Charged dinuclear Cu(I) complexes for solution-processed single-emitter warm white organic light-emitting devices

Article abstract of DOI:10.1016/j.dyepig.2017.04.036

A series of charged dinuclear Cu(I) complexes were developed by functionalization of both central bipyrimidine-based ligands and organic phosphine ligands. The chemical modification of the ligands can effectively affect the absorption, emission and thermal stability properties of the Cu(I) complexes. The decomposition temperatures (T_d) are improved by using bulky ligands. Interestingly, the complexes with the triphenylamine group on the bipyrimidine ligand show higher photoluminescence quantum yield (PLQY) than the complexes bearing the triphenylphosphine oxide group, while the complexes having the triphenylamine group on the phosphine ligand display lower PLQY. The functional groups also show an obvious influence on the redox behaviors of these complexes. Most importantly, the organic light-emitting diodes (OLEDs) based on selected dinuclear Cu(I) complexes show impressive EL features. The best performance is achieved by the device based on complex Cu-MD-1 with the maximum external quantum efficiency (EQE) of 6.09%, current efficiency (CE) of 12.78?cd?A^?1 and power efficiency (PE) of 5.93 lm W^?1, representing the state-of-the-art EL efficiencies reported for the charged Cu(I) complexes. In addition, the OLEDs based on these Cu(I) complexes can emit warm white light with Color Rendering Index (CRI) as high as 88 and Commission Internationale Ed I'eclairage (CIE) coordinates close to (0.40, 0.46), showing the great potential of these Cu(I) complexes in fabricating single-emitter warm WOLEDs.

Full text of DOI:10.1016/j.dyepig.2017.04.036

Dyes and Pigments 143 (2017) 151e164
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
Charged dinuclear Cu(I) complexes for solution-processed single-
emitter warm white organic light-emitting devices
,
*
Xiaolong Yang a, Xiaogang Yan a, Haoran Guo a, Boao Liu a, Jiang Zhao a, Guijiang Zhou a
,
,
,
, d, ****
Yong Wu a **, Zhaoxin Wu b ***, Wai-Yeung Wong c
a MOE Key Laboratory for Nonequilibrium Synthesis and Modulation of Condensed Matter and Institute of Chemistry for New Energy Materials, Department
of Chemistry, School of Science, State Key Laboratory for Mechanical Behavior of Materials, Xi'an Jiaotong University, Xi'an 710049, China
b Key Laboratory of Photonics Technology for Information, School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an 710049, China
c Institute of Molecular Functional Materials and Department of Chemistry, Hong Kong Baptist University, Waterloo Road, Kowloon Tong, Hong Kong, China
d Department of Applied Biology and Chemical Technology, The Hong Kong Polytechnic University, Hung Hom, Hong Kong, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of charged dinuclear Cu(I) complexes were developed by functionalization of both central
bipyrimidine-based ligands and organic phosphine ligands. The chemical modification of the ligands can
effectively affect the absorption, emission and thermal stability properties of the Cu(I) complexes. The
decomposition temperatures (Td) are improved by using bulky ligands. Interestingly, the complexes with
the triphenylamine group on the bipyrimidine ligand show higher photoluminescence quantum yield
(PLQY) than the complexes bearing the triphenylphosphine oxide group, while the complexes having the
triphenylamine group on the phosphine ligand display lower PLQY. The functional groups also show an
obvious influence on the redox behaviors of these complexes. Most importantly, the organic light-
emitting diodes (OLEDs) based on selected dinuclear Cu(I) complexes show impressive EL features.
The best performance is achieved by the device based on complex Cu-MD-1 with the maximum external
quantum efficiency (EQE) of 6.09%, current efficiency (CE) of 12.78 cd Aꢀ1 and power efficiency (PE) of
5.93 lm Wꢀ1, representing the state-of-the-art EL efficiencies reported for the charged Cu(I) complexes.
In addition, the OLEDs based on these Cu(I) complexes can emit warm white light with Color Rendering
Index (CRI) as high as 88 and Commission Internationale Ed I'eclairage (CIE) coordinates close to (0.40,
0.46), showing the great potential of these Cu(I) complexes in fabricating single-emitter warm WOLEDs.
© 2017 Published by Elsevier Ltd.
Received 21 February 2017
Received in revised form
29 March 2017
Accepted 19 April 2017
Available online 20 April 2017
Keywords:
Dinuclear Cu(I) complex
Photoluminescence
Single-emitter
OLED
1. Introduction
complexes tend to show weak emission with metal-to-ligand
charge transfer (MLCT) character due to structural distortions in
The long-lived emission properties of Cu(I) complexes with N^N
type ligands have attracted the attention of the researchers since
1978 [1]. These Cu(I) complexes can show great potential in several
fields, including luminescence-based sensing, solar-energy con-
version and as bio-probes, etc. [2e4] However, the tetrahedral Cu(I)
their excited states, which will enhance the non-radiative decay
rate constant (knr) to give low photoluminescence quantum yields
(PLQYs) [5,6]. Fortunately, the optimized Cu(I) complexes with both
bulky N^N ligand and bidental P^P ligand can show greatly
enhanced PLQYs in solution [7e10]. In addition, the emission of the
Cu(I) complexes can be enhanced in the solid state due to the
absence of the exciplex with the solvent molecules [10,11]. Gener-
ally, the emissions of these Cu(I) complexes mainly originate from
triplet MLCT states (3MLCT) [12], which are very similar to those of
cyclometalated phosphorescent Ir(III) and Pt(II) complexes [13e16].
However, considering the high cost of using Ir(III) and Pt(II) com-
plexes to fabricate OLEDs, the emissive Cu(I) complexes with low
cost and abundant supply are very competitive candidate as high-
performance phosphorescent emitters for making low-cost OLEDs.
Since the first OLED fabricated using the Cu(I) complex Cu-1
* Corresponding author.
** Corresponding author.
*** Corresponding author.
**** Corresponding author. Institute of Molecular Functional Materials and
Department of Chemistry, Hong Kong Baptist University, Waterloo Road, Kowloon
Tong, Hong Kong, China.
E-mail addresses: zhougj@mail.xjtu.edu.cn (G. Zhou), specwy@mail.xjtu.edu.cn
(Y. Wu), zhaoxinwu@mail.xjtu.edu.cn (Z. Wu), rwywong@hkbu.edu.hk, wai-yeung.
wong@polyu.edu.hk (W.-Y. Wong).
http://dx.doi.org/10.1016/j.dyepig.2017.04.036
0143-7208/© 2017 Published by Elsevier Ltd.

152
X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
(Scheme 1) as emitter by Ma et al. [17], the EL efficiency of OLEDs
using phenanthroline-based Cu(I) complex Cu-2 with bis[2-
(diphenylphosphino)phenyl]ether has been improved to ca.
11 cd Aꢀ1 by Wang and Li, respectively [12,18]. Recently, novel
phosphine-based Cu(I)X (X ¼ Cl, Br, I) complexes can show
outstanding electroluminescence (EL) performance [19,20]. For
example, the three-coordinate Cu(I) complex Cu-3 prepared by
Osawa et al. gave the EL efficiencies over 65 cd Aꢀ1 and 21% in
vacuum-deposited devices, which are among the highest EL effi-
ciencies ever achieved by the Cu(I) complexes [19]. In addition, Liu
et al. have reported that the emissive Cu(I) complex Cu-4 can be
formed in-situ by vacuum co-deposition to show high EL effi-
ciencies in simple EL devices, providing an important outlet to
construct high performance devices based on Cu(I) complexes [21].
Some high performance neutral dinuclear Cu(I) complexes have
also been developed as emitters for OLEDs. Bian et al. reported that
achieved an extremely high current efficiency (CE) of 76 cd Aꢀ1 and
external quantum efficiency (EQE) of 23% [23]. However, although
the charged dinuclear Cu(I) complexes (e.g. Cu-7 and Cu-8) have
also been synthesized to investigate their structural and emissive
features, they are rarely employed in OLEDs [24e26]. Therefore, we
synthesized a series of charged dinuclear Cu(I) complexes bearing
functional groups to explore not only their structure-property
relationship, but also assess their potential in the field of OLEDs.
The experimental results show that the thermally stable, photo-
physical and redox properties of these dinuclear Cu(I) complexes
can be conveniently tuned by adjusting or modifying the ligands.
Furthermore, solution-processed OLEDs based on selected dinu-
clear Cu(I) complexes can display satisfactory EL performance with
the highest EQE of 6.09%, current efficiency (CE) of 12.78 cd Aꢀ1 and
power efficiency (PE) of 5.93 lm Wꢀ1, setting a benchmark of EL
efficiencies for the charged Cu(I) complexes. More encouragingly,
devices based on these charged Cu(I) complexes can emit intense
warm white light with high CRI of 88. Generally, white light can be
generated by mixing the red, green and blue (RGB) colors or by
the
m-iodino-bridged dinuclear Cu(I) complex Cu-5 could show
high EL efficiencies of 51.6 cd Aꢀ1 and 15.7% [22]. Baumann et al.
demonstrated that the fully bridged dinuclear Cu(I) complex Cu-6
Scheme 1. The molecular structures of some Cu(I) complexes.

