Neutral copper(i) phosphorescent complexes from their ionic counterparts with 2-(2′-quinolyl)benzimidazole and phosphine mixed ligands

Article abstract of DOI:10.1039/c0dt01031f

Neutral mononuclear Cu^I complexes and their counterparts with counterion, i.e. Cu(qbm)(PPh₃)₂, Cu(qbm)(DPEphos), [Cu(Hqbm)(PPh₃)₂](BF₄) and [Cu(Hqbm)(DPEphos)] (BF₄), where Hqbm

Full text of DOI:10.1039/c0dt01031f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 686
www.rsc.org/dalton
PAPER
Neutral copper(I) phosphorescent complexes from their ionic counterparts
with 2-(2¢-quinolyl)benzimidazole and phosphine mixed ligands†
Junhui Min,a,c Qisheng Zhang,b Wei Sun,a,c Yanxiang Cheng*a,b and Lixiang Wang*a
Received 17th August 2010, Accepted 26th October 2010
DOI: 10.1039/c0dt01031f
Neutral mononuclear CuI complexes and their counterparts with counterion, i.e. Cu(qbm)(PPh3)2,
Cu(qbm)(DPEphos), [Cu(Hqbm)(PPh3)2](BF4) and [Cu(Hqbm)(DPEphos)](BF4), where Hqbm =
2-(2¢-quinolyl)benzimidazole, DPEphos = bis[2-(diphenylphosphino)phenyl]ether, have been
synthesized and characterized by X-ray structure analyses. All of the four complexes in solid state
exhibit a strong phosphorescence band in the orange spectral region at room temperature. The
photophysical properties of these complexes in both methylene chloride solution and poly(methyl
methacrylate) film have been studied. Compared to the related cationic complexes, the neutral ones
show blue-shifted emissions and longer lifetimes that can be attributed to the additional ligand-centered
p–p* transition beside traditional metal-to-ligand charge-transfer (MLCT). By doping these complexes
in N-(4-(carbazol-9-yl)phenyl)-3,6-bis(carbazol-9-yl) carbazole (TCCz), multilayer organic
light-emitting diodes (OLEDs) were fabricated with the device structure of ITO/PEDOT/TCCz: CuI
(10 wt%)/BCP/Alq3/LiF/Al. The neutral complex Cu(qbm)(DPEphos) exhibits a higher current
efficiency, up to 8.87 cd A-1, than that (5.58 cd A-1) of its counterpart [Cu(Hqbm)(DPEphos)](BF4).
injection.3 Similar phenomenon has also been found in the OLEDs
based on ionic CuI complexes.4 Therefore, the charge-neutral
Introduction
Photoluminescent CuI complexes based on a,a¢-diimine and
triphenylphosphine mixed ligand have attracted much attention
recently because of their highly phosphorescence from metal-
to-ligand charge-transfer (MLCT) excited state and possible
utilization in electroluminescent devices,1 for which they exhibit
remarkable performance with the external quantum efficiency
comparable to those third-row noble IrII, PtII, and OsII complexes.2
However, these cationic CuI emitters suffer two disadvantages
of the device fabrication. First, most of these mixed-ligand CuI
complexes exhibit poor volatility due to their ionic nature and
can’t fabricate the small-molecule organic light-emitting diodes
(OLEDs) by the conventional vacuum deposition method, which
always achieve more excellent device performance than those
polymer light-emitting diodes (PLEDs) fabricated by spin-coating.
Second, it is generally recognized that the small counterions of
ionic complex presented in the luminescent layer would decrease
the performance of the device by affecting the balance of charges
CuI phosphorescent complexes without counterions have been
expected to improve the performance of the OLEDs device.
Recently, several research groups have undertaken the photo-
physical study of neutral mononuclear, dinuclear and trinuclear
CuI complexes supported by individual bidentate, tridentate
and mixed ligands. For example, Omary et al. reported that
a neutral trinuclear CuI complex with fluorinated pyrazolate
ligands exhibits multicolor bright phosphorescent emissions that
are sensitive to temperature, solvent and concentration.5 More
recently, Peter et al. reported an example of neutral dinuclear CuI
complex, [Cu(PNP-tBu)]2, using tridentate PNP ligand ([PNP]- =
bis(2-(diisobutylphosphino)phenyl)amide), which exhibits excel-
lent quantum yield even in polar solvents due to a strongly rigid
structure resulting from steric hindrance of the ligand and the
absence of a counterion,6 and a kind of neutral mononuclear
CuI complex supported by amidophosphine ligands with quantum
efficiency in the range of 0.16–0.70 and lifetimes as long as 150 ms in
benzene solution.7 More exciting, the maximum external quantum
efficiencies (EQE) of 16.1% has been reported by using the above
complex, [Cu(PNP-tBu)]2, while submitting our manuscript.6b
However, neutral CuI complexes still remain rare compared to
a great deal of cationic CuI compounds.
