Hybrid heterocycle-containing electron-transport materials synthesized by regioselective Suzuki cross-coupling reactions for highly efficient phosphorescent OLEDs with unprecedented low operating voltage

Article abstract of DOI:10.1021/cm303075m

A series of hybrid heterocyle-containing electron-transport materials (ETMs) with different pyridine substitution positions in a single molecule were successfully synthesized by regioselective sequential palladium-catalyzed Suzuki cross-coupling reactions. B3LYP calculations show that the experimental regioselectivity is consistent with the trends in calculated bond dissociation energies of the carbon-halogen (C-X) bonds of the reactants and intermediates. The two carbon species associated with C-N and C-C can be detected by X-ray photoelectron spectroscopy. Extremely low turn-on voltages (V_on) of 2.1 V for electroluminescence, which are readily 0.2-0.3 V lower than the minimum value of the emitted photon energy (hv)/e, were experimentally achieved by utilizing the developed ETMs as an electron-transport and hole/exciton-block layer for the classical fac-tris(2-phenylpyridine) iridium (Ir(PPy) ₃)-based green phosphorescent organic light-emitting devices (OLEDs). In addition, hitherto the lowest operating voltages of 2.39, 2.72, and 3.88 V for 100, 1000, and 10 000 cd m^-2 were achieved with simultaneously improved external quantum efficiency (η_ext) to give a high power efficiency, and the operating voltage for 100 cd m^-2 is already corresponding to the value of hv/e. Atomic force microscopy measurements reveal morphological fine structure of the thin film samples of ETMs with a pyrimidine core, which possibly contributes to the low driving voltages that were achieved for the devices that are based on these ETMs. These findings indicate that aside from the ultralow operating voltage well-balanced carriers can be also achieved with a finely synthesized hybrid heterocyle-containing ETM as a nondoped electron-transport layer, and the B3LYP calculation is an effective method to predict the regioselective cross-coupling reactions for fine design of such an ETM.

Full text of DOI:10.1021/cm303075m

Article
pubs.acs.org/cm
Hybrid Heterocycle-Containing Electron-Transport Materials
Synthesized by Regioselective Suzuki Cross-Coupling Reactions for
Highly Efficient Phosphorescent OLEDs with Unprecedented Low
Operating Voltage
Ming Liu,† Shi-Jian Su,*,† Min-Cherl Jung,‡ Yabing Qi,*,‡ Wei-Ming Zhao,§ and Junji Kido∥
†State Key Laboratory of Luminescent Materials and Devices and Institute of Polymer Optoelectronic Materials and Devices, South
China University of Technology, Guangzhou 510640, China
‡Energy Materials and Surface Sciences Unit, Okinawa Institute of Science and Technology Graduate University, 1919-1 Tancha,
Onna-son, Kunigami-gun, Okinawa 904-0495, Japan
§Dongguan LITEWELL Technology Incorporation, Dongguan 523656, China
∥Department of Organic Device Engineering, Graduate School of Science and Engineering, Yamagata University, 4-3-16 Jonan,
Yonezawa, Yamagata 992-8510, Japan
ABSTRACT: A series of hybrid heterocyle-containing electron-transport materials (ETMs) with different pyridine substitution
positions in a single molecule were successfully synthesized by regioselective sequential palladium-catalyzed Suzuki cross-
coupling reactions. B3LYP calculations show that the experimental regioselectivity is consistent with the trends in calculated
bond dissociation energies of the carbon−halogen (C−X) bonds of the reactants and intermediates. The two carbon species
associated with C−N and C−C can be detected by X-ray photoelectron spectroscopy. Extremely low turn-on voltages (Von) of
2.1 V for electroluminescence, which are readily 0.2−0.3 V lower than the minimum value of the emitted photon energy (hv)/e,
were experimentally achieved by utilizing the developed ETMs as an electron-transport and hole/exciton-block layer for the
classical fac-tris(2-phenylpyridine) iridium (Ir(PPy)3)-based green phosphorescent organic light-emitting devices (OLEDs). In
addition, hitherto the lowest operating voltages of 2.39, 2.72, and 3.88 V for 100, 1000, and 10 000 cd m−2 were achieved with
simultaneously improved external quantum efficiency (ηext) to give a high power efficiency, and the operating voltage for 100 cd
m−2 is already corresponding to the value of hv/e. Atomic force microscopy measurements reveal morphological fine structure of
the thin film samples of ETMs with a pyrimidine core, which possibly contributes to the low driving voltages that were achieved
for the devices that are based on these ETMs. These findings indicate that aside from the ultralow operating voltage well-
balanced carriers can be also achieved with a finely synthesized hybrid heterocyle-containing ETM as a nondoped electron-
transport layer, and the B3LYP calculation is an effective method to predict the regioselective cross-coupling reactions for fine
design of such an ETM.
KEYWORDS: regioselective, Suzuki cross-coupling reaction, phosphorescent OLEDs, electron-transport material, low operating voltage
voltage generally gives higher power conversion efficiency. As
such, lowering operating voltage is also crucially important to
reduce their electrical power consumption.3 It has been
reported that OLEDs with p-doped hole-transport layer
(HTL) and n-doped electron-transport layer (ETL), called p-
i-n OLEDs, generally exhibit low driving voltage as well as high
power efficiency.4 However, carrier and exciton block layers are
absolutely essential to prevent exciton quenching in the
INTRODUCTION
■
Organic light-emitting devices (OLEDs) have become of
considerable commercialization interest in recent years for
their applications as eco-friendly solid-state lighting and next-
generation full-color flat-panel displays. In the last two decades,
many studies have focused on improving external quantum
efficiency (ηext) of OLEDs,1 and phosphorescent OLEDs have
attracted tremendous attention since nearly 100% internal
quantum efficiency of electroluminescence is readily achieved
by harvesting both electro-generated singlet and triplet
excitons.2 It is well-known that at a given ηext lower operating
Received: March 25, 2012
Published: September 27, 2012
© 2012 American Chemical Society
3817
dx.doi.org/10.1021/cm303075m | Chem. Mater. 2012, 24, 3817−3827

Chemistry of Materials
Article
Thermogravimetric analysis (TGA) was undertaken using a SEIKO
EXSTAR 6000 TG/DTA 6200 unit under nitrogen atmosphere at a
heating rate of 10 °C min−1. UV−vis absorption spectra were
measured using a Shimadzu UV-3150 UV−vis-NIR spectrophotom-
eter. Steady-state photoluminescent spectra were obtained at room
temperature using a FluoroMax-2 (Jobin−Yvon−Spex) luminescence
spectrometer. Ionization potentials were determined by atmospheric
ultraviolet photoelectron spectroscopy (Rikken Keiki AC-3). The
ETMs were transported in sealed containers that were backfilled with
dry nitrogen gas to Okinawa Institute of Science and Technology
Graduate University for further analysis. The ETMs were kept in a dry
N2 storage box prior to use to minimize exposure to air. A two-
chamber ultrahigh vacuum (UHV) system (Omicron NanoTechnol-
ogy GmbH) was used to perform in situ X-ray photoelectron
spectroscopy (XPS). The ETMs were introduced to the preparation
chamber of the UHV system in quartz crucibles held by tungsten wires
for resistive heating. A quartz crystal microbalance was used to
monitor evaporation rate and film thickness. After thorough
outgassing, thin film samples (thickness ∼30 nm) were prepared by
evaporating ETMs on precleaned indium−tin oxide (ITO) glass
substrates (Geomatec Co. Ltd., 5 Ω/sq) at evaporation rates of ∼0.2
Å/s.
emissive layer (EML) by the p and n dopants, and the internal
energy barriers to overcome the carrier and exciton block layers
may induce voltage drop and thus higher operating voltage.
