekhe, S. A.
Electron Transport Materials for Organic Light-Emitting Diodes.
Chem. Mater. 2004, 16, 4556−4573.
(12) Hughes, G.; Bryce, M. R. Electron-transporting materials for
organic electroluminescent and electrophosphorescent devices. J.
Mater. Chem. 2005, 15, 94−107.
(13) Hamada, Y.; Adachi, C.; Tsutsui, T.; Saito, S. Blue-Light-
Emitting Organic Electroluminescent Devices with Oxadiazole Dimer
Dyes as an Emitter. Jpn. J. Appl. Phys. 1992, 31, 1812−1816.
(14) Shirota, Y.; Kageyama, H. Charge Carrier Transporting
Molecular Materials and Their Applications in Devices. Chem. Rev.
2007, 107, 953−1010.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.chemma-
ter.8b02632.
Oxidative and reductive cyclic voltammograms; absorp-
tion and photoluminescence spectra in cyclohexane,
toluene, dichloromethane, and chloroform; transient
photoluminescence in air-saturated toluene solution;
TGA curves; structures for compounds used in OLEDs;
1H, 13C{1H}, 13C{19F}, and 19NMR spectra (PDF)
(15) Leung, M.-k.; Yang, C.-C.; Lee, J.-H.; Tsai, H.-H.; Lin, C.-F.;
Huang, C.-Y.; Su, Y. O.; Chiu, C.-F. The Unusual Electrochemical
and Photophysical Behavior of 2,2’-Bis(1,3,4-oxadiazol-2-yl)biphenyls,
Effective Electron Transport Hosts for Phosphorescent Organic Light
Emitting Diodes. Org. Lett. 2007, 9, 235−238.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: Seth.Marder@chemistry.gatech.edu.
ORCID
Xiaoqing Zhang: 0000-0003-1930-623X
Canek Fuentes-Hernandez: 0000-0001-7958-3699
Stephen Barlow: 0000-0001-9059-9974
Bernard Kippelen: 0000-0002-8417-7051
Seth R. Marder: 0000-0001-6921-2536
(16) Tao, Y.; Wang, Q.; Yang, C.; Wang, Q.; Zhang, Z.; Zou, T.;
Qin, J.; Ma, D. A Simple Carbazole/Oxadiazole Hybrid Molecule: An
Excellent Bipolar Host for Green and Red Phosphorescent OLEDs.
Angew. Chem. 2008, 120, 8224−8227.
(17) Chaskar, A.; Chen, H.-F.; Wong, K.-T. Bipolar Host Materials:
A Chemical Approach for Highly Efficient Electrophosphorescent
Devices. Adv. Mater. 2011, 23, 3876−3895.
(18) Zhang, Y.; Zuniga, C.; Kim, S.-J.; Cai, D.; Barlow, S.; Salman,
Notes
́
S.; Coropceanu, V.; Bredas, J.-L.; Kippelen, B.; Marder, S. Polymers
The authors declare no competing financial interest.
with Carbazole-Oxadiazole Side Chains as Ambipolar Hosts for
Phosphorescent Light-Emitting Diodes. Chem. Mater. 2011, 23,
4002−4015.
ACKNOWLEDGMENTS
■
(19) Guan, M.; Chen, Z.; Bian, Z.; Liu, Z.; Gong, Z.; Baik, W.; Lee,
H.; Huang, C. The host materials containing carbazole and oxadiazole
fragment for red triplet emitter in organic light-emitting diodes. Org.
Electron. 2006, 7, 330−336.
This material is based upon work supported in part by the
Samsung Advanced Institute of Technology (SAIT) GRO
Program, and upon work supported by the U.S. Department of
Energy’s Office of Energy Efficiency and Renewable Energy
(EERE) under the Solid-State Lighting Program (Award
Number De-EE0008205).
̈
(20) Brunner, K.; van Dijken, A.; Borner, H.; Bastiaansen, J. J. A. M.;
Kiggen, N. M. M.; Langeveld, B. M. W. Carbazole Compounds as
Host Materials for Triplet Emitters in Organic Light-Emitting
I
DOI: 10.1021/acs.chemmater.8b02632
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
Diodes: Tuning the HOMO Level without Influencing the Triplet
Energy in Small Molecules. J. Am. Chem. Soc. 2004, 126, 6035−6042.
(21) Guan, M.; Qiang Bian, Z.; Feng Zhou, Y.; You Li, F.; Jun Li, Z.;
Hui Huang, C. High-performance blue electroluminescent devices
based on 2-(4-biphenylyl)-5-(4-carbazole-9-yl)phenyl-1,3,4-oxadia-
zole. Chem. Commun. 2003, 2708−2709.
(22) Justin Thomas, K. R.; Lin, J. T.; Tao, Y.-T.; Chuen, C.-H. New
Carbazole−Oxadiazole Dyads for Electroluminescent Devices:
Influence of Acceptor Substituents on Luminescent and Thermal
Properties. Chem. Mater. 2004, 16, 5437−5444.
