Improved performance of organic light-emitting diodes using a metal-phthalocyanine hole-injection layer

Article abstract of DOI:10.1149/1.2189267

In this paper, we systematically investigated the physical characteristics of the various metal phthalocyanines (MPcs) and the influence of the MPcs hole-injection layer on the electroluminescence performance of indium tin oxide/MPc/naphthylphenylbiphenyl diamine (NPB)/ Al q3 LiFAl devices. The characteristics were measured at room temperature with a thickness variation of the MPc layer. The individual highest occupied and lowest unoccupied molecular orbital (HOMO and LUMO) energies of MPcs were derived from the photoelectron emission and the optical absorption measurements. The results showed that the HOMO and LUMO level energies of MPcs are dependent on their central metal atoms. The turn-on voltage for the devices is lowered by inserting MPc layers and remains virtually the same as the MPc layer thickness is adjusted in the range of 5-15 nm. In addition, the turn-on voltage decreases significantly with the increase of the HOMO levels of the MPc films, demonstrating that the MPc/NPB interface instead of the ITO/MPc interface plays an important role in the hole injection.

Full text of DOI:10.1149/1.2189267

H122
Journal of The Electrochemical Society, 153 ͑6͒ H122-H126 ͑2006͒
0
013-4651/2006/153͑6͒/H122/5/$20.00 © The Electrochemical Society
Improved Performance of Organic Light-Emitting Diodes
Using a Metal-Phthalocyanine Hole-Injection Layer
a
a,z
b
a
Po-Ching Kao, Sheng-Yuan Chu, Shyh-Jiun Liu, Zong-Xian You, and
b
Chan-An Chuang
a
Department of Electrical Engineering, National Cheng Kung University, Tainan, 701-01, Taiwan
b
Department of Natural Science Education, National University of Tainan, Tainan, 700, Taiwan
In this paper, we systematically investigated the physical characteristics of the various metal phthalocyanines ͑MPcs͒ and the
influence of the MPcs hole-injection layer on the electroluminescence performance of indium tin oxide/MPc/
naphthylphenylbiphenyl diamine ͑NPB͒/Alq /LiF/Al devices. The characteristics were measured at room temperature with a
3
thickness variation of the MPc layer. The individual highest occupied and lowest unoccupied molecular orbital ͑HOMO and
LUMO͒ energies of MPcs were derived from the photoelectron emission and the optical absorption measurements. The results
showed that the HOMO and LUMO level energies of MPcs are dependent on their central metal atoms. The turn-on voltage for
the devices is lowered by inserting MPc layers and remains virtually the same as the MPc layer thickness is adjusted in the range
of 5–15 nm. In addition, the turn-on voltage decreases significantly with the increase of the HOMO levels of the MPc films,
demonstrating that the MPc/NPB interface instead of the ITO/MPc interface plays an important role in the hole injection.
©
2006 The Electrochemical Society. ͓DOI: 10.1149/1.2189267͔ All rights reserved.
Manuscript submitted October 31, 2005; revised manuscript received February 1, 2006. Available electronically April 13, 2006.
Organic light-emitting diodes ͑OLEDs͒ are currently considered
CuPc, SnPc, NiPc, MnPc, FePc, and SnPc͒. The OLEDs were fab-
ricated using an insertion of MPc thin film as a hole-injection layer
͑HIL͒ and the device properties were investigated.
as promising candidates for large-area, full-color, flat-panel displays
due to their prominent advantages such as ease in fabrication and
1
convenience of use in applications. A lot of effort has therefore
been made to improve device structures and to understand their
operating mechanisms. In OLEDs, the injection efficiency is a criti-
cal parameter and depends, to a great extent, on the work function of
the electrode. The potential barrier between indium tin oxide ͑ITO͒
and the organic layer severely limits the efficiency of hole injection.