aState Key Laboratory of Polymer Physics and Chemistry, Changchun
Institute of Applied Chemistry, Chinese Academy of Science, Changchun,
130022, P. R. China. E-mail: yanxiang@ciac.jl.cn, lixiang@ciac.jl.cn;
Fax: 86-431-85684937; Tel: +86-431-85262106
bState Key Laboratory of Structural Chemistry, Fujian Institute of Research
on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, 350002,
P. R. China
In this paper, we report two pairs of mononuclear di-
imine/diphosphine mixed-ligand CuI complexes with (ionic form)
and without counterions (neutral form), which possess completely
identical skeletons but quite different photophysical properties in
each pair. The first ligand used in this study, as a key factor, is
cThe Graduate School, Chinese Academy of Sciences, Beijing, 100039,
P. R. China
† CCDC reference numbers 724902–724905. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/c0dt01031f
686 | Dalton Trans., 2011, 40, 686–693
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Scheme 1 The synthesis route of mononuclear copper(I) complexes.
2-(2¢-quinolyl)benzimidazole (Hqbm), which can offer two kinds
of coordination features. One is from two sp2 N atoms located in
quinolyl and benzimidazole ring, respectively, to form the cationic
CuI complexes. Another is from a sp2 N atom in a quinolyl ring
and a sp3 N atom of NH group in a benzimidazole ring to prepare
neutral complexes by removing the active H atom in the presence
of a strong base.
Results and discussion
Synthesis and characterization
The synthesis of neutral qbm CuI complex with triph-
enylphosphine (PPh3) or bis[2-(diphenylphosphino)phenyl]ether
(DPEphos) as the second ligand via a two step reaction is outlined
in Scheme 1. Reaction of [Cu(NCCH3)4](BF4) with Hqbm and
phosphine ligands gives ionic complex [Cu(Hqbm)(PPh3)2](BF4)
(1a) and [Cu(Hqbm)(DPEphos)](BF4) (1b) in high yield first.
By treating 1a or 1b with NaOH mixture in methanol, the
neutral Cu(qbm)(PPh3)2 (2a) or Cu(qbm)(DPEphos) (2b) can be
subsequently isolated in 80 and 59% yield by recrystallization in
chloroform–methanol, respectively. All of the CuI complexes are
Fig.
1
ORTEP representation of the structure of the cation in
1
stable in air and characterized by H and 31P NMR spectra, in
[Cu(qbm)(PPh3)2](BF4)·2CH2Cl2 (1a). Thermal ellipsoids are drawn at
the 30% probability level. Solvent molecules, the anion and H atoms are
omitted for clarity.
which both 1a and 1b show the proton signals of imidazole rings
at about 12 ppm, and 31P signals (about -1.0 ppm for 1a and 2a,
about -12 ppm for 1b and 2b) shift down field compared with
those of free phosphine ligands (-4.60 for PPh3 and -16.40 for
DPEphos).
of neutral CuI complexes shorten due to the stronger bonding
ability of negatively charged benzimidazole. For example; the Cu–
˚
Neutral complexes can simply convert back into ionic complexes
while adding light excess 40% HBF4 solution, monitored by H
N distances of 2.064 and 2.118 A in neutral complex 2a are
1
˚
obviously shorter than those of 2.087 and 2.133 A in 1a, while
NMR. The pKa values measured by UV-visible spectrometry are
9.47 and 9.11 for complexes 1a and 1b, respectively, while the
free ligand, Hqbm, shows no shift in its spectrum under the same
conditions. However, the longer-wavelength peak red-shifts from
333 to 356 nm in the methanol of KOH instead of 6 : 4 methanol–
water solution, indicating the pKa value of the free ligand is bigger
than that of water. The smaller pKa of the complexes, relative to the
free ligand, indicates that the coordination decreases the acidity
of the proton on the imidazole ring.
the Cu–P distances are similar both in 1a and 2a. Second, the
negative charge on benzimidazole ring reshapes the five-membered
rings dramatically and the bond distances of C–N in imidazole
rings tend to average for neutral CuI complexes. Third, the torsion
angles N1–C7–C8–N2 of 1a and 1b (8.25 and 4.35◦) are obviously
reduced relative to 2a and 2b (9.12 and 11.34◦). The latter two facts
reveal that there is a stronger aromaticity on the NN chelating
ligand of neutral CuI complexes.
