Page 1
Fabrication of nickel oxide thin films and application thereof in organic electronics
by
Leonid Mordoukhovski
A thesis submitted in conformity with the requirements for the degree of Master of Applied Science
School of Graduate Studies, Department of Materials Science & Engineering University of Toronto
© Copyright by Leonid Mordoukhovski 2010
Page 2
ii
Fabrication of nickel oxide thin films
and application thereof in organic electronics
Leonid Mordoukhovski
Master of Applied Science
School of Graduate Studies, Department of Materials Science & Engineering
University of Toronto
2010
Abstract
This work investigates fabrication methods of nickel oxide thin films and their use in
organic electronics. Two fabrication techniques were studied: UV-ozone oxidation of pure nickel
films and reactive RF magnetron sputtering. The former was used to produce Ni/Ni2O3 bi-layer
anodes to use as a substitute for the de facto standard ITO anode. OLEDs fabricated using
Ni/Ni2O3 bi-layer anodes exhibited comparable device performance to standard ITO devices.
UV-ozone oxidation was also used to fabricate Ni2O3 buffer layers for OPVs. Solar cells
fabricated using Ni2O3 coated ITO exhibited an enhanced power conversion efficiency of up to
90%. RF magnetron sputtering was used to produce NiOx buffer layers with tunable conductivity
and optical transparency for OPVs. Solar cells fabricated using NiOx coated ITO exhibited an
enhanced power conversion efficiency of up to 60%. Nickel oxide films have been characterized
with various techniques: sheet resistance measurements, optical transmission, XPS, UPS, AFM,
and TEM.
Page 3
iii
Acknowledgements
I would like to thank the members of the Lu Group for helping me with various stages of
my research: Michael Helander, Zhibin Wang, Mark Greiner. My appreciation also goes to two
members who I worked on joined experiments with: Graham Murdoch and Dong Gao. A special
thank you to Dr. Daniel Grozea for working with me side-by-side on a number of experiments
and developing a patent application.
Also, I would like to show gratitude to various research facilities and people that run them
that I used and worked with throughout my two years as a Master‟s student: Adrienne Tanur and
Claudia Grozea at Prof. Gilbert Walker‟s lab at the Department of Chemistry for access to and
help with AFM characterization, Dr. Ilia Gourevich and Dr. Neil Combs for the use of STEM,
SEM and TEM facilities at the Department of Chemistry Centre for Nanostructure Imaging; Prof.
Ted Sargent‟s lab facility for the use of photospectrometer equipment at the Department of
Electrical Engineering; Peter Brodersen at Surface Institute of Ontario for use of profilometer
equipment at the Department of Chemical Engineering; and Sal Boccia for use of SEM
equipment at the Department of Materials Science and Engineering.
I would like to thank the Department of Materials Science & Engineering, Lu Group,
NSERC and Vale INCO for financial support of this research at various stages.
Finally I, would like extend gratitude to my thesis supervisor and the director of Lu
Group, Prof. Zheng-Hong Lu, for putting together a state of the art fabrication and
characterization equipment that made this research physically possible, for guiding me through
the tough world of research and development of cutting-edge technologies that can have a real
Page 4
iv
economic impact on the market place and society, and for many valuable lessons and advices in
the field of organic electronics technology, commercialization of new ideas and innovations, and
life in general.
Page 5
v
Table of Contents
Abstract .......................................................................................................................................... ii
Acknowledgements ...................................................................................................................... iii
List of Tables .............................................................................................................................. viii
List of Figures ............................................................................................................................... ix
List of Units ................................................................................................................................. xii
List of Abbreviations ................................................................................................................. xiv
Chapter 1: Introduction ................................................................................................................1
1.1 Motivation ............................................................................................................................1
1.2 Organic Light Emitting Diodes (OLEDs) ............................................................................2
1.3 Organic Photovoltaic Devices (OPVs) ................................................................................7
1.4 Nickel and Nickel Oxide Materials......................................................................................9
Chapter 2: Fabrication and characterization of nickel oxide thin films…..…………………11
2.1 Review of fabrication techniques.......................................................................................11
2.2 Reactive RF sputtering technique, NiOx ............................................................................14
2.2.1 First “shotgun” trial experiment.............................................................................15
Page 6
vi
2.2.2 Fabrication of NiOx at 150 W constant deposition power .....................................18
2.2.3 XPS and work function measurements of NiOx .....................................................20
2.2.4 Sputtering of NiOx at different target-to-substrate distances .................................21
2.3 UV-ozone oxidation technique, Ni2O3 ...............................................................................24
2.3.1 Description of fabrication and characterization methods ......................................24
2.3.2 Characterization of Ni2O3 films .............................................................................26
Chapter 3: Ni/Ni2O3 bi-layers as anodes for OLEDs……………………………………..….. 32
3.1 Organic Materials...............................................................................................................32
3.2 OLED Fabrication ..............................................................................................................33
3.3 OLED Performance ...........................................................................................................36
3.4 Charge carrier injection......................................................................................................39
3.5 UPS and work function measurements ..............................................................................42
Chapter 4: Nickel oxide buffer layers for OPVs .......................................................................45
4.1 Organic Materials...............................................................................................................46
4.2 Fabrication of OPVs ..........................................................................................................46
4.3 OPVs with NiOx buffer layers ...........................................................................................47
4.4 OPVs with Ni2O3 Buffer layers .........................................................................................50
Page 7
vii
4.5 Electron Blocking Action ..................................................................................................52
Chapter 5: Conclusions……………………………………………………………….………...54
5.1 OLEDs and Ni2O3 ..............................................................................................................54
5.2 OPVs and nickel oxide buffer layers .................................................................................55
Chapter 6: Future work……………………………………………………...…………………56
6.1 OLEDs ...............................................................................................................................56
6.2 OPVs ..................................................................................................................................56
References .....................................................................................................................................58
Appendix A: OLED with CuPc Layer .......................................................................................63
Appendix B: Contributions to Research and Development .....................................................64
Page 8
viii
List of Tables
Table 1. NiOx fabrication parameters and characterization. ................................................... 16
Table 2: Summary of sputtering parameters (dry run)............................................................ 17
Table 3: Deposition of 50 nm NiOx films at 150 W constant deposition power. ................... 18
Table 4: Sheet Resistance Characteristics (Ω/□). ................................................................... 26
Table 5: % Transmission at 525 nm. ...................................................................................... 27
Table 6: Solar cell characteristics (NiOx buffer). ................................................................... 48
Table 7: Solar cell characteristics (Ni2O3 buffer).................................................................... 51
Page 9
ix
List of Figures
Figure 1. Typical OLED structure. NPB and Alq3 are respectively the hole transport layer (HTL)
and the electron transport layer (ETL). Alq3:C545T is the emissive layer, where C545T serves as
a fluorescent dopant. ................................................................................................................. 3
Figure 2. Band diagram of a typical OLED structure. .............................................................. 4
Figure 3. SONY XEL-1 OLED TV. ......................................................................................... 5
Figure 4. Operating mechanism of an OPV. ............................................................................. 8
Figure 5. RF sputter system used in this work. ....................................................................... 15
Figure 6. Transmission of NiOx films (1st deposition experiment). ........................................ 16
Figure 7. Conductivity and optical transmission as a function of deposition pressure........... 19
Figure 8. Ni 2p core-level XPS spectra of pure Ni and NiOx thin films. ................................ 21
Figure 9. SEM micrographs of NiOx films produced at: (top) 8 cm, (middle) 12 cm and (bottom)
19 cm target-to-substrate distance. ......................................................................................... 23
Figure 10. Fabrication of Ni2O3 thin films.............................................................................. 24
Figure 11. Schematic of anode pattern on glass substrate. ..................................................... 25
Figure 12. Ni 2p XPS spectra of pure and oxidized Ni films. ................................................ 28
Figure 13. PS bulk 5 vs. surface 45 ) profile of the oxidized Ni film. ............................ 29
Page 10
x
Figure 14. Timeline of nickel oxide growth. .......................................................................... 30
Figure 15. AFM of Ni and Ni2O3 thin films. .......................................................................... 31
Figure 16. Kurt J. Lesker cluster tool. .................................................................................... 34
Figure 17. Operating OLED Ni/Ni2O3 anode. ........................................................................ 35
Figure 18. Vacuum cryostat. ................................................................................................... 36
Figure 19. Power efficiency vs. luminance characteristics of ITO and Ni/Ni2O3 based OLEDs.