X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
153
ratiometric fine-tuning of two complementary colors of light.
2.3. OLED fabrication and measurements
However, these two approaches require an elaborate design of
device structures as well as the delicate balance between each
emitting material, which will not favor the practical application at
low cost. Therefore, the development of WOLEDs based on a single
emitter becomes more attractive [27,28]. Our new Cu(I) complexes
nicely show the great potential in fabricating single-emitter warm
WOLEDs.
The pre-cleaned ITO glass substrates were treated with ozone
for 20 min. Then, a 35 nm thick PEDOT:PSS was deposited on the
ITO glass substrates by spin-coating. After drying 30 min at 100 ꢁC,
the emission layer of 50 nm was constructed by spin-coating the
solution of the mixture of TCTA and Cu(I) complex in 1,2-
dichloroethane. After drying 20 min at 50 ꢁC, TPBI, LiF and Al
were evaporated successively at a base pressure of less than
10ꢀ6 Torr. The EL spectra and CIE coordinates were measured with a
PR650 Spectra colorimeter. The JeVeL curves of the devices were
recorded using a Keithley 2400/2000 source meter and the lumi-
nance was measured using a PR650 SpectraScan spectrometer. All
of the experiments and measurements were carried out at room
temperature under ambient conditions.
2. Experimentals
2.1. Methods
All reactions were conducted under a nitrogen atmosphere and
no special precautions were required during the workup. Solvents
were carefully dried and distilled from appropriate drying agents
prior to use. Commercially available reagents were used without
further purification unless otherwise stated. All reactions for the
preparation of the organic compounds were monitored by TLC with
Merck pre-coated aluminum plates. Flash column chromatography
was carried out using silica gel from Shenghai Qingdao (200e300
mesh). UVevis spectra were recorded on a Shimadzu UV-2250
spectrophotometer. Emission spectra and lifetimes of these com-
pounds were recorded on an Edinburgh Instruments Ltd (FLSP920)
fluorescence spectrophotometer using the software package pro-
vided by Edinburgh Instruments. The PLQYs of these dinuclear
Cu(I) complexes in doped PMMA film on quartz substrate with ca.
10 wt % doping level were measured in an integration sphere
associated with FLSP920 fluorescence spectrophotometer. The
lifetimes were measured by a single photon counting spectrometer
from Edinburgh Instruments FLS920 with a nano-LED as the exci-
tation source. The data analysis was conducted by iterative
convolution of the luminescence decay profile with the instrument
response function using the software package provided by Edin-
burgh Instruments. Fast atom bombardment (FAB) mass spectra
were collected with Finnigan MAT SSQ710 system. The thermal
gravimetric analysis (TGA) was conducted on a NETZSCH STA 409C
2.4. Synthesis
The synthesis of [(CH3CN)4Cu]BF4 [34], 4-(diphenylamino)phe-
nylboronic, PPO-Br and (4-bromophenyl)diphenylphosphine were
prepared by the literature methods [35]. DEPHOS and XANTPHOS
were purchased from Acros. The dinuclear Cu(I) complexes have
been prepared by referring to the synthetic method employed for
the preparation of mononuclear analogs [36].
2.4.1. Synthesis of MD-Br
2,20-Bipyrimidine (0.5 g, 3.16 mmol), Br2 (3 mL) and nitroben-
zene (1.5 mL) were charged into a sealed reaction tube. The reaction
mixture was heated to 160 ꢁC for 24 h. Then, the reaction mixture
was poured into 1 M Na2S2O3 (30 mL) and the pH of the aqueous
phase was adjusted to ca. 10. Then, the mixture was extracted with
CH2Cl2 (3 ꢂ 30 mL). The organic phase was collected and dried over
Mg2SO4. The crude product was purified by column chromatog-
raphy on silica gel with a mixture of ethyl acetate (EA) and meth-
anol (MeOH) (VEA/VMeOH ¼ 5:1) as eluent to give the product as an
off-white solid (0.19 g, Yield: 25%). 1H NMR (400 MHz, CDCl3):
d
(ppm) 9.03 (s, 2H), 9.00 (d, J ¼ 4.8 Hz, 2H), 7.44 (t, J ¼ 4.8 Hz, 1H);
13C NMR (100 MHz, CDCl3):
d (ppm) 161.61, 160.24, 158.70, 158.08,
121.66, 121.59; FAB-MS (m/z): 236, 238 [M]þ; Anal. Calcd. for
C8H5BrN4: C, 40.53; H, 2.13; N, 23.63; found: C, 40.28; H, 2.38; N,
23.47.
instrument under nitrogen with a heating rate of 20 K minꢀ1
.
Electrochemical measurements were measured using a Princeton
Applied Research model 2273A potentiostat with a scan rate of
100 mV sꢀ1. A conventional three-electrode configuration consist-
ing of a glassy carbon working electrode, a Pt-sheet counter elec-
trode, and a Pt-wire reference electrode was used. The supporting
electrolyte was 0.1 M [Bu4N]PF6 in MeCN. Ferrocene was added as a
calibrant after each set of measurements, and all potentials re-
ported are quoted with reference to the ferroceneeferrocenium
(Fc/Fcþ) couple.
2.4.2. Synthesis of PPO-B
PPO-Br (0.36 g, 1.0 mmol), bis(pinacolato)diboron (0.31 g,
1.2 mmol) potassium acetate (0.49 g, 5.0 mmol) were added to
dioxane (15 mL) in an inert atmosphere and stirred at 100 ꢁC for
18 h. After cooling to room temperature, water (20 mL) was added.
The mixture was extracted with CH2Cl2 (3 ꢂ 20 mL). The collected
organic phase was dried over anhydrous MgSO4 and the solvent
was removed under reduced pressure. The residue was chromato-
graphed on a silica column using CH2Cl2 (DCM) and petroleum
ether (PE) (5:1 v/v) as eluent to afford the product as a white solid
2.2. Computational details
(0.34 g, 83%). 1H NMR (400 MHz, CDCl3):
7.70e7.62 (m, 6H), 7.55e7.51 (m, 2H), 7.46e7.42 (m, 4H). 13C NMR
(100 MHz, CDCl3): (ppm) 134.56, 134.44, 132.11, 132.01, 131.93,
131.91, 131.21, 131.11, 128.51, 128.39, 84.17, 24.82. 31P NMR
(162 MHz, CDCl3):
(ppm) 29.15; FAB-MS (m/z): 404 [M] þ. Anal.
d (ppm) 7.89e7.87 (m, 2H),
Density functional theory (DFT) calculations using the B3LYP
were performed for all the Cu(I) complexes and their geometries
were obtained by theoretical optimization during the computation.
The basis set used for C, H, N and O atoms was 6-31G while effective
core potentials with a LanL2DZ basis set were employed for P and
Cu atom [29,30]. The energies of the excited states of the complexes
were computed by time-dependent (TD) DFT (TD-DFT) based on all
the ground-state geometries. All calculations were carried out us-
ing the Gaussian 09 program [31]. Mulliken population analyses
were performed using MullPop [32]. Frontier molecular orbitals
obtained from the DFT calculations were plotted using the Molden
3.7 program [33].
d
d
calcd. for C24H26BO3P: C, 71.31; H, 6.48; found: C, 71.19; H, 6.53.
2.4.3. Synthesis of MD-TPA
Under
a N2 atmosphere, 4-(diphenylamino)phenylboronic
(0.094 g, 0.325 mmol), MD-Br (0.0516 g, 2.17 mmol) and Pd(PPh3)4
(0.025 g, 0.021 mmol) were added to the mixture of THF (5 mL) and
2 M K2CO3 (3 mL). The reaction mixture was stirred at 110 ꢁC for
16 h. After cooling to room temperature, water (20 mL) was added