(23) Linton, K. E.; Fisher, A. L.; Pearson, C.; Fox, M. A.; Palsson, L.-
O.; Bryce, M. R.; Petty, M. C. Colour tuning of blue electro-
luminescence using bipolar carbazole-oxadiazole molecules in single-
active-layer organic light emitting devices (OLEDs). J. Mater. Chem.
2012, 22, 11816−11825.
(37) Nobuyasu, R. S.; Ren, Z.; Griffiths, G. C.; Batsanov, A. S.; Data,
P.; Yan, S.; Monkman, A. P.; Bryce, M. R.; Dias, F. B. Rational Design
of TADF Polymers Using a Donor−Acceptor Monomer with
Enhanced TADF Efficiency Induced by the Energy Alignment of
Charge Transfer and Local Triplet Excited States. Adv. Opt. Mater.
2016, 4, 597−607.
(38) Dias, F. B.; Bourdakos, K. N.; Jankus, V.; Moss, K. C.;
Kamtekar, K. T.; Bhalla, V.; Santos, J.; Bryce, M. R.; Monkman, A. P.
Triplet Harvesting with 100% Efficiency by Way of Thermally
Activated Delayed Fluorescence in Charge Transfer OLED Emitters.
Adv. Mater. 2013, 25, 3707−3714.
(39) Etherington, M. K.; Gibson, J.; Higginbotham, H. F.; Penfold,
T. J.; Monkman, A. P. Revealing the spin−vibronic coupling
mechanism of thermally activated delayed fluorescence. Nat. Commun.
2016, 7, 13680.
(24) Tao, Y.; Wang, Q.; Yang, C.; Zhong, C.; Zhang, K.; Qin, J.; Ma,
D. Tuning the Optoelectronic Properties of Carbazole/Oxadiazole
Hybrids through Linkage Modes: Hosts for Highly Efficient Green
Electrophosphorescence. Adv. Funct. Mater. 2010, 20, 304−311.
(25) Wu, X.; Wang, L.; Hua, Y.; Wang, C.; Batsanov, A. S.; Bryce, M.
R. A carbazole−oxadiazole diad molecule for single-emitting-
component white organic light-emitting devices (WOLEDs).
Tetrahedron 2014, 70, 2015−2019.
(26) Lee, J.; Shizu, K.; Tanaka, H.; Nomura, H.; Yasuda, T.; Adachi,
C. Oxadiazole- and triazole-based highly-efficient thermally activated
delayed fluorescence emitters for organic light-emitting diodes. J.
Mater. Chem. C 2013, 1, 4599−4604.
(27) Luo, J.; Xie, G.; Gong, S.; Chen, T.; Yang, C. Creating a
thermally activated delayed fluorescence channel in a single polymer
system to enhance exciton utilization efficiency for bluish-green
electroluminescence. Chem. Commun. 2016, 52, 2292−2295.
(28) Zhang, Q.; Li, B.; Huang, S.; Nomura, H.; Tanaka, H.; Adachi,
C. Efficient blue organic light-emitting diodes employing thermally
activated delayed fluorescence. Nat. Photonics 2014, 8, 326−332.
(29) Wex, B.; Kaafarani, B. R. Perspective on carbazole-based
organic compounds as emitters and hosts in TADF applications. J.
Mater. Chem. C 2017, 5, 8622−8653.
(30) Zhang, D.; Cao, X.; Wu, Q.; Zhang, M.; Sun, N.; Zhang, X.;
Tao, Y. Purely organic materials for extremely simple all-TADF white
OLEDs: a new carbazole/oxadiazole hybrid material as a dual-role
non-doped light blue emitter and highly efficient orange host. J. Mater.
Chem. C 2018, 6, 3675−3682.
(40) Kim, D. Effect of Topology on the Singlet−Triplet Energy
Difference and their Natures: A Density Functional Theory Study of
Carbazolyl-Phthalonitrile Derivatives. Bull. Korean Chem. Soc. 2017,
38, 899−903.
(41) Geng, Y.; Cui, L.-S.; Kim, J. U.; Nakanotani, H.; Adachi, C.
Molecular Design for Blue Thermal Activated Delayed Fluorescence
Materials: Substitution Position Effect. Chem. Lett. 2017, 46, 1490−
1492.
(42) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J.
Ab Initio Calculation of Vibrational Absorption and Circular
Dichroism Spectra Using Density Functional Force Fields. J. Phys.
Chem. 1994, 98, 11623−11627.
(43) Frisch, M. J.; Trucks, G.W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson,
G. A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A.; Bloino, J.;
Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J.
V.; Izmaylov, A. F.; Sonnenberg, J. L.; Williams-Young, D.; Ding, F.;
Lipparini, F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson,
T.; Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.;
Liang, W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.;
Throssell, K.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.;
Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V.
N.; Keith, T.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell,
A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Millam, J. M.;
Klene, M.; Adamo, C.; Cammi, R.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Farkas, O.; Foresman, J. B.; Fox, D. J. Gaussian 09,
Revision A.02; Gaussian, Inc.: Wallingford, CT, 2016.
