Optics and Photonics Journal, 2013, 3, 256-259
doi:10.4236/opj.2013.32B060 Published Online June 2013 (http://www.scirp.org/journal/opj)
Dependence Study of Optoelectronics Performance on
Carefully Differed LiF Thickness in Alq 3 Based OLEDs
Shengxun Su, Changxiao Pan, Xi Luo, Wei Chen, Jiarong Lian*
Key Laboratory of Optoelectronic Devices and Systems of Ministry of Education and Guangdong province,
Shenzhen University, No.3688, Nanhai Road, Nanshan District, Shenzhen 518060, China
Received 2013
ABSTRACT
The effect of LiF thickness on the electrical and luminescent characteristics in OLEDs has been studied by carefully
varying thickness value range from 0 nm to 1.2 nm. It’s interesting to find that the device with 0.2 nm LiF layer per-
forms the largest current and comparative lower luminescent efficiency, while the one with 0.6 nm LiF performs an-
other current peak (lower than that of device with 0.2 nm LiF layer) but the highest luminescent efficiency in all devices.
Here the much enhanced electron injection and destructive efficiency for 0.2 nm LiF device are understood by the
chemical interaction model at cathode interface, while the fairly increased electron injection and much improved effi-
ciency for 0.6 nm LiF device would be interpreted by other mechanisms, and LiF plays a protective part in preventing
the deposition-indu ce photoluminescence from quenching by Al cathode.
Keywords: OLEDs; Thickness; LiF; Chemical Interaction; Luminescent Quenching
1. Introduction
Since the report of Tang and VanSlyke [1] of multilayer
organic light-emitting devices (OLEDs), tremendous
efforts have been made to improve the device perform-
ance. To improve the performance of OLEDs, carrier
injection and transport mechanisms for organic materials
have been widely investigated. A famous approach for
efficient electron injection into Alq3 was first reported by
Hung et al. [2], who used Al metal together with a thin
interlayer of LiF. Although the effects of this layer cath-
ode on electron injection have been well recognized, the
role of the LiF electron injection layer (EIL) has long
been in dispute [2-4]. A number of mechanisms have
been proposed in the literatures, including: (a) the insu-
lating buffer model, which attributes electron injection
enhancement to the tunneling effect. The biased LiF
layer would take a large electric field because of its ex-
cellent insulatin g prop erty (with wid e b and gap o f 12eV),
and shift the electronic states of electrode to a higher
level thus a more direct electron injection into the lowest
unoccupied molecular orbital of the electron transporting
material can be obtained [4-6]; (b) Chemical reaction
model, in which chemical interaction takes place at the
interface of LiF/Al or Aluminum 8-hydroxyquinoli-
nate(Alq3)/LiF/Al [7-9]. The subsequent observed phe-
nomena include the LiF dissociation and Alq3 anion for-
mation with Li doping. The XPS results further indicate
that this reaction does not occur at LiF/Al interface in the
absence of Alq3, but in the presence of it [9]; (c) Bang
bending model, which makes the band bended to a cer-
tain extent, leading it easier for electron to inject into the
organic material [10]. The significance of these studies
lies in that they each support one of the different catego-
rized models. For example, using a LiF-doped Al com-
posite cathode to replace the conventional layer electrode
of LiF/Al exhibits the almost identical beneficial effect
on the I-V and L-I characteristics, and the enhanced elec-
tron injection in devices with a composite cathode can be
attributed neither to tunneling through an in sulator nor to
the band bending at the metal-organic interface [11]. An-
other study by Wang et al., however, opposed the result
and showed that a Ag/LiF(3 nm) cathode can also increase
the electron injection by varying the LiF thickness [12].
In this paper, we have researched the optical and elec-
trical performances dependence of the OLEDs on the
thickness of LiF interlayer. Although other researchers
have published articles to indicate that different optical
properties would be obtained when the LiF thickness
changes, the step of LiF thickness change in their re-
searches are much larger than that as we did. In our ex-
periments, it is interesting to find that there are two cur-
rent peaks when carefully increasing LiF thickness from
0 nm to 1.2 nm, which is different with that of others.
And, a conclusion can be safely drawn that different
*Corresponding author.

S. X. SU ET AL. 257
mecha- nisms for enhanced electron injection would play
a part when LiF thickness changes.
