AN ABSTRACT OF THE THESIS OF
Meena Suhanya Rajachidambaram for the degree of Master of Science in Chemical
Engineering presented on December 22, 2011.
Title : Investigation of Oxide Semiconductor Based Thin Films: Deposition,
Characterization, Functionalization, and Electronic Applications.
ABSTRACT APPROVED:
Gregory S. Herman
ABSTRACT
Nanostructured ZnO films were obtained via thermal oxidation of thin films
formed with metallic Zn-nanoparticle dispersions. Commercial zinc nanoparticles used
for this work were characterized by microscopic and thermal analysis methods to analyze
the Zn-ZnO core shell structure, surface morphology and oxidation characteristics. These
dispersions were spin-coated on SiO2/Si substrates and then annealed in air between 100
and 600 °C. Significant nanostructural changes were observed for the resulting films,
particularly those from larger Zn nanoparticles. These nanostructures, including
nanoneedles and nanorods, were likely formed due to fracturing of ZnO outer shell due to
differential thermal expansion between the Zn core and the ZnO shell. At temperatures
above 227 °C, the metallic Zn has a high vapor pressure leading to high mass transport
through these defects. Ultimately the Zn vapor rapidly oxidizes in air to form the ZnO
nanostructures. We have found that the resulting films annealed above 400 °C had high
electrical resistivity. The zinc nanoparticles were incorporated into zinc indium oxide
solution and spin-coated to form thin film transistor (TFT) test structures to evaluate the
potential of forming nanostructured field effect sensors using simple solution processing.
The functionalization of zinc tin oxide (ZTO) films with self-assembled
monolayers (SAMs) of n-hexylphosphonic acid (n-HPA) was investigated. The n-HPA
modified ZTO surfaces were characterized using contact angle measurement, x-ray
photoelectron spectroscopy (XPS) and electrical measurements. High contact angles were
obtained suggesting high surface coverage of n-HPA on the ZTO films, which was also
confirmed using XPS. The impact of n-HPA functionalization on the stability of ZTO
TFTs was investigated. The n-HPA functionalized ZTO TFTs were either measured
directly after drying or after post-annealing at 140 °C for 48 hours in flowing nitrogen.
Their electrical characteristics were compared with that of non-functionalized ZTO
reference TFTs fabricated using identical conditions. We found that the nonfunctionalized devices had a significant turn-on voltage (VON) shift of ~0.9 V and ~1.5V
for the non-annealed and the post-annealed conditions under positive gate bias stress for
10,000 seconds. The n-HPA modified devices showed very minimal shift in VON (0.1 V),
regardless of post-thermal treatment. The VON instabilities were attributed to the
interaction of species from the ambient atmosphere with the exposed ZTO back channel
during gate voltage stress. These species can either accept or donate electrons resulting in
changes in the channel conductance with respect to the applied stress.
©Copyright by Meena Suhanya Rajachidambaram
December 22, 2011
All Rights Reserved
Investigation of Oxide Semiconductor Based Thin Films: Deposition, Characterization,
Functionalization, and Electronic Applications.
by
Meena Suhanya Rajachidambaram
A THESIS
submitted to
Oregon State University
In partial fulfillment of
the requirements for the
degree of
Master of Science
Presented December 22, 2011
Commencement June 2012
Master of Science thesis of Meena Suhanya Rajachidambaram presented on December
22, 2011
APPROVED:
Major Professor, representing Chemical Engineering
Head of School of Chemical, Biological & Environmental Engineering
Dean of the Graduate School
I understand that my thesis will become a part of the permanent collection of Oregon
State University libraries. My signature below authorizes release of my thesis to any
reader upon request
Meena Suhanya Rajachidambaram, Author
ACKNOWLEDGMENTS
First and foremost, I’d like to express my sincere thanks to my parents, Mr. and Mrs.
Rajachidambaram for their endless moral support throughout my graduate life. I want to
thank my elder sister Priya, my brother-in law Gopinath for their enormous
encouragement and advice, and their new born baby Akanshaa for cheering me up during
the month of my defense.
I would like to specifically thank my major advisor, Dr. Gregory S. Herman for the great
opportunity to perform research in his laboratory with his encouragement, financial
support and guidance. As a part of my learning experience, it was good to be one of his
first two students watching the lab grow day by day.
My sincere thanks to Dr.Kenneth J.Williamson and Janet Mosley for their kindness in
offering me a scholarship during my first year of graduate school.
I would like to thank my group members Brendan Flynn and Richard Oleksak for
reviewing my thesis documents and all the helpful discussions towards my research
work. I wish to thank Dr. Seung-Yeol Han for his guidance during my initial stages of
research. My sincere thanks to Dr. Chih-Hung Chang for allowing me to use some of the
capabilities in his lab. I would like to acknowledge Chris Tasker for his availability and
expertise in dealing with any questions related to the research tools in the cleanroom. I
would like to thank Jeremy Campbell for offering me tool time on the UV-Ozone
cleaning system when requested. My heartfelt thanks to Ken Hoshino for his guidance in
the bias stress stability testing.
A portion of the work was performed at the Environmental Molecular Sciences
Laboratory (EMSL) at Pacific Northwest National Laboratory (PNNL) in Richland, WA.
My special thanks to Dr. S. Thevuthasan for providing me an Alternate Sponsored
Fellowship at EMSL. I would like to extend my special thanks to Dr. V. Shutthanandan
for his training and guidance in using the Helium Ion Microscope (HIM), Tamas Varga
and Libor Kovarik for their help with acquiring XRD data and TEM images. My sincere
thanks to Dr. Nachimuthu and his team members Archana Pandey and Subramanian
Vilayurganapathy for their help with XPS data acquisition. My special thanks to Rahul
Sanghavi for providing me the instructions and lab materials for working in cleanroom at
EMSL.
This research work was funded by Oregon Nanoscience and Microtechnologies
Institute (ONAMI) and the Office of Naval Research (ONR) under contract number
200CAR262.
TABLE OF CONTENTS
Page
CHAPTER 1 INTRODUCTION TO OXIDE MATERIALS AND DEVICES……… 1
1.1 Thin film transistor history……………………………………………....... 1
1.2 Thin film transistors based on oxide semiconductor materials…………… 3
1.2.1 Tin Oxide (SnO2)……………………………………………….. 4
1.2.2 Zinc Oxide (ZnO)………………………………………………. 5
1.2.3 Indium Gallium Zinc Oxide (IGZO)……………………………. 7
1.2.4 Zinc Tin Oxide (ZTO)…………………………………………... 8
1.2.5 Indium Zinc Oxide (IZO)……………………………………….. 10
1.2.6 Indium Tin Oxide (ITO)………………………………………… 11
1.2.7 Indium Gallium Oxide (IGO)…………………………………… 12
1.3 Thin film transistor (TFT) overview……………………………………... 13
1.4 Oxide materials for sensors……………………………………………….. 22
1.5 References………………………………………………………………… 26
CHAPTER 2 EXPERIMENTAL METHODS……………………………………….. 32
2.1 Surface treatments………………………………………………………… 32
2.1.1 Ultraviolet-Ozone (UV-O3) cleaning…………………………… 32
2.2 Thin film processing……………………………………………………… 33
2.2.1 Thin film deposition techniques………………………………… 33
2.2.1.1 Spin-coating………………………………………..…. 33
2.2.1.2 RF magnetron sputter deposition……………………... 36
2.2.1.3 Thermal Evaporator…………………………………… 37
2.2.2 Post-deposition thermal annealing……………………………… 39
2.2.3 Photolithography………………………………………………... 40
TABLE OF CONTENTS (Continued)
Page
2.3 Thin film characterization………………………………………………… 42
2.3.1 Goniometer for contact angle measurement……………………. 42
2.3.2 Thermogravimetric Analysis (TGA)……………………………. 44
2.3.3 Scanning Electron Microscopy (SEM)…………………………..46
2.3.4 Helium Ion Microscopy (HIM)…………………………………. 49
2.3.5 Transmission Electron Microscopy (TEM)…………………….. 50
2.3.6 Micro-X-ray diffraction (XRD)………………………………….52
2.3.7 Ellipsometry…………………………………………………….. 54
2.3.8 X-ray photoelectron spectroscopy (XPS)……………………….. 56
2.4 Electrical characterization………………………………………………… 57
2.4.1 Four-point probe………………………………………………… 57
2.4.2 Semiconductor Parameter Analyzer (SPA)……………………... 58
2.5 References………………………………………………………………… 60
CHAPTER 3: FORMATION AND CHARACTERIZATION OF ZINC OXIDE
FILMS USING ZINC NANOPARTICLE DISPERSIONS…………………………... 62
3.1 Introduction……………………………………………………………….. 62
3.2 Experimental methods…………………………………………………….. 64
3.3 Results and discussions…………………………………………………… 68
3.4 Conclusions……………………………………………………………….. 82
3.5 References………………………………………………………………… 83
CHAPTER 4: MOLECULAR PASSIVATION OF AMORPHOUS ZINC TIN
OXIDE THIN FILM TRANSISTORS USING N-HEXYLPHOSPHONIC ACID…... 85
4.1 Introduction……………………………………………………………….. 85
TABLE OF CONTENTS (Continued)
Page
4.2 Experimental……………………………………………………………… 88
4.3 Results and discussions…………………………………………………… 92
4.4 Conclusions……………………………………………………………….. 106
4.5 References………………………………………………………………… 108
CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE
WORK………………………………………………………………………………...114
5.1 Conclusions……………………………………………………………….. 114
5.2 Recommendations for future work………………………………………... 115
BIBLIOGRAPHY…………………………………………………………………….. 117
LIST OF FIGURES
Figure
Page
Figure 1.1: Four basic TFT structure layouts: (a) Staggered bottom-gate, (b)
Staggered top-gate (c) Co-planar bottom-gate, (d) Co-planar top-gate………………. 14
Figure 1.2: Energy band diagrams for a typical n-type, accumulation mode TFT, (a)
Equilibrium energy band diagram i.e at zero bias at gate contact (b) Energy band
diagram when a negative voltage is applied to the gate contact. (c)Energy band
diagram when a positive voltage is applied to the gate contact………………………. 15
Figure 1.3: Typical log ID vs VGS plot illustrating drain current ID, turn on voltage
VON and the drain current ON-OFF ratio (ION-OFF). The VON is ~ -3 V and ION-OFF is
~107. Voltage applied on the drain VDS=1V and the zinc tin oxide (ZTO) channel
width-to-length ratio W/L = 5………………………………………………………… 17
Figure 1.4: The extracted average and incremental (µavg and µinc ) mobilities as a
function of gate bias for a TFT with ZTO channel W/L = 5. The maximum
mobilities are µavg = 13 cm2/V sec and µinc= 22 cm2/V sec………………………….. 19
Figure 1.5: A schematic illustration of the impact of O2- and H2O+ adsorption on
band bending at amorphous oxide semiconductor (AOS) surfaces…………………... 20
Figure 2.1: A sample was irradiated with ultraviolet radiation with wavelengths of
184.9 nm and 253.7nm, to produce O3 inside the apparatus………………………….. 33
Figure 2.2: A schematic of a typical spin-coating process: a) deposition b) spin-up c)
spin- off d) evaporation……………………………………………………………….. 35
Figure 2.3: Schematic indicating basic aspects of RF sputtering……………………... 36
Figure 2.4: Schematic of a thermal evaporator showing the evaporation source and
substrate placement………………………………………………………………….... 38
Figure 2.5: A typical temperature profile used for post-deposition annealing of
semiconductor thin films……………………………………………………………… 39
Figure 2.6: Typical photolithography process steps (for negative photoresist)………. 41
Figure 2.7: A FTA 32 Contact angle goniometer set up……………………………… 42
Figure 2.8: A Schematic of TGA instrumentation showing a furnace and a sample
assembly………………………………………………………………………………. 45
LIST OF FIGURES (Continued)
Figure
Page
Figure 2.9: A typical TGA profile for Zn nanoparticle oxidation in air. A) Loss of
moisture from the surface B) Sample metal is oxidized in air C) Sample metal is
completely oxidized. …………………………………………………………………. 46
Figure 2.10: An SEM image of a ZnO film obtained by sol-gel chemistry………….. 48
Figure 2.11: A secondary electron image of ZnO film subjected to FIB milling…….. 49
Figure 2.12: A schematic of difference in interaction volume for SEM & HIM……... 50
Figure 2.13: A TEM image of an individual Zn nanoparticle showing atomic lattice
fringes…………………………………………………………………………………. 51
Figure 2.14: Schematic showing the process of x-ray diffraction occurring by
constructive interference……………………………………………………………… 53
Figure 2.15: Ellipsometer set-up for measurement of thickness of films and optical
constants………………………………………………………………………………. 55
Figure 2.16: A typical XPS illustration (left) and a process that describes an ejection
of electron (photoelectron) from an innermost shell when interacting with a photon... 57
Figure 2.17: A typical representation of a four-point probe…………………………...58
Figure 3.1: Structure of a bottom-gate thin film transistor (TFT) test structure in
which the Zn nanoparticles are deposited by spin-coating and then oxidized to ZnO
by annealing in air…………………………………………………………………….. 68
Figure 3.2: SEM images showing size distribution of zinc nanoparticles with
estimated average diameters of (a) 162nm (b) 234nm………………………………...69
Figure 3.3: HIM images of Zn nanoparticles, davg = 162nm (a-g) and davg = 234nm
(h-n) films that have been spin-coated and annealed between 100-600°C…………… 71
Figure 3.4: TGA graph showing oxidation of 162nm and 234nm diameter Zn
nanoparticles………………………………………………………………………....... 72
Figure 3.5: X-ray diffraction plots of intensity vs 2θ for (a) Zn nanoparticles, davg =
162nm and (b) Zn nanoparticles, davg = 234nm based on deposited films and films
annealed from 100 °C-600 °C………………………………………………………… 73
LIST OF FIGURES (Continued)
Figure
Page
Figure 3.6: ZnO weight % composition as determined by micro-XRD for Zn
nanoparticles, davg = 162 nm and Zn nanoparticles, davg = 234nm films after
annealing to the indicated temperatures in air………………………………………… 76
Figure 3.7: Calculated thicknesses of ZnO outer shells for Zn nanoparticles, davg =
162nm and Zn nanoparticles, davg = 234nm versus temperature for films annealed in
air……………………………………………………………………………………… 77
Figure 3.8: Electrical characterization of TFTs deposited and annealed to 400°C,
500°C & 600°C by spin coating (a) ZIO films (b) a bottom ZIO layer with a 5%
solution of davg = 162 nm Zn dispersed in ZIO. (c) A bottom ZIO layer with a 5%
solution of davg = 234 nm Zn dispersed in ZIO……………………………………….. 79
Figure 4.1: a) Chemical structure of n-hexylphosphonic acid (n-HPA). Schematic
view of staggered bottom gate ZTO TFT b) without n-HPA c) with n-HPA…............ 90
Figure 4.2: Contact angle (degree) vs n-HPA exposure time (hrs) for each individual
ZTO film exposed to n-HPA at specified intervals of time. The reference film was
exposed for 24 hours. The contact angle directly after UV treatment was 26.6 °…….. 93
Figure 4.3: Zn 2p3/2 XPS spectra obtained from ZTO films with/without n-HPA and
with/without post-annealing…………………………………………………………... 94
Figure 4.4: Sn 3d5/2 XPS spectra obtained from ZTO films with/without n-HPA and
with/without post-annealing…………………………………………………………... 95
Figure 4.5: O 1s XPS spectra obtained from ZTO films with/without n-HPA and
with/without post-annealing…………………………………………………………... 97
Figure 4.6: C 1s XPS spectra obtained from ZTO films with/without n-HPA and
with/without post-annealing…………………………………………………………... 98
Figure 4.7: P 2p XPS spectra obtained from ZTO films with/without n-HPA and
with/without post-annealing…………………………………………………………... 99
Figure 4.8: Plot of ID vs VGS curves for various conditions of the ZTO TFTs as a
function of applied bias stress time. (W/L = 1000 µm/200 µm). a) Without n-HPA,
not post-annealed, b) Without n-HPA, 140 °C annealed, c) With n-HPA, not postannealed, d) With n-HPA, 140 °C annealed…………………………………………... 101
LIST OF FIGURES (Continued)
Figure
Page
Figure 4.9: Shift in VON (V) versus bias stress time (sec) for ZTO n-HPA
functionalized devices that were not subjected to post-annealing, and a fit of the
stretched exponential model with the measured values of the TFTs with no n-HPA ... 103
Figure 4.10: Shift in VON (V) versus bias stress time (sec) for ZTO n-HPA
functionalized devices that were not subjected to post-annealing, and a fit of the
stretched exponential model with the measured values of the TFTs with no n-HPA ... 104
LIST OF TABLES
Table
Page
Table 3.1: Resistivity measurements of Zn nanoparticles, davg = 162 nm and Zn
nanoparticles, davg = 234 nm films spin-coated using dispersions made up of
chloroform/ methanol and n-octylamine as a dispersant and after annealing at 300
°C, 400°C, 500°C & 600°C in air…………………………………………………….. 78
Table 4.1: Characteristic trapping time of carriers, τ and stretched exponential
exponent, β for the ZTO TFTs (without n-HPA) either left in room temperature or
post-annealed to 140 °C for 48 hours…………………………………………………. 104
1
CHAPTER 1 INTRODUCTION TO OXIDE MATERIALS AND DEVICES
This chapter provides a historical review of thin film transistors (TFT), including
device operation. Furthermore, a summary of oxide semiconductor based TFTs is
provided to allow a comparison to results presented in the thesis. Finally, an overview of
oxide materials for sensors is given at the end of this chapter.
1.1 Thin film transistor history
The very first TFT was invented by Paul.K.Weimer of the RCA Laboratories and
was published in the proceedings of the IEEE in 1962 [1]. This TFT was fabricated with
cadmium sulfide (CdS) as the channel layer with Au source and drain contacts, which
were patterned using shadow masks. The transistors operated by the control of injected
majority carriers in the semiconductor where an insulator (silicon monoxide) was used
between the gate and the channel. A maximum drift mobility µd of ~ 140 cm2/V sec was
obtained. In 1968, Brody et al. from Westinghouse Research Laboratory integrated TFTs
with cadmium selenide (CdSe) channels into electroluminescent (EL) display panels [2].
In a TFT-EL panel, the multiplexing and the controlling functions of the panel are
performed by a TFT matrix circuit, while the individual EL elements emit light according
to the ON-OFF conditions of the control TFTs. This invention led directly to the
integration of TFTs as a fundamental component of active-matrix liquid crystal displays
(AMLCDs) in the 1970's. Researchers began to explore TFTs with amorphous Si (a-Si:H)
channels for AMLCDs [3]. Amorphous silicon became the dominant TFT technology for
AM-LCD due to its relatively low cost, fabrication flexibility and high uniformity. The
potential of a-Si:H TFTs for flexible electronics have also been demonstrated by several
groups where mobilities of ~1 cm2/V sec were obtained [4, 5]. With the aim of
2
developing TFTs with higher performance, for system on panel applications, led to the
development of TFTs with polysilicon channels. The first TFT-LCD display was
commercialized by Seiko Epson during 1983. The pocket color TV proposed by Seiko
Epson used microcolor filters and poly-Si TFTs formed on quartz substrates due to the
required high temperature processing [6]. In another study, polycrystalline Si TFTs
devices were fabricated by depositing double layers of amorphous silicon by LPCVD on
an n-type monosilicon wafer at 550 °C and then crystallized by annealing at 600 °C for
12 hours [7]. Silicon dioxide was used as a gate insulator and aluminum was used as
source-drain contacts. The devices showed an incremental mobility µFE of up to 100
cm2/V sec. High quality low-temperature polycrystalline Si (LTPS) films have been
obtained on glass substrates by laser crystallization of dehydrogenated amorphous a-Si
[8]. High performance TFTs can be obtained by reducing the channel length in order to
reduce the number of grain boundaries within the channel. By doing this, mobilities of
410-510 cm2/V sec have been obtained. Although LTPS and CdSe technologies have
relatively high electrical performance, extensive process development is still necessary to
improve uniformity over large areas, which is why LTPS is used primarily for small
displays [9].
More recently, there have been significant developments in the TFTs that use
organic materials as channel layer [10]. Historically organic thin film transistors (OTFTs)
have low mobilities (< 1 cm2/Vs) and are not suitable for use in applications that require
very high switching speeds. However, the electrical performance of OTFTs is good
enough for a variety of applications that require low cost, low temperature processing,
3
and structural flexibility. One of the main advantages of OTFTs is that they can be
processed at or near room temperature and are compatible with plastic substrates [10].
Polythiophene is a polymeric OTFT channel material and recent results have indicated
very low mobilities (as low as 10-5 cm2/V sec) [10, 11, 12]. However, it is not necessary
that organic materials are limited to such poor performance. For example, pentacene is a
small molecule OTFT channel material and has recently been demonstrated as a flexible
OTFT with silk fibroin gate dielectric and gold top electrodes. The field effect mobility
for this device was 23.2 cm2/V sec. Typically, organic semiconductor materials tend to be
p-type although, n-type organic materials are being developed as well. For example,
Fujisaki et al. demonstrated n-type OTFTs with a field effect mobility over 0.1 cm2/V sec
[13]. In order to improve the chemical and physical properties of OTFTs, hybrid
semiconductors have been developed where mixtures of inorganic and organic materials
are used. An example of a hybrid channel material is tin(II) iodide perovskite,
(C6H5C2H4NH3)SnI4, that has channel mobility 0.61 cm2/V sec [14].
1.2 Thin film transistors based on oxide semiconductor materials
The following subsections provide a summary of TFTs that utilize transparent
semiconducting oxides as the channel material. The class of materials that were primarily
used were from the transparent conducting oxide literature. These materials can have
high conductivity and optical transparency and find application in solar cells and
AMLCDs. Initial investigations were performed on polycrystalline binary oxides
including tin oxide (SnO2) [16-19] and zinc oxide (ZnO) [20-28]. More recently, Hosono
et al. proposed that transparent conducting oxides (TCOs) can be formed by using multi-
4
component oxides composed of heavy metal cations, where these cations have a (n1)d10ns0 (n≥4) electronic configuration [15]. Hosono et al. have also demonstrated that
these materials can also be used as the TFT channel material when one reduces the carrier
concentration (i.e., makes the material less conductive). Consequently, a variety of
amorphous oxide semiconductors have been used for TFT applications including, indium
gallium zinc oxide (IGZO) compounds [29-31], zinc tin oxide (ZTO) [32-38] , indium
zinc oxide (IZO) [39-43], indium tin oxide (ITO) [44-45], and indium gallium
oxide(IGO) [46-48]. Since ZnO, IZO and ZTO were used in our studies, they will be
explained in more detail below.
1.2.1 Tin Oxide (SnO2)
Aoki et al. fabricated a coplanar TFT using SnO2 films as a channel where
aluminum was used for the gate, source and drain [16]. The SnO2 films were deposited
using a vapor phase reaction, and the gate dielectric was a double layer composed of SiO
and nitrocellulose. The SnO2 channel layer was a few microns thick and the device had a
transconductance of 0.3 m ohm [16].
Sb-doped n-type SnO2 TFTs were demonstrated where PbZr0.2Ti0.8O3 and SrRuO3
were used as the ferroelectric gate insulator and gate electrode respectively [17]. The gate
electrode, gate insulator, and the channel layer were all deposited on a strontium titanate
(SrTiO3) substrate by pulsed laser deposition (PLD). The use of ferroelectric gate
insulator allowed the TFT to have an intrinsic memory. The channel conductance at zero
5
voltage was dependant on polarity of the prior gate bias due to ferroelectric polarization.
The best TFTs had a mobility of 5 cm2/V sec and a threshold voltage of -2 V.
Presley et al. fabricated a TFT with SnO2 channel layer deposited by RF
magnetron sputtering on a multilayer aluminum oxide/titanium oxide (ATO) dielectric
and an ITO gate [18]. Rapid thermal annealing of the SnO2 channel was done at 600 °C in
an O2 ambient, followed by deposition of ITO source-drain. Both enhancement mode and
depletion mode devices were demonstrated. The mobility of the best enhancement mode
device was 0.8 cm2/V sec and that for the best depletion mode device was 2 cm2/V sec.
Recently, Huh et al. fabricated TFTs with a tin oxide channel deposited by ultralow
pressure sputtering and achieved µFE as high as ~ 30 cm2/V sec, demonstrating that
relatively low values of sputtering pressure allows reduction and/or control of the carrier
concentration and produces high quality films with reasonably high mobilities [19].
1.2.2 Zinc Oxide (ZnO)
Polycrystalline ZnO has been used as a channel material for TFTs by many
groups. A variety of methods have been used to fabricate ZnO TFTs including ion beam
sputtering, radio frequency (RF) sputtering and pulsed laser deposition (PLD).
Transparent thin film transistors (TTFT) have been demonstrated with a ZnO channel
layer deposited via RF sputtering [20]. The gate dielectric for these devices was ATO and
the gate and source-drain electrodes were ITO. It was necessary to anneal the devices up
to 600-800 °C in O2 to improve the crystallinity of the ZnO channel, increase the channel
resistivity, and improve the electrical quality of ZnO/insulator interface. The devices had
6
channel mobilities and threshold voltages in the range of 0.3-2.5 cm2/V sec and a 10-20
V, respectively [20]. Other groups have focused on reducing the thermal budget for ZnO
TTFT [21]. Studies have indicated that controlling the oxygen pressure during deposition
reduces the maximum process temperature to 450 °C. Maximum channel mobilities of
0.97 cm2/V sec and threshold voltage of -1V were obtained [21]. TFTs using room
temperature deposited ZnO films have resulted in devices, which do not require a postdeposition anneal. In this study, the ZnO films were deposited on heavily doped n-type Si
wafers with ~100nm thick oxide layer and the devices used Ti/Au source and drain
electrodes [22]. Mobilities of ~ 2 cm2/V sec and a threshold voltage of 0 V were
obtained. More recent studies have shown that low temperature processed sputter
deposited ZnO channels can have mobilities of 50- 70 cm2/V sec [23, 24].