154
X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
and the mixture was extracted with CH2Cl2 (3 ꢂ 20 mL). The
organic phase was collected and dried over Mg2SO4. After removing
the solvent, the residue was purified by column chromatography on
silica gel with a mixture of ethyl acetate (EA) and methanol (MeOH)
(VEA/VMeOH ¼ 3:1) as eluent to give the product as a white solid
2.4.6.1. Cu-MD-1 [24](yield: 94%) 1H NMR (400 MHz, CDCl3).
d
(ppm) 8.74 (d, J ¼ 5.2 Hz, 4H), 7.91 (t, J ¼ 5.2 Hz, 2H), 7.31 (t,
J ¼ 7.2 Hz, 12H), 7.15 (t, J ¼ 7.2 Hz, 24H), 7.06 (br, 24H); 31P NMR
(162 MHz, CDCl3): d
(ppm) 3.81 (s); FAB-MS (m/z): 1332 [M-2BF4]þ;
Anal. Calcd. for C80H66B2Cu2F8N4P4: C, 63.72; H, 4.41; N, 3.72;
found: C, 63.53; H, 4.48; N, 3.63.
(0.071 g, Yield: 81%). 1H NMR (400 MHz, CDCl3):
d (ppm) 9.18 (s,
2H), 9.04 (d, J ¼ 4.8 Hz, 2H), 7.55 (d, J ¼ 8.8 Hz, 2H), 7.44 (t,
J ¼ 4.4 Hz, 1H), 7.33e7.29 (m, 4H), 7.21e7.15 (m, 6H), 7.10 (t,
2.4.6.2. Cu-MD-2 [37](yield: 91%) 1H NMR (400 MHz, CDCl3).
J ¼ 4.4 Hz, 2H); 13C NMR (100 MHz, CDCl3):
d
(ppm) 158.07, 155.20,
d
(ppm) 8.91 (d, J ¼ 4.8 Hz, 4H), 7.95 (t, J ¼ 5.2 Hz, 2H), 7.29e7.22 (m,
149.26, 147.04, 129.50, 127.78, 125.12, 123.86, 122.99; FAB-MS (m/z):
401 [M]þ; Anal. Calcd. for C26H19N5: C, 77.79; H, 4.77; N, 17.44;
found: C, 77.58; H, 4.53; N, 17.23.
12H), 7.17 (t, J ¼ 7.2 Hz, 16H), 7.00e6.95 (m, 24H), 6.71 (m, 4H); 13
C
NMR (100 MHz, CDCl3):
d (ppm) 158.26, 134.41, 132.80, 132.34,
130.72, 129.28, 125.42; 31P NMR (162 MHz, CDCl3):
(s); FAB-MS (m/z): 1360 [M-2BF4]þ; Anal. Calcd. for
80H62B2Cu2F8N4O2P4: C, 62.56; H, 4.07; N, 3.65; found: C, 62.37; H,
d
(ppm) ꢀ10.65
2.4.4. Synthesis of MD-PPO
C
Under a N2 atmosphere, PPO-B (0.14 g, 0.35 mmol), MD-Br
(0.07 g, 0.30 mmol) and Pd(PPh3)4 (0.035 g, 0.03 mmol) were added
to the mixture of THF (10 mL) and 2 M K2CO3 (3 mL). The reaction
mixture was stirred at 110 ꢁC for 16 h. After cooling to room tem-
perature, water (20 mL) was added and the mixture was extracted
with CH2Cl2 (3 ꢂ 30 mL). The organic phase was collected and dried
over Mg2SO4. After removing the solvent, the residue was purified
by column chromatography on silica gel with a mixture of ethyl
acetate (EA) and methanol (MeOH) (VEA/VMeOH ¼ 5:1) as eluent to
give the product as a white solid (0.10 g, Yield: 80%). 1H NMR
3.89; N, 3.47.
2.4.6.3. Cu-MD-3 (yield: 90%) 1H NMR (400 MHz, CDCl3).
8.54 (d, J ¼ 5.2 Hz, 4H), 7.86 (t, J ¼ 5.2 Hz, 2H), 7.68 (d, J ¼ 7.6 Hz,
4H), 7.22 (t, J ¼ 7.6 Hz, 8H), 7.18 (t, J ¼ 7.6 Hz, 4H), 7.07 (t, J ¼ 7.2 Hz,
16H), 6.90 (m, 16H), 6.61 (m, 4H); 1.77 (s, 12H); 13C NMR (100 MHz,
d (ppm)
CDCl3):
d (ppm) 158.05, 155.03, 134.06, 132.40, 131.48, 130.50,
130.29, 130.11, 129.93, 129.25, 129.20, 129.16, 128.05, 126.77, 125.37,
124.57, 118.62, 118.47, 118.31 36.11, 28.27; 31P NMR (162 MHz,
CDCl3):
d
(ppm) ꢀ13.13 (s); FAB-MS (m/z): 1440 [M-2BF4]þ; Anal.
(400 MHz, CDCl3):
7.90e7.85 (m, 2H), 7.79e7.77 (m, 2H), 7.73e7.68 (m, 4H), 7.60e7.56
(m, 2H), 7.52e7.45 (m, 5H); 13C NMR (100 MHz, CDCl3):
(ppm)
d
(ppm) 9.23 (s, 2H), 9.05 (d, J ¼ 4.8 Hz, 2H),
Calcd. for C86H70B2Cu2F8N4O2P4: C, 63.91; H, 4.37; N, 3.47; found: C,
63.79; H, 4.50; N, 3.35.
d
162.04, 161.58, 158.14, 156.10, 137.24, 134.53, 133.51, 133.29, 133.19,
132.58, 132.23, 132.20, 132.10, 132.00, 131.54, 128.72, 128.60, 127.25,
2.4.6.4. Cu-MD-4 (yield: 91%) 1H NMR (400 MHz, CDCl3).
d (ppm)
8.85 (d, J ¼ 5.2 Hz, 4H), 7.97 (t, J ¼ 5.2 Hz, 2H), 7.40 (d, J ¼ 8.0 Hz,
127.13, 121.63; 31P NMR (162 MHz, CDCl3):
d (ppm) 28.43 (s); FAB-
8H), 7.34 (d, J ¼ 8.4 Hz, 8H), 7.27e7.23 (m, 24H), 7.15e7.03 (m, 72H);
MS (m/z): 434 [M]þ; Anal. Calcd. for C26H19N4OP: C, 71.88; H, 4.41;
13C NMR (100 MHz, CDCl3):
d (ppm) 158.31, 148.08, 147.34, 143.15,
N, 12.90; found: C, 71.76; H, 4.48; N, 12.79.
133.61, 132.65, 132.46, 131.03, 130.80, 129.35, 128.07, 127.75, 127.22,
126.92, 124.76, 123.32, 123.22; 31F NMR (162 MHz, CDCl3):
(ppm)
2.91 (s); FAB-MS (m/z): 2304 [M-2BF4]þ; Anal. Calcd. for
152H118B2Cu2F8N8P4: C, 73.58; H, 4.79; N, 4.52; found: C, 73.63; H,
d
2.4.5. Synthesis of TPA-PPP
Under
a
N2 atmosphere, 4-(diphenylamino)phenylboronic
C
(0.82 g, 2.84 mmol), (4-bromophenyl)diphenylphosphine (0.80 g,
2.35 mmol) and Pd(PPh3)4 (0.15 g, 0.13 mmol) were added to the
mixture of THF (20 mL) and 2 M K2CO3 (6 mL). The reaction mixture
was stirred at 110 ꢁC for 16 h. After cooling to room temperature,
water (30 mL) was added and the mixture was extracted with
CH2Cl2 (3 ꢂ 60 mL). The organic phase was collected and dried over
Mg2SO4. After removal of the solvent, the residue was purified by
column chromatography on silica gel with a mixture of CH2Cl2
(DCM) and petroleum ether (PE) (bp 60e90 ꢁC) (VDCM/VPE ¼ 1:4) as
eluent to give the product as a white solid (0.98 g, Yield: 85%). 1H
4.84; N, 4.35.
2.4.6.5. Cu-MD-5 (yield: 90%) 1H NMR (400 MHz, CDCl3).
d (ppm)
8.60 (br, 4H), 7.72 (br, 1H), 7.37 (t, J ¼ 8.0 Hz, 4H), 7.30e7.27 (m,
10H), 7.21e7.14 (m, 60H); 13C NMR (100 MHz, CDCl3):
d
(ppm)
158.09, 153.15, 150.85, 148.04, 147.36, 146.29, 142.95, 133.58, 132.99,
130.79, 129.74, 129.40, 129.26, 128.36, 127.73, 127.17, 125.99, 124.80,
123.39, 123.13, 121.37; 31P NMR (162 MHz, CDCl3):
d (ppm) 4.48 (s);
FAB-MS (m/z): 1575 [M-2BF4]þ; Anal. Calcd. for C98H79B2Cu2F8N5P4:
C, 67.21; H, 4.55; N, 4.00; found: C, 67.05; H, 4.39; N, 3.82.
NMR (400 MHz, CDCl3):
d
(ppm) 7.54 (dd, J ¼ 1.4, 8.0 Hz, 2H), 7.47
(d, J ¼ 8.8 Hz, 2H), 7.37e7.33 (m, 10H), 7.29e7.25 (m, 6H), 7.14e7.11
2.4.6.6. Cu-MD-6 (yield: 92%) 1H NMR (400 MHz, CDCl3).
8.72 (br, 4H), 7.82 (br, 1H), 7.34e7.02 (m, 126H); 13C NMR (100 MHz,
CDCl3): (ppm) 147.96, 147.34, 132.95, 130.71, 129.68, 129.35,
129.22, 127.70, 127.09, 125.88, 124.73, 123.31, 123.14; 31P NMR
(162 MHz, CDCl3):
(ppm) 3.49 (s); FAB-MS (m/z): 2548 [M-2BF4]þ;
d (ppm)
(m, 6H), 7.05e7.17 (m, 2H); 31P NMR (162 MHz, CDCl3):
d
(ppm) ꢀ6.19 (s); FAB-MS (m/z): 505 [M]þ; Anal. Calcd. for
d
C
36H28NP: C, 85.52; H, 5.58; N, 2.77; found: C, 85.47; H, 5.49; N,
2.79.
d
Anal. Calcd. for C170H131B2Cu2F8N9P4: C, 74.94; H, 4.85; N, 4.63;
found: C, 74.69; H, 4.58; N, 4.37.
2.4.6. General procedure for the synthesis of the dinuclear Cu(I)
complexes
Under a N2 atmosphere, [(CH3CN)4Cu]BF4 (1.0 equiv) and the
organic phosphine ligand (2.0 equiv for PPh3 and TPA-PPP; 1.0
equiv for DPEPHOS and XANTPHOS) were mixed in CH2Cl2. After
stirring for 2 h at room temperature, the corresponding 2,2'-
bipyrimidine-based ligand (0.5 equiv) was added. The reaction was
allowed to proceed at room temperature for 4 h. The solvent was
removed under reduced pressure. The colored residue was dis-
solved in CH2Cl2 (2 mL) and precipitated in diethyl ether. The
precipitation was collected by filtration and washed with diethyl
ether three times. The title Cu(I) complexes were obtained as
colored solids in high yields.
2.4.6.7. Cu-MD-7 (yield: 93%) 1H NMR (400 MHz, CDCl3).
d (ppm)
8.69 (d, J ¼ 29.6 Hz, 4H), 7.95 (t, J ¼ 4.8 Hz, 2H), 7.79e7.74 (m, 4H),
7.59e7.52 (m, 8H), 7.33e7.13 (m, 61H); 13C NMR (100 MHz, CDCl3):
d
(ppm) 158.12, 154.84, 137.07, 133.48, 133.38, 132.97, 132.30, 132.12,
132.02, 130.79, 129.24, 128.88, 128.76, 127.94, 127.82, 126.56; 31P
NMR (162 MHz, CDCl3): d (ppm) 28.81, 4.61 (s); FAB-MS (m/z): 1608
[M-2BF4]þ; Anal. Calcd. for C98H79B2Cu2F8N4OP5: C, 65.97; H, 4.46;
N, 3.14; found: C, 65.79; H, 4.32; N, 2.89.
2.4.6.8. Cu-MD-8 (yield: 88%) 1H NMR (400 MHz, CDCl3).
8.80 (d, J ¼ 16.0 Hz, 4H), 7.88 (br, 2H), 7.77e7.86 (m, 4H), 7.43e7.36
d (ppm)