2. Experiment
In our experiments, the structures of reference OLEDs
are consist of indium tin oxide(ITO) substrates as anodes,
molybdenum oxide(MoO3) as hole injecting layers, NPB
as hole transport layers, A lq3 as electron transport layers,
LiF as electron injection layers and Al as cathodes. The
sequences of each layer are as follows: glass substrate/
ITO/MoO3 (10 nm)/NPB (40 nm)/Alq3 (60 nm)/LiF
(Xnm) /Al (100 nm), of which chooses X value between
0 nm to 1.2 nm with an increasing step of 0.1 nm. The
ITO/glasses substrates were used after routinely clean
process without oxidization treatment [13]. The devices
were prepared by thermal evaporation from resistively
heated crucibles in three connected vacuum chambers. In
one of the chambers, organic films of Alq3 and NPB are
deposited at a pressure of 2 × 10-4 pa at a deposition rate
of 0.1 nm/s, MoO3 and LiF layers were deposited in the
adjacent evaporation chamber at a pressure of 4 × 10-4 pa,
at rates of 0.03 nm/s and 0.01 nm/s, and metal Al elec-
trodes were deposited at a pressure of 8 × 10-4 pa, at a
rate of 1.5 nm/s. The rates of evaporation and thicknesses
of the film were monitored via Model TDM-200 measure
unit produced by Weitai Company. All samples were
rotated at uniform velocity during deposition so as to
guarantee the homogeneous films.
The current density (J) – luminance (L) – voltage (V)
characteristics of the devices were measured by a Keith-
ley SMU 2400 source-measure unit and BM-8 multi-
function radiant measurement meter in ambient conditio n
after device fabrication. Photoluminescence quenching
experimental was tested via LS 45 Fluorescence Spec-
trometer (PerkinElmer).
3. Results and Discussion
Figure 1 displays the typical current characteristics of
the OLEDs mentioned above, and the specific current
values at given voltages are also plotted in the insert of
Figure 1. As is shown in Figure 1, all devices with a LiF
interlayer perform obviously larger current than that of
standard devices, illustrating that all LiF layers with
thickness less than 1.2 nm improve electron injection
effectively, which is consistent with the reference reports
[2]. However, we can also find that devices show different
current performances with the LiF thickness change. To
show the changing law of the current more clearly, we
compared the current density at given voltage 12 V, 10 V
and 8 V, respectively. As is shown in insert, all three
curves at the given voltage have the same profile. Here
the one at 12 V is chosen as an example. When LiF
thick- ness increases from 0 nm to 0.2 nm, the current
density in- creases sharply and reaches its maximal value
(678 mA/cm2) at 0.2 nm. Then, it decreases rapidly when go
on increasing LiF thickness from 0.2 nm to 0.4 nm. While
further increasing LiF thickness from 0.4 nm to 0.6 nm,
current density of the device increases again and reaches
another maximal value (490 mA/cm2) at 0.6 nm. Lastly,
the current density decreases slowly while LiF thickness
further increases to 1.2 nm. Therefore, two maximal
values of current density can be found with the increase
of LiF thickness. None of the existing injection
mechanism, including chemical reaction model, band
bending, tunneling model etc. can explain the
two-maximal-value phenomenon independently. Thus, a
hypothesis can be easily drawn that there may be more
than one electron injection mechanisms are needed to
explain the two-maximal-value phenomenon.
In order to further investigate the influence of LiF
thickness on the luminescent characteristics of devices,
the current efficiency curves in devices are also plotted in
Figure 2. It can be seen that the standard device without
LiF layer, which performs the lowest current efficiency,
exhibits the maximal value of only 1.5 cd/A. After add-
ing a LiF interlayer, luminous efficiency of all the de-
vices are improved obviously, which probably because
the
Figure 1. the current-voltage and current-LiF thickness char-
acteristic for OLEDs with different thickness of LiF layer
between the Al cathode and light-emitting-layer.
Figure 2. The EL efficiency-current characteristic for OLEDs
with different thickness of LiF layer between Al cathode
and Alq3.
improved electron injection making the carrier more

S. X. SU ET AL.
258
balance in the emitting layer. As we all known, not only
the injection barrier of the hole is much lower than that
of the electron, but also the mobility of the hole trans-
porting layer is far greater than that of electron, thus the
device is in a hole-rich status [14]. Therefore, devices
with LiF layer help to improve the electron injection,
making carriers of the light-emitting layer more balanced
and reducing the hole-current leakage of devices, then
the luminous efficiency is improved. However, it also
shows that not all the devices follow the law that the
more balanced the carriers are, the higher luminous effi-
ciencies are performed. For example, the current of de-
vice reaches its maximal value when the LiF layer is 0.2
nm but its luminous efficiency is only 3.9 cd/A, which is
not the maximal value and is even lower than that of
other devices with LiF interlayer. As for comparison, the
device with LiF thickness of 0.6 nm, has the highest
maximal efficiency of 6.5 cd/A. Therefore, it indicates
that a large part of exciton in light emittin g layer must be
quenched when LiF thickness is 0.2 nm.