The treatment of the ITO surface usually has a strong influence on
Experimental
The ITO-coated glass substrates ͑Merck Display Technologies͒
used have a film thickness of 0.2 ␮m and sheet resistance of ap-
proximately 10 ⍀/ᮀ. The MPc powders used in this work are pre-
pared directly from phthalonitrile using the metal ion as a template
for the cyclotetramerization, followed by purification through wash-
ing and sublimation. Prior to the organic films being deposited, the
substrates were initially scrubbed in a detergent solution ͑Merck
Extran͒. They were then immersed sequentially in a heated ultra-
sonic bath of deionized ͑DI͒ water, isopropyl alcohol, and ethanol
for 15 min each followed by being rinsed in DI water. Finally, the
substrates were blown dry with nitrogen gas and then treated by O2
plasma for 2 min prior to use. Electroluminescent devices fabricated
in this work have the following configurations: ITO/MPc/NPB
͑60 nm͒/tris͑8-hydroxyquinoline͒ aluminum ͑Alq3͒ ͑75 nm͒/
lithium fluoride ͑LiF͒ ͑1 nm͒/Al ͑150 nm͒. In the devices, MPc was
2-6
the performance of the OLED device. For example, the O plasma
2
treatment increases the work function of ITO and removes its sur-
1
6
7
face contaminants, thus enhancing the hole injection. However, the
work function of the treated ITO is not high enough, and the energy
barrier exists for the hole injection from ITO to the organic layer.
Another problem with the ITO anode is the diffusion of indium into
the organic layer during device operations, which is correlated with
8
the decay of device performance.
An important improvement of device stability was achieved by
depositing a copper phthalocyanine ͑CuPc͒ hole-injection layer
9,10
͑
HIL͒ on the ITO anode.
The improved stability could be con-
tributed to the good match of its highest-occupied molecular orbital
1
1
used as the HIL, NPB as the HTL, Alq as the electron-transport
͑
HOMO͒ level to the work function of ITO and the improved
3
1
2
layer and emission layer, and LiF/Al layers served as the composite
cathode to enhance the electron injection. All organic layers and the
cathode were successively deposited using vacuum vapor evapora-
tion at room temperature. The deposition rate was controlled by a
calibrated quartz crystal oscillator and was maintained at
.5 ± 0.2 Å/s for the organic materials, 0.1 Å/s for LiF, and
± 2 Å/s for Al. The base pressure in the chamber was about
ϫ 10− Torr. A shadow mask was used for the deposition of the
wetting property of organic materials on ITO. Moreover, CuPc has
very weak absorption of light with wavelengths from
4
00 to 500 nm, making it suitable for use in blue and green
1
3
14
OLEDs. Aziz et al. also reported that inserting CuPc between the
ITO anode and hole transport layer ͑HTL͒ could reduce the ITO-
induced degradation on the HTL materials.
0
5
5
Compared to other HIL materials, most metal phthalocyanines
6
͑
MPcs͒ are water and air stable, thermally stable, and nontoxic.
cathode. The active area of the devices is 2 ϫ 2 mm.
The structural and crystalline characteristics of MPc complexes
were examined using X-ray diffraction ͑XRD͒ and scanning electron
microscopy ͑SEM͒. The optical bandgaps and energy levels of the
MPc thin films were determined by measuring optical absorption
and photoelectron emission. The former was measured with a Hita-
chi U-4100 and the latter was measured with a Riken Keiki AC-2
system.
The current density-voltage ͑J-V͒ and luminance-voltage ͑L-V͒
charact eristics of the devices were both measured with a source
meter ͑Keithley-2400͒ and a luminance meter ͑Minolta CS-100A͒.
Electroluminescence ͑EL͒ was measured using a luminescence spec-
trometer ͑SPEX 1681͒ and a source meter. All measurements were
carried out at room temperature and in ambient lighting without any
protective coating.
They can be sublimed or sputtered with highly uniform, thin films
15
on a variety of substrates. The literature on their synthesis is vast,
1
5
and they are also relatively inexpensive and easy to prepare. The
chemical structure of MPc allows tuning its ionization potential ͑or
HOMO levels͒ by altering the central atom in Pc macrocycles.
Therefore, a systematic change in the energies of the HOMO levels
of MPcs enables us to study the influence of the ITO/MPc/
naphthylphenylbiphenyl diamine ͑NPB͒ interface barrier on the
characteristics of OLEDs.