More importantly, the approach of using n dopant-like alkaline
metals might be hampered by stability issues and leads to
manufacturing difficulties. It is thus of interest to utilize a single
carrier transport layer instead of p- or n-doped layer and
carrier/exciton block layer. In the past years, a series of
multifunctional pyridine-containing electron-transport materials
(ETMs) have been developed as an ETL as well as a hole/
exciton block layer for the phosphorescent OLEDs, leading to a
simplified device structure and an improved ηext.7 Most
recently, operating voltages that are readily lower than those
of the p-i-n OLEDs were also achieved by utilizing 1,3,5-
triazine-containing ETMs.7d A general character for all these
ETMs is that they have pyridine rings on the periphery of the
molecule and the substituted positions of the pyridine rings in a
single molecule are identical. Since the nitrogen atom position
of the substituted pyridine ring was found to be a key factor in
determining the electron injection and transport property and
thus the OLED performance,8 it is of interest to synthesize
pyridine-containing compounds with different pyridine sub-
stitution positions in a single molecule to further tune the
property of the pyridine-containing ETMs. Besides this,
simultaneous introduction of other heterocycles to give hybrid
heterocycle-containing compounds is also an attractive
approach to tune their property.
A general synthetic method for the pyridine-containing
ETMs is the Suzuki cross-coupling reaction. To obtain
pyridine-containing compounds with different substituted
positions in a single molecule, selective sequential cross-
coupling reactions involving multihalogenated aromatics should
be an effective approach. For the Suzuki cross-coupling
reactions, the nature of the halogen that is being displaced
has been reported as one source of selectivity. In this article, we
report on pyridine-containing ETMs with different pyridine
substitution positions in a single molecule that are successfully
synthesized by regioselective palladium-catalyzed Suzuki cross-
coupling reactions of organoboron compounds and aryl halides,
although the halogens being displaced are identical for the
reactants and the intermediates. In addition, pyrimidine ring
was simultaneously introduced to give hybrid heterocycle-
containing ETMs. Bond dissociation energies (BDEs) of the
aryl halides were calculated with B3LYP, and the trends in the
calculated BDEs are consistent with the experimental
regioselectivity. As proven by the experimental and calculation
results, their energy levels can be finely tuned by the different
pyridine substitution positions as well as the introduction of
pyrimidine. In addition, their structure−property relationships
were thoroughly studied by thermal analysis (TA), X-ray
photoelectron spectroscopy (XPS) analysis, atomic force
microscopy (AFM), and device application. Simultaneous
improvement of ηext and unprecedented low operating voltage
were achieved for the classical fac-tris(2-phenylpyridine)
iridium (Ir(PPy)3) based green phosphorescent OLEDs by
utilizing the developed hybrid heterocycle-containing ETMs.
An unmonochromated Mg Kα (1253. Six eV) X-ray source and
EA125 energy analyzer with single channeltron were used for XPS
measurements. The pass energy of electron analyzer and the energy
resolution of XPS measurements are 20 and 0.9 eV, respectively. The
chemical states of all samples were obtained for the C 1s and N 1s
regions. The binding energies were calibrated with respect to the Au 4f
level (at a binding energy of 84.0 eV).9 Following the XPS
7/2
measurements, the samples were taken out of the UHV system to
investigate their surface morphology properties. Topography images of
the samples were acquired in air by atomic force microscopy
(ASYLUM Research MFP 3D) in the tapping mode using Al coated
silicon cantilevers with spring constants of 48 N/m at resonant
frequency of 190 kHz.
Synthesis. All the reagents were purchased from Sigma-Aldrich
and were used without further purification. 3-(3-(4,4,5,5-Tetramethyl-
1,3,2-dioxaborolan-2-yl)phenyl)pyridine (15) was synthesized accord-
ing to the literature procedures.7b 1−6 were synthesized according to
regioselective palladium-catalyzed Suzuki cross-coupling reactions of
organoboron compounds and aryl halides.
1,3-Dibromo-5-pyrid-3-yl-benzene (9). A mixture of 1,3,5-
tribromobenzene (7) (18.9 g, 60 mmol), 3-(4,4,5,5-tetramethyl-
[1,3,2]dioxaborolanyl)-pyridine (8) (12.3 g, 60 mmol), tetrakis-
(triphenylphosphine) palladium(0) (Pd(PPh3)4) (0.693 g, 0.6
mmol), and 2 M Na2CO3 (70 mL) in toluene (200 mL) and ethanol
(67 mL) was stirred at 85 °C for 12 h under nitrogen atmosphere.
After cooling to room temperature, the mixture was poured into water
and then extracted with toluene. The combined organic phase was
washed with brine and dried over MgSO4. The subjection of the crude
mixture to silica gel chromatography (eluent: chloroform/ethyl acetate
= 1/1) afforded 9 (9.33 g, 50%) as white powder, accompanied by a
1
16% yield of 1-bromo-3,5-dipyrid-3-yl-benzene (18) (2.93 g). 9 H
NMR (270 MHz, CDCl3, δ): 8.80 (d, J = 2.4 Hz, 1H), 8.66 (dd, J =
1.6 and 5.0 Hz, 1H), 7.85−7.81 (m, 1H), 7.72 (t, J = 1.6 Hz, 1H), 7.66
1
(d, J = 1.6 Hz, 2H), 7.42−7.37 (m, 1H). 18 H NMR (270 MHz,
CDCl3, δ): 8.88 (d, J = 2.4 Hz, 2H), 8.67 (dd, J = 1.0 and 4.8 Hz, 2H),
7.93−7.89 (m, 2H), 7.77 (t, J = 1.0 Hz, 2H), 7.69 (s, 1H), 7.45−7.40
(m, 2H).
1,3-Bis(3,5-dichlorophenyl)-5-pyrid-3-yl-benzene (11). A mixture
of 9 (1.79 g, 5.72 mmol), 3,5-dichlorophenyl boronic acid (10) (2.62
g, 13.7 mmol), Pd(PPh3)4 (0.661 g, 0.572 mmol), and 2 M K2CO3 (50
mL) in toluene (100 mL) and ethanol (40 mL) was stirred at 85 °C
for 22 h under nitrogen atmosphere. After cooling to room
temperature, the mixture was poured into water and then extracted
with chloroform. The combined organic phase was washed with brine
and dried over MgSO4. The subjection of the crude mixture to silica
gel chromatography (eluent: chloroform) afforded 11 (1.69 g, 66%) as
EXPERIMENTAL SECTION
■
1
General. H and 13C NMR spectra were recorded on Varian 500
(500 MHz) spectrometer. Mass spectra were obtained using a JEOL
JMS-K9 mass spectrometer. Differential scanning calorimetry (DSC)
was performed using a Perkin-Elmer Diamond DSC Pyris instrument
under nitrogen atmosphere at a heating rate of 10 °C min−1.
1
white powder. H NMR (400 MHz, CDCl3, δ): 8.933 (d, J = 2.3 Hz,
3818
dx.doi.org/10.1021/cm303075m | Chem. Mater. 2012, 24, 3817−3827

Chemistry of Materials
Article
1H), 8.682 (d, J = 5.0 Hz, 1H), 7.981−7.962 (m, 1H), 7.740 (d, J = 1.8
Hz, 2H), 7.701 (s, 1H), 7.552 (s, 4H), 7.458−7.422 (m, 3H).
1,3-Bis(3,5-dipyrid-3-ylphenyl)-5-pyrid-3-yl-benzene (1). A mix-
ture of 11 (1.69 g, 3.80 mmol), 8 (4.05 g, 19.8 mmol),
tris(dibenzylideneacetone) dipalladium(0) (Pd2(dba)3) (139 mg,
0.152 mmol), tricyclohexylphosphine (PCy3) (102 mg, 0.365
mmol), and K3PO4 (15.5 g, 73 mmol) in 1,4-dioxane (300 mL) and
water (70 mL) was stirred at 90 °C for 24 h under nitrogen
atmosphere. After cooling to room temperature, the mixture was
poured into water and then extracted with chloroform. The combined
organic phase was washed with brine and dried over MgSO4. The
subjection of the crude mixture to silica gel chromatography (eluent:
chloroform/methanol = 15/1) afforded 1 (1.92 g, 82%) as white
powder. 1H NMR (500 MHz, CDCl3, δ): 9.008−9.001 (m, 1H),
8.980−8.973 (m, 4H), 8.683−8.659 (m, 5H), 8.054−7.984 (m, 6H),
7.924−7.919 (m, 6H), 7.820−7.811 (m, 2H), 7.466−7.427 (m, 5H).