To confirm the quenching phenomenon in the light
emitting layer, we have further conducted photolumines-
cence (PL) experiment, which compares the PL intensity
of four kinds of films with structures of Alq3 (30 nm),
Alq3 (30 nm)/A l (2 nm), Alq3 (30 nm)/LiF (0.2 nm)/Al(2
nm) and Alq3 (30 nm)/LiF (0.7 nm)Al(2 nm). As is
shown in Figure 3, the largest PL intensity of 629 A.U.
at a wave- length of 505 nm has performed in the neat
Alq3 film. After evaporating a 2 nm Al on the Alq3 layer,
the PL intensity of the light-emitting layer reduced to 390
A.U. and about 38% of total intensity is lost, which re-
veals that Al atoms have a severe quenching effect on the
light-emitting layer. When inserting a 0.7 nm LiF layer,
the PL intensity is increased to 491 A.U., wh ich indicates
that the quenching effect of Al has been prevented obvi-
ously and the LiF layer plays the protective function. In
Figure 3. The photoluminescence (PL) curves for the films
of Alq3, Alq3/Al, Alq3/0.2nm LiF/Al and Alq3/0.2nm LiF/Al,
which the thickness of Alq3 and Al is 30nm and 2nm, re-
spectively.
fact, quenching effect of Al has been almost prevented
by the 0.7 nm LiF layer, because the Alq3/LiF (0.7 nm)
film plays the comparable PL intensity (489 A.U) with
that of Alq3/LiF (0.7 nm)/Al (2 nm), of which the further
deposition of Al layer takes near no effect on PL inten-
sity. On the contrast, when inserting a 0.2 nm LiF layer
between Alq3 and Al, the PL intensity is only 348 A.U.,
which is even lower than that of Alq3 (30 nm)/Al (2 nm).
Given that both of the LiF layers with thickness value of
0.2 nm and 0.7 nm have the same chemical states, the
LiF layer with 0.2 nm thickness would prevent the diffu-
sion of the Al atom and reduce the PL quenching effect
in a certain extent as the one with 0.7 nm thickness does,
thus improve the PL intensity, which is opposite to the
experimental results. Therefore a different chemical state
can be deduced for LiF layer at Alq3/Al interface when
its thickness changes from 0.7 nm to 0.2 nm, another
quenching factor may also be introduced into Alq3 layer.
Here the chemical reaction mechanism is the most prob-
able explanation for the lowest PL intensity in Alq3 (30
nm)/LiF(0.2 nm)/Al(2 nm) film. According to the chemi-
cal reaction model, Li was reported being released when
the chemical reaction took place at the ternary interface
of Alq3/LiF/Al, and diffusing in to Alq 3 layer, which would
exert certainly destructive effect on luminescent charac-
teristics [15].
4. Conclusions
In summary, we have carefully changed the thickness of
LiF electron injection layer and studied the optical and
electrical characteristics changes of OLEDs. Two peaks
in current density were founded by differing LiF thick-
ness from 0 nm to 1.2 nm. The device with 0.2 nm LiF
layer performs the largest current and comparative lower
luminescent efficiency, while the one with 0.6 nm LiF
performs another current peak (lower than that of device
with 0.2 nm LiF layer) but the highest luminescent effi-
ciency in all devices. Here the much enhanced electron
injection and destructive efficiency for device with a 0.2
nm LiF layer are understood by the chemical interaction
model at cathode interface, while the fairly increased
electron injection and much improved efficiency in the
one with a 0.6 nm LiF layer, would be interpreted by the
other mechanisms, and LiF taking a protective function
in preventing the deposition-induce photoluminescence
from quenching by Al cathode.
5. Acknowledgements
This work was supported by Natural Science Foundation
of China (Grand No. 61106094 and 20972097), Natural
Science Foundation of Guangdong Province (Grand No.
S2012020011003), Educational Commission of Guang-
dong Province (Grant No. LYM11111), Shenzhen Sci-
ence Foundation (Grant No. JC201005280458A).