For low-cost electronics, other methods have been evaluated for the non-vacuum
based deposition of ZnO for TFTs. To improve crystallization and device performance,
high temperature post-deposition annealing is typically performed on solution deposited
ZnO films. For example, Norris et al. fabricated a TTFT that had a ZnO channel layer
formed by chemical solution deposition using spin-coating [25]. In this study, a zinc
nitrate based solution was spun onto an ATO/ITO substrate and then annealed in air for
10 minutes at 600 °C to convert to ZnO. The films were then annealed to 700 °C to
further improve crystallinity using rapid thermal annealing. These ZnO TFTs used ITO
source, drain, and gate electrodes. The n-type enhancement mode ZnO TFTs had a VON
of 3 V and an effective mobility µeff (~0.2 cm2/V sec). ZnO nanorods have also been
used for TFTs where the nanorods were spin-coated on Si/SiO2 substrates and post-
7
annealed to 230 °C for 30 min [26]. The devices used gold-chromium source-drain
electrodes and had a field effect mobility of 0.61 cm2/V sec. Solution processed TFTs
using thin films of ZnO nanoparticles with various particle shapes including nanospheres
and nanorods have also been fabricated [27]. The ZnO nanoparticles were prepared by
polyol method in which zinc acetate dihydrate was dissolved in diethylene glycol. The
films were spin-coated on Si/SiO2 substrates followed by annealing in either air or
oxygen in order to improve the connectivity between nanoparticles in the films. The
TFTs with channel mobility of up to 4 x 10-4 cm2/V sec were obtained using nanorod
films annealed to 600 °C for 5 hours in oxygen. Solution-processed ZnO TFTs with
channel mobility of 13 cm2/V sec have also been demonstrated in literature [28].
1.2.3 Indium Gallium Zinc Oxide (IGZO)
Indium gallium zinc oxide (IGZO) is an n-type semiconductor with a band-gap of
~3.5 eV and a stoichiometry that can be given by the following expression In1-x
Ga1+xO3(ZnO)k where 0<x<1. In recent years, IGZO has been widely investigated for
TFT applications [29-31]. The initial work on this system was by Nomura et al., where
TTFTs were fabricated by the epitaxial growth of an IGZO superlattice on single crystal
substrates [29]. A lattice matched single- crystal yttria stabilized zirconia substrate was
used and a hafnium oxide gate insulator was grown on top of IGZO. To improve the
quality of the single crystalline IGZO, thermal annealing was performed at 1400 °C for
30 minutes in an atmospheric electric furnace. The µFE and VT obtained for these TTFTs
were 80 cm2/V sec and 3 V, respectively. Nomura et al. also prepared IGZO TFTs by
depositing amorphous IGZO films via pulsed laser deposition at room temperature on
8
polyethylene terephthalate (PET) substrates [30]. These films were determined to be
amorphous by XRD, and TFTs were fabricated using Y2O3 as the gate insulator and ITO
(Sn :10%) as the source, drain and gate electrodes. The deposited IGZO films exhibited a
Hall effect mobility exceeding 10 cm2/V sec, while the best performance for the TTFTs
was µFE and VT of 6-9 cm2/V sec and 1.6 V, respectively.
Higher performance a-IGZO TFTs have been fabricated on glass substrates. For
example, Jeong et al. fabricated a-IGZO TFTs by RF-magnetron sputtering by depositing
200 nm MoW on a SiO2/glass substrate with a SiNx gate dielectric [31]. The sputtering
was carried out with a gas mixing ratio of Ar/O2 = 65/35, while the chamber pressure was
varied from 1 to 5 mTorr. The IZO source and drain electrodes were deposited in the
same chamber at room temperature. Finally the samples were annealed at 350 °C for 1
hour in an N2 atmosphere. The IGZO films that were deposited at low pressure (1 mTorr)
had a µFE of 21.8 cm2/V sec. It was suggested that lower deposition pressure leads to
greater densification of the a-IGZO films resulting in better device performance [31].
1.2.4 Zinc Tin Oxide (ZTO)
Zinc tin oxide (ZTO) is a wide band gap n-type semiconductor that initially has
been studied as a TCO, and has the general formula (ZnO)x-(SnO2)1-x where 0<x<1.
Chiang et al. fabricated bottom gate TFTs and TTFTs with amorphous ZTO as the
channel layer [32]. The ZTO channel layer was deposited via RF magnetron sputtering on
either an ATO/ITO glass substrate or a heavily doped Si wafer with a thermally grown
SiO2 gate insulator. XRD analysis was performed on these ZTO films and it was
9
confirmed that films annealed at a temperature below 650 °C were amorphous and those
annealed above 650 °C were polycrystalline. ITO source and drain electrodes were used
to complete fabrication of the devices. It was found that the µFE for the devices annealed
to 300 °C was 5-15 cm2/V sec and for those annealed to 600 °C was 20-50 cm2/V sec.
The effects of annealing and stoichiometry on ZTO based TFTs was explored in
detail by Hoffman [33]. Sputter targets with various stoichiometries of Zn/(Zn +Sn)
between 0 and 1 were used. It was found that devices with channel Zn/(Zn +Sn) ratios of
0.33, 0.5 and 0.67 with post-deposition anneals between 400-600 °C yielded, µFE of ~30
cm2/V sec. Also, the turn on voltage decreased with increasing annealing temperature and
decreasing Zn/(Zn +Sn) ratio.
Hong et al. evaluated methods to passivate ZTO channel TFTs [34]. His best
results was for ~100nm thermally evaporated SiO2 on top of ZTO. A post deposition
anneal was performed after deposition of the ZTO and second annealing was done after
passivation. The µinc for the unpassivated TFT annealed at 600 °C was found to be 20
cm2/V sec and that for the SiO2 passivated TFTs with no post-passivation annealing, 300
°C annealing and 600 °C annealing were found to be 20, 26 and 15 cm2/V sec,
respectively. Triska et al. studied the bias stress stability of ZTO TFTs fabricated with
Al2O3 dielectrics deposited by atomic layer deposition (ALD) or SiO2 gate dielectrics
deposited by plasma enhanced chemical vapor deposition (PECVD) [35]. These devices
had a positive parallel shift of the transfer data under bias stress. A larger shift was
observed for Al2O3 dielectrics compared to that of SiO2. To improve the bias stress
stability of the Al2O3 devices, a thin layer of a SiO2 was deposited by PECVD between
10
the Al2O3 and ZTO films. This greatly improved the bias stress stability of the devices
suggesting that the Al2O3/ZTO interface was a dominant source of charge trapping.
Recently, bias stress stability of ZTO as a function of various dielectrics, passivation
layers, illumination, and solvent mediated exposures has been reported [36-38].
1.2.5 Indium Zinc Oxide (IZO)
Recently, IZO has been introduced as a robust binary metal oxide material for
TFT and TTFT applications. IZO is sometimes referred to as zinc indium oxide (ZIO).
IZO is typically amorphous as deposited which limits grain boundary scattering. Dehuff
et al. fabricated n-type IZO TTFT that had very high mobilities [39]. RF magnetron
sputtering was used to deposit IZO on an AZO/ITO glass substrate. The ITO source-drain
was deposited after annealing the channel to either 300 °C or 600 °C. The 300 °C
annealed devices operated in enhancement mode with µinc of 10-30 cm2/V sec and VT of
0-10 V, while those annealed to 600 °C operated in depletion mode with µinc of 45-55
cm2/V sec and VT of -20 to -10V. More recently, bottom gate IZO TTFTs with no post
deposition annealing was demonstrated using RF magnetron sputtering [40]. The devices
were fabricated on glass substrates coated with AZO/ITO layer and used IZO sourcedrain contacts. In this study, the channel thickness was varied between 15 nm to 60 nm
and it was found that the channel mobility varied with channel thickness. The best
performing devices had a mobility of 40 cm2/V sec with threshold voltages of 10 V for
IZO thicknesses of 15 nm. The threshold voltage range from 3 V and 10 V was observed
to vary for the thicker and thinner channels, respectively [40].
11
As mentioned previously for ZnO, solution processed TFTs offer the potential for
low cost when compared to the vacuum-deposited TFTs [41]. Choi et al. fabricated
TTFTs with IZO channel layer produced by spin-coating. The IZO solution was
formulated with zinc acetate dihydrate and indium acetate in 2-methoxyethanol, and was
spin-coated at 8000 rpm on Si substrate with 120nm SiO2 gate dielectric. The devices
were annealed for 1 hour in air at 500 °C. For these devices Al source/drain were
deposited by evaporation. This depletion mode TFT demonstrated a field effect mobility
of 7.3 cm2/V sec with a VT of 2.5 V [41]. Lee et al. used metal-halide precursors
dissolved in acetonitrile and formed uniform and continuous thin films using ink-jet
printing and spin coating for TFT applications [42]. The devices had µFE of 7.4 cm2/sec
and 16.1 cm2/V sec with Von of -26 V and -32 V were obtained for the TFTs with ink-jet
printed and spin-coated channels, respectively. Han et al., demonstrated solution
processed IZO TFTs using a metal-halide and acetonitrile chemistry that was described
above [43]. Devices that were annealed at 500 °C in air had a µFE of 6.1 cm2/V sec and
Von of ~ 0 V while those annealed to 280 °C had µFE of 0.84 cm2/V sec and Von of ~ 6 V.
1.2.6 Indium Tin Oxide (ITO)
Indium Tin Oxide (ITO) has been used as gate and source-drain electrode for
various TFT and TTFT applications [20, 21, 32]. The most commonly used form of ITO
has 10% Sn by weight and is highly conductive. Tahar et al. reported that ITO films
formed under varying processing conditions can have a wide range of electrical
properties [44]. The deposition techniques used were physical vapor deposition (PVD),
chemical vapor deposition (CVD) and sol-gel processing. The electrical properties can be
12
controlled by several processing parameters including film thickness, substrate heating,
deposition technique, and oxygen partial pressure. For example, the carrier mobilities for
ITO films were found to vary between 15-41 cm2/V sec for DC and RF magnetron
sputtering, 16-103 cm2/V sec for vacuum evaporation, 5-70 cm2/V sec for CVD, and 921cm2/V sec for various sol-gel procedures [44].
Jiang et al. has demonstrated the use of ITO as both the channel and the sourcedrain layers where the layers were simultaneously deposited using one step magnetron
sputtering onto low cost paper substrates while patterning with a single shadow mask
[45]. The devices had high performance with a large on/off ratio of 8 x 105, a low
operating voltage of 2.0 V, and a µFE estimated to be 22.4 cm2/V sec.
1.2.7 Indium Gallium Oxide (IGO)
Indium gallium oxide (IGO) has been explored for its potential as a TCO and for
TFTs [46, 47]. This n-type semiconductor has a stoichiometry which can be best
described as In1+x-Ga1-x-O3 where, 0 < x < 1. Chiang et al. fabricated high performance
bottom-gate TFTs employing amorphous indium gallium oxide (a-IGO) as a channel
material [47]. About 50 nm of IGO was sputter deposited on p-type heavily doped Si
substrates with ~100 nm thermally grown SiO2 that functions as a gate dielectric. Films
of various ratios of In/Ga were deposited and these films were annealed between 200-800
°C. ITO source and drain electrodes were deposited after the annealing step. It was found
that decreasing oxygen partial pressures during IGO deposition increased the µinc with a
corresponding decrease in Von. Likewise, an increase in indium concentration of the
13
channel material resulted in an increase of µinc with a corresponding decrease in Von. The
best performing devices were annealed at 600 °C and had a µinc of 27 cm2/V sec and a
Von of -14 V. The IGO devices annealed to lower temperatures (< 200 °C) had a µinc of
19 cm2/V sec and a Von of 2 V [47].
Goncalves et al. further studied the effect of oxygen partial pressure for sputter
deposited IGO thin films [48]. The IGO films were deposited on a Si substrate that had
100 nm SiO2 layer deposited by PECVD. It was shown that the oxygen partial pressure
could result in amorphous films (when O2 was present) or polycrystalline films (when O2
was not present). The best performing TFTs with un-annealed IGO had µsat ~ 43 cm2/V
sec and VT ~ 3.05 V.
1.3 Thin film transistor overview
There are four basic layouts for a TFT, as illustrated in the Figure 1.1 namely, (a)
staggered bottom-gate (b) staggered top-gate (c) co-planar bottom gate, and (d) co-planar
top gate. The source/drain and gate placement defines these various structures. The
device is a top gate structure if the gate is located above the channel layer. In contrast, the
device is in bottom gate structure, if the gate is located below the channel layer. In a
staggered configuration, the source/drain are located on the opposite side of the channel
layer from the gate. Likewise, in a co-planar configuration, the source/drain terminals are
on located on the same side of the channel layer as the gate.
Furthermore, TFTs can be classified based on their mode of operation where
enhancement-mode TFTs require application of gate bias to form a channel and induce
14
cu
urrent flow (i.e
( to turn th
he device "on"), and deppletion-modee TFTs requiire applicatioon of
gate bias to reestrict the flo
ow of curren
nt through thhe channel ( i.e to turn thhe device "offf").
Figure
F
1.1: Four
F
basic TF
FT structures: (a) Staggeered bottom--gate, (b) Staaggered top--gate
(cc) Co-planarr bottom-gatte, (d) Co-plaanar top-gate.
Figuree 1.2 shows energy band
d diagrams fo
for an n-type accumulatioon mode TFT
hat operates under three modes. Wheen no voltagge is applied to the gate, tthe energy bbands
th
arre assumed to
t be at flat-b
band as illusstrated in Figgure 1.2 (a). When a neggative voltagge is
ap
pplied to thee gate, the deelocalized electrons that are present in the channnel layer are
reepelled from
m the semicon
nductor/dieleectric interfaace which prroduces a deepletion regioon
th
hat is positiv
vely charged, implying th
hat the regioon is depletedd of negativee charge. Thhe
co
ollection of positive
p
chaarge near the interface, caauses an upw
ward band-bbending of thhe
co
onduction an
nd valence bands
b
as show
wn in Figuree 1.2 (b). Buut, when a poositive voltagge is
ap
pplied to thee gate contacct, the delocaalized electroons that are present in thhe channel laayer
15
arre attracted to
t the semico
onductor/dieelectric interrface, which produces ann accumulatiion
reegion that is negatively charged.
c
Thiis accumulattion of negattive charge nnear the interrface
caauses a negaative or down
nward band--bending of tthe conductiion and valence bands ass
illlustrated in the
t Figure 1.2 (c).
Figure
F
1.2: Energy
E
band diagrams fo
or a typical nn-type, accum
mulation moode TFT, (a)
Equilibrium
E
energy
e
band
d diagram i.ee at zero biass at gate (b) E
Energy bandd
diagram
d
when a negativee voltage is aapplied to thee gate. (c) Energy band
diagram
d
when a positive voltage is appplied to thee gate.
Equattions 1.1 and
d 1.2, represeents "square--law" whichh can be usedd to model thhe
op
peration of a metal-oxid
de semicondu
uctor field-e ffect transisttor (MOSFE
ET) [49]. Forr an
n-type transisstor, the accu
umulation off delocalizedd electrons nnear the
16
semiconductor/dielectric interface, is described as the formation of a "channel", and is
dependent on the applied voltage and the polarizability of the dielectric. The application
of positive voltage to the drain electrode results in the extraction of delocalized electrons
into the accumulation layer from the channel. This results in a high electron
concentration, which can result in a current flow between source and the drain, called the
drain current (ID), when a positive drain voltage is applied. It is necessary for the applied
gate bias to be larger than the voltage needed to turn the device "on" (i.e, VGS ≥ VON)
where VGS is the gate-to-source voltage and VON is the turn-on voltage. In general, VON is
defined as the VGS at which the drain current (ID) increases above the leakage current (IG)
for small values of drain voltage (VDS) [50]. At small positive drain voltages, i.e.,
voltages less than or equal to the gate voltage minus the turn-on voltage VDS ≤ VGS-VON,
the drain current conduction for a TFT can be expressed by equation 1.1 [51],
= μ
(
−
)
−
( . )
where µ is the channel mobility, COX is the oxide capacitance, W is the width of the
channel, and L is the length of the channel.
At large positive drain voltage, electrons are depleted at the channel region that is
close to the drain. This is termed the channel pinch-off point where the drain current
becomes independant of VDS (i.e the drain current saturates). This occurs when VDS =
VDS,SAT , where the saturation drain voltage VDS,SAT = VGS-VON by which equation 1.1
becomes,
17
,
=
µ
(
) ( . )
−
The square law designation arises from the quadratic dependence displayed in
equation 1.2 in which the saturation drain current (ID,SAT) is proportional to the square of
the applied gate voltage in excess of the turn-on voltage. The entire derivation for the
log (ID/(W/L)) (A)
square law model can be found elsewhere [51].
-3
-4
-5
-6
-7
-8
-9
-10
-11
-12
-13
ZTO channel
W/L = 5
VDS = 1 V
ION/OFF
VON
-20 -15 -10
-5
0
5
10
15
20
VGS (V)
Figure 1.3 : Typical log ID vs VGS plot illustrating drain current ID, turn on voltage VON
and the drain current ON-OFF ratio (ION-OFF). The VON is ~ -3 V and ION-OFF is
~107. Voltage applied on the drain VDS=1 V and the zinc tin oxide (ZTO)
channel width-to-length ratio W/L = 5.
Electrical characterization of TFTs involves data collection and analysis which is
used to identify the performance of the device. Transfer characteristics are measured
where the drain voltage is fixed, and the gate voltage is scanned. Both the drain current
18
and gate current are monitored during these scans. The three main device characteristics
evaluated from the transfer curves are the turn-on voltage, drain current on-off ratio, and
channel mobility. The turn-on voltage is where the gate voltage VGS at which the drain
current ID starts to increase for small values of drain voltage (VDS) on a log ID vs VGS
plot. The drain current on-off ratio (ION_OFF) is the ratio of maximum to minimum drain
current at a high value of VDS to ensure that the device is operating in the saturation
regime [50]. The typical values for most of the ZTO TFTs fabricated for this thesis are
ION_OFF =107 and VON is ~ -3 V as shown in Figure 1.3.
Channel mobility is one of the most common parameters used for characterizing a
TFT. Hoffman et al. has given a detailed description on methods to extract channel
mobilities and precise meanings that are important to the electrical data presented in the
paper. Using the nomenclature of Hoffman, the average mobility (µavg) is defined as the
average mobility of all the carriers present in the channel and is directly proportional to
the channel conductance (GCH) [50].
μ
(
)=
(
)
(
−
)
( . )
where,
(
)=
→ │
( . )
19
Likewise, the incremental mobility (µinc) is the mobility of the carriers added to the net
channel charge when increasing the gate voltage. This is proportional to the differential
channel conductance as follows.
μ
(
)=
′ (
)
( . )
For this analysis, VON replaces the threshold voltage (Vth), that is commonly used
for ideal field effect transistors (FET), where µavg and µinc are nearly identical to the
effective mobility µeff and field effect mobility µFE, respectively [52]. Figure 1.4 shows
extracted values of µinc and µavg as a function of gate bias for the device shown in Figure
1.3. The maximum of µinc is 22 cm2/V sec and the maximum of µavg is 13 cm2/V sec.
25
µinc
µavg
Series1
µ (cm2/V s)
20
Series3
15
10
5
0
-30
-20
-10
0
10
20
30
VGS (V)
Figure 1.4: The extracted average and incremental mobilities as a function of gate bias
for a TFT with ZTO channel W/L = 5. The maximum mobilities are µavg =
13 cm2/V sec and µinc = 22 cm2/V sec.
20
Metall oxides are sensitive
s
to the
t absorptioon of molecuules found inn the ambiennt
h can act as either a donoor or an acceeptor of elecctrons
attmosphere (ee.g., O2 and H2O), which
to
o/from the su
urface [4]. For example, O2 from thee atmospheree can adsorbb onto the surrface
of a metal oxiide and captture an electrron from thee conductionn band forminng [O2]-. This
O2]- species leads
l
to surfface depletio
on of carrierss and results in a decreassed channel
[O
co
onductivity. Similarly, H2O can don
nate an electrron to the chhannel forminng [H2O]+ w
which
leeads to an acccumulation of carriers and
a results inn an increaseed channel conductivity.
Other
O
molecu
ules such as H2, CO2 can
n also react w
with the metaal oxide surfface to produuce
ch
hanges in ch
hannel condu
uctivity. A scchematic illuustration of tthe impact oof O2- and H2O+
ad
dsorption on
n band bendiing at amorp
phous oxide ssemiconducttor (AOS) suurface is shoown
in
n Figure 1.5.. A TFT who
ose channel is
i left un-pa ssivated wouuld have a shhift in transiistor
Vth or VON if the surface concentratio
c
on of these sppecies changge during devvice operatioon. A
TFT
T channel can thereforre be passivaated using innorganic or oorganic mateerials, whichh
co
ould possibly act either as
a a protection layer or a sensor layeer.
Figure
F
1.5: A schematic illustration of
o the impacct of O2- and H2O+ adsorpption on bannd
bending
b
at am
morphous ox
xide semiconnductor (AO
OS) surfaces (Adapted froom
th
he study in [53].
21
One of the ways that TFT device stability can be assessed is by constant-voltage
DC bias stress testing. For these measurements the TFTs are stressed at a constant VGS
and VDS over a period of time. The transfer measurements are performed and VON is
determined with respect to stress time. For our bias stress experiments we set VGS = 10V
and VDS = 0 V. Before and after stability testing, transfer characteristics log10 ( ID) -VGS at
VDS = 1V were measured and plotted.
Our constant voltage bias stress testing was executed by first performing two
log10(ID) -VGS transfer curves : Initial transfer curve and the pre-stress curve measured
with VDS = 1V by double sweeping the gate voltage from either -10 to 15 V or -15 to 10
V depending on the value of VON. After acquiring these curves, transfer characteristics
were obtained with the same conditions after 10, 100, 1000, and 10000 seconds of
applied bias on the gate.
A detailed numerical analysis of data associated with bias stress stability
measurements can be performed. Stretched exponential equation has been developed to
model the change in threshold voltage (ΔVth) by charge trapping mechanisms in a-Si
TFTs [54]. The stretched-exponential model best describes the ΔVth by equation 1.6,
|
| =|
|
−
−
( . )
,
( )−
=
=
,
−
( )( . )
( )( . )
22
where, ΔV0 is the effective voltage drop across the gate insulator (or also called effective
stress voltage), VG,stress is the voltage applied to the gate during bias stressing, Vth(0) is the
initial threshold voltage, β is the stretched exponential exponent and τ is the characteristic
trapping time of carriers expressed by τ = τ0 exp (Eτ/kT). Where, Eτ is the average
effective energy barrier that electrons in the TFT channel need to overcome before they
can enter the insulator or near the interface region with τ0 being the thermal pre-factor for
emission over the barrier.
For our research, ΔVth and ΔV0 were obtained from the transfer curves associated
with the pre- and post-stress measurements corresponding to the 0, 10, 100, 1000, and
10000 sec. The time dependence of ΔVth can be well fitted with the stretched-exponential
equation 1.6 by appropriately determining τ and β. A larger τ and/or a larger β correspond
to more stable TFT.
1.4 Oxide materials for sensors
In the1960's, it was discovered that the adsorption and desorption of gaseous
constituents on the surface of metal oxide semiconductors can significantly modulate
their conductivity. It was found that semiconductor materials with wide band-gap have
several advantages for gas sensing applications, including the ability to operate at high
temperatures and having high environmental stability [55]. Metal oxide semiconductor
materials having large surface area is highly desirable since they can adsorb as much
target material as possible thereby leading to a higher measurable response [56]. This
phenomenon was first demonstrated by Seiyama et al. using thin films of zinc oxide
23
(ZnO) [57]. The authors found that at high temperatures (~ 400 °C), the adsorption and
successive desorption of gaseous species on the semiconductor surface took place rapidly
and were correlated to the observed changes in electrical conductivity of the
semiconductor material. It was proposed that thin films of ZnO can be used for the
detection of gaseous constituents [57]. The conductivity changes in SnO2 has also been
observed when CO, CO2, NO, and NO2 are exposed to the surface. Many other oxide
materials have been used for sensing including TiO2 [58] for CO, ITO [59] and BaTiO3
[60] for CO2[61] and WO3[61] and ZnO (at 400 °C) for NOx gas detection, with many
others reported in literature [56] Metal oxide films for gas sensors are commonly made
by chemical vapor deposition (CVD), physical vapor deposition (PVD) including
evaporation and sputtering, and sol-gel techniques [56].
Recently, oxide materials have also been investigated for liquid phase sensing. In
2003, Chibirova et al. fabricated In2O3 sensors with Pt electrodes on an insulating surface
and exposed the sensors to various mediums including water, sodium chloride solutions,
and Luria broth without and with E-coli bacteria before and after UV irradiation, for
evaluation of dissolved oxygen content [62]. The technique revealed high sensitivity.
Zhou et al. proposed that ZnO can be used as biosensors due to its properties including
biodegradability and biocompatibility [63]. In this study, ZnO nanowires with a
hexagonal cross-section and high crystalline quality were produced by vapor-solid
growth process and exposed to various solutions with moderate pH values including
ammonia, sodium hydroxide solution, horse blood serum, and de-ionized water to
observe the etching and dissolving behavior of ZnO. The authors anticipated that the
24
biosensors made with ZnO will have a certain time to perform the device function and
once completing the sensing, the ZnO nanowires can eventually dissolve into ions and
can be completely absorbed by the body, especially in the case of horse blood serum [63].
Followed by these studies, Mason et al. examined approaches to functionalize the surface
of ZnO nanowires [64]. One of the main limitations of oxide-based sensors is their poor
selectivity to different components in the medium to be sensed. Surface functionalization
of the oxide material with specific chemical properties is essential to improve the
selectivity. ZnO nanowires were formed by vapor-solid growth method, and used to
fabricate ZnO nanobridge sensor devices on Si substrates with 300nm thick oxide
insulator layer, after patterning and carbonizing at 900 °C. The ZnO nanowires were
functionalized with biotin. Device operation could not be determined since biotinylated
ZnO NW surfaces were found to dissolve during exposure to DI water. Parylene-A, a
common moisture barrier, was coated via CVD to prevent dissolution of the ZnO
nanowires. The authors have proposed that parylene-A is an excellent encapsulation for
ZnO nanowires and can improve selective sensing of target species in the liquid or vapor
phase [64].
Self-assembled monolayers (SAM) have been used for surface functionalization
of materials and to achieve stable surfaces for sensing [65]. These SAMs are usually
defined by the chemistry that binds them to the surface and can consist of silanes, thiols,
carboxylic acids and phosphonic acids as molecular anchors [66]. Among them,
phosphonic acid and alkyl phosphonates have been demonstrated to form well-packed
SAMs by strongly binding to a wide range of metal oxides when compared to carboxylic
25
acids [65]. This suggests that phosphonic acid based chemistries can be utilized to
functionalize metal oxide surfaces and form stable sensors.
26
1.5 REFERENCES
1. P.K. Weimer, "The TFT - A New Thin Film Transistor", Proc. IRE, 50, 1462 (1962).
2. T.P. Brody, F.C. Luo, Z.P. Szepesi, D.H. Davies, "A 6x6-in 20-lpi Electroluminescent
Display Panel", IEEE Trans.Electron Devices, ED-22, 739 (1975).
3. M.J. Powell, B.C. Easton, O.F. Hill, "Amorphous silicon-silicon nitride thin-film
transistors", Appl. Phys. Lett., 38, 794 (1981).
