Content uploaded by Hyunbok Lee
Author content
All content in this area was uploaded by Hyunbok Lee on Jan 08, 2014
Content may be subject to copyright.
The origin of the hole injection improvements at indium tin
oxide/molybdenum trioxide/N,N′-bis(1-naphthyl)-N,N′-diphenyl-1,1′-
biphenyl- 4,4′-diamine interfaces
Hyunbok Lee, Sang Wan Cho, Kyul Han, Pyung Eun Jeon, Chung-Nam Whang et al.
Citation: Appl. Phys. Lett. 93, 043308 (2008); doi: 10.1063/1.2965120
View online: http://dx.doi.org/10.1063/1.2965120
View Table of Contents: http://apl.aip.org/resource/1/APPLAB/v93/i4
Published by the AIP Publishing LLC.
Additional information on Appl. Phys. Lett.
Journal Homepage: http://apl.aip.org/
Journal Information: http://apl.aip.org/about/about_the_journal
Top downloads: http://apl.aip.org/features/most_downloaded
Information for Authors: http://apl.aip.org/authors
Downloaded 01 Oct 2013 to 202.116.1.149. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
The origin of the hole injection improvements at indium tin oxide/
molybdenum trioxide/N,N⬘-bis„1-naphthyl…-N,N⬘-diphenyl-1,1⬘-biphenyl-
4,4⬘-diamine interfaces
Hyunbok Lee,1Sang Wan Cho,1,a兲Kyul Han,1Pyung Eun Jeon,1Chung-Nam Whang,1
Kwangho Jeong,1Kwanghee Cho,2and Yeonjin Yi3,b兲
1Institute of Physics and Applied Physics, Yonsei University, 134 Sinchon-dong, Seodaemoon-Gu, Seoul
120-749, Republic of Korea
2Research and Development Division, Hynix Semiconductor Inc., San 136-1, Ami-Ri Bubal-Eub Icheon-Si,
GyeungGi-Do 467-701, Republic of Korea
3Division of Advanced Technology, Korea Research Institute of Standards and Science, 209 Gajeong-Ro,
Yuseong-Gu, Daejeon 305-340, Republic of Korea
共Received 9 June 2008; accepted 8 July 2008; published online 1 August 2008兲
We investigated the interfacial electronic structures of indium tin oxide 共ITO兲/molybdenum trioxide
共MoO3兲/N,N⬘-bis共1-naphthyl兲-N,N⬘-diphenyl-1,1⬘-biphenyl-4, 4⬘-diamine 共NPB兲using in situ
ultraviolet and x-ray photoemission spectroscopy to understand the origin of hole injection
improvements in organic light-emitting devices 共OLEDs兲. Inserting a MoO3layer between ITO and
NPB, the hole injection barrier was remarkably reduced. Moreover, a gap state in the band gap of
NPB was found which assisted the Ohmic hole injection at the interface. The hole injection barrier
lowering and Ohmic injection explain why the OLED in combination with MoO3showed improved
performance. © 2008 American Institute of Physics.关DOI: 10.1063/1.2965120兴
Since the development of organic light-emitting devices
共OLEDs兲with thin-multilayer structures,1many studies on
OLEDs have been conducted by academic and industrial re-
searchers. However, more efforts are needed to solve invet-
erate problems such as high operation voltage, luminance
efficiency, and lifetime. These problems could be solved by
implementing proper interfaces among the layers composing
the organic devices. In the case of the interface between an
anode and a hole transport layer 共HTL兲, various organic or
inorganic interlayers have been adopted such as copper
phthalocyanine 共CuPc兲,23,4,9,10 perylenetetracarboxylic
dianhydride,3SiO2,4SiOxNy,5TiO2共Ref. 6兲, and several
metal oxides.7In general, it has been considered that these
materials act as a buffer layer to balance the concentrations
of holes and electrons and/or as a hole injection layer 共HIL兲
to reduce the hole injection barrier between anode and HTL.
In addition, the interlayer makes good contact between an
anode and a HTL by forming a smooth surface on the anode.
These interlayers have shown markedly improved device
properties.