13C NMR (500 MHz, CDCl3, δ): 149.118, 148.465, 148.435, 142.549,
142.373, 139.764, 139.699, 136.201, 136.117, 134.736, 134.702,
126.199, 126.085, 125.944, 125.677, 123.769. EIMS (m/z): [M+]
calcd, 815.72; found, 816. Anal. Calc. for C43H29N5 (%): C, 83.88; H,
4.75; N, 11.37. Found: C, 83.65; H, 4.82; N, 11.48.
4,6-Bis(3,5-di(pyridin-3-yl)phenyl)-2-(pyridin-3-yl)pyrimidine (3).
3 was synthesized in a manner similar to that of 1. Yield: 87%. H
1
NMR (500 MHz, CDCl3, δ): 9.959 (d, J = 2.0 Hz, 1H), 9.030 (d, J =
2.0 Hz, 4H), 8.991−8.967 (m, 1H), 8.799 (dd, J = 2.0 and 5.0 Hz,
1H), 8.717 (dd, J = 2.0 and 5.0 Hz, 4H), 8.523 (d, J = 1.5 Hz, 4H),
8.246 (s, 1H), 8.079−8.055 (m, 4H), 7.968 (t, J = 1.8 Hz, 2H), 7.536−
7.509 (m, 1H), 7.505−7.478 (m, 4H). 13C NMR (500 MHz, CDCl3,
δ): 164.671, 163.366, 151.785, 150.316, 149.382, 148.504, 139.967,
138.830, 135.847, 134.722, 133.078, 128.779, 125.857, 123.842,
123.522, 111.357. EIMS (m/z): [M+] calcd, 617.70; found, 618.
Anal. Calc. for C41H27N7 (%): C, 79.72; H, 4.41; N, 15.87. Found: C,
79.54; H, 4.52; N, 15.65.
4,6-Bis(3,5-di(pyridin-4-yl)phenyl)-2-(pyridin-3-yl)pyrimidine (4).
1
4 was synthesized in a similar manner to that of 1. Yield: 82%. H
NMR (500 MHz, CDCl3, δ): 9.980 (d, J = 2.5 Hz, 1H), 8.989−8.965
(m, 1H), 8.823−8.793 (m, 9H), 8.582 (d, J = 2.0 Hz, 4H), 8.226 (s,
1H), 8.057 (t, J = 1.8 Hz, 2H), 7.684 (dd, J = 2.0 and 4.5 Hz, 8H),
7.552−7.527 (m, 1H). 13C NMR (500 MHz, CDCl3, δ): 164.625,
163.488, 151.907, 150.698, 150.373, 147.303, 140.490, 138.994,
135.805, 132.959, 128.512, 126.532, 123.553, 121.893, 111.445.
EIMS (m/z): [M+] calcd, 617.70; found, 618. Anal. Calc. for
C41H27N7 (%): C, 79.72; H, 4.41; N, 15.87. Found: C, 79.58; H,
4.55; N, 15.60.
1,3-Bis(3,5-dipyrid-4-ylphenyl)-5-pyrid-3-yl-benzene (2). 2 was
1
synthesized in a similar manner to that of 1. Yield: 41%. H NMR
4,6-Bis(3,5-di(pyridin-3-yl)phenyl)-2-(3-(pyridin-3-yl)phenyl)-
(500 MHz, CDCl3, δ): 9.012 (d, J = 2.5 Hz, 1H), 8.767−8.755 (m,
8H), 8.690 (dd, J = 1.5 and 5.0 Hz, 1H), 8.047−8.025 (m, 1H),
8.001−7.998 (m, 4H), 7.954 (t, J = 1.5 Hz, 1H), 7.931−7.928 (m,
4H), 7.652−7.640 (m, 8H), 7.469−7.444 (m, 1H). 13C NMR (500
MHz, CDCl3, δ): 150.586, 149.285, 148.542, 147.565, 142.667,
142.255, 140.287, 139.871, 136.037, 134.641, 126.749, 126.165,
126.081, 125.391, 123.746, 121.854. EIMS (m/z): [M+] calcd,
815.72; found, 816. Anal. Calc. for C43H29N5 (%): C, 83.88; H,
4.75; N, 11.37. Found: C, 83.60; H, 4.68; N, 11.57.
pyrimidine (5). 5 was synthesized in a similar manner to that of 1.
1
Yield: 60%. H NMR (500 MHz, CDCl3, δ): 9.035−9.030 (m, 4H),
8.995−8.990 (m, 1H), 8.949 (t, J = 1.8 Hz, 1H), 8.801−8.780 (m,
1H), 8.710 (dd, J = 1.5 and 4.5 Hz, 4H), 8.644 (dd, J = 2.0 and 5.0 Hz,
1H), 8.513 (d, J = 1.5 Hz, 4H), 8.205 (s, 1H), 8.075−8.051 (m, 4H),
8.036−8.013 (m, 1H), 7.959 (t, J = 1.8 Hz, 2H), 7.799−7.778 (m,
1H), 7.713 (t, J = 7.5 Hz, 1H), 7.499−7.472 (m, 4H), 7.444−7.417
(m, 1H). 13C NMR (500 MHz, CDCl3, δ): 164.947, 149.566, 148.983,
148.738, 148.670, 140.125, 139.461, 138.759, 138.534, 136.753,
136.143, 134.949, 134.762, 130.054, 129.761, 128.845, 128.590,
127.495, 126.148, 124.069, 123.920, 111.618. EIMS (m/z): [M+]
calcd, 693.80; found, 694. Anal. Calc. for C47H31N7 (%): C, 81.36; H,
4.50; N, 14.13. Found: C, 81.22; H, 4.57; N, 13.95.
2-Chloro-4,6-bis(3,5-dichlorophenyl)pyrimidine (14). A mixture of
2,4,6-trichloropyrimidine (13) (5.50 g, 30 mmol), 10 (12.0 g, 63
mmol), bis(triphenylphosphine)palladium(II) chloride
(PdCl2(PPh3)2) (842 mg, 1.2 mmol), and 2 M K2CO3 (60 mL) in
acetonitrile (150 mL) was stirred at 50 °C for 20 h under nitrogen
atmosphere. After cooling to room temperature, the mixture was
poured into water and then extracted with chloroform. The combined
organic phase was washed with brine and dried over MgSO4. The
subjection of the crude mixture to silica gel chromatography (eluent:
chloroform/n-hexane = 2/1) afforded 14 (11.2 g, 92%) as white
4,6-Bis(3,5-di(pyridin-4-yl)phenyl)-2-(3-(pyridin-3-yl)phenyl)-
pyrimidine (6). 6 was synthesized in a manner similar to that of 1.
1
Yield: 74%. H NMR (500 MHz, CDCl3, δ): 9.011−9.007 (m, 1H),
8.964 (s, 1H), 8.791 (dd, J = 1.5 and 4.5 Hz, 8H), 8.660 (dd, J = 1.5
and 5.0 Hz, 1H), 8.572 (d, J = 2.0 Hz, 4H), 8.183 (s, 1H), 8.054−
8.021 (m, 3H), 7.810 (d, J = 7.5 Hz, 1H), 7.730 (t, J = 7.5 Hz, 1H),
7.683 (dd, J = 1.5 and 4.5 Hz, 8H), 7.445−7.420 (m, 1H). EIMS (m/
z): [M+] calcd, 693.80; found, 694. Anal. Calc. for C47H31N7 (%): C,
81.36; H, 4.50; N, 14.13. Found: C, 81.19; H, 4.59; N, 14.02.
Computations. BDEs of carbon−halogen (C−X) bonds of aryl
halides were calculated using the Gaussian suite of programs (Gaussian
03W) with UB3LYP/6-31G(d) method. BDE is defined as the change
in enthalpy, ΔH, for homolysis of the C−X bond.
1
powder. H NMR (500 MHz, CDCl3, δ): 8.041 (d, J = 2.0 Hz, 4H),
7.909 (s, 1H), 7.565 (t, J = 1.8 Hz, 2H). EIMS (m/z): [M+] calcd,
404.51; found, 404.