4. H. Gleskova, P.I. Hsu, Z. Xi, J.C. Sturm, Z. Suo, S. Wagner, "Field-effect mobility of
amorphous silicon thin-film transistors under strain", J Non-Cryst Solids., 338–340,
732 (2004).
5. A. Kuo, T.K. Won, J. Kanicki, "Advanced Multilayer Amorphous Silicon Thin-Film
Transistor Structure: Film Thickness Effect on Its Electrical Performance and Contact
Resistance", Jpn J Appl Phys., 47, 3362 (2008).
6. S. Morozumi, K. Oguchi, S. Yazawa, T. Kodaira, H. Ohshima, T. Mano, "B/W and
color LC Video displays addressed by Poly Si TFTs", SID Digest, 83, 156 (1983).
7. L. Pichon, K. Mourgues, F. Raoult, T. Mohammed-Brahim, K. Kis-Sion, D. Briand, O.
Bonnaud, "Thin film transistors fabricated by in situ doped unhydrogenated polysilicon
films obtained by solid phase crystallization" Semicond. Sci. Technol, 16, 918 (2001).
8. Y. Helen, R. Dassow, M. Nerding, K. Mourgues, F. Raoult, J.R. Kohler, T.
Mohammed-Brahim, R. Rogel, O. Bonnaud, J.H. Werner, H.P. Strunk, "High mobility
thin film transistors by Nd:YVO4-laser crystallization" Thin Solid Films, 383, 143
(2001).
9. Y. Kuo, "Thin film transistor technologies V", The Electrochemical Society
Proceedings, 2000-31,121 (2001).
10. C.D. Dimitrakopoulos, D.J. Mascaro, "Organic thin-film transistors: A review of
recent advances", IBM J. of. Res and Dev , 45, 11 (2011).
11. C-H. Wang, C-Y. Hsieh, J-C. Hwang, "Flexible Organic Thin-Film Transistors with
Silk Fibroin as the Gate Dielectric", Adv Mater., 23, 1630 (2011).
12. Q. Meng, H. Dong, W. Hu, D. Zhu, "Recent progress of high performance organic thin
film field-effect transistors", J. Mater. Chem., 21, 11708 (2011).
13. Y. Fujisaki, Y. Nakajima, D. Kumaki, T. Yamamoto, S. Tokito, T. Kono, J-i Nishida,
and Y. Yamashita, "Air-stable n-type organic thin-film transistor array and high gain
complementary inverter on flexible substrate", Appl. Phys. Lett., 97, 133303 (2010).
27
14. D.B. Mitzi, K. Chondroudis, C.R. Kagan, "Organic thin film transistors: A review of
recent advances", IBM J. of. Res and Dev., 45, 3, (2001).
15. H. Hosono, "Working hypothesis to explore novel wide band gap electrically
conducting amorphous oxides and examples", J Non-Cryst Solids., 198-200, 165
(1996).
16. A. Aoki, H. Sasakura, "Tin oxide thin film transistors", Japan. J. Appl. Phys., 9, 582
(1970).
17. M.W.J. Prins, K.-O. Grosse-Holz, G. Muller, J.F.M. Cillessen, J.B. Giesbers, R. P.
Weening, R. M. Wolf, "A ferroelectric transparent thin-film transistor" Appl. Phys.
Lett., 68, 3650 (1996).
18. R.E. Presley, C.L. Munsee, C-H. Park, D. Hong, J.F. Wager, D.A. Keszler, "Tin oxide
transparent thin-film transistors", J. Phys. D: Appl. Phys., 37, 2810 (2004).
19. M.S. Huh, B.S. Yang, S. Oh, J-h Kim, B.D. Ahn, J-H. Lee, J. Kim, J.K. Jeong, C.S.
Hwang, H.J. Kim "Improving the Performance of Tin Oxide Thin-Film Transistors by
Using Ultralow Pressure Sputtering", J. Electrochemical Society, 157, H425 (2010).
20. R.L. Hoffman, B.J. Norris, J. F. Wager, "ZnO-based transparent thin-film transistors"
Appl. Phys. Lett., 82, 733 (2003).
21. S. Masuda, K. Kitamura, Y. Okumura, S. Miyatake, H. Tabata, T. Kawai,
"Transparent thin film transistors using ZnO as an active channel layer and their
electrical properties", J Appl. Phys., 93, 1624 (2003).
22. P.F. Carcia, R.S. McLean, M.H. Reilly, G. Nunes, Jr., "Transparent ZnO thin-film
transistor fabricated by rf magnetron sputtering", Appl. Phys. Lett., 82, 1117 (2003).
23. E. Fortunato, A. Pimentel, L. Pereira, A. Goncalves, G. Lavareda, H. Aguas, I.
Ferreira, C.N. Carvalho, R. Martins, "High field-effect mobility zinc oxide thin film
transistors produced at room temperature", J Non-Cryst Solids.,338–340, 806 (2004).
24. L. Zhang, J. Li , X.W. Zhang , X.Y. Jiang, Z.L. Zhang, "High-performance ZnO thin
film transistors with sputtering SiO2/Ta2O5/SiO2 multilayer gate dielectric" Thin Solid
Films., 518, 6130 (2010).
25. B.J. Norris, J. Anderson, J.F. Wager, D.A. Keszler, "Spin-coated zinc oxide
transparent transistors", J. Phys. D: Appl. Phys., 36, L105 (2003).
26. B. Sun, H. Sirringhaus "Solution-Processed Zinc Oxide Field-Effect Transistors Based
on Self-Assembly of Colloidal Nanorods", Nano Lett., 5, 2408 (2005).
28
27. S. Lee, Y. Jeong, S. Jeong, J. Lee, M. Jeon, J. Moona, "Solution-processed ZnO
nanoparticle-based semiconductor oxide thin-film transistors", Superlattices
Microstruct., 44, 761 (2008).
28. H. Bong, W.H. Lee, D.Y. Lee, B.J. Kim, J.H. Cho, K. Cho, "High-mobility lowtemperature ZnO transistors with low-voltage operation", Appl. Phys. Lett., 96,
192115 (2010).
29. K. Nomura, H. Ohta, K. Ueda, T. Kamiya, M. Hirano, H. Hosono, " Thin-Film
Transistor Fabricated in Single-Crystalline Transparent Oxide Semiconductor",
Science, 300, 1269 (2003).
30. K. Nomura, H. Ohta, A. Takagi, T. Kamiya, M. Hirano, H. Hosono, "Roomtemperature fabrication of transparent flexible thin-film transistors using amorphous
oxide semiconductors", Nature, 432, 488 (2004).
31. J.H. Jeong, H.W. Yang, J-S. Park, J. K. Jeong, Y-G. Mo, H-D. Kim, J-Song, C.S.
Hwang, "Origin of Subthreshold Swing Improvement in Amorphous Indium Gallium
Zinc Oxide Transistors", Electrochem. Solid-State Lett., 11, H157 (2008).
32. H.Q. Chiang, J.F. Wager, R.L. Hoffman, J. Jeong, D.A. Keszler, "High mobility
transparent thin-film transistors with amorphous zinc tin oxide channel layer ", Appl.
Phys. Lett, 86, 013503 (2005).
33. R.L. Hoffman, "Effects of channel stoichiometry and processing temperature on the
electrical characteristics of zinc tin oxide thin-film transistors", Solid-State Electron.,
50, 784 (2006).
34. D. Hong, J.F. Wager, "Passivation of zinc–tin–oxide thin-film transistors", J. Vac. Sci.
Technol. B., 23, L25 (2005).
35. J. Triska, J.F. Conley, Jr., R. Presley, J.F. Wager, "Bias stress stability of zinc-tinoxide thin-film transistors with Al2O3 gate dielectrics", J. Vac. Sci. Technol. B., 28,
C5I1 (2010).
36. W.B. Jackson, R. Hoffman, B. Yeh, T. Emery, T. Koch, C. McConica, O. Kwon
"Metastability in multicomponent oxide transistors", Phys. Status Solidi A., 207, 695
(2010).
37. Y-H. Kim, H.S. Kim, J-I. Han, S.K. Park, "Solvent-mediated threshold voltage shift
in solution-processed transparent oxide thin-film transistors", Appl. Phys. Lett, 97,
092105 (2010).
29
38. S-J. Seo, S.C. Yang, J-H. Ko, B-S. Bae, "Effects of Sol-Gel Organic-Inorganic
Hybrid Passivation on Stability of Solution- Processed Zinc Tin Oxide Thin Film
Transistors", Electrochem. Solid-State Lett., 14, H375 (2011).
39. N.L. Dehuff, E.S. Kettenring, D. Hong, H.Q. Chiang, J.F. Wager, R.L. Hoffman, C.H. Park, D.A. Keszler, "Transparent thin-film transistors with zinc indium oxide
channel layer", J Appl. Phys., 97, 064505 (2005).
40. P. Barquinha, A. Pimentel, A. Marques, L. Pereira, R. Martins, E. Fortunato,
"Influence of the semiconductor thickness on the electrical properties of transparent
TFTs based on indium zinc oxide", J Non-Cryst Solids., 352, 1749 (2006).
41. C.G. Choi, S-J. Seo, B-S. Bae, "Solution-Processed Indium-Zinc Oxide Transparent
Thin-Film Transistors", Electrochem. Solid-State Lett., 11, H7 (2008).
42. D-H.Lee, Y-J.Chang, G. S. Herman, C-H.Chang, "A General Route to Printable
High-Mobility Transparent Amorphous Oxide Semiconductors", Adv Mater., 19, 843
(2007).
43. S-Y. Han, G. S. Herman, C-H.Chang, "Low Temperature, High-Performance,
Solution-Processed Indium Oxide Based Thin Film Transistors", ECS Transactions,
33, 275 (2010).
44. R.B.H. Tahar, T. Ban, Y. Ohya, Y. Takahashi, "Tin doped indium oxide thin films:
Electrical properties", J Appl. Phys, 83, 2631 (1998).
45. J. Jiang, J. Sun, W. Dou, B. Zhou, and Q. Wan, "In-plane-gate indium-tin-oxide thinfilm transistors self-assembled on paper substrates", Appl. Phys. Lett, 98, 113507
(2011).
46. R.J. Cava, J.M. Phillips, J. Kwo, G. A. Thomas, R. B. van Dover, S. A. Carter, J. J.
Krajewski, W.F. Peck, Jr., J. H. Marshall, and D. H. Rapkine, "GalnO3 - A new
transparent conducting oxide", Appl. Phys. Lett., 64, 2071 (1994).
47. H.Q. Chiang, D. Hong, C. M. Hung, R. E. Presley, J.F. Wager, C.-H. Park, D.A.
Keszler, and G.S. Herman, "Thin-film transistors with amorphous indium gallium
oxide channel layers", J. Vac. Sci. Technol. B, 24, 2702 (2006).
48. G. Goncalves, P. Barquinha, L. Pereira, N. Franco, E. Alves, R. Martins, E.
Fortunato, "High Mobility a-IGO Films Produced at Room Temperature and Their
Application in TFTs", Electrochem. and Solid-State Lett., 13, H20 (2010).
49. J.P.Uyemura, “CMOS Logic Circuit Design”, Kluwer Academic Publishers, 15,
(1999).
30
50. R.L. Hoffman, "ZnO-channel thin film transistors: Channel mobility," J. Appl. Phys.
95, 5813 (2004).
51. D. Hong, G. Yerubandi, H. Q. Chiang, M. C. Spiegelberg, J. F. Wager, "Electrical
Modeling of Thin-Film Transistors", Critical Reviews in Solid State and Materials
Sciences, 33, 101, (2008).
52. D.K. Schroder, "Semiconductor Materials and Device Characterization," New
York: John Wiley & Sons, Inc., 2 ed. (1998).
53. J.F. Conley, "Instabilities in Amorphous Oxide Semiconductor Thin-Film
Transistors", IEEE Transactions on Device and Materials Reliability", 10, 460,
(2010).
54. F.R. Libsch, J. Kanicki, "Bias-stress-induced stretched-exponential time dependence
of charge injection and trapping in amorphous thin-film transistors", Appl. Phys. Lett,
62, 1286, (1993).
55. J.S. Wright, W. Lim, D.P. Norton, S.J. Pearton, F. Ren, J.L. Johnson, A. Ural,
"Nitride and oxide semiconductor nanostructured hydrogen gas sensors",
Semiconductor. Sci. Technol., 25, 1 (2010).
56. G.F. Fine, L.M. Cavanagh, A. Afonja, R. Binions, "Metal Oxide Semi-Conductor Gas
Sensors in Environmental Monitoring", Sensors., 10, 5469 (2010).
57. T. Seiyama, A. Kato, K. Fujiishi, M. Nagatani, "A New Detector for Gaseous
Components Using Semiconductive Thin Films", Anal Chem, 34, 1502 (1962).
58. X.Li, R.Ramasamy, P.K.Dutta, “Study of the resistance behavior of anatase and rutile
thick films towards carbon monoxide and oxygen at high temperatures and
possibilities for sensing applications”, Sens Actuators. B., 143, 308 (2009).
59. N.G.Patel, K.K.Makhija, C.J.Panchal, “Fabrication of carbon dioxide gas sensor and
its alarm system using indium tin oxide (ITO) thin films”, Sens Actuators. B: Chem.,
21, 193 (1994).
60.M.S.Lee, J-U Meyer, “A new process for fabricating CO2 -sensing layers based on
BaTiO3 and additives”, Sens. Actuators. B, 68, 293 (2000).
61.A.A.Tomchenko, G.P.Harmer, B.T.Marquis, J.W.Allen, “Semiconducting metal oxide
sensor array for the selective detection of combustion gases”, Sens. Actuators. B, 93,
126, (2003).
31
62 F.K. Chibirova, A.K. Avetisov, S.A. Kazakov, L.A. Obvintseva, M.V. Strobkova,
"Semiconductor Metal Oxide Sensors in Water and Water Based Biological Systems",
Sensors, 3, 451 (2003).
63. J. Zhou, N. Xu, Z. L. Wang, "Dissolving Behavior and Stability of ZnO Wires in
Biofluids: A Study on Biodegradability and Biocompatibility of ZnO Nanostructures",
Adv. Mater., 18, 2432 (2006).
64. A.D. Mason, C.-C. Huang, S. Kondo, M.T. Koesdjojo, Y.H. Tennico, V.T. Remcho,
J.F. Conley Jr., "Synthesis, functionalization, and environmental stabilization of ZnO
nanobridge transducers for gas and liquid-phase sensing", Sens. Actuators. B, 155,
245, (2011).
65 B. Zhang, T. Kong, W. Xu, R. Su, Y. Gao, and G. Cheng, "Surface Functionalization
of Zinc Oxide by Carboxyalkylphosphonic Acid Self-Assembled Monolayers",
Langmuir, 26, 4514, (2010).
66 F. Schreiber, "Structure and growth of self-assembling monolayers", Prog. Surf. Sci.,
65, 151 (2000).
32
CHAPTER 2 EXPERIMENTAL METHODS
This chapter introduces the various instruments and techniques used in this
research for thin film deposition, processing and characterization.
2.1. Surface treatment
2.1.1 Ultraviolet-Ozone (UV-O3) cleaning
The main purpose of UV-O3 cleaning is to remove organic contaminants from
substrates to achieve high surface energies for subsequent processes. The two major
wavelengths of UV radiation from common low pressure mercury vapor lamps are 184.9
and 253.7 nm. Atmospheric O2 absorbs 184.9 nm UV radiation to form O3 according to
the following reaction [1].
O → O + O
O+O →O
Furthermore, O3 absorbs 253.7 nm UV and decomposes via the reverse reaction. Thus,
during the formation and decomposition of O3 (via UV exposure), atomic oxygen is
generated which then reacts with the sample surface. Simultaneously, organic compounds
on the sample surface exposed to UV radiation photolyze to form ions, free radicals and
excited/neutral molecules which react with atomic oxygen to form common molecules
with high vapor pressures, such as CO2, H2O, N2 and O2, which then desorb from the
surface. The removal of organic contaminants from the surface can then be evaluated
33
using water contact angle (θ) measurements. A clean (high energy) surface will exhibit
small contact angles while a contaminated (low energy) surface will exhibit large contact
angles with water. A schematic of a UV Ozone cleaning system is shown in Figure 2.1.
Aluminum reflector plate
184.9 nm UV rays
253.7 nm
Low pressure mercury lamp
Sample stage
Ozone
Figure 2.1: A sample was irradiated with ultraviolet radiation with wavelengths of 184.9
nm and 253.7nm, to produce O3 inside the apparatus
For this research, a Novascan PSD UVT UV-Ozone system with a flow controller
for oxygen was used. The distance between the substrate and light source is maintained at
~ 5 mm to ensure consistent surface treatments from run to run.
2.2 Thin film processing
2.2.1 Thin film deposition techniques
2.2.1.1 Spin-coating
Spin-coating has been used for several decades for thin film deposition. It is
widely used in micro-fabrication [2], capable of depositing thin films with thicknesses
below 10 nm [3]. One of the main applications for this technique is the deposition of
photoresists for photolithography. Due to the high uniformity of film thickness that can
34
be obtained by this method it is used in a variety of industrial applications to deposit
resins, polymers, epoxy, and sol-gel solutions on metal, glass, ceramic, paper, plastic and
semiconductor substrates.
The theory behind this deposition technique involves the equilibrium between the
centrifugal forces created by the rapid spinning and viscous forces determined by the
viscosity of the solution [4]. The final film thickness and other properties depends on a
variety of solution characteristics including viscosity, drying rate, percent solids and
surface tension. The adjustable parameters for the spinning process include final rotation
speed, acceleration and spinning time.
Spin-coating can be classified into four main stages namely, deposition, spin-up,
spin-off and evaporation, as shown in Figure 2.2. The deposition stage involves
dispensing a small puddle of fluid onto the center of the flat substrate held in position by
vacuum. Two common dispense methods are static dispense and dynamic dispense. For
static dispense, the substrate is stationary while applying the fluid. For dynamic dispense,
the fluid is applied to the rotating substrate. Higher viscosity and/or larger substrates
typically require larger quantities of fluid to ensure full coverage when using a high speed
spin step. After dispensing the fluid, the substrate is accelerated to a relatively high speed
(1500-6000 rpm) where the thickness of the fluid is reduced to near its final value. In this
spin-up stage, the liquid flows radially outward, driven by centrifugal forces. The
substrate is then held at this speed during the spin-off stage, allowing liquid to flow to the
substrate edges and then be removed. As the film thins, liquid run-off slows down and
evaporation of the liquid begins to dominate the film thinning behavior, thereby
35
in
ncreasing thee viscosity and
a concentraation of the ddeposited fillm. It has beeen found
em
mpirically th
hat film thick
kness (t) is in
nversely prooportional too the square-root of the sspin
sp
peed (ω) [4],
=
√
( . )
Thus,
T
higher spin speeds (and longer spin times) lead to the fformation off thinner film
ms.
Figure
F
2.2: A schematic of a typical spin-coatingg process: a)) deposition bb) spin-up c)
sp
pin-off d) ev
vaporation [A
Adapted from
m Seung-Yeeol Han, Ph.D
D dissertatioon,
Oregon
O
Statee University, 2010].
The major
m
advantages of spin coating are its low cost, ease of usee, and
reeproducibilitty for creatin
ng uniform th
hin films onn a variety off substrates. There are
seeveral disadv
vantages of spin
s
coating
g, including tthe limitationn that the process can onnly
be applied to smooth and flat substrattes, the low ccoating efficciency wheree only ~2-5%
% of
th
he dispensed
d material is utilized, and
d the remaindder generatees additionall costs for waaste
disposal, mak
king the overrall cost for the
t spin coat
ating processs 160% that oof the actuall
material
m
conssumed [5]. Thus,
T
it is imp
portant to acccount for w
waste producttion and
asssociated dissposal costs when consid
dering spin ccoating as a technique.
36
2.2.1.2 RF magnetron sputter deposition
Sputtering is a physical vapor deposition process in which the surface of a target
is struck by accelerated positive ions, thereby ejecting atoms, ions and clusters which are
subsequently deposited onto the surface of a substrate, forming a thin film. A schematic
of a RF sputtering process is shown in Figure 2.3.
RF cathode
Dark space shield
Target
Dark space
plasma
Target particles “sputtered”
onto substrate
Deposited thin film
Acceleration of positive ions
towards target
Acceleration of negative ions
towards substrate
Substrate
Figure 2.3: Schematic indicating basic aspects of RF sputtering [Adapted from Eric
Steven Sundholm, Masters thesis, Oregon State University, 2010]
In a typical process, positive and negative ions are generated due to ionization of
gas (usually Argon) when a glow discharge (plasma) is created between substrate and the
target using a radio frequency power supply. The ions then strike the target with enough
energy to eject target atoms and these are then deposited on the substrate. To improve
37
quality and deposition rate of the deposited film, RF sputtering is carried out under
vacuum, typically between 1 and 50 mTorr. Magnetron sources are set behind the target
to generate a magnetic field that can be used to capture secondary electrons close to the
target. Electrons around the magnetic field lines follow helical paths, undergoing an
increasing number of ionizing collisions near the target which in-turn increases the
sputter rate. A more detailed description of plasmas and sputtering can be found in
Lieberman [6].
For our work, an AJA International RF magnetron sputter deposition system was
used. The system used a turbo molecular pump and had a base pressure of ~ 1 x 10-8
mbar. Reactive sputter deposition with a mixture of argon and oxygen gas was used to
deposit thin films for TFT applications.
2.2.1.3 Thermal Evaporator
Thermal evaporation is a physical vapor deposition technique, which is
commonly used to deposit aluminum (Al). This process was used to deposit contacts for
TFTs. The process consists of three basic steps: 1) vaporization of the solid target
material, 2) transport of material to the substrate and 3) deposition onto the substrate
surface. The transformation of solid to vapor could be through sublimation or via melting
followed by evaporation. The heating source and material being deposited dictates the
type of process that occurs for vaporization. The thermal evaporation process is typically
carried out under high vacuum (~ 4 x 10-5 mbar) to reduce contaminant incorporation into
the film and to increase the mean free path of the vaporized molecules for uniform
38
deposition. Although,
A
thiss technique is
i relatively inexpensivee and can obttain high
uitable for deepositing a single elemennt, such as A
Al.
deposition rattes it is typiccally only su
This
T is due to
o the variatio
on in vapor pressure
p
at a given tempeerature for each constituuent
of compound
d materials su
uch as ITO or
o IGZO i.e eeach elemennt will vaporrize at unequual
g to formation
n of films with
w undesirab
able stoichiom
metry. A typpical thermall
raates, leading
ev
vaporator sy
ystem is show
wn in Figuree 2.4. A polaaron model E
E6100 therm
mal evaporatiion
sy
ystem was used
u
for Al deposition.
d
The main chaamber consissts of a bell jjar assemblyy in
which
w
high vaacuum is ach
hieved using
g a diffusion pump. Al inn the form off individual clips
iss placed in a refractory crucible,
c
boaat or a wire bbasket comm
monly made oof molybdennum
(M
Mo) or tungssten (W). Th
he boat or thee wire baskeet is directly heated by ann electric cuurrent
an
nd results in
n aluminum evaporation
e
when a highh enough currrent is appliied.
Figure
F
2.4 : Schematic
S
off a thermal evaporator
e
shhowing the eevaporation source and
substrate plaacement. [Ad
dapted from Eric Stevenn Sundholm, Masters theesis,
Oregon
O
State University
y, 2010]
39
2.2.2 Post-d
deposition
n thermal annealingg
Many
y thin films reequire therm
mal annealingg after depossition to impprove variouus
prroperties, including physical, electro
onic, catalytiic, or opticall. For exampple, the turn--on
voltage and mobility
m
of a TFT devicee is strongly influenced bby post-depoosition anneaaling
nealing is gen
nerally accom
mplished byy controllingg the temperaature
of the channeel layer. Ann
of the substraate, pressure in the furnacce, and comp
mposition of tthe ambient. Thin films ccan
be annealed in
n either inerrt or reactivee environmennts. Commonn reactive ennvironmentss use
eiither an oxid
dizing or red
ducing ambieent. Typicallly, the goal oof thermal annnealing of tthin
fiilms is to ind
duce crystalllization via solid-state
s
diiffusion, thouugh several other changes
may
m also occu
ur, such as generating
g
orr reducing fiilm stress duue to thermall
ex
xpansion/con
ntraction.
Figure
F
2.5: A typical tem
mperature pro
ofile used foor post-depossition annealling of
semiconducto
or thin films.
40
Devices fabricated in this work were annealed using a Thermofisher scientific
model quartz tube furnace. Samples were placed in a ceramic crucible and were inserted
into a quartz tube within the furnace. For inert conditions, a nitrogen flow was utilized
where vacuum was used on the output to control the pressure in the furnace. Otherwise,
samples were annealed in ambient air. The typical temperature profile of a postdeposition annealing process is shown in Figure 2.5. The temperature slope of the
decreasing part of the profile will show a slight difference with respect to the specified
ramp down rate since there is no active cooling implemented in the tube furnace used.
2.2.3 Photolithography
Overview of photolithography
Photolithography is a process used in semiconductor processing in which a
specific pattern from a mask is transferred onto a thin-film layer of photoresist (organic
polymer) via exposure to high intensity short-wavelength light which drives
photochemical reactions in the resist (cross-linking). The resist then goes through a
develop step which leaves the photoresist in the desired locations on the substrate but
removes the photoresist from all others. This process is followed by selective etching to
remove the film from regions not covered by the photoresist. Finally the photoresist is
removed from the substrate leaving the patterned film. There are two basic types of
photoresist: positive and negative photoresist. Positive photoresist transfers an exact copy
of the pattern in the mask to the film so that the exposed photoresist is removed during
the developing process. A negative photoresist transfers an inverted mask pattern to the
film so that the exposed photoresist is not removed while being developed. For this work
41
a negative ph
hotoresist waas used and a stainless steeel shadow m
mask definees the patternns
T channell.
used for the TFT
Figure
F
2.6: Typical
T
photo
olithography
y process steeps (for negaative photoreesist).
Processing
P
stteps in phottolithograph
hy
The photolithograaphy processsing steps us ed in this woork are show
wn in Figure 2.6.
The
T first step was to spin coat the Microchem SU
U-8 2010 neggative photorresist on thee
su
ubstrate wheere rotation speeds
s
of 1000 to 4000 rrpm were used. This wass followed bby an
in
nitial bake att 95 °C [7]. The
T next step
p was to preeferentially eexpose to UV
V light throuugh a
ph
hotomask. This
T was folllowed by a post-exposur
p
re bake at 955 °C. For phootoresist bakking,
a hot plate tem
mperature off 95 °C was used mainlyy to avoid exxcessive craccking,
42
dehydration or
o shrinking of photoresiist. The phottoresist was then prefereentially remooved
by
y soaking in
n developer solution
s
for ~ 1 min. Thee substrate w
was then rinssed with
issopropyl alco
ohol (IPA) and
a blow dried with nitroogen. Finallyy the underlyying unexpoosed
seemiconducto
or material was
w wet etched in a dilut ed acid folloowed by strippping off thee UV
ex
xposed photoresist by ussing removerr PG solutioon.