Another structural character is the existence of intramolecular
The molecular structures of four CuI complexes have been
determined by X-ray diffraction. Fig. 1–4 show the perspective
drawing of all complexes and Table 1 lists selected bond distances
and bond angles. Although all complexes exhibit similar distorted
tetrahedral coordination geometries, several differences stem from
the loss of the counterion and active H. First, the Cu–N distances
p–p interactions between benzimidazolyl and phenyl rings with
˚
˚
3.391 A of C41–N1 in 1b and 3.642 A of C42–N1 in 2b, respectively,
while DPEphos as the second ligand. The similar feature was
observed in the compound [Cu(phen)(dppf)](BF4), [where dppf =
1,1¢-bis(diphenylphosphino)ferrocene],8 due to the restricted bite
angle of the bidentate ligand, which always results in the more
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 686–693 | 687
©

View Article Online
◦
˚
Table 1 Selected bond distances (A) and bond angles ( ) for all complexes
Complexes
1a
1b
2a
2b
Cu–N1
Cu–N2
Cu–P1
Cu–P2
N1–C1
N1–C7
C7–C8
N3–C7
N3–C6
C6–C1
N2–C8
2.087 (3)
2.133 (3)
2.074 (3)
2.122 (3)
2.064 (2)
2.118 (2)
2.048 (5)
2.086 (5)
2.2223 (15)
2.2741 (15)
1.424 (8)
1.344 (8)
1.434 (9)
1.359 (7)
1.410 (9)
1.409 (9)
1.338 (8)
2.2633 (10) 2.2915 (11) 2.2630 (7)
2.2676 (10) 2.2227 (11) 2.2757 (8)
1.393 (4)
1.332 (4)
1.455 (5)
1.374 (4)
1.391 (4)
1.407 (4)
1.324 (4)
1.399 (5)
1.317 (5)
1.455 (6)
1.382 (5)
1.372 (5)
1.406 (6)
1.342 (5)
1.382 (4)
1.343 (4)
1.44 1(5)
1.356 (4)
1.372 (4)
1.416 (4)
1.345 (4)
N1–Cu–N2
P1–Cu–P2
N1–Cu–P1
N1–Cu–P2
N2–Cu–P1
N2–Cu–P2
N1–C7–C8–
N2
79.31 (11)
124.24 (3)
111.23 (8)
109.49 (8)
109.97 (8)
113.68 (8)
9.12
80.50 (12)
112.40 (4)
101.71 (9)
80.76 (10)
124.00 (3)
113.16 (7)
81.0 (2)
117.08 (6)
124.01 (14)
104.65 (13)
120.36 (14)
103.02 (13)
4.35
130.56 (10) 107.59 (7)
106.65 (9)
119.71 (9)
11.34
109.88 (7)
113.21 (7)
8.25
Fig. 3 ORTEP representation of the structure of Cu(qbm)(PPh3)2·
2CH3OH (2a). Thermal ellipsoids are drawn at the 30% probability level.
Solvent molecules and H atoms are omitted for clarity.
Fig.
2
ORTEP representation of the structure of the cation in
[Cu(qbm)(POP)](BF4)·1.5CH3OH·1.5CH2Cl2·1.5H2O (1b). Thermal ellip-
soids are drawn at the 30% probability level. Solvent molecules, the anion
and H atoms are omitted for clarity.
distorted tetrahedral geometries in CuI complexes. In contrast,
there is no intramolecular p–p interaction in 1a and 2a, and their
coordination geometries tend towards an ideal tetrahedron. In
addition, compounds 1a, 2a and 1b display intermolecular p–p
interactions between the quinolyl rings of two adjacent molecules
˚
with 3.417 (C6A–C6B), 3.574 (C11A–C11B) and 3.495 A (C11A–
C13B, Fig. 5), respectively.
Fig. 4 ORTEP representation of the structure of Cu(qbm)(POP)·CH3OH
(2b). Thermal ellipsoids are drawn at the 30% probability level. Solvent
molecule and H atoms are omitted for clarity.
Photophysical and electrochemical properties
The absorption and emission spectra for four CuI complexes in
CH2Cl2 are shown in Fig. 6. All complexes display an intense
p–p* transition centered on 2-(2¢-quinolyl)benzimidazolyl at 359
and 355 nm for 1a and 1b, 387 and 383 nm for 2a and 2b,
respectively. The red shift of the p–p* absorption of the neutral
complexes is not surprising in comparison to that of cationic
forms, which is consistent with the stronger conjugation ligands
in the neutral complexes after the removal of proton from 2-
(2¢-quinolyl)benzimidazole. The characteristic MLCT absorption
bands (e ª 3000 M-1 cm-1) can be observed clearly in cationic
688 | Dalton Trans., 2011, 40, 686–693
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Fig. 5 Crystal packing diagram between adjacent molecules of the cation in [Cu(qbm)(POP)](BF4)·1.5CH3OH·1.5CH2Cl2·1.5H2O (1b). Solvent
molecules, the anion and H atoms are omitted for clarity.
than DPEphos (1b), quite like the similar 2,2¢-biquinoline (bq)
complexes reported previously. What’s more, the radiative decay
rate (kr = f/t) of these two complexes is about 3.7–3.9 ¥ 103 s-1,
also very close to the value of bq complexes (3.6 ¥ 103 s-1 for
[Cu(bq)(PPh3)2]+ and 3.3 ¥ 103 s-1 for [Cu(bq)(DPEphos)]+) due to
the close NN ligand structures.2c Comparing the neutral complexes
to their parent cationic complexes, the quantum efficiency of 2b
(f = 0.015) is much higher than that of 1b (f = 0.0020), while that
of 2a (f = 0.0035) is slightly lower than that of 1a (f = 0.0042).
