Substituted azomethine-zinc complexes: Thermal stability, photophysical, electrochemical and electron transport properties

Article abstract of DOI:10.1016/j.ica.2008.10.016

A series of zinc complexes with salicylidene-aniline and its derivatives as ligands have been designed and synthesized for electron transport in organic light-emitting diodes (OLEDs). A systematic study on their thermal, photophysical, electrochemical and

Full text of DOI:10.1016/j.ica.2008.10.016

Inorganica Chimica Acta 362 (2009) 2327–2333
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Substituted azomethine–zinc complexes: Thermal stability, photophysical,
electrochemical and electron transport properties
*
Liang Chen, Juan Qiao , Junfeng Xie, Lian Duan, Deqiang Zhang,
Liduo Wang, Yong Qiu
Key Lab of Organic Optoelectronics & Molecular Engineering of Ministry of Education, Department of Chemistry,
Tsinghua University, Beijing 100084, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 August 2006
Received in revised form 13 October 2008
Accepted 15 October 2008
Available online 25 October 2008
A series of zinc complexes with salicylidene–aniline and its derivatives as ligands have been designed and
synthesized for electron transport in organic light-emitting diodes (OLEDs). A systematic study on their
thermal, photophysical, electrochemical and electron transport properties has been carried out and dem-
onstrated that the substitution of ꢀCH3, ꢀOCH3, ꢀCN and ꢀN(CH3)2 on aniline ring of ligands can finely
tune the properties of the corresponding zinc complexes. The density functional theory calculations of
location and distribution of the frontier molecular orbital states unveiled the relationships between
the substituents and the photophysical and electrochemical properties of these complexes. OLEDs with
bis(salicylidene-p-methylaniline)zinc(II) (Zn(sama)2) as the electron transport layer exhibited high cur-
rent efficiency, indicating its great potential as a useful electron transport material for OLEDs.
Ó 2008 Elsevier B.V. All rights reserved.
Keywords:
Salicylidene–aniline
Zinc complex
Electron transport
OLEDs
1. Introduction
(Znq2) [5,9] and bis(2-(1-hydroxyphenyl)benzothiazolate)zinc
(Zn(BTZ)2) [10,11] have proven to possess higher electron mobili-
Vacuum-deposited organic light-emitting diodes (OLEDs) have
made great progress since the first report by Tang and VanSlyke
of strong green electroluminescence from tris(8-hydroquino-
line)aluminum (Alq3) [1]. Alq3 is the most widely used emitting
material and electron transport host because of its environmental
stability with high glass transition temperature (Tg) and ease of
sublimation to form amorphous thin film [2,3]. Up to now, many
other organic metal chelates have been reported as emitting mate-
rials or electron transport hosts. By introducing other central metal
atoms (Ga, In, Sc [4], Be, Mg, Zn [5]. . .), a variety of metal com-
plexes with 8-hydroquinoline have been designed and synthesized
for use in OLEDs [6].
It is well known that the electron mobility of Alq3 is much lower
than the hole mobility of conventional hole transport materials
such as N,N0-diphenyl-N,N0-bis(3-methyl)-benzidine (TPD) and
N,N0-diphenyl-N,N0-bis(1-naphthyl)-benzidine (NPB) [7]. The
imbalance of two charge carriers reduces the quantum efficiency
for light emitting [8], then new electron transport materials with
higher electron mobility are highly desirable. Recently, zinc com-
plexes have been introduced to OLEDs and recognized as useful
electron transport materials, e.g. bis(8-hydroquinoline)zinc
ties than Alq3. Azomethine–zinc complexes with Schiff-base li-
gands have previously been studied about their structures and
reactions [12–16]. Though they were used as the emitting layer
in OLEDs research by Hamada and co-workers [17] in 1993, more
reports on material properties and corresponding device perfor-
mance about these zinc complex derivatives are needed. In this pa-
per, we synthesized five zinc complexes with different Schiff-base
ligands, bis(salicylidene–aniline)zinc(II) (Zn(saa)2), bis(salicylid-
ene-p-methylaniline)zinc(II) (Zn(sama)2), bis(salicylidene-p-met
hoxyaniline)zinc(II) (Zn(saoa)2), bis(salicylidene-p-cyanoaniline)
zinc(II) (Zn(saca)2), and bis[salicylidene(4-dimethylamino) ani-
line]zinc(II) (Zn(sada)2), as shown in Fig. 1a. Amongst these com-
pounds, Zn(saa)2, Zn(sama)2, Zn(saoa)2 and Zn(sada)2 have
previously been reported by different groups [13,15,18,19]. Here,
a detailed investigation of the molecular structure, thermal stabil-
ity as well as the photoluminescent, electrochemical and electron
transport properties was carried out. It was found that the steric
structure and electronegativity of substitution groups have a large
effect on the electrochemical and photophysical behaviors of the
corresponding complexes. With the assistance of quantum chemi-
cal calculation, we further studied the distribution of molecular
orbitals and the effect of substitution on the electronic structures
of these zinc compounds. Finally, the device performance demon-
strated that Zn(sama)2 might be a useful electron transport mate-
rial for OLEDs.