2.3 Thin fillm characcterization
n
2.3.1 Gonio
ometer forr contact angle
a
measurement
Concept
C
of contact
c
angle
The wettability
w
off a liquid is defined
d
as thhe contact anngle betweenn a droplet of the
liiquid in therm
mal equilibriium on a horrizontal surfface. The shaape of a wateer droplet onn a
so
olid surface is a function
n of the surfaace energy, aand thus the contact angle of the drooplet
on
n a substratee can be used
d to estimatee its surface energy.
Figure
F
2.7: A FTA 32 Co
ontact angle goniometer set up
[h
http://cnst.niist.gov/nanofab/pdf/goniiometer_mannual.pdf]
43
In 1805, Thomas Young discovered that the interfacial effects among solids,
liquids and vapors at the edge of the liquid droplet had significantly different properties
than the bulk core, and could be better characterized by considering the contact angle [8].
A schematic of the goniometer used to measure contact angles is shown in Figure 2.7.
For a liquid droplet sitting on a solid surface, each interface (solid-liquid, solidvapor, liquid-vapor) has a characteristic surface energy density,γSL, γSV and γLV. Under
equilibrium conditions, these interfacial surface energy densities are related to the contact
angle, θ, as follows.
=
−
( . )
where, a contact angle of zero corresponds to complete wetting of the surface while a
non-zero value corresponds to partial wetting of the surface.
Contact angle measurement
A FTA32 goniometer was used to perform contact angle and surface tension
measurements to ensure clean surfaces with good wettability and adhesion. Contact
angles are generally measured by appropriately fitting the mathematical expression
(Equation 2.2) to the shape of the liquid droplet on the surface and then calculating the
slope of the tangent to the liquid drop at the liquid-solid-vapor (LSV) interface line. In
recent years, computer software has been used to better determine the liquid drop shape.
The set up consists of three parts: a digital camera, an illuminator, and image processing
software. For these studies, a single droplet of liquid on the order of a few µL was
44
dispensed with a micropipette and the contact angle was found by connecting several
small lines to fit a circle to the cross section of the drop. This was followed by
determining the intersection with the projection of the surface using the appropriate tools
in the computer software.
2.3.2 Thermogravimetric analysis (TGA)
Concept of thermogravimetric analysis
Thermogravimetric analysis (TGA) is an analytical technique used to monitor the
mass changes that occurs when a sample is heated. This information can then be used to
determine a material's thermal stability, oxidative stability, and fraction of volatile
compounds present. TGA measurements are generally carried out in air, nitrogen or an
inert atmosphere such as helium or argon and the weight change is recorded as a function
of temperature. Because of the nature of the process, this technique requires a high degree
of precision in temperature and mass change measurements.
TGA instrumentation and data analysis
Figure 2.8 shows the basic set-up of a typical TGA instrument. The sample is
loaded into an empty aluminum pan which is pre-weighed inside the TGA system for
reference. The furnace is closed and the temperature is increased at a given ramp rate.
The mass gain/loss with respect to temperature increase may be attributed to a variety of
mechanisms including release of moisture and oxidation/reduction (depending on the
furnace gas used) among others.
45
Figure 2.9 shows a typical TGA curve for our work with Zn nanoparticles. The
mass loss at low temperatures (A) is attributable to the loss of volatile species from the
sample. The mass gain in region (B) is due to oxidation of zinc, and no further weight
increase is seen for region (C), suggesting particles have been completely oxidized.
Balance Assembly
Temperature probe
Sample
Computer
Furnace
Figure 2.8: A Schematic of TGA instrumentation showing a furnace and a sample
assembly. [ Morgan L.Ferguson, Ph.D dissertation, Oregon State University,
2011]
Material inhomogeneity and heating rate could act as potential sources of error for
this technique [9]. TGA experiments were carried out on several samples from the same
batch of material and they had TGA data that were not consistent. It is possible that
impure materials or materials that change over time may lead to variations in the TGA
experiments. For this reason, data was collected for at least three specimens of a
46
particular batch to acquiree reliable daata. Heating rrates used foor TGA meaasurements aare
he range of 10-20 °C/miin. It was obbserved that hheating rate had a
generally in th
prronounced effect
e
on the onset of oxiidation whicch can be attrributed to low rates of heeat
co
onduction fo
or nanomaterrials [9]. Forr unpurified samples, combustion waas observed for
raamp rates at or above 5 °C/min
°
[9].
Figure
F
2.9: A typical TG
GA profile fo
or Zn nanopaarticle oxidattion in air. A
A) Loss of
moisture
m
from
m the surfacee B) Samplee metal is oxiidized in airr C) Sample
metal
m
is comp
pletely oxidiized.
2.3.3 Scann
ning electrron microsscopy (SE
EM)
Overview
O
of SEM
The sccanning elecctron microscope is a verrsatile instruument capablle of high
magnification
m
n imaging by
y scanning a sample withh a beam of high energyy electrons (00.5 –
40 keV). Elecctrons that in
nteract with atoms
a
presennt in the sam
mple producee a variety of
47
signals which can be used to obtain information about the sample morphology, surface
topography, chemical composition and electrical conductivity [10].
Fundamental principles and instrumentation
The major components of an SEM include the electron gun, electromagnetic
lenses, detectors and data output devices. In order to prevent contamination of the
electron gun, and scattering of electrons by molecules, a high vacuum (~10-7 mbar) is
maintained using mechanical, diffusion, ion or turbo pumps. Electrons emitted by the
electron gun are focused through a series of electromagnetic lenses on to a sample
mounted on a mobile stage. When the primary electron beam bombards a sample surface,
energetic electrons penetrate into the sample and interact with matter for a given volume,
which is defined as the interaction volume. This interaction of primary beam electrons
with the sample gives rise to secondary electrons, back-scattered electrons, visible light
and characteristic X-rays which can be collected by various detectors and can be further
processed and displayed in the output device. Of all these methods, imaging via detection
of secondary electrons was primarily used for our studies. A typical secondary electron
image obtained using 3D Quanta Dual Beam SEM system is shown in Figure 2.10. The
sample was imaged using an operating voltage of 15 kV and spot size of 3.0.
Another additional capability that is combined with some SEM instrumentation is
focused ion beam (FIB) milling. For FIB milling samples were mounted on a motorized
five-axis stage inside the chamber. When ions (Ga+ ions in our case) with energy of 1-30
keV strike the surface of the solid sample with high current (~15 nA), they lose energy to
48
the electrons and atoms in the sample [11]. The physical effects of this energy transfer
includes milling (sputtering of neutral and ionized substrate atoms), emission of
electrons, displacement of atoms in the solid and photon emission. When FIB milling is
completed, the samples are polished at a relatively low ion current (~1 nA) to obtain a
smooth and defect free area of interest. The major applications of FIB is for TEM sample
preparation and formation of cross-sections for defect analysis in semiconductor
industries. For our work, we used FIB for characterizing the structure of the films we
have coated and for obtaining thicknesses of the nanostructured films. A secondary
electron image of a ZnO film subjected to FIB milling is shown in Figure 2.11.
Figure 2.10: An SEM image of a ZnO film obtained by sol-gel chemistry.
49
ZnO
Figure
F
2.11: A secondary
y electron im
mage of ZnO
O film subjeccted to FIB m
milling.
2.3.4 Heliu
um ion miccroscopy (HIM)
Helium
m ion microscopy is sim
milar to SEM
M, but it utilizzes helium ioons for imagging
an
nd surface analysis
a
insteead of electro
ons. Helium
m ions have hhigh brightneess and low
en
nergy, allow
wing for a fin
nely focused beam and m
much smallerr interaction volume thann for
ellectrons. Thu
us, HIM is capable
c
of prroducing higgh resolutionn images withh improved
depth of focu
us and good material
m
con
ntrast [12]. A schematic oof interaction volumes inn
SEM & HIM is compared
d in Figure 2.12.
2
The io
on source is a sharp need
dle maintain ed at high vaacuum and ccryogenic
teemperatures.. A sufficien
ntly strong ellectric field iis applied to the needle sso that helium
m
attoms in close proximity are ionized via
v electron tunneling [113]. The resuulting emitteed
positive heliu
um ions are directed
d
dow
wn the micro scope colum
mn through a series of
50
fo
ocusing, alig
gnment and scanning
s
elements. Like SEM, the beam is rasterred across thhe
saample and generates secondary electtrons which can be deteccted to obtain high qualiity
im
mages. An ad
dvantage of HIM over SEM
S
includee the high seccondary elecctron yield ( i.e.
SEM producees 1 secondaary electron for
f each incooming electrron whereas HIM producces
y electrons per
p impingin
ng ion). Alsoo HIM is prefferred comppared to other ion
3-9 secondary
microscopes
m
since it prod
duces less sam
mple damagge due to relaatively light mass of heliium
io
ons compareed to other io
on sources.
Figure
F
2.12: A schematicc of difference in interacction volumee for SEM & HIM.
2.3.5 Transsmission electron
e
microscopy
m
y (TEM)
TEM is a microsccopy techniqu
ue in which a beam of eelectrons is trransmitted
th
hrough an ulltrathin speciimen, wheree the electronn beam interracts with thee material ass it
passes throug
gh [14]. A staandard TEM
M chamber iss evacuated tto low pressuure on the orrder
of 10-6 mbar to
t increase th
he mean freee path of beaam of electroons. The electron sourcee
onsists of a LaB
L 6 cathod
de. Heating of
o the cathodde results in emitted elecctrons whichh are
co
th
hen focused using electro
omagnetic leenses and meetal aperturees. The electtron beam
51
impinges on the sample, and are elastically scattered through the sample. The resulting
electrons that are transmitted through the sample are then magnified and focused on to an
imaging device such as a fluorescent screen which is then collected by a CCD camera.
This technique is capable of much higher resolution than light microscopes due to the
small de Broglie wavelength of electrons, allowing the user to investigate atomic
resolution features within a sample.
For our work, a JOEL TEM 2010 system with 200 kV accelerating voltage was
used. The nanoparticle samples were deposited on carbon coated TEM grids and the
images were obtained with low electron doses to avoid beam induced damage and
potential sample oxidation. A typical TEM image of a zinc nanoparticle used in our
studies with observable atomic fringes is shown in Figure 2.13.
Figure 2.13: A TEM image of an individual Zn nanoparticle showing atomic lattice
fringes
52
TEM can operate in several modes that can allow for the system to observe
electronic structure, crystal orientation, chemical identity of various samples. TEM can
also be used for particle size determination.
2.3.6 Micro-x-ray diffraction (XRD)
Principle of XRD
X-ray diffraction is a technique used to identify atomic structure of materials and
is based on elastic scattering of x-rays from the electron density of planes of atoms in a
material. Diffracted x-rays undergo constructive and destructive interference for
particular incident angles due to the spacing of atomic planes as shown in Figure 2.14.
This intensity is recorded while a variety of angles are scanned, resulting in diffraction
spectra, which can be related to crystal structure.
The spacing of crystalline planes (d) can be related to the angle at which constructive
interference occurs (θ) for a given x-ray wavelength (λ) using Bragg's law [15].
=
( . )
where, n is an integer corresponding to the order of reflection. This d-spacing information
can be used for identification of crystallographic structures. XRD is a fast, nondestructive technique capable of providing information about the crystallographic phase,
orientation, and chemical composition of materials.
The XRD data obtained for our work was collected using a Rigaku D/Max Rapid
II instrument with a 2D image plate detector. X-rays were generated with a MicroMax
53
007HF generaator fitted with
w a rotating
g Cr anode ( = 2.2897 Å
Å) focused oon the specim
men
hrough a 300
0 µm diametter collimaorr. The Zn nan
anoparticle saamples weree applied to oone
th
siide of doublee-sided tape that was plaaced onto a fflat sample hholder, and ffixed onto thhe
saample stage.. The 2DP, Rigaku
R
2D Data
D Processiing Softwaree (Ver. 1.0, R
Rigaku, 20007)
was
w used to in
ntegrate the diffraction rings
r
captureed by the 2-D
D image platte detector.
Figure 2.14: Schematic
S
sh
howing the process
p
of x--ray diffractiion occurringg by construuctive
interference
i
Crystallite
C
siize and com
mposition an
nalysis
The crrystallite sizze in a materiial can be esstimated usinng Scherrer'ss relation [155],
=
( . )
where
w
L repreesents the crrystallite sizee, K is the shhape factor ((~1 for our m
materials), β (rad)
iss the peak wiidth at half of
o maximum
m intensity (F
FWHM) andd θ (deg) is thhe angle betw
ween
54
the incident beam and the normal to the reflecting plane. The analysis of diffraction data
was carried out using JADE 9.1.5 (Materials Data Inc.) and PDF4+ database (ICSD).
Quantitative phase analysis was carried out using JADE’s whole pattern refinement
option using the published crystal structures of minerals related to the expected materials
for this study.
2.3.7 Ellipsometry
Ellipsometry is an optical technique which measures the change in polarization of
a linearly polarized beam of light upon reflection from a thin film specimen as illustrated
in the Figure 2.15. These polarization changes are highly sensitive to the physical
dimensions (e.g. thickness) and optical properties of the film, making the technique very
powerful for determining film thickness and optical properties.
In a variable angle spectroscopic ellipsometer, measurements are taken for a range
of incident angles and wavelengths. Psi (ψ) and delta (Δ) represent the raw measurement
from the ellipsometer and correspond to the change in amplitude and phase of the
polarized light, respectively, and are related by,
=
( )
∆
( . )
where, Rp and Rs are the complex Fresnel reflections for the p- and s- directions,
respectively [16].
55
Film thickness,
t
reefractive indeex, surface rroughness, annd other useeful propertiees of
d by constru
ucting modells relating thhe raw data too properties of
a sample can be estimated
th
he material. For
F examplee, the refracttive index off a transparennt material ccan be modeled
using the Cau
uchy relation
nship,
Figure
F
2.15 :Ellipsomete
:
er set-up for measuremennt of thickneess and opticcal constantss.
( ) =
+
+
( . )
where
w
n repreesents the reffractive indeex of the matterial,
reprresents waveelength of ligght
an
nd A, B and C are constaants [16]. Th
hus, by fittinng a model too experimenntal data by
ad
djusting the constants, one is able to
o approximatte the film reefractive inddex. The moddel
caan also be ussed to fit oth
her material properties
p
siimultaneouslly, for exam
mple the film
th
hickness.
56
For our work, a J.A.Woolam variable angle spectroscopic ellipsometer (VASE)
system with a rotating analyzer configuration was used. A Cauchy model was employed
to determine the thickness of the oxide films. Data was collected after deposition of each
layer in the device in order to add the known thickness for use in modeling to determine
thickness of the new layer.
2.3.8 X-ray photoelectron spectroscopy (XPS)
Principle of XPS
X-ray photoelectron spectroscopy is a quantitative surface spectroscopic
technique which provides information about structure and chemical composition of the
surface of a sample [17]. The two common x-ray sources used are Al Kα (1486.6 eV) and
Mg Kα (1253.6 eV). XPS spectra are obtained by irradiating a material with x-rays and
measuring the quantity and kinetic energy of electrons that escape from the top 1-10nm
of the material. XPS must be performed under ultra-high vacuum (UHV) in order to
minimize electron scattering before detection. A typical XPS process is illustrated in
Figure 2.16. An ejection of photoelectron from an inner shell of an atomic orbital when it
interacts with a photon is also depicted in the figure.
Analysis and quantification
The kinetic energy of an ejected electron (Ek) can be related to its binding energy
(Eb) as follows, =
−
− ( . )
57
where,
w
hυ is the
t energy of the photos and Φ is thee work functtion of the sppectrometer
which
w
is consstant for a given XPS insstrument [188].
For th
he XPS data analysis, thee peaks of biinding energgy versus intensity were fitted
using XPSPE
EAK 4.1 softtware. The main
m peak fittting parameeters include the positionn of
th
he peak (bind
ding energy)), area underr the curve, ffull width at half maxim
mum (FWHM
M)
an
nd % Lorenttzian-Gaussiian (shape off curve). Thee peaks weree fitted with a linear
background.
Figure
F
2.16: A typical XP
PS illustratio
on (left) andd a process thhat describess an ejectionn of
electron (ph
hotoelectron
n) from an innnermost sheell when inteeracting withh a
photon
2.4 Electriccal characcterization
n
2.4.1 Four-po
oint probe
A four point probee is an apparratus used foor measuringg the resistivvity of conduuctive
nductor) sam
mples. Substraate/film resi stivity can bbe measured by passing
(aand semicon
58
current through two outer probes and measuring the voltage through the two inner probes.
A Jandel model RM2 four point probe was used for our work. The resistivity of a film
can be computed using the measured voltage with respect to the applied current using the
following relation
( ) =
∗ ∗
( . )
( )∗
where t is the thickness of film in cm, I is the applied current in µA and V is the measured
voltage in mV. A schematic of a typical 4-point probe is shown in Figure 2.17.
I
V
L
L
L
L – Probe spacing
Substrate/film
Thickness, t
Figure 2.17: A typical representation of a four-point probe
2.4.2 Semiconductor parameter analyzer (SPA)
All of the electrical characterization of TFTs presented in this thesis used a probe
station with Agilent 4155C precision semiconductor parameter analyzer. This instrument
59
acquires measured data using four source monitor units (SMU's) (-40 V ≤ Vsource ≤ 40V),
(1 fA ≤ Isource ≤ 100mA). Three DC probes are used with tungsten tips, one each for the
gate, source and drain contacts. This instrument was used for electrical characterization
of our TFT devices by measuring the transfer characteristics, where we were then able to
extract the turn-on-voltage, mobility and drain current on-to-off ratio from the data. Bias
stress stability measurements on the TFTs were obtained with this instrument as well.
60
2.5 REFERENCES
1. J.R. Vig, J.W.L. Bus, "UV/Ozone cleaning of surfaces", IEEE Transactions on Parts,
Hybrids and Packaging, PHP-12, 365 (1976).
2. S.Franssila, “Introduction to micro-fabrication”, John Wiley & Sons, Ltd., 2nd edition,
(2010).
3. S. Middleman, “The effect of induced airflow on the spin coating of viscous liquids”,
J. Appl. Phys. 62, 2530 (1987).
4. D. Mitzi, "Solution processing of inorganic materials", New Jersey : John Wiley &
Sons, Inc., (2009).
5. F.E.H.Tay, “Materials & process Integration for MEMS”, Kluwer academic
publishers, (2002).
6. M.A. Lieberman, A.J. Lichtenberg, "Principles of plasma discharges and
materials processing", New York: John Wiley and Sons Inc., first ed., (1994).
7. Microchem, "SU-8 2000, permanent epoxy negative photoresist",
www.microchem.com.
8. J.C. Berg, “Wettability”, New York: Basel and Hong Kong: Marcel Dekker Inc.,
(1993).
9. Thermogravimetric analysis - www.scribd.com/doc/13729612/TGA.
10. L. Reimer, "Scanning Electron Microscope-Physics of image formation and
microanalysis", New York : Springer-Verlag Berlin Heidelberg, 2nd edition, (1998).
11. S. Reyntjens, R. Puers, "A review of focussed ion beam applications in microsystem
technology", J. Micromech. Microeng. 11, 287, (2001).
12. J. Morgan, J. Notte, R. Hill, B. Ward, "An introduction to Helium ion microscope",
Microscopy Today, 14, 24, (2006).
13. N. Economou, B. Ward, J. Morgan, J. Notte, "Helium ion microscopy: an
introduction", Innovations in Pharmatceutical Technology, 24, (2008).
14. B. Voutou, and E.C. Stefanaki, "Electron Microscopy: The Basics", Physics of
Advanced Materials Winter School, 1, (2008).
15. I. Chorkendorff, and J.W. Niemantsverdriet, "Concepts of Modern Catalysis and
Catalysis", WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, 2nd Edition, (2007).
61
16. J.A.Woollam, B.Johs, C.M.Herzinger, J.Hilfiker, R.Synowicki, and
C.L.Bungay,“Overview of Variable Angle Spectroscopic Ellipsometry (VASE), Part I:
Basic Theory and Typical Applications”, Critical Reviews of Optical Science and
Technology CR72, 3 (1999).
17. D.A. Skoog, F.J. Holler, and S.R. Crouch, "Principles of instrumental analysis",
Thompson Brooks/cole, 6th edition, (2007).
18. J.W. Robison, E.M.S. Frame, and G.M. Frame II, "Undergraduate instrumental
analysis", New York : Marcel Dekker, 6th edition, (2005).
62
CHAPTER 3 - FORMATION AND CHARACTERIZATION OF ZINC OXIDE
FILMS USING ZINC NANOPARTICLE DISPERSIONS
3.1 INTRODUCTION
Zinc Oxide (ZnO) is an economical and environmentally benign transparent wide
bandgap (3.37eV) semiconductor with a number of interesting chemical, electrical,
piezoelectrical, and optical properties making it uniquely suited for a variety of thin-film
device applications. For example, ZnO has been used in gas sensors,2,3 photodiodes,4
varistors,5 piezoelectric devices,3 and thin film transistors (TFT).1,6 Typically ZnO films
are deposited by vacuum-based methods which include sputter deposition,7,8
evaporation,9,10 chemical vapor deposition,11 and atomic layer deposition.12 To produce
films at potentially lower costs, efforts have focused on solution-based methods,
including sol-gel13,14 and oxide nanoparticle dispersions.1,6,15
Recently, research has focused on using solution-processed ZnO as the active
material for TFTs. ZnO nanoparticle approaches have been investigated by several
groups, where stable dispersion of ZnO nanoparticles have been deposited and
subsequently annealed.16 For these devices it was necessary to anneal to 500 °C in air to
remove surface impurities including hydroxyls, and to obtain good device performance
where mobilities as high as 0.104 cm2/Vs were obtained. One issue with the nanoparticle
approach is that high temperatures are typically necessary to increase grain size and
density of the films, and this may limit its applicability for low-temperature processing
for flexible electronics on polymeric substrates. Meyers et al. have recently produced
63
dense, high quality polycrystalline ZnO films by inkjet printing and spin-coating for the
channel layer in TFTs.17 It was found that large grain ZnO films were formed after postannealing the printed films to 300 °C or spin-coated films to 150 °C. The resulting
devices had good electronic properties with incremental mobilities of 4-6 cm2/V sec,
while the spin- coated films had incremental mobilities of 1.8 cm2/V sec. Cheng et al.
have fabricated bottom gate TFTs with a ZnO channel layer produced using chemical
bath deposition from a zinc nitrate precursor at 60 °C which was then dried at 100 °C
with no further annealing.18 These devices had a channel mobility of 0.248 cm2/V sec.
Another interesting property of ZnO is that numerous methods have already been
demonstrated on how to form ZnO nanostructures with different morphologies such as
nanorods,1,15,19 nanoneedles,20,21 nanowires,3,20,21 and nanobelts.3,22 Recently it has been
demonstrated that nanostructured ZnO films can be formed by the thermal oxidation of
Zn nanoparticle films.7,9 For example, Kim et al. deposited Zn nanowires by RF
magnetron sputtering and fabricated coaxial Zn/ZnO nanocables by thermal oxidation
between 100-400 °C. These studies indicated that controlled thermal oxidation of Zn
nanowires results in substantial variation in the Zn/ZnO core-shell structures.7 Alivov et
al. reported that thermal oxidation of Zn films, deposited by electron-beam evaporation,
can have significant impact on film properties including roughness, resistivity, mobility
and luminescence.9 It is fairly common to find oxide shells on the surface of metal
nanoparticles, and for Zn the ZnO shell can significantly influence the nanoparticle's
electrical, chemical, and physical properties. Furthermore, it has been shown that the ZnO
shell can strongly influence the oxidation of Zn nanoparticles.23 For example, Gui et al.
64
have investigated the formation of nanostructures on Zn nano- and microparticles by a
two-step oxidation process where pre-oxidation with H2O2 at different Zn/H2O2 ratios
was followed by thermal oxidation.23 In the case of nanoparticles, the melting and
oxidation of zinc led to the formation of nanostructures which occured simultaneously
during the thermal oxidation step. In the case of microparticles, the oxidation process
started only after melting of the zinc occured.
Although there have been a number of studies investigating the formation of
nanostructured ZnO by annealing Zn films, there has been limited characterization of the
relative compositions of Zn and ZnO during annealing Zn nanoparticle films in air.
Furthermore, no electrical characterization of ZnO test structures formed by the oxidation
of metallic Zn films has been reported. In this chapter we present results on the formation
of ZnO films via the thermal oxidation of spin-coated zinc nanoparticle films where the
nanoparticles had vendor specified average diameters of 35 and 130 nm. The Zn
nanoparticles were characterized using thermogravimetric analysis (TGA), while the Zn
nanoparticle films were characterized using helium ion microscopy (HIM) and x-ray
diffraction. Finally, the Zn nanoparticles were incorporated into zinc indium oxide (ZIO)
solutions which were then used to fabricate test structures to evaluate the potential of
forming nanostructured field-effect sensors using simple solution processing.
3.2 EXPERIMENTAL METHODS
Zn nanoparticles with vendor specified average diameters of 35 and 130 nm were
purchased from Nanostructured and Amorphous Materials Inc. Particles were stored in an
65
inert environment prior to formation of dispersions to prevent the nanoparticles from
incidental oxidation by minimizing exposure to oxygen (O2) and water vapor (H2O) from
the ambient.
Characterization of the commercial Zn nanoparticles
A TA Instruments modulated TGA 2950 was used for thermogravimetric analysis
of the 35 nm and 130 nm Zn nanoparticles. Analysis was performed using an air flow rate
of 60 cm3/min. In each experiment, ~25 mg of particles were loaded into aluminum
sample pans, heated at a rate of 5 °C/min up to 150 °C, and held isothermally for 10 min.
The samples were then heated at the same rate up to 600 °C and again held isothermal for
10 min.
A Rigaku D/Max Rapid II micro-x-ray diffraction (micro-XRD) system was used
to characterize the crystallographic phase and composition ratio of Zn:ZnO. The microXRD has a rotating Cr anode (λ = 2.2897 Å) with an operating power of 875 W, where a
0.3 mm collimator was used to define the analysis area.
Preparation of zinc nanoparticle dispersions
Stable dispersions of ~5 wt% were prepared by adding 1.0 g of Zn nanoparticles
into 1:3 volume ratio of methanol:chloroform. 15 0.35 ml of either n-octylamine or nethylmethylamine was added as a dispersant to stabilize the nanoparticle solution. The
solutions were sonicated using a Branson digital sonifier 450 model ultrasonic probe.