In the current study, we systematically investigated the physico-
chemical characteristics of the various MPc complexes ͑CoPc,
z
E-mail: chusy@mail.ncku.edu.tw

Journal of The Electrochemical Society, 153 ͑6͒ H122-H126 ͑2006͒
H123
*
were assigned as the first and second ␲-␲ transitions, respectively.
Another band within the 200–300 nm range which resulted from
2
0
b ͑d ͒ to b ͑␲͒ was called the N band.
2
g
xy
1n
2
A plot of the absorption coefficient ␣ as a function of the photon
energy h␯ for MPc films is shown in Fig. 1b. The extrapolation of
the straight line ␣ = 0 gives the value of the optical bandgap. The
HOMO level energy can be deduced by using the photoelectron
emission technique. The lowest unoccupied molecular orbital
2
21
͑
LUMO͒ level energy is then estimated on the basis of the HOMO
level energy and the optical bandgap of the Q band of the MPc
films. The optical bandgap E and energy levels obtained for the
g
MPc films are shown in Table I. The results show that the HOMO
and LUMO level energies of MPc films are dependent on their cen-
tral metal atoms. When these results are compared to those obtained
2
2-25
from other researchers,
some small differences in some MPc
complexes were found. Such differences may be caused by the mea-
2
2-24
25
suring methods used,
or the deposition rate, heat-treatment, or
time of exposure to the air.
Luminance-current
density-voltage
characteristics
of
OLEDs.— Layer optimization.— The J–V of the devices with re-
spect to the NiPc thickness under a forward bias are shown in Fig.
2
a. The forward J–V curves have two regimes, a low-current regime
with weak voltage dependence and a high-current regime with a
steeper current rise as a function of the voltage. As can be clearly
seen, with the insertion of the NiPc film, the threshold voltage de-
creases markedly. Figure 2b shows L–V of the NiPc devices. Simi-
2
larly, the turn-on voltage ͑at 1 cd/m ͒ of the devices decreases when
the NiPc layer is inserted. However, the turn-on voltage remains
virtually the same when we alter the NiPc layer thickness in the
range of 5–15 nm ͑see the insert of the Fig. 2b͒. In these cases, the
values are around 4.3 V, while the device without a NiPc layer has
the value of 5.3 V. Figure 3c shows the luminance–current density
characteristics ͑L–J͒ of the devices mentioned above. Because the
slope of each line in Fig. 2c is proportional to its external quantum
efficiency, the efficiency is sensitive to the thickness of the NiPc
layer. The results show that the insertion of a 10 nm NiPc layer
gives a higher efficiency with maximum value of 4.9 cd/A, com-
pared with 3.3 cd/A for OLEDs without a NiPc layer.
Figure 1. ͑a͒ Transmission spectra of MPc thin films on the quartz substrate.
2
͑
b͒ Variation of the absorption coefficient ␣ as a function of photon energy
h␯.
Results and Discussion
2
6
Structural, optical properties and energy levels of the MPc com-
The above results are consistent with the work of Hill et al. and
9
14
plexes.— The XRD patterns for MPc thin films showed no obvious
differences. Only one major peak ͓200͔ was found and these results
showed that all samples belonged to the ␣-form structure. No
Vestweber and Rieß, but contrary to that of Aziz et al. for OLEDs
with a CuPc layer. These discrepancies are mainly due to the process
conditions as ITO and CuPc are very subtle materials and their
2
7
␤
-form structure was found because the substrate temperature was
properties may change a lot. Thus, depending on different surface
17
2
8
lower than the phase transition temperature. From SEM observa-
tions, all vacuum-deposited MPc films showed rod-like crystallites
with a width of approximately 30 nm, which were similar to those
treatments, ITO has different work functions. Nuesch et al. also
reported that the hole injection into diamine compounds could be
either lowered or increased by using a CuPc layer in different sur-
face treatments of the ITO anode. These observations elucidate the
apparent contradiction in the literature concerning hole injection into
HTLs.
18
shown in the previous study.
