Selective and Non-selective Synthesis of Carbon Nanotubes (CNTs) by
Chemical Vapor Deposition (CVD) Characterization: Catalysts and
Underlayers E ects on Field Emission Properties
by
Chung-Nan Tsai
A thesis submitted to the Graduate Faculty of
Auburn University
in partial ful llment of the
requirements for the Degree of
Master of Science
Auburn, Alabama
Dec 08, 2012
Keywords: Carbon Nanotube, Chemical Vapor Deposition, Field Emission
Copyright 2012 by Chung-Nan Tsai
Approved by
Hulya Kirkici, Chair, Professor of Electrical and Computer Engineering
Guofu Niu, Professor of Electrical and Computer Engineering
Michael Baginski, Associate Professor of Electrical and Computer Engineering
Abstract
Since the discovery of carbon nanotubes (CNTs), they have been attracted much at-
tention with abundant potential applications based on their outstanding properties. CNTs
are well-known for their superior mechanical strength and low weight, excellent heat con-
ductance, and varying electronic properties depending on their helicity and diameter. In
particular, the recent research studies have reported that CNTs have excellent electrical
 eld emission properties, with high emission currents at low electric  eld strength due to the
high aspect ratio (small diameter and relatively long length). As a result, CNTs are consid-
ered as one of the promising materials as cold-cathode  eld emission sources, especially for
application requiring high-current densities and lightweight packaging.
In this research work, the selective and non-selective multi-wall CNTs (MWCNTs) are
grown by using chemical vapor deposition (CVD) technique. Then, their  eld emission
properties are examined in a high pressure vacuum chamber of around 10 7 to 10 6 Torr.
MWCNTs are grown onto various underlying layers such as SiO2,Ti ,and W-coated silicon
substrates. Thermal CVD furnace containing gas mixtures of acetylene and argon is used to
grow CNTs. The growth conditions such as catalyst types and thickness, gas  ow rate and
deposition temperature are discussed. E ects of di erent catalysts with various underlayers
on the  eld emission properties of CNTs are studied and results are presented. The mea-
surement results indicate that CNTs have signi cant  eld emission capabilities to be used
as cold cathode materials.
ii
Acknowledgments
First of all, with a deep sense of gratitude, I wish to thank my thesis advisor, Prof. Hulya
Kirkici, for her guidance throught my research work, for her encougement of my professional
development, and for her advise on organizing and writing this thesis. I am also very grateful
to Prof. Guofu Niu and Prof. Michael Baginski for their time to review my thesis.
During the period of my M.S. study in Auburn, I would like to thank Dr. Yu-Chun
(Brad) Chen who spent enormous time with me to discuss my research in details and was
willing to share his life experiences. I also thank all my friends for their friendship, help and
support, who are Dr. Haitao Zhao, Ms. Huirong Li, Mr. Zhenhong Li, Mr. Ming Zhang
and Mr. George Hernandez.
Most of all, I thank my parents, Hsueh-Jung Tsai and I-Min Liao, for their love, encour-
agement, dedication, and support. I also thank my girlfriend, Yueh-Chuan Liu, who always
believe in me, encourage me to try out my own life, and give me all her support whenever I
need it. I would like dedicate this thesis to my family.
iii
Table of Contents
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x
1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 LITERATURE REVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.1 Material properties of carbon nanotubes . . . . . . . . . . . . . . . . . . . . 5
2.1.1 Structure of carbon nanotubes . . . . . . . . . . . . . . . . . . . . . . 5
2.1.2 Electrical properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1.3 Mechanical properties . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1.4 Other properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2 Synthesis techniques of carbon nanotubes . . . . . . . . . . . . . . . . . . . . 11
2.2.1 Arc discharge and laser ablation . . . . . . . . . . . . . . . . . . . . 11
2.2.2 Chemical vapor deposition . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3 Growth mechanisms of CNTs . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.4 Field emission of electrons from CNTs . . . . . . . . . . . . . . . . . . . . . 18
2.5 Potential applications of CNTs . . . . . . . . . . . . . . . . . . . . . . . . . 21
3 Carbon Nanotubes: FABRICATION AND CHARACTERIZATION . . . . . . . 25
3.0.1 Growth process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.0.2 Selective grown CNTs synthesis processes . . . . . . . . . . . . . . . 28
3.1 Electron  eld emission measurements and experimental setup . . . . . . . . . 30
4 RESULTS AND DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.1 In uence of catalyst layer thickness . . . . . . . . . . . . . . . . . . . . . . . 32
iv
4.2 In uence of underlying layer . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.3 In uence of the deposition temperature . . . . . . . . . . . . . . . . . . . . . 33
4.4 Fowler-Nordheim curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.5 Scanning electron microscope image . . . . . . . . . . . . . . . . . . . . . . . 35
4.5.1 SEM images of CNTs with di erent catalysts sputtering time and un-
derlayers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5 SUMMARY AND FUTURE DIRECTIONS . . . . . . . . . . . . . . . . . . . . 70
v
List of Figures
1.1 Schematic of an individual layer of honeycomb-like carbon called graphene, rolling
into CNT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Atomic structures of carbon nanotubes . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 Base and tip growth of CNTs rooted in a nanoporous (e.g. zeolite) substrate . . 4
2.1 sp2 hybridization of carbon and its derived materials. (a) The three sp2 hy-
bridized orbital are in-plane, with 2p orbital orthogonal to the plane,  and  ?