2-Pyrid-3-yl-4,6-bis(3,5-dichlorophenyl)pyrimidine (16). A mixture
of 14 (4.05 g, 10 mmol), 8 (2.26 g, 11 mmol), PdCl2(PPh3)2 (140 mg,
0.2 mmol), and 2 M K2CO3 (50 mL) in 1,4-dioxane (120 mL) was
stirred at 100 °C for 18 h under nitrogen atmosphere. After cooling to
room temperature, the mixture was poured into water and then
extracted with chloroform. The combined organic phase was washed
with brine and dried over MgSO4. The subjection of the crude mixture
to silica gel chromatography (eluent: chloroform) afforded 16 (2.84 g,
64%) as white powder. 1H NMR (500 MHz, CDCl3, δ): 9.872 (d, J =
1.5 Hz, 1H), 8.915−8.891 (m, 1H), 8.796 (dd, J = 1.5 and 4.5 Hz,
1H), 8.156 (d, J = 2.0 Hz, 4H), 7.938 (s, 1H), 7.572 (t, J = 1.8 Hz,
2H), 7.521−7.495 (m, 1H). EIMS (m/z): [M+] calcd, 447.14; found,
448.
A−X(g) → A(g) + X(g)
(1)
Density functional theory (DFT) calculations were also performed
for the calculation of HOMO, LUMO, and triplet energy levels by
using the Gaussian suite of programs (Gaussian 03W). The ground
state (S0) structures were optimized with B3LYP/6-31G(d) method,
and the single-point energies were calculated with B3LYP/6-
311+G(d,p) method. In comparison, the lowest-energy triplet excited
state (T1) structure optimization and single-point energy calculation
were performed with UB3LYP/6-31G(d) and UB3LYP/6-311+G(d,p)
methods, respectively.
Device Fabrication and Characterization. OLEDs were grown
on a glass substrate precoated with a ∼110 nm thick layer of ITO
having a sheet resistance of 15 Ω/square. The substrates were cleaned
with ultrapurified water and organic solvents and then dry-cleaned for
20 min by exposure to an UV−ozone ambient. To improve the hole
injection from the anode, poly(arylene amine ether sulfone)-
containing tetraphenylbenzidine (TPDPES) doped with 10% (by
weight) tris(4-bromophenyl) aminium hexachloroantimonate
(TBPAH) was spun onto the precleaned substrate from its
4,6-Bis(3,5-dichlorophenyl)-2-(3-(pyridin-3-yl)phenyl)pyrimidine
(17). 17 was synthesized in a manner similar to that of 16. Yield: 67%.
1H NMR (500 MHz, CDCl3, δ): 8.993 (d, J = 2.0 Hz, 1H), 8.828 (t, J
= 1.8 Hz, 1H), 8.697 (dd, J = 1.5 and 7.5 Hz, 1H), 8.672−8.659 (m,
1H), 8.130−8.113 (m, 4H), 8.032−8.008 (m, 1H), 7.878 (s, 1H),
7.769 (dd, J = 1.5 and 7.5 Hz, 1H), 7.684 (t, J = 7.5 Hz, 1H), 7.554−
7.543 (m, 1H), 7.462−7.437 (m, 1H). 13C NMR (500 MHz, CDCl3,
δ): 164.709, 162.897, 148.798, 148.504, 139.822, 138.395, 138.010,
136.427, 135.927, 134.577, 130.961, 130.060, 129.481, 128.363,
127.127, 125.780, 123.686, 110.716. EIMS (m/z): [M+] calcd,
523.24; found, 524.
3819
dx.doi.org/10.1021/cm303075m | Chem. Mater. 2012, 24, 3817−3827

Chemistry of Materials
Article
Scheme 1. Synthetic Routes of Pyridine-Containing Compounds 1 and 2. i. Pd(PPh3)4, 2M K2CO3, or Na2CO3, Toluene/
Ethanol; ii. Pd2(dba)3, PCy3, K3PO4, 1,4-Dioxane/Water
Scheme 2. Synthetic Routes of Pyridine- and Pyrimidine-Containing Compounds 3−6. i. PdCl2(PPh3)2, K2CO3, Acetonitrile/
Water; ii. PdCl2(PPh3)2, K2CO3, Dioxane/Water; iii. Pd2(dba)3, PCy3, K3PO4, 1,4-Dioxane/Water
dichloroethane solution to form a 20-nm-thick polymer buffer layer.7
Then, a 30-nm-thick HTL of 1,1-bis[4-[N,N-di(p-tolyl)-amino]-
phenyl]cyclohexane (TAPC) was deposited. Next, 8% (by weight)
Ir(PPy)3 was codeposited with N,N′-dicarbazolyl-4,4′-biphenyl (CBP)
to form a 10-nm-thick EML. Finally, a 50-nm-thick ETL of 1−6 was
deposited to block holes and to confine excitons in the emissive zone.
Cathodes consisting of a 0.5-nm-thick layer of LiF followed by a 100-
nm-thick layer of Al were patterned using a shadow mask with an array
of 2 mm × 2 mm openings. The electroluminescent (EL) spectra were
taken by an optical multichannel analyzer, Hamamatsu PMA 11. The
3820
dx.doi.org/10.1021/cm303075m | Chem. Mater. 2012, 24, 3817−3827

Chemistry of Materials
Article
current density and luminance versus driving voltage characteristics
were measured by Keithley source-measure unit 2400 and Konica
Minolta chroma meter CS-200, respectively. External quantum
efficiencies were calculated from the luminance, current density, and
EL spectrum, assuming a Lambertian distribution.
Although G3B3 was reported to be highly accurate and also has
the fewest convergence problem, it is practically limited to only
10 heavy atoms (non-hydrogen atoms) and might not be
applicable for most of the current reactants and intermediates.14
Here, B3LYP was used to determine BDEs of the C−X bonds
for the reactants and the intermediates.15 Calculations
demonstrate that 7 has a C−Br BDE of 82.1 kcal/mol, which
is 0.5 kcal/mol lower than that of 9 coupled with a pyridine ring
(Chart 1). It is predicted that 8 reacts primarily with 7 rather
RESULTS AND DISCUSSION
■
Synthesis. The developed ETMs are comprised of benzene,
pyridine, and pyrimidine as building blocks, and they are
combined with each other at their meta-positions to break their
π conjugation. All the nitrogen atoms of pyridine rings are at
the meta-position (refer to the combination sites) for 1, 3, and
5. In contrast, one nitrogen atom is at the meta-position and the
others are at the para-position for 2, 4, and 6. Different from 1
and 2, there is a pyrimidine ring in the center of 3−6, and 5 and
6 are structurally different from 3 and 4 in one more phenyl
inserted at the α-position to two nitrogen atoms of the central
pyrimidine, respectively.
Chart 1. Carbon−halogen (C−X) bond dissociation energies
(BDEs) (kcal/mol) of reactants and intermediates calculated
by B3LYP
As shown in Scheme 1, the cross-coupling reaction of 1,3,5-
tribromobenzene 7 with 1 equiv of 3-pyridine boronate ester 8
in the presence of catalytic amounts of Pd(PPh3)4 and excess
Na2CO3 in a toluene/ethanol/water solvent system at 80 °C
led to a 50% yield of 1,3-dibromo-5-pyrid-3-yl-benzene 9,
accompanied by a 16% yield of two-substituted byproduct of 1-
bromo-3,5-dipyrid-3-yl-benzene 18. Reaction of 9 with 3,5-
dichlorophenyl boronic acid 10 led to a 66% yield of
tetrachloride 11. 11 was further coupled with an excess of 8,
in the presence of more effective catalyst system of Pd2(dba)3
and PCy3, and excess K3PO4 in a 1,4-dioxane/water solvent
system to afford 1 in a yield of 82%.10 Similarly, 2 was
synthesized in a similar manner to that of 1 in a yield of 41% by
coupling with 4-pyridine boronate ester 12.