* Corresponding author. Tel.: +86 10 62773109; fax: +86 10 62795137.
E-mail addresses: qjuan@mail.tsinghua.edu.cn (J. Qiao), qiuy@mail.tsinghua.
edu.cn (Y. Qiu).
0020-1693/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2008.10.016

2328
L. Chen et al. / Inorganica Chimica Acta 362 (2009) 2327–2333
Zn(sama)2: MS (EI, 70 eV): [m/z] 484. Elemental Anal. Calc. for
a
C28H24N2O2Zn (485.9): C, 69.21; H, 4.98; N, 5.77. Found: C, 69.36;
H, 4.98; N, 5.99%. 1H NMR (400 MHz, [d6]DMSO, 25 °C, TMS),
d = 2.23 (s, 6H), 6.65 (m, 2H), 6.73 (d, 2H, J = 8.5 Hz), 7.14 (d, 4H,
J = 8.4 Hz), 7.19 (d, 4H, J = 8.4 Hz), 7.35 (ddd, 2H, J = 8.7, 7.0,
1.8 Hz), 7.50 (dd, 2H, J = 8.0, 1.6 Hz), 8.78 (s, 2H).
Zn(saoa)2: MS (EI, 70 eV): [m/z] 516. Elemental Anal. Calc. for
C28H24N2O4Zn (517.9): C, 64.94; H, 4.67; N, 5.41. Found: C, 64.72;
H, 4.64; N, 5.81%. 1H NMR (400 MHz, [d6]DMSO, 25 °C, TMS),
d = 3.70 (s, 6H), 6.65 (m, 2H), 6.91 (m, 4H), 7.27 (m, 4H), 7.34
(ddd, 2H, J = 8.5, 6.9, 1.6 Hz), 7.50 (dd, 2H, J = 7.9, 1.5 Hz), 8.77 (s,
2H).
R
N
O
Zn
O
N
R
Zn(saca)2: MS (EI, 70 eV): [m/z] 506. Elemental Anal. Calc. for
C28H18N4O2Zn (507.8): C, 66.22; H, 3.57; N, 11.03. Found: C,
66.02; H, 3.60; N, 11.23%. 1H NMR (400 MHz, [d6]DMSO, 25 °C,
TMS), d = 6.59 (t, 2H, J = 8.6 Hz), 7.33 (ddd, 2H, J = 8.6, 7.8,
1.7 Hz), 7.41 (dd, 2H, J = 7.8, 1.6 Hz), 7.54 (m, 4H), 7.84 (d, 4H,
J = 7.7 Hz), 8.57 (s, 2H).
b
The single crystal structure of Zn(sama)2 was analyzed by X-ray
diffraction and the result is similar with the previous report by
other group.[15] The ORTEP drawing of Zn(sama)2 with 35% prob-
ability ellipsoids is illustrated in Fig. 1b to show the stereochemis-
try of this complex.