................................................................................................................................................. 37
Figure 20. Luminance vs. voltage characteristics of ITO and Ni/Ni2O3 based OLEDs. ........ 38
Figure 21. Normalized emission intensities of ITO and Ni/Ni2O3 based anodes: constant current
applied for 1.5 hrs. .................................................................................................................. 39
Figure 22. I-V characteristics of single carrier hole-only devices with ITO, Ni and Ni/Ni2O3
anodes. .................................................................................................................................... 40
Figure 23. Temperature dependent IV characteristics of single carrier hole-only devices with
ITO, Ni and Ni/Ni2O3. ............................................................................................................. 41
Figure 24. He I valence band spectra for Ni and Ni2O3 thin films. ........................................ 43
Figure 25. Schematic of BHJ solar cell device. ...................................................................... 45
Figure 26. Solar cells JV characteristics (NiOx buffer). ......................................................... 48
Figure 27. Solar cell JV characteristics (Ni2O3 buffer). .......................................................... 51
Page 11
xi
Figure 28. A band diagram schematic of a typical planar OPV structure. Because the LUMO
level of the donor is much lower than the HOMO level of the acceptor, the flow of electrons
towards the anode is blocked. ................................................................................................. 52
Figure 29. Band Diagram for OPVs with NiOx or Ni2O3 buffer layers. The flow of electrons
towards the anode is inhibited by the nickel oxide buffer layer, which has conduction band much
lower than the LUMO of the acceptor. ................................................................................... 53
Figure 30. a) Power efficiency vs. luminance characteristics for ITO, Ni and Ni/Ni2O3 based
OLEDs with CuPc. b) Luminance vs. voltage for ITO, Ni and Ni/Ni2O3 based OLEDs with
CuPc. ....................................................................................................................................... 63
Page 12
xii
List of Units
Ωcm ohms times centimeter (bulk resistivity)
Ω-1
cm-1
(bulk conductivity)
eV electron volt (energy)
K degree Kelvin (temperature)
nm nanometer (distance)
C degree Celsius (temperature)
Torr (vacuum pressure)
mTorr millitorr (pressure)
SCCM standard cubic centimeters per second (gas flow rate)
W watt (power)
A amperes (current)
Å/s angstrom per second (rate of deposition)
Ω/□ ohms per square (sheet resistance)
cm centimeter (distance)
min minutes (time)
s seconds (time)
hr hours (time)
m meters (distance)
Page 13
xiii
mm millimeter (distance)
cd/m2 candela per meter squared (luminance)
lm/w lumens per watt (power efficiency)
V volts (voltage)
A/cm2 amperes per centimeter squared (current density)
mA/cm2 millamperes per centimeter squared (current density)
a.u. arbitrary units
mW/cm2 milliwatts per centimeter squared (solar illumination power flux)
Page 14
xiv
List of Abbreviations
AFM Atomic Force Microscopy
Alq3 Tris-(8-hydroxy-quinolinato)aluminum
AZO Aluminum Zinc Oxide
BHJ Bulk Heterojunction (solar cell)
C545T 2,3,6,7-Tetrahydro-1,1,7,7,-tetramethyl-1H, 5H,11H-10-
(2-benzothiazolyl)quinolizino-[9,9a,1gh]coumarin
CNL Charge Neutrality Level
CuPc Copper Phthalocyanine
DC Direct Current
EL Emission Layer
ETL Electron Transport Layer
FF Fill Factor
FTO Fluorine Doped Tin Oxide
HTL Hole Transport Layer
HOMO Highest Occupied Molecular Orbital
ITO Indium Tin Oxide
Jsc Short Circuit Current (solar cell)
LUMO Lowest Unoccupied Molecular Orbital
Page 15
xv
MSR Mean Surface Roughness
NPB 4,4‟-Bis[N-(1-napthyl)-N-phenyl-amino] biphenyl
OLED Organic Light Emitting Device
OPV Organic Photovoltaic
P3HT Poly(3-hexylthiophene), donor for OPVs
PCBM Phenyl-C61-butyric acid methyl ester, acceptor for OPVs
PLD Pulsed Laser Deposition
RF Radio Frequency
Rs Series Parasitic Resistance (solar cell)
Rsh Shunt Parasitic Resistance (solar cell)
SEM Scanning Electron Microscopy
SMU Source Measurement Unit
STEM Scanning Transmission Electron Microscopy
TCO Transparent Conducting Oxide
TEM Transmission Electron Microscopy
UPS Ultraviolet Photoelectron Spectroscopy
UV-Vis-NIR Ultraviolet-Visible-Near Infrared
Voc Open Circuit Voltage (solar cell)
XPS X-ray Photoelectron Spectroscopy
Page 16
1
Chapter 1 : Introduction
1.1 Motivation
In recent years thin film organic semiconductors have become a viable alternative to
conventional inorganic semiconductors, such as Si or GaAs, due to their potential application
in low-cost and high-performance electronics, such as organic light emitting diodes (OLEDs)
for flat-panel displays and solid state lighting1, organic photovoltaics (OPVs)
2, organic thin-
film transistors (OTFTs)3, and organic memory devices (OMDs)
4. While the research and
commercialization in these areas intensify5, one of the major challenges for industry is to find
new materials suitable for electrodes, primarily for semi-transparent anodes.
Up to date indium-tin oxide (ITO) has been extensively used as the de facto standard
anode for most organic and inorganic optoelectronic devices due to its high optical
transparency and moderate electrical conductivity. However, the upcoming launch of mass
produced organic electronic devices requires overcoming the shortcomings of ITO. For
example, the work function of ITO is known to vary dramatically depending on the surface
treatment method and conditions6. It has been reported that, during normal operation of the
device, various species from ITO will diffuse into the organic semiconductor layers, leading
to degradation in device performance7. The use of ITO on flexible plastic substrates is also
restricted due to the deposition and post-deposition annealing treatments, which require
temperatures in excess of 200o
C. This severely limits ITO as anode in low-cost roll-to-roll
processing on plastic substrates. In conjunction with the aforementioned disadvantages of
ITO, the limited world reserves of indium8
, combined with its ever increasing price, present
Page 17
2
an additional problem. Therefore, alternative anode materials for optoelectronics are highly
desirable.
Various metal and metal-oxide materials, and combinations thereof, have been
demonstrated to work as anodes in OLEDs9-14
. In particular, tin oxide and zinc oxide based
transparent conducting oxides (TCOs), such as fluorine doped tin oxide (FTO) and aluminum
doped zinc oxide (AZO), have been extensively studied as potential replacements for ITO.
However, tin or zinc based TCOs are typically n-type semiconductors and therefore are not
well suited as anode materials due to their low work function. For example the work function
of FTO has been reported to be only 4.4 eV15
, which is even lower than ITO. Ideally a p-type
semiconductor, which possesses high work function, should be used as anode. Many
transition metal oxides, such as NiO or Ni2O3, are p-type semiconductors and therefore
should be suitable materials as high work function anode buffer layers or as direct
replacements for the ITO anode.
1.2 Organic Light Emitting Diodes (OLEDs)
A typical organic light-emitting device includes an anode, an active light-emitting
zone comprised of one or more electroluminescent organic materials, and a cathode (Figure
1). One of the electrodes is optically transmissive while the other one is optically reflective.
The function of the anode is to inject positively charged particles referred to as holes into the
light-emitting zone, and that of the cathode is to inject electrons into the emission zone.
Under an applied electrical bias the injected holes and electrons are transported to the
emission layer (EL) by hopping from molecule-to-molecule in the hole transport layer (HTL)
and electron transport layer (ETL) respectively. Once a hole and electron reach the EL they
Page 18
3
may come into close proximity with each other, such that they feel a Coulombic attraction to
each other. This Coulombically bound electron-hole pair is referred to as an exciton. Once
formed, an exciton can defuse a finite distance before it becomes trapped on an
electroluminescent molecule and subsequently radiatively recombines to emit light. The
emitted light is reflected by the metal cathode and escapes through the aforementioned
optically transmissive anode. Because of its high optical transparency and moderate electrical
conductivity, ITO has been used exclusively as the standard anode layer despite many of its
shortcomings.
Figure 1. Typical OLED structure. NPB and Alq3 are respectively the
hole transport layer (HTL) and the electron transport layer (ETL).
Alq3:C545T is the emissive layer, where C545T serves as a fluorescent
dopant.
Although an OLED can consist of a single organic thin film, often additional organic
layers are introduced: an HTL to facilitate hole transport from the anode to the emissive layer
and to block electrons leakage from the cathode to the anode; and an ETL, which serves the
purpose of facilitating transport of electrons from the cathode into the emissive layer and
Page 19
4
blocking leakage of holes from anode to cathode. Figure 2, shows a typical band diagram of a
simple tri-layer OLED structure.
Figure 2. Band diagram of a typical OLED structure.
OLED technology is considered to deliver the next generation of light sources. Just as
the incandescent lamp quickly replaced fire as a primary source of light, OLEDs are expected
to eventually replace inefficient incandescent lamps. Currently, compact fluorescent lamps
and inorganic LEDs are starting to replace incandescent lamps. Some consider lighting based
on conventional semiconductor LED to be next generation solid-state lighting sources16
.
Others consider both LED and OLED technologies as two competing alternatives17
.
However, compact fluorescent lamps and inorganic LEDs are intermediate stopgaps at best
as they suffer from several critical disadvantages. For example, compact fluorescent lamps
use highly toxic mercury vapor. Inorganic LEDs also use many toxic heavy metal
compounds during manufacturing and are also very expensive to fabricate. OLEDs, on the
other hand, have the potential to be manufactured at extremely low cost using roll-to-roll
processing and printing techniques. OLEDs can also be fabricated using environmentally
friendly organic compounds, which are free from heavy metals. Finally, organic materials are
Page 20
5
highly efficient at emitting light and in many cases can be more efficient than their inorganic
counterparts. OLEDs therefore have enormous potential as low-cost high efficiency light
sources.
Figure 3. SONY XEL-1 OLED TV21
.
Not surprisingly, many of the advantages of OLEDs for lighting, also make them
ideal as a display technology. OLEDs can be made extremely thin - less than 1 micron in
device total thickness, allowing for slick product design in TVs and computer screens, and,
potentially reducing the weight of products such as laptops. In addition, many research
groups are developing flexible OLEDs, paving the way to products with rolled-up displays18
.
With respect to the quality of the picture, OLED technology offers more vivid colours and
faster refresh rates, which contribute to improved viewing of display products19,20
. Lastly,
OLED technology has the potential to be ecologically friendly. First of all, the amount of
material used to produce OLED devices on a mass scale is very minimal. Second, the organic
materials have the potential for recycling. And third, the OLED devices exhibit higher
Page 21
6
luminance efficiencies which may lead to lower power consumption of products with OLED
technology. The word “organic” in OLED definitely does help to market the technology as a
“green” and environmentally friendly. In fact, SONY, one of the leaders in trying to
commercialize OLED technology, has introduced recently a small OLED prototype TV,
where the product was marketed as “organic” and “green” (see Figure 3)21
.
Before OLED technology comes to every home and office in the form of light sources
or as flat panel displays, several major challenges must first be overcome. The lifetime of
OLED devices has to be increased to the order of conventional inorganic semiconductor
LEDs lifetimes, approximately a half-life of 50,000 – 100,000 hours. Also, new materials to
replace the de facto standard ITO are required. Due to the limited supply of indium the price
of ITO is too expensive for mass produced OLEDs. Also, ITO is incompatible with plastic
substrates, a necessity for low cost manufacturing using roll-to-roll processing, due to the
high annealing temperatures required to achieve adequate conductivity.
In this work, the use of a Ni/Ni2O3 bi-layer material system has been studied as an
alternative anode for OLEDs. Sheet resistance, transmission and hole injection properties
have been investigated and compared to the characteristics of pure Ni and ITO thin films.
The Ni/Ni2O3 anode exhibited superior hole injection properties compared to ITO and pure
Ni films. In addition, OLEDs with Ni/Ni2O3, pure Ni and ITO anodes have been fabricated
and tested. Devices based on Ni/Ni2O3 bi-layer materials system exhibited electrical
characteristics similar to that of ITO devices, whereas power efficiency was slightly higher
for ITO devices. Additionally, Ni/Ni2O3 anode based devices demonstrated better device
stability.
Page 22
7
1.3 Organic Photovoltaic Devices (OPVs)
Harvesting solar energy has become one of the primary research foci in recent years.
Issues associated with depletion of fossil fuels and other environmental concerns are
troubling governments and business throughout the globe. This leads to increases in funding
and research interests in developing alternative energy sources. Photovoltaic devices occupy
a major role in the alternative energy mix. Similar to OLEDs, organic photovoltaic (OPV)
technology lags in its maturity compared with conventional inorganic semiconductor solar
cell technologies. C.W. Tang reported one of the first organic solar cells in 1986 with
efficiency of solar-to-electrical power conversion at 1% based on an organic donor-acceptor
heterojunction22
. At that time, inorganic solar cells based on crystalline Si achieved
efficiencies of up to 20%23
. Now efficiencies of inorganic solar cells are reported as high as
40%24
, while OPVs reach about 5%25
. It is generally believed that OPVs must reach 10%
efficiency before their cost-to-performance ratio matches that of inorganic solar cell
technologies26
.