Different from 1 and 2, there is a pyrimidine ring in the
center of 3−6, started from 2,4,6-trichloropyrimidine 13
(Scheme 2). Cross-coupling reaction of 13 with two equivalent
of 10 in the presence of catalytic amounts of PdCl2(PPh3)2 and
excess K2CO3 in an acetonitrile/water solvent system at 50 °C
led to a 92% yield of 2-chloro-4,6-bis(3,5-dichlorophenyl)-
pyrimidine 14. Reaction of 14 with 8 in a 1,4-dioxane/water
solvent system at an elevated temperature led to a 64% yield of
tetrachloride 16. Finally, 16 was coupled with an excess of
pyridine boronate esters 8 and 12 in a similar manner to that of
1 and 2 to afford 3 in a yield of 87% and 4 in a yield of 82%,
respectively. Similarly, reaction of 14 with pyrid-3-yl benzene
boronate ester 15 led to a 67% yield of tetrachloride 17. 17 was
further coupled with an excess of 8 and 12 to afford 5 in a yield
of 60% and 6 in a yield of 74%, respectively.
Bond Dissociation Energy. Selective sequential cross-
coupling reactions involving multihalogenated aromatics is a
particularly attractive approach to polysubstituted heterocycles
that are of importance in pharmaceutical and manufacturing
applications.11 Knowles et al. reported that one source of
selectivity is the nature of the halogen that is being displaced;
iodine reacts before bromine, and bromine reacts before
chlorine.12 This originates from the significant differences in the
carbon−halogen (C−X) bond strengths. Considering that the
halogens being displaced are identical for the current reactants
and intermediates, such as bromine for 4 and 6, and chlorine
for 8, 11, and 12, there must be another factor that drives the
regioselectivity in the current Suzuki cross-coupling reactions.
Houk et al. reported that the regioselectivities in oxidation
additions are controlled by the energy to distort a C−X bond to
the transition state geometry, a factor related to the BDE.13
than 9 to give a higher yield of 9 than 18, which is consistent
with the experimental result. In addition, 10 has a C−Cl BDE
of 90.3 kcal/mol, which is much higher than the C−Br BDE of
9, further indicating bromine reacts before chlorine to give a
relatively high yield of the object intermediate 11.
Calculations demonstrate that the 4- and 6-positions of 13
have lower C−Cl bond BDE of 85.8 kcal/mol than the 2-
position (86.9 kcal/mol), and this is similar to that of 2,4-
dichloropyrimidine,14 indicating higher reactivity of C4−Cl and
C6−Cl than C2−Cl. Considering C−Cl BDE of 10 is further
higher than ones of 13, 10 undergoes cross-coupling with 13
and reacts primarily with its 4- and 6-positions rather than its 2-
position to give 14 in a high yield. For 14, the C−Cl bond at
the α-position to two nitrogen atoms of pyrimidine has a BDE
of 87.0 kcal/mol, which is quite similar to that of 13 at C2 and
is 2.8 kcal/mol lower than that at the benzene ring. As a result,
cross-coupling reactions of 14 with functionalized boranes of 8
and 15 give the α-position coupled products of 16 and 17 in
relatively high yield with the same catalyst system but higher
reaction temperature. Similar to 10 and 14, C−Cl BDEs at the
benzene ring of 16 and 17 are much higher than that at the
pyrimidine, a more effective catalyst system such as Pd2(dba)3,
PCy3, and K3PO4 is necessary for their cross-coupling reactions
with boronate esters. As a result, cross-coupling reactions of
tetrachloride of 11, 16, and 17 with pyridine boronate esters
give 1−6 in relatively high yield. Benefit from the sequential
regioselective cross-coupling reactions, pyridine-containing
compounds with different pyridine substitution positions in a
single molecule, such as 2, 4, and 6, were successfully
synthesized for the first time. We have reported that the
nitrogen atom position of the substituted pyridine ring is a key
factor in determining the electron injection and transport
property and thus the OLED performance.8 The current
3821
dx.doi.org/10.1021/cm303075m | Chem. Mater. 2012, 24, 3817−3827

Chemistry of Materials
Article
achievements would enable us to finely tune the property of the
heterocycle-containing ETMs and to thoroughly understand
their structure−property relationships.
Physical Properties. As an ideal ETM that can act as an
ETL as well as a hole-block layer, it should possess a low-lying
lowest unoccupied molecular orbital (LUMO) energy level to
give a low electron injection barrier and a low-lying highest
occupied molecular orbital (HOMO) energy level to block hole
leakage from the EML. The calculated HOMO and LUMO
energy levels of 1−6 are summarized in Table 1. For 1
Table 1. Calculated HOMO and LUMO Energy Levels,
HOMO−LUMO Band Gaps (Eg), and Adiabatic Triplet
Energies (T1−S0) of 1−6 (B3LYP/6-311+G(d,p)//B3LYP/
6-31G(d)) (All Values Reported in eV)
cmpd
HOMO
LUMO
Eg
T1−S0
1
2
3
4
5
6
6.66
6.90
6.81
7.13
6.52
6.78
1.83
2.08
2.47
2.69
2.49
2.65
4.83
4.82
4.33
4.43
4.03
4.13
2.95
2.98
2.94
2.96
2.92
2.96
Figure 1. UV−vis absorption spectra of the vacuum-deposited films of
□
◇
△
▽
○
●
1 ( ), 2 ( ), 3 ( ), 4 ( ), 5 ( ), and 6 ( ) on quartz substrates.
containing a meta-terphenyl core and five pyrid-3-yls combined
at its meta-positions as the periphery, its LUMO energy level is
lower-lying than that of a general electron-transport and hole-
block material of 2,9-dimethyl-4,7-diphenylphenanthroline
(BCP) (LUMO = 1.71 eV), and it also has deeper HOMO
energy level than BCP (HOMO = 6.07 eV),16 predicting
improved electron-injection and hole-block property compared
with BCP from the calculations. For 2 containing a pyrid-3-yl
and four pyrid-4-yls, its HOMO and LUMO energy levels are
0.24 and 0.25 eV lower-lying than those of its analugue 1,
respectively, by solely introducing four pyrid-4-yls instead of
pyrid-3-yls. By introducing a pyrimidine ring instead of their
central benzene, further lower-lying HOMO and LUMO
energy levels are obtained for 3 and 4. LUMO energy levels
of 5 and 6 are slightly influenced by introducing a phenyl at the
α-position to two nitrogen atoms of pyrimidine in comparison
with 3 and 4, respectively, although their HOMO energy levels
are ∼0.3 eV higher-lying. Moreover, adiabatic triplet energies
(T1−S0) of 1−6 are estimated to be over 2.9 eV, which are ca.
0.3 eV higher than that of BCP,8b suggesting improved triplet
energy confinement when used as an ETL as well as an exciton-
block layer adjacent to the EML of the phosphorescent OLEDs.
UV−vis absorption spectrum of 1 exhibits a peak at 262 nm,
which can be attributed to its π−π* transition (Figure 1).
Different from 1, there are absorption shoulders at 320 and 342
nm in combination with an absorption peak at 254 nm for 3,
which can be attributed to the n−π* transitions according to
the central diphenylpyrimidine skeleton. The absorption
shoulders due to the n−π* transitions shift to the longer
wavelengths of 326 and 346 nm for 5 due to one more phenyl
inserted at the α-position to two nitrogen atoms of pyrimidine
that induces an elongated π conjugation. Narrower optical band
gaps (Egopt) were obtained in the order of 1 > 3 > 5 due to the
bathochromic shift in UV−vis absorption spectrum (Table 2),
which is consistent with the calculation results (Table 1).
Similar phenomenon was also found for the compounds 2, 4,
and 6 containing pyridine rings with different substituted
positions, but the substituted positions of the pyridine rings
Table 2. Physical Properties of 1−6
a
a
b
c
d
e
Tm
Tg
(°C)
Td
HOMO
(eV)
LUMO
(eV)
Egopt
cmpd
(°C)
(°C)
(eV)
1
2
3
4
5
6
278
269
309
359
300
378
136
157
139
170
129
n.d.
497
506
498
505
507
517
6.68
6.42
6.57
6.51
6.65
6.62
2.68
2.42
3.10
3.01
3.21
3.17
4.00
4.00
3.47
3.50
3.44
3.45
a
Glass transition temperature (Tg) and melting temperature (Tm)
obtained from differential scanning calorimetry (DSC) measurement.