2.2. Equipment
EI-MS spectra were obtained from a TRIO-2000 mass spectro-
graph. Element analysis data were obtained from a Elementar Var-
io EL CHN element analysis instrument. 1H NMR spectra were
recorded on a Burker ARX400 NMR spectrometer with tetrameth-
ylsilane as internal standard. Thermal analysis was determined
by differential scanning calorimetry (DSC) and thermal gravimetric
analysis (TGA) performed using TA DSC 2910 Modulated
instrument and TA TGA 2050 instrument, respectively. Absorption
spectra were recorded with an Agilent 8453 UV–Vis spectropho-
tometer, and PL spectra were obtained from a HORABA Fluoro-
Max-3 spectrofluorimeter, respectively. PL quantum efficiencies
(UPL) were determined from dimethyl formamide (DMF) solutions
by controlling the concentration of the sample so that the absor-
bance at 390 nm (excitation wavelength) was lower than 0.2
absorption unit. All the UPL of samples were calculated with the
known value for Alq3 in DMF (UPL = 0.116) [20] as reference. Cyclic
voltammetry measurements were conducted on a model CH 600
voltammetric analyzer with a platinum plate as the working elec-
trode, a silver wire as the pseudo-reference electrode, a polished
platinum wire as the counter electrode, and ferrocene as an inter-
nal reference [21], at a scan rate of 50 mV/s. The supporting elec-
trolyte was 0.1 mol/L tetrabutylammonium tetrafluoroborate in
DMF, the solution was deoxygenated with bubbling nitrogen for
15 min. The energy levels of highest occupied molecular orbital
(HOMO) and lowest unoccupied molecular orbital (LUMO) were
calculated from the oxidation and reduction potentials by adding
4.4, respectively [22].
Fig. 1. (a) Molecular formulas of the zinc complexes used in this study. R = H,
Zn(saa)2; R = CH3, Zn(sama)2; R = OCH3, Zn(saoa)2; R = CN, Zn(saca)2; R = N(CH3)2,
Zn(sada)2. (b) ORTEP drawing of Zn(sama)2 with 35% probability ellipsoids.
2. Experimental
2.1. Synthesis and structure characterization
The preparation and purification of Schiff-base ligands and cor-
responding zinc complexes were similar to synthesis of Zn(sada)2
described in our first paper [19] about Zn(sada)2.
The Schiff-base ligand was synthesized by refluxing a mixture
of salicylaldehyde and corresponding substituted aniline in a 1:1
molar ratio and the following recrystallization with ethanol gave
a precipitate with a yield of about 90%.
The zinc complex was prepared through a reaction between
ZnCl2 and corresponding Schiff-base ligand. A solution of ZnCl2
(0.6815 g, 5 mmol) in ethanol (30 ml) was gradually added to a
mixed solution of Schiff-base ligand (5 mmol) and piperidine
(1.0 ml, 10 mmol) in 120 ml ethanol. The mixture while being
heated at reflux for 0.5 h, and then stirred at room temperature
for 24 h. All the products are light yellow precipitates. The crude
products were collected by filtration and washed with ethanol,
and finally dried under an infrared lamp. The crude product yields
are around 70–87%. The materials were further purified by vacuum
train sublimation before analysis and the fabrications of devices.
Zn(saa)2: MS (EI, 70 eV): [m/z] 457. Elemental Anal. Calc. for
C26H20N2O2Zn (457.8): C, 68.21; H, 4.40; N, 6.12. Found: C, 68.01;
H, 4.32; N, 6.33%. 1H NMR (400 MHz, [d6]DMSO, 25 °C, TMS),
d = 6.52 (t, 2H, J = 7.3 Hz), 6.62 (d, 2H, J = 8.5 Hz), 7.28 (m, 4H),
7.43 (m, 6H), 7.62 (d, 4H, J = 7.9 Hz), 8.54 (s, 2H).
2.3. Density functional theory calculations
The density functional theory (DFT) was employed by using
GAUSSIAN 03 package [23] in this study. The restricted B3LYP (Becke
three-parameter Lee–Yang–Parr, for neutral molecules) and re-
stricted-open B3LYP (for ions) exchange-correlation functional,
which combines the Becke three-parameter exchange functional
[24] with the gradient-corrected correlation functional of Lee
et al. [25], was used for all calculations. We chose the basis 6-
31g(d) and performed the full optimizations for the geometrical
structures and electron distributions of HOMO and LUMO of neu-
tral Zn(saa)2, Zn(sama)2 and Zn(sada)2. In addition, cation and an-

L. Chen et al. / Inorganica Chimica Acta 362 (2009) 2327–2333
2329
ion of Zn(sama)2 (the neutral molecule in the presence of an extra
4
3
2
1
0
Zn(saa)2
hole and that of an extra electron, respectively) were also calcu-
lated. The initial geometric parameters employed in our calcula-
tions are from the crystal structure data [15,19].