Figure 1a shows the transmission spectra curve of MPc films
deposited on the quartz substrate. The direct electronic transition
*
from a ͑␲͒ to e ͑␲ ͒ within the 300–450 nm range resulted in an
The reduction in turn-on voltage can be attributed to a decrease
in the hole-injection barrier. This is because the HOMO level energy
of the NiPc is between that of the NPB and the Fermi level of the
2
u
g
1
9
intense B band. The Q band within the 600–750 nm range which
resulted from a ͑␲͒ to e ͑␲ ͒ yielded two levels, Q and Q , that
*
1
u
g
x
y
Table I. The optical bandgap, HOMO, LUMO, and MPc/NPB barrier of MPc complexes. Data were collected in thin films by aUV-visible
spectrophotometer and bAC-2 photoelectron spectrometer.
B Bandgap
N bandgap
HOMO
͑eV͒
LUMO
͑eV͒
MPc/NPB
barrier
Q Bandgap ͑eV͒a
͑eV͒a
͑eV͒a
b
Complex
CoPc
CuPc
ZnPc
NiPc
MnPc
FePc
SnPc
NPB
1.61͑Qx͒
1.77͑Qy͒
1.70͑Qy͒
1.77͑Qy͒
1.80͑Qy͒
1.66͑Qy͒
1.73͑Qy͒
1.64͑Qy͒
3.17
3.08
3.20
3.14
3.14
3.11
3.09
5.10
4.96
4.96
5.06
5.00
5.02
5.00
−5.36
−5.28
−5.18
−5.16
−5.07
−5.00
−4.88
−5.41
−3.75
−3.64
−3.57
−3.53
−3.44
−3.34
−3.26
−2.51
0.05
0.13
0.23
0.25
0.34
0.41
0.53
1.64͑Qx͒
1.61͑Qx͒
1.63͑Qx͒
1.63͑Qx͒
1.66͑Qx͒
1.62͑Qx͒

H124
Journal of The Electrochemical Society, 153 ͑6͒ H122-H126 ͑2006͒
Figure 2. ͑a͒ J-V, ͑b͒ L-V, and ͑c͒ L-J characteristics for
ITO/NiPc͑x nm͒/NPB͑60 nm͒/Alq ͑75 nm͒/LiF͑1 nm͒/Al͑200 nm͒
de-
3
vices. ͑b Inset͒ Turn-on voltage ͑V ͒ vs thickness ͑d͒ of the NiPc layer.
t
Figure 3. ͑a͒ J-V, ͑b͒ L-V, and ͑c͒ L-J characteristics for
ITO/MPc͑10 nm͒/NPB͑60 nm͒/Alq ͑75 nm͒/LiF͑1 nm͒/Al͑200 nm͒ de-
3
ITO anode. When the NiPc layer is inserted between the ITO and
NPB layer, a ladder effect makes it easier for holes to inject into the
organic layer, which in turn leads to increased current density. Fur-
vices.
28
thermore, Nuesch et al. used a Kelvin probe technique to investi-
gate the interface formation between ITO and CuPc layer by layer.
They demonstrated that a decrease in the overall barrier height to
hole injection depends weakly on the CuPc thickness. Similar results
2
9
materials. This effect might occur in the CuPc layer and result in
lower mobility of holes, which could bring the opposite carrier
1
2
12
were also obtained using UV photoemission by Tadayyon et al.
populations into balance. The same mechanism is also responsible
Thus, the turn-on voltage of the devices may remain unchanged with
alteration of the MPc layer thickness in the range of 5–15 nm. The
possible oxygen diffusion from ITO into the CuPc layer may have a
countereffect on the charge mobility. The amount of oxygen may be
large enough to increase the trap level. In addition, it has also been
shown that the negatively charged oxygen ions diffused from ITO
behave as shallow traps for positive charge carriers in organic
for the lower mobility of holes and shallow traps formed in the NiPc
layer. Therefore, according to these two facts, we speculated that a
thickness-dependent property may be responsible for the changes in
device efficiency. Based on our experimental results, the optimum
thickness of the NiPc buffer layer is around 10 nm. Further investi-
gations show that 10 nm is also the optimum thickness for other
3
0
MPc buffer layers.

Journal of The Electrochemical Society, 153 ͑6͒ H122-H126 ͑2006͒
H125
Table II. The turn-on voltage, maximum efficiency, and lumi-
nance characteristics for the devices with non-MPc or different
MPc layers.