denotes the bonding and anti bonding orbital. (b) Graphene as the source of
three di erent materials, fullerence (left), carbon nanotube (center) and bulk
graphite (right) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Chiral structure of carbon nanotube . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 A two-dimensional honeycomb lattice of graphene sheet. The vectors OA and
OB de ne the chiral vector Ch. Unitary vector a1 and a2 are to determine the
rolling direction expressed by vector Ch. There are several ways to roll it up As
a result, di erent types of tubules can be formed . . . . . . . . . . . . . . . . . 8
2.4 Schematic diagram of the arc apparatus where the nanotubes are formed from
the plasma between the two carbon rods . . . . . . . . . . . . . . . . . . . . . . 12
2.5 Schematic diagram of a laser ablation set-up . . . . . . . . . . . . . . . . . . . . 13
2.6 Schematic diagram for thermal CVD reactor . . . . . . . . . . . . . . . . . . . . 16
vi
2.7 (a) tip-growth model (b) base-growth model . . . . . . . . . . . . . . . . . . . . 18
2.8 Typical set-up for  eld emission: a potential di erence is applied between a
nanotube and a counter electrode . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.9 TEM image of an individual SWNT tip produced by controlling carefully the
catalyst density. The scale bar equals 10 nm . . . . . . . . . . . . . . . . . . . . 23
2.10 Schematic process for the fabrication of a CNTFET with a suspended graphene
gate without any Si3N4 protective layer: The CNTFET can be gated either by
the suspended graphene gate or by the Si substrate acting as a back gate . . . . 24
3.1 Schematic diagram for thermal CVD reactor . . . . . . . . . . . . . . . . . . . . 27
3.2 Schematic diagram of CNTs fabrication processing (a) silicon wafer cleaned and
prepared, (b) mask is aligned to remove parts of the photoresist for patterned
deposition of the iron catalyst. (c) the spattering process is conducted. (d)
photoresist is removed from the wafer exposing only the sections with catalyst
Fe. The sample is now ready to be moved to the CVD chamber for CNT growth
process. (e) the CNTs are selectively grown on the substrate. . . . . . . . . . . 29
3.3 Schematic diagram for the  eld emission measurement setup . . . . . . . . . . . 31
4.1 1min Fe sputtering time on SiO2 underlayer . . . . . . . . . . . . . . . . . . . . 36
4.2 5min Fe sputtering time on SiO2 underlayer . . . . . . . . . . . . . . . . . . . . 36
4.3 5min Fe with Carbon 8 min sputtering time on SiO2 underlayer . . . . . . . . . 37
4.4 5min Fe sputtering time on SiO2 underlayer . . . . . . . . . . . . . . . . . . . . 37
4.5 15 sec Cobalt sputtering time on SiO2 underlayer . . . . . . . . . . . . . . . . . 38
vii
4.6 30 sec Cobalt sputtering time on SiO2 underlayer . . . . . . . . . . . . . . . . . 38
4.7 Cobalt sputtering time above 1min on SiO2 underlayer . . . . . . . . . . . . . . 39
4.8 SEM image of vertically-aligned multi-wall carbon nanotubes (VA-MWCNTs). . 40
4.9 Cross-sectional SEM image of randomly-oriented MWCNTs. . . . . . . . . . . . 41
4.10 SEM image of patterned CNTs array (a) 25 m circle array with 25 m spacing,
(b) 25 m diameter circle array with 50 m spacing, (c) 50 m  50 m square
array with 25 m spacing, (d) a close-up image of a single CNT bundle (from a
circle array sample) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.11 Field emission curve for growing CNTs on SiO2 using Fe as catalyst w/o pre-
treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.12 Fowler-Nordeim curve of CNT grown on SiO2 using Fe as catalyst w/o pre-
treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.13 Field emission curve for growing CNTs on SiO2 using Fe as catalyst with pre-
treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4.14 Fowler-Nordeim curve of CNT grown on SiO2 using Fe as catalyst with pre-
treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.15 Field emission curve for growing CNTs on SiO2 using Co as catalyst w/o pre-
treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.16 Fowler-Nordeim curve of CNT grown on SiO2 using Co as catalyst w/o pre-
treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
4.17 Field emission curve for growing CNTs on plain silicon using Fe as catalyst w/o
pre-treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
viii
4.18 Fowler-Nordeim curve of CNT grown on plain silicon using Fe as catalyst w/o
pre-treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
4.19 Field emission curve for growing CNTs on plain silicon using Fe as catalyst with
pre-treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.20 Fowler-Nordeim curve of CNT grown on plain silicon using Fe as catalyst with
pre-treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.21 Field emission curve for growing CNTs on Ti underlying layer using Fe as catalyst
w/o pre-treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.22 Fowler-Nordeim curve of CNT grown on plain silicon using Fe as catalyst w/o
pre-treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.23 Field emission curve for growing CNTs on Ti underlying layer using Fe as catalyst
with pre-treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.24 Fowler-Nordeim curve of CNT grown on Ti underlying layer using Fe as catalyst
with pre-treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.25 Field emission curve for growing selective CNTs on plain silicon using Fe as
catalyst w/o pre-treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.26 Fowler-Nordeim curve of selective CNT grown on plain silicon using Fe as catalyst
w/o pre-treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
ix
List of Tables
4.1 Various Structural Properties of Catalyst-Underlying-Layer Combinations . . . 32
4.2 Iron catalyst vs. sputtering time . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.3 Cobalt catalyst vs. sputtering time . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.4 Various Structural Properties of Catalyst-Underlying-Layer Combinations . . . 43
4.5 Various Structural Properties of Catalyst-Underlying-Layer Combinations . . . 44
4.6 MWNTs growth on the catalyst-coated(Fe) substrate w/ SiO2 underlayer . . . . 45
4.7 MWNTs growth on the catalyst-coated(Co) substrate w/ SiO2 underlayer . . . 50
4.8 MWNTs growth on the catalyst-coated(Fe) substrate w/o SiO2 underlayer . . . 53
4.9 MWNTs growth on the catalyst-coated(Co) substrate w/o SiO2 underlayer . . 58
4.10 MWNTs growth on the catalyst(Fe)-underlying Ti layer-coated substrate . . . . 59
4.11 MWNTs growth on the catalyst(Co)-underlying Ti layer-coated substrate . . . . 64
4.12 MWNTs growth on the catalyst(Fe)-underlying W layer-coated substrate . . . . 65
4.13 MWNTs growth on the catalyst(Co)-underlying W layer-coated substrate . . . 66
4.14 Selective growth of MWNTs on the catalyst(Fe)-coated substrate w/o SiO2 . . 67
x
Chapter 1
INTRODUCTION
Since the  rst amazing discovery of carbon nanotubes (CNTs) by Japanese physicist
Iijima in 1991 using arc discharge method [1], they have attracted considerable interest
because of their extraordinary properties. CNTs are known for their superior mechanical
strength, low weight, good heat conductance, and varying electronic properties depending
on their helicity and diameter. In particular, the recent research studies [2{4] have reported
that CNTs have excellent electrical  eld emission properties, with high emission currents at
low electric  eld strength. The turn-on voltage of CNTs can be as low as 1-3 V/ m and
emission current can be as high as 0.1mA from a single nanotube [5]. As a result, CNT are
considered as one of the promising materials as cold-cathode  eld emission sources, especially
for application requiring high-current densities and lightweight packaging.