Decomposition temperature (Td) obtained from thermogravimetric
analysis (TGA). HOMO energy level approximated by atmospheric
b
c
d
ultraviolet photoelectron spectroscopy (Rikken Keiki AC-3). LUMO
energy level approximated from HOMO energy level and optical band
e
gap (Egopt). Optical band gap (Egopt) estimated from the lowest
energy absorption edge of the UV−vis absorption spectrum.
seem influence the absorption spectra and thus the energy band
gaps vary slightly.
The experimental results indicate that the LUMO energy
level of 1 is 2.68 eV, which is higher-lying than that of 3.
Considering the only structural difference between 1 and 3 is
the central pyrimidine ring. The lower-lying LUMO energy
level achieved for 3 can be attributed to strong electron affinity
of the central pyrimidine. The density functional theory
(DFT)15 calculations indicate that LUMO of 3 is mainly
located at the central diphenyl pyrimidine skeleton and HOMO
is mainly located at the outer dipyridyl phenyl skeleton (Figure
2), proving that the lower-lying LUMO energy level of 3 is due
to the strong electron affinity of the central pyrimidine.
Similarly, a reduced LUMO energy level is achieved for 4
compared with 2. By introducing one more phenyl at the α-
position to two nitrogen atoms of pyrimidine, further reduced
LUMO energy levels are achieved for 5 and 6, accompanied
with slightly reduced HOMO energy levels. Although the
3822
dx.doi.org/10.1021/cm303075m | Chem. Mater. 2012, 24, 3817−3827

Chemistry of Materials
Article
Figure 2. Calculated frontier orbitals of the HOMO (bottom) and LUMO (top) of 1−6.
Figure 3. (a) C 1s and (b) N 1s regions of the XPS spectra for thin film samples of ETMs 1−6.
Figure 4. (a) Curve fitting for C 1s of thin film sample of ETM 3. (b) Peak area of the two C species (C−N and C−C) for thin film samples of
ETMs 1−6.
introduction of one more phenyl for 5 and 6, similar to 3 and 4,
their LUMOs are also mainly located at the central diphenyl
pyrimidine skeleton. However, their HOMOs are mainly
located at the pyridyl phenyl combined at the α-position to
two nitrogen atoms of pyrimidine.
current compounds prove the high morphologic stability of the
amorphous phase in a deposited film, which is a prerequisite for
their applications in OLEDs.
In Figure 3, parts a and b show C 1s and N 1s regions of the
XPS spectra for thin film samples of ETMs 1−6. In Figure 3a,
thin film samples of ETMs 1, 2, 3, and 4 show the binding
energy of 285.9 eV at the maximum intensity. However, thin
film samples of ETMs 5 and 6 show the binding energies of
285.7 and 286.0 eV, respectively. In Figure 3b, the binding
energies of N 1s peaks at the maximum intensity are 400.1,
400.0, 399.9, 399.8, 399.8, and 400.0 eV, respectively.
To further understand the XPS data, we performed curve
Improved thermal properties were also achieved to give
higher melting temperature (Tm) and glass transition temper-
ature (Tg) by introducing pyrimidine as the core, and it can be
attributed to the increased polarity and thus stronger molecular
interaction. It was previously reported that no Tg was detected
for the similar compounds containing pyridine due to the
strong C − H···N hydrogen bonding interaction in the solid
state.8b,16,17 It is of interest that all the current compounds
show Tg higher than 120 °C except for 6. It can be partly
attributed to the introduction of pyrid-3-yl or pyrid-3-yl phenyl
at the α-position to two nitrogen atoms of pyrimidine to give a
reduced molecular symmetry and thus a reduced intermolecular
hydrogen bonding interaction. The high Tg values of the
̆
́
fitting for C 1s spectra using Doniach-Sunjic curves convoluted
with Gaussian distribution of 0.6 eV full-width at half-
maximum.18 The background due to inelastic scattering was
subtracted by the Shirley (or integral) method.19 Figure 4a
shows an example of curve fitting for the thin film sample of
ETM 3. The red circles in Figure 4a represent the raw data.
3823
dx.doi.org/10.1021/cm303075m | Chem. Mater. 2012, 24, 3817−3827

Chemistry of Materials
Article
The Shirley background (gray solid line in Figure 4a) was
subtracted, followed by the curve fitting with two peaks
corresponding to carbon species in 3, that is, C−N (red line)
and C−C (blue line). Because the electronegativity of the N
atom (3.04) is significantly stronger than that of the C atom
(2.55), the peak corresponding to C−N is expected to have a
higher binding energy than the peak corresponding to C−C.
Lorenzian width and Gaussian width are fixed to be 0.90 and
0.64 eV, respectively, based on the Au 4f curve fitting for an
independent Au thin film sample. The peak position and
intensity of the two peaks are used as fitting parameters to
obtain the final total signal (black line in Figure 4a). We
performed similar curve fitting for all samples. Note that, for 3−
6, there exists a third carbon species N−C−N. However, only
one carbon atom (out of 41 carbon atoms per molecule) has
this configuration, which is beyond the detection limit of the
XPS system used in this study. Therefore, only two major
carbon species are considered in the discussion hereafter.
Figure 4b shows the integrated area under the two carbon
peaks (in percent). As expected, the two curves change in
opposite directions to maintain the sum of the two area values
to be 100%. Black squares in Figure 4b represent the calculated
C−N peak area. There is a striking similarity between the
experimental curve and the theory curve, although they are
offset by approximately 13% possibly due to strong
intermolecular interactions in solid state films.
Figure 5. Luminance (L) and current density (J) (inset) versus
operating voltage (V) characteristics of the devices: ITO/TPDPES (20
nm)/TAPC (30 nm)/CBP: 8 wt % Ir(PPy)3 (10 nm)/ETMs (50
□
◇
△
▽
nm)/LiF (0.5 nm)/Al (100 nm). ETMs: 1 ( ), 2 ( ), 3 ( ), 4 ( ),
○
●
5 ( ), and 6 ( ).
As shown in Figure 6, electroluminescent (EL) spectra of all
the current devices are identical with a peak at 514 nm and a
Device Application. It has been reported that the green
phosphorescent OLEDs based on Ir(PPy)3 exhibit the lowest
driving voltage among the previously reported phosphorescent
OLEDs. To study the electron injection and transport property
of the developed materials, devices in an architecture of ITO/
TPDPES (20 nm)/TAPC (30 nm)/CBP: 8 wt % Ir(PPy)3 (10
nm)/1−6 (50 nm)/LiF (0.5 nm)/Al (100 nm) were fabricated
by using 1−6 as a nondoped ETL as well as a hole/exciton-
block layer; that is, there is no other hole/exciton-block layer
inserted between the EML and the ETL, which is indispensable
for the p-i-n OLEDs. In addition, a 0.5-nm-thick layer of LiF
followed by a 100-nm-thick layer of Al is used as the cathode.
The same as our previously reported green phosphorescent
OLEDs, TPDPES is used as the hole-injection buffer layer, and
TAPC is used as the HTL.7d
Figure 6. Electroluminescence (EL) spectra of the devices: ITO/
TPDPES (20 nm)/TAPC (30 nm)/CBP: 8 wt % Ir(PPy)3 (10 nm)/
ETMs (50 nm)/LiF (0.5 nm)/Al (100 nm). ETMs: 1−6.
For the device based on 1, a threshold voltage (Vth) of 2.5 V
for current and a turn-on voltage (Von) of 2.7 V for
electroluminescence (luminance of 1 cd m−2 was detected)
were obtained. In comparison with the device based on 1,
although Vth and Von for the device based on 2 are 0.1 V lower,
driving voltages for higher luminance are much higher due to
the higher-lying LUMO energy level of 2 and thus a high
electron injection barrier. Driving voltages of the devices
significantly decreased by utilizing 3 and 4 as an ETL (Figure
5). An ultralow Vth of 2.0 V and Von of 2.2 V were achieved for
the device based on 3, both are 0.5 V lower than those of the
device based on 1. Considering the only structural difference
between 1 and 3 is the central pyrimidine ring in 3, the reduced
driving voltage can be attributed to the lower-lying LUMO
energy level of 3 according to the pyrimidine ring with a strong
electron affinity. By introducing four pyrid-4-yls instead of the
pyrd-3-yls combined with the phenyls in 3 to give 4, further
reduced Von of 2.1 V was achieved for the device based on 4.