Zn(sama)2
Zn(saoa)2
Zn(saca)2
Zn(sada)2
2.4. Fabrication of devices
Devices were fabricated on a glass slides coated with indium tin
oxide (ITO) with 30
X/square using a conventional vacuum vapor
deposition in a vacuum of around 10ꢀ3 Pa. ITO substrates were
cleaned by ultrasonication in ethanol–acetone (1:1) and de-ionized
water successively, and were then dried under infrared lamp for
2 h before use. After deposition of the organic layers, Mg and Ag
were coevaporated from separate evaporation boats in a 10:1
atomic ratio (Mg:Ag) to form a 150 nm alloy layer, then followed
by a 50 nm Ag cap without breaking the vacuum. The lumi-
300
350
400
450
500
Wavelength/nm
nance–current–voltage characteristics were measured with
a
Keithley 4200 semiconductor characterization system. All the mea-
surements were carried out in air without encapsulation of the
devices.
1.2
0.8
0.4
0.0
Zn(saa)2
Zn(sama)2
Zn(saoa)2
Zn(saca)2
Zn(sada)2
3. Results and discussion
3.1. Thermal stability
Thermal events observed in DSC and TGA curves can be attrib-
uted to several phase transitions of materials in heating process.
The weight loss 3% temperature can be obtained in TGA curve,
and the melting point (Tm), glass transition point (Tg) and recrystal-
lization point (Tc1) were measured from DSC cycle, respectively.
Results are presented in Table 1.
Despite of the similarity of ligands and coordination environ-
ments, the complexes with –CN, –OCH3, –CH3 as substitutions ex-
hibit relatively lower weight loss 3% temperatures. The trend in
thermal sublimation temperatures increases in the order of
Zn(saca)2<Zn(saoa)2<Zn(sama)2<Zn(sada)2ꢁZn(saa)2.
400
450
500
550
600
650
700
Wavelength/nm
The strong polar group ꢀCN or ꢀN(CH3)2 increases the molecu-
lar dipole and the Van der Waals force inner crystal and glass state,
as a result, Zn(saca)2 and Zn(sada)2 have much higher melting and
glass transition points. Among these compounds, Zn(saa)2 is the
only one without observable glass transition. It is suggested that
Zn(saa)2 can not form stable glass state. The surface morphology
study of vapor-deposited films on ITO substrates by atomic force
microscope (AFM) was explored subsequently and demonstrated
that the thin films of these zinc complexes except Zn(saa)2, exhibit
an entirely amorphous, uniform surface and a negligible change
after being exposed in air at room temperature for 5 days.
In contrast to Zn(sama)2 and Zn(saca)2, the second heating cy-
cles of Zn(saoa)2 and Zn(sada)2 do not reveal a recrystallization
exotherm (Tc1) following glass transition. Such recrystallization of
glass state is one of the most potential approaches to reduce the
stability of amorphous thin films. Non-planar substitution groups
(ꢀOCH3 and ꢀN(CH3)2) may increase the vibrational and rotational
degrees of freedom of the whole molecules, thereby preventing
easy packing of molecules and hence ready crystallization.
Fig. 2. Absorption and PL emission spectra of zinc complexes in DMF solutions.
Table 2
Photophysical data in DMF solution.
zinc
complexes
Abs. kmax
Abs. shift
(nm)b
Emission kmax
Emission shift
(nm)b
UPL
(nm)a
(nm)a
Zn(saa)2
400
401
400
409
412
ꢀ
487
480
480
520
553
ꢀ
0.057
0.031
0.011
0.012
0.007
Zn(sama)2
Zn(saoa)2
Zn(saca)2
Zn(sada)2
+1
0
ꢀ7
ꢀ7
+33
+66
+9
+12
a
All data were obtained in DMF solution (10ꢀ5 mol/L).
Shift from Zn(saa)2.
b
Table 3
Electrochemical data in DMF solution.
Zinc complexes
HOMO (eV)
LUMO (eV)
Eg (eV)
Zn(saa)2
ꢀ5.8
ꢀ5.8
ꢀ5.6
ꢀ5.8
ꢀ5.1
ꢀ3.1
ꢀ2.9
ꢀ2.7
ꢀ3.2
ꢀ2.5
2.7
2.9
2.9
2.6
2.6
Table 1
Thermal analysis data.
Zn(sama)2
Zn(saoa)2
Zn(saca)2
Zn(sada)2
Zinc complexes Weight loss 3% temperature (°C) Tm (°C) Tg (°C)a Tcl (°C)a
Zn(saa)2
317
301
282
279
317
186
208
191
250
274
n.
82
80
119
112
n.
144
n.
170
n.