CNTs are essentially tubes with the diameter of nanometer and made of carbon. One
could think of it as a graphene sheet rolled upinto tubes. The structure of a graphene sheet
is shown in Figure 1.1 [6], which is a planar sheet of carbon atoms that are densely packed in
a honeycomb crystal lattice. CNTs are distinguished by their numbers of concentric layers
(\wall"), with spacing  0:34nm, and their chirality (wrapping angle). They are typically
categorized as single-wall CNTs (SWCNTs), double-wall CNTs (DWCNTs), and multi-wall
CNTs (MWCNTs) with respect to the numbers of graphitic layer, as shown in Figure 1.2 [7].
The typical diameter of SWCNTs are 0.4 to 5nm, and MWCNTs can be up to 100nm. The
number of sidewalls of CNTs can can be precisouly controlled by the preparation of catalyst
that serves as the CNT growth sites that is deposited onto the substrate prior to the growth.
There are two general growth mechanisms that have been proposed [8]. When the
catalyst particles remain anchored to the substrate, this is known as \base-growth" model
1
Figure 1.1: Schematic of an individual layer of honeycomb-like carbon called graphene,
rolling into CNT [6]
(Figure 1.3 [8]). On the other hand, the growth follows a \tip-growth" model when the
metal catalyst particles lift o from the substrate and remain at the top of the CNTs during
growth. In both cases, carbon is added at the catalyst site for CNTs to grow.
Currently, there are several synthesis techniques proposed for growing CNTs. These
can be divided into methods where CNTs grow from transition metal catalyst particles, and
method where CNTs grow without catalyst particles. They include carbon arc discharge [9],
laser ablation [10], and chemical vapor deposition [11].
Numbers of potential applications of CNTs have been widely proposed. Among these,
CNTs have been reported to be a promosing materials for cold-cathode applications [12].
Because of their high aspect ratio, electrons can be emitted easily from the tips of CNTs in a
vacuum ambient under a relatively low applied electric  eld. Also, the high current density
of CNTs is the other important property making them attractive [13].
2
Figure 1.2: Atomic structures of carbon nanotubes [7]
The purpose of this study is to investigating the growth condition of CNTs on di er-
ent underlaying layers and catalyst types and characterize their  eld emission properties.
From the experimental results, It is found that underlaying layers play an criticle role for
CNTs growth. SiO2, Titanium and Tungsten are used. Each of them has substantially
di erent performances as a substrate for CNTs growth. In addition to investigation of un-
derlayers, the transitation metals, Iron and Cobalt, are utilized as catalysts for depositing
CNTs. Randomly-oriented and vertically-aligned CNTs can be grown by controlling the ex-
perimental parameters such as catalyst type, catalyst thickness, and deposition temperature.
Moreover, di erent sizes of patterned CNTs are also successfully grown by utilizing basic
photolithography steps.
3
Figure 1.3: Base and tip growth of CNTs rooted in a nanoporous (e.g. zeolite) substrate [8]
4
Chapter 2
LITERATURE REVIEW
2.1 Material properties of carbon nanotubes
2.1.1 Structure of carbon nanotubes
Carbon, the group IV element, is very active in producing many molecular compounds
and crystalling solids. A carbon atom has six electrons which occupy 1s2, 2s2, and 2p2 atomic
orbital and can hybridize in sp, sp2 or sp3 forms. It has been used for centries, but yet it
has been stimulated in the nanotechnology  eld by the discovery of these unique nanometer
sizes of sp2 carbon-bonded materials such as fullerene, carbon nanotube and graphite, as
shown in Figure 2.1 [14].
Figure 2.1: sp2 hybridization of carbon and its derived materials. (a) The three sp2 hy-
bridized orbital are in-plane, with 2p orbital orthogonal to the plane,  and  ? denotes the
bonding and anti bonding orbital. (b) Graphene as the source of three di erent materials,
fullerence (left), carbon nanotube (center) and bulk graphite (right) [14]
5
Carbon nanotube (CNT), a new form of pure carbon, can be thought as a hexagonal
graphene sheet rolled up to form a cylinder that is capped by pentagonal carbon rign.
Depending on the manner in which the graphene sheet is rolled up, the arrangement of
carbon atoms along the cylinder circumference can be \arm-chair", \zig-zag", or several
di erent intermediate chiral structure (Figure 2.2) [15].
Figure 2.2: Chiral structure of carbon nanotube [15]
CNTs are distinguished by their numbers of concentric layers called the "walls", with
spacing of sub-nm, and their chirality. There are several types of CNTs: Single-Wall CNTs
(SWCNTs), Double-Wall CNTs (DWCNTs), and Multi-Wall CNTs (MWCNTs) with respect
to the number of graphitic layers (Figure 1.2). Unlike graphene, CNTs exhibit di erent phys-
ical properties depending upon their structure. The crystal structure of a nanotube depends
upon the axis along which the cylinder is formed from the graphene sheet. Figure 2.3 [16]
shows the vectors on graphene plane that are important in understanding the formation of
6
a nanotube . Here, we will follow the established notations, which can be found in litera-
ture [17]. The electrical character of a nanotube is speci ed by two numbers that determine
the chirality of the nanotube. The vector OB perpendicular to the nanotube axis is called
chiral vector  !Ch. The vector OA, which is parallel to the axis is termed as the translational
vector T, this is the unit vector of 1-D nanotube. The chiral vector is !Ch = n !a1 + m !a2 and
the chiral numbers n and m are integers (the chirality convention requires 0 jmj n). The
length of the unit vectors is a and the angle they enclose is 60 . As a result, the diameter of
the nanotube can be expressed as:
d =
 !C
h
 =
apm2 +mn+n2
 
and the chiral angle from the  gure with the expression:
cos =
 !C
h !a1
j !Chjj !a1 j =
2n+m
2pn2 +m2 +nm
Due to the hexagonal symmetry of the lattice, the chiral angle can only take on values
between 0 and 30 . The electron wavevector along the circumference of the CNT is quantized
while the graphene sheet is rolled. It is shown that this condition leads to a metallic nanotube,
zero band gap, when the di erence n-m is a multiple of 3 [17].