As formerly mentioned, the ideal operating voltage of
OLEDs should approximately correspond to the photon energy
(hv) of the emitted photons. In fact, most of the ever reported
classical OLEDs were far above in terms of operating voltage.
shoulder at 550 nm. Note that the emitted photon energy
corresponding to the peak emission wavelength of 514 nm is
2.41 eV, and the average photon energy of the emitted light for
the Ir(PPy)3-based device is ∼2.3 eV. It is of interest that the
threshold voltage Von for electroluminescence of the current
OLEDs is readily 0.2−0.3 V lower than the minimum value of
hv/e, and a similar phenomenon has also been experimentally
found for the inorganic light-emitting diodes.20 By introducing
a phenyl at the α-position to two nitrogen atoms of pyrimidine
to give 5, although the on-set voltages for current and
electroluminescence are almost the same as those of the device
based on 3, at a display-relevant luminance of 100 cd m−2,
decreased driving voltage of 2.47 V was achieved for the device
based on 5. In comparison, the lowest operating voltage for a
typical luminance of 100 cd m−2 for the Ir(PPy)3-based green
phosphorescent p-i-n OLEDs is 2.65 V,6 which is nearly 0.2 V
higher than the current one. Even at an illumination-relevant
luminance of 1000 cd m−2, the operating voltage is as low as
2.83 V. By introducing four pyrid-4-yls instead of four pyrid-3-
yls of 5 to give 6, further reduced driving voltages of 2.39 and
3824
dx.doi.org/10.1021/cm303075m | Chem. Mater. 2012, 24, 3817−3827

Chemistry of Materials
Article
Figure 7. AFM images of the bare ITO substrate and the thin films of 1−6 vacuum deposited on ITO.
2.72 V were achieved for the device based on 6 at 100 and 1000
cd m−2, respectively, which are the lowest driving voltages for
the green phosphorescent OLEDs hitherto,7d including the ever
reported p-i-n OLEDs.6 Although the current driving voltage
for 100 cd m−2 is already corresponding to the emitted photon
energy, it is not the theoretical limit yet. As inspected by Leo et
al., in an ideal device the quasi-Fermi-level spread is identical to
the voltage applied to the contacts of the device, the theoretical
voltage necessary to reach 1 and 100 cd m−2 for green light is as
low as 1.83 and 1.95 V, respectively.5 In comparison, the turn-
on voltage and the operating voltage for 100 cd m−2 for the
device based on 6 are only 0.27 and 0.44 V above the
thermodynamic limits, respectively. Considering that the
exciton binding energy might need some extra energy, the
driving voltage achieved by 6 is very close to the
thermodynamic limit, especially at low brightness. Moreover,
even at a much brighter luminance of 10 000 cd m−2, a driving
voltage as low as 3.88 V was also achieved.
shown in Figure 8, improved current efficiency (CE) is
achieved for the device based on 3 with a pyrimidine core in
It is well-known that, aside from the LUMO energy level and
the electron injection barrier, the electron mobility may also
play an important role in reduced operating voltage.
Unfortunately, we failed in obtaining clear transient current
by using general time-of-flight (TOF) technique and thus
electron mobility of the current ETMs. Nevertheless, we have
previously found that electron mobility as high as 10−4 ∼ 10−3
cm2 V−1 s−1 can be achieved for the pyridine-containing ETMs
like triphenylbenzene derivatives.8a Figure 7 shows the AFM
images of the bare ITO substrate after rigorous cleaning and
the thin films of ETMs 1−6. It can be seen that the
morphologies of the films of 1 and 2 are nearly conformal to
the underlying ITO substrates. In contrast, for the films of 3−6
with a pyrimidine core fine structures can be identified, which
may result from the well-ordered molecular domain structures
in these samples. Such structures are expected to be favorable
for electron hopping due to the strong intermolecular
interaction such as π−π stacking. Interestingly, lower driving
voltages were achieved for the devices that are based on the
ETMs containing a pyrimidine core.
Figure 8. Power efficiency (PE) and current efficiency (CE) versus
luminance (L) characteristics of the devices: ITO/TPDPES (20 nm)/
TAPC (30 nm)/CBP: 8 wt % Ir(PPy)3 (10 nm)/ETMs (50 nm)/LiF
□
◇
▲
▽
○
(0.5 nm)/Al (100 nm). ETMs: 1 ( ), 2 ( ), 3 ( ), 4 ( ), 5 ( ),
●
and 6 ( ).
comparison with 1 with a benzene core due to the improved
carrier balance. Considering the reduced operating voltage and
the improved current efficiency, aside from the improved
electron injection and transport, the hole injection and
transport must also be improved when 3 is used as the ETL.
The improved electron injection may induce an aggregation of
electron at the cathode side and thus an increased internal
electric field. The hole injection and transport from the anode
should be accelerated due to the increased electric field despite
the same HTL and buffer layer utilized. Although CE of the
device based on 5 slightly decreased compared with the device
based on 3, it is still higher than the device based on 1 at the
luminance lower than 5000 cd m−2. A similar phenomenon was
also found for the devices based on the ETMs of 2, 4, and 6
containing pyridine rings with different substitution positions.
Among these ETMs, the highest CE of 102 and 94.9 cd A−1,
Besides the unprecedented low operating voltage, an
improved carrier balance was also achieved for the current
devices based on the ETMs containing a pyrimidine ring. As
3825
dx.doi.org/10.1021/cm303075m | Chem. Mater. 2012, 24, 3817−3827

Chemistry of Materials
Article
Table 3. Summary of Performance Data of the Devices: ITO/TPDPES (20 nm)/TAPC (30 nm)/CBP: 8 wt % Ir(PPy)3 (10
nm)/ETMs (50 nm)/LiF (0.5 nm)/Al (100 nm). ETMs: 1−6
at 100 cd m−2
at 1000 cd m−2
PE (lm W1− CE (cd A−1
ETMs
Von (V)
V (V)
PE (lm W1− CE, (cd A−1
)
)
ηext (%)
V (V)
)
)
ηext (%)
1
2
3
4
5
6
2.7
2.6
2.2
2.1
2.2
2.1
3.07
3.32
2.49
2.45
2.47
2.39
90.3
79.8
88.1
84.3
24.6
23.5
28.5
27.2
26.8
26.9
3.48
4.11
2.97
2.98
2.83
2.72
72.8
58.3
80.7
76.2
94.9
88.4
87.5
89.0
22.5
21.3
26.4
24.6
24.3
24.6
129
102
100
125
123
128
97.6
96.5
97.0
93.3
97.1
103
corresponding to ηext of 28.5% and 26.4%, were achieved at 100
and 1000 cd m−2, respectively, for the 3-based device.
Accompanied with the low operating voltage, high power
efficiency (PE) of 129 and 100 lm W1− were achieved at 100
and 1 000 cd m−2, respectively. In comparison, the devices
based on 4−6 exhibit relatively lower CE of ∼97 and ∼88 cd
A−1 at 100 and 1 000 cd m−2, respectively. Although CE of the
device based on 6 is somewhat lower than that of the device
based on 3, a comparable PE of 128 lm W1− was achieved at
100 cd m−2 due to the reduced driving voltage. In addition, at
an illumination-relevant luminance of 1000 cd m−2, a higher PE
of 103 lm W1− was achieved (Table 3). Compared with the
previously reported green phosphorescent OLEDs in the same
architecture that exhibit very low operating voltage,7d the device
based on 6 exhibits further lower operating voltage and higher
ηext, further proving simultaneous reduced operating voltage
and improved carrier balance can be achieved by the
development of an ETM with improved electron injection
and transport. Note that no light-out coupling technology was
used in this work, and all the efficiencies were recorded from
the forward view. This means that there is still a big room for
further improvement of electroluminescent efficiency and
decrement of operating voltage of the current devices by
various light-out coupling technologies.21 Considering that the
ηext of the device with the lowest driving voltage is still
somewhat lower than the device based on 3, there is also some
other room for further improvement of power efficiency by the
improvement of ηext to the high limit without any voltage drops,
and further optimization of hybrid heterocycle-containing
ETMs should be an effective route to realize it.
morphological fine structure of the thin film samples of ETMs
with a pyrimidine core, which possibly contributes to the low
driving voltages that were achieved for the devices that are
based on these ETMs. These findings indicate that aside from
the ultralow operating voltage well-balanced carriers can be also
achieved with a finely synthesized ETM as a nondoped ETL,
and the B3LYP calculation is an effective method to predict the
regioselective cross-coupling reactions for fine design of such an
ETM.