S. X. SU ET AL.
259
REFERENCES
[1] C. W. Tang and S. A. VanSlyke, “Organic Electrolumi-
nescent Diodes,” Applied Physics Letters, Vol. 51, No.12,
1987, pp. 913-915. doi: 10.1063/1.98799
[2] L. S. Hung, C. W. Tang and M. G. Mason, “Enhanced
Electron Injection in Organic Electroluminescence De-
vices Using an Al/LiF Electrode,” Applied Physics Let-
ters, Vol. 70, No. 5, 1997, pp. 152-154.
doi: 10.1063/1.118344
[3] Y. Takahiro, Y. Daisuke, I. Eisuke, I. Hisao, O. Yukio
and S. Kazuhiko, “Energy Level Alignment at
Alq3/LiF/Al Interfaces Studied by Electron Pectroscopies:
Island Growth of LiF and Size- Dependence of the Elec-
tronic Structures,” Journal of Applied Physics, Vol. 42,
No. 6A, 2003, pp. 3666-3675.doi: 10.1143/JJAP.42.3666
[4] G. E. Jabbour, Y. Kawabe, S. E. Shaheen, J. F. Wang and
M. M. Morrell, “Highly Efficient and Bright Organic
Electroluminescent Devices with An Aluminum Cath-
ode,” Applied Physics Letters, Vol. 71, No. 13, 1997, pp.
1762-1764. doi:10.1063/1.119392
[5] R. Schlaf, B. A. Parkinson, P. A. Lee, K. W. Nebesny, G.
Jabbour, “Photoemission Spectroscopy of LiF Coated Al
and Pt electrodes,” Journal of Applied Physics, Vol. 84,
No. 12, 1998, pp. 6729-6736. doi: 10.1063/1.369000
[6] Y. E. Kim, H. Park and J. J. Kim, “Enhanced Quantum
Efficiency in Polymer Electroluminescence Devices by
Inserting A Tunneling Barrier Formed by Lang-
muir-Blodgett Films,” Applied Physics Letters, Vol. 69,
No. 5, 1996, pp. 599-601. doi: 10.1063/1.117919
[7] M. G. Mason, C. W. Tang, L. S. Hung, P. Raychaudhuri
and J. Madathil, “Interfacial Chemistry of Alq3 and LiF
with Reactive Metals,” Journal of Applied Physics, Vol.
89, No. 5, 2001, pp. 2756-2765.
doi: 10.1063/1.1324681
[8] H. Heil, J. Steiger, S. Karg, M. Gastel and H. Ortner,
“Mechanisms of Injection Enhancement in Organic Light-
emitting Diodes through An Al/LiF Electrode,” Journal
of Applied Physics, Vol. 89, No. 1, 2001, pp. 420-424.
doi: 10.1063/1.1331651
[9] Z. T. Xie, W. H. Zhang, B. F. Ding, X. D. Gao and Y. T.
You, “Interfacial Reactions at Al/LiF and LiF/Al,” Ap-
plied Physics Letters, Vol. 94, No.6, 2009, pp.
063302-01- 063302-03. doi: 10.1063/1.3077167
[10] K. Ihm, T.-H. Kang, K.-J. Kim, C.-C. Hwang and Y.-J.
Park, “Band Bending of LiF/Alq3 Interface in Organic
Light-Emitting Diodes,” Applied Physics Letters, Vol.
83, No. 14, 2003, pp. 2949-2952. doi: 10.1063/1.1616977
[11] L. S. Hung, R. Q. Zhang, P. He and M. G. Mason, “Con-
tact Formation of LiF/Al Cathodes in Alq3-Based Organic
Light-Emitting Diodes,” Journal of Physics D: Applied-
Physics, Vol. 35, No. 2, 2002, pp. 103-107.
doi：10.1088/0022-3727/35/2/302
[12] X. J. Wang, J. M. Zhao, Y. C. Zhou, X. Z. Wang and S. T.
Zhang, “Enhancement of Electron Injection in Organic
Light-Emitting Devices Using An Ag/LiF Cathode,”
Journal of Applied Physics, Vol. 95, No. 7, 2004, pp.
3828-3830. doi: 10.1063/1.1655676
[13] J. R. Lian, Y. W. Liu, F. F. Niu and P. J. Zeng, “Im-
proved Electron Injection of OLEDs with A Thin PBD
Layer At Alq3/Cs2CO3 Interface,” Applied Surface Sci-
ence, Vol. 257, No. 10, 2011, pp. 4608-4611.
doi:10.1016/j.apsusc.2010.12.092
[14] G. E. Jabbour, J.–F. Wang and N. Peyghambarian,
“High-Efficiency Organic Electrophosphorescent Devices
through Balance of Charge Injection,” Applied Physics
Letters, Vol. 80, No. 11, 2002, pp. 2026-2028.
doi: 10.1063/1.1458687
[15] J. J. Kido and Toshio Matsumoto, “Bright Organic Elec-
troluminescent Devices Having A Metal-Doped Electron
Injecting Layer,” Applied Physics Letters, Vol. 73, No. 20,
1998, pp. 2866-2868. doi: 10.1063/1.122612