OPVs are constructed similar to OLEDs – two or more organic layers are sandwiched
between cathode and electrode. The electrodes are often the same thin films as used in
OLEDs: reflective Al cathode (~100 nm in thickness) and transparent ITO anode. The
operating mechanism of an OPV is the reverse process to that of an OLED. Light enters the
device through the semitransparent anode. The light is subsequently absorbed by the active
organic layers. The absorption of light excites the molecule into the excited states, resulting
in the formation of an exciton. Excitons will then diffuse towards the donor-acceptor
interface. At the interface the exciton is dissociated into an unbound hole and electron, which
Page 23
8
then drift under the built-in electric field to the electrodes, resulting in an electrical current.
This process is schematically depicted in Figure 4.
Figure 4. Operating mechanism of an OPV27
.
Similar to OLEDs there is also an interest in replacing the de facto standard ITO
anode with an alternative material. Some groups have reported the use of TCO anode
materials other than ITO as anode, such as FTO and AZO28-30
. However, similar to OLEDs
these TCOs are not optimal since they are n-type semiconductors and therefore have low
work functions. Alternatively, another avenue of OPV performance improvement that has
Page 24
9
been extensively investigated is the incorporation of a high work function buffer layer
between anode and hole transport layers. This strategy would allow for example the use of
low work function materials as anode. Shrotriya et al.31
reported the use of V2O5 and MoO3
buffer layers for polymer photovoltaic cells, improving efficiency of OPVs in comparison to
standard structures with ITO anode. As mentioned previously, nickel oxides are p-type
semiconductors with high work function. There may therefore be suitable as anode buffer
layers in OPVs. In this work, the use of two different kinds of nickel oxide thin films (i.e.,
Ni2O3 and NiOx) as buffer layers is demonstrated, in each case, the power conversion
efficiency of the OPVs is doubled using the nickel oxide buffer layer in comparison to
devices with standard bare ITO anodes.
1.4 Nickel and Nickel Oxide Materials
The selection of nickel as a precursor material occurred due to many reasons. Nickel
is a readily available material and very abundant in the world. This metal also has good
conductivity, second to only a few other metals: Cu, Au, Ag and Pt. In addition, a very good
adhesion of nickel to glass results in a very uniform surface of deposited thin films.
NiO, or Ni(II) oxide, is an insulator at room temperature with resistivity well
exceeding ~106 Ωcm, with values as high as ~10
13 Ωcm
32,33. NiO has a wide band gap of 3.6
– 4.0 eV. Below the Neel temperature of TN = 523K, NiO is type 2 anti-ferromagnetic. The
NiO has face-centered rombohedral crystal structure with lattice constants: a = 4.1 Å, α =
0 3.8′ at room temperature 34
. NiO can also exhibit small deviations from strict
stoichiometry, for example: Ni1-δ0, where 0 < δ < 5×10-3
. At δ close to the upper boundary,
the material is opaque with black/brownish colour, while with δ closer to zero the material
Page 25
10
becomes green and translucent (usually upon annealing). The resistivity can be lowered by
inclusion of Ni3+
ions either through Ni vacancies and/or interstitial oxygen in NiO
crystallites31
. The actual charge conduction mechanism in nickel oxide is still not clear. Some
argue that it could be due to thermal excitation of holes from shallow acceptor levels or due
to thermal excitation of polarons (small polaron hopping).35
Ni2O3, or Ni(III) oxide, has been reported, but has not been well characterized36
. One
of primary goals of this work is the fabrication and characterization of Ni2O3 thin films.
2 : Fabrication and characterization of nickel oxide thin films
Page 26
11
Chapter 2 : Fabrication and characterization of
nickel thin films
2.1 Review of fabrication techniques
A variety of methods exist for production of nickel oxide materials, including single
crystal fabrication methods. Most of the earlier reported methods were in the realm of
metallurgical sciences - fabrication techniques such as flame fusion, arc image furnace
floating zone, plasma torch, solar furnace melting, and arc transfer37
all require high
temperature processes. Methods like halide decomposition on MgO substrates38
and chemical
vapor transport are carried out at much lower temperatures39
. More recently, Vaidyan et al.
reported fabrication of NiO and Ni2O3 thin films by thermal annealing of vacuum deposited
thin films at 573 K – 823 K in a muffle furnace40
. Kim et al. reported fabrication of NiO thin
films via oxidation at 500 C and its use in OLEDs41
. Arof et al. reported on a less common
method of producing nickel oxide films using a spray pyrolysis technique42
. Ogura et al.
reported an electrochemical deposition method of nickel oxide thin films43
.
Overall, for application as anodes in organic optoelectronic devices, the
aforementioned methods are not suitable for fabrication of nickel oxide thin films with
nanometer thicknesses for a few reasons. First, methods that require some kind of high
temperature process are not appropriate because high temperature may damage the
underlying anode or substrate. Second, high temperature methods are not economically
feasible for mass production due to high thermal budgets. Third, most of the aforementioned
methods do not have the precision needed to produce thin films with thickness in the 1 nm –
Page 27
12
10 nm range. Finally, some of these methods produce films with a high degree of impurities.
For application in organic electronics, the purity of components used is crucial and largely
influences device performances and lifetimes.
Some fabrication methods have been proven to be effective in producing functional
nickel oxide layers. For example, Irwin et al.44
have demonstrated a Pulsed-Laser Deposition
technique (PLD) producing nickel oxide buffer layers used in improving efficiency of bulk-
heterojunction (BHJ) polymer photovoltaic devices. The PLD technique allows fabrication of
high quality thin films. However, it is quite expensive and not readily available in many
research groups.
Alternatively, there are various sputtering methods used to produce nickel oxide thin
films for use in organic electronics. In the literature there are few sputtering methods
reported. The common theme of all methods is the need to fine-tune the deposition process
by varying one or more critical parameters in order to achieve the desired extent of oxidation
of nickel. For instance, a number of different groups reported deposition of nickel oxide thin
films by sputtering of NiO targets. In case of the work of Yang et al.45
, a NiO target was RF
sputtered in the presence of oxygen gas. The control parameter was the temperature of the
substrate, which was varied between 250 C – 400 C. Depending on the substrate temperature
during deposition, nickel oxide films exhibited different atomic percentage of oxygen and
therefore different resitivities and optical properties. A different RF sputtering control
experiment was performed by Sung et al.46
, where the control parameter was
deposition/working pressure. The pressure was changed between 2.7 mTorr and 20 mTorr,
producing films with varying density, microstructure and optical characteristics. In another
example, a DC sputtering method was used by Yoo et al.47
, where the percentage of oxygen
Page 28
13
gas was varied to produce nickel oxide thin films with different microstructural and electrical
characteristics. Although all of these methods are deemed practical, there is one major
disadvantage - NiO targets are expensive (about 10 times more expensive than pure Ni metal
targets). Furthermore, sputtering NiO oxide does not necessarily produce NiO thin films with
desired extent of nickel oxidation. Virtually in all cases of sputtering NiO targets, a fine-
tuning of deposition parameters is required in order to produce nickel oxide thin films with
desired oxidation extent and electrical and optical properties.
An alternative to sputtering of NiO, nickel oxide thin films can also be produced by
sputtering pure Ni metal target with argon gas in the presence of oxygen gas as a reactive
agent. In such case, the oxidation of nickel is controlled by varying either the concentration
of oxygen, deposition pressure or rate of deposition. Often these parameters are inter-related.
For example, addition of more oxygen gas increases deposition pressure, which in turn
decreases the molecules mean free-path, increases the number of collisions and decreases
deposition rate. At lower deposition rate oxygen molecules have more time to react with
nickel atoms, and therefore, a higher oxidation state of nickel is achieved. Below are a few
reports presenting experiments, where nickel oxide thin films were produced using reactive
RF sputtering of Ni metal targets.
Lu et al.48
showed that the oxidation of nickel can be controlled by varying the
substrate temperature. In this case, the temperature of the substrate was varied between 100 C
and 400 C. Similarly to the work of Yang et al. with sputtering of NiO, the varying of the
substrate temperature produced films with wide variations in grain microstructure, resistivity
and optical transmissivity. An even simper method was used by Hong et al.49
, where a pure
Ni metal target was sputtered in the presence of oxygen gas only. In their work, a particular
Page 29
14
combination of deposition parameters was used to produce nickel oxide buffer layers used to
enhance the performance of OLEDs.
In summary, there are many variations of sputtering of NiO and Ni targets available
for fabrication of nickel oxide thin films. The use of Ni targets is a more practical choice,
because Ni targets are much less expensive than NiO targets. Fine-tuning of one or more
deposition parameters is required for both kinds of targets.
2.2 Reactive RF sputtering technique, NiOx
In order to produce thin film buffer layers for application in organic photovoltaic
devices the reactive RF sputtering method has been chosen for the reasons stated in the
previous section. A 99.999% Ni target with 2-inch diameter was sputtered with Ar gas in the
presence of oxygen. The experimental setup available for utilizing this method was designed
and constructed without ability to change substrate temperature and change target to substrate
distance. Therefore, the only control parameters available are deposition power and
deposition pressure. The deposition power is manually adjusted on the control panel of the
sputtering system. The deposition pressure is controlled by varying Ar gas concentration
using a precision mass flow controller. The sputter chamber is part of the Kurt J. Lesker
cluster tool in Prof. Lu‟s lab. The same cluster tool is also used for the fabrication of OLED
devices. The sputter chamber is depicted in Figure 5. The sputter chamber is pumped with a
cryo-pump and can achieve a base pressure of 10-8
Torr with ~ 24 hours pump down time.
Samples are loaded into the sputter chamber via the central distribution chamber of the
cluster tool.
Page 30
15
Figure 5. RF sputter system used in this work.
2.2.1 First “shotgun” trial experiment
Although the sputter system has previously been used to deposit various oxide layers,
the system was never utilized to produce nickel oxide thin films. Hence, the initial
experiment required “probing” the system at various combinations of deposition power and
pressure. As demonstrated by many previous studies, fine tuning the deposition parameters is
necessary to achieve the desired film properties. In this experiment, 150 nm nickel oxide thin
films have been fabricated and tested with respect to their optical and electrical properties.
Transmission properties were measured using a UV-Vis-NIR spectrophotometer. Sheet
resistance was measured using an in-house 4-point probe setup. Table 1 summarizes the
results of these initial test experiments.
Page 31
16
Table 1. NiOx fabrication parameters and characterization.