Zn(sama)2
Zn(saoa)2
Zn(saca)2
Zn(sada)2
3.2. Photophysical properties
The electron transition responsible for photoluminescence in
these zinc complexes is originated from Schiff-base ligands. The
a
n. = not exist or not observable.

2330
L. Chen et al. / Inorganica Chimica Acta 362 (2009) 2327–2333
absorption and PL spectra of the zinc complexes in DMF solutions
are shown in Fig. 2, and the main photophysical data are list in
Table 2. The kmax of absorption and luminescence of the corre-
sponding complexes have the similar red-shift trend in the order
of Zn(saoa)2 = Zn(sama)2<Zn(saa)2<Zn(saca)2, in accordance with
the sequence of the electron donating ability of substitution groups
Fig. 3. Plots of the (a) HOMO of Zn(saa)2 molecule, (b) LUMO of Zn(saa)2 molecule, (c) HOMO of Zn(sama)2 molecule, (d) LUMO of Zn(sama)2 molecule, (e) HOMO of Zn(sada)2
molecule, and (f) LUMO of Zn(sada)2 molecule obtained at the RB3LYP/6-31g(d) level. All the MO surfaces correspond to an isocontour value of |
w| = 0.03 a.u. Here the light
grayish and darkish colors stand for the positive and negative phase of MO wavefunctions, respectively.

L. Chen et al. / Inorganica Chimica Acta 362 (2009) 2327–2333
2331
ꢀOCH3>ꢀCH3>ꢀH>ꢀCN. There is an exception that Zn(sada)2 with
the strongest electron donating group ꢀN(CH3)2 exhibits the larg-
est red-shift of PL spectrum in contrast to non-substituted
Zn(saa)2.
According to Table 2, like the reported Zn(sada)2 [19], the UPL of
these zinc complexes are very low, showing low potentials as the
emitters in OLEDs. However, it is worth noting that non-substi-
tuted Zn(saa)2 exhibits relative higher UPL. It seems that the substi-
tutions bring a negative effect on the UPL, because the vibrational
and rotational degrees of freedom due to the substituents increase
the probability of non-radiative relaxation process.
tuted aniline ring. In addition, dimethylamino substitution group
is relatively easy to be oxidized, thus resulting in an abnormal high
HOMO.
To reveal the effects of charge injection on the molecular con-
formational stability, we have also calculated the electronic prop-
erties of both anion and cation of Zn(sama)2. As can be seen in
Fig. 4, the distribution of the extra hole in the Zn(sama)2 cation cal-
culated as excess spin densities is completely localized on the sal-
icylaldehyde ring, which is similar to HOMO of neutral Zn(sama)2,
and the extra electron densities piled on the aniline ring is consid-
erable, which is also similar to LUMO of neutral Zn(sama)2. It is
suggested that the delocalization of extra negative charge in Zn(sa-
ma)2 is more effective, which may stabilize the anion and facilitate
electron injection and transport.
3.3. Electrochemical properties
Energy level is an important factor for emitting and charge
transport materials used in OLEDs. The injection and transport bar-
rier of charge carriers, which is one of the key conditions that influ-
ence the operating voltage and efficiency of OLEDs, is mainly due
to energetic differences between the work functions of the elec-
trodes and HOMO or LUMO energy levels of organic materials.
Experimental results are listed in Table 3.
The calculated adiabatic ionization potential (IP) and electron
affinity (EA) of Zn(sama)2 (IP: 6.50 eV; EA: 0.61 eV) are compared
with those of Alq3 (IP: 6.27 eV; EA: 0.52 eV). Notice that the IP
and EA data of Alq3 were reported in our previous work [26] using
the same calculation method. Since the adiabatic EA value of Zn(sa-
ma)2 is larger than that of Alq3, i.e., Zn(sama)2 releases more en-
ergy to create an anion than Alq3. We may expect that Zn(sama)2
is more likely to accept an electron and should be a more active
electron transporter.
The energy levels of HOMOs of these zinc complexes are all
about 5.7 eV except Zn(sada)2. The energy levels of LUMOs increase
in the order of Zn(saca)2<Zn(saa)2<Zn(sama)2<Zn(saoa)2<Zn(sa-
da)2, in accord with the electronegativities of substitution groups.
Among these substituent, ꢀCN is the strongest electron withdraw-
ing group and ꢀN(CH3)2 is the strongest electron donating group.