X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
155
(m, 24H), 7.28e6.99 (m, 97H); 13C NMR (100 MHz, CDCl3):
d
(ppm)
NMR spectra of the dinuclear Cu(I) complexes, the two resonance
sets with peaks at 9.0 ppm and 8.0 ppm indicate the protons on the
bipyrimidine core of the central ligands. The single resonance peak
located at ca. 3.0e4.0 ppm in the 31P NMR spectra can be assigned to
the phosphine moieties in either PPh3 or TPA-PPP from Cu-MD-1,
Cu-MD-4, Cu-MD-5 and Cu-MD-6. Owing to the influence of the
ether bond in the bis(2-diphenylphosphinophenyl)ether (DPE-
PHOS) and 4,5-bis(diphenylphosphino)-9,9-dimethylxanthene
(XANTPHOS), the chemical shifts for the phosphine moiety in Cu-
MD-2 (ꢀ10.65 ppm) and Cu-MD-3 (ꢀ13.13 ppm) are in the negative
region. In the 31P NMR spectra for both Cu-MD-7 and Cu-MD-8,
there are two singlet peaks which can be assigned to the ePOPh2
(29.0 ppm) and ePPh2 (4.0 ppm), respectively.
147.96, 147.32, 142.96, 133.63, 133.41, 132.79, 132.06, 131.96, 130.71,
129.34, 128.84, 128.72, 127.99, 127.87, 127.67, 127.10, 126.62, 124.71,
123.28, 123.18; 31P NMR (162 MHz, CDCl3):
d
(ppm) 28.58, 3.61 (s);
FAB-MS (m/z): 2582 [M-2BF4]þ; Anal. Calcd. for C170H131B2Cu2F8
-
N8OP5: C, 74.05; H, 4.79; N, 4.06; found: C, 73.93; H, 4.68; N, 3.98.
3. Results and discussion
3.1. Synthesis
The synthetic route for these complexes are shown in Scheme 2
and Scheme 3. The ligands, i.e., MD-TPA, MD-PPO and TPA-PPP,
were prepared through Suzuki cross-coupling reaction in high
yields. In the 31P NMR spectra, the single resonance peak at ca.
28.43 ppm and ꢀ6.19 ppm can be assigned to the ePOPh3 unit in
MD-PPO and ePPh3 unit in TPA-PPP, respectively.
The bipyrimidine-based dinuclear Cu(I) complexes can be easily
prepared in high yields by firstly treating the precursor Cu(I) com-
plex [(CH3CN)4Cu]BF4 with the corresponding organic phosphine
ligand, and then adding the 2,2'-bipyrimidine-based ligand at room
temperature. The target bipyrimidine-based dinuclear Cu(I) com-
plexes can be purified by precipitation in diethyl ether. In the 1H
3.2. Thermal stability
The thermal stability of these bipyrimidine-based dinuclear
Cu(I) complexes was characterized by thermogravimetric analysis
(TGA) (Table 1). For the complexes bearing small-size phosphine
ligand PPh3, Cu-MD-1, Cu-MD-5 and Cu-MD-7 exhibit noticeably
lower decomposition temperature (Td < 290 ꢁC) than the com-
plexes bearing either bidentate phosphine ligand DPEPHOS and
XANTPHOS (Cu-MD-2 Td ¼ 380 ꢁC and Cu-MD-3 Td ¼ 385 ꢁC) or
Scheme 2. Synthesis of the ligands.

156
X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
Scheme 3. Synthesis of the dinuclear Cu(I) complexes.
Table 1
Decomposition temperatures (Td) for these dinuclear Cu(I) complexes.
Complex
Cu-MD-1
Cu-MD-2
Cu-MD-3
Cu-MD-4
Cu-MD-5
Cu-MD-6
Cu-MD-7
Cu-MD-8
Td (ꢁC)
285
380
385
359
282
366
260
366
large-size phosphine ligand TPA-PPP (Cu-MD-4 Td ¼ 359 ꢁC, Cu-
MD-6 Td ¼ 366 ꢁC and Cu-MD-8 Td ¼ 366 ꢁC) (Table 1). It seems that
chemical tailoring of the central bipyrimidine ligand might not
exert a sound influence on the thermal stability of these complexes.
Hence, these TGA results clearly indicate that the thermal stability
of these dinuclear Cu(I) complexes is mainly decided by the nature
of the phosphine ligands. According to their structural features, the
decomposition of these dinuclear Cu(I) complexes should be
induced by the cleavage of the phosphine ligands. Reasonably, the
large-size and bidentate phosphine ligands are difficult to cleave
from the Cu(I) centers, in line with the aforementioned TGA results.
3.3. Photophysical properties and theoretical calculations
The UVeVis absorption spectra of these dinuclear Cu(I) com-
plexes in CH2Cl2 are shown in Fig. 1. All these dinuclear Cu(I)
complexes typically exhibit two sets of absorption bands. The
intense UV absorption bands before ca. 350 nm can be ascribed to
the
p-p* transition processes from both bridging bipyrimidine-
based ligands and the organic phosphine ligands. For Cu-MD-4,
Cu-MD-6 and Cu-MD-8, there is an intense absorption band located
at ca. 360 nm (Fig. 1 and Table 2), which can be assigned to the
p-p*
transition associated with the triphenylamine unit. This result has