7
Figure 2.3: A two-dimensional honeycomb lattice of graphene sheet. The vectors OA and OB
de ne the chiral vector Ch. Unitary vector a1 and a2 are to determine the rolling direction
expressed by vector Ch. There are several ways to roll it up As a result, di erent types of
tubules can be formed [16].
8
2.1.2 Electrical properties
CNTs can be either metallic or semiconducting depending on the con guration and
molecular structure [18]. For example, SWCNT is metallic if the structure is armchair. On
the other hand, zigzag tube as well as chiral-type tube (m,n) with 2m+n=3N (N: positive
integer), is a narrow-gap semiconductive [19]. In contrast, the band structures of MWCNTs
are more sophisticated because of the interlayer coupling. However, theoretical discussion
by Saito et al. emphasized that the interlayer has little e ect on the electronic properties of
individual tubes [20]. As a result, two coaxial armchair nanotubes yield a DWCNT. Coaxial
metallic-semiconducting and semiconducting-metallic tube will retain their respective char-
acters when interlayer interaction is introduced [19].
Several techniques have been utilized in determining the properties of CNTs, such as
raman spectroscopy [21], electron energy loss spectroscopy (EELS) [22], electron spin reso-
nance (ESR) [23], scanning tunneling microscopy (STM) [24], four-point probe, and atomic
force microscopy (AFM).
Many groups worked on characterizing the conductivity of CNTs. Ebbesen et al [25]
measured the resistivities of MWCNTs using four-point probe technique and determined the
resistivties of nanotubes ranged between 8m -m and 0.051   -m. The results show that
the electrical resistivity of CNTs can vary greatly according to their porposed structure.
The resistivity of SWCNTs have also been measured by means of four-probe arrangement
by Smalley et al. [26], and ranged from 0.34  -m to 1.0  -m. The range of resistivities for
SWCNTs is smaller than MWCNTs because it is not very conductive between the multi-
layers of MWCNTs.
2.1.3 Mechanical properties
CNT are among the strongest materials in nature, especially in the axial direction.
Because of their high strength-to-weight (STW) ratio, CNTs have been concluded as one of
the sti est materials. Also, traditional carbon nano bers have the the STW ratio of 40 times
9
greater than steel, however, CNTs have STW ratio at least 2 orders of magnitude greater
than steel [19]. According to Lourie et. al [27], SWCNTs have Young?s modulus of 2.8 - 3.6
TPa, and MWCNTs for 1.7 - 2.4 TPa. In addition, direct tensile loading tests of SWCNTs
and MWCNTs have been reported by Yu et al [28] . The Yong?s modulus can be obtained
ranging from 320 to 1470 GPa for SWCNTs and from 270 to 950 GPa for MWCNTs [17].
Theoretical studies have suggested that SWCNTs have Young?s modulus as high as 1 - 5
TPa [29]. For MWCNTs, the strength would be a ected by the sliding of individual graphene
cylinders with respect to each other. Treacy et al. [30] were the  rst to report  tting Young?s
modulus of MWCNTs to experimental data. For a total of 11 MWCNTs? Young?s modulus
values were reported as ranging from 0.4 to 4.15 TPa with a mean of 1.8 TPa. A similar
experimental research on SWCNTs was reported by Krishnan et al [31], who presented an
average Young?s modulus of 1.3 TPa from measured amplitudes of 27 SWNTs.
2.1.4 Other properties
In addition to the electrical and mechanical properties, there are other outstanding
properteis that make CNTs attractive such as optical properties, high thermal conductivity
,and high resistance to chemical attacks. As a result, CNTs can be applied on a variety of
technological applications. For example, the researchers shows that the low-density vertcially
aligned CNTs arrays can be engineered to have an extremely low index of refraction and
combined with the nanoscale surface roughness of the arrays , can produce a near-perfect
optical absoprtion materials [32]. Moreover, CNTs have been demonstrated that can be the
fabricated as the darkest materials. [33]. As a result, CNTs will be the excellent candidate
as the next-generation solar cell materials.
10
2.2 Synthesis techniques of carbon nanotubes
Currently, CNTs can be synthesized by using a variety of techniques such as arc-
discharge, laser ablation, chemical vapor depsotion, plasma-enhanced CVD, and so on. In
this section, the methods to produce MWCNTs and SWCNTs are summarized.
2.2.1 Arc discharge and laser ablation
Arc discharge and laser ablation methods for growing CNTs have been widely pursued
in the past years. Both synthesis techniques are based on condensing carbon atoms from
evaporation of solid carbon sources onto substrate. Temperatures in both methods are close
to the melting temperature of graphite, 3000 4000 C.
In arc-discharge method, carbon atoms are evaporated by plasma of helium gas ignited
by high currents passed through opposing carbon anode and cathode. The arc-discharge
technique has been developed into an excellent method for producing both high quality
MWCNTs and SWCNTs. In 1992, the DC-arc technique was  rst done by Ebbesen et
al. [34] and become a common scienti c method for synthesizing MWCNTs. MWCNTs have
lengths of on the order of ten microns and diameter in the range of 5 - 30nm. A schemtaic
diagram of DC arc-discharge method used for synthesizing CNTs is shown in Figure 2.5 [16].
The arc is generated between two pure graphite electrodes. The positive electrode must be
replaced with a new one before doing any new deposition because it is consumed in the arc.
Helium is a typical gas used as  lling up the chamber because it has low ionization potential.