Ignoring the difference in emission efficiency, the lifetimes of the
neutral complexes are both several times longer than their parent
cationic ones.
The solid of four CuI complexes display bright yellow-orange
luminescence under a hand-held UV lamp. Their photophysical
data in poly(methyl methacrylate) (PMMA) film (20 wt% in
PMMA) were also studied (Table 2). The absorption spectra
of four CuI complexes in solid film are similar to those in
DCM, while their emission spectra blue shift 40–60 nm. In
addition, both quantum yield and lifetime significantly increased
in polymer films because the rigid medium can sterically pre-
Fig. 6 The absorption (left) and emission spectra (right) of four CuI
complexes in CH2Cl2 at room temperature.
vent the excited-state distortion of phosphorescent heavy metal
complexes and therefore decease the non-radiative decay, as
previously reported in RuII,9 IrIII,10 and CuI systems.2a,2c,11 As in
solution, the emission spectra blue-shift and the lifetimes have
almost doubled when comparing the neutral complexes with their
parent cationic complexes. However, the quantum efficiencies of
all complexes in solid film are very close within the range of
0.12–0.16.
With reference to previous work,12,13 it can be deduced that
the lowest excited state of the cationic complexes 1a and 1b
is principally attributed to the MLCT state. However, the case
in the neutral complexes 2a and 2b are more complicated.
First, the emissive excited state of the neutral complexes still
has MLCT character, because their photophysical characteristic
complexes 1a and 1b at 415 and 425 nm, respectively, when
these, even as the shoulder in neutral 2a and 2b disappear along
with red shift of p–p* bands. The emission spectra of the four
complexes show a broad orange-red emission band in DCM at
room temperature. The emission spectra of neutral complexes
show lmax at 596 nm for 2a and 612 nm for 2b, both of which
blue-shift 26 nm compared to their parent cationic complexes 1a
and 1b.
The quantum efficiency and lifetime of each complex in
degassed DCM varies widely depending on the auxiliary ligand
as well as the electronegativity of NN ligand, as listed in Table 2.
For the cationic complexes based on Hqbm, PPh3 complex
(1a) exhibits a 2-fold greater quantum efficiency and lifetime
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 686–693 | 689
©

View Article Online
Table 2 Photophysical and electrochemical data for CuI complexes at room temperature
Absorption in DCM
max/nm (e/M-1cm-1)
Emission in DCM
Emission in 20 wt% PMMA film
a
b
a
c
Compound
l
l
max/nm
f
t /ms
l
max/nm
f
t /ms
DEd/eV HOMOe/LUMOf/eV
1a
1b
2a
2b
359 (19 263), 415 (2522)
355 (18 933), 425 (3285)
387 (18 972)
622
638
596
612
0.0042
0.0020
0.0035
0.015
1.09
0.54
5.20
1.92
570
579
559
564
0.14
0.12
0.12
0.16
26.8
16.1
53.8
33.8
2.59
2.51
2.69
2.67
5.62/3.03
5.55/3.04
5.62/2.93
5.55/2.88
383 (18 671)
a Error 10%. b Fitted by single-exponential. c Since the decay is best-fit by double-exponentials, a weighted-average lifetime (tave ) was used and calculated
ꢀ
ꢀ
by the equation tave
=
(Aiti/ Ai), where Ai is the pre-exponential for the lifetime ti. d Calculated from the absorption edge. e Estimated by cyclic
voltammetry in dichloromethane with ferrocene (4.8 eV) as the reference.15 f Derived from the relation, DE = HOMO - LUMO.
strongly depend on the rigidity of the medium, which is often
observed in well defined MLCT CuI complexes.2a,2c,11 We speculate
that the MLCT bands disappear in the absorption spectra of
the neutral complexes can be explained in terms of that the
MLCT bands blue shift and superimpose on the red shifted
p–p* band. Therefore the lowest excited state of the neutral
complexes may be attributed to the MLCT plus LC (ligand-
centered p–p* transition) state. Generally, for those IrIII or OsII
phosphorescent complexes dominated by a mixed state involving
both MLCT and LC character, the larger extent of the LC
character will blue shift their emission spectra as well as prolong
their lifetimes (or decrease kr).14 The relatively longer lifetimes
and blue-shifted emissions observed in neutral complexes 2a
and 2b further confirm that the lowest energy excited state of
these neutral CuI complexes should contain a proportion of LC
character.
For comparison to the cationic CuI complexes, the corre-
sponding neutral one exhibits a larger HOMO–LUMO gap
(Table 2) and a higher energy emission. In order to find out
which orbital, HOMO or LUMO, increases the MLCT energy
gap, the electrochemical properties of the four complexes were
investigated by cyclic voltammetry at room temperature in dried
and argon purged CH2Cl2 solutions. As revealed by the data in
Table 2, each pair of complexes have the same HOMO energy level
calculated from the oxidation potential onset, -5.62 eV for 1a and
2a, -5.55 eV for 1b and 2b, respectively. These results suggest that
the negative charge in qbm ligand pushes up the energy level of
LUMO, leading to a larger MLCT (p*(qbm) ← d10(Cu)) energy
and a higher energy emission.