Recently, transition metal oxides, such as molybdenum
oxide 共MoO3兲, vanadium oxide 共V2O5兲, and tungsten oxide
共WO3兲have attracted much attention due to their hole injec-
tion barrier lowering properties8,9and charge generation abil-
ity in tandem OLEDs.10–12 You et al.13 reported significantly
improved OLEDs with a MoO3interlayer. The device with a
MoO3layer showed better performance than the device with
a CuPc layer which has been used commonly as a HIL to
enhance both luminance-voltage characteristics and life-
time. In addition, Matsushima et al.14 reported Ohmic hole
injection by inserting the MoO3layer between indium tin
oxide 共ITO兲and N,N⬘-bis共1-naphthyl兲-N,N⬘-diphenyl-1,1⬘-
biphenyl-4,4⬘-diamine 共NPB兲. However, the mechanisms for
the hole injection barrier lowering and the Ohmic injection
from ITO to NPB are not clear yet. The interfacial electronic
structures and the energy level alignments at ITO/MoO3/
NPB interfaces would play a crucial role in the improve-
ments in device performance. We performed in situ ultravio-
let photoemission spectroscopy 共UPS兲and x-ray photoemis-
sion spectroscopy 共XPS兲measurements to study the
ITO/MoO3/NPB interfaces hence clarifying the origin of the
improvement in device performance. UPS and XPS allowed
us to understand the electronic structures of the interfaces
from the information of the vacuum level, highest occupied
molecular orbital 共HOMO兲, and core levels.
We fabricated a multilayer film of ITO /MoO3/NPB
structure and performed in situ UPS and XPS measurements.
Silver paste was employed at the contact region between an
ITO-coated glass substrate and the sample holder in order to
secure the electrical contact between the ITO and analyzer. A
clean ITO surface was obtained with Ar+ion sputtering so
the surface carbon was adequately reduced. We exposed the
ITO surface to the Ar+ions for as short as possible time in
order to maintain its pristine stoichiometry. MoO3and NPB
were thermally evaporated onto the ITO substrate in a depo-
sition chamber maintained below 5⫻10−8 Torr. The deposi-
tion rates of both MoO3and NPB were controlled to be
0.01 nm/s with evaporation temperatures of 430 and 130 °C,
respectively. To investigate the interface formation,
MoO3共5.0 nm兲and NPB 共0.1, 0.3, 0.5, 1.0, and 2.0 nm兲
were deposited on the clean ITO in a stepwise manner. Be-
fore and after each deposition step, the sample was trans-
ferred to the analysis chamber and analyzed without break-
ing the vacuum. The analysis chamber was composed of a
hemispherical electron energy analyzer 共PHI 5700 spectrom-
eter兲, monochromatic x-ray 共Al K
␣
, 1486.6 eV兲, and ultra-
violet 共He I, 21.21 eV兲discharge source. The UPS spectra
a兲Present address: Department of Physics, Boston University, 590 Common-
wealth Avenue, Boston, Massachusetts 02215, USA.
b兲Author to whom correspondence should be addressed. Electronic mail:
yeonjin@kriss.re.kr.
APPLIED PHYSICS LETTERS 93, 043308 共2008兲
0003-6951/2008/93共4兲/043308/3/$23.00 © 2008 American Institute of Physics93, 043308-1
Downloaded 01 Oct 2013 to 202.116.1.149. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
were obtained with a sample bias of −15 V in normal emis-
sion geometry to measure the secondary electron cutoff. We
plotted all spectra with respect to the weak but well defined
Fermi level 共EF兲of ITO. We also prepared an ITO/NPB
multilayer film and carried out similar measurements as a
control sample.
Figure 1shows the XPS core level spectra of each ele-
ment during the step-by-step deposition. The In 3dpeak 共a兲
was measured from the sputter cleaned ITO surface and it
attenuated completely after deposition of the 5 nm MoO3
layer, implying that the ITO surface was fully covered with
MoO3. The Mo 3dpeak position 关232.9 eV; Fig. 1共b兲兴
showed good agreement with the MoO3phase compared to
literature.15,16 The intensities of the Mo 3dpeaks attenuated
gradually with the deposition of NPB. However, they did not
show significant shifts or shape changes during the NPB
deposition, which shows there was neither band bending 共Vb兲
nor chemical reactions in the MoO3layer during the NPB
deposition. Figure 1共c兲shows the series of C 1sspectra dur-
ing the NPB layer deposition followed by the MoO3deposi-
tion on ITO. There is no significant carbon contamination on
the sputter cleaned ITO and MoO3deposited. The C 1s
peaks appeared and intensified as the NPB layer thickened.
The Vbof the NPB layer 共0.15 eV兲was estimated from the
shift of the C 1speaks during the step-by-step measure-
ments and it was cross-checked with the HOMO shift ob-
tained from the UPS spectra since the band bending should
occur throughout all the energy levels.
Figures 2共a兲–2共c兲show the secondary electron cutoff re-
gion, full range, and HOMO region of UPS spectra, respec-
tively. Figure 2共b兲shows spectral changes clearly with the
deposition of MoO3and NPB on ITO. The MoO3valence
band structures are clearly seen at 3–10 eV 共Ref. 17兲
with the MoO3共5兲spectrum and the spectral features of NPB
dominating the spectra as the NPB layer thickens. In Fig.