It  ows though the chamber from the inlet a at a desired pressure of around 500 torr. Since
plasma is necessary in this methode, the type and pressure of the gas surrounding the arc
becomes critical. The typical applied DC voltage is 20 V. The positive electrode is brought
closer to negative one until arcing occurs. A plasma with a temperature up to 3700 C is
formed. Consequently, the variation of temperature might cause the large size distribution
of CNTs. The deposit forms on the negative electrode where there is a current  owing. It
11
has been shown that lower the current  ow, the better the yields of CNTs are. With this
process, a following puri cation procedure is needed to extract CNT samples [35].
Figure 2.4: Schematic diagram of the arc apparatus where the nanotubes are formed from
the plasma between the two carbon rods [16]
12
The growth of high-quality SWCNTs at the 1 - 10g scale was achieved by Smalley using a
laser ablation method. The schematic diagram is shown in Figure 2.6 [36]. The method took
advantage of intense laser pulses to ablate a carbon target which contains 0.5 atomic percent
of nickel and cobalt. The target was palced in a tube-furnace heated to 1200 C. During the
processing of laser ablation, a  ow of inert gas was passed through the chamber to carry the
grown nanotubes downstream to be collected on a cold  nger. The synthesized SWCNTs
are mostly in the form of ropes consisting of tens of individual nanotubes close-packed into
hexagonal crystals via Vander Waals interactions. In growth SWCNTs by means of arc-
discharge and laser ablation, typical by-products including fullerenes, graphitic polyhedrons
with enclosed metal particles, and amorphous carbon in the form of particles or overcoating
on the sidewalls of nanotubes have been seen [17].
Figure 2.5: Schematic diagram of a laser ablation set-up [36]
13
2.2.2 Chemical vapor deposition
Among these synthetic techniques, chemical vapor deposition (CVD) method is espe-
cially attractive because it can be easily scaled up. The synthesis of CNTs by CVD method
requires the presence of a gaseous phase activated carbon. It is common to use gaseous carbon
sources including methane, acetylene, and carbon monoxide. But also alcohols and carbon
clusters derived from solid carbon forms can be used. The activation of the molecules or of
the nanostructured fragments is achieved using a variety of methods which can be roughly
categorized as: [37]
1) Plasma CVD
2) Thermal CVD
CNTs synthesis by CVD is a two-step process consisting of a preliminary catalyst prepa-
ration step followed by actual synthesis of the nanotubes. In general, CVD technique tends
to produce nanotubes with fewer carbonaceous impurities with respect to the other synthetic
techniques and the residual particles of the metal catalyst are frequently found at an extrem-
ity of the nanotube, making their elimination by post-synthesis chemical processes easier.
The choice of metal cayalyst, usually a  rst-row transition metal such as Ni, Fe, or Co can
drive the process toward the preferential growth of single rather than multiwall nanotubes
and to control the formation of individual or bundled nanotubes. The dimensions of the
catalyst are very important: large particles can produce MWCNTs, but if the particle size
of the metal or metallic alloy is too large, carbon  laments or  bers can be produced instead
of nanotubes [37].
In plasma enhanced CVD technique, the plasmas is used to decompose and activate
the reactants in the gas phase. Plasma is often generated by hot- lament or by electrical
discharges at di erent frequencies (DC, RF, MW). Plasma-enhanced CVD (PECVD) is a
method that can be easily scaled-up. In PECVD, the catalyst is supported by the substrate
14
and can be prepared by wet chemistry or by a sputtering process. In the  rst case, a
solution containing a metal compound is deposited by casting onto the substrate surface. In
the second case, a layer of metal is deposited on the substrate by sputtering. Both processes
are followed by either chemical etching or thermal annealing to induce catalyst clustering
and particle formation on the substrate. The reactive carbon species in the gas phase then
di use towards the substrate, which is generally heated between 650 to 1500 C temperature.
The pressure in the deposition reactors is typically low (< 100 Torr).
In a hot  lament reactor, the reactants are activated by a heated  lament that induces
the formation of radicals and active species: the reaction product condense on a substrate
forming a deposit. The fundamental components of a typical hot  lament CVD (HFCVD)
system for the growth of nanotubes are: 1) A vcauum chamber that the pressure of the
reaction gases is maintained at values between 10 and 300 Torr. 2) A heated  lament located
at a distance of a few mm from the substrate on which the material must be deposited. 3)
An additional system for heating the substrate up to 800 1500 C.
The thermal CVD method involves the decomposition of a gaseous or volatile compound
of carbon, catalyzed by metallic nanoparticles, which also serve as nucleation sites for the
initiation of nanotube growth. Abundant carbon gas mixed with argon or nitrogen gases are
used, but sometimes the starting material can be a liquid that is consequently vaporized.
The catalyst can be either in solid form, supported on a previously coated substrate, or
mixed with the feed gas, and  owed into the reactor. The deposition reactors are typically
maintained under atmospheric pressure, even if some intereting processes can be carried out
under high-pressure condition. It has the advantage of being remarkably cheap and o er the
posibility of being easily scaled-up. One disadvantage is the presence of deposits of a large
amount of residual metallic catalyst. The general scheme of a thermal CVD apparatus for
carbon nanotubes synthesis is shown in Figure 2.8 [38].
15
Figure 2.6: Schematic diagram for thermal CVD reactor [38]
In thermal CVD, the catalyst has a strong e ect on the growth rate and on the  nal
nanotube yield [39]. Therefore, various techniques have been developed for the preparation of
catalyst. It is often used in bulk quantities, and to the choice of a suitable catalyst support.
For example, the use of metallic alloys to catalyze for the nanotube growth often results
in an enhancement of the yields. Nanoparticles of bimetallic alloys, as for exmple Fe and
Co, give 10 - 100 times higher yield of SWNTs than pure Fe [40]. The e ect of deposition
temperature and of the metal particle concentration on the deposit morphology has also
been investigated [41]. When growing CNTs by thermal CVD on a catalyst supported by a
substrate, the CNT diameters are often found to be dependent on the  lm thickness or on
the particle size. For instance, a research group reported [42] that using substrates coated
by a metal  lm with a thickness of 13 and 27nm, the diameter distribution resulted in the
ranges of 30 - 40 nm and 100 - 200 nm respectively.