X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
157
and the CueP
s-bonding orbital, while the LUMOs are contributed
by the -antibonding orbital of the central bipyrimidine-based li-
p
gands in Cu-MD-1, Cu-MD-2, Cu-MD-3 and Cu-MD-7. Hence, their
S1 states exhibit obvious metal-to-ligand charge transfer (MLCT)
character. Accordingly, the low-energy weak CT absorption bands
in Cu-MD-1, Cu-MD-2, Cu-MD-3 and Cu-MD-7 should be induced
by the MLCT transition. Differently, the H/L transition in Cu-MD-5
displays inter-ligand charge transfer (ILCT, i.e. from the triphenyl-
amine unit to the bipyrimidine segment) property with very minor
involvement from the Cu(I) centers. Thus, the S1 states from Cu-
MD-5 possess ILCT feature and its weak CT absorption band is
mainly assigned to ILCT as well. From the TD-DFT results (Table 3),
the oscillator strength (f) of the S0/S1 transition in Cu-MD-5 (ca.
0.6619) is nearly ten times higher than that for Cu-MD-1 (ca.
0.0699), Cu-MD-2 (ca. 0.0839), Cu-MD-3 (ca. 0.0549) and Cu-MD-7
(ca. 0.0664). The upshot is that Cu-MD-5 should show much
stronger CT absorption bands, which is in good agreement with the
experimental results in Fig.1. These results also indicate the validity
of the theoretical computational results in interpreting the ab-
sorption behavior of these dinuclear Cu(I) complexes. As for Cu-
MD-4 and Cu-MD-8, they have similar H/L transition patterns
(Fig. 2), indicating that their S0/S1 transitions are dominated by
the ILCTcharacter. Therefore, it is reasonable that Cu-MD-4 and Cu-
MD-8 show similar absorption spectra (Fig. 1).
The photoluminescence (PL) properties of these Cu(I) complexes
were characterized in doped PMMA film with 10 wt % doping level.
Fig. 3 shows the effect of the organic phosphine ligands on the
emission properties of the Cu(I) complexes. Compared with the
reference complex Cu-MD-1 displaying emission maximum at ca.
619 nm, both Cu-MD-2 bearing DPEPHOS and Cu-MD-3 bearing
XANTPHOS show red-shifted emissions with peaks at ca. 622 and
650 nm, respectively. Considering their large Stokes shift
(>100 nm), the shape of the decay curve and long lifetime in the
Fig. 1. UVeVis absorption spectra of the bipyrimidine-based dinuclear Cu(I)
complexes.
also been observed in other complexes bearing triphenylamine or
similar units [38,39]. Besides the intense absorption bands in the
UV region, the weak and broad absorption bands located in the long
wavelength region have been detected except for Cu-MD-4, Cu-
MD-6 and Cu-MD-8 (Fig. 1). According to the reported results for
Cu(I) complexes, these weak absorption bands can be ascribed to
the charge transfer (CT) transitions. The indiscernible absorptions
of Cu-MD-4, Cu-MD-6 and Cu-MD-8 in the long wavelength region
may be caused by the overlap between the weak absorption bands
and the strong p-p* transition bands from the triphenylamine unit.
Fig. 1 also indicates that the MLCT absorption bands of these Cu(I)
complexes can be red-shifted with an enhanced intensity by
introducing an electron-donating triphenylamine moiety to the
bridging bipyrimidine ligand. However, it seems that the electron-
withdrawing triphenylphosphine oxide unit shows an inappre-
ciable influence on the CT absorption bands of these Cu(I) com-
plexes as indicated by the similar CT absorption features of Cu-MD-
7 and Cu-MD-8 with respect to that of the reference complex Cu-
MD-1.
In order to gain insight of the absorption behaviors of these Cu(I)
complexes, the transition features of the representative complexes
have been obtained theoretically by the time-dependent density
functional theory (TD-DFT) calculations (Table 3). Due to its large
contribution (>90%) to the lowest singlet excited states (S1), the
HOMO/LUMO transition (H/L) features can represent the char-
acter of the S0/S1 transition in these dinuclear Cu(I) complexes. As
shown in Fig. 2, the HOMOs are mainly located on the Cu(I) center
order of microsecond (ms), these emissions were common phos-
phorescence. According to the TD-DFT results (Table 3), the char-
acter of the T1 states in Cu-MD-1, Cu-MD-2 and Cu-MD-3 can be
reflected by the features of the H/L transitions due to their high
contribution (>80%) to the T1 states. Hence, the frontier molecular
orbital (MO) patterns of Cu-MD-1, Cu-MD-2 and Cu-MD-3 can be
employed to rationalize the phosphorescent characters. The HO-
MOs of Cu-MD-1, Cu-MD-2 and Cu-MD-3 are mainly located on the
Cu(I) centers and the organic phosphine units. Clearly, the electron-
donating ether unit in both Cu-MD-2 and Cu-MD-3 will elevate
their HOMO level to facilitate the H/L transition and hence lower
the energy level of the T1 states. As a result, the phosphorescent
peaks of Cu-MD-2 and Cu-MD-3 are located in the longer wave-
length region compared with that of the reference complex Cu-
MD-1 (Fig. 3). Besides the ether linker, the phosphine ligands in Cu-
MD-3 contain another electron-donating C(CH3)2 linker, which will
Table 2
Photophysical data for bipyrimidine-based dinuclear Cu(I) complexes.
PLQYc
a
b
Complex
Absorption labs (nm)
Emission lemb (nm)
tp
(m
s)
Cu-MD-1
Cu-MD-2
Cu-MD-3
Cu-MD-4
Cu-MD-5
Cu-MD-6
Cu-MD-7
Cu-MD-8
248 (4.89), 285 (4.48), 362 (3.71)
619
622
650
421
671
676
662
422
1.1
3.6
4.5
ed
0.45
0.10
0.19
ed
246 (4.78), 288 (4.47), 376 (3.79), 505 (3.48)
248 (4.80), 286 (4.50), 376 (3.53), 505 (3.16)
248 (5.04), 293 (4.85), 358 (5.07)
258 (4.74), 296 (4.51), 460 (4.17), 545 (3.82)
260 (5.27), 296 (5.06), 357 (5.03), 463 (4.16), 550 (3.83)
258 (4.76), 325 (4.22), 365 (3.66)
5.7
5.4
3.2
ed
0.15
0.06
0.01
ed
246 (5.10), 295 (4.95), 357 (5.11)
a
Measured in CH2Cl2 at a concentration of 10ꢀ5 M and logε values are shown in parentheses.
Measured in doped PMMA film with the thickness of ca. 200 nm on quartz and the doping-level of 10-wt%.
Measured in an integrating sphere with 10-wt% doped PMMA film on quartz as sample and the excitation wavelength was set at 360 nm.
Cannot be detected properly.
b
c
d

158
X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
Table 3
Theoretical calculation results for the selected dinuclear Cu(I) complexes.a
Complex Contribution of dp
Contribution of dp
Calculated energy of T1
state (in wavelength)
Contribution of H/L Calculated energy of S1
Contribution of H/L
to the S1 state
f
orbitals to HOMOb
orbitals to LUMOb
to the T1 state
state (in wavelength)
(S0/S1)
Cu-MD- 29.21%
1
Cu-MD- 28.25%
2
Cu-MD- 28.18%
3
Cu-MD- 0.32%
5
1.77%
2.04%
1.50%
1.37%
1.57%
618.06 nm
661.00 nm
678.48 nm
887.67 nm
607.78 nm
83.3%
563.09 nm
97.0%
95.2%
93.2%
97.5%
96.9%
0.0699
0.0839
0.0549
0.6619
0.0664
83.1%
82.2%
93.6%
84.5%
599.48 nm
532.81 nm
623.32 nm
556.95 nm
Cu-MD- 29.06%
7
a
The TD-DFT results cannot be obtained properly for Cu-MD-4, Cu-MD-6 and Cu-MD-8 due to their large molecular size.
The data have been obtained by exporting DFT results with the software AOMix.
b
further rise the HOMO level of Cu-MD-3, leading to the more red-
shifted phosphorescent emission of Cu-MD-3 (Fig. 3).
strong electron-accepting capacity associated with the bipyr-
imidine segment will definitely facilitate the H/L transition in Cu-
MD-5 to lower the energy-level of its T1 states. Hence, its phos-
phorescent emission from the T1 states is located in the longer
wavelength region (Fig. 6). The complex Cu-MD-6 shows the H/L
transition with inter-ligand charge transfer from the triphenyl-
amine unit in TPA-PPP to the bipyrimidine segment. However, the
HOMOe4 of Cu-MD-6 is still located on the triphenylamine unit
attached to the bipyrimidine (Fig. S1). The theoretical results
indicate that the HOMOe4 of Cu-MD-6 also contributes to the
features of its S1 and T1 states. From this point of view, it is
reasonable that Cu-MD-5 and Cu-MD-6 should show similar ab-
sorption and emission properties, which have been supported by
the results in Figs.1 and 6. Indicated by the very low PLQYof ca. 0.01
for Cu-MD-7, introduction of an electron-withdrawing triphenyl-
phosphine oxide to the bipyrimidine segment will seriously
weaken the phosphorescent ability of these Cu(I) complexes. Ac-
cording to the MO patterns of Cu-MD-7 (Fig. 2), the LUMO is mainly
located on the bipyrimidine unit because the triphenylphosphine
oxide group can enhance electron-withdrawing ability of the
bipyrimidine unit. It will lower the LUMO level of Cu-MD-7 to
narrow the band gap to facilitate the aforementioned non-radiative
decay process of the T1 states (Fig. 5a). Hence, Cu-MD-7 only emits
very weak phosphorescence with a red-shifted peak at 662 nm
(Fig. 6). The locations of these experimental emission peaks are in
good agreement with the trend of the calculated phosphorescent
wavelengths (Table 3).
However, only very weak emission band located at ca. 420 nm
was detected for Cu-MD-4 using TPA-PPP as the organic phosphine
ligand. The PLQY and lifetime cannot be determined properly. We
tentatively assign this emission to the ligand-centered fluorescence
because a similar emission is apparent in TPA-PPP, which means
that the TPA-PPP ligand should disfavor the phosphorescent
emission from Cu(I) complexes. This conclusion is properly sup-
ported by the emission properties of Cu-MD-6 and Cu-MD-8 which
also contain TPA-PPP ligand. For Cu-MD-8, nearly no phospho-
rescent emission can be observed as well (Fig. 4). Despite its
phosphorescent emission, Cu-MD-6 still shows much lower PLQY
of ca. 0.06 with respect to its analogue Cu-MD-5 (PLQY ¼ ca. 0.15)
without TPA-PPP ligand (Table 2). As the theoretical calculation
results indicated (see Table S1 in Supporting Information), the
complexes bearing TPA-PPP possess much narrower energy gaps
(Eg < 0.7 eV), which implies that the potential-energy profile of S0
state will get much closer to that of the S1 state in Cu-MD-4, Cu-
MD-6 and Cu-MD-8. This might provide the chance to form a cross-
point between the potential-energy profiles of S0 and T1 states
(Fig. 5a), which greatly increases the opportunity for the non-
radiative decay process from the T1 states to the S0 states through
the cross-point (Fig. 5a). The non-radiative decay process will
definitely weaken the phosphorescent emission. Therefore, Cu-
MD-4, Cu-MD-6 and Cu-MD-8 just exhibit weak or even no phos-
phorescence (Fig. 4). Different from Cu-MD-4, Cu-MD-6 and Cu-
MD-8, the other complexes possess wider band gap in the range
from 2.28 to 2.98 eV (Table S1) and hence there should be no cross-
point between the potential-energy profiles of their T1 and S0 states
(Fig. 5b). Therefore, their T1 states can effectively decay to the S0
states radiatively to induce phosphorescent emission (Figs. 3 and
5). What should be noted is that introducing TPA-PPP ligand can
effectively confine the transitions of Cu-MD-4, Cu-MD-6 and Cu-
MD-8 on their organic ligands, which will greatly weaken the spin-
orbital coupling effect to reduce the quantum yield of T1 states. This
might be another reason for the low phosphorescent ability of Cu-
MD-4, Cu-MD-6 and Cu-MD-8.
3.4. Electrochemical properties
The electrochemical properties of these bipyrimidine-based
dinuclear Cu(I) complexes were characterized by cyclic voltam-
metry (CV) using ferrocene/ferrocenium (Fc/Fcþ) redox couple as
an internal reference under a nitrogen atmosphere. The results are
summarized in Table 4. Except for Cu-MD-1 and Cu-MD-7, all the
other complexes show two anodic potentials (Ea). For the com-
plexes Cu-MD-4, Cu-MD-5, Cu-MD-6 and Cu-MD-8, the first Ea
located in the low potential region from ca. 0.40 V to 0.46 V should
be assigned to the reversible oxidation process of the triphenyl-
amine group according to our previous results for the IrIII com-
plexes [35,40]. The first Ea for Cu-MD-5 (ca. 0.46 V) is slightly more
positive with respect to that for Cu-MD-4, Cu-MD-6 and Cu-MD-8.
This result can be ascribed to the fact that the electron-
withdrawing bipyrimidine will make the attached triphenylamine
group in Cu-MD-5 more reluctant to be oxidized. However, as the
analogue of Cu-MD-5, Cu-MD-6 exhibits the smaller value (ca.
0.4 V) for its first Ea. According to what aforementioned, the tri-
phenylamine group attached to the bipyrimidine in Cu-MD-6
The effect of tuning the structure of the bridging bipyrimidine
ligand on the PL behavior of these dinuclear Cu(I) complexes can be
seen clearly from Fig. 6. Compared with that of their reference
complex Cu-MD-1 (lem ¼ 619 nm), the emission maxima for Cu-
MD-5 (ca. 671 nm) and Cu-MD-6 (ca. 676 nm) show an obvious
bathochromic effect due to the introduction of an electron-
donating triphenylamine unit to the bipyrimidine ligand. As
aforementioned, the H/L transition in Cu-MD-5 displays ILCT (i.e.
from triphenylamine to bipyrimidine) features. The strong
electron-donating ability of triphenylamine unit together with the