Hereby, Zn(sada)2 possesses the highest HOMO and LUMOs. With
the aid of quantum calculations, the electronic structures of the
representative Zn(saa)2, Zn(sama)2 and Zn(sada)2 were investi-
gated. As seen in Fig. 3, the frontier molecular orbitals of these zinc
complexes are mainly dominated by atomic orbitals originating
from Schiff-base ligands in all cases and the contribution of Zn2+
ions is vanishingly small. For the parent complex Zn(saa)2, both
HOMO and LUMO are mainly localized on the salicylaldehyde ring
of Schiff-base ligand, just a little bit of LUMO on the aniline ring. So,
the substitutions of ꢀCH3, ꢀOCH3 and ꢀCN on the aniline ring have
little effect on HOMOs, but a noticeable effect on LUMOs. However,
when the substitution is the strong electron donating group
ꢀN(CH3)2, the electron distribution of Zn(sada)2 has largely chan-
ged, especially HOMO is delocalized on the whole ligand, not only
on the salicylaldehyde ring, but also a little more on the substi-
3.4. Electron transport properties
Because of its good stability of vapor-deposited amorphous thin
film and potential electron injection and transport ability accord-
ing to quantum calculation, Zn(sama)2 has been used for device
investigation. The ‘‘electron only” devices are used to investigate
the electron transport ability of electron transport materials [27].
In our study, two ‘‘electron only” devices were fabricated
with the configurations of ITO/4,7-diphenyl-1,10-phenanthroline
(BPhen, 10 nm)/Zn(sama)2 (60 nm)/Mg:Ag (device A) and ITO/
BPhen (10 nm)/Alq3 (60 nm)/Mg:Ag (device B), where BPhen was
used as a hole blocking material to block hole injection from the
anode. Fig. 5 shows the dependence of the current density on volt-
age (I–V) for devices A and B. The threshold voltages of current
density (defined as the voltage required to give a current density
of 1 A/m2) for devices A and B are about 3.4 V and 14.1 V, respec-
tively, indicating that Zn(sama)2 is more effective in electron injec-
tion than Alq3 when the metal cathodes of OLEDs are conventional
Fig. 4. Spin density surfaces of the cation (a) and anion (b) of Zn(sama)2 obtained at the ROB3LYP/6-31g(d) level. All the MO surfaces correspond to an isocontour value of
| = 0.03 a.u. Here the light grayish and darkish colors stand for the positive and negative phase of MO wavefunctions, respectively.
|
w

2332
L. Chen et al. / Inorganica Chimica Acta 362 (2009) 2327–2333
104
103
102
101
100
10-1
10-2
10-3
10-4
a
3000
(30nm)/Zn(sama)
C: ITO/NPB/Alq
(30nm)/Mg:Ag
2
A:
/Mg:Ag
ITO/BPhen/Zn(sama)2
3
D: ITO/NPB/Alq 3 (60nm)/Mg:Ag
B: ITO/BPhen/Alq3/Mg:Ag
2500
2000
1500
1000
500
0
0
2
4
6
8
10
12
0
3
6
9
12
15
Voltage/V
Voltage/V
Fig. 5. Logarithmic plots of the current density–voltage characteristics of the
‘‘electron only” devices. Device A: ITO/BPhen (10 nm)/Zn(sama)2 (60 nm)/Mg:Ag;
device B: ITO/BPhen (10 nm)/Alq3 (60 nm)/Mg:Ag.
8000
6000
4000
2000
0
b
C: ITO/NPB/Alq3(30nm)/Zn(sama)2 (30nm)/Mg:Ag
D: ITO/NPB/Alq3 (60nm)/Mg:Ag
Mg:Ag. It is well known that the electron injection barrier is deter-
mined by the difference between the work function of the metal
cathode and the LUMO energy level of the used electron transport
material. In this case, the work function of Mg:Ag is 3.7 eV, the en-
ergy level of LUMOs of Zn(sama)2 and Alq3 are 2.9 eV and 2.8 eV
[28], respectively. It suggests that the OLEDs with Zn(sama)2 as
the ETL have a little lower electron injection barrier in ETL/cathode
interfaces. Furthermore, at a fixed voltage, the current density of
device A with Zn(sama)2 as the ETL is significantly larger than de-
vice B with Alq3 as the ETL. For example, at a constant voltage of
10.0 V, the current density of device A is 152 A/m2, whereas device
B shows only 5.55 ꢂ 10ꢀ2 A/m2. We considered that the enhance-
ment of current density of device A is due to not only the better
injection at the ETL/cathode contact, but also the intrinsic high
electron mobility within Zn(sama)2.