Turn-on voltage at Maximum efficiency Maximum
2
1
cd/m
͑V͒
␩c ͑␩p͒
͑cd/A͒ ͑lm/w͒
luminance
͓cd/m ͑V͔͒
2
MPc
CoPc
CuPc
ZnPc
NiPc
MnPc
4.3
4.3
4.5
4.6
4.8
5.0
5.3
5.3
3.9͑1.1͒
5.1͑1.9͒
4.4͑1.7͒
4.9͑1.8͒
3.8͑1.2͒
3.9͑1.4͒
3.6͑1.1͒
3.3͑1.0͒
53,700͑14.8͒
36,830͑14.8͒
38,560͑15.0͒
41,690͑15.5͒
30,660͑15.5͒
27,110͑15.5͒
21,130͑15.5͒
21,030͑14.5͒
FePc
SnPc
Without MPc
Influence of the MPc layers on OLEDs.— The bright green emis-
sion ͑␭ = 525 nm͒ was observed for all the devices with different
MPc layers under forward bias. The J-V and L-V characteristics of
the devices with the inserted MPc layers are shown in Fig. 3a and b,
respectively. The turn-on voltage of the devices is lowered signifi-
cantly when CoPc or CuPc layers are inserted. In contrast, for the
devices with SnPc or FePc layers, the values are 5.0 and 5.3 V,
respectively. Figure 3c shows the L-J characteristics of the devices
mentioned above. Compared to the non-MPc device, higher emis-
sion efficiency was observed in all MPc devices and the CuPc de-
vice has the highest. In other words, for the same emission intensity,
the much higher emission efficiency suggests a much lower current
density is required. This may decrease the joule heating during de-
vice operation and then increase the operating durability of the de-
vice. Durability is one of the most critical issues in all issues asso-
ciated with industrial applications of OLEDs. Table II shows the
turn-on voltage, maximum efficiency, and luminance characteristics
for the devices whether or not the MPc buffer layer is applied.
The energy-level diagram of our devices, as shown in Fig. 4, was
constructed by measuring the optical absorption and photoelectron
emission. It can be seen that all MPcs possess low ionization poten-
tial and thus lower the potential barrier for hole injection. The small
barrier for hole injection may increase the stability of barriers at the
interface, consequently improving the durability of the devices. In
addition, the insertion of the MPc layer produces two new inter-
Figure 5. Turn-on voltage vs HOMO levels of the MPcs for
ITO/MPc/NPB/Alq /LiF/Al devices.
3
faces, the ITO/MPc and MPc/NPB interfaces. In principle, the bar-
riers at both of these interfaces will affect the turn-on voltage. How-
ever, the front interface ͑inorganic/organic͒ was different from the
back interface ͑organic/organic͒. Figure 5 shows the relation of the
turn-on voltages vs the HOMO levels of the MPcs for ITO/MPc
͑
10 nm͒/NPB ͑60 nm͒/Alq ͑75 nm͒/LiF ͑1 nm͒/Al ͑200 nm͒ de-
3
vices. The results show that the turn-on voltage of the devices de-
creases significantly with the increase of the HOMO levels against a
vacuum of the MPc films. Therefore, this finding gives a clear indi-
cation that the energy barrier for hole injection at the MPc/NPB
interface instead of the ITO/MPc interface predominately deter-
mines the turn-on voltage.
The existence of the internal energy barriers can cause the accu-
mulation of space charges at the interface in the multilayer OLEDs.
31
In order to find the limiting barrier, Riel et al. fabricated OLEDs
with graded interfaces and showed that the abrupt CuPc/NPB inter-
face is the dominant barrier where most of the holes accumulate.
32
Masenelli et al. reported that the limiting factor for hole injection
from ITO to HTL is the CuPc/HTL interface rather than the ITO/
CuPc interface or CuPc layer. They found that a large number of
holes are accumulated at CuPc/HTL and these holes cause the ener-
getic barrier to decrease, while those holes increase with the external
2
8
field. Nuesch et al. also showed that effective ITO work function is
pinned at the energy level of the HOMO level of CuPc, despite
differently treated ITO substrates with different initial work func-
tions. They ascribed this effect to oxygen doping from the ITO
electrode. In other words, MPc can help to either increase or lower
the effective ITO work function with different surface treatment of
the ITO anode. Therefore, we infer that the MPc/NPB interface
͑
rather than the ITO/MPc interface͒ plays an important role in hole
injection, and the optimum energy level between ITO and NPB
should be a HOMO approaching that of NPB.