Electroluminescence
By spin-coating the emitting layer, multilayer OLEDs with the
configuration of ITO/PEDOT/emitting layer (~50 nm)/BCP
(20 nm)/Alq3 (40 nm)/LiF (1 nm)/Al (~100 nm) were fabricated
to evaluate the electrophosphorescent properties of the cuprous
complexes, as shown in Fig. 7. In this configuration, BCP
(2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline) and Alq3 [tris(8-
hydroxyquinolinato) aluminium] act as the exciton blocking and
electron transporting materials, respectively.
Similar to those [Cu(NN)(PP)]+ systems,2a,2c the electrolumines-
cent (EL) spectra of 1b or 2b are identical to photoluminescent
(PL) ones in PMMA film. Fig. 8 shows PL spectra and EL spectra
of 1b and 2b in TCCz host as examples. In contrast to the PL
spectra, no emission from the host was found in EL spectra,
indicating that charge trap occurred in the device operation beside
the energy-transfer process.2c
The data of OLEDs performance are listed in Table 3. Although
the PL quantum yield of 1a/1b or 2a/2b in PMMA is almost the
same, the devices based on 1a and 2a exhibit much poorer EL
performance. This may be associated with their poorer thermal
stability (as revealed by thermal gravimetric analysis) and the
ligand exchange reaction in solution during fabricating device
with mono-dentate ligand PPh3.1,2 With a doping concentration of
10 wt% in TCCz, device with a turn-on voltage of 7.5 V, maximum
current efficiency of 5.58 cd A-1, maximum brightness of 2820 cd
m-2 has been realized for 1b.
As for the neutral CuI complex 2b, devices with different doping
concentrations varying from 2, 5, 10 and 20 wt% were fabricated
Fig. 7 The device architecture (left) and the molecular structures of the relevant compounds (right).
690 | Dalton Trans., 2011, 40, 686–693 This journal is The Royal Society of Chemistry 2011
©

View Article Online
Table 3 Turn-on voltage (VT), current efficiency (hc), maximum brightness (Bmax), emission wavelength (lmax) of the OLEDs based on CuI complexes
c,max/cd A-1
hc /cd A-1
hc /cd A-1
B
max/cd m-2
lmax/nm
a
b
Emitting layer
D C (wt%)
VT/V
h
1a
2a
1b
2b
10
10
10
2
5
10
20
11.9
15.1
7.5
9.5
9.1
7.7
9.7
1.39
0.46
5.58
1.19
3.89
8.87
3.68
1.32
0.35
4.11
1.10
2.40
6.00
2.32
0.88
0.14
2.16
0.67
0.98
2.35
0.96
885
20
2820
292
210
465
260
571
547
562
555
564
555
555
a Measured at 1 mA cm-2. b Measured at 10 mA cm-2.
2b descends more sharply than 1b at high current density since it
exhibits longer decay lifetime thus tends to undergo triplet–triplet
annihilation.16
Unfortunately, an attempt to fabricate multilayer phosphores-
cent OLEDs using the vacuum deposition technique is unsuc-
cessful, though the thermolysis temperatures of neutral 1b and
2b are over 350 ◦C. Decomposition of the neutral complexes,
in company with the dissociation of the ligand proved by the
presence of the ligand emissive peak, was observed during the
alone subliming, indicating their poor volatility due to their high
formula weight. Therefore, synthesis of fluorinated5 or similar
neutral CuI complexes6 that exhibit increased volatility should
be desirable for deposited OLEDs.
Fig. 8 Electroluminescence (EL) and photoluminescence (PL) spectra of
1b and 2b in TCCz.
Conclusion
In summary, this paper describes the dramatic emission properties
of neutral mononuclear CuI complexes relative to the correspond-
ing cationic complexes. These compounds can act as models
for understanding the relationship between structural factors
and photophysical properties due to their completely identical
skeletons. The increased aromaticity and the decreased electron
accepting ability of the negatively charged NN ligand result in a
blue-shift emission from MLCT plus LC excited state in neutral
CuI complexes. The results obtained from the study of PL and EL
have demonstrated that the neutral CuI complexes possess similar
luminescent performance to ionic ones and are a kind of more
effective phosphorescent emitter for OLEDs.
to improve its electroluminescent property. As shown in Table 3,
the current efficiencies of the devices increase to the highest with
increasing the concentration to 10 wt%, and further enhancement
of the concentration results in reduction of the current efficiency.