Typically the sonifier was operated at 400W and 50 % amplitude for ~30 min to break up
nanoparticle agglomerations. During ultrasonication, the vials were placed in an ice bath
66
to minimize heating of the solutions and evaporation of volatile solvents. The resulting
dispersions were found to be stable for several days.
To form semiconducting zinc indium oxide (ZIO) films with embedded Zn
nanoparticles we initially formed ZIO precursors using indium chloride (InCl3) and zinc
chloride (ZnCl3) dissolved in acetonitrile (0.025 M InCl3/ZnCl3).24 Ethylene glycol (EG)
was then added to this solution in a 1:50 EG:acetonitrile volume ratio. The EG helps
increase solution viscosity and decrease evaporation rate, enabling formation of uniform
thin films.25 Finally, Zn nanoparticles were added to this solution to form a Zn loading of
5 wt%.
Deposition, oxidation and characterization of films
The prepared 35 nm and 130 nm Zn nanoparticle dispersions were spin-coated at
500 rpm for 30 sec on thermally oxidized Si substrates (100 nm SiO2). Prior to coating,
the substrates were O2 plasma treated to remove residual organics and then exposed to
hexamethylene disilazane (HMDS) vapor to create a hydrophobic surface prior to
coating. 15 Film thicknesses from 35 nm and 130 nm particle dispersions were ~1.5 µm
and ~2.5 µm, respectively, as determined by profilometry and scanning electron
microscopy (SEM) cross-sections. These films were then annealed in a tube furnace for 1
hour in air at temperatures between 100-600 °C with 100 °C steps.
Images of the un-annealed films were obtained using HIM to investigate the film
morphology and to analyze the nanoparticle size distribution. Using ImageJ 1.45I
67
software, the diameter of the nanoparticles were determined to be 162 ± 86 nm and 234 ±
112 nm for the vendor specified 35nm and 130 nm, respectively.
Micro-XRD analysis was performed to determine crystallographic phase and the
relative composition of the Zn nanoparticles (e.g., Zn metal and ZnO) using Jade 9.1.5
(Materials Data, Inc.) software. The resistivities (ρ) of the annealed films were measured
using a Jandel model RM2 four point probe where the current, I, was set to 0.1µA and
voltage, V (mV), was measured to determine the resistivity with respect to thickness, t
(cm), of the films using the relation
=
∗ ∗
( . )
( )∗
A set of TFT test structures were fabricated with channel layers formed by spincoating ZIO that did not contain Zn nanoparticles followed by the 5 wt% solutions of
IZO with both Zn nanoparticle sizes. The spin-coating was carried out at 3000 rpm for 30
sec on Si/SiO2 substrates. Prior to spin-coating, the substrates were made hydrophilic by
ultrasonicating in a 1 M NaOH solution. Film thickness was ~600 nm as determined by
profilometry. These films were annealed to 400, 500, and 600 °C in air for two hours. For
these samples patterning was performed using photolithography, where a 0.5 mM HCl
etch was used for the Zn nanoparticles followed by an oxalic acid etch to pattern the
underlying ZIO layer. The TFT test structures were completed by evaporating 500 nm Al
through a shadow mask to form source-drain contacts. A schematic of the test structure is
shown in Figure 3.1. Another batch of TFT test structures with only spin-coated ZIO
68
solution processed under similar conditions were used as a reference for comparing the
electrical characteristics between the devices fabricated for this work.
Figure 3.1: Structure of a bottom-gate thin film transistor (TFT) test structure in which
the Zn nanoparticles are deposited by spin-coating and then oxidized to ZnO
by annealing in air.
3.3 RESULTS AND DISCUSSIONS
In Figure 3.2 we show SEM images of the 162 and 234 nm Zn nanoparticle films
formed on Si/SiO2 substrates. As can be seen, there is a wide size distribution of
69
nanoparticles and the particles are seen to aggregate. Small rumple structures and cavities
can be seen, which are primarily responsible for the large specific surface area making
the particles quite active and susceptible to oxidation. The 234 nm nanoparticles are seen
to have a larger general range of particle sizes than the 162 nm nanoparticles and appear
to be more porous and less densely packed as well.
(a) Zn davg= 162 nm
2 µm
(b) Zn davg= 234 nm
2 µm
Figure 3.2: SEM images showing size distribution of zinc nanoparticles with estimated
average diameters of (a) 162nm (b) 234nm.
It has been proposed that the Zn core is typically surrounded by a transparent
layer of ZnO formed by native oxidation in air and that during annealing, Zn from the
core diffuses to the outer surface of the nanoparticles, while oxygen diffuses from the
surface to the core of the nanoparticles. 26,27,28,29 This process effectively increases the
ZnO shell thickness until all of the Zn atoms in the core have reacted and only a hollow
particle of ZnO exists.
70
HIM images of the spin-coated Zn nanoparticle films annealed at different
temperatures are shown in Figure 3.3 a-n. Films formed from 162 nm nanoparticles had
minimal structural changes for annealing temperatures below 400 °C, but at this
temperature and above coalescence of the nanoparticles occurs. Furthermore, significant
structural changes were observed for films annealed to 500 and 600 °C, where a variety
of morphologies are seen including various nanostructures including ribbons and rods.
We found that the larger particles had structural changes beginning at 300 °C, and
showed similarly variable morphology for anneal temperatures 400 °C and higher,
including the formation of nanostructures such as nanoribbons, nanoneedles20,21 and
nanorods.1,15,19 However, these nanostructures were formed only in specific regions of the
films which may be related to regions with initially larger Zn nanoparticles.
Prior studies have proposed that nanorod formation from the oxidation of Zn
nanoparticles is related to the Zn/ZnO core-shell structure, where large stresses are
created for increasing temperature due to the large difference in thermal expansion
coefficients of Zn (60.8 x 10-6/degree) and ZnO (4 x 10-6/degree).28 This results in a large
amount of stress at high temperatures which causes the ZnO outer shell to fracture. At
these high temperatures the evaporation of metallic zinc can occur through these
fractures, and this Zn vapor can be rapidly oxidized by air resulting in the formation of
the ZnO nanostructures.
71
Figure
F
3.3: HIM
H images of Zn nanop
particles, davgg = 162nm (aa-g) and davgg = 234nm (h
h-n)
films
fi
that hav
ve been spin
n-coated and annealed beetween 100-6600°C
72
125
Zn
Zn davg162nm
davg= 162nm
Zn
Zn 234nm
davg= 234nm
Mass Gain (%)
120
115
Zn oxidation
119.4%
115.1%
110
105
100
95
0
100 200 300 400 500 600 700
Temperature (°C)
Figure 3.4: TGA graph showing oxidation of 162nm and 234nm diameter Zn
nanoparticles.
In Figure 3.4 we show thermogravimetric data for 162 and 234 nm Zn
nanoparticles to identify the onset of oxidation, the temperature at which the
nanoparticles are completely oxidized, and their respective mass gains. In the first step,
Zn nanoparticles were heated in air to 150 ºC at a ramp rate of 5 °C/min and held at 150
ºC for 10 min to remove moisture. The data indicates that there was no mass change
during this step. Particles were then heated to 600 °C at the same ramp rate, and a
significant mass increase was observed. This mass increase can be attributed to the
oxidation of the Zn nanoparticles, with the onset of oxidation occurring at ~250 °C for
73
both sizes of nanoparticlees, but with much
m
higherr mass increaase rates (~22x) being
ob
bserved for the
t larger paarticles. Duee to their highh surface/voolume ratio, nnanoparticlees of
both sizes weere completely oxidized by
b ~500 ºC. TGA resullts indicated a mass gainn of
15.1 and 19.4
4 % for the 162 and 234 nm particless, respectively. These vaalues can be
co
ompared to the
t expected
d mass gain for
f Zn to ZnnO, which is 24.5%, sugggesting that tthe
sm
maller zinc nanoparticle
n
s initially haave a larger rrelative ratioo of ZnO:Zn compared to the
laarger particlees. This is no
ot unexpecteed since the nnative oxidee thickness shhould be
in
ndependent for
f the particcle sizes und
der investigaation, and larrger particless will have a
laarger volumee of Zn comp
pared to the volume of thhe ZnO shelll.
74
Figure
F
3.5: X-ray
X
diffracction plots off intensity vss 2θ for (a) Z
Zn nanopartiicles, davg =
162nm and (b
b) Zn nanoparticles, davgg = 234nm based on depoosited films and
films
fi
annealeed from 100 °C-600 °C.
In Fig
gure 3.5 a an
nd b we show
w XRD specttra obtained for Zn nanooparticle film
ms
an
nnealed betw
ween100-600
0 °C. The peeaks at 2θ = 48, 52.12 annd 55.12° coorrespond to the
(1
100), (002) and
a (101) plaanes of the hexagonal
h
ZnnO while thee peaks at 2θθ = 55.12, 599.48
an
nd 66.36° co
orrespond to the (002), (100) and (1001) planes off hexagonal metallic Zn.
XRD
X
patternss from unann
nealed Zn fillms are dom
minated by peeaks corresponding to
metallic
m
Zn, but
b have low
w intensity peeaks correspponding to ZnnO, suggestiing the preseence
75
of a native oxide shell on the Zn nanoparticles. Peaks corresponding to ZnO are less
intense for annealing temperatures up to 300 °C, before gradually increasing for
temperatures between 300-500 °C and becoming quite prominent for 400 and 500 °C, as
shown by the intense peak corresponding to the (101) plane of ZnO. Likewise, peaks
corresponding to Zn show high intensity for annealing temperatures up to 300 °C,
decreasing for 400 °C, and ultimately disappearing for 500 °C, as shown by the intense
peak corresponding to the (101) plane of Zn metal. The peak at 2θ = 88.72° in the XRD
plot for larger nanoparticle films annealed to 600 ºC corresponds to Si (311) peak from
the substrate. These data indicate the complete oxidation of Zn to ZnO occurs between
400 to 500° C. Furthermore, films formed from 162 and 234 nm Zn nanoparticles had
similar behavior, but with slightly varying relative intensities.
Quantitative phase analysis was carried out using JADE 9.1.5 analysis software
where the whole pattern refinement option was used with the published crystal structures
of Zn (PDF #01-071-3764) and ZnO (PDF #01-074-9939) to estimate the relative
compositions of Zn and ZnO in the films. The gradual decrease in Zn composition and
accompanying increase in ZnO composition for increasing annealing temperature is
shown in Figure 3.6. The weight percent Zn:ZnO ratio for room temperature films was
determined to be 70.1%:29.9% for the smaller nanoparticles and 76.9%:23.1% for larger
nanoparticles indicating that films with larger nanoparticles have a larger metallic
component than smaller nanoparticles, which is consistent with the TGA analysis
described above. Furthermore, Zn was not detectable in the films with smaller
nanoparticles annealed to 500 °C, whereas in the films with larger nanoparticles, Zn was
76
sttill detectablle but was beelow 1% at 500
5 °C. How
wever, a highher rate of oxxidation wass
ob
bserved with
h the larger nanoparticle
n
e films betweeen 200-300 °C, during w
which the ZnnO
co
omposition increased
i
fro
om 29.8 to 65.1%
6
compaared to the loower increasse from 40.00% to
60.9% that was observed for smaller nanoparticlee films in thee same tempperature rangge.
Again,
A
the diffferent relatiive rates of oxidation
o
forr the 162 andd 234 nm naanoparticles aare
in
n good agreeement with th
hose observeed using TG
GA. The % coomposition oof ZnO at eaach
teemperature was
w taken fro
om the micro
o-XRD analyysis and useed to estimatte the % masss
gain for comp
parison to th
he TGA data. The mass ggains were estimated to bbe 10.0 % annd
14
4.3 %, respeectively, whiich compares reasonablyy well with thhe experimeental mass gaains
determined frrom the TGA
A data.
Figure
F
3.6: ZnO
Z weight % compositiion as determ
mined by miicro-XRD foor Zn
nanoparticles
n
s, davg = 162 nm and Zn nnanoparticlees, davg = 2344nm films affter
annealing
a
to the
t indicated
d temperaturres in air.
77
From this data wee have estimaated the thiccknesses of thhe ZnO shellls for varyinng
teemperatures in Figure 3.7. For these calculationss, the particlees are assum
med to be
sp
pherical, wh
here the initiaal diameters of the particcles, as obtaiined by HIM
M, are taken tto be
162 nm and 234
2 nm. These values incclude both thhe diameter of the Zn coore and twicee the
hickness of the
t ZnO shelll. When the weight % vvalues from tthe XRD datta is combined
th
with
w the partiicles diameteers it can be determined that the smaaller particles have a Zn core
diameter of 123 nm and a ZnO shell thickness
t
off 19nm. Likeewise, the larrger
nanoparticles have a Zn core
c
diameteer of 190 nm
m and a ZnO shell thickneess of 22 nm
m. To
esstimate the thickness
t
of the ZnO sheell for the annnealed sampples we usedd the changess in
weight
w
% for the Zn and ZnO,
Z
obtained using Miicro-XRD, affter annealinng to the various
teemperatures.. This data provides
p
both
h the volum
me of the Zn ccore and the ZnO shell,
which
w
can theen be converrted to the th
hickness of thhe ZnO shelll.
Figure
F
3.7: Calculated
C
th
hicknesses off ZnO outer shells for Znn nanoparticcles, davg =
162nm and Zn
Z nanoparticcles, davg = 2234nm versuus temperaturre for films
annealed
a
in air.
a
78
Film resistivity measurements using a 4-point probe are summarized in Table 3.1.
Films formed from the 162 nm Zn nanoparticles had a measured resistivity of ~11 Ω cm
after annealing to 300 °C, and increased to ~ 350 - 650 Ω cm after annealing to 400-600
°C. In contrast, films formed from the 234 nm nanoparticles had resistivity values of
~105 Ω cm after annealing to 300 °C, and increased to nearly 1,100 Ω cm after annealing
to 500 and 600 °C. The reason for the low values of resistivity for 300 and 400 °C is
likely due to the presence of a Zn core, whereas films annealed to 500 °C and 600 °C are
almost completely oxidized (as indicated by the micro-XRD data) resulting in higher
resistivities.
Temperature °C
Resistivity (Ω cm)
Zn davg= 162nm Zn davg= 234nm
300
11.0
105.2
400
355.2
373.7
500
574.0
1059.7
600
658.8
1093.5
Table 3.1: Resistivity measurements of Zn nanoparticles, davg = 162 nm and Zn
nanoparticles, davg = 234 nm films spin-coated using dispersions made up of
chloroform/ methanol and n-octylamine as a dispersant and after annealing at
300 °C, 400°C, 500°C & 600°C in air.
79
Thin film transistor test structures were prepared with spin coated ZIO films, as
well as with films consisting of one layer of 0.025 M ZIO followed by an additional layer
of ~5 wt% zinc nanoparticle solution using either 162 or 234 nm Zn nanoparticles
dispersed in 0.025M ZIO. The devices fabricated without nanoparticles were used as a
reference. The ZIO devices were annealed to 400, 500 and 600 °C and the transfer
characteristics of the resulting devices are shown in Figure 3.8 a, where the gate voltage
(VGS) was scanned from -40 V to 40 V and back to -40 V with a fixed drain voltage (VDS)
of 1 V. The ZIO devices annealed to 400, 500 and 600 °C had turn on voltage (VON) of 7, 0 and -17 V with Ion/Ioff ratios of ~106, 107 and 105. The average mobility (µavg) were
log (ID/(W/L) (A))
determined to be 1.72, 1.78 and 0.71 cm2/V sec, respectively.
-3
-4
-5
-6
-7
-8
-9
-10
-11
-12
-13
(a)
400 °C
500 °C
600 °C
-40 -30 -20 -10 0 10
VGS (V)
20
30
40
log (ID/(W/L) (A))
80
-3
-4
-5
-6
-7
-8
-9
-10
-11
-12
-13
(b)
400 °C
500 °C
600 °C
-40 -30 -20 -10
0
10
20
30
40
-40 -30 -20 -10 0 10
VGS (V)
20
30
40
log (ID/(W/L) (A))
VGS (V)
-3
-4
-5
-6
-7
-8
-9
-10
-11
-12
-13
(c)
400 °C
500 °C
600 °C
Figure 3.8: Electrical characterization of TFTs deposited and annealed to 400°C, 500°C
& 600°C by spin coating (a) ZIO films (b) a bottom ZIO layer with a 5%
solution of davg = 162 nm Zn dispersed in ZIO. (c) A bottom ZIO layer with a
5% solution of davg = 234 nm Zn dispersed in ZIO.
81
Transfer characteristics, with the same VGS range and VDS, were obtained for
devices that were formed with ZIO precursor containing Zn nanoparticles after a 400 °C
anneal. For these devices we observed a combination of both conducting characteristics
associated with Zn nanoparticles and semiconducting characteristics of the underlying
ZIO layer, as shown in Figure 3.8 b and 3.8 c. For negative VGS the channel conductance
is dominated by the Zn nanoparticle films, likely due to the large thickness of the films
and the relatively low resistivities of the Zn nanoparticle films. For positive VGS the 400°
C films are observed to have field enhanced conduction, likely due to the semiconducting
ZIO layer. Both nanoparticle based devices exhibited VON of ~ -10V which is in good
agreement with their corresponding reference devices. However, the Ion/Ioff ratio was in
the range of 101 to 102, much lower than that of the reference ZIO devices. The average
mobilities on these devices could not be accurately calculated due to their high channel
conductance. The I-V characteristics were also obtained for devices with nanoparticle
films annealed to 500 and 600 °C. In the case of films with 162 nm nanoparticles, the
drain current versus gate voltage was nearly a straight line in the resistive region for both
500 and 600 °C annealed devices. This suggests that the flow of current in the channel is
dominated by the relatively low resistivity of the ZnO films, which was confirmed by the
associated four-point probe measurements. The films with 234 nm had similar
characteristics. Since there was no defined Ion/Ioff characteristics observed, the VON,
average mobility and Ion/Ioff ratio could not be determined.
82
3.4 CONCLUSIONS
In this chapter, we reported on a novel method for obtaining nanostructured ZnO
films via thermal oxidation of spin-coated metallic Zn nanoparticles. Thermal analysis
indicates that the onset of oxidation for the Zn nanoparticles occurs at ~250 °C. Weight
gains determined by TGA corresponding to oxidation of particles were 15.1% and 19.4%
for 162 and 234 nm Zn nanoparticles, respectively. HIM imaging showed formation of a
variety of nanostructures for increasing annealing temperatures, particularly for larger
particles. This is likely due to the fracture of the ZnO shell and rapid oxidation of the Zn
metal vapor that passes through these defects. Micro-XRD analysis of the films indicates
the existence of both Zn and ZnO phases for samples annealed at lower temperatures,
with increasing intensity of ZnO diffraction peaks and decreasing intensity of Zn
diffraction peaks for increasing temperature until the films are completely oxidized at
~500°C. Likewise, micro-XRD results were used to estimate the relative compositions of
metallic Zn and ZnO of the films, which compared reasonably with TGA analysis.
Resistivity measurements indicate that the films are highly resistive. However, electrical
characterization of the TFTs fabricated with Zn nanoparticle and ZIO precursors were
dominated by the conducting nature of the thick Zn nanoparticle films.
83
3.5 REFERENCES
1
B. Sun, and H. Sirringhaus, Nano Letters. 5, 2408 (2005).
2
B. L. Zhu, C. S. Xie, A. H. Wang, J .Wu, R. Wu, and J. Liu, J. Mater. Sci. 42, 5416
(2007).
3
Z. L. Wang, Appl. Phys. A. 88, 7 (2007).
4
L. Luo, Y. Zhang, S. S. Mao, and L. Lin, Sensors and Actuators A 127, 201 (2006).
5
S. A. Shojaee, M. M. Shahraki, M. A. F. Sani, A. Nemati, and A. Yousefi, J. Mater. Sci :
Material Electron. 21, 571 (2010).
6
S. Lee, Y. Jeong, S. Jeong, J. Lee, M. Jeon, and J. Moon, Superlattices and
Microstructures 44, 761 (2008).
7
S. Kim, M. C. Jeong, B-Y. Oh, W. Lee, and J. M. Myoung, Journal of Crystal Growth
290, 485 (2006).
8
T. Minami, H. Nanto, and S. Takata, Appl .Phys. Lett. 41, 958 (1982).
9
Ya. I. Alivov, A. V. Chernykh, M. V. Chukichev, and R. Y. Korotkov, Thin Solid Films
473, 241 (2005).
10
W. S. Khan, C. Cao, G. Nabi, R. Yao, and S. H. Bhatti, Journal of Alloys and
Compounds 506, 666 (2010).
11
J . J. Wu, and S. C. Liu, Adv. Mater. 14, 215 (2002).
12
S. J. Lim, S. Kwon and, H. Kim, Thin Solid Films 516, 1523 (2008).
13
Y. S. Kim, W. P. Tai, and S. J. Shu, Thin Solid Films 491, 153 (2005).
14
L. Znaidi, G. J. A. A. Soler Illia, S. Benyahia, C. Sanchez, and A. V. Kanaev, Thin
Solid Films 428, 257 (2003).
15
B. Sun, R. L. Peterson, H. Sirringhaus, and K. Mori, J. Phys. Chem. C. 111, 18831
(2007).
16
H-C. Huang, and T-E. Hsieh, Nanotechnology, 21, 295707 (2010).
17
S. T. Meyers, J. T. Anderson, C. M. Hung, J. Thompson, J. F. Wager, and D. A.
Keszler, J. Am. Chem. Soc. 130, 17603 (2008).
18
H. C. Cheng, C. F. Chen, and C. C. Lee, Thin Solid Films 498, 142 (2006).
19
Y. W. Chen, Q. Qiao, Y. C. Liu, and G. L. Yang, J. Phys. Chem. C. 113, 7497 (2009).
84
20
T. Y. Kim, J. Y. Kim, K. M. Senthil, E. K. Suh, and K. S. Nahm, J. Crystal Growth 270,
491 (2004).
21
C. F. Guo, Y. Wang, P. Jiang, S. Cao, J. Miao, Z. Zhang, and Q. Liu, Nanotechnology
19, 445710 (2008).
22
X. Y. Kong, Y. Ding, and Z. L. Wang, J. Phys. Chem. B. 108, 570 (2004).
23
Y. Gui, C. Xie, Q. Zhang, M. Hu, J. Yu, and Z. Weng, Journal of Crystal Growth 289,
663 (2006).
24
S. Y. Han, G. S. Herman, and C. H. Chang, ECS Transactions 33, 275 (2010).
25
S. Y. Han, G. S. Herman, and C. H. Chang, J. Am. Chem. Soc. 133, 5166 (2010).
26
R. Nakamura, J. G. Lee, D. Tokozakura, H. Mori, and H. Nakajima, Materials Letters
61, 1060 (2007).
27
R. Wu, C. Xie, H. Xia, J. Hu, and A.Wang, Journal of Crystal Growth 217, 274 (2000).
28
R. Wu, C. Xie, J. Hu, H. Xia and A. Wang, Scripta Mater.43, 841 (2000).
29
L. Gunawan, and G. P. Johari, J. Phys. Chem C. 112, 20159 (2008).
85
CHAPTER-4 MOLECULAR PASSIVATION OF AMORPHOUS ZINC TIN
OXIDE THIN FILM TRANSISTORS USING N-HEXYLPHOSPHONIC ACID
4.1 INTRODUCTION
Transparent amorphous oxide semiconductors (TAOS) have gained much
attention for potential use as the active semiconductor in thin film transistors (TFTs) for
large area, high definition displays. 1 These materials, which are composed of heavy
metal cations (HMCs) with (n-1)d10ns0 (n ≥ 4), such as indium gallium zinc oxide
(IGZO) and zinc tin oxide (ZTO), constitute an interesting class of semiconductors that
can be used as either conductors or active materials for TFTs, since they possess
relatively high electron mobilities despite their amorphous nature. 2-4 It has been
demonstrated that oxide TFTs have electrical characteristics suitable for use as select
transistors for pixels in an active-matrix liquid crystal display (AMLCD). 5 However,
good stability of the TFTs under bias stress is necessary since any shift in threshold
voltage (Vth) will cause display non-uniformities by changing the brightness of individual
pixels. 6-11 This threshold voltage instability is attributed not only to charge trapping at
the channel/insulator interface, but also to the interaction between water, oxygen and/or
other molecules in the ambient atmosphere with the exposed channel back-surface during
gate voltage stress for bottom gate TFTs. 10, 12-17
Various passivation layers have been used for oxide TFTs, such as SiOx, SiNx,
AlOx, and poly (methyl methacrylate) (PMMA), to improve bias stress stability by
minimizing the interaction with molecules in the atmosphere and the back channel
86
layer.15, 18-21 Inorganic passivation has been reported to impact the performance of
unstressed devices. For example, inorganic passivated devices have shown negative
shifted Vth when compared to unpassivated devices, with no change in field effect
mobility (µFE) or sub-threshold swing (S). 22 It was suggested that desorption of O2 from
the surface during bias stressing results in higher carrier densities and a negative shift in
the threshold voltage. 22 In another study it was reported that SiO2 could be used to
passivate ZTO TFTs, but that additional annealing steps were necessary to minimize
negative shifts in the turn on voltage (Von) when compared to unpassivated TFTs.18 On
the other hand, hybrimer passivation using methacrylate or epoxy gave much better Von
and S stability compared to poly(methyl methacrylate) (PMMA). It was proposed that the
higher TFT instability for PMMA is due to the diffusion of H2O molecules through the
polymer layer during bias stress.23
Organic molecules are considered attractive functional materials due to their
structural variability and have recently led to fabrication of hybrid organic-inorganic
structures with a wide range of material properties. 23, 24 Such hybrid materials have been
developed in the form of nanoparticles, thin films, and self-assembled monolayers
(SAMs). As mentioned above, methacrylate/epoxy hybrimer perform well as an organicinorganic hybrid passivation layer due to the excellent diffusion barrier formed for O2
and H2O. It was demonstrated that ZTO TFTs with hybrimer passivation layers had much
improved gate bias stability and resilience to environmental degradation.23 Organic
SAMs are commonly used to modify metal/metal oxide surfaces and tune their electronic
and interfacial properties. The molecular design of SAMs allows for the specific tuning
87
of surface properties for wide-ranging applications including coatings for corrosion
protection 25, 26 and as active elements in sensors.27 For example, SAMs patterned via
micro-contact printing can be used to define regions on transparent conducting oxide
films that will not be wet etched, and may provide a low cost method for fabricating
display structures.28
Silanes, 29, 30 thiols, 31-33 carboxylic acids, 34 and phosphonic acids (PA) 35-46 have
all been used as molecular anchors on metal oxide surfaces. While thiols have shown
some promise, silanes are only appropriate for specific applications and have shown
some limitations due to the dependence on surface hydroxyl concentration and the
tendency to hydrolyze and polymerize in solution. 40 Likewise, carboxylic acids do not
bind strongly to the metal oxide surface. 34, 44 Phosphonic acids (PA) have been used to
successfully modify a variety of metal oxide surfaces with high packing densities and
good stability. Furthermore, metal oxide surfaces modified with alkane–based
phosphonic acid species are insensitive to water as a result of the low surface energy due
to the alkane groups and the strong bonding between the phosphonic group and the oxide
surface. 47 It has been suggested that improved stability of phosphonic SAM films can
occur through activation of bonding by thermal annealing after absorption of the SAM
layer on the substrate. 48, 49 It was proposed that in the unheated film phosphonic acid
molecules are only hydrogen bonded to the substrate and to neighboring molecules.