16
2.3 Growth mechanisms of CNTs
Growth mechanism is always an interesting topic that has been debatable right from
its discovery. Several research groups have reported a variety of possibilities which are often
contradicting. As a result, there is no well established CNT growth mechanism. However,
there are two widely accepted mechanism can be categorized as tip-growth model and base-
growth model. Hydrocarbon vapor when comes in contact with the hot metal nanoparticels,
 rst decomposes into carbon and hydrogen species; hydrogen  ies away and carbon gets
dissolved into the metal. After reaching the carbon-solubility limit in the metal at that
temperatutre, as-dissolved carbon precipitates out and crystallizes in the form of a cylindrical
network having no dangling bonds [43]. When the catalyst-substrate interaction is weak,
hydrocarbon decomposes on the top surface of the metal; carbon di uses down through the
metal; and CNT precipitates out across the metal bottom, pushing the whole metal particle
o the substrate. This is described in picture 2.9 (a) As long as the metal?s top is open
for fresh hydrocarbon decomposition, CNT continues to grow longer and longer. Once the
metal is fully covered with excess carbon, its catalytic activity ceases and the CNT growth
is stopped. This is known as tip-growth model (Figure 2.9(a) [43]).
On the other hand, when the catalyst-substrate interaction is strong, initial hydrocarbon
decomposition and carbon di usion take place similar to that in the tip-growth case, but
the CNT precipitation fails to push the metal particle up; so the precipitation is compelled
to emerge out from the metal?s apex. First, carbon crystallizes out as a hemispherical dome
which then extends up in the form of seamless graphitic cylinder. Subsequent hydrocarbon
deposition takes place on the lower peripheral surface of the metal, and as-dissolved carbon
di uses upward. CNT grows up with the catalyst particle rooted on its base; hence, this is
known as base-growth model (Figure 2.9(b) [43]). CNT synthesis involves many parameters
such as hydrocarbon, catalyst, temperature, pressure, gas  ow rate, deposition time and
reactor geometry.
17
Figure 2.7: (a) tip-growth model (b) base-growth model [43]
2.4 Field emission of electrons from CNTs
Local electric  eld is the  eld that surface experiences and proportional to the applied
 eld, when applied  eld is enhaced due to \sharp" surface irregulation. Field emission is a
process that the electrons are emitted from a cold solid surface under the action of a strong
electric  eld. It involves the extraction of electrons from a solid by tunneling through the
surface potential barrier. The potential barrier is square when no electric  eld is applied as
shown in Figure 2.8 [44]. Its shape becomes triangular when a negative potential is applied to
the solid, with a slope that depends on the amplitude of the local electic  eld, F, just above
the surface. Field emission from CNTs follows the same physics. For a single nanotube with
sharp tip, the local electic  eld cannot be simply calculated by dividing the applied voltage
18
by the gap distance between the tip of the nanotube and the anode electrode. The local  eld
will be higher by a factor  , which ampli es the  eld and is determined by the geometrical
shape of the emitter. Therefore, the electric  eld is written as: F =  E =  V=d0, where
E is the applied  eld. The  eld emission characteristic of SWCNT have been studied [45].
Most single SWCNT emitters with a closed tip as well as opened are capable of emitting
over an incredibly large current. Field emission from individual MWCNTs has studied by
Satio et al [46]. Also, they have used closed and opened tubes as the sources in a  eld
emission microscopy. The motivation behind these studied is to explore the possibility of
using individual nanotube  eld emitters in cathod ray tubes or electron guns for electron
microscopy.
19
Figure 2.8: Typical set-up for  eld emission: a potential di erence is applied between a
nanotube and a counter electrode [44]
20
The emission current increases linearly with applied  eld for a  eld smaller than a critical
 eld which is designated as the turn-on  eld, E0, and then the current increases exponentially
with the applied  eld for a  eld stronger than turn-on  eld. The J-E characteristics of the
CNTs were analyzed using Fowler-Nordheim (F-N) model [47]:
J =a E2 exp
  b
E
 
, where a=A  
2=d2
1:1 exp
 1:44 10 7B
 1=2
 
, b=0:95B 
3=2d
 
, where A is 1.54 10 6 eV V 2 and B is 6.83  107 eV 3=2V cm 1 derived from quantum
statistics,  is the work function of CNTs,  is the emission e etive area, and  is the
 eld enhancement factor, which is directly related to the geometry and surface properties of
the CNT samples. The applied electric  eld is de ned as E=V/d, where V is the applied
voltage and d is the distance between the electrodes. The localized electric  eld experienced
by one CNT is expected to be  V/d. The parameters in the F-N equation can be deduced
by  tting the ln(I/V2) vs. 1/V curve, the F-N plot. Fowler-Nordheim model shows that the
dependence of the emitted current on the local electric  eld E and the work function  , is
exponential like.
2.5 Potential applications of CNTs
Many potential applications have been reported for CNTs including their use as rein-
forcement in composite materials, as transparent and  exible electrodes [48], AFM/STM
tips (Figure 2.9 [49]), and electron- eld emitters [50]. The SWCNT  lms can exhibit con-
ductivity/transmittance values comparable to those of low-temperature ITO. Transparent
conducting SWCNT coatings on  exible substrates such as polyethylene terephthalate (PET)
outperform ITO/PET electrodes in terms of chemical and mechanical stability and exhbit
a wider electrochemical window. Moreover, the application of SWCNTs as  eld-emission
electron sources for use in  at-panel displays [17], gas-discharge tubes [17] and microwave
generators [51] has been widely explored. The advantages of these devices over those made
from metals such as tungsten and molybdenum are the following: relatively easy manufac-
turing/fabrication process, less deterioration in high vacuum (10 8Torr), and high current
21
densities of 109 A/cm2. Other desirable properties that make CNTs promising materials
as  eld emitters are their nanosize diameter, structural integrity, high electrical conductiv-
ity, and high chemical stability. Studies on a  eld-e ect transistor (Figure 2.10 [52]) made
from a semiconducting SWCNT showed it to have the ability to be switched from a con-
ducting to an insulating state. Logic switches, the basic components of computers, can be
constructed by coupling such CNT transistors. Recently, the application of random CNT
Networks (CNTNs) as semiconducting materials for thin- lm transistors (TFT) [53] has at-
tracted interest due to their superior performance compared to that of organic TFTs and
potentially low-cost fabrication. Uniformity of CNTN properties is achieved by statistical
averaging over the large number of individual tubes that make up the network. Various
devices and components based on CNTNs have been successfully demonstrated, including
diodes, logic circuit elements, solar cells, displays, and sensors [53].