0
2
4
6
8
10
12
Voltage/V
Two additional light emitting devices were designed and fabri-
cated in order to further investigate the electron transport charac-
teristics of Zn(sama)2 in OLED: ITO / NPB (60 nm) / Alq3 (30 nm) /
Zn(sama)2 (30 nm) / Mg:Ag (device C), and ITO / NPB (60 nm) / Alq3
(60 nm) / Mg:Ag (device D), where NPB was used as a hole trans-
port material. Strong green light is emitted from both devices at
520 nm, which is the characteristic luminescence from Alq3. Fig.
6a and b contains the plots of the current density and brightness
characteristics of device C and D. Device C with Zn(sama)2 as the
electron transport layer gives both higher current density and
brightness at a fixed voltage, indicating that Zn(sama)2 with
matched energy level and high electron mobility enhances the
injection and transport of electrons and improves the electron-hole
balance. Although Zn(sama)2 shows better electron injection and
transport property than Alq3, the current luminescent efficiency
of device C (maximum: 2.60 cd/A) is lower than that of device D.
Actually, the Eg of Zn(sama)2 is 2.9 eV, which is comparable with
Alq3 with an Eg of 2.8 eV [28], therefore Zn(sama)2 could not block
the excitons generated in the emissive layer completely. The exci-
tons injected into Zn(sama)2 layer may quench in non-radiative
relaxation process, resulting in non-effective recombination and
low current efficiency.
4
3
2
1
0
c
E: ITO/NPB/Alq /B-Alq(5nm)/Zn(sama) (30nm)/Mg:Ag
3
2
C: ITO/NPB/Alq /Zn(sama) (30nm)/Mg:Ag
3
2
1800
600
900
1200
1500
300
2
Current Density/(A/m )
Fig. 6. (a) The current density versus voltage characteristics of devices C and D; (b)
the brightness versus voltage characteristics of devices C and D; (c) the current
efficiency versus current density characteristics of devices E and C. Device C: ITO/
NPB (60 nm)/Alq3 (30 nm)/Zn(sama)2 (30 nm)/Mg:Ag; device D: ITO/NPB (60 nm)/
Alq3 (60 nm)/Mg:Ag; device E: ITO/NPB (60 nm)/Alq3 (30 nm)/B-Alq (5 nm)/
Zn(sama)2 (25 nm)/Mg:Ag.
In order to prevent the injection of holes and excitons from EML
to ETL, a blocking layer aluminum (III) bis(2-methyl-8-quinolinato)
4-phenylphenolate (B-Alq) was introduced into our study. The de-
vice configuration is ITO/NPB (60 nm)/Alq3 (30 nm)/B-Alq (5 nm)/
Zn(sama)2 (25 nm)/Mg:Ag (device E). Current efficiency versus cur-
rent density comparison between devices E and C are plotted in
Fig. 6c. Contrast with device C, at the current density of 1340 A/
m2, the current efficiency of device E gives a maximum value of
3.19 cd/A. It is noted that there is no strong efficiency roll-off at
high current density for device E. That means the introduction of
blocking layer plays an important role in the device performance.

L. Chen et al. / Inorganica Chimica Acta 362 (2009) 2327–2333
[3] L.S. Hung, C.H. Chen, Mater. Sci. Eng. R 39 (2002) 143.
2333
4. Summary
[4] P.E. Burrows, L.S. Sapochak, D.M. McCarty, S.R. Forrest, M.E. Thompson, Appl.
Phys. Lett. 64 (1994) 2718.
A systematic study of (salicylidene–aniline)zinc(II) complex and
its derivatives on the thermal, photophysical and electrochemical
and electron transport properties has been performed. These com-
plexes are monomers of tetracoordinated zinc with distorted tetra-
hedron geometry. Our research demonstrated that the molar
weight, conformation and electronegativity of the different substi-
tution groups on the Schiff-base ligand can finely tune the thermal,
photophysical and electrochemical properties of the corresponding
Zn complexes. The sequence of the electron donating ability of sub-
stitution groups ꢀN(CH3)2>ꢀOCH3>ꢀCH3>ꢀH>ꢀCN is in accor-
dance with the height of the LUMO levels of corresponding
complexes. Analysis of the electronic structure of these complexes
calculated by B3LYP/6-31(d) method reveals the distribution of
frontier molecular orbitals in these zinc compounds and a localiza-
tion shift of HOMO in Zn(sada)2.