Thermal annealing after the SAM formation promotes strong chemical bonding of
phosphonic acid molecules to the surface. In a recent study, Perkins compared both
hexanethiol and n-hexylphosphonic acid as a surface modifier for polycrystalline ZnO
88
thin films and ZnO crystals. 35 It was shown that improved corrosion resistance against
Brönsted acids was achieved with monolayers formed from n-HPA compared to
monolayers formed from hexanethiol.35
Although a tremendous amount of data has been published on the interaction of
SAMs with oxide surfaces, there have been no studies on the surface functionalization of
ZTO with SAMs. In this article, we have studied the interaction of n-HPA with
amorphous ZTO. The surface was characterized using contact angle measurements and
X-ray photoelectron spectroscopy (XPS). Furthermore, we have fabricated bottom-gate
TFTs with and without n-HPA passivation, and measured the corresponding electrical
characteristics for these devices.
4.2 EXPERIMENTAL
N-hexylphosphonic acid (n-HPA, C6H15O3P) was purchased from STREM
chemicals, stored in a desiccator, and used without further purification. A structural
representation of n-HPA is shown in Figure 4.1 (a). Solutions of 2 mM n-HPA were
prepared by diluting in 95% ACS grade ethanol.
Surface modification of blanket ZTO substrates
The blanket and patterned ZTO films used for our work were deposited using RF
magnetron sputter deposition. Prior to deposition Si substrates with thermally grown SiO2
(100nm) were cleaned with acetone, isopropyl alcohol (IPA) and DI water followed by
exposure to UV-Ozone for 1 hour to remove organic contaminants. A 3 inch diameter
ceramic ZTO target (2:1 ZnO:SnO2) was used for these studies and was purchased from
89
AJA International Inc. Typical sputter deposition conditions used an RF power of 100 W,
a chamber pressure of ~ 4 mTorr, 20 SCCM flow rate that included a mixture of 1:19
O2:Ar. The deposition time was adjusted to obtain a desired ZTO film thickness of ~ 40
nm. During deposition the substrate was rotated to ensure thickness uniformity. For
TFTs, the channel layer was patterned using a shadow mask during the deposition
process. The as-deposited ZTO films were post-annealed in air with a ramp rate of 10
°C/min up to 600 °C in a tube furnace. Staggered bottom gate TFT type structures were
finished by depositing ~500 nm Al by thermal evaporation through a shadow mask to
define the source and drain contacts. Fabricated TFT width and length were 1000 µm by
200 µm, respectively.
The ZTO blanket films were cleaned using UV ozone for 1 hour to remove
organic contaminants and to increase the surface energy to assist in uniform binding of nHPA to the ZTO surface. Within 15 minutes of UV ozone cleaning, the substrates were
immersed in a glass petri dish containing 100 mL of 2 mM n-HPA solution and allowed
to functionalize for an amount of time ranging from 10 seconds to 1 day. After the
specified exposure time, the immersed ZTO substrates were removed from solution and
thoroughly rinsed with 95% ethanol and blown dried with N2.
Contact angle measurements
An FTA32 goniometer was used to perform video-based contact angle
measurements. Contact angles from the ZTO substrates were obtained within 15 minutes
of functionalization using 0.015 mL of water as the probe liquid. Two droplets were
90
dispensed at different
d
locations on thee substrate aand images w
were taken w
within a few
seeconds for contact anglee measuremeents.
Surface mod
dification of ZTO TFTss
Figure
F
4.1: a) Chemical structure
s
of n-hexylphos
n
sphonic acidd (n-HPA). S
Schematic view
of
o staggered bottom
b
gate ZTO TFT bb) without n--HPA c) withh n-HPA
ZTO TFTs
T
preparred for this work
w
were prrocessed usinng identical UV-Ozone
trreatment, n-H
HPA exposu
ure, and rinsiing as descriibed for the bblanket subsstrates. The T
TFTs
91
were allowed to functionalize in n-HPA for ~16 hours followed by rinsing and blow
drying. For comparison, TFTs were either measured with no further treatment or after a
post-annealing at 140 °C for 48 hours. For these annealing studies, samples were placed
in a quartz tube with flowing dry nitrogen. Post annealing was performed to activate the
chemical bonding at the surface. 36 To ensure that no physisorbed n-HPA molecules
remained on the surface after the annealing step, a base rinse was performed by
sonicating the samples in 5% triethylamine/ethanol solution for 30 minutes followed by
rinsing with ethanol and drying under nitrogen flow. In Figure 4.1, a schematic of ZTO
TFTs (b) without and (c) with n-HPA is schematically illustrated.
X-ray photoelectron spectroscopy (XPS)
Surface analysis was performed on the ZTO TFTs using X-ray photoelectron
spectroscopy (XPS). The XPS instrument used for this study was a Phi 5000 VersaProbe,
which used monochromatic Al Kα radiation with a photon energy of 1486.6eV. All
spectra were obtained with a pass energy of 23.5 eV and high resolution data were
obtained with a 0.2 eV step size. All XPS measurements were carried out at a take-off
angle of 45°. To correct for sample charging the C 1s peak for all spectra were set to
284.6 eV. 50
Electrical measurements
All TFT bias-stress measurements were performed at room temperature in air
using an Agilent 4155C precision semiconductor parameter analyzer. Transistors were
measured in a dark box to separate bias effects from light induced shifts of Vth. Single
92
sweep source-to-drain current versus gate voltage (IDS-VGS) transfer curves were
measured with the drain voltage (VDS) set to 1 V. The stress measurements were
performed up to 104 sec using an applied VGS of 10 V (1 MV/cm) and VDS set to 0 V. The
bias stressing was interrupted at pre-determined times to measure the transfer
characteristics. A single sweep consisted of ramping VGS from either -10 to 15 V or -15
to 10 V with 0.1 V steps, depending on the Von for each transistor.
4.3 RESULTS AND DISCUSSIONS
Water contact angle measurements were performed to determine the saturation
coverage of n-HPA monolayers on ZTO. The average contact angles of the UV-O3
cleaned blanket ZTO were ~26.6° due to the removal of organic contaminants and
increased activation of surface oxygen species. The functionalization with n-HPA
increased the contact angle for the surface indicating the surface has a lower surface
energy after the n-HPA treatment. A summary of contact angle measurements for blanket
ZTO films exposed to n-HPA for various times is shown in Figure 4.2. The contact angle
was found to increase with increasing n-HPA exposure time, where a maximum contact
angle of 104.8° was obtained at 16 hours. This contact angle was in reasonable agreement
with values obtained for n-HPA functionalized ZnO surfaces using similar conditions in a
prior study.35 A reference substrate was also exposed to n-HPA for 24 hours to confirm
that contact angle did not continue to increase with exposure time and that the surface
had reached saturation.
93
104.8 °
Figure
F
4.2: Contact
C
anglee (degree) veersus n-HPA
A exposure tiime (hours) for each
in
ndividual ZT
TO film expo
osed to n-HP
PA at a speccific interval of time. Thee
reeference film
m was expossed for 24 hoours. The coontact angle directly afteer UV
trreatment was 26.6 °
XPS was
w used to study
s
the changes in surrface chemiccal state of thhe ZTO TFT
T
h n-HPA andd post therm
mal treatmentts. A survey
fiilms due to the functionaalization with
sp
pectrum wass initially obtained to dettermine the ssurface com
mposition andd assess surfa
face
im
mpurity conccentrations. It
I was found
d that the onlly impurity oon the un-fuunctionalizedd
su
urface was carbon.
c
For the
t n-HPA fu
unctionalizedd surfaces addditional peaaks due to
ph
hosphorous were observ
ved, along with
w an increaase in intensiity for the caarbon peaks..
High-resoluti
H
ion spectra were
w obtained
d for the Zn 2p3/2, Sn 3dd5/2, O 1s, C 1s and P 2p
peaks. These spectra weree obtained fo
or samples w
with and withhout n-HPA
A, and with annd
without
w
the post-annealin
ng step.
94
Figure
F
4.3: Zn
Z 2p3/2 XPS spectra obtaained from Z
ZTO films w
with/without n-HPA and
with/without
w
post-annealing.
XPS peak
p
fitting was
w perform
med using XP
PSPEAK 4.11. The peaks were fit witth a
liinear backgro
ound and du
uring this anaalysis, the peeak area, fulll width at haalf maximum
m
(F
FWHM), %L
Lorentzian-G
Gaussian (%L-G) and binnding energies were adjusted to obtaain a
good fit to thee experimen
ntal data. Theese parameteers were heldd constant foor the variouus
XPS
X peaks an
nd surface treatments. Th
he FWHM ffor the Zn 2pp3/2 and Sn 3dd5/2 were 1.665
an
nd 1.38eV and the corresponding %L-G were 400% and 45%
%, respectively. The Zn 22p3/2
an
nd Sn 3d5/2 spectra
s
for ZTO
Z
TFT film
ms are show
wn in Figuress 4.3 and 4.4, respectivelly.
The
T binding energy
e
for Zn
Z 2p3/2 and Sn
S 3d5/2 for th
the un-functiionalized sam
mple with noo-
95
post annealing and after charge
c
comp
pensation waas found to bbe 1021.84 eV and 486.661
V. These vallues are closse to those reeported for ooxide films ccontaining ziinc and tin aand
eV
co
orrespond to
o Zn2+ and Sn
n4+ oxidation
n states 51,52. We found tthat the bindding energiess
were
w nearly id
dentical for films irrespeective of the presence off n-HPA or tthe post therm
mal
trreatments.
Figure
F
4.4: Sn
S 3d5/2 XPS spectra obtaained from Z
ZTO films w
with/without n-HPA and
with/without
w
post-annealing.
The O1s
O spectra for
f ZTO TFT
T films are shhown in Figgure 4.5. We found that iit was
u three peaaks to adequ
uately fit the spectra, wheere a FWHM
M of 1.42eV and
necessary to use
96
%L-G of 10 was used for the main peak at 530.4eV which we assign to oxygen
coordinated with zinc and tin in the ZTO film. For the samples that are not functionalized
we assign the shoulder at 531.7eV to surface hydroxyl groups, while the third peak at
532.9 eV is assigned to H2O absorbed on to the surface. 52 In prior studies, it has been
shown that the O 1s peak changes in shape due to the binding of phosphonic acid groups
to the oxide surface. This is due to the reaction between surface hydroxyl groups leading
to the formation of P-O-M bonds. 36 The O 1s peaks for the ZTO films with n-HPA are
also shown in Figure 4.5. We found that it was again necessary to use three peaks to fit
the data, but that we required a slightly lower FWHM of 1.36 when compared to the
surfaces with no HPA. Both un-annealed and 140 °C annealed n-HPA/ZTO surfaces
showed a main O 1s peak at 530.1 eV which we assign to oxygen coordinated with zinc
and tin in the ZTO. In comparison to un-functionalized devices, the main O 1s peak is
reduced in intensity for the samples functionalized with n-HPA. This can be attributed to
the formation of molecular monolayers on the surface that attenuates photoelectrons
originating from the ZTO. Similar attenuation of the substrate signal for surfaces
functionalized with SAMS has been shown in a prior study. 36 The second peak was
determined to be at 531.29eV and 531.35eV for the SAM/ZTO devices that were not
post-annealed and post-annealed to 140 °C. These peaks can be attributed to P-O-Zn/PO-Sn and P=O. It should also be noted that this second peak increases in relative intensity
compared to the surfaces without n-HPA suggesting a higher surface concentration of
these species. The third peak at a higher binding energy of 532.5 eV can be assigned to
the P-OH species. These O 1s binding energies are consistent with literature values for
97
ph
hosphonic acid species bonding
b
to oxide
o
surfacees. 53,54 In alll these casess, the FWHM
M of
th
he second an
nd the third peaks
p
for thee O 1s spectrra were assiggned a slighttly higher vaalue
of 1.47 eV to take into acccount inhom
mogenities asssociated wiith the hydrooxyl, water, aand
n-HPA speciees on the ZT
TO surface.
Figure
F
4.5: O 1s XPS speectra obtaineed from ZTO
O films with//without n-H
HPA and
with/without
w
post-annealing.
The C 1s spectra obtaained for the ZTO surfacce at various conditions iis shown in
Figure 4.6. Th
he ZTO film
m without HP
PA that was not post-annnealed had a C 1s peak,
98
which
w
was fittted with a siingle compo
onent at 284.6 eV using a FWHM off 1.32 and % L-G
of 40. These fitting
f
param
meters were kept
k constannt for all otheer sample prreparation
co
onditions. Th
his peak is attributed
a
to aliphatic carrbon (C-C orr C-H bondss). 55 In addittion
to
o the main peeak, the sam
mples annealeed to 140 °C
C had a seconnd broad peaak at 286.2 eeV
which
w
can be assigned to the carbon bound
b
to oxyygen (C-O) oor carbon boound to
ph
hosphorus (C
C-P, has an expected
e
C 1s binding eenergy of 2866-286.4 eV).55, 56
Figure
F
4.6: C 1s XPS speectra obtaineed from ZTO
O films with//without n-H
HPA and
with/without
w
post-annealing.
99
Figure
F
4.7: P 2p XPS speectra obtaineed from ZTO
O films with//without n-H
HPA and
with/without
w
post-annealing.
The P 2p spectra is
i shown in Figure
F
4.7 annd was fittedd as a singlee spin-orbit ssplit
doublet with a binding en
nergy differeence of 0.86 eV 57 and a fixed-area raatio of 2:1. For
th
he ZTO samp
ple with n-H
HPA that wass not annealeed after funcctionalization a FWHM of
1.46 eV was used
u
where the
t binding energy
e
for thhe P 2p3/2 annd P 2p1/2 weere determineed to
100
be 132.98 and 133.84 eV, respectively. For the post-functionalization annealed samples
similar fitting parameters were used except for the FWHM which was significantly
reduced to 1.18 eV and the binding energy for the P 2p3/2 and P 2p1/2 were determined to
be 132.94 and 133.80 eV, respectively. As expected the P 2p peak was not observed for
ZTO films which were not functionalized with n-HPA. The significant reduction in
FWHM in the P 2p spectra for the post-functionalization annealed samples suggests
higher uniformity of the n-HPA bonding to the ZTO surface, possibly due to bond
activation.
Figure 4.8 shows ID versus VGS transfer characteristics for the ZTO TFTs, after
various treatments, as a function of gate bias stress time. Both un-functionalized and nHPA functionalized ZTO TFTs were characterized to investigate the effect of backchannel molecular passivation on gate-bias stress stability. For these studies a gate bias of
10 V was applied while the source and drain electrodes were grounded. As the duration
of applied gate bias increased, positive threshold voltage shifts were observed during
which the sub-threshold slope and the average mobility (µavg) remained unchanged. The
room temperature un-functionalized TFTs had an average mobility (µavg) of 10.8 cm2/Vs
and a maximum turn-on voltage shift (ΔVON) of 0.8 V for 10000 sec of applied stress,
whereas the un-functionalized devices annealed to 140°C had an µavg of 12.6 cm2/Vs and
ΔVON of 1.5 V. A variety of mechanisms have been proposed for bias stress induced
shifts of ZTO TFTs including charge injection and trapping at the dielectric/ZTO
interface, trapping in the ZTO channel, or from the absorption/desorption of molecules
from the atmosphere during bias stress. 17, 58, 59, 60
101
Figure
F
4.8: Plot
P of ID vs VGS curves for
f various cconditions off the ZTO TFTs as a funnction
of
o applied biaas stress time. (W/L = 10000 µm/2000 µm)
a) Witthout n-HPA
A, not post-aannealed, b) W
Without n-H
HPA, 140 °C
C annealed,
c) Witth n-HPA, not
n post-anneealed, d) Witth n-HPA, 140 °C anneaaled.
s
in VONN for un-funcctionalized Z
ZTO TFTs w
were
The positive gate--bias stress shifts
fo
ound to be well
w describeed by the streetched exponnential equattion, 58
∆
=
−
−
( . )
102
where ΔVo = VG,stress - Vthi , which is approximately the effective voltage drop across the
gate oxide, τ is the characteristic trapping time of carriers, and β is the stretched
exponential exponent. In using this relationship, we use VON in the place of Vth due to the
higher precision for obtaining VON. The characteristic trapping time is thermally activated
and is related to an effective energy barrier that carriers in the semiconductor channel
must overcome before entering charge trap states. Although this model is primarily used
to describe VON shifts due to trapping at the dielectric interface or due to metastability in
the channel material we have found that molecular passivation has a profound effect on
reducing bias induced VON shifts where there were no other changes in the system. We
also found that the logarithmic time dependence model did not adequately fit the
experimental data as have other groups evaluating the effect of molecular absorption on
bias stress instabilities for TAOS TFTs. 17
The shift in VON (V) versus bias stress time (sec) for ZTO devices with and
without n-HPA and with and without post-annealing treatment are shown in Figure 4.9
and 4.10, along with fits to the stretched exponential model for the un-functionalized
TFTs. The values of VON were obtained by using log (ID/(W/L))(A) equal to -11 from the
corresponding log ID versus VGS plots. The extracted values of τ and β for different
thermal treatments are summarized in Table 4.1. The τ values were determined to be 1.4
x 108 and 9.0 x 105 sec while β values were 0.30 and 0.59 for the un-functionalized
devices that were not annealed and post-annealed to 140 °C, respectively. The
characteristic trapping times estimated from the stretched exponential model are in
reasonably good agreement with prior studies for various TFT materials and structures.
103
For example, Libsch and Kanichi rep
ported the chharacteristic ttrapping tim
me of ~108 seec in
morphous Sii TFTs using
g a SiNx gatee insulator. 5 8 For organiic TFTs withh a variety off
am
acctive semico
onductor layeers and therm
mally grownn SiO2 gate innsulator, vallues of τ cann
raange between 103 to 107 sec. 61 Receently a τ of 33.9 x 105 secc was obtaineed for RF-spputter
deposited ZTO with a PE
ECVD SiO2 gate
g insulatoor, 10and a τ oof 6.0 x 104 ssec was obtaained
fo
or an ink-jet printed ZTO
O TFT with an
a SiO2 gatee insulator deeposited by PECVD. 11
Figure
F
4.9 Sh
hift in VON (V
V) versus biias stress tim
me (sec) for Z
ZTO n-HPA functionalizzed
deevices that were
w not subjected to posst-annealing,, and a fit off the stretched
ex
xponential model
m
with th
he measured values of thhe TFTs withh no n-HPA.
104
Figure 4.10 Sh
hift in VON (V)
( versus biias stress tim
me (sec) for Z
ZTO n-HPA
A functionalizzed
deevices that were
w not subjected to posst-annealing at a temperaature of 140 °C,
an
nd a fit of thee stretched exponential
e
m
model with tthe measuredd values of tthe
TF
FTs with no n-HPA.
ZTO- Withou
ut n-HPA- annealing
Room
R
Tem
mperature
140 °C for 48 hours
τ
(sec)
β
1. 4 E+08
9. 0 E+05
0.30
0.59
Table
T
4.1: Ch
haracteristic trapping tim
me of carrier s, τ and strettched exponnential exponnent,
β for ZTO TFT
Ts (without n-HPA), eith
ther left in rooom temperaature or posttan
nnealed to140 °C for 48 hours.
h
105
The TFTs with n-HPA had a maximum ΔVON of 0.1V irrespective of the postfunctionalization annealing treatment. One major difference is that the TFTs with n-HPA
that was not annealed had a shift of 0.1 V after 100 sec of gate bias stress, whereas the
TFTs with n-HPA that was annealed to 140 °C had a shift of 0.1 V only after 10000 sec.
This suggests that ZTO TFTs with post-functionalization thermal annealing are even
more stable than those without. This could be attributed to the fact that assembling the nHPA from solution on the zinc tin oxide surface followed by thermal annealing to
moderate temperature could possibly lead to formation of self-assembled monolayers
with more uniform surface-bound phosphonic acid-anchored alkyl chains. This is
consistent with the reduced with in the FWHM of the P 2p peak fitting parameters for the
N-HPA functionalized surface that had the activation anneal.
As mentioned above, these results suggest that the threshold voltage shifts can be
attributed to the back channel surface and not the dielectric/semiconductor interface since
the functionalization of the ZTO makes a tremendous difference in the observed bias
stress shifts. This may be due to the adsorption/desorption of oxygen and water vapor on
the exposed back channel layer by the application of a positive gate bias.
For example, Several studies have suggested that oxygen molecules absorbed on a
TAOS back-channel can accept electrons from the conduction band resulting in
superoxide formation (O2-) according to the following reaction equation,
O + e → O
106
This reaction leads to the formation of a depletion layer at the TAOS surface, and a space
charge width that can approach the thickness of the channel layer, which can result in a
positive shift of VON. 15, 62, 63
Likewise it has been shown that the absorption of H2O on a TAOS surface can
donate electrons to the conduction band, as well as provide electron traps. 15 This reaction
leads to the formation of an accumulation layer at the TAOS surface which can result in a
negative shift in VON. To obtain a positive VON shift, as observed in the experiments for
the un-functionalized surfaces, it would be necessary for water molecules to desorb
during positive gate bias stress. Likewise, if water desorption does occur upon positive
gate bias it would be expected that the sub-threshold slope would be reduced during this
process due to the reduction of water induced traps. 15 Since we have not observed any
change in S we believed that field-induced chemisorption of oxygen on the backchannel
was the dominant mechanism leading to the VON shift for un-functionalized surfaces
during positive bias stressing. This effect has been proposed by other groups as well for
IGZO surfaces. 15,64
4.4 CONCLUSIONS
High-performance stable ZTO TFTs were fabricated through the molecular
passivation of the back channel surface with n-hexylphosphonic acid. Contact angle
measurements indicate that SAMs formed using n-HPA was hydrophobic, suggesting
good coverage. XPS confirms that the phosphonic acid chemically binds to the ZTO
surface and that the surface bonding is activated by a post 140 °C anneal. All ZTO TFTs,
107
regardless of functionalization or post-annealing treatment had positive turn-on voltage
shifts with application of a positive gate bias stress. The time dependence of the turn-on
voltage shifts for the un-functionalized ZTO devices was best described by a stretched
exponential model. It was found that n-HPA functionalized ZTO TFTs with and without
post-annealing had significant lower ∆VON compared to un-functionalized ZTO TFTs.
These data suggest that n-HPA forms very stable back channel surfaces resulting in very
stable devices. Finally, we propose that these well-packed SAMs protect the ZTO backchannel from interaction with ambient O2/H2O, thus preventing the modulation of
channel conductance.
108
4.5 REFERENCES
1. K. Nomura, H. Ohta, A. Takagi, T. Kamiya, M. Hirano and H. Hosono: Roomtemperature fabrication of transparent flexible thin-film transistors using amorphous
oxide semiconductors. Nature 432, 488 (2004).
2. H. Hosono, N. Kikuchi, N. Ueda and H. Kawazoe: Working hypothesis to explore
novel wide band gap electrically conducting amorphous oxides and examples. Journal
of Non-Crystalline Solids 200, 165 (1996).
3. S. Narushima, M. Orita, M. Hirano and H. Hosono: Electronic structure and transport
properties in the transparent amorphous oxide semiconductor 2 CdO center dot GeO2.
Physical Review B 66, 8 (2002).
4. H.Q. Chiang, J.F. Wager, R.L. Hoffman, J. Jeong and D.A. Keszler: High mobility
transparent thin-film transistors with amorphous zinc tin oxide channel layer. Applied
Physics Letters 86, 3 (2005).
5. R.L. Hoffman, B.J. Norris and J.F. Wager: ZnO-based transparent thin-film
transistors. Applied Physics Letters 82, 733 (2003).
6. P. Gorrn, M. Sander, J. Meyer, M. Kroger, E. Becker, H.H. Johannes, W. Kowalsky
and T. Riedl: Towards see-through displays: Fully transparent thin-film transistors
driving transparent organic light-emitting diodes. Advanced Materials 18, 738 (2006).
7. P. Gorrn, P. Holzer, T. Riedl, W. Kowalsky, J. Wang, T. Weimann, P. Hinze and S.
Kipp: Stability of transparent zinc tin oxide transistors under bias stress. Applied
Physics Letters 90, 3 (2007).
8. P. Gorrn, F. Ghaffari, T. Riedl and W. Kowalsky: Zinc tin oxide based driver for
highly transparent active matrix OLED displays. Solid-State Electronics 53, 329
(2009).
9. Y. Jeong, K. Song, D. Kim, C.Y. Koo and J. Moon: Bias Stress Stability of SolutionProcessed Zinc Tin Oxide Thin-Film Transistors. Journal of the Electrochemical
Society 156, H808 (2009).
10. J. Triska, J.F. Conley, R. Presley and J.F. Wager: Bias stress stability of zinc-tin-oxide
thin-film transistors with Al(2)O(3) gate dielectrics. Journal of Vacuum Science &
Technology B 28, C5I1 (2010).
109
11. C. Avis and J. Jang: A High Performance Inkjet Printed Zinc Tin Oxide Transparent
Thin-Film Transistor Manufactured at the Maximum Process Temperature of 300
degrees C and Its Stability Test. Electrochemical and Solid State Letters 14, J9 (2011).
12. R.B.M. Cross and M.M. De Souza: Investigating the stability of zinc oxide thin film
transistors. Applied Physics Letters 89, 3 (2006).
13. Y. Vygranenko, K. Wang and A. Nathan: Stable indium oxide thin-film transistors
with fast threshold voltage recovery. Applied Physics Letters 91, 3 (2007).
14. J.M. Lee, I.T. Cho, J.H. Lee and H.I. Kwon: Bias-stress-induced stretched-exponential
time dependence of threshold voltage shift in InGaZnO thin film transistors. Applied
Physics Letters 93, 3 (2008).
15. J.K. Jeong, H.W. Yang, J.H. Jeong, Y.G. Mo and H.D. Kim: Origin of threshold
voltage instability in indium-gallium-zinc oxide thin film transistors. Applied Physics
Letters 93, 3 (2008).