22
Figure 2.9: TEM image of an individual SWNT tip produced by controlling carefully the
catalyst density. The scale bar equals 10 nm [49]
23
Figure 2.10: Schematic process for the fabrication of a CNTFET with a suspended graphene
gate without any Si3N4 protective layer: The CNTFET can be gated either by the suspended
graphene gate or by the Si substrate acting as a back gate [52].
24
Chapter 3
Carbon Nanotubes: FABRICATION AND CHARACTERIZATION
In this study, CNTs are synthesized using thermal chemical vapor deposition technique
(CVD). The CVD method has advantages of simple equipment setup and excellent unifor-
mity of thin- lm deposition over large area. Silicon wafers are used as the substrate for CNTs
growth and several underlayers and catalyst materials deposited on the silicon wafers using
DC/RF magnetron supttering system. Once the synthesis process is completed, CNTs are
examined used scanning electron microscopy (SEM) and their electrical properties are char-
acterized by  eld emission measurements in vacuum. Synthezied parameters are optimized
by using sputtering and growth conditions.
3.0.1 Growth process
Two metal catalysts were used. Fe as the catalyst, the growth of CNTs is carried at
a temperaute of 700 C, however, Co as the catlyst, the temperature was risen from 700 to
850 C. All the experiments were running at the pressure of 70 Torr and the growth time are
all 20 min. Figure 3.1 shows the schematic diagram for the thermal CVD reactor used for the
growth of CNTs. A resistive heater is used to heat the quartz substrate inside the furnace. A
thermocouple is connected to the substrate holder to measure the temperature. Flowmeters
connected between the chamber and the gas cylinders are used to measure the gas  ow into
the CVD chamber. The pressure in the chamber is controlled by a throttle valve, which is
connected between mechanical pump and the pressure gauge. The gas mixture of acetylene
and argon with 20 and 75sccm respectively were used as the feed gas in the chamber. For
growing the vertically-aligned MWCNTs (VA-MWCNTs), after the deposition of the iron
catlayst, it was received the other graphite deposition. After two steps deposition processes,
25
the sample was oxidized in air at 300 C for 8 hours. Thermal CVD of CNTs was performed
in a vacuum furnace  lled with a gas mixture of acetylene and argon at a pressure of 70
torr. Before the gas mixture was fed into the furnace, the substrate was heated to 700 C in
vacuum and the temperature was remained constant during the 20 min growth of MWNTs.
VA-MWNTs grew on the substrate iwht a uniform length approximately 20  m (Figure 4.8).
26
Figure 3.1: Schematic diagram for thermal CVD reactor [54]
27
3.0.2 Selective grown CNTs synthesis processes
The fabrication process for synthesizing selectively grown CNTs sample is carried out
in detail as shown in Figure 3.3. First, the plain n-type (100) silicon wafers are cleaned
by the standard RCA (Radio Corporation of Americ) cleaning procedure. They are dipped
into a Bu ered Oxide Etching (BOE) solution in order to remove native oxide and chemical
impurities, followed by deionized water (DI water) rinse for a couples of minutes. A dehy-
dration bake step for 20min is performed before priming and spin-coating a wafer with resist.
Following the dehydration bake, the silicon wafer is primed with a pre-resist coating of a
material designed to improve adhesion in this case. Hexamethyldisilazane (HMDS) enhances
the adhesion between the Si wafer and the photoresist (PR), and bahaves as surface-linking
adhesion promoter. 10min of HMDS priming allows good adhesion. Following cleaning,
dehydration baking, and priming, the silicon substrate is coated with photoresist (PR) at
a speed of 3000 rpm for 30 sec, and soft-baked on the hotplate at 105 C for 1min. After
a wafer has been coated with resist, it is subjected to a temperature step, called soft-bake
(or pre-bake), and it is ready to be exposed to some form of radiation in order to create a
latent image in the resist. After exposure, then, the patterned substrate is subject to another
temperature step, called hot-bake, putting on the hotplate at 120 C for 1min. After this
baking process, the substrates are then sputtered with iron forming a 7nm thick Fe (purity
99.99%) catalysis  lm in a sputtering vacuum chamber. The sputtering system is a 2-inch
DC magnetron sputtering system with a power 100W through shadow masks, containing
di erent patterned openings with the sizes of 25 m to 0.5 cm at pitch distances of 25 m to
0.5 cm. When the catalyst are deposited onto the substrate, it is dipped into the aceton for
couple minutes to remove PR, then the patterned  lm are successfully obtained.
28
Figure 3.2: Schematic diagram of CNTs fabrication processing (a) silicon wafer cleaned and
prepared, (b) mask is aligned to remove parts of the photoresist for patterned deposition of
the iron catalyst. (c) the spattering process is conducted. (d) photoresist is removed from
the wafer exposing only the sections with catalyst Fe. The sample is now ready to be moved
to the CVD chamber for CNT growth process. (e) the CNTs are selectively grown on the
substrate.
29
3.1 Electron  eld emission measurements and experimental setup
Once the growth process of the CNTs is completed,  eld emission properties are mea-
sured in vacuum. These CNTs samples, one at a time, are loaded into a high vacuum
chamber. The chamber is pumped by a turbomolecular pump to a vacuum level of 3 10 6
Torr. The measurements are performance at room temperature. The distance between the
CNTs sample (the cathode) and the anode is maintained constant by a glass spacer with a
thickness of 140 m. A rod made by tungsten with a diameter of 0.4 cm was used as the
anode. A dc variable high-voltage supply is used to bias the electrodes. The voltage is con-
tinuously increased with intervals between the cathode and the anode. The emission current
was recorded by a Keithley picoammeter which was connected to a computer through GPIB
card. The applied voltage and the  eld emission current are recorded and then analyzed.