2共a兲, the cutoff position shifted toward lower energy by 2.35
eV immediately after the 5 nm MoO3layer was deposited.
This shows the formation of a huge interface dipole18 on ITO
since the MoO3has very large work function.17 As NPB was
deposited on the MoO3共5nm兲covered ITO, the position of
secondary cutoff gradually moved back toward higher bind-
ing energies and the shift was saturated with 2 nm NPB
deposition. The total secondary cutoff shift during NPB
deposition was estimated to be 1.90 eV. Figure 2共c兲shows
the HOMO region spectra with background removal corre-
sponding to each deposition step. The HOMO onset posi-
tions were determined by line fitting of the HOMO onset
after considering the analyzer broadening 0.1 eV for our
system.19 The Vbwas determined from the shift of the
HOMO onset position as NPB layer thickened. The deter-
mined Vbwas 0.15 eV which accords well with the core level
shift 共see C 1speaks in Fig. 1兲.
The most remarkable changes in the spectra are the for-
mation of gap state in the gap of NPB with the MoO3layer
insertion. The gap state at 0.15 eV is seen clearly at the first
共0.1 nm兲and second 共0.3 nm兲deposition steps of NPB on
MoO3. The gap state was not observed at the interface of
ITO/NPB 共not shown兲. As the NPB layer thickened over 0.5
nm, the gap state disappeared, implying that the gap state
exists at very short range from the interface between MoO3
and NPB. This gap state could be generated by weak inter-
action since the XPS spectra did not show any strong chemi-
cal reactions.20
Combining the information of HOMO and vacuum level
shifts in combination with the band bending, we drew the
energy level diagrams of ITO/NPB and ITO/MoO3/NPB in
Figs. 3共a兲and 3共b兲.共The detailed procedures are presented
elsewhere.19兲The energy level diagram of ITO/NPB was
evaluated with the similar UPS and XPS measurements. The
position of the lowest unoccupied molecular orbital was es-
timated from a previously reported energy gap of 3.10 eV for
NPB.21 The ionization energy of NPB 共Eion兲showed good
agreement for both cases compared to previous results within
the error margin of our measurement 共0.05– 0.1 eV兲.22 In the
case of the ITO/NPB interface 共a兲, the interface dipole 共eD兲
was negligible 共0.03 eV兲and the hole injection barrier 共⌽h兲
was determined to be 1.64 eV. On the other hand, an ex-
tremely large interface dipole was induced after the deposi-
tion of the MoO3layer. Although the vacuum level was low-
ered again with the subsequent NPB deposition, the amount
of the lowering did not exceed the initial vacuum level rising
FIG. 1. 共Color online兲Measured XPS core level spectra of 共a兲In 3d,共b兲
Mo 3d,and共c兲C1sin between the MoO3and NPB layer depositions.
Each spectrum was obtained from sputter cleaned ITO, ITO/MoO3共5nm兲
and ITO/MoO3共5nm兲/NPB 共0.1, 0.3, 0.5, 1.0, and 2.0 nm兲.
FIG. 2. 共Color online兲The measured UPS spectra of the 共a兲secondary
electron cutoff region normalized for easy comparison, 共b兲full range, and
共c兲HOMO region with background removal before and after the MoO3and
NPB layer deposition. Each spectra was obtained from sputter cleaned ITO,
ITO/MoO3共5nm兲and ITO/MoO3共5nm兲/NPB 共0.1, 0.3, 0.5, 1.0, and 2.0
nm兲. The fitting procedures for determining the HOMO onset is indicated.
043308-2 Lee et al. Appl. Phys. Lett. 93, 043308 共2008兲
Downloaded 01 Oct 2013 to 202.116.1.149. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions
with the MoO3layer. Therefore, the vacuum level was effec-
tively raised, resulting in the ⌽hof 0.89 eV, greatly reduced
from its previous value, with MoO3insertion 共b兲. From the
view point of the charge injection, Matsushima et al.14 re-
ported Ohmic hole injection with the MoO3interlayer be-
tween ITO and NPB. We directly observed a gap state in the
gap of NPB indicated with a dotted line. This state could be
attributed to the charge transfer.14 This state was very close
to the Fermi level so that the charge carriers could be in-
jected easily from ITO to NPB; i.e., Ohmic hole injection
could occur. This explains the origin of the Ohmic hole in-
jection at the interface.
In summary, we investigated the role of a MoO3layer
between ITO and NPB using in situ photoemission measure-
ments. The hole injection barrier was highly reduced with the
insertion of MoO3layer compared to the interface without
MoO3. Furthermore, the formation of the gap state in the
NPB gap assisted the Ohmic hole injection from the ITO
electrode. The hole injection barrier lowering and Ohmic
hole injection explain why the OLED in combination with
MoO3showed excellent performance.