AUTHOR INFORMATION
Corresponding Author
*E-mail: mssjsu@scut.edu.cn (S.J.S.), yabing.qi@oist.jp
(Y.B.Q.).
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
S.J.S. greatly appreciates the financial support from the National
Natural Science Foundation of China (51073057), the Ministry
of Science and Technology (2009CB930604 and
2011AA03A110), the Ministry of Education (NCET-11-
0159), and the Fundamental Research Funds for the Central
Universities (2011ZZ0002). Y.B.Q. and M.C.J. acknowledge
the support from an internal grant of Okinawa Institute of
Science and Technology Graduate University. The authors
acknowledge Chemipro Kasei Kaisha Limited for TPDPES.
REFERENCES
■
(1) (a) Kido, J.; Iizumi, Y. Appl. Phys. Lett. 1998, 73, 2721.
(b) Okumoto, K.; Kanno, H.; Hamaa, Y.; Takahashi, H.; Shibata, K.
Appl. Phys. Lett. 2006, 89, 063504.
CONCLUSIONS
■
In summary, a series of hybrid heterocycle-containing ETMs
with different pyridine substitution positions in a single
molecule were developed by regioselective sequential palla-
dium-catalyzed Suzuki cross-coupling reactions. The two
carbon species associated with C−N and C−C can be detected
by X-ray photoelectron spectroscopy. B3LYP calculations
exhibit the experimental regioselectivity is conceivably con-
sistent with the quantitative trends in BDEs of the C−X bonds
of aryl halides. Similar to the inorganic light-emitting diodes,
extremely low turn-on voltage for electroluminescence, which is
0.2−0.3 V lower than the minimum value of hv/e, was
experimentally achieved by utilizing the developed ETMs as an
electron-transport and hole/exciton-block layer for the classical
Ir(PPy)3-based green phosphorescent OLEDs. Moreover,
hitherto the lowest operating voltage of 2.39 V for 100 cd
m−2 that is already corresponding to the emitted photon energy
(hv) was achieved with simultaneously improved external
quantum efficiency to give high power efficiency by utilizing 6
as an ETL. Atomic force microscopy measurements reveal
(2) (a) Baldo, M. A.; O’Brien, D. F.; You, Y.; Shoustikov, A.; Sibley,
S.; Thompson, M. E.; Forrest, S. R. Nature (London) 1998, 395, 151.
(b) Baldo, M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.;
Forrest, S. R. Appl. Phys. Lett. 1999, 75, 4. (c) Wong, W.-Y.; Ho, C.-L.
Coord. Chem. Rev. 2009, 253, 1709. (d) Wong, W.-Y.; Ho, C.-L. J.
Mater. Chem. 2009, 19, 4457.
(3) (a) Kido, J.; Matsumoto, T. Appl. Phys. Lett. 1998, 73, 2866.
(b) Watanabe, S.; Agata, Y.; Tanaka, D.; Kido, J. J. Photopolym. Sci.
Technol. 2005, 18, 83.
(4) (a) Huang, J.; Pfeiffer, M.; Werner, A.; Blochwitz, J.; Leo, K. Appl.
Phys. Lett. 2002, 80, 139. (b) Pfeiffer, M.; Forrest, S. R.; Leo, K.;
Thompson, M. E. Adv. Mater. 2002, 14, 1633. (c) Matsushima, T.;
Adachi, C. Appl. Phys. Lett. 2006, 89, 253506. (d) Meerheim, R.;
Walzer, K.; Pfeiffer, M.; Leo, K. Appl. Phys. Lett. 2006, 89, 061111.
(5) Meerheim, R.; Walzer, K.; He, G.; Pfeiffer, M.; Leo, K. Proc. SPIE
2006, 6192, 61920P.
(6) He, G.; Pfeiffer, M.; Leo, K.; Hofmann, M.; Birnstock, J.; Pudzich,
R.; Salbeck, J. Appl. Phys. Lett. 2004, 85, 3911.
(7) (a) Su, S.-J.; Chiba, T.; Takeda, T.; Kido, J. Adv. Mater. 2008, 20,
2125. (b) Su, S.-J.; Tanaka, D.; Li, Y.-J.; Sasabe, H.; Takeda, T.; Kido,
3826
dx.doi.org/10.1021/cm303075m | Chem. Mater. 2012, 24, 3817−3827

Chemistry of Materials
Article
J. Org. Lett. 2008, 10, 941. (c) Xiao, L.; Su, S.-J.; Agata, Y.; Lan, H.;
Kido, J. Adv. Mater. 2009, 21, 1271. (d) Su, S.-J.; Sasabe, H.; Pu, Y.-J.;
Nakayama, K.; Kido, J. Adv. Mater. 2010, 22, 3311.
(8) (a) Su, S.-J.; Takahashi, Y.; Chiba, T.; Takeda, T.; Kido, J. Adv.
Funct. Mater. 2009, 19, 1260. (b) Sasabe, H.; Tanaka, D.; Yokoyama,
D.; Chiba, T.; Pu, Y.-J.; Nakayama, K.; Yokoyama, M.; Kido, J. Adv.
Funct. Mater. 2011, 21, 336.
(9) Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D.
Handbook of X-ray Photoelectron Spectroscopy; Physical Electronics:
Eden Prairie, MN, 1995.
(10) Kudo, N.; Perseghini, M.; Fu, G. C. Angew. Chem., Int. Ed. 2006,
45, 1282.
(11) (a) Schroter, S.; Stock, C.; Bach, T. Tetrahedron 2005, 61, 2245.
̈
(b) Fairlamb, I. J. S. Chem. Soc. Rev. 2007, 36, 1036.
(12) Knowles, J. P.; Whiting, A. Org. Biomol. Chem. 2007, 5, 31.
(13) Legault, C. Y.; Garcia, Y.; Merlic, C. A.; Houk, K. N. J. Am.
Chem. Soc. 2007, 129, 12664.
(14) Garcia, Y.; Schoenebeck, F.; Legault, C. Y.; Merlic, C. A.; Houk,
K. N. J. Am. Chem. Soc. 2009, 131, 6632.
(15) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J.
B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R.
E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J.
J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.;
Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision
D.01; Gaussian, Inc.: Wallingford, CT, 2004.
(16) Sasabe, H.; Chiba, T.; Su, S.-J.; Pu, Y.-J.; Nakayama, K.; Kido, J.
Chem. Commun. 2008, 5821.
(17) Tanaka, D.; Sasabe, H.; Li, Y.-J.; Su, S.-J.; Takeda, T.; Kido, J.
Jpn. J. Appl. Phys., Part 2 2007, 46, L10.
̆
́
(18) Doniach, S.; Sunjic, M. J. Phys. C 1970, 3, 285.
(19) Shirley, D. A. Phys. Rev. B 1972, 5, 4709.
(20) Schubert, E. F. Light-Emitting Diodes, 2nd ed.; Cambridge
University Press: Cambridge, 2006; pp 83−85.
(21) (a) Sun, Y.; Forrest, S. R. Nat. Photonics 2008, 2, 483.
(b) Reineke, S.; Lindner, F.; Schwartz, G.; Seidler, N.; Walzer, K.;
Lussem, B.; Leo, K. Nature (London) 2009, 459, 234. (c) Helander, M.
̈
G.; Wang, Z. B.; Qiu, J.; Greiner, M. T.; Puzzo, D. P.; Liu, Z. W.; Lu,
Z. H. Science 2011, 332, 944.
3827
dx.doi.org/10.1021/cm303075m | Chem. Mater. 2012, 24, 3817−3827