A schematic diagram of an experimental setup used for theses measurements is shown in
Figure. 3.3 [6].
For the  eld emission study, once the voltage and current data are collected, the elec-
tric  eld and the current density data are calculated from the raw data with the following
equations:
E = V  IRd and j = IS
where R is the resistance of the current limiting resistor with a value of 3 M , d is the gap
distance, which is the thickness of the glass spacer, and S is the emission area which is the
exposed area of the collector rod to the CNTs. In these equations d = 140  m and S =
0.0316 cm2 are used. For each applied voltage,  ve current data with 1 second intervals are
recorded and then averaged to form one data point, to assure the test accuracy.
30
Figure 3.3: Schematic diagram for the  eld emission measurement setup [6]
31
Chapter 4
RESULTS AND DISCUSSION
Two di erent morphologies of CNTs, randomly oriented and vertically aligned, were
successfully deposited on silicon substrate with di erent catalysts and underlying layers by
thermal chemical vapor deposition. The catalysts are Iron (Fe) and Cobalt (Co) and under-
lying layers are SiO2, Titanium (Ti) and Tungsten (W). The  eld emission characteristics
of the CNTs specimens are examined in a high pressure vacuum chamber. The e ect of
catalysts and underlying layers on the  eld emission characteristics of CNTs are studied.
Table 4.1: Various Structural Properties of Catalyst-Underlying-Layer Combinations
Substrate Underlayer Catalyst
N/A FeCo
Silicon dioxide (SiO2) FeCo
Silion Titanium (Ti) Fe
Co
Tungstun (W) FeCo
Selectively growth CNTs w/o underlayer Fe
4.1 In uence of catalyst layer thickness
It is known that CNTs grown by low temperature CVD procedures requires a transition
metal catalyst [43], and numerous studies have reported the use of Ni, Fe, Co. [55{57]. In
this study, the CNTs are grown on silicon p-type (100) substrates. A thin catalyst  lm (Fe,
Co) is deposited on substrates by DC magnetron sputtering. Table 4.2 and 4.3 show that
32
the relationship between sputtering time and metal catalysts (Fe, Co) thickness, separately.
From the experimental results, for both Fe and Co catalysts, CNTs can by synthesized well
onto the plain silicon substrates by depositing the thickness around 7nm-12nm. However,
once the catalyst  lms are thicker than 15nm, CNTs are not found at any substrate, instead
of graphite. The catalyst thinkess was measured by pro lometer.
Table 4.2: Iron catalyst vs. sputtering time
Sputtering time 1min 3min 5min 8min 10min 15min
Iron thickness t  7nm  7nm  10nm  15nm
Table 4.3: Cobalt catalyst vs. sputtering time
Sputtering time 15sec 30sec 45sec 1min >1min
Cobalt thickness t  7nm  15nm  25nm >25nm
4.2 In uence of underlying layer
CNTs are grown ever di erent underlying layers of SiO2, Ti and W. We investigated
the Fe and Co catalysts and various underlayers in their respective e ectivenss. For using Fe
catalyst, the results indicates that nanotubes are grown well on plain Si and SiO2 underlayer
at a certain growth condition. On the other hand, using the Ti or W as underlayer, the
high-quality nanotubes cannot be form. For Co catalyst, nanotubes can be only grown
on SiO2 underlayers. The Ti, W and plain silicon are not found nanotubes formation. For
plain silicon, the reason could be the formation of silicide, CoSix during the high temperature
processing [58] . As a result, a barrier layer such as SiO2 is useful to prevent silicide formation
when using Co as catalyst.
4.3 In uence of the deposition temperature
In this study, nanotubes were synthesized at two di erent deposition temperatures while
keeping all other parameters constant. For Fe catalyst, our experimental results indicate
33
when temperature is below 700 C, it may not be su cient for nucleating nanotubes. The
temperature may be too low to decompose C2H2, hindering growth. Furthermore, the low
temperature also limits the graphitization, which increases the condensation of a-C [59]. The
nanotubes grown at 700 C are uniform and dense in size. Nevertheless, when the deposition
temperature is increased to about 900 C, the nanotubes are not grown. For Co catalyst,
it is found that the nanotubes cannot be grown at 700 C. However, while increasing the
deposition temperature to 850 C, the nanotubes are successfully grown. According to C.H.
Lin et al. [60], it is believed that the e ect of the substrate temperature is to minimize the
thermal energy to activate catalysts to precipitate carbon atoms to form nanotubes.
4.4 Fowler-Nordheim curves
In section 2.4, the  eld emission properteis of CNTs has been introduced. During the
 eld emission process, when the injected electrons pass through the surface of the CNTs and
are emitted into the vacuum towards the anode from the CNTs, the local electric  eld is
enhanced due to the sharp needle like emitting structures of the CNTs compared to the  at
structures. According to Fowler-Nordheim (F-N) theory [61], the dependence of the emission
current, I(A) on the work function  (eV) of the emitting surface and the local electric  eld
just above the emitter surface, F= E= V/d (V/ m), is exponential, and described by the
F-N equation written as:
I=k1 
2E2s
 exp
0
@ k2 
3
2
 
1
A
where s is the emitter area (cm2) and  is the enhancement factor, determined by the
geometric shape of the emitter. The constants k1 = 1:54 10 6(A-eV/V2) and k2 = 6:83 107
(eV 3=2-V/cm) are given in the literature [62]. In general, the plot of ln(1/E2) vs 1=E yields
a straight line and the slope of the line is proportional to the  eld enhancement factor,  of
the emitters and the intercept gives the emitter area.
34
4.5 Scanning electron microscope image
In this section, the scanning electron microscope (SEM) images of CNTs samples with
di erent growth conditions and underlying layers are presented. The following SEM images
show that the samples with 1 and 5 of iron sputtering time on SiO2 and 15, 30sec and
1min of cobalt sputtering time on SiO2. The sample with 5 min iron and 8 min graphite
sputtering time can yield very good-quality MWNTs on both plain silicon and SiO2, as
shown in Figure 4.3. Figure 4.5 and 4.6 were used cobalt as catlayst to deposit CNTs. Using
cobalt to synthesize CNTs