This work was supported by Institute of Physics and
Applied Physics 共IPAP兲, Yonsei University, and BK21
project of the Korea Research Foundation 共KRF兲.
1C. W. Tang and S. A. VanSlyke, Appl. Phys. Lett. 51, 913 共1987兲.
2S. A. Van Slyke, C. H. Chen, and C. W. Tang, Appl. Phys. Lett. 69,2160
共1996兲.
3L. Chkoda, C. Heske, M. Sokolowski, and E. Umbach, Appl. Phys. Lett.
77, 1093 共2000兲.
4Z. B. Deng, X. M. Ding, S. T. Lee, and W. A. Gambling, Appl. Phys. Lett.
74, 2227 共1999兲.
5C. O. Poon, F. L. Wong, S. W. Tong, R. Q. Zhang, C. S. Lee, and S. T.
Lee, Appl. Phys. Lett. 83, 1038 共2003兲.
6Z. Zhi-Feng, D. Zhen-Bo, L. Chun-Jun, Z. Meng-Xin, and X. Deng-Hui,
Displays 24, 231 共2003兲.
7S. Tokito, K. Noda, and Y. Taga, J. Phys. D 29,2750共1996兲.
8J. Li, M. Yahiro, K. Ishida, H. Yamada, and K. Matsushige, Synth. Met.
151, 141 共2005兲.
9T.-Y. Chu, J.-F. Chen, S.-Y. Chen, C.-J. Chen, and C. H. Chen, Appl. Phys.
Lett. 89, 053503 共2006兲.
10C.-C. Chang, J.-F. Chen, S.-W. Hwang, and C. H. Chen, Appl. Phys. Lett.
87, 253501 共2005兲.
11C.-W. Chen, Y.-J. Lu, C.-C. Wu, E. H.-E. Wu, C.-W. Chu, and Y. Yang,
Appl. Phys. Lett. 87, 241121 共2005兲.
12F. Guo and D. Ma, Appl. Phys. Lett. 87, 173510 共2005兲.
13H. You, Y. Dai, Z. Zhang, and D. Ma, J. Appl. Phys. 101, 026105 共2007兲.
14T. Matsushima, Y. Kinoshita, and H. Murata, Appl. Phys. Lett. 91, 253504
共2007兲.
15M. A. Bica de Moraes, B. C. Trasferetti, F. P. Rouxinol, R. Landers, S. F.
Durrant, J. Scarminio, and A. Urbano, Chem. Mater. 16, 513 共2004兲.
16J. F. Moudler, W. F. Stickle, P. E. Sobol, and K. D. Bomben, Handbook of
X-Ray Photoemission Spectroscopy 共Perkin-Elmer, Minnesota, 1992兲.
17R. Tokarz-Sobieraj, K. Hermann, M. Witko, A. Blume, G. Mestl, and R.
Schlogl, Surf. Sci. 489, 107 共2001兲.
18H. Ishii, K. Sugiyama, E. Ito, and K. Seki, Adv. Math. 11,605共1999兲.
19R. Schlaf, P. G. Schroeder, M. W. Nelson, B. A. Parkinson, C. D. Merritt,
L. A. Crisafulli, H. Murata, and Z. H. Kafafi, Surf. Sci. 450,142共2000兲.
20Y. Yi, S. J. Kang, K. Cho, J. M. Koo, K. Han, K. Park, M. Noh, C.-N.
Whang, K. Jeong, and E. J. Hahn, Appl. Phys. Lett. 86, 113503 共2005兲.
21E. W. Forsythe, V. E. Choong, T. Q. Le, and Y. Gao, J. Vac. Sci. Technol.
A17, 3429 共1999兲.
22S. W. Cho, K.-H. Yoo, K. Jeong, C.-N. Whang, Y. Yi, and M. Noh, Appl.
Phys. Lett. 91, 052102 共2007兲.
FIG. 3. 共Color online兲Energy level diagrams of 共a兲ITO/NPB and 共b兲
ITO/MoO3/NPB. ⌽eand ⌽hare the electron and hole injection barrier. Eion
and eD are the ionization energy of NPB and interface dipole. The band
bending 共Vb兲was evaluated from the shift of HOMO and C 1slevel during
the deposition. Comparing with the diagrams, the hole injection barrier of
ITO/MoO3/NPB is much lower than that of ITO/NPB. Noticeable gap state
was observed with the insertion of MoO3layer.
043308-3 Lee et al. Appl. Phys. Lett. 93, 043308 共2008兲
Downloaded 01 Oct 2013 to 202.116.1.149. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://apl.aip.org/about/rights_and_permissions