Among the hole-injection materials utilized, CuPc, NiPc, and
ZnPc complexes exhibit greater enhancement in improving OLED
performance, including the turn-on voltage and efficiency. The
smaller threshold against hole injection should increase the imbal-
ance between electron and hole carriers, and hence decrease the
efficiency. Thus, it is rather unexpected that the devices with the
insertion of MPc layers show better efficiency than those devices
3
3
without MPc layers. However, values reported in the literature
indicate that hole mobility and trapping concentration in CuPc can
change by two orders of magnitude depending on the deposition
conditions. Thus, at lower substrate temperature, higher background
pressure for deposition, or even increased rate of evaporation, hole
3
3
mobility can be lower and trapping concentration can be higher in
3
4
Figure 4. Schematic energy-level diagram of ITO/MPc/NPB/Alq /LiF/Al
device.
the CuPc layer than in the NPB. If trapping and lower mobility of
the holes occurs in MPc layers, such an effect can prevent holes
3

H126
Journal of The Electrochemical Society, 153 ͑6͒ H122-H126 ͑2006͒
5
6
7
. S. F. J. Appleyard and M. R. Willis, Opt. Mater. (Amsterdam, Neth.), 9, 120
͑1998͒.
. H. Y. Yu, X. D. Feng, D. Grozea, Z. H. Lu, R. N. S. Sodhi, A.-M. Hor, and H. Aziz,
Appl. Phys. Lett., 78, 2595 ͑2001͒.
. S. T. Lee, Z. Q. Gao, and L. S. Hung, Appl. Phys. Lett., 75, 1404 ͑1999͒.
from reaching the cathode. This in turn increases the percentage of
holes that form excitons, thereby increasing device efficiency.
1
2
Conclusions
In this article, we have systematically investigated the physical
characteristics of the various MPc complexes and continued with the
application of these layers in reference to OLEDs. The HOMO and
LUMO level energies of MPcs are dependent on their central metal
atoms. Compared with the EL characteristics of the non-MPc de-
vices, the devices with the MPc layer inserted between the ITO
anode and HTL significantly improved the turn-on voltage and effi-
ciency of OLEDs. The turn-on voltage remains virtually the same
when the MPc layer thickness is adjusted in the range of 5–15 nm;
however, the efficiency is sensitive to the thickness of the MPc
layer. The determined energy level alignments measured by using
the photoemission and UV-visible techniques have provided direct
evidence to reveal that the origins of the improved turn-on voltage
of the devices can be attributed to the reduction of the barrier at the
MPc/NPB interface. This demonstrates that the MPc/NPB interface
instead of the ITO/MPc interface plays an important role in hole
injection and the optimum energy level between ITO and NPB
should be a HOMO, approaching that of NPB.
8. A. R. Schlatmann, D. Wilms Floet, A. Hilberer, F. Garten, P. J. M. Smulders, T. M.
Klapwijk, and G. Hadziioannou, Appl. Phys. Lett., 69, 1764 ͑1996͒.
9
. H. Vestweber and W. Rieß, Synth. Met., 91, 181 ͑1997͒.
0. J. Szuber and L. Grzadziel, Thin Solid Films, 391, 282 ͑2001͒.
1. W.-L. Yu, Y. Cao, J. Pei, W. Huang, and A. J. Heeger, Appl. Phys. Lett., 75, 3270
1
1
͑1999͒.
1
2. S. M. Tadayyon, H. M. Grandin, K. Griffiths, P. R. Norton, H. Aziz, and Z. D.
Popovic, Org. Electron., 5, 157 ͑2004͒.
3. J. C. Scott, J. H. Kaufman, P. J. Brock, R. DiPertro, J. Salem, and J. A. Gotia, J.
Appl. Phys., 79, 2745 ͑1996͒.