X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
159
Fig. 2. HOMO (down) and LUMO (up) patterns for these dinuclear Cu(I) complexes.
should be more difficult to be oxidized compared with the ones
with strong electron-donating triphenylphosphine in the ligand
TPA-PPP. Hence, the first oxidation process in the CV measurement
of Cu-MD-6 should be induced by the four triphenylamine blocks
from the TPA-PPP ligands. Reasonably, the strong electron-
donating triphenylphosphine unit will greatly facilitate oxidation
of the triphenylamine unit of the ligand TPA-PPP in Cu-MD-6. In
addition, different from Cu-MD-4 and Cu-MD-8, the electron-
donating triphenylamine unit attached to the central ligand of
Cu-MD-6 will weaken the electron-accepting ability of bipyr-
imidine from Cu(I) centers and the triphenylphosphine unit in TPA-
PPP ligands. Therefore, the triphenylphosphine units in TPA-PPP
ligands of Cu-MD-6 can exhibit stronger electron-donating ability
to the attached triphenylamine units than those in Cu-MD-4 (ca.
0.42 V) and Cu-MD-8 (ca. 0.45 V). Hence, Cu-MD-6 exhibits the
lowest potential for its first Ea (Table 4). On this basis, it is also easy
to understand the order for the first Ea of Cu-MD-8 > Cu-MD-
4 > Cu-MD-6. Without triphenylamine units, both Cu-MD-2 and
Cu-MD-3 can show an oxidation process at a lower potential region
as well (Table 4). It might be induced by the ether-linked units
showing electron-donating and conjugation features.
Besides the oxidation wave at a low potential, all the dinuclear

160
X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
Fig. 3. PL spectra for Cu-MD-1, Cu-MD-2, Cu-MD-3 and Cu-MD-4 in doped PMMA film
with 10 wt % doping level at 298 K.
Fig. 6. PL spectra for Cu-MD-1, Cu-MD-5, Cu-MD-6 and Cu-MD-7 in doped PMMA film
with 10 wt % doping level at 298 K.
Cu(I) complexes can show irreversible oxidation process with much
higher Ea in the range from ca. 0.58 V to 0.88 V (Table 4), which can
be assigned to the oxidation of the Cu(I) centers. The second Ea
located at the much more positive potential region for Cu-MD-1
and Cu-MD-7 suggests that introducing electron-donating unit to
either central or phosphine ligands can effectively reduce the
oxidation potential of the Cu(I) centers. However, it seems that
introducing electron-donating unit to the phosphine ligands is
more effective in promoting the oxidation of the Cu(I) centers, since
Cu-MD-4, Cu-MD-6 and Cu-MD-8 show the second Ea (<0.64 V)
with much lower value with respect to other complexes. Even with
relatively weaker electron-donating ether linker, Cu-MD-2 and Cu-
MD-3 still possess lower Ea (ca. 0.71 V and 0.69 V, respectively) than
that of the reference complex Cu-MD-1 (ca. 0.85 V). Among all the
complexes, Cu-MD-7 possesses the highest Ea for the Cu(I) centers
(ca. 0.88 V) due to the fact that the electron-withdrawing triphe-
nylphosphine oxide unit can reduce the electron density on the
Cu(I) centers to make them more difficult to be oxidized.
Fig. 4. PL spectra for Cu-MD-4, Cu-MD-6 and Cu-MD-8 in doped PMMA film with
10 wt % doping level at 298 K.
According to the MO patterns of these dinuclear Cu(I) complexes
(Fig. 2), their reduction process should be induced by the
Fig. 5. Different patterns for the potential-energy profiles involved in these dinuclear Cu(I) complexes. (a) With cross-point between potential-energy profiles of T1 and S0 states; (b)
Without cross-point between potential-energy profiles of T1 and S0 states.

X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
161
Table 4
Redox properties of the bipyrimidine-based dinuclear Cu(I) complexes.
Complex
Ea (V)
Ec (V)
EHOMOc (eV)
ELUMOc (eV)
Cu-MD-1
Cu-MD-2
Cu-MD-3
Cu-MD-4
Cu-MD-5
Cu-MD-6
Cu-MD-7
Cu-MD-8
0.85a
ꢀ1.79a, ꢀ2.04b, ꢀ2.35b
ꢀ5.45
ꢀ5.27
ꢀ5.26
ꢀ5.22
ꢀ5.26
ꢀ5.20
ꢀ5.64
ꢀ5.25
ꢀ3.81
ꢀ2.64
ꢀ2.63
ꢀ3.22
ꢀ3.11
ꢀ3.16
ꢀ3.76
ꢀ3.91
0.47b, 0.71a
0.46b, 0.69a
0.42b, 0.61a
0.46b, 0.66a
0.40b, 0.58a
0.88a
ꢀ2.47a
ꢀ2.49a
ꢀ2.47a
ꢀ2.51a, ꢀ2.69a
ꢀ2.52a
ꢀ1.33a, ꢀ1.70a, ꢀ2.14a, ꢀ2.99a
0.45b, 0.63a
ꢀ1.35a, ꢀ1.72a, ꢀ2.15b, ꢀ3.12b
a
Irreversible or quasi-reversible. The value was derived from the anodic or cathodic peak potential.
b
c
Reversible. The value was set as E1/2
.
HOMO levels are calculated according to the E1/2 of the first reversible oxidation wave or the onset potential of the first irreversible oxidation wave, respectively. LUMO
levels are derived from the onset potential of the first irreversible reduction wave.
bipyrimidine unit. Compared with the reference complex Cu-MD-1
with the first cathodic potential (Ec) at ca. ꢀ1.79 V, the complexes
Cu-MD-2 e Cu-MD-6 exhibit the Ec in a more negative potential
region (from ca. ꢀ2.47 V to ꢀ2.52 V) due to the presence of
electron-donating ligands which make the bipyrimidine units more
difficult to be reduced. Bonded directly with strong electron-
donating triphenylamine group, the bipyrimidine units in Cu-
MD-5 and Cu-MD-6 are even more difficult to be reduced. Hence,
their first Ec is located after ca. ꢀ2.50 V. For Cu-MD-7 and Cu-MD-8,
they possess a reduction process at ca. ꢀ1.3 V, which can be
assigned to the reduction of the triphenylphosphine oxide units.
Owing to the electron-withdrawing triphenylphosphine oxide
units attached to the bipyrimidine units, the second Ec of Cu-MD-7
and Cu-MD-8 is located in a less negative potential region
(ca. ꢀ1.72 V) with reference to that of the reference complex Cu-
MD-1.
dopants by solution processing approach due to their higher PLQYs.
The devices based on the Cu(I) phosphorescent emitters possess
the configuration of ITO/PEDOT:PSS (35 nm)/TCTA: Cu(I) complex
(x wt %, 50 nm)/TPBI (40 nm)/LiF (1 nm)/Al (100 nm) (Fig. 7). In
these OLEDs, 4,40,400-tri(N-carbazolyl)triphenylamine (TCTA) is the
host material and 1,3,5-tris (N-phenylbenzimidazole-2-yl)benzene
(TPBI) serves the function of electron-transporting and hole-
blocking because of its relatively low HOMO level. With the aim
to optimize their EL performances, these phosphorescent emitters
have been doped at different doping levels (Fig. 7). Table 5 sum-
marizes the important EL performance data for these devices.
After applying a proper voltage, these OLEDs exhibit intense
electroluminescence. As shown in Fig. 8, the EL spectral profile of
each optimized device is very similar to the PL spectrum of the
corresponding Cu(I) complex, suggesting that the EL emission is
indeed originated from the triplet excited state of the Cu(I) com-
plex. Additionally, no residual emission from TCTA is observed in
these devices, implying the efficient forward energy transfer from
the host excitons to the phosphorescent Cu(I) complexes.
3.5. Electroluminescence properties
As listed in Table 5, the OLEDs with the doping level at 10 wt %,
i.e., devices A2, B2 and C2, show the highest EL efficiencies. The
current densityevoltageeluminance (JꢀVꢀL) characteristics and EL
In order to characterize the electroluminescence properties of
these bipyrimidine-based dinuclear Cu(I) complexes, the OLEDs
have been constructed using Cu-MD-1, Cu-MD-3 and Cu-MD-5 as
Fig. 7. Configuration of the OLEDs and chemical structures for the materials involved.