(44) Zhao, Y.; Truhlar, D. G. Hybrid Meta Density Functional
Theory Methods for Thermochemistry, Thermochemical Kinetics,
and Noncovalent Interactions: The MPW1B95 and MPWB1K
Models and Comparative Assessments for Hydrogen Bonding and
van der Waals Interactions. J. Phys. Chem. A 2004, 108, 6908−6918.
(45) Zou, L. H.; Mottweiler, J.; Priebbenow, D. L.; Wang, J.;
Stubenrauch, J. A.; Bolm, C. Mild Copper-Mediated Direct Oxidative
Cross-Coupling of 1,3,4-Oxadiazoles with Polyfluoroarenes by Using
Dioxygen as Oxidant. Chem. - Eur. J. 2013, 19, 3302−3305.
(46) Zuniga, C. A.; Abdallah, J.; Haske, W.; Zhang, Y.; Coropceanu,
I.; Barlow, S.; Kippelen, B.; Marder, S. R. Crosslinking Using Rapid
Thermal Processing for the Fabrication of Efficient Solution-
Processed Phosphorescent Organic Light-Emitting Diodes. Adv.
Mater. 2013, 25, 1739−1744.
(31) Mei, L.; Hu, J.; Cao, X.; Wang, F.; Zheng, C.; Tao, Y.; Zhang,
X.; Huang, W. The inductive-effect of electron withdrawing
trifluoromethyl for thermally activated delayed fluorescence: tunable
emission from tetra- to penta-carbazole in solution processed blue
OLEDs. Chem. Commun. 2015, 51, 13024−13027.
(32) Lee, K.; Kim, D. Local-Excitation versus Charge-Transfer
Characters in the Triplet State: Theoretical Insight into the Singlet−
Triplet Energy Differences of Carbazolyl-Phthalonitrile-Based Ther-
mally Activated Delayed Fluorescence Materials. J. Phys. Chem. C
2016, 120, 28330−28336.
́
(33) Kim, D.; Coropceanu, V.; Bredas, J.-L. Design of Efficient
Ambipolar Host Materials for Organic Blue Electrophosphorescence:
Theoretical Characterization of Hosts Based on Carbazole Deriva-
tives. J. Am. Chem. Soc. 2011, 133, 17895−17900.
́
(34) Salman, S.; Kim, D.; Coropceanu, V.; Bredas, J.-L. Theoretical
Investigation of Triscarbazole Derivatives As Host Materials for Blue
Electrophosphorescence: Effects of Topology. Chem. Mater. 2011, 23,
5223−5230.
(47) Karon, K.; Lapkowski, M. Carbazole electrochemistry: a short
review. J. Solid State Electrochem. 2015, 19, 2601−2610.
(48) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.;
Springer: New York, 2010.
(49) Noda, H.; Nakanotani, H.; Adachi, C. Excited state engineering
for efficient reverse intersystem crossing. Sci. Adv. 2018, 4, eaao6910.
(50) Zhang, X.; Fuentes-Hernandez, C.; Zhang, Y.; Cooper, M. W.;
Barlow, S.; Marder, S. R.; Kippelen, B. High performance blue-
emitting organic light-emitting diodes from thermally activated
delayed fluorescence: A guest/host ratio study. J. Appl. Phys. 2018,
124, 055501.
́
(35) Kim, D.; Zhu, L.; Bredas, J.-L. Electronic Structure of
Carbazole-Based Phosphine Oxides as Ambipolar Host Materials for
Deep Blue Electrophosphorescence: A Density Functional Theory
Study. Chem. Mater. 2012, 24, 2604−2610.
́
(36) Samanta, P. K.; Kim, D.; Coropceanu, V.; Bredas, J.-L. Up-
Conversion Intersystem Crossing Rates in Organic Emitters for
Thermally Activated Delayed Fluorescence: Impact of the Nature of
Singlet vs Triplet Excited States. J. Am. Chem. Soc. 2017, 139, 4042−
4051.
J
DOI: 10.1021/acs.chemmater.8b02632
Chem. Mater. XXXX, XXX, XXX−XXX

Chemistry of Materials
Article
(51) Lee, D. R.; Choi, J. M.; Lee, C. W.; Lee, J. Y. Ideal Molecular
Design of Blue Thermally Activated Delayed Fluorescent Emitter for
High Efficiency, Small Singlet−Triplet Energy Splitting, Low
Efficiency Roll-Off, and Long Lifetime. ACS Appl. Mater. Interfaces
2016, 8, 23190−23196.
(52) We note that EQEmax for the 2,3,4,6CzDPO and
2,3,4,5,6CzDPO device is larger than would be expected based on
the PLQY for the corresponding film, assuming isotropic emission,
and a 20% outcoupling efficiency. Such apparent discrepancies have
been seen in other TADF devices (for example, in ref 49) and may
arise from anisotropic fluorophore orientation, from improved
outcoupling, or from underestimated PLQY values due to partial
quenching by traces of O2.
K
DOI: 10.1021/acs.chemmater.8b02632
Chem. Mater. XXXX, XXX, XXX−XXX