Run Deposition
Pressure,
mTorr
Ar flow
rate SCCM
O2 flow
rate SCCM
Power,
W
Dep. rate
(A/s)
%T At
525 nm
Sheet Resistance,
Ω/□
Control 1 1.6 5 5 200 0.65 0.29 12.375
Control 2 3.5 10 10 200 0.1 75.9 too high*
Control 3 6 15 15 200 0.4 0.96 2.25
Control 4 7.6 20 20 200 0.4 0.25 1.3725
Trial A 5.6 15 15 150 0.15 57.3 too high
Trial B 4 10 10 100 0.1 54.8 too high
*voltage was outside of the sensitivity range of the multi-meter used
The produced films showed wide differences in their properties. For example, the
Control 2 film was 75.9% transparent at 525 nm wavelength with sheet resistance
immeasurable with the standard 4-point probe setup (the sheet resistance was most likely in
the order of 10 K‟s of Ω/□), while the Control 3 thin film was only 0.96% transparent with
sheet resistance just over 20 Ω/□. Figure 6 displays optical transmission of the films
Figure 6. Transmission of NiOx films (1st deposition experiment).
Page 32
17
To further investigate the capabilities of the sputtering system and to avoid
unnecessary waste of the Ni target, the sputtering system was quickly run “dry”: without
deposition of films. Various parameters were tried out for short periods of time in order to
optimize the deposition rate as a function of RF power and gas flows. Deposition powers
were investigated in the range of 50 W to 150 W and various combinations of flow rates of
Ar and O2 gases were sampled. Table 2 summarizes this experiment‟s results. Some
combinations of parameters were then used to fabricate the actual films for electrical and
optical testing.
Table 2: Summary of sputtering parameters (dry run).
RUN Power,
W
Ar flow
rate
SCCM
O2 flow
rate
SCCM
Dep.
rate
(A/s)
Deposition
Pressure,
mTorr
1 200 5 5 0.7 1.6
2 200 10 10 0.6 3.4
3 200 15 15 0.5 5.5
4 200 20 20 0.4 7.1
5 200 25 25 0.3 8.9
6 200 30 30 0.3 14
7 200 5 10 0.8 1.9
8 200 5 20 0.8 2.2
9 200 5 30 0.8 2.4
10 200 0 30 0.6 1
11 100 10 10 0.3 4
12 120 10 10 0.4 4
13 140 10 10 0.4 4
14 160 10 10 0.5 4
15 180 10 10 0.6 3.9
16 200 10 10 0.6 3.9
17 80 10 10 0.3 3.9
18 60 10 10 0.2 3.9
19 40 10 10 0.1 3.9
Page 33
18
Table 3: Deposition of 50 nm NiOx films at 150 W constant deposition power.
Film Sheet
Resistance, Ω/□
Bulk
resistivity,
Ωcm
Conductivity,
Ω1cm
-1
%T at
525nm
Flow
rate,
SCCM
Pressure,
mTorr
Deposition
rate,
(A/s)
1 6.48E+01 1.47E-03 6.81E+02 3 5 6.5 0.59
2 4.65E+01 1.05E-03 9.51E+02 2.6 7 8.6 0.56
3 5.59E+02 1.27E-02 7.90E+01 9.3 9 10.71 0.57
4 7.75E+02 1.76E-02 5.70E+01 14.2 10 11.64 0.54
5 3.33E+08 7.54E+03 1.33E-04 80.8 11 12.94 0.56
6 3.97E+08 9.00E+03 1.11E-04 88.9 12.5 13.83 0.52
7 4.20E+08 9.51E+03 1.05E-04 87 15 17.36 0.41
8 8.08E+07 1.83E+03 5.46E-04 74.6 20 23.55 0.19
Consequently, a deposition power of 150 W was chosen for fabrication of NiOx films
for the use in OPVs, after it was found that fabrication at 50 W and 100 W produced films
with a mixture of pure nickel metal and metal oxide yielding thin films with high
conductivity and low optical transmission characteristics, while deposition at 200 W induced
aggressive oxidation yielding films with very rough surfaces.
2.2.2 Fabrication of NiOx at 150 W constant deposition power
With one deposition parameter fixed, the only variable left was the deposition
pressure, which is controlled by varying Ar and O2 flow rates. An experiment was designed
to optimize these parameters: eight nickel oxide thin films with thickness of 50 nm were
fabricated at 150 W constant deposition power and flow rates varying from 5 SCCM to 20
SCCM. The ratio of Ar:O2 flow rates was kept constant. An important observation was made
regarding the gas cylinders. In order to have meaningful control experiment, where flow rate
is a control parameter, it is important to have the same output pressure from the gas cylinder.
In this case, the output pressure on both Ar and O2 gas cylinders was kept at 50 kPa. If these
Page 34
19
pressures are not constant throughout all experiments, results become not reproducible. All
things being equal the output pressure is, for instance, on the O2 gas cylinder will affect the
concentration of O2 gas during deposition, even at the same flow rates. Nevertheless, the
results of this control experiment are presented in Table 3, while Figure 7 displays the
dependence of optical and electrical properties of deposited films on deposition pressure.
Figure 7. Conductivity and optical transmission as a function of deposition pressure.
For the actual fabrication of NiOx buffer layers the following parameters were
chosen: fabrication was performed at 150 W deposition power, 13.83 mTorr deposition
pressure with Ar to O2 flow ratio of 1:1 at 12.5 SCCM each. The rate of NiOx deposition was
0.53 Å/s. From Figure 7 it can be seen that fabrication below 12 mTorr produces under-
oxidized nickel films – low transmission and high conductivity indicate that there is a
significant presence of pure nickel metal. Analogously, fabrication above 15 mTorr produces
Page 35
20
entirely oxidized films, where the transition between metallic and metal oxide behavior lays
in the 12 mTorr – 15 mTorr region. The 13.83 mTorr point was chosen to produce buffer
layers for OPVs, because at this pressure nickel oxide films were found to be the most
transparent and least conductive. Higher optical transmission is desired in order to facilitate
incidence of light onto the active organic region, while low conductivity is beneficial to
inhibiting lateral current leakage in OPVs.
2.2.3 XPS and work function measurements of NiOx
X-ray photoelectron spectroscopy (XPS) and ultra-violet photoelectron spectroscopy
(UPS) measurements were carried out to investigate the oxidation state of nickel oxide thin
films and the work function, where the latter is an important parameter for the contact
formation with organic materials. XPS and UPS measurements were performed using a PHI
5500 Multi-Technique system with monochromated Al Kα x-rays (hν = 1486.7 eV) and He
Iα UV (hν = 21.22 eV). The nominal analysis area is 0.5 mm2, the sampling depth from the
surface is 7 - 8 nm and for „bulk‟ measurement electron take-off angle is 75 degrees. The
spectra were calibrated by referencing to the Fermi edge of sputtered Au foil. The sample
was biased at -9V with respect to ground for the UPS measurements in order to separate the
work function of the sample from that of the spectrometer.
Figure 8 shows the Ni 2p XPS core-level spectra of NiOx (green curve) and pure Ni
metal (blue curve) thin films. The binding energy of the Ni 2 p3/2 peak can be used to
determine the chemical state of Ni in the films. For metallic Ni the main feature of the Ni 2
p3/2 peak is located at a binding energy of 852.7 eV, while for the NiOx film the main feature
of the 2 p3/2 peak is relatively broad, spanning from approximately 854.0 eV to 856.5 eV.
Page 36
21
This finding indicates that the NiOx film is comprised of mixed nickel oxide species50
. Using
UPS, the work function of ITO modified with 3 nm NiOx was found to increase from 4.7 eV
to 5.2 eV. The higher work function of NiOx was also found to be quite stable, in contrast to
many of the other surface treatments of ITO that yield only metastable work function
improvements (e.g., UV ozone treatment). The higher work function of the NiOx buffer layer
is expected to significantly enhance device performance.
Figure 8. Ni 2p core-level XPS spectra of pure Ni and NiOx thin films.
2.2.4 Sputtering of NiOx at different target-to-substrate distances
Although the experimental sputtering setup does not allow for changing the substrate-
to-target distance, an experiment was performed that does exactly that. G.M. Murdoch et al.51
reported a method where AZO thin films were fabricated at 8, 12, 16 and 19 cm center-to-
Page 37
22
center distance between the target and the substrate in the same chamber that is used in this
work. The substrates were placed on a long vacuum grade steel rod the “tower” . The rod
was erected upwards, such that the position of substrates was vertical, which is rotated 90
degrees from the usual horizontal position.
The same method was used to produce NiOx films at the same center-to-center
distances between the substrate and the target. The films were fabricated with the deposition
parameters corresponding to film #6 in Table 3. Overall, the films produced were not suitable
for further investigation, because it was found that the deviation of optical and electrical
properties throughout short distances (1 cm) was too large.
In addition, the “tower” setup could not be permanently used, because it was blocking
the line of sight of the quartz thickness monitor for at least one of the target positions, and
therefore, no accurate film thickness could be determined.
However, one interesting fact was found: the varying of target-to-substrate distance
produced changes in the microstructure. Scanning electron microscopy (SEM) analysis
showed that the grain sizes of films produced at various distances were different. Figure 9
shows SEM micrographs of films produced at 8, 12 and 19 cm target-to-substrate distances.
The surface structure visible on the micrographs appears to be grains. The “spikes” seem to
have clearly defined edges and facets, indicating that the structures may be grains. It is
apparent that the grain size does not appear to increase or decrease proportionally to increase
in target-to-substrate distance. The films deposited at 12 cm appear to have much larger grain
structure. There could be many explanations for this. One factor to consider is that the angle
of incidence of molecules is different in each case of deposition geometry arrangement.
Page 38
23
Figure 9. SEM micrographs of NiOx films produced at: (top) 8 cm,
(middle) 12 cm and (bottom) 19 cm target-to-substrate distance.
Page 39
24
2.3 UV-ozone oxidation technique, Ni2O3
An alternative fabrication technique of nickel oxide thin films has been developed in
the course of this work, for which a provisional patent application was submitted. The
method involves vacuum deposition of pure nickel films followed by UV-ozone oxidation.
2.3.1 Description of fabrication and characterization methods
Both Ni2O3 anode buffer films for organic solar cells and Ni/Ni2O3 anode films for
OLEDs were fabricated in a two step process: thermal evaporation of pure nickel metal
followed by low temperature UV-ozone treatment.
Thermal evaporation was performed in a Kurt J. Lesker cluster tool using stainless
steel shadow masks to define the device structure. Ni films were deposited in the
metallization chamber using high purity (99.999%) nickel pellets from an alumina coated
molybdenum boat. The UV-ozone treatment was carried out in a chamber equipped with a
photo surface processor, model PL16-110 from Sen Lights corporation (Figure 10).
Figure 10. Fabrication of Ni2O3 thin films.
Page 40
25
Ni2O3 buffer layers for organic solar cells were produced by deposition of 3 nm thick
Ni films on top of commercially patterned ITO coated glass (25 mm × 25 mm) with a sheet
resistance at 21.5Ω/□ followed by ex situ oxidation using low temperature UV-ozone
treatment for 30 minutes. Ni/Ni2O3 anodes films for OLEDs were produced by deposition of
10 nm Ni films on top of glass substrate followed by ex situ oxidation using low temperature
UV ozone treatment for 30 minutes. The substrate layout is shown in Figure 11.