16. K. Hoshino, D. Hong, H.Q. Chiang and J.F. Wager: Constant-Voltage-Bias Stress
Testing of a-IGZO Thin-Film Transistors. Ieee Transactions on Electron Devices 56,
1365 (2009).
17. S.Y. Sung, J.H. Choi, U.B. Han, K.C. Lee, J.H. Lee, J.J. Kim, W. Lim, S.J. Pearton,
D.P. Norton and Y.W. Heo: Effects of ambient atmosphere on the transfer
characteristics and gate-bias stress stability of amorphous indium-gallium-zinc oxide
thin-film transistors. Applied Physics Letters 96, 3 (2010).
18. D. Hong and J.F. Wager: Passivation of zinc-tin-oxide thin-film transistors. Journal of
Vacuum Science & Technology B 23, L25 (2005).
19. S.K. Park, Y.H. Kim, H.S. Kim and J.I. Han: High Performance Solution-Processed
and Lithographically Patterned Zinc-Tin Oxide Thin-Film Transistors with Good
Operational Stability. Electrochemical and Solid State Letters 12, H256 (2009).
20. J.S. Park, T.S. Kim, K.S. Son, K.H. Lee, W.J. Maeng, H.S. Kim, E.S. Kim, K.B. Park,
J.B. Seon, W. Choi, M.K. Ryu and S.Y. Lee: The influence of SiO(x) and SiN(x)
passivation on the negative bias stability of Hf-In-Zn-O thin film transistors under
illumination. Applied Physics Letters 96, 3 (2010).
21. D.A. Mourey, D.A.L. Zhao, J. Sun and T.N. Jackson: Fast PEALD ZnO Thin-Film
Transistor Circuits. Ieee Transactions on Electron Devices 57, 530 (2010).
22. P. Gorrn, T. Riedl and W. Kowalsky: Encapsulation of Zinc Tin Oxide Based Thin
Film Transistors. Journal of Physical Chemistry C 113, 11126 (2009).
110
23. S.J. Seo, S. Yang, J.H. Ko and B.S. Bae: Effects of Sol-Gel Organic-Inorganic Hybrid
Passivation on Stability of Solution-Processed Zinc Tin Oxide Thin Film Transistors.
Electrochemical and Solid State Letters 14, H375 (2011).
24. Y. Jung, T. Jun, A. Kim, K. Song, T.H. Yeo and J. Moon: Direct photopatternable
organic-inorganic hybrid gate dielectric for solution-processed flexible ZnO thin film
transistors. Journal of Materials Chemistry 21, 11879 (2011).
25. J. Scherer, M.R. Vogt, O.M. Magnussen and R.J. Behm: Corrosion of alkanethiolcovered Cu(100) surfaces in hydrochloric acid solution studied by in-situ scanning
tunnelling microscopy. Langmuir 13, 7045 (1997).
26. H. Amar, J. Benzakour, A. Derja, D. Villemin, B. Moreau and T. Braisaz: Piperidin-1yl-phosphonic acid and (4-phosphono-piperazin-1-yl) phosphonic acid: A new class of
iron corrosion inhibitors in sodium chloride 3% media. Applied Surface Science 252,
6162 (2006).
27. H.J. Lim, D.Y. Lee and Y.J. Oh: Gas sensing properties of ZnO thin films prepared by
microcontact printing. Sensors and Actuators a-Physical 125, 405 (2006).
28. T.L. Breen, P.M. Fryer, R.W. Nunes and M.E. Rothwell: Patterning indium tin oxide
and indium zinc oxide using microcontact printing and wet etching. Langmuir 18, 194
(2002).
29. C.G. Allen, D.J. Baker, J.M. Albin, H.E. Oertli, D.T. Gillaspie, D.C. Olson, T.E.
Furtak and R.T. Collins: Surface Modification of ZnO Using Triethoxysilane-Based
Molecules. Langmuir 24, 13393 (2008).
30. C. Haensch, S. Hoeppener and U.S. Schubert: Chemical modification of selfassembled silane based monolayers by surface reactions. Chemical Society Reviews
39, 2323 (2010).
31. H.P. Zhang and S. Baldelli: Alkanethiol monolayers at reduced and oxidized zinc
surfaces with corrosion proctection: A sum frequency generation and electrochemistry
investigation. Journal of Physical Chemistry B 110, 24062 (2006).
32. C. Nogues and P. Lang: Self-assembled alkanethiol monolayers on a Zn substrate:
Structure and organization. Langmuir 23, 8385 (2007).
33. T.C. Monson, M.T. Lloyd, D.C. Olson, Y.J. Lee and J.W.P. Hsu: Photocurrent
Enhancement in Polythiophene- and Alkanethiol-Modified ZnO Solar Cells. Advanced
Materials 20, 4755 (2008).
111
34. C.E. Taylor and D.K. Schwartz: Octadecanoic acid self-assembled monolayer growth
at sapphire surfaces. Langmuir 19, 2665 (2003).
35. C.L. Perkins: Molecular Anchors for Self-Assembled Monolayers on ZnO: A Direct
Comparison of the Thiol and Phosphonic Acid Moieties. Journal of Physical
Chemistry C 113, 18276 (2009).
36. P.J. Hotchkiss, M. Malicki, A.J. Giordano, N.R. Armstrong and S.R. Marder:
Characterization of phosphonic acid binding to zinc oxide. Journal of Materials
Chemistry 21, 3107 (2011).
37. I. Gouzman, M. Dubey, M.D. Carolus, J. Schwartz and S.L. Bernasek: Monolayer vs.
multilayer self-assembled alkylphosphonate films: X-ray photoelectron spectroscopy
studies. Surface Science 600, 773 (2006).
38. S.E. Koh, K.D. McDonald, D.H. Holt, C.S. Dulcey, J.A. Chaney and P.E. Pehrsson:
Phenylphosphonic acid functionalization of indium tin oxide: Surface chemistry and
work functions. Langmuir 22, 6249 (2006).
39. R. Luschtinetz, A.F. Oliveira, J. Frenzel, J.O. Joswig, G. Seifert and H.A. Duarte:
Adsorption of phosphonic and ethylphosphonic acid on aluminum oxide surfaces.
Surface Science 602, 1347 (2008).
40. C.W. Hsu, H.R. Liou, W.F. Su and L.Y. Wang: Self-assembled monolayers of 2(thienyl)hexylphosphonic acid on native oxide surface of silicon fabricated by airliquid interface-assisted method. Journal of Colloid and Interface Science 324, 236
(2008).
41. P.B. Paramonov, S.A. Paniagua, P.J. Hotchkiss, S.C. Jones, N.R. Armstrong, S.R.
Marder and J.L. Bredas: Theoretical Characterization of the Indium Tin Oxide Surface
and of Its Binding Sites for Adsorption of Phosphonic Acid Monolayers. Chemistry of
Materials 20, 5131 (2008).
42. O. Acton, G.G. Ting, P.J. Shamberger, F.S. Ohuchi, H. Ma and A.K.Y. Jen: Dielectric
Surface-Controlled Low-Voltage Organic Transistors via n-Alkyl Phosphonic Acid
Self-Assembled Monolayers on High-k Metal Oxide (vol 2, pg 511, 2010). Acs
Applied Materials & Interfaces 2, 2963 (2010).
43. B.B. Zhang, T. Kong, W.Z. Xu, R.G. Su, Y.H. Gao and G.S. Cheng: Surface
Functionalization of Zinc Oxide by Carboxyalkylphosphonic Acid Self-Assembled
Monolayers. Langmuir 26, 4514 (2010).
44. M. Dubey, T. Weidner, L.J. Gamble and D.G. Castner: Structure and Order of
Phosphonic Acid-Based Self-Assembled Monolayers on Si(100). Langmuir 26, 14747
(2010).
112
45. G. Fonder, I. Minet, C. Volcke, S. Devillers, J. Delhalle and Z. Mekhalif: Anchoring
of alkylphosphonic derivatives molecules on copper oxide surfaces. Applied Surface
Science 257, 6300 (2011).
46. M. Maxisch, P. Thissen, M. Giza and G. Grundmeier: Interface Chemistry and
Molecular Interactions of Phosphonic Acid Self-Assembled Monolayers on
Oxyhydroxide-Covered Aluminum in Humid Environments. Langmuir 27, 6042
(2011).
47. P.H. Mutin, G. Guerrero and A. Vioux: Hybrid materials from organophosphorus
coupling molecules. Journal of Materials Chemistry 15, 3761 (2005).
48. E.S. Gawalt, G. Lu, S.L. Bernasek and J. Schwartz: Enhanced bonding of
alkanephosphonic acids to oxidized titanium using surface-bound alkoxyzirconium
complex interfaces. Langmuir 15, 8929 (1999).
49. E.S. Gawalt, M.J. Avaltroni, N. Koch and J. Schwartz: Self-assembly and bonding of
alkanephosphonic acids on the native oxide surface of titanium. Langmuir 17, 5736
(2001).
50. G.A. Diaz-Quijada, C. Maynard, T. Comas, R. Monette, C. Py, A. Krantis and G.
Mealing: Surface Patterning with Chemisorbed Chemical Cues for Advancing
Neurochip Applications. Industrial & Engineering Chemistry Research 50, 10029
(2011).
51. M.G. Kim, H.S. Kim, Y.G. Ha, J.Q. He, M.G. Kanatzidis, A. Facchetti and T.J.
Marks: High-Performance Solution-Processed Amorphous Zinc-Indium-Tin Oxide
Thin-Film Transistors. Journal of the American Chemical Society 132, 10352 (2010).
52. S.J. Seo, Y.H. Hwang and B.S. Bae: Postannealing Process for Low Temperature
Processed Sol-Gel Zinc Tin Oxide Thin Film Transistors. Electrochemical and Solid
State Letters 13, H357 (2010).
53. B. Adolphi, E. Jahne, G. Busch and X.D. Cai: Characterization of the adsorption of
omega-(thiophene-3-yl alkyl) phosphonic acid on metal oxides with AR-XPS.
Analytical and Bioanalytical Chemistry 379, 646 (2004).
54. G. Mani, D.M. Johnson, D. Marton, V.L. Dougherty, M.D. Feldman, D. Patel, A.A.
Ayon and C.M. Agrawal: Stability of self-assembled monolayers on titanium and gold.
Langmuir 24, 6774 (2008).
55. Y. Xie, X. Zhao, Y. Chen, Q. Zhao and Q. Yuan: Preparation and characterization of
porous C-modified anatase titania films with visible light catalytic activity. Journal of
Solid State Chemistry 180, 3576 (2007).
113
56. N. Adden, L.J. Gamble, D.G. Castner, A. Hoffmann, G. Gross and H. Menzel:
Phosphonic acid monolayers for binding of bioactive molecules to titanium surfaces.
Langmuir 22, 8197 (2006).
57. Y. Sun, Z. Liu, F. Machuca, P. Pianetta and W.E. Spicer: Optimized cleaning method
for producing device quality InP(100) surfaces. Journal of Applied Physics 97, 7
(2005).
58. F.R. Libsch and J. Kanicki: Bias-stress-induced stretched-exponential timedependence of charge injection and trapping in amorphous thin-film transistors.
Applied Physics Letters 62, 1286 (1993).
59. M.J. Powell, C. Vanberkel and J.R. Hughes: Time and temperature-dependence of
instability mechanisms in amorphous-silicon thin-film transistors. Applied Physics
Letters 54, 1323 (1989).
60. A. Suresh and J.F. Muth: Bias stress stability of indium gallium zinc oxide channel
based transparent thin film transistors. Applied Physics Letters 92, 3 (2008).
61. S.G.J. Mathijssen, M. Colle, H. Gomes, E.C.P. Smits, B. de Boer, I. McCulloch, P.A.
Bobbert and D.M. de Leeuw: Dynamics of threshold voltage shifts in organic and
amorphous silicon field-effect transistors. Advanced Materials 19, 2785 (2007).
62. Y.C. Chen, T.C. Chang, H.W. Li, S.C. Chen, J. Lu, W.F. Chung, Y.H. Tai and T.Y.
Tseng: Bias-induced oxygen adsorption in zinc tin oxide thin film transistors under
dynamic stress. Applied Physics Letters 96, 3 (2010).
63. M. Batzill and U. Diebold: The surface and materials science of tin oxide. Progress in
Surface Science 79, 47 (2005).
64. J.S. Park, J.K. Jeong, H.J. Chung, Y.G. Mo and H.D. Kim: Electronic transport
properties of amorphous indium-gallium-zinc oxide semiconductor upon exposure to
water. Applied Physics Letters 92, 3 (2008).
114
CHAPTER -5 CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK
5.1 CONCLUSIONS
This thesis focusses on deposition, functionalization, physical and electrical
characterization of two types of oxide-based semiconductor films used for various
applications.
Firstly we have shown new approach for obtaining nanostructured ZnO films via
thermal oxidation of Zn nanoparticle films that were obtained by spin coating metallic Zn
nanoparticles. SEM images of commercial nanoparticles of two different sizes were
imported into imageJ software and the size distributions were found to be 162 ± 86 nm
and 234 ± 112 nm for the vendor specified 35 nm and 130 nm respectively. The Zn
nanoparticles were stable with an onset of oxidation of ~250 °C during the thermal
analysis using TGA and the weight gains corresponding to oxidation of nanoparticles
were 15.1% and 19.4% for 162 and 234 nm Zn nanoparticles, respectively. Imaging with
HIM showed formation of a variety of nanostructures for increasing annealing
temperatures, especially for larger particles. The existence of Zn and ZnO phases on the
un-annealed and annealed samples were confirmed using Micro-XRD. Decreasing
intensity of Zn diffraction peaks with increasing intensity of ZnO diffraction peaks were
observed for increasing temperature upto 500°C at which the samples oxidized
completely. The relative compositions of metallic Zn and ZnO of the films were also
determined using micro-XRD. The wt% of Zn and ZnO on these films were used to
determine thickness of ZnO outer shell on the Zn nanoparticles at temperatures between
115
100-600 °C. Resistivity measurements indicate that the films are highly resistive.
However, electrical characterization of the TFTs fabricated with Zn nanoparticle and ZIO
precursors were dominated by the conducting nature of the thick Zn nanoparticle films.
Secondly, stable TFTs with sputter deposited ZTO were fabricated via
functionalization of the channel surface with SAMs by exposure to n-HPA. Contact angle
measurements suggested that SAMs formed using n-HPA were hydrophobic, ensuring
good coverage. Chemical binding of phosphonic acid to the ZTO surface was confirmed
using XPS. With application of positive bias at the gate, all ZTO TFTs, regardless of
functionalization or post-annealing treatment, exhibited a positive turn-on voltage shift.
Stretched exponential equation was used to obtain a best fit to the experimental data
suggesting that the VON shift is due to absorption of molecules from the ambient as well
as the charge trapping at the semiconductor/oxide interface. After 104 sec of gate bias, nHPA functionalized ZTO TFTs with and without post-annealing showed significant
improvement in device characteristics compared to un-functionalized ZTO TFTs. The
ZTO TFTs with n-HPA exhibited a ΔVON as low as 0.1 V which indicated that the
surfaces are very stable and the well-packed SAMs protected the ZTO back-channel from
ambient molecules whose interaction would have altered the channel conductance.
5.2 RECOMMENDATIONS FOR FUTURE WORK
For further studies, bias stress stability measurements can be done on the unfunctionalized/functionalized ZTO TFTs by varying the substrate temperatures and also
by testing in different ambients like air, nitrogen, vacuum, H2 and O2. Light effects during
116
the bias stress measurements can also be studied. The morphology of the n-HPA
functionalized ZTO films can be analyzed using atomic force microscopy (AFM). Fourier
transform infrared spectroscopy (FTIR) can be done to find the stretching vibrations of PO, P=O, P-OH and identify whether the phosphonic acid is bound to the ZTO surface in a
mono-dentate, bi-dentate or tri-dentate fashion.
117
BIBLIOGRAPHY
1. P.K. Weimer,"The TFT - A New Thin Film Transistor", Proc. IRE, 50, 1462 (1962).
2. T.P. Brody, F.C. Luo, Z.P. Szepesi, D.H. Davies, "A 6x6-in 20-lpi Electroluminescent
Display Panel", IEEE Trans.Electron Devices, ED-22, 739 (1975).
3. M.J. Powell, B.C. Easton, O.F. Hill, "Amorphous silicon-silicon nitride thin-film
transistors", Appl. Phys. Lett., 38, 794 (1981).
4. H. Gleskova, P.I. Hsu, Z. Xi, J.C. Sturm, Z. Suo, S. Wagner, "Field-effect mobility of
amorphous silicon thin-film transistors under strain", J Non-Cryst Solids., 338–340,
732 (2004).
5. A. Kuo, T.K. Won, J. Kanicki, "Advanced Multilayer Amorphous Silicon Thin-Film
Transistor Structure: Film Thickness Effect on Its Electrical Performance and Contact
Resistance", Jpn J Appl Phys., 47, 3362 (2008).
6. S. Morozumi, K. Oguchi, S. Yazawa, T. Kodaira, H. Ohshima, T. Mano, "B/W and
color LC Video displays addressed by Poly Si TFTs", SID Digest, 83, 156 (1983).
7. L. Pichon, K. Mourgues, F. Raoult, T. Mohammed-Brahim, K. Kis-Sion, D. Briand, O.
Bonnaud, "Thin film transistors fabricated by in situ doped unhydrogenated polysilicon
films obtained by solid phase crystallization" Semicond. Sci. Technol, 16, 918 (2001).
8. Y. Helen, R. Dassow, M. Nerding, K. Mourgues, F. Raoult, J.R. Kohler, T.
Mohammed-Brahim, R. Rogel, O. Bonnaud, J.H. Werner, H.P. Strunk, "High mobility
thin film transistors by Nd:YVO4-laser crystallization" Thin Solid Films, 383, 143
(2001).
9. Y. Kuo, "Thin film transistor technologies V", The Electrochemical Society
Proceedings, 2000-31,121 (2001).
10. C.D. Dimitrakopoulos, D.J. Mascaro, "Organic thin-film transistors: A review of
recent advances", IBM J. of. Res and Dev , 45, 11 (2011).
11. C-H. Wang, C-Y. Hsieh, J-C. Hwang, "Flexible Organic Thin-Film Transistors with
Silk Fibroin as the Gate Dielectric", Adv Mater., 23, 1630 (2011).
12. Q. Meng, H. Dong, W. Hu, D. Zhu, "Recent progress of high performance organic thin
film field-effect transistors", J. Mater. Chem., 21, 11708 (2011).
13. Y. Fujisaki, Y. Nakajima, D. Kumaki, T. Yamamoto, S. Tokito, T. Kono, J-i Nishida,
and Y. Yamashita, "Air-stable n-type organic thin-film transistor array and high gain
complementary inverter on flexible substrate", Appl. Phys. Lett., 97, 133303 (2010).
118
14. D.B. Mitzi, K. Chondroudis, C.R. Kagan, "Organic thin film transistors: A review of
recent advances", IBM J. of. Res and Dev., 45, 3, (2001).
15. H. Hosono, "Working hypothesis to explore novel wide band gap electrically
conducting amorphous oxides and examples", J Non-Cryst Solids., 198-200, 165
(1996).
16. A. Aoki, H. Sasakura, "Tin oxide thin film transistors", Japan. J. Appl. Phys., 9, 582
(1970).
17. M.W.J. Prins, K.-O. Grosse-Holz, G. Muller, J.F.M. Cillessen, J.B. Giesbers, R. P.
Weening, R. M. Wolf, "A ferroelectric transparent thin-film transistor" Appl. Phys.
Lett., 68, 3650 (1996).
18. R.E. Presley, C.L. Munsee, C-H. Park, D. Hong, J.F. Wager, D.A. Keszler, "Tin oxide
transparent thin-film transistors", J. Phys. D: Appl. Phys., 37, 2810 (2004).
19. M.S. Huh, B.S. Yang, S. Oh, J-h Kim, B.D. Ahn, J-H. Lee, J. Kim, J.K. Jeong, C.S.
Hwang, H.J. Kim "Improving the Performance of Tin Oxide Thin-Film Transistors by
Using Ultralow Pressure Sputtering", J. Electrochemical Society, 157, H425 (2010).
20. R.L. Hoffman, B.J. Norris, J. F. Wager, "ZnO-based transparent thin-film transistors"
Appl. Phys. Lett., 82, 733 (2003).
21. S. Masuda, K. Kitamura, Y. Okumura, S. Miyatake, H. Tabata, T. Kawai,
"Transparent thin film transistors using ZnO as an active channel layer and their
electrical properties", J Appl. Phys., 93, 1624 (2003).
22. P.F. Carcia, R.S. McLean, M.H. Reilly, G. Nunes, Jr., "Transparent ZnO thin-film
transistor fabricated by rf magnetron sputtering", Appl. Phys. Lett., 82, 1117 (2003).
23. E. Fortunato, A. Pimentel, L. Pereira, A. Goncalves, G. Lavareda, H. Aguas, I.
Ferreira, C.N. Carvalho, R. Martins, "High field-effect mobility zinc oxide thin film
transistors produced at room temperature", J Non-Cryst Solids.,338–340, 806 (2004).
24. L. Zhang, J. Li , X.W. Zhang , X.Y. Jiang, Z.L. Zhang, "High-performance ZnO thin
film transistors with sputtering SiO2/Ta2O5/SiO2 multilayer gate dielectric" Thin Solid
Films., 518, 6130 (2010).
25. B.J. Norris, J. Anderson, J.F. Wager, D.A. Keszler, "Spin-coated zinc oxide
transparent transistors", J. Phys. D: Appl. Phys., 36, L105 (2003).
26. B. Sun, H. Sirringhaus "Solution-Processed Zinc Oxide Field-Effect Transistors Based
on Self-Assembly of Colloidal Nanorods", Nano Lett., 5, 2408 (2005).
119
27. S. Lee, Y. Jeong, S. Jeong, J. Lee, M. Jeon, J. Moona, "Solution-processed ZnO
nanoparticle-based semiconductor oxide thin-film transistors", Superlattices
Microstruct., 44, 761 (2008).
28. H. Bong, W.H. Lee, D.Y. Lee, B.J. Kim, J.H. Cho, K. Cho, "High-mobility lowtemperature ZnO transistors with low-voltage operation", Appl. Phys. Lett., 96,
192115 (2010).
29. K. Nomura, H. Ohta, K. Ueda, T. Kamiya, M. Hirano, H. Hosono, " Thin-Film
Transistor Fabricated in Single-Crystalline Transparent Oxide Semiconductor",
Science, 300, 1269 (2003).
30. K. Nomura, H. Ohta, A. Takagi, T. Kamiya, M. Hirano, H. Hosono, "Roomtemperature fabrication of transparent flexible thin-film transistors using amorphous
oxide semiconductors", Nature, 432, 488 (2004).
31. J.H. Jeong, H.W. Yang, J-S. Park, J. K. Jeong, Y-G. Mo, H-D. Kim, J-Song, C.S.
Hwang, "Origin of Subthreshold Swing Improvement in Amorphous Indium Gallium
Zinc Oxide Transistors", Electrochem. Solid-State Lett., 11, H157 (2008).
32. H.Q. Chiang, J.F. Wager, R.L. Hoffman, J. Jeong, D.A. Keszler, "High mobility
transparent thin-film transistors with amorphous zinc tin oxide channel layer ", Appl.
Phys. Lett, 86, 013503 (2005).
33. R.L. Hoffman, "Effects of channel stoichiometry and processing temperature on the
electrical characteristics of zinc tin oxide thin-film transistors", Solid-State Electron.,
50, 784 (2006).
34. D. Hong, J.F. Wager, "Passivation of zinc–tin–oxide thin-film transistors", J. Vac. Sci.
Technol. B., 23, L25 (2005).
35. J. Triska, J.F. Conley, Jr., R. Presley, J.F. Wager, "Bias stress stability of zinc-tinoxide thin-film transistors with Al2O3 gate dielectrics", J. Vac. Sci. Technol. B., 28,
C5I1 (2010).
36. W.B. Jackson, R. Hoffman, B. Yeh, T. Emery, T. Koch, C. McConica, O. Kwon
"Metastability in multicomponent oxide transistors", Phys. Status Solidi A., 207, 695
(2010).
37. Y-H. Kim, H.S. Kim, J-I. Han, S.K. Park, "Solvent-mediated threshold voltage shift
in solution-processed transparent oxide thin-film transistors", Appl. Phys. Lett, 97,
092105 (2010).
120
38. S-J. Seo, S.C. Yang, J-H. Ko, B-S. Bae, "Effects of Sol-Gel Organic-Inorganic
Hybrid Passivation on Stability of Solution- Processed Zinc Tin Oxide Thin Film
Transistors", Electrochem. Solid-State Lett., 14, H375 (2011).
39. N.L. Dehuff, E.S. Kettenring, D. Hong, H.Q. Chiang, J.F. Wager, R.L. Hoffman, C.H. Park, D.A. Keszler, "Transparent thin-film transistors with zinc indium oxide
channel layer", J Appl. Phys., 97, 064505 (2005).
40. P. Barquinha, A. Pimentel, A. Marques, L. Pereira, R. Martins, E. Fortunato,
"Influence of the semiconductor thickness on the electrical properties of transparent
TFTs based on indium zinc oxide", J Non-Cryst Solids., 352, 1749 (2006).
41. C.G. Choi, S-J. Seo, B-S. Bae, "Solution-Processed Indium-Zinc Oxide Transparent
Thin-Film Transistors", Electrochem. Solid-State Lett., 11, H7 (2008).
42. D-H.Lee, Y-J.Chang, G. S. Herman, C-H.Chang, "A General Route to Printable
High-Mobility Transparent Amorphous Oxide Semiconductors", Adv Mater., 19, 843
(2007).
43. S-Y. Han, G. S. Herman, C-H.Chang, "Low Temperature, High-Performance,
Solution-Processed Indium Oxide Based Thin Film Transistors", ECS Transactions,
33, 275 (2010).
44. R.B.H. Tahar, T. Ban, Y. Ohya, Y. Takahashi, "Tin doped indium oxide thin films:
Electrical properties", J Appl. Phys, 83, 2631 (1998).
45. J. Jiang, J. Sun, W. Dou, B. Zhou, and Q. Wan, "In-plane-gate indium-tin-oxide thinfilm transistors self-assembled on paper substrates", Appl. Phys. Lett, 98, 113507
(2011).
46. R.J. Cava, J.M. Phillips, J. Kwo, G. A. Thomas, R. B. van Dover, S. A. Carter, J. J.
Krajewski, W.F. Peck, Jr., J. H. Marshall, and D. H. Rapkine, "GalnO3 - A new
transparent conducting oxide", Appl. Phys. Lett., 64, 2071 (1994).