[5] Y. Hamada, T. Sano, M. Fujii, T. Fujii, Y. Nishio, K. Shibata, Jpn. J. Appl. Phys. 32
(1993) L514.
[6] C.H. Chen, J.M. Shi, Coord. Chem. Rev. 171 (1998) 161.
[7] W. Brütting, S. Berleb, A.G. Mückl, Org. Electron. 2 (2001) 1.
[8] Y. Qiu, Y. Gao, P. Wei, L.D. Wang, Appl. Phys. Lett. 80 (2002) 2628.
[9] L. Sapochak, F. Benincasa, R. Schofield, J. Baker, K. Riccio, D. Fogarty, H.
Kohlmann, K. Ferris, P. Burrows, J. Am. Chem. Soc. 124 (2002) 6119.
[10] Y. Hamada, T. Sano, M. Fujii, Y. Nishio, H. Takanashi, K. Shibata, Jpn. J. Appl.
Phys. Part 2 35 (1996) L1339.
[11] G. Yu, S.W. Yin, Y.Q. Liu, Z.G. Shuai, D.B. Zhu, J. Am. Chem. Soc. 125 (2003)
14816.
[12] M. Dreher, H. Elias, Z. Naturforsch. 42b (1987) 707.
[13] T. Sogo, J. Romero, A. Sousa, A. de Blas, M.L. Duran, Z. Naturforsch. 43b (1988)
611.
[14] F.A. Bottino, P. Finocchiaro, E. Libertini, G. Mattern, Z. Kristallogr. 187 (1989)
71.
[15] X.L. Liu, H.L. Chen, F.M. Miao, Youji Huaxue 12 (1992) 522.
[16] F.A. Bottino, P. Finocchiaro, Polyhedron 4 (1985) 1507.
[17] Y. Hamada, T. Sano, M. Fujii, Y. Nishio, K. Shibata, Jpn. J. Appl. Phys. 32 (1993)
L511.
[18] K.P. Dubey, C.N. Cachru, B.L. Wazir, J. Chin. Chem. Soc.-Taip. 26 (1979) 37.
[19] J.F. Xie, J. Qiao, L.D. Wang, J. Xie, Y. Qiu, Inorg. Chim. Acta 38 (2005) 4451.
[20] F.E. Lytle, D.R. Story, M.E. Juricich, Spectrochim. Acta Part A 29A (1973) 1357.
[21] R.R. Gange, C.A. Koval, G.C. Lisensky, Inorg. Chem. 19 (1980) 2854.
[22] S. Janietz, D.D.C. Bradley, M. Grell, C. Giebeler, M. Inbasekaran, E.P. Woo, Appl.
Phys. Lett. 73 (1998) 2453.
The studies of the ‘‘electron only” devices and OLEDs using
Zn(sama)2 as the electron transport layer have confirmed the elec-
tron transport ability of Zn(sama)2 which is predicted by quantum
calculation. Compared with typical material Alq3, the devices with
Zn(sama)2 as the ETL show a better performance.
[23] M.J. Frisch, L.W. Trucks, H.B. Schlegel, et al., GAUSSIAN 03, Gaussian Inc.,
Pittsburgh, PA, 2003.
[24] A.D. Becke, Phys. Rev. A 38 (1988) 3098.
Acknowledgements
[25] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[26] J. Qiao, H. Tan, Y. Qiu, K. Balasubramanian, J. Chem. Phys. 124 (2006) 024719.
[27] N. Donze, P. Pechy, M. Grätzal, M. Schaer, L. Zuppirioli, Chem. Phys. Lett. 305
(1999) 405.
[28] M.T. Lee, H.H. Chen, C.H. Liao, C.H. Tsai, C.H. Chen, Appl. Phys. Lett. 85 (2004)
3301.
This work is supported by the National Nature Science Founda-
tion of China (Nos. 50403001 and 50325310). We thank Center for
Advanced Study, Tsinghua University for their help.
References
[1] C.W. Tang, S.A. Van Slyke, Appl. Phys. Lett. 51 (1987) 913.
[2] M. Brinkmann, G. Gadret, M. Muccini, C. Taliani, N. Masciocchi, A. Sironi, J. Am.
Chem. Soc. 122 (2000) 5147.