162
X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
Table 5
The EL performance of these dinuclear Cu(I) complexes.
Lmax (cd mꢀ2
)
EQE (%)
CE (cd Aꢀ1
)
PE (lm Wꢀ1
)
lmaxd (nm)
CRI/CCTe
a
Device
Dopant
Vturn-on (V)
A1
Cu-MD-1 (8 wt %)
4.9
3318 (13.6)
1.84 (8.6)a
1.11b
3.90 (8.6)
2.31
2.80
1.53 (7.7)
1.30
1.52
572
(0.43, 0.47)
75/3541 K
1.79c
A2
A3
B1
B2
B3
C1
C2
C3
Cu-MD-1 (10 wt %)
Cu-MD-1 (12 wt %)
Cu-MD-3 (8 wt %)
Cu-MD-3 (10 wt %)
Cu-MD-3 (12 wt %)
Cu-MD-5 (8 wt %)
Cu-MD-5 (10 wt %)
Cu-MD-5 (12 wt %)
4.3
4.8
4.6
4.3
4.7
4.8
4.3
4.5
5379 (11.3)
943 (11.5)
3131 (13.6)
3697 (10.8)
1248 (11.2)
462 (12.1)
2466 (11.1)
670 (11.0)
6.09 (6.8)
3.95
6.06
2.87 (6.7)
2.24
2.71
2.34 (6.3)
2.16
2.12
4.03 (6.6)
3.00
3.80
1.54 (6.9)
1.50
1.43
1.00 (7.9)
0.98
0.95
2.17 (6.9)
2.01
2.12
1.32 (6.5)
1.23
12.78 (6.8)
7.85
12.74
6.15 (6.7)
4.79
5.81
4.96 (6.3)
4.51
4.43
8.78 (6.6)
6.56
8.27
3.33 (6.9)
3.17
3.11
1.55 (7.9)
1.54
1.47
3.22 (6.9)
3.09
3.14
1.72 (6.5)
1.58
5.93 (6.5)
4.51
5.79
2.94 (6.5)
2.56
2.50
2.52 (6.0)
2.45
1.80
4.27 (6.3)
3.59
3.38
1.56 (6.7)
1.55
1.25
0.68 (6.4)
0.64
0.50
1.53 (6.4)
1.50
1.30
0.85 (6.2)
0.62
574
(0.44, 0.47)
75/3524 K
75/3674 K
78/4039 K
77/4037 K
76/4042 K
80/2518 K
82/2439 K
88/2342 K
574
(0.43, 0.47)
558
(0.40, 0.46)
558
(0.40, 0.47)
558
(0.40, 0.46)
634
(0.50, 0.46)
637
(0.50, 0.45)
652
(0.50, 0.44)
1.02
1.33
0.44
a
Maximum values of the devices. Values in parentheses are the voltages at which they were obtained.
Values were collected at 20 cd mꢀ2
b
c
.
Values collected at 100 cd mꢀ2
.
d
e
Values were collected at 8 V and CIE coordinates (x, y) are shown in parentheses.
Color Rendering Index (CRI) and Correlated Color Temperature (CCT) obtained at 8 V.
Fig. 9. The current densityevoltageeluminance (JꢀVꢀL) curves for the optimized
OLEDs.
Fig. 8. EL spectra for the optimized devices based on these dinuclear phosphorescent
Cu(I) complexes at about 8 V.
emitters typically show low EL efficiencies. However, device C2
with the EL maximum at ca. 640 can still show good EL efficiencies
with CE of 3.22 cd Aꢀ1, PE of 1.53 lm Wꢀ1 and EQE of 2.17%.
Although these charged dinuclear Cu(I) complexes exhibit inferior
EL performance with respect to some neutral dinuclear Cu(I) ana-
logues [23,41], complex Cu-MD-1 can still outperform the
advanced charged mononuclear Cu(I) counterparts [42,43]. The
charged green-emitting mononuclear Cu(I) complex bearing 1,10-
phenanthroline-type ligand can show high CE of 11.0 cd Aꢀ1 [12],
while the analogue with quinoline-based ligand can show CE of
5.58 cd Aꢀ1 [42]. The impressive EL performances achieved by these
dinuclear Cu(I) emitters should be ascribed to the enhanced
charge-trapping ability afforded by both bipyrimidine ligand and
efficiencyeluminance curves for the optimized OLEDs are shown in
Fig. 9 and Fig. 10. The optimized device (device A2) doped with Cu-
MD-1 can exhibit the best EL performance with a turn-on voltage
(Vturn-on
) of 4.3 V, and the maximum luminance (Lmax) of
5379 cd mꢀ2 at 11.3 V, the peak current efficiency (CE) of
12.78 cd Aꢀ1, power efficiency (PE) of 5.93 lm Wꢀ1 and external
quantum efficiency (EQE) of 6.09% (Table 5 and Fig. 10), repre-
senting impressive EL efficiencies for the OLEDs based on charged
dinuclear phosphorescent Cu(I) complexes. The device B2 with Cu-
MD-3 as emitter can also give decent EL performances with Lmax of
3697 cd mꢀ2 at 10.8 V, CE of 8.78 cd Aꢀ1, PE of 4.27 lm Wꢀ1 and EQE
of 4.03%. According to the energy gap law, long-wavelength

X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
163
4. Conclusion
Through introduction of the functional moieties with different
electronic features to the bipyrimidine unit, a series of charged
bipyrimidine-based dinuclear Cu(I) complexes bearing different
phosphine ligands have been developed. The chemical structures of
both central bipyrimidine ligands and organic phosphine ligands of
the dinuclear Cu(I) complexes can exert a great influence on their
thermal stability as well as photophysical properties. In addition,
the functional groups can also promote the charge-trapping of
these dinuclear Cu(I) complexes to benefit their EL ability. Hence,
these bipyrimidine-based dinuclear Cu(I) complexes can exhibit
impressive EL performances with Lmax of 5379 cd mꢀ2, EQE of
6.09%, CE of 12.78 cd Aꢀ1, and PE of 5.93 lm Wꢀ1, representing the
top-ranking EL efficiencies ever achieved by the charged Cu(I)
complexes. More importantly, the OLEDs based on these dinuclear
Cu(I) complexes can emit intense warm white light with CRI as high
as 88, showing the great potential of these dinuclear Cu(I) com-
plexes in achieving single-emitter WOLEDs. All these results will
not only establish the structure-property relationship of these
dinuclear Cu(I) complexes, but also provide valuable information
for developing highly efficient Cu(I) phosphorescent emitters for
improving the EL performance of the OLEDs based on charged Cu(I)
complexes.
Acknowledgments
This work was supported by the National Natural Science
Foundation of China (Nos. 20902072, 1572176 and 21602170), the
Natural Science Foundation of Shaanxi Province (No. 2016JQ2011),
Fundamental Research Funds for the Central Universities (Nos.
cxtd2015003 and 01091191320074), the China Postdoctoral Science
Foundation (Nos. 20130201110034 and 2015M580831), Creative
Scientific Research Team for Yulin City in Shannxi Province. W.-Y.W.
acknowledges the financial support from the National Basic
Research Program of China (973 Program, grant no.
2013CB834702), the Areas of Excellence Scheme, University Grants
Committee of HKSAR, China (AoE/P-03/08), Hong Kong Research
Grants Council (HKBU 12304715), Hong Kong Baptist University
(FRG1/14-15/084) and the Hong Kong Polytechnic University.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.dyepig.2017.04.036.
References
[1] Buckner MT, McMillin DR. Photoluminescence from copper(I) complexes with
low-lying metal-to-ligand charge transfer excited states. J Chem Soc Chem
Commun 1978;(17):759e61.
[2] Scaltrito DV, Thompson DW, O'Callaghan JA, Meyer GJ. MLCT excited states of
cuprous bis-phenanthroline coordination compounds. Coord Chem Rev
2000;208(1):243e66.
Fig. 10. The relationship between the EL efficiency and luminance for the optimized
OLEDs. (a) Device A2, (b) device B2 and (c) device C2.
[3] Armaroli N. Photoactive mono- and polynuclear Cu(I)-phenanthrolines.
a
the functional groups as indicated by the CV results (Table 4).
Interestingly, the OLEDs based on these charged dinuclear Cu(I)
complexes can show very broad EL spectra with CRI > 75 and
maximum at 88 (Fig. 8 and Table 5). As a result, the CIE coordinates
for the devices based on Cu-MD-1 and Cu-MD-3 are very close to
that of some warm WOLEDs based on complementary-color strat-
egy with CIE of (0.395, 0.452) or (0.376, 0.467) [44]. These results
demonstrate that the bipyrimidine-based dinuclear Cu(I) com-
plexes are very promising for fabricating single-emitter warm
WOLEDs, which can greatly reduce the complexity of the fabrica-
tion process as well as the cost.
viable alternative to Ru(II)-polypyridines? Chem Soc Rev 2001;30(2):113e24.
[4] McMillin DR, McNett KM. Photoprocesses of copper complexes that bind to
DNA. Chem Rev 1998;98(3):1201e20.
[5] Chen LX. Probing transient molecular structures in photochemical processes
using laser-initiated time-resolved X-ray absorption spectroscopy. Annu Rev
Phys Chem 2005;56(1):221e54.
[6] Everly RM, McMillin DR. Concentration-dependent lifetimes of Cu (NN)2þ
systems: exciplex quenching from the ion pair state. Photochem Photobiol
1989;50(6):711e6.
[7] Cuttell DG, Kuang S-M, Fanwick PE, McMillin DR, Walton RA. Simple Cu(I)
complexes with unprecedented excited-state lifetimes.
J Am Chem Soc
2002;124(1):6e7.
[8] Harkins SB, Peters JC. A highly emissive Cu2N2 diamond core complex sup-
ported by a [PNP]-ligand. J Am Chem Soc 2005;127(7):2030e1.
[9] Miller AJM, Dempsey JL, Peters JC. Long-lived and efficient emission from