Device with a turn-on voltage of 7.7 V, maximum current efficiency
of 8.87 cd A-1, maximum brightness of 465 cd m-2 has been realized
for 2b. Furthermore, a noticeable finding is that below the current
density of about 12 mA cm-2 (corresponding brightness of about
240 cd m-2), the current efficiency of device 2b is higher than that of
its counterpart 1b at the doping concentration of 10 wt% (shown in
Fig. 9). The lower efficiency of the devices of 1b may be attributed
to the increased current density, which was caused by the drifting
-
and accumulating of BF4 counterions towards the anode (ITO)
upon application of a bias.2a In addition, the current efficiency of
Experimental
General
All experiments were carried out without special treatments.
[Cu(CH3CN)4](BF4)17 and 2-(2-quinolyl)benzimidazole (Hqbm)18
were prepared by literature procedures. Triphenylphosphine
(PPh3) and bis(2-diphenylphosphinophenyl)ether (DPEphos)
were purchased from Aldrich Chemical Co.
Characterization. 1H NMR spectra with tetramethylsilane
(TMS) as internal reference and 31P NMR spectra with 85%
H3PO4 as external reference were recorded on a Bruker Avance
300 NMR instrument. The film samples for photophysical studies
were prepared by spin-coating a mixture of CuI complex (20 wt%)
and poly(methyl methacrylate) (80 wt%) in methylene chloride
onto a glass slide. UV-vis absorption and photoluminescence
spectra were recorded by a Perkin-Elmer Lambda 35 UV/VIS
Fig. 9 The current density–efficiency plot of the device for 1b and 2b.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 686–693 | 691
©

View Article Online
spectrophotometer and a HORIBA Jobin-Yvon FluoroMax-4
spectrometer, respectively. The solution PL quantum yield was
measured with [Ru(bpy)3]2+ as the reference in degassed water.19
The film PL quantum yield and the lifetime of both solution and
film were measured according to the method reported previously.11
Thermal gravimetric analysis (TGA) was carried out using the
Perkin-Elmer-DSC7 instrument under a nitrogen atmosphere
TGA 277.2 ◦C. Anal. Calcd. for C52H41BCuF4N3P2: C, 67.9; H,
4.5; N, 4.6%. Found: C, 67.8; H, 4.1; N, 4.5%. 1H NMR (CDCl3,
300 MHz): d 12.72 (s, 1H), 8.64 (d, 1H, J = 8.58 Hz), 8.52 (d, 1H,
J = 8.58 Hz), 8.00 (d, 1H, J = 8.19 Hz), 7.93 (t, 2H, J = 8.37 Hz),
7.56 (t, 1H, J = 7.17 Hz), 7.38–7.28 (m, 9H), 7.19–7.12 (m, 24H),
7.04 (d, 1H, J = 8.22 Hz). 31P NMR: -1.10 (s). pKa, 9.47 (25 ◦C,
in 6 : 4 methanol–water).
◦
at a heating rate of 20 C min-1. Elemental analyses (C, H
and N) were determined using a Bio-Rad elemental analysis
system. Cyclic voltammetry were performed on an EG&G 283
(Princeton Applied Research) potentiostat/galvanostat system at
room temperature with a conventional three-electrode system
consisting of a platinum working electrode, a platinum counter
electrode and an Ag/AgCl reference electrode. The measurements
were carried out in dried and argon purged CH2Cl2 solutions
with 1.0 mM of CuI complexes at a scan rate of 0.1 V s-1.
The supporting electrolyte was 0.1 M of n-tetrabutylammonium
hexafluorophosphate (TBAH). The potentials are quoted against
the ferrocene internal standard. The pKa values were measured
using spectra methods with KH2PO4/KCl as buffer solution in
6 : 4 methanol–water (v/v) at 25 ◦C.
[Cu(Hqbm)(DPEphos)](BF4) (1b). A dichloromethane solu-
tion (10 ml) of [Cu(CH3CN)4](BF4) (0.314 g, 1.0 mmol), DPEphos
(0.538 g, 1.0 mmol) and Hqbm (0.245 g, 1.0 mmol) was stirred
at room temperature for 5 h to give an orange solution. The
solution was then filtered and evaporated to dryness. The residue
was dissolved in dichloromethane and diffusion of diethyl ether
vapor into its concentrated solution gave orange crystals of 1b
(0.80 g, 86%).
◦
TGA 355.6 C. Anal. Calcd. for C52H39BCuF4N3OP2: C, 66.9;
H, 4.2; N, 4.5%. Found: C, 66.8; H, 4.0; N, 4.4%. 1H NMR (CDCl3,
300 MHz): d 12.19 (s, 1H), 8.43 (s, 2H), 7.99 (d, 1H, J = 8.58 Hz),
7.91 (d, 1H, J = 7.92 Hz), 7.80 (d, 1H, J = 8.13 Hz), 7.58 (d, 1H,
J = 8.16 Hz), 7.47 (m, 1H), 7.40–7.16 (m, 9H), 7.13 (t, 6H, J =
7.32 Hz), 7.05 (t, 4H), 6.89 (m, 4H), 6.75 (t, 4H), 6.69 (dd, 4H).
31P NMR: -11.32 (s). pKa, 9.11 (25 ◦C, in 6 : 4 methanol–water).