A B
C D
Figure 11. Schematic of anode pattern on glass substrate.
Corning® 1737 glass (25 mm × 25 mm) was used for devices with Ni/Ni2O3 anodes.
All substrates were ultrasonically cleaned with a standard regiment of cleaning detergent
Alconox®, acetone, methanol and de-ionized (DI) water, followed by ultraviolet (UV) ozone
treatment for 15 minutes.
Page 41
26
Sheet resistance was measured using a Keithley® 2400 source measurement unit
(SMU) in a 4-point probe configuration. Optical transmittance measurements were
performed using a Varian Carry 5000 UV-Vis-NIR spectrophotometer.
2.3.2 Characterization of Ni2O3 films
2.3.2.1 Sheet resistance measurements with a 4-point probe
Table 4 shows results of sheet resistance measurements, carried out with a 4-point
probe, of 34 different Ni2O3 thin films. In the experiment, four sets of Ni films with different
thickness (3 nm, 5 nm, 10 nm and 15 nm) were fabricated, where each set contained five
identical films. Then, samples from each set were treated with UV ozone oxidation method at
various times (0 s, 10 s, 30 s, 1 min, 5 min, 30 min, 60 min and 120 min).
Table 4: Sheet Resistance Characteristics (Ω/□).
Time 0 s 10 s 30 s 1 min 5 min 30 min 60 min 120 min
3 nm 1097 1458 1505 1635 2652 5465 11429 25007
5 nm 170 182 192 200 227 234 261 349
10 nm 42 41 40 41 44 40 48 56
15 nm 23 23 24 24 25 22 24 29
For a bottom emitting OLED structure a low sheet resistance is needed to facilitate
efficient lateral transport of charge carriers between the external contact and the light
emitting part of the device. This plays an even more significant role in solid-state lighting
applications based on OLEDs, where charge carriers must transport through large areas with
Page 42
27
orders of magnitude of tens of centimeters. An important trend was observed, the sheet
resistance increased with longer UV ozone treatment time, which suggests that a more
resistive oxide layer (thicker film) grows with longer UV ozone treatment times. Overall, 3
nm and 5 nm films, treated at all different times, exhibited sheet resistances not suitable for
functional OLEDs. Typically, the sheet resistance of the anode should be below 50 Ω/□ in
order to ensure lateral conductivity in the anode. The 10 nm and 15 nm films displayed
moderately low sheet resistances, which saturated rapidly with treatment time (almost no
change beyond 1 min). Hence, based on sheet resistance evidence alone, 10 nm and 15 nm
films are good candidates to be used as anodes in OLED.
2.3.2.2 Optical transmission measurements
Table 5 shows results of transmission analysis of 32 Ni2O3 thin films, where Ni2O3
films exhibited partial transmission with thinner films exhibiting higher transmission
coefficient.
Table 5: % Transmission at 525 nm.
Time, min 0 s 10 s 30 s 1 min 5 min 30 min 60 min 120 min
3 nm 70.3 73.3 73.7 74.0 73.2 73.5 77.6 79.4
5 nm 56.2 57.3 57.9 58.2 57.7 57.9 60.5 62.8
10 nm 32.9 34.0 33.0 34.4 33.9 33.9 35.4 37.0
15 nm 20.7 21.0 21.3 21.4 21.0 21.2 21.3 23.0
Since in OLEDs the anode layer is positioned at the bottom of the OLED, it must be
transparent to the wavelength of the OLED light output. In this work, the standard OLED
Page 43
28
structure displays green electroluminescence emission with emission peak at 525 nm. Also,
transmission increased with longer UV ozone treatment time, which can be explained by the
progressive growth of Ni oxide layer. Ni oxide, being a p-type semiconductor, is more
transparent than pure metallic Ni. The 3 nm and 5 nm films were deemed unusable in OLEDs
due to high sheet resistance characteristics, whereas 15 nm films were too opaque. This was
confirmed by experimental data where failed or poor performing OLED structures were
fabricated with the above films. The best recipe for efficient Ni based anode was
experimentally identified as 10 nm Ni films treated with UV ozone for times between 5 and
60 minutes. The OLED device performance analysis is based on 10 nm Ni thin film anodes
treated for 30 minutes with UV ozone.
Figure 12. Ni 2p XPS spectra of pure and oxidized Ni films.
Page 44
29
2.3.2.3 Chemical composition: XPS
XPS analysis was performed to the identify chemical state of Ni in the as-deposited
and UV ozone treated (oxidized) films. Figure 12 shows the Ni 2p XPS spectra of oxidized
Ni3+
(dashed line) and pure Ni0 (dotted line) thin films. For a pure Ni specimen the main
feature of its 2p3/2 peak is located at around 852.7 eV. As for the UV ozone treated film the
main feature of the 2p3/2 peak is located at 856.0 eV, indicating the presence of mostly Ni2O3
oxide.
Figure 13 ) profile of the oxidized Ni film.
Figure 13 shows a bulk profile analysis of the UV ozone treated Ni films, which was
performed to investigate the extent of oxidation. The bulk and surface spectra, measured
respectively at 75 and 45 degrees photoelectron take-off angle, are nearly identical with the
Page 45
30
bulk spectrum showing a small knee at the location of the Ni0 2p3/2 peak, suggesting that
there is a pure metallic Ni layer underneath the Ni2O3 oxide. This indicates that the UV
ozone treated films are bi-layers consisting of pure a Ni layer covered with Ni2O3.
2.3.2.4 Oxidation timeline
XPS analysis was further used to investigate the timeline of nickel oxide growth
under UV-ozone treatment. The same thickness pure nickel films were subjected to 0 s, 5 s,
10 s, 1 min and 5 min and treatment. Figure 14 shows XPS Ni 2p spectra of the films. From
the figure it is visible that within 5 min of oxidation the Ni0 peak disappears and the oxide
peak completely forms. The comparison of films treated for 5 min with films treated for
longer times did not show significant differences, indicating that within 5 min the top layer of
the films completely oxidizes within the depth sensitivity of XPS.
Figure 14. Timeline of nickel oxide growth.
Page 46
31
2.3.2.5 Surface characterization: AFM
For OLEDs Ni and Ni/N2O3 thin films exhibited very smooth morphology with mean
surface roughness (MSR) of Ni and Ni/N2O3 samples calculated by the atomic force
microscopy (AFM) software to 0.565 nm and 0.469 nm respectively (Figure 15). The films
were subjected to scotch tape pull-off test to verify this property. The removal of nickel and
nickel oxide thin films was difficult to implement using scotch tape, indicating a very good
adhesion of Ni to glass surface. Excellent adhesion and smooth morphology make Ni based
thin-film electrodes attractive for use in organic electronics. In comparison to ITO‟s mean
surface roughness of 4 – 5 nm, nickel oxide thin films are much smoother. This may be
beneficial to induce more ordered packing of organic molecules eliminating voids and
dislocations, which can act as charge traps.
Ni: MSR = 0.565 nm Ni2O3: MSR = 0.496 nm
Figure 15. AFM of Ni and Ni2O3 thin films.
3 : Ni/Ni2O3 bi-layers as anodes for OLEDs
Page 47
32
Chapter 3 : Ni/Ni2O3 bi-layers as anodes for
OLEDs
Ni/Ni2O3 bi-layer thin films have been studied as an alternative to ITO as anode for
OLEDs. The OLEDs have been fabricated with the standard structure previously used in the
group: anode/CuPc/NPB/Alq3:C545T/Alq3/cathode, where CuPc is copper phthalocyanine,
NPB is 4,4‟-bis[N-(1-napthyl)-N-phenyl-amino] biphenyl, Alq3 is tris-(8-hydroxy-
quinolinato)aluminum, and C545T is 2,3,6,7-tetrahydro-1,1,7,7,-tetramethyl-1H, 5H,11H-10-
(2-benzothiazolyl)quinolizino-[9,9a,1gh]coumarin (Appendix A). It was found that
elimination of CuPc hole injection layer produced devices with better operational
performances for nickel oxide anodes. In what follows is a comparison study between ITO
and Ni/Ni2O3 anodes for devices with structure not including the CuPc layer.
3.1 Organic Materials
The structures and physical properties of the various organic materials used in the
OLEDs are summarized below52
:
Page 48
33
3.2 OLED Fabrication
OLEDs and single carrier devices were fabricated in a Kurt J. Lesker cluster tool (see
Figure 16) using stainless steel shadow masks to define the device structure. The standard
Page 49
34
OLED device structure is as follows: anode/ NPB (45 nm)/ Alq3 doped with 1.0 wt.% C545T
(30nm)/ Alq3 (15 nm)/ LiF (1 nm)/ Al (100nm). ITO, Ni and Ni/Ni2O3 were used anodes.
Figure 16. Kurt J. Lesker cluster tool.
The standard single carrier (hole-only) device structure is as follows: anode (as above)/NPB
(500 nm)/Au (25 nm). Au was used as the cathode for its high work function (i.e., 5.1 eV) in
order to create a sufficiently large electron injection barrier to minimize the electron current
in the hole-only devices. Commercially patterned ITO coated glass (25 mm × 25 mm) with
sheet resistance less than 15 Ω/□ was used for devices with ITO anodes. Corning® 1737
glass (25 mm × 25 mm) was used for devices with Ni and Ni/Ni2O3 anodes. Substrates were
ultrasonically cleaned with a standard regiment of cleaning detergent Alconox®, acetone,
methanol and DI water, followed by UV ozone treatment for 15 minutes. The various organic
molecules (and LiF) were deposited from alumina crucibles in a dedicated organic deposition
chamber with a base pressure of ~ 10-8
Torr. The Al cathode lines (2 mm wide) were
Page 50
35
deposited orthogonally to the anode lines (1 mm wide) from a pyrolytic boron nitride
crucible, in a separate metallization chamber with a base pressure of ~ 10-8
Torr. Ni anode
lines (1 mm wide) for the Ni and Ni/Ni2O3 anodes were deposited in the same metallization
chamber using high purity (99.999%) nickel pellets from an alumina coated molybdenum
boat. Ni/Ni2O3 films were fabricated from pure Ni films by ex situ oxidation using low
temperature UV ozone treatment for 30 minutes. The UV ozone treatment was carried out in
a chamber equipped with a photo surface processor, model PL16-110 from Sen Lights
corporation. Film thicknesses were monitored using a calibrated quartz crystal microbalance.
Since the thickness of the organic layers is critical in injection/transport studies, the thickness
of the NPB layers in the single carrier devices were further verified (for each device) using
both a stylus profilometer (KLA Tencor P-16+) and capacitance-voltage (CV) measurements
(Agilent 4294A). The intersection of each cathode and anode line yields one OLED pixel.
The active area for all devices was 2 mm2. A total of four different device structures (four
different substrates) were fabricated during each fabrication run to eliminate possible run-to-
run variability caused by slight variations in process conditions. Figure 17 shows an
operating OLED with Ni/Ni2O3 anode.