164
X. Yang et al. / Dyes and Pigments 143 (2017) 151e164
mononuclear amidophosphine complexes of copper. Inorg Chem
2007;46(18):7244e6.
light-emitting device with an external quantum efficiency over 20%. Adv
Mater 2013;25(18):2573e6.
[10] Czerwieniec R, Yu J, Yersin H. Blue-light emission of Cu(I) complexes and
singlet harvesting. Inorg Chem 2011;50(17):8293e301.
[28] Chen Y-H, Tang K-C, Chen Y-T, Shen J-Y, Wu Y-S, Liu S-H, et al. Insight into the
mechanism and outcoupling enhancement of excimer-associated white light
generation. Chem Sci 2016;7(6):3556e63.
[29] Wadt WR, Hay PJ. Ab initio effective core potentials for molecular calculations.
Potentials for main group elements Na to Bi. J Chem Phys 1985;82(1):284e98.
[30] Hay PJ, Wadt WR. Ab initio effective core potentials for molecular calculations.
Potentials for K to Au including the outermost core orbitals. J Chem Phys
1985;82(1):299e310.
[11] Wei F, Qiu J, Liu X, Wang J, Wei H, Wang Z, et al. Efficient orange-red phos-
phorescent organic light-emitting diodes using an in situ synthesized cop-
per(i) complex as the emitter. J Mater Chem C 2014;2(31):6333e41.
[12] Zhang Q, Zhou Q, Cheng Y, Wang L, Ma D, Jing X, et al. Highly efficient green
phosphorescent organic light-emitting diodes based on CuI complexes. Adv
Mater 2004;16(5):432e6.
[13] Yang X, Zhou G, Wong W-Y. Functionalization of phosphorescent emitters and
their host materials by main-group elements for phosphorescent organic
light-emitting devices. Chem Soc Rev 2015;44(23):8484e575.
[14] Hay PJ. Theoretical studies of the ground and excited electronic states in
cyclometalated phenylpyridine Ir(III) complexes using density functional
theory. J Phys Chem A 2002;106(8):1634e41.
[15] Lamansky S, Djurovich P, Murphy D, Abdel-Razzaq F, Lee HE, Adachi C, et al.
Highly phosphorescent bis-cyclometalated iridium complexes: synthesis,
photophysical characterization, and use in organic light emitting diodes. J Am
Chem Soc 2001;123(18):4304e12.
[16] Brooks J, Babayan Y, Lamansky S, Djurovich PI, Tsyba I, Bau R, et al. Synthesis
and characterization of phosphorescent cyclometalated platinum complexes.
Inorg Chem 2002;41(12):3055e66.
[17] Ma Y, Che C-M, Chao H-Y, Zhou X, Chan W-H, Shen J. High luminescence
gold(I) and copper(I) complexes with a triplet excited state for use in light-
emitting diodes. Adv Mater 1999;11(10):852e7.
[18] Che G, Su Z, Li W, Chu B, Li M, Hu Z, et al. Highly efficient and color-tuning
electrophosphorescent devices based on CuI complex. Appl Phys Lett
2006;89(10):103511.
[31] Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, et al.
Gaussian 09, revision A.02. Wallingford CT: Gaussian, Inc; 2009.
[32] R. Pis Diez, MullPop, The National University of La Plata, Argentina.
[33] Schaftenaar G. Molden v3.7. Toernooiveld, Nijmegen, The Netherlands: CAOS/
CAMM Center Nijmegen; 2001.
[34] Kubas GJ, Monzyk B, Crumbliss AL. Tetrakis(Acetonitrile)Copper(I) hexa-
fluorophosphate. Inorg Synth: John Wiley & Sons, Inc.; 2007. p. 90e2.
[35] Zhou GJ, Ho CL, Wong WY, Wang Q, Ma DG, Wang LX, et al. Manipulating
charge-transfer character with electron-withdrawing main-group moieties
for the color tuning of iridium electrophosphors. Adv Funct Mater 2008;18(3):
499e511.
[36] Kuang S-M, Cuttell DG, McMillin DR, Fanwick PE, Walton RA. Synthesis and
structural characterization of Cu(I) and Ni(II) complexes that contain the bis
[2-(diphenylphosphino)phenyl]ether ligand. novel emission properties for the
Cu(I) species. Inorg Chem 2002;41(12):3313e22.
[37] Linfoot CL, Richardson P, Hewat TE, Moudam O, Forde MM, Collins A, et al.
Substituted [Cu(I)(POP)(bipyridyl)] and related complexes: synthesis, struc-
ture, properties and applications to dye-sensitised solar cells. Dalton Trans
2010;39(38):8945e56.
[19] Osawa M, Hoshino M, Hashimoto M, Kawata I, Igawa S, Yashima M. Appli-
cation of three-coordinate copper(I) complexes with halide ligands in organic
light-emitting diodes that exhibit delayed fluorescence. Dalton Trans
2015;44(18):8369e78.
[38] Yang X, Zhao Y, Zhang X, Li R, Dang J, Li Y, et al. Thiazole-based metal-
lophosphors of iridium with balanced carrier injection/transporting features
and their two-colour WOLEDs fabricated by both vacuum deposition and
solution processing-vacuum deposition hybrid strategy.
J Mater Chem
[20] Zhang J, Duan C, Han C, Yang H, Wei Y, Xu H. Balanced dual emissions from
tridentate phosphine-coordinate copper(I) complexes toward highly efficient
yellow OLEDs. Adv Mater 2016;28(28):5975e9.
[21] Liu ZW, Qayyum MF, Wu C, Whited MT, Djurovich PI, Hodgson KO, et al.
A codeposition route to CuI-pyridine coordination complexes for organic
light-emitting diodes. J Am Chem Soc 2011;133(11):3700e3.
2012;22(15):7136e48.
[39] Xu X, Yang X, Wu Y, Zhou G, Wu C, Wong WY. Tris-heteroleptic cyclo-
metalated iridium(III) complexes with ambipolar or electron injection/trans-
port features for highly efficient electrophosphorescent devices. Chem Asian J
2015;10(1):252e62.
[40] Xu X, Yang X, Dang J, Zhou G, Wu Y, Li H, et al. Trifunctional Ir-III ppy-type
asymmetric phosphorescent emitters with ambipolar features for highly
efficient electroluminescent devices. Chem Commun 2014;50(19):2473e6.
[41] Zhang QS, Komino T, Huang SP, Matsunami S, Goushi K, Adachi C. Triplet
exciton confinement in green organic light-emitting diodes containing
luminescent charge-transfer Cu(I) complexes. Adv Funct Mater 2012;22(11):
2327e36.
[22] Liu Z, Qiu J, Wei F, Wang J, Liu X, Helander MG, et al. Simple and high effi-
ciency phosphorescence organic light-emitting diodes with codeposited
copper(I) emitter. Chem Mater 2014;26(7):2368e73.
ꢀ
[23] Volz D, Chen Y, Wallesch M, Liu R, Flechon C, Zink DM, et al. Bridging the
efficiency gap: fully bridged dinuclear Cu(I)-complexes for singlet harvesting
in high-efficiency OLEDs. Adv Mater 2015;27(15):2538e43.
[24] Vogler C, Hausen H-D, Kaim W, Kohlmann S, Kramer HEA, Rieker J. Copper(I)-
assisted formation of an “organic” sandwich structure: structural pre-
[42] Min J, Zhang Q, Sun W, Cheng Y, Wang L. Neutral copper(I) phosphorescent
complexes from their ionic counterparts with 2-(2[prime or minute]-
quinolyl)benzimidazole and phosphine mixed ligands. Dalton Trans
2011;40(3):686e93.
[43] Sun W, Zhang Q, Qin L, Cheng Y, Xie Z, Lu C, et al. Phosphorescent cuprous
complexes with N, O ligands-synthesis, photoluminescence, and electrolu-
minescence. Eur J Inorg Chem 2010;2010(25):4009e17.
[44] Wu H, Zhou G, Zou J, Ho C-L, Wong W-Y, Yang W, et al. Efficient polymer
white-light-emitting devices for solid-state lighting. Adv Mater 2009;21(41):
4181e4.
requisites for luminescence of the dinuclear complexes [(m-Bipyrimidine)
{Cu(PR3)2}2]X2. Angewandte Chemie Int Ed Engl 1989;28(12):1659e60.
[25] Chen J-L, Wu B, Gu W, Cao X-F, Wen H-R, Hong R, et al. Synthesis and char-
acterization of mono- and dinuclear copper(I) complexes with 3-(2-
pyrimidinyl)-1,2,4-triazine. Transit Met Chem 2011;36(4):379e85.
[26] Tan C-H, Ma X, Zhu Q-L, Huang Y-H, Fu R-B, Hu S-M, et al. Syntheses, crystal
structures and properties of dinuclear copper(I) complexes. J Mol Struct
2012;1007:26e30.
[27] Fleetham T, Ecton J, Wang ZX, Bakken N, Li J. Single-doped white organic