14. H. Aziz, Z. D. Popovic, N.-X. Hu, A.-M. Hor, and G. Xu, Science, 283, 1900
͑1999͒.
5. T. L. Anderson, G. C. Komplin, and W. J. Pietro, J. Phys. Chem., 97, 6577 ͑1993͒.
6. S. J. Liu, C. A. Chuang, S. Y. Chu, and P. C. Kao, J. Mater. Chem., Submitted.
7. N. B. Mckeown, Phthalocyanine Materials Synthesis, Structure and Function, pp.
32–47, Cambridge University Press, New York ͑1998͒.
18. L. Grzadziel, J. Żak, and J. Szuber, Thin Solid Films, 436, 70 ͑2003͒.
9. E. A. Ough, J. M. Stillman, and K. A. M. Creber, Can. J. Chem., 71, 1898 ͑1993͒.
0. A. Davidson, J. Chem. Phys., 77, 168 ͑1982͒.
1
1
1
1
1
2
2
2
1. J. Bardeen, F. Salt, and L. Hall, Photoconductivity Conference, p. 146 ͑1956͒.
2. A. Wolberg and J. Manassen, J. Am. Chem. Soc., 92, 2982 ͑1970͒.
23. D. Schlettwein and N. R. Armstrong, J. Phys. Chem., 98, 11771 ͑1994͒.
4. D. R. Kearns and M. Calvin, J. Chem. Phys., 34, 2026 ͑1961͒.
25. A. J. Ikushima, T. Kanno, S. Yoshida, and A. Maeda, Thin Solid Films, 273, 35
1996͒.
6. I. G. Hill, D. Milliron, J. Schwartz, and A. Kahn, Proc. SPIE, 3797, 276 ͑1999͒.
2
Acknowledgments
͑
Financial support was provided by the National Science Council,
Republic of China, under grant no. NSC-94-2215-E-006-008. We
also thank the Center for Micro-NanoTechnology, National Cheng
Kung University, for providing a single-side mask aligner ͑OAI,
J500͒.
2
27. S. M. Tadayyon, H. M. Grandin, K. Griffiths, L. L. Coatsworth, P. R. Norton, H.
Aziz, and Z. D. Popovic, Org. Electron., 5, 199 ͑2004͒.
8. F. Nuesch, M. Carrara, M. Schaer, D. B. Romero, and L. Zuppiroli, Chin. Phys.
Lasers, 347, 311 ͑2001͒.
9. B. Boudjema, G. Guillaud, M. Gamoudi, M. Maitrot, J. Andre, M. Martin, and J.
Simon, J. Appl. Phys., 56, 2323 ͑1984͒.
0. P. C. Kao, S. Y. Chu, Z. X. You, S. J. Liu, and C. A. Chuang, Thin Solid Films, to
be published.
2
2
3
3
National Cheng Kung University assisted in meeting the publication
costs of this article.
1. H. Riel, W. Brutting, T. Beierlein, E. Haskal, P. Muller, and W. Rieß, Synth. Met.,
References
111, 303 ͑2000͒.
1
2
. S. R. Forrest, Chem. Rev. (Washington, D.C.), 97, 1793 ͑1997͒.
. F. Li, H. Tang, J. Shinar, O. Resto, and S. Z. Weisz, Appl. Phys. Lett., 70, 2741
32. B. Masenelli, D. Berner, M. N. Bussac, F. Nüesch, and Zuppiroli, Appl. Phys. Lett.,
79, 4438 ͑2001͒.
͑
1997͒.
33. Z. Bao, A. J. Lovinger, and A. Dodabalapur, Appl. Phys. Lett., 69, 3066 ͑1996͒.
34. Z. Deng, S. T. Lee, D. P. Webb, Y. C. Chan, and W. A. Gambling, Synth. Met., 107,
107 ͑1999͒.
3
4
. C. C. Wu, C. I. Wu, J. C. Sturm, and A. Kahn, Appl. Phys. Lett., 70, 1348 ͑1997͒.
. I. M. Chan, W. C. Cheng, and F. C. Hong, Appl. Phys. Lett., 80, 13 ͑2002͒.