Figure 17. Operating OLED Ni/Ni2O3 anode.
Page 51
36
IV characteristics were measured using an HP4140B picoammeter in ambient air.
Luminance measurements were taken using a Minolta LS-110 Luminance meter.
Temperature dependent electrical characterization was conducted in a closed-loop He
cryostat (see Figure 18) with a base pressure of ~ 10-7
Torr at room temperature. Sheet
resistance was measured using a Keithley® 2400 source measurement unit (SMU) in a 4-
point probe configuration. Optical transmittance measurements were performed using a
Varian Carry 5000 UV-Vis-NIR spectrophotometer.
Figure 18. Vacuum cryostat.
3.3 OLED Performance
The power efficiency (PE) is a benchmark characteristic of OLED performance. It
indicates the amount of power consumption of the device for unit of light output produced,
lm/W. Power efficiency is an important consideration in OLED technology, because power
efficient devices are more attractive for mass commercialization. Three different films were
used as anodes with identical OLED structure. Device A was composed of a standard OLED
Page 52
37
architecture with ITO anode. The ITO films used in this experiment exhibited sheet
resistance of ~20 Ω/□ and normal transmission of about 90% at 525 nm. Device B was
constructed identical to A but the ITO anode was replaced with vacuum deposited 10 nm
pure Ni film. Device C was constructed with 10 nm Ni/Ni2O3 anode film. Figure 19
illustrates the relative PE for the three devices. The ITO and Ni/Ni2O3 anode based devices
exhibited comparable PE. Device B exhibited very poor PE, which can be explained by the
inferior charge carrier injection properties of pure Ni.
Figure 19. Power efficiency vs. luminance characteristics of ITO and Ni/Ni2O3 based OLEDs.
Figure 20 illustrates luminance vs. voltage characteristics of the three different anode
structures. The driving voltage is largely dependent on the hole-injecting properties of an
anode. The Ni/Ni2O3 anodes exhibit nearly identical luminance vs. driving voltage behavior.
Whereas, it has been previously verified that pure Ni is a poor hole-injecting material, which
is consistent with the high driving voltage and low luminance.
Page 53
38
Figure 20. Luminance vs. voltage characteristics of ITO and Ni/Ni2O3 based OLEDs.
Another important characteristic of OLED is its operational stability. One of the
typical ways to test this is to measure luminance output of a device under constant voltage.
The measurement is performed over a prolonged period of time (e.g., 10 mins, 2 hrs, etc),
where the luminance will decrease over time due to various factors: degradation of organic
material, oxidation and other. Two devices were subjected to the above analysis: ITO and
Ni/Ni2O3 based devices. Devices were tested at 9 V applied forward bias and the luminance
was recorded at 10 second intervals.
Figure 21 shows the normalized emission intensity of the two devices over the period
of 1.5 hr. The Ni/Ni2O3 based device exhibited better device stability with luminance
decreasing by about 11%. ITO based device exhibited decrease of luminance by about 16%
during the same period of time. This indicates that Ni/Ni2O3 OLEDs may demonstrate longer
Page 54
39
life time than the ones based on ITO. The reasons for poorer performance of ITO OLEDs
may be various. For example, as already mentioned before, various chemical species diffuse
from ITO into the organic layers degrading the device. Both devices were tested without
encapsulation and therefore degraded very fast.
Figure 21. Normalized emission intensities of ITO and Ni/Ni2O3 based
anodes: constant current applied for 1.5 hrs.
3.4 Charge carrier injection
In the standard OLED structure the injection of electrons from the cathode into the
ETL is optimized with LiF/Al cathode system. The best operational efficiencies are achieved
by balancing the injection of these electrons with the same amount of holes injected from the
anode. If there is an imbalance of charges on either side the efficiency will decrease
Page 55
40
dramatically. Since LiF/Al is a good electron injector the anode must also be a good hole
injector. ITO is known for its good hole injection characteristics already53
, whereas Ni and
Ni2O3 material systems had not been extensively evaluated. To investigate their hole
injecting properties a series of experiments have been performed.
Figure 22. I-V characteristics of single carrier hole-only devices with
ITO, Ni and Ni/Ni2O3 anodes.
Organic semiconductors are typically disordered van der Waal solids with highly
localized states. As a result organic semiconductors contain almost no intrinsic charge
carriers, and hence all of the holes and electrons must be injected from the anode and cathode
respectively54
. In order to reduce the driving voltage and increase the efficiency in organic
optoelectronic devices excellent injection contacts are required at the electrodes. In the case
Page 56
41
of OLEDs, Ohmic contacts, which eliminate the junction resistance, are highly desirable.
Single carrier hole-only devices were studied in order to demonstrate the superior hole
injection from Ni/Ni2O3 bi-layer into α-NPD, the most commonly used hole transport
molecule.
Figure 23. Temperature dependent IV characteristics of single carrier
hole-only devices with ITO, Ni and Ni/Ni2O3.
Figure 22 shows the IV characteristics of single carrier hole-only devices with ITO,
Ni and Ni/Ni2O3 anodes. Since a high work function metal (i.e., Au) is used as the cathode,
electron injection is effectively blocked, leading to total current consisting of mostly holes.
Hence, an increase in current density, at a given voltage, is representative of the hole
injection barrier at the anode; higher current density corresponds to a lower hole injection
barrier. It is found that the hole current, over a large voltage range, is ~ 3 orders of magnitude
higher for Ni/Ni2O3 than for ITO. Remarkably, the Ni/Ni2O3 anode also has ~ 9 orders of
Page 57
42
magnitude higher current density than pure Ni. Clearly, the thin surface oxide layer of Ni2O3
dramatically enhances hole injection into NPD; a significant benefit for high performance
OLEDs.
Figure 23 shows the temperature dependent IV characteristics of single carrier hole-
only devices with ITO, Ni and Ni/Ni2O3 anodes at a given voltage of 10 V. All of the anode
structures exhibited higher hole current densities at higher temperature, following an
Arrhenius type activation energy with respect to the hole injection current. However, the hole
injection current in devices with Ni anodes appears to be nearly independent of temperature.
This finding is consistent with a small density of deep interfacial trap states, which dominate
the injection current for high barrier55
.
3.5 UPS and work function measurements
From the IV characteristics of the single carrier devices it is evident that Ni/Ni2O3
exhibits superior hole injection compared to both pure Ni and ITO anodes. However, the
origins of the reduced hole injection barrier are still unclear. UPS analysis was performed to
characterize the valance band features and work function of the different anode materials.
Figure 24 shows the He I valence band spectra for Ni (native oxide) and Ni2O3 (UV
ozone) samples. The secondary electron cutoff was found to shift to lower binding energy for
the Ni2O3 samples, which is indicative of a higher work function. The measured work
functions were 4.4 eV and 5.2 eV for the Ni and Ni2O3 samples respectively. Following the
Schottky-Mott rule (vacuum level alignment) the estimated difference in hole inject barrier
between Ni and Ni2O3 is ~ 0.8eV. However, it is well known that the Schottky-Mott limit is
Page 58
43
rarely observed at real organic semiconductor contacts. Often, a large dipole is formed
between organic molecules and various metal and metal-oxide surfaces.
Figure 24. He I valence band spectra for Ni and Ni2O3 thin films.
Previously, it was shown, by Helander et al., how this effect can be accounted for by
interface dipole theory56
. The theory describes the formation of an interfacial dipole due to
the charging of intrinsic interface states, which tends to drive the band alignment towards a
zero net dipole charge. A charge neutrality level CNL is defined as the point at which the
interface states are equally donor- and acceptor-like. As a result, the charge neutrality level of
the organic and the Fermi level of the anode will tend to align, the difference being
accommodated by an interface dipole. Based on this theory, the dipole between the Ni/Ni2O3
anode and the adsorbed NPB molecules is given by:
1 m CNLS (1)
Page 59
44
where is the interfacial dipole, S is the interface slope parameter, m is the work function
of the pristine anode and CNL is the charge neutrality level of the interface states. S and
CNL are intrinsic properties of the organic semiconductor and are taken as 0.71 and 4.2
respectively57
. Using Eq. (1) the magnitude of the interface dipole between Ni/Ni2O3 and
NPB is estimated to be ~ 0.3eV.
4 : Nickel oxide buffer layers for OPVs
Page 60
45
Chapter 4 : Nickel oxide buffer layers for OPVs
Two different nickel oxide thin films have been used to demonstrate improvement of
OPV efficiencies. First, Ni2O3 buffers were fabricated using the same UV-ozone oxidation
technique as in the OLED work. The only difference is that the thickness of the buffer layers
was between 1 – 3 nm. Thin films with such thicknesses reach complete oxidation
throughout the film, based on the observed extent of oxidation of nickel films (see chapter 2).
Second, RF reactive magnetron sputtering method was used to produce NiOx buffer layers.
“NiOx” refers to the mixture of different nickel oxide species: Ni II and Ni III oxides.
Figure 25. Schematic of BHJ solar cell device.
In this work, the solar cells under investigation had BHJ architecture based on
P3HT:PCBM materials. Where P3HT donor is poly(3-hexylthiophene), and PCBM acceptor
is phenyl-C61-butyric acid methyl ester. The idea of BHJ is in mixing the donor and the
acceptor materials in order to increase the active surface of solar cell device. In contrast, in
Page 61
46
the case of planar OPVs, where the active region is only located on the interface between the
donor and the acceptor layers. Figure 25 shows the typical BHJ device structure.
4.1 Organic Materials
The structures and physical properties of the various organic materials used in the
OPVs are summarized below52
:
4.2 Fabrication of OPVs
Organic solar cells were fabricated with the following structure: ITO / buffer (3 nm) /
PCBM:P3HT (70 nm)/ LiF (1 nm) /Al cathode (100 nm).
Page 62
47
ITO anodes came from commercially ITO coated glass (25 mm × 25 mm) with sheet
resistance of 21.5 Ω/□. Substrates were ultrasonically cleaned with a standard regiment of
cleaning detergent Alconox®, acetone, methanol and DI water, followed by UV ozone
treatment for 15 minutes.
PCBM (17 mg/mL):P3HT (17 mg/mL) solution was prepared in 1,2-dichlorobenzene
in nitrogen-filled glove box, then stirred in the dark overnight >8 hours at 50 C. A solution
layer was spin-coated on treated substrate at 500 rpm for 9 s, then at 1100 rpm for 45 s. The
polymer layer was dried in dark over 24 hours at room temperature in glove box. The film
thickness was measured to be 70 nm. LiF was deposited from alumina crucibles in a
dedicated organic deposition chamber with a base pressure of ~ 10-8
Torr. The Al cathodes (2
mm diameter) were deposited from a pyrolytic boron nitride crucible, in a separate
metallization chamber with a base pressure of ~ 10-8
Torr.