X-Ray crystallography†. The crystals of the complexes were
obtained via a solution-diffusion method. All diffraction inten-
sities were collected at 186 K on a Bruker Smart APEX CCD
˚
area-detector diffractometer (MoKa, l = 0.71073 A). The crystal
structure was solved using the SHELXTL program and refined
using full matrix least squares.20 The hydrogen atom positions
were calculated theoretically and included in the final cycles of
refinement in a riding model along with attached carbon atoms.
Cu(qbm)(PPh3)2 (2a). A mixture of 1a (0.230 g, 0.25 mmol)
and KOH (0.20 g, 5 mmol) in methanol (100 ml) was stirred for
10 h at room temperature. After evaporating to dryness, the solid
residue was extracted with dichloromethane. Subsequent diffusion
of methanol vapor into its concentrated solution gave orange
crystals of 2a (0.17 g, 80%).
Devices fabrication and characterization. To fabricate OLEDs,
a 50 nm thick poly(ethylenedioxythiophene)/poly (styrene sul-
fonic acid) (PEDOT : PSS, purchased from H.C. Starck) film
was first deposited on the pre-cleaned ITO-glass substrates with
a sheet resistance of 100 X/square and then cured at 120 ◦C in
air for 30 min. Then the emitting layer was prepared by spin-
coating a dichloromethane solution of a mixture of the host
material and the cuprous complexes [TCCz : CuI = 10 : 1 wt.] at a
concentration of 6.0 mg mL-1, respectively, forming a film around
50 nm. Successively, 20 nm of BCP, 40 nm of Alq3, 1 nm of LiF
and 100 nm of Al were evaporated at a base pressure less than 10-6
Torr (1 Torr = 133.32 Pa) through a shadow mask with an array
of 9 mm2 opening. The electroluminescence (EL) spectra were
recorded on a Horiba Jobin-Yvon FluoroMax-4 spectrometer. The
current–voltage and brightness–voltage curves of the devices were
recorded on a Keithley 2400/2000 source meter and a calibrated
silicon photodiode. All measurements on the devices were carried
out at room temperature under ambient conditions.
TGA 252.5 ◦C. Anal. Calcd. for C52H40CuN3P2: C, 75.0; H,
4.8; N, 5.1%. Found: C, 75.2; H, 4.6; N, 5.0%. 1H NMR (CDCl3,
300 MHz): d 8.71 (m, 1H), 8.21 (m, 1H), 7.86 (m, 1H), 7.77 (dd,
2H), 7.34–7.24 (m, 4H), 7.19–7.07 (m, 30H), 6.92 (br, 1H). 31P
NMR: -1.07 (s).
Cu(qbm)(DPEphos) (2b). A mixture of 1b (0.093 g, 0.10 mmol)
and KOH (0.06 g, 1.5 mmol) in methanol (40 ml) was stirred for
12 h at room temperature. After evaporating to dryness, the solid
residue was extracted with dichloromethane. Subsequent diffusion
of diethyl ether vapor into its concentrated solution gave orange
crystals of 2b (0◦ .05 g, 59%).
TGA 350.1 C. Anal. Calcd. for C52H38CuN3OP2: C, 73.8; H,
4.5; N, 5.0%. Found: C, 73.8; H, 4.3; N, 4.8%. 1H NMR (CDCl3,
300 MHz): d 8.59 (d, 1H, J = 8.55 Hz), 8.18 (d, 1H, J = 8.52 Hz),
7.87 (t, 2H, J = 8.49 Hz), 7.69 (t, 2H, J = 8.49 Hz), 7.49 (m, 4H),
7.26–6.74 (m, 30H). 31P NMR: -13.43 (s).
Synthesis
[Cu(Hqbm)(PPh3)2](BF4) (1a). A dichloromethane solution
(10 ml) of [Cu(CH3CN)4](BF4) (0.314 g, 1.0 mmol), PPh3 (0.525 g,
2.0 mmol) and Hqbm (0.245 g, 1.0 mmol) was stirred at room
temperature for 5 h to give an orange solution. The solution
was then filtered and evaporated to dryness. The residue was
dissolved in dichloromethane and diffusion of methanol vapor
into its concentrated solution gave orange crystals of 1a (0.77 g,
84%).
Acknowledgements
This work is supported by the National Natural Science Founda-
tion of China (No. 20874098 and 50673088), the Natural Science
Foundation of Fujian Province (No. 2007F3116), Science Fund
for Creative Research Groups (No. 20621401) and 973 Project
(2009CB623600).
692 | Dalton Trans., 2011, 40, 686–693
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
9 P. C. Alford, M. J. Cook, A. P. Lewis, G. S. G. McAuliffe, V. Skarda,
A. J. Thomson, J. L. Glasper and D. J. Robbins, J. Chem. Soc., Perkin
Trans. 2, 1985, 705.