47. H.Q. Chiang, D. Hong, C. M. Hung, R. E. Presley, J.F. Wager, C.-H. Park, D.A.
Keszler, and G.S. Herman, "Thin-film transistors with amorphous indium gallium
oxide channel layers", J. Vac. Sci. Technol. B, 24, 2702 (2006).
48. G. Goncalves, P. Barquinha, L. Pereira, N. Franco, E. Alves, R. Martins, E.
Fortunato, "High Mobility a-IGO Films Produced at Room Temperature and Their
Application in TFTs", Electrochem. and Solid-State Lett., 13, H20 (2010).
49. J.P.Uyemura, “CMOS Logic Circuit Design”, Kluwer Academic Publishers, 15,
(1999).
121
50. R.L. Hoffman, "ZnO-channel thin film transistors: Channel mobility," J. Appl. Phys.
95, 5813 (2004).
51. D. Hong, G. Yerubandi, H. Q. Chiang, M. C. Spiegelberg, J. F. Wager, "Electrical
Modeling of Thin-Film Transistors", Critical Reviews in Solid State and Materials
Sciences, 33, 101, (2008).
52. D.K. Schroder, "Semiconductor Materials and Device Characterization," New
York: John Wiley & Sons, Inc., 2 ed. (1998).
53. J.F. Conley, "Instabilities in Amorphous Oxide Semiconductor Thin-Film
Transistors", IEEE Transactions on Device and Materials Reliability", 10, 460,
(2010).
54. F.R. Libsch, J. Kanicki, "Bias-stress-induced stretched-exponential time dependence
of charge injection and trapping in amorphous thin-film transistors", Appl. Phys. Lett,
62, 1286, (1993).
55. J.S. Wright, W. Lim, D.P. Norton, S.J. Pearton, F. Ren, J.L. Johnson, A. Ural,
"Nitride and oxide semiconductor nanostructured hydrogen gas sensors",
Semiconductor. Sci. Technol., 25, 1 (2010).
56. G.F. Fine, L.M. Cavanagh, A. Afonja, R. Binions, "Metal Oxide Semi-Conductor Gas
Sensors in Environmental Monitoring", Sensors., 10, 5469 (2010).
57. T. Seiyama, A. Kato, K. Fujiishi, M. Nagatani, "A New Detector for Gaseous
Components Using Semiconductive Thin Films", Anal Chem, 34, 1502 (1962).
58. X.Li, R.Ramasamy, P.K.Dutta, “Study of the resistance behavior of anatase and rutile
thick films towards carbon monoxide and oxygen at high temperatures and
possibilities for sensing applications”, Sens Actuators. B., 143, 308 (2009).
59. N.G.Patel, K.K.Makhija, C.J.Panchal, “Fabrication of carbon dioxide gas sensor and
its alarm system using indium tin oxide (ITO) thin films”, Sens Actuators. B: Chem.,
21, 193 (1994).
60. M.S.Lee, J-U Meyer, “A new process for fabricating CO2 -sensing layers based on
BaTiO3 and additives”, Sens. Actuators. B, 68, 293 (2000).
61. A.A.Tomchenko, G.P.Harmer, B.T.Marquis, J.W.Allen, “Semiconducting metal
oxide sensor array for the selective detection of combustion gases”, Sens. Actuators.
B, 93, 126, (2003).
122
62. F.K. Chibirova, A.K. Avetisov, S.A. Kazakov, L.A. Obvintseva, M.V. Strobkova,
"Semiconductor Metal Oxide Sensors in Water and Water Based Biological Systems",
Sensors, 3, 451 (2003).
63. J. Zhou, N. Xu, Z. L. Wang, "Dissolving Behavior and Stability of ZnO Wires in
Biofluids: A Study on Biodegradability and Biocompatibility of ZnO
Nanostructures", Adv. Mater., 18, 2432 (2006).
64. A.D. Mason, C.-C. Huang, S. Kondo, M.T. Koesdjojo, Y.H. Tennico, V.T. Remcho,
J.F. Conley Jr., "Synthesis, functionalization, and environmental stabilization of ZnO
nanobridge transducers for gas and liquid-phase sensing", Sens. Actuators. B, 155,
245, (2011).
65. B. Zhang, T. Kong, W. Xu, R. Su, Y. Gao, and G. Cheng, "Surface Functionalization
of Zinc Oxide by Carboxyalkylphosphonic Acid Self-Assembled Monolayers",
Langmuir, 26, 4514, (2010).
66. F. Schreiber, "Structure and growth of self-assembling monolayers", Prog. Surf. Sci.,
65, 151 (2000).
67. J.R. Vig, J.W.L. Bus, "UV/Ozone cleaning of surfaces", IEEE Transactions on Parts,
Hybrids and Packaging, PHP-12, 365 (1976).
68. S.Franssila, “Introduction to micro-fabrication”, John Wiley & Sons, Ltd., 2nd
edition, (2010).
69. S. Middleman, “The effect of induced airflow on the spin coating of viscous liquids”,
J. Appl. Phys. 62, 2530 (1987).
70. D. Mitzi, "Solution processing of inorganic materials", New Jersey : John Wiley &
Sons, Inc., (2009).
71. F.E.H.Tay, “Materials & process Integration for MEMS”, Kluwer academic
publishers, (2002).
72. M.A. Lieberman, A.J. Lichtenberg, "Principles of plasma discharges and
materials processing", New York: John Wiley and Sons Inc., first ed., (1994).
73. Microchem, "SU-8 2000, permanent epoxy negative photoresist",
www.microchem.com.
74. J.C. Berg, “Wettability”, New York: Basel and Hong Kong: Marcel Dekker Inc.,
(1993).
75. Thermogravimetric analysis - www.scribd.com/doc/13729612/TGA.
123
76. L. Reimer, "Scanning Electron Microscope-Physics of image formation and
microanalysis", New York : Springer-Verlag Berlin Heidelberg, 2nd edition, (1998).
77. S. Reyntjens, R. Puers, "A review of focussed ion beam applications in microsystem
technology", J. Micromech. Microeng. 11, 287, (2001).
78. J. Morgan, J. Notte, R. Hill, B. Ward, "An introduction to Helium ion microscope",
Microscopy Today, 14, 24, (2006).
79. N. Economou, B. Ward, J. Morgan, J. Notte, "Helium ion microscopy: an
introduction", Innovations in Pharmatceutical Technology, 24, (2008).
80. B. Voutou, and E.C. Stefanaki, "Electron Microscopy: The Basics", Physics of
Advanced Materials Winter School, 1, (2008).
81. I. Chorkendorff, and J.W. Niemantsverdriet, "Concepts of Modern Catalysis and
Catalysis", WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, 2nd Edition,
(2007).
82. J.A.Woollam, B.Johs, C.M.Herzinger, J.Hilfiker, R.Synowicki, and
C.L.Bungay,“Overview of Variable Angle Spectroscopic Ellipsometry (VASE), Part
I: Basic Theory and Typical Applications”, Critical Reviews of Optical Science and
Technology CR72, 3 (1999).
83. D.A. Skoog, F.J. Holler, and S.R. Crouch, "Principles of instrumental analysis",
Thompson Brooks/cole, 6th edition, (2007).
84. J.W. Robison, E.M.S. Frame, and G.M. Frame II, "Undergraduate instrumental
analysis", New York : Marcel Dekker, 6th edition, (2005).
85. B. Sun, and H. Sirringhaus, Nano Letters. 5, 2408 (2005).
86. B. L. Zhu, C. S. Xie, A. H. Wang, J .Wu, R. Wu, and J. Liu, J. Mater. Sci. 42, 5416
(2007).
87. Z. L. Wang, Appl. Phys. A. 88, 7 (2007).
88. L. Luo, Y. Zhang, S. S. Mao, and L. Lin, Sensors and Actuators A 127, 201 (2006).
89. S. A. Shojaee, M. M. Shahraki, M. A. F. Sani, A. Nemati, and A. Yousefi, J. Mater.
Sci : Material Electron. 21, 571 (2010).
90. S. Lee, Y. Jeong, S. Jeong, J. Lee, M. Jeon, and J. Moon, Superlattices and
Microstructures 44, 761 (2008).
91. S. Kim, M. C. Jeong, B-Y. Oh, W. Lee, and J. M. Myoung, Journal of Crystal Growth
290, 485 (2006).
124
92. T. Minami, H. Nanto, and S. Takata, Appl .Phys. Lett. 41, 958 (1982).
93. Ya. I. Alivov, A. V. Chernykh, M. V. Chukichev, and R. Y. Korotkov, Thin Solid
Films 473, 241 (2005).
94. W. S. Khan, C. Cao, G. Nabi, R. Yao, and S. H. Bhatti, Journal of Alloys and
Compounds 506, 666 (2010).
95. J . J. Wu, and S. C. Liu, Adv. Mater. 14, 215 (2002).
96. S. J. Lim, S. Kwon and, H. Kim, Thin Solid Films 516, 1523 (2008).
97. Y. S. Kim, W. P. Tai, and S. J. Shu, Thin Solid Films 491, 153 (2005).
98. L. Znaidi, G. J. A. A. Soler Illia, S. Benyahia, C. Sanchez, and A. V. Kanaev, Thin
Solid Films 428, 257 (2003).
99. B. Sun, R. L. Peterson, H. Sirringhaus, and K. Mori, J. Phys. Chem. C. 111, 18831 (2007).
100. H-C. Huang, and T-E. Hsieh, Nanotechnology, 21, 295707 (2010).
101. S. T. Meyers, J. T. Anderson, C. M. Hung, J. Thompson, J. F. Wager, and D. A.
Keszler, J. Am. Chem. Soc. 130, 17603 (2008).
102. H. C. Cheng, C. F. Chen, and C. C. Lee, Thin Solid Films 498, 142 (2006).
103. Y. W. Chen, Q. Qiao, Y. C. Liu, and G. L. Yang, J. Phys. Chem. C. 113, 7497
(2009).
104. T. Y. Kim, J. Y. Kim, K. M. Senthil, E. K. Suh, and K. S. Nahm, J. Crystal Growth
270, 491 (2004).
105. C. F. Guo, Y. Wang, P. Jiang, S. Cao, J. Miao, Z. Zhang, and Q. Liu,
Nanotechnology 19, 445710 (2008).
106. X. Y. Kong, Y. Ding, and Z. L. Wang, J. Phys. Chem. B. 108, 570 (2004).
107. Y. Gui, C. Xie, Q. Zhang, M. Hu, J. Yu, and Z. Weng, Journal of Crystal Growth
289, 663 (2006).
108. S. Y. Han, G. S. Herman, and C. H. Chang, ECS Transactions 33, 275 (2010).
109. S. Y. Han, G. S. Herman, and C. H. Chang, J. Am. Chem. Soc. 133, 5166 (2010).
110. R. Nakamura, J. G. Lee, D. Tokozakura, H. Mori, and H. Nakajima, Materials Letters
61, 1060 (2007).
111. R. Wu, C. Xie, H. Xia, J. Hu, and A.Wang, Journal of Crystal Growth 217, 274
(2000).
125
112. R. Wu, C. Xie, J. Hu, H. Xia and A. Wang, Scripta Mater.43, 841 (2000).
113. L. Gunawan, and G. P. Johari, J. Phys. Chem C. 112, 20159 (2008).
114. K. Nomura, H. Ohta, A. Takagi, T. Kamiya, M. Hirano and H. Hosono: Roomtemperature fabrication of transparent flexible thin-film transistors using amorphous
oxide semiconductors. Nature 432, 488 (2004).
115. H. Hosono, N. Kikuchi, N. Ueda and H. Kawazoe: Working hypothesis to explore
novel wide band gap electrically conducting amorphous oxides and examples. Journal
of Non-Crystalline Solids 200, 165 (1996).
116. S. Narushima, M. Orita, M. Hirano and H. Hosono: Electronic structure and transport
properties in the transparent amorphous oxide semiconductor 2 CdO center dot GeO2.
Physical Review B 66, 8 (2002).
117. H.Q. Chiang, J.F. Wager, R.L. Hoffman, J. Jeong and D.A. Keszler: High mobility
transparent thin-film transistors with amorphous zinc tin oxide channel layer. Applied
Physics Letters 86, 3 (2005).
118. R.L. Hoffman, B.J. Norris and J.F. Wager: ZnO-based transparent thin-film
transistors. Applied Physics Letters 82, 733 (2003).
119. P. Gorrn, M. Sander, J. Meyer, M. Kroger, E. Becker, H.H. Johannes, W. Kowalsky
and T. Riedl: Towards see-through displays: Fully transparent thin-film transistors
driving transparent organic light-emitting diodes. Advanced Materials 18, 738 (2006).
120. P. Gorrn, P. Holzer, T. Riedl, W. Kowalsky, J. Wang, T. Weimann, P. Hinze and S.
Kipp: Stability of transparent zinc tin oxide transistors under bias stress. Applied
Physics Letters 90, 3 (2007).
121. P. Gorrn, F. Ghaffari, T. Riedl and W. Kowalsky: Zinc tin oxide based driver for
highly transparent active matrix OLED displays. Solid-State Electronics, 53, 329
(2009).
122. Y. Jeong, K. Song, D. Kim, C.Y. Koo and J. Moon: Bias Stress Stability of SolutionProcessed Zinc Tin Oxide Thin-Film Transistors. Journal of the Electrochemical
Society 156, H808 (2009).
123. J. Triska, J.F. Conley, R. Presley and J.F. Wager: Bias stress stability of zinc-tinoxide thin-film transistors with Al(2)O(3) gate dielectrics. Journal of Vacuum
Science & Technology B 28, C5I1 (2010).
124. C. Avis and J. Jang: A High Performance Inkjet Printed Zinc Tin Oxide Transparent
Thin-Film Transistor Manufactured at the Maximum Process Temperature of 300
126
degrees C and Its Stability Test. Electrochemical and Solid State Letters 14, J9
(2011).
125. R.B.M. Cross and M.M. De Souza: Investigating the stability of zinc oxide thin film
transistors. Applied Physics Letters 89, 3 (2006).
126. Y. Vygranenko, K. Wang and A. Nathan: Stable indium oxide thin-film transistors
with fast threshold voltage recovery. Applied Physics Letters 91, 3 (2007).
127. J.M. Lee, I.T. Cho, J.H. Lee and H.I. Kwon: Bias-stress-induced stretchedexponential time dependence of threshold voltage shift in InGaZnO thin film
transistors. Applied Physics Letters 93, 3 (2008).
128. J.K. Jeong, H.W. Yang, J.H. Jeong, Y.G. Mo and H.D. Kim: Origin of threshold
voltage instability in indium-gallium-zinc oxide thin film transistors. Applied Physics
Letters 93, 3 (2008).
129. K. Hoshino, D. Hong, H.Q. Chiang and J.F. Wager: Constant-Voltage-Bias Stress
Testing of a-IGZO Thin-Film Transistors. Ieee Transactions on Electron Devices 56,
1365 (2009).
130. S.Y. Sung, J.H. Choi, U.B. Han, K.C. Lee, J.H. Lee, J.J. Kim, W. Lim, S.J. Pearton,
D.P. Norton and Y.W. Heo: Effects of ambient atmosphere on the transfer
characteristics and gate-bias stress stability of amorphous indium-gallium-zinc oxide
thin-film transistors. Applied Physics Letters 96, 3 (2010).
131. D. Hong and J.F. Wager: Passivation of zinc-tin-oxide thin-film transistors. Journal
of Vacuum Science & Technology B 23, L25 (2005).
132. S.K. Park, Y.H. Kim, H.S. Kim and J.I. Han: High Performance Solution-Processed
and Lithographically Patterned Zinc-Tin Oxide Thin-Film Transistors with Good
Operational Stability. Electrochemical and Solid State Letters 12, H256 (2009).
133. J.S. Park, T.S. Kim, K.S. Son, K.H. Lee, W.J. Maeng, H.S. Kim, E.S. Kim, K.B.
Park, J.B. Seon, W. Choi, M.K. Ryu and S.Y. Lee: The influence of SiO(x) and
SiN(x) passivation on the negative bias stability of Hf-In-Zn-O thin film transistors
under illumination. Applied Physics Letters 96, 3 (2010).
134. D.A. Mourey, D.A.L. Zhao, J. Sun and T.N. Jackson: Fast PEALD ZnO Thin-Film
Transistor Circuits. Ieee Transactions on Electron Devices 57, 530 (2010).
135. P. Gorrn, T. Riedl and W. Kowalsky: Encapsulation of Zinc Tin Oxide Based Thin
Film Transistors. Journal of Physical Chemistry C 113, 11126 (2009).
127
136. S.J. Seo, S. Yang, J.H. Ko and B.S. Bae: Effects of Sol-Gel Organic-Inorganic
Hybrid Passivation on Stability of Solution-Processed Zinc Tin Oxide Thin Film
Transistors. Electrochemical and Solid State Letters 14, H375 (2011).
137. Y. Jung, T. Jun, A. Kim, K. Song, T.H. Yeo and J. Moon: Direct photopatternable
organic-inorganic hybrid gate dielectric for solution-processed flexible ZnO thin film
transistors. Journal of Materials Chemistry 21, 11879 (2011).
138. J. Scherer, M.R. Vogt, O.M. Magnussen and R.J. Behm: Corrosion of alkanethiolcovered Cu(100) surfaces in hydrochloric acid solution studied by in-situ scanning
tunnelling microscopy. Langmuir 13, 7045 (1997).
139. H. Amar, J. Benzakour, A. Derja, D. Villemin, B. Moreau and T. Braisaz: Piperidin1-yl-phosphonic acid and (4-phosphono-piperazin-1-yl) phosphonic acid: A new class
of iron corrosion inhibitors in sodium chloride 3% media. Applied Surface Science
252, 6162 (2006).
140. H.J. Lim, D.Y. Lee and Y.J. Oh: Gas sensing properties of ZnO thin films prepared
by microcontact printing. Sensors and Actuators a-Physical 125, 405 (2006).
141. T.L. Breen, P.M. Fryer, R.W. Nunes and M.E. Rothwell: Patterning indium tin oxide
and indium zinc oxide using microcontact printing and wet etching. Langmuir 18,
194 (2002).
142. C.G. Allen, D.J. Baker, J.M. Albin, H.E. Oertli, D.T. Gillaspie, D.C. Olson, T.E.
Furtak and R.T. Collins: Surface Modification of ZnO Using Triethoxysilane-Based
Molecules. Langmuir 24, 13393 (2008).
143. C. Haensch, S. Hoeppener and U.S. Schubert: Chemical modification of selfassembled silane based monolayers by surface reactions. Chemical Society Reviews
39, 2323 (2010).
144. H.P. Zhang and S. Baldelli: Alkanethiol monolayers at reduced and oxidized zinc
surfaces with corrosion proctection: A sum frequency generation and
electrochemistry investigation. Journal of Physical Chemistry B 110, 24062 (2006).
145. C. Nogues and P. Lang: Self-assembled alkanethiol monolayers on a Zn substrate:
Structure and organization. Langmuir 23, 8385 (2007).
146. T.C. Monson, M.T. Lloyd, D.C. Olson, Y.J. Lee and J.W.P. Hsu: Photocurrent
Enhancement in Polythiophene- and Alkanethiol-Modified ZnO Solar Cells.
Advanced Materials 20, 4755 (2008).
128
147. C.E. Taylor and D.K. Schwartz: Octadecanoic acid self-assembled monolayer growth
at sapphire surfaces. Langmuir 19, 2665 (2003).
148. C.L. Perkins: Molecular Anchors for Self-Assembled Monolayers on ZnO: A Direct
Comparison of the Thiol and Phosphonic Acid Moieties. Journal of Physical
Chemistry C 113, 18276 (2009).
149. P.J. Hotchkiss, M. Malicki, A.J. Giordano, N.R. Armstrong and S.R. Marder:
Characterization of phosphonic acid binding to zinc oxide. Journal of Materials
Chemistry 21, 3107 (2011).
150 I. Gouzman, M. Dubey, M.D. Carolus, J. Schwartz and S.L. Bernasek: Monolayer vs.
multilayer self-assembled alkylphosphonate films: X-ray photoelectron spectroscopy
studies. Surface Science 600, 773 (2006).
151. S.E. Koh, K.D. McDonald, D.H. Holt, C.S. Dulcey, J.A. Chaney and P.E. Pehrsson:
Phenylphosphonic acid functionalization of indium tin oxide: Surface chemistry and
work functions. Langmuir 22, 6249 (2006).
152. R. Luschtinetz, A.F. Oliveira, J. Frenzel, J.O. Joswig, G. Seifert and H.A. Duarte:
Adsorption of phosphonic and ethylphosphonic acid on aluminum oxide surfaces.
Surface Science 602, 1347 (2008).
153. C.W. Hsu, H.R. Liou, W.F. Su and L.Y. Wang: Self-assembled monolayers of 2(thienyl)hexylphosphonic acid on native oxide surface of silicon fabricated by airliquid interface-assisted method. Journal of Colloid and Interface Science 324, 236
(2008).
154. P.B. Paramonov, S.A. Paniagua, P.J. Hotchkiss, S.C. Jones, N.R. Armstrong, S.R.
Marder and J.L. Bredas: Theoretical Characterization of the Indium Tin Oxide
Surface and of Its Binding Sites for Adsorption of Phosphonic Acid Monolayers.
Chemistry of Materials 20, 5131 (2008).
155. O. Acton, G.G. Ting, P.J. Shamberger, F.S. Ohuchi, H. Ma and A.K.Y. Jen:
Dielectric Surface-Controlled Low-Voltage Organic Transistors via n-Alkyl
Phosphonic Acid Self-Assembled Monolayers on High-k Metal Oxide (vol 2, pg 511,
2010). Acs Applied Materials & Interfaces 2, 2963 (2010).
156. B.B. Zhang, T. Kong, W.Z. Xu, R.G. Su, Y.H. Gao and G.S. Cheng: Surface
Functionalization of Zinc Oxide by Carboxyalkylphosphonic Acid Self-Assembled
Monolayers. Langmuir 26, 4514 (2010).
129
157. M. Dubey, T. Weidner, L.J. Gamble and D.G. Castner: Structure and Order of
Phosphonic Acid-Based Self-Assembled Monolayers on Si(100). Langmuir 26, 14747
(2010).
158. M. Maxisch, P. Thissen, M. Giza and G. Grundmeier: Interface Chemistry and
Molecular Interactions of Phosphonic Acid Self-Assembled Monolayers on
Oxyhydroxide-Covered Aluminum in Humid Environments. Langmuir 27, 6042
(2011).
159. P.H. Mutin, G. Guerrero and A. Vioux: Hybrid materials from organophosphorus
coupling molecules. Journal of Materials Chemistry 15, 3761 (2005).
160. E.S. Gawalt, G. Lu, S.L. Bernasek and J. Schwartz: Enhanced bonding of
alkanephosphonic acids to oxidized titanium using surface-bound alkoxyzirconium
complex interfaces. Langmuir 15, 8929 (1999).
161. E.S. Gawalt, M.J. Avaltroni, N. Koch and J. Schwartz: Self-assembly and bonding of
alkanephosphonic acids on the native oxide surface of titanium. Langmuir 17, 5736
(2001).
162. G.A. Diaz-Quijada, C. Maynard, T. Comas, R. Monette, C. Py, A. Krantis and G.
Mealing: Surface Patterning with Chemisorbed Chemical Cues for Advancing
Neurochip Applications. Industrial & Engineering Chemistry Research 50, 10029
(2011).
163. M.G. Kim, H.S. Kim, Y.G. Ha, J.Q. He, M.G. Kanatzidis, A. Facchetti and T.J.
Marks: High-Performance Solution-Processed Amorphous Zinc-Indium-Tin Oxide
Thin-Film Transistors. Journal of the American Chemical Society 132, 10352 (2010).
164. S.J. Seo, Y.H. Hwang and B.S. Bae: Postannealing Process for Low Temperature
Processed Sol-Gel Zinc Tin Oxide Thin Film Transistors. Electrochemical and Solid
State Letters 13, H357 (2010).
165. B. Adolphi, E. Jahne, G. Busch and X.D. Cai: Characterization of the adsorption of
omega-(thiophene-3-yl alkyl) phosphonic acid on metal oxides with AR-XPS.
Analytical and Bioanalytical Chemistry 379, 646 (2004).
166. G. Mani, D.M. Johnson, D. Marton, V.L. Dougherty, M.D. Feldman, D. Patel, A.A.
Ayon and C.M. Agrawal: Stability of self-assembled monolayers on titanium and
gold. Langmuir 24, 6774 (2008).
167. Y. Xie, X. Zhao, Y. Chen, Q. Zhao and Q. Yuan: Preparation and characterization of
porous C-modified anatase titania films with visible light catalytic activity. Journal of
Solid State Chemistry 180, 3576 (2007).
130
168. N. Adden, L.J. Gamble, D.G. Castner, A. Hoffmann, G. Gross and H. Menzel:
Phosphonic acid monolayers for binding of bioactive molecules to titanium surfaces.
Langmuir 22, 8197 (2006).
169. Y. Sun, Z. Liu, F. Machuca, P. Pianetta and W.E. Spicer: Optimized cleaning method
for producing device quality InP(100) surfaces. Journal of Applied Physics 97, 7
(2005).
170. F.R. Libsch and J. Kanicki: Bias-stress-induced stretched-exponential timedependence of charge injection and trapping in amorphous thin-film transistors.
Applied Physics Letters 62, 1286 (1993).
171. M.J. Powell, C. Vanberkel and J.R. Hughes: Time and temperature-dependence of
instability mechanisms in amorphous-silicon thin-film transistors. Applied Physics
Letters 54, 1323 (1989).
172. A. Suresh and J.F. Muth: Bias stress stability of indium gallium zinc oxide channel
based transparent thin film transistors. Applied Physics Letters 92, 3 (2008).
173. S.G.J. Mathijssen, M. Colle, H. Gomes, E.C.P. Smits, B. de Boer, I. McCulloch, P.A.
Bobbert and D.M. de Leeuw: Dynamics of threshold voltage shifts in organic and
amorphous silicon field-effect transistors. Advanced Materials 19, 2785 (2007).
174. Y.C. Chen, T.C. Chang, H.W. Li, S.C. Chen, J. Lu, W.F. Chung, Y.H. Tai and T.Y.
Tseng: Bias-induced oxygen adsorption in zinc tin oxide thin film transistors under
dynamic stress. Applied Physics Letters 96, 3 (2010).
175. M. Batzill and U. Diebold: The surface and materials science of tin oxide. Progress in
Surface Science 79, 47 (2005).
176. J.S. Park, J.K. Jeong, H.J. Chung, Y.G. Mo and H.D. Kim: Electronic transport
properties of amorphous indium-gallium-zinc oxide semiconductor upon exposure to
water. Applied Physics Letters 92, 3 (2008).
131