Organic solar cell current density versus voltage (JV) characteristics were measured
using a Keithley® 6430 sub-femptoamp meter under 100 mW/cm2 simulated AM1.5G solar
illumination. The active area was 4.45 mm2 for all devices.
4.3 OPVs with NiOx buffer layers
The performance of any new material is ultimately judged on its effect on
performance in real devices. Figure 26 shows the JV characteristics from PCBM:P3HT solar
cells with ITO and ITO modified with a 3 nm thick NiOx buffer layer anodes. The power
conversion efficiency of the standard devices was increased by ~ 60% with addition of the
NiOx buffer layer. The higher power conversion efficiency for NiOx is a result of a much
Page 63
48
higher short circuit current (Jsc) and also slightly increased open circuit voltage (VOC) and fill
factor (FF). Table 6 summarizes the device performance parameters of the devices.
Figure 26. Solar cells JV characteristics (NiOx buffer).
Table 6: Solar cell characteristics (NiOx buffer).
Device Anode VOC
(V)
JSC
(mA/cm2)
FF ηP
(%)
RS
(Ωcm2)
Rsh
(Ωcm2)
A ITO 0.54 5.36 0.60 1.74 8.53 516
B ITO/NiOx 0.57 7.58 0.64 2.77 10.5 886
In BHJ OPVs, VOC is dependent on the difference between the highest occupied
molecular orbital (HOMO) level of the donor and the lowest unoccupied molecular orbital
(LUMO) level of the accepter58,59
(HOMO and LUMO are analogous to respectively valence
Page 64
49
and conduction bands in inorganic semiconductor theory). However, in this case, the organic
material is the same in both cases and should not contribute to any change in VOC. It has been
shown that VOC can be lowered by charge loss at the electrodes60
. One way to minimize such
losses is to provide for a better energy matching between the HOMO level and the anode
work function at the anode/organic interface61,62
The work function of ITO modified with
NiOx was found to be 0.5 eV higher than that of bare ITO, providing better energy-level
matching between the HOMO of P3HT and the work function of the buffered solar cell
anode. UPS measurements were performed: ITO was found to have work function of 4.7 eV
and ITO/NiOx exhibited work function of 5.2 eV. Therefore, the increase in VOC can be
attributed to the higher work function of NiOx at 5.2 eV. All work function measurements
were done by the fellow group member Mark Greiner.
From Figure 26, the slope of the IV curves to the left of the maximum power point is
smaller for the NiOx buffered devices, indicating a larger shunt resistance (Rsh); the slope of
the IV curves to the right of the maximum power point is also higher for buffered cells
indicating a smaller series resistance (Rs). Both, larger Rsh and smaller Rs contribute to a
higher FF. Rs is a parasitic resistance resulting from the resistance of the organic material,
contacts and electrodes to current flow. For NiOx buffered ITO, Rs is slightly larger than in
the case of bare ITO (10.5 Ωcm2 and 8.53 Ωcm
2 respectively). This is expected since the
NiOx buffer layer is an insulating material and adds resistance to current flow.
On the other hand, Rsh is indicative of current leakage in the device around the edges
of the device and between contacts. Effectively, lower leakage results in higher Rsh. For the
NiOx buffered ITO, Rsh was higher than for the bare ITO standard (886 Ωcm2 and 516 Ωcm
2
Page 65
50
respectively). This can be explained by the effect of the addition of the NiOx layer. The layer
makes the surface smoother, which reduces the chance of charge carriers to flow in alternate
paths. Another possible explanation of lower Rsh in the NiOx buffered devices may be the
effect of NiOx on the packing of the active layer near the interface. It has been reported that
various buffer layers inserted into organic solar cells induced different packing of organic
molecules resulting in decreased current leakage63
. AFM analysis of ITO and ITO/NiOx
surfaces showed that NiOx buffer decreases mean-surface roughness by about 50%, allowing
for better packing of organic material on the anode surface. ITO surface roughness was
measured at 5.6 nm RMS and ITO/NiOx exhibited roughness of 3.0 nm RMS.
The NiOx buffer layers enhance the performance of OPVs in comparison to standard
ITO-only devices. The insertion of a 3 nm thick buffer layer was found to increase the power
conversion efficiency by ~60% in comparison to standard ITO devices. XPS analysis showed
that the NiOx films consisted of a mixture of various nickel oxide species (i.e., Ni(II) and
Ni(III) oxidation states). AFM analysis demonstrated that the NiOx produces a much
smoother surface in comparison to the bare ITO surface, which in turn contributed to better
device performance.
4.4 OPVs with Ni2O3 Buffer layers
Solar cells structures were fabricated and compared with respect to their device
performance: anode(A and B) / PCBM:P3HT / LiF / Al. Structure A was fabricated with ITO
anode. Structure B incorporated nickel oxide buffer layer, ITO/ Ni2O3. Figure 27 shows JV
curves for the two devices. Table 7 summarizes device performance measures for two
different structures.
Page 66
51
Figure 27. Solar cell JV characteristics (Ni2O3 buffer).
Table 7: Solar cell characteristics (Ni2O3 buffer).
Device
Anode
VOC,
volts
JSC,
mA/cm2
FF
ηP
(%)
RS
(Ωcm2)
Rsh
(Ωcm2)
A ITO 0.55 5.68 0.41 1.29 23.6 214
B ITO/Ni2O3 0.50 6.90 0.69 2.39 6.6 1126
The power conversion efficiency of ITO devices was increased after addition of Ni2O3 buffer
layer by ~90%. This increase is a result of higher JSC and larger FF achieved with Ni2O3
buffer layer. Similarly to the case with NiOx buffer layers the shunt resistance was higher
than for ITO device. However, unlike in case of NiOx, devices with Ni2O3 buffer layers had
higher series resistance.
Page 67
52
4.5 Electron Blocking Action
In planar OPVs, when an electron-hole pair is created, both charge carries are
separated and directed towards the electrodes. If electrons reach the anode, they can
recombine with holes, canceling both charge carriers. This, in turn, decreases the solar cell
power conversion efficiency. The donor layer is typically selected to have LUMO level much
lower than the LUMO level of the acceptor layer. This is needed in order to prevent electron
transport towards the anode (see Figure 28).
Figure 28. A band diagram schematic of a typical planar OPV
structure. Because the LUMO level of the donor is much lower than the
HOMO level of the acceptor, the flow of electrons towards the anode is
blocked.
In the case of BHJ solar cells, the current leakage to the anode is unavoidable. The
acceptor material is distributed though the whole device, from one electrode to another. This
makes it easy for electrons to flow towards the anode.
Page 68
53
An addition of a buffer layer creates a barrier for electrons. Nickel oxide is a known
p-type semiconductor with wide band gap (around 3.5 eV). The conduction band of nickel
oxide is approximately around 1.8 eV. In comparison, the LUMO level of PCBM is at 4.0 eV
(Irwin et al.). Due to the high barrier it is not energetically favorable for electrons to flow
though the nickel buffer layer and reach the anode (see Figure 29). This electron blocking
action is the main factor in improvement of BHJ solar cell performance when nickel oxide
buffer layers are added. The fact that the Jsc current is increased when either of the two
different nickel oxide layers are added supports the electron blocking action theory.
Figure 29. Band Diagram for OPVs with NiOx or Ni2O3 buffer layers.
The flow of electrons towards the anode is inhibited by the nickel oxide
buffer layer, which has conduction band much lower than the LUMO
of the acceptor.
5 Conclusions
Page 69
54
Chapter 5 : Conclusions
5.1 OLEDs and Ni2O3
An attempt to substitute ITO with Ni/Ni2O3 as an anode for OLEDs has produced
devices with comparable operational performance. Nickel oxide based devices displayed the
same output and better device stability in comparison to standard ITO devices. The power
efficiency, however, was still slightly higher for ITO devices, despite the fact that nickel
oxide anode displayed better injection of charger carriers. This is explained by relatively low
transmission of the Ni/Ni2O3 bi-layer anode (~34%), which blocks more than 60% of the
light produced inside the device, between the electrodes.
The attempts to increase transmission by making the anodes thinner (e.g., 9 nm – 5
nm) did not yield positive results, because thinner anodes exhibited higher sheet resistance,
which made lateral conductivity of the anodes difficult. Therefore, because changing the
nature of the anode material will always result in a tradeoff between optical transmission and
conductivity, a different approach is needed.
In any case, in order for nickel oxide OLED anode technology to have a real chance
to replace ITO in mass production of OLED products the hurdle of low optical transmission
has to be overcome. Before that happens, all advantages of nickel oxide technology, such as
ease of processing and price, do not outweigh higher efficiencies possible with standard ITO
anodes.
Page 70
55
5.2 OPVs and nickel oxide buffer layers
The experiments with both kinds of nickel oxide buffer layers showed that the device
performances can be increased almost 2-fold in comparison to standard ITO-only devices.
The two fabrication methods were produced as alternatives to earlier reported PLD
fabrication technique. The overall efficiencies of buffered cells at 2.39% and 2.77% were still
not at the same level to compete with top OPVs efficiencies reported in the literature
devices59
. At the moment of the experiments, the polymer layer fabrication was just
introduced in the group and was not optimized. It is expected that once the fabrication of
polymer layer and devices optics are optimized, OPVs base efficiencies may be increased to
2 - 3%. Furthermore, incorporation of nickel buffer layers may yield efficiencies above 5%
level.
6 Future work
Page 71
56
Chapter 6 : Future work
6.1 OLEDs
In order to couple out the light produced inside an OLED a few methods may be
explored. First, an introduction of some kind of photonic crystal between the anode and the
substrate may help induce the desired effect. When light is absorbed by metallic films, a
portion of the energy is converted into surface plasmon modes, which then dissipate in the
film via thermal vibrations. A photonic crystal can help to capture these surface plasmon
modes and convert them into light65
. Hence, more useful light will be produced and emitted
by an OLED device. Another method that can be explored is construction of a MIM (metal-
insulator-metal)66
, where one of the metal layers is the nickel based anode. The problem with
both of the above methods is that in each case extra material or layers have to go between the
main device and the glass substrate. This introduces new problems with organic molecule
packing. For example, the anode would have to be deposited on something other than the
glass substrate. This in turn will produce a surface for organic deposition, which is different
than the surface produced by nickel thin film lying directly on top of the glass substrate.
6.2 OPVs
It has been shown in this work that nickel oxide buffer layers increase the power
efficiencies of OPVs. Future work may consist of producing nickel buffer layers with
different deposition parameters (e.g., different deposition power for sputtered thin films). For
the sputtering fabrication method, NiO target may be tried instead of the pure Ni target that
Page 72
57
was used in this work. Also, further optimization of the polymer active layer is needed in
order to show comparable results to literature.