10 R. D. Costa, E. Ort´ı, H. J. Bolink, S. Graber, S. Schaffner, M.
Neuburger, C. E. Housecroft and E. C. Constable, Adv. Funct. Mater.,
2009, 19, 3456.
References
1 (a) N. Armaroli, G. Accorsi, F. Cardinali and A. Listorti, Top. Curr.
Chem., 2007, 280, 69; (b) A. Barbieri, G. Accorsi and N. Armaroli,
Chem. Commun., 2008, 2185.
2 (a) Q. Zhang, Q. Zhou, Y. Cheng, L. Wang, D. Ma, X. Jing and F.
Wang, Adv. Mater., 2004, 16, 432; (b) Q. Zhang, Q. Zhou, Y. Cheng,
L. Wang, D. Ma, X. Jing and F. Wang, Adv. Funct. Mater., 2006, 16,
1203; (c) Q. Zhang, J. Ding, Y. Cheng, L. Wang, Z. Xie, X. Jing and F.
Wang, Adv. Funct. Mater., 2007, 17, 2983; (d) N. Armaroli, G. Accorsi,
M. Holler, O. Moudam, J.-F. Nierengarten, Z. Zhou, R. T. Wegh and
R. Welter, Adv. Mater., 2006, 18, 1313; (e) G. Che, Z. Su, W. Li, B. Chu
and M. Li, Appl. Phys. Lett., 2006, 89, 103511.
3 P.-T. Chou and Y. Chi, Eur. J. Inorg. Chem., 2006, 3319.
4 Q. Zhang, Ph.D. Dissertation, Changchun Institute of Applied Chem-
istry, Chinese Academy of Sciences, 2005.
5 H. V. Rasika Dias, H. V. K. Diyabalanage, M. A. Rawashdeh-Omary,
M. A. Franzman and M. A. Omary, J. Am. Chem. Soc., 2003, 125,
12072; H. V. Rasika Dias, H. V. K. Diyabalanage, M. G. Eldabaja, O.
Elbjeirami, M. A. Rawashdeh-Omary and M. A. Omary, J. Am. Chem.
Soc., 2005, 127, 7489.
11 L. Qin, Q. Zhang, W. Sun, J. Wang, C. Lu, Y. Cheng and L. Wang,
Dalton Trans., 2009, 9388.
12 S.-M. Kuang, D. G. Cuttell, D. R. McMillin, P. E. Fanwick and R. A.
Walton, Inorg. Chem., 2002, 41, 3313.
13 T. McCormick, W.-L. Jia and S. Wang, Inorg. Chem., 2006, 45,
147.
14 (a) L. He, J. Qiao, L. Duan, G. Dong, D. Zhang, L. Wang and Y. Qiu,
Adv. Funct. Mater., 2009, 19, 2950; (b) Y.-L. Chen, S.-W. Li, Y. Chi,
Y.-M. Cheng, S.-C. Pu, Y.-S. Yeh and P.-T. Chou, ChemPhysChem,
2005, 6, 2012; (c) Y.-M. Cheng, G.-H. Lee, P.-Tai Chou, L.-S. Chen,
Y. Chi, C.-H. Yang, Y.-H. Song, S.-Y. Chang, P.-I. Shih and C.-F. Shu,
Adv. Funct. Mater., 2008, 18, 183.
15 (a) J. Pommerehne, H. Vestweber, W. Guss, R. F. Mahrt, H. Bassler, M.
Porch and J. Daub, Adv. Mater., 1995, 7, 551; (b) M. Thelakkat and
H. W. Schmidt, Adv. Mater., 1998, 10, 219.
6 (a) S. B. Harkins and J. C. Peters, J. Am. Chem. Soc., 2005, 127, 2030;
(b) J. C. Deaton, S. C. Switalski, D. Y. Kondakov, R. H. Young, T. D.
Pawlik, D. J. Giesen, S. B. Harkins, A. J. M. Miller, S. F. Mickenberg
and J. C. Peters, J. Am. Chem. Soc., 2010, 132, 9499.
16 M. A. Baldo, C. Adachi and S. R. Forrest, Phys. Rev. B: Condens.
Matter Mater. Phys., 2000, 62, 10967.
17 G. J. Kubas, Inorg. Synth., 1979, 19, 90.
18 T. Hisano, M. Ichikawa and K. Tsumoto, Chem. Pharm. Bull., 1982,
30, 2996.
7 A. J. M. Miller, J. L. Dempsey and J. C. Peters, Inorg. Chem., 2007, 46,
7244.
19 J. Van Houten and R. J. Watts, J. Am. Chem. Soc., 1976, 98,
8 N. Armaroli, G. Accorsi, G. Bergamini, P. Ceroni, M. Holler, O.
Moudam, C. Duhayon, B. Delavaux-Nicot and J. F. Nierengarten,
Inorg. Chim. Acta, 2007, 360, 1032.
4853.
20 G. M. Sheldrick, SHELXTL, version 5.1, Bruker Analytical X-ray
Systems, Inc., Madision, WI 1997.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 686–693 | 693
©