Page 73
58
References
1 F. So, J. Kido and P. Burrows, MRS Bull. 33, 663 (2008)
2 E. Shaheen, D.S. Ginle, G.E. Jabbour, MRS Bull. 30, 10 (2005)
3 P. Peumans, S. Uchida, S.R. Forrest, Nature 425, 158 (2003)
4 H. Kim, K.S. Yook, J.Y. Lee, J. Jang, Appl. Phys. Lett. 93, 053306 (2008)
5 R.F. Service, Science 310, 1762 (2005)
6 H. Y. Yu, X. D. Feng, D. Grozea, Z. H. Lu, R. N. S. Sodhi, A-M. Hor, H. Aziz, Appl.
Phys. Lett. 78, 2595 (2001)
7 P. Melpignano, A. Baron-Toaldo, V. Biondo, S. Priante, R. Zamboni, M. Murgia, S.
Caria, L. Gregoratti, A. Barinov, M. Kiskinova, Appl. Phys. Lett. 86, 041105 (2005)
8 J.D. Jorgenson, M.W. George, US Geol. Surv., Open File Report 2004-1300, Indium
(2004)
9 K. Tamano, D.C. Cho, T. Mori, T. Mizutani, M. Sugiyama, Thin Sol. Films 438-439,
182 (2003)
10 C. Qiu, H. Peng, H. Chen, Z. Xie, M. Wong, H.S. Kwok, IEEE Trans. on Elec. Dev. 51,
1207 (2004)
11 C.J. Yang, C.L. Lin, C.C. Wu, Y.H. Yeh, C.C. Cheng, Y.H. Kuo, T.H. Chen, Appl.
Phys. Lett. 87, 143507 (2005)
12 I.M. Chan, T.Y. Hsu, F.C. Hong, Appl. Phys. Lett. 81, 1899 (2002)
13 H.M. Zhang, W.C.H. Choy, J. Phys. D: Appl. Phys. 41, 062003 (2008)
Page 74
59
14 G.B. Murdoch, M. Greiner, M.G. Helander, Z.B. Wang, Z.H. Lu, Appl. Phys. Lett. 93,
083309 (2008)
15 A. Andersson, N. Johansson, P. Bröms, N. Yu, D. Lupo, and W. R. Salaneck, Adv.
Mater. 10, 859 (1998)
16 E.F. Schubert, J.K. Kim, Science 308, 1274 (2005)
17 R.F. Service, Science 310, 1762 (2005)
18 M. Mizukami, N. Hirohata, T.Iseki, K. Ohtawara, T. Tada, S. Yagyu, T. Abe, T .
Suzuki, Y. Fujisaki, Y. Inoue, S. Tokito, T. Kurita, IEEE Elec. Dev. Lett. 27, 249
(2006)
19 B. Geffroy, P. le Roy, C. Prat, Polym. Int. 55, 572 (2006)
20 J.K. Borchardt, Materials Today September, 43 (2004)
21 http://www.sonystyle.com, accessed September 23 2009
22 C.W. Tang, Appl. Phys. Lett. 48, 183 (1986)
23 M.A. Green, Prog. Photovolt: Red. Appl. 17, 183 (2009)
24 R.R. King, D.C. Law, K.M. Edmondson, C.M. Fetzer, G.S. Kinsey, H. Yoon, R.A.
Sherif, N.H. Karam, Appl. Phys. Lett. 90, 183516 (2007)
25 J. Xue, B. P. Rand, S. Uchida, S. R. Forrest, Appl. Phys. Lett. 85, 5757 (2004)
26 G. Dennler, M. C. Scharber, C. J. Brabec, Adv. Mater. 21, 1323 (2009)
27 C.J. Brabec, Solar Energy Materials & Solar Cells 83, 273 (2004)
Page 75
60
28 V. Bosle, J.T. Prater, F.Yang, D.Burk, S.R. Forrest, J. Narayan J. Appl. Phys. 90,
5661(2001)
29 J.C. Bernede, Y. Berredjem, L. Cattin, M. Morsli, Appl. Phys. Lett. 92, 083304 (2008)
30 K. Schulze, B. Maening, Y. Tomita, C. May, J. Hupkes, E. Brier, E. Reinold,
P.Bauerle, K . Leo, Appl. Phys. Lett. 91, 073521 (2207)
31 V. Shrotriya, G. Li, Y. Yao, C.W. Chu, Y. Yang, Appl. Phys. Lett. 88, 073508 (2006)
32 J.E. Keem, J.M. Honig, “Selected Electrical and Thermal Properties of Undoped Nickel
Oxide”, CINDAS REPORT 52, US Department of Defense, August 1978
33 E. Antolini, J. Mater. Sci. 27, 3335 (1992)
34 H. Hosoi, K. Sueoka, K. Hayakawa, K. Mukasa, Appl. Surf. Sci. 157 218 (2000)
35 P. Lunkenheimer, A. Loidl, C. R. Ottermann, K. Bange, Phys.Rev. B 44, 5927 (1991)
36 N. E. Greenwood, and A. Earnshaw, Chemistry of the Elements (2nd ed.), Oxford:
Butterworth-Heinemann (1997)
37 J.E. Keem, J.M. Honig, “Selected Electrical and Thermal Properties of Undoped Nickel
Oxide”, CINDAS REPORT 52, US Department of Defense, August 1 8
38 R.E. Cech, E.I. Allessandrini, Trans. ASM 51, 56 (1959)
39 K. Kurosawa, S. Saito, T. Takemoto, Japanese J. App. Phys. 11, 1230 (1972); 14, 887
(1975)
40 B. Sasi, K.G. Gopchandran, P.K. Manoj, P. Koshy, P.P. Rao, V.K. Vaidayn, Vaccum
68, 149 (2003)
Page 76
61
41 H.C. Im, D.C. Choo, T.W. Kim, J.H. Kim, J.H. Seo, Y.K. Kim, Thin Solid Films 515,
5099 (2007)
42 P. Pusparajan, S. Radhakrishna, A.K. Arof, Journal of Materials Science 32, 3001
(1997)
43 K. Nakaoka, J. Ueyama, K. Ogura, J. Electroanal. Chem. 571, 93 (2004)
44 M.D. Irwin, D,B, Buchholz, A.W. Hains, R.P.H. Chang, T.J. Marks, PNAS 195, 2783
(2008)
45 Y.M. Lu, W.S. Hwang, J.S. Yang, Surface and Coatings Technology 155, 231 (2002)
46 Kwang-soon Ahn, Yoon-Chae Nah, Young-Eun Sung, Appl. Phys. Lett. 93, 1268
(2002)
47 Myoungjae Lee, Sunae Seo, David Seo, Eunju Jeong, I.K. Yoo, Integrat. Ferroelect.
68, 19 (2004)
48 Y.M. Lu, W.S. Hwang, J.S. Yang, H.C. Chuang, Thin Solid Films 420-421, 54 (2002)
49 I-Min Chan, Tsung-Yi Hsu and Franklin C. Hong, Appl. Phys. Lett. 81, 1899 (2002)
50 J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray
Photoelectron Spectroscopy (Physical Electronics, Eden Prairie, MN, 1995)
51 G.B. Murdoch, S. Hinds, E.H. Sargent, S.W. Tsang, L. Mordoukhovski, Z.H. Lu, Appl.
Phys. Lett. 94, 213301 (2009)
52 Lumtec Catalog 2009-2010, Luminescence Technology Corp. (2009)
53 M.G. Mason, L.S. Hung, C.W. Tang, S.T. Lee, K.W. Wong, M. Wang, J . Appl .Phys.
86, 1688 (1999)
Page 77
62
54 J. C. Scott,G. G. Malliaras, Chem. Phys. Lett. 299, 115 (1999)
55 Z. B. Wang, M. G. Helander, S. W. Tsang, Z. H. Lu, Phys. Rev. B. 78, 193303 (2008)
56 M. G. Helander, Z. B. Wang, J, Qiu, Z. H. Lu, App. Phys. Lett. 93, 193319 (2008)
57 S. Günes, Chem. Rev. 107, 1324 (2007)
58 C.J. Brabec, A. Cravino, D. Meissner, N.S. Saricifti, T.Fromherz M. Minse, L. sanchez,
J.C. Hummelen, Adv. Mater. 11, 374 (2201)
59 M. Scharber, D. Muhlbacher, M . Koppe, P. Denk, C. Waldauf, A.J. Heeger, C. Brabec.
Adv. Mater. 18, 789 (2006)
60 G.G. Mallairas, J.R. Salem, P.J. Brock, J.C. Scott, J. Appl. Phys. 84, 1583 (1998)
61 L. Liu, Y. Shi, Y. Yang, Adv. Funct. Mater. 11, 420 (2001)
62 C.C. Wu, C.I. Wu, J.C. Strum, A. Kahn, Appl. Phys. Lett. 70, 1348 (1997)
63 J.C. Bernede, Y. berredjem, L. Cattin, M. Morsli, Appl. Phys. Lett. 92, 083304 (2008)
64 J.Y. Kim, K. Lee, N.E. Coates, D. Moses, T.Q Nguyen, M. Dante, A.J. Heeger, Science
317, 222 (2007)
65 C. Genet, T.W. Ebbesen, Nature 445, 39 (2007)
66 J. Feng, T. Okamoto, J. Simonen, S. Kawata, Appl. Phys. Lett. 90, 081106 (2007)
Page 78
63
Appendix A: OLED with CuPc Layer
In this experiment, ITO, Ni and Ni/Ni2O3 anodes were used to fabricate OLEDs with
the standard device structure, incorporating the CuPc layer (see Chapter 3). The results are
presented below. Figure 30 shows power efficiency vs. luminance and luminance vs. voltage
relationships:
Figure 30. a) Power efficiency vs. luminance characteristics for ITO, Ni
and Ni/Ni2O3 based OLEDs with CuPc. b) Luminance vs. voltage for
ITO, Ni and Ni/Ni2O3 based OLEDs with CuPc.
The devices made with Ni/ Ni2O3 anodes exhibited power efficiency equal to 60% of
the efficiency of ITO based devices. Consequently, the same experiment was performed
where the device structure did not contain CuPc layer. In this case, Ni/ Ni2O3 anodes
exhibited power efficiency equal to 80% of the efficiency of ITO devices (see Chapter 3).
Page 79
64
Appendix B: Contributions to Research and Development
Articles published in refereed journals
[1] G.B. Murdoch, S. Hinds, E.H. Sargent, S.W. Tsang, L. Mordoukhovski, and Z.H.
Lu, "Aluminum doped zinc oxide for organic photovoltaics", Appl. Phys. Lett. 94,
213301 (2009).
[2] M. G. Helander, Z. B. Wang, L. Mordoukhovski, and Z. H. Lu, “Comparison of
Alq3/alkali-metal fluoride/Al cathodes for organic electroluminescent devices”, J.
Appl. Phy. 104, 094510 (2008).
Patents and copyrights
[1] L. Mordoukhovski, D. Grozea, and Z.H. Lu, “Low Temperature Fabrication Method
of NICKEL(III) OXIDE and Applications Thereof”, US Provisional Application, 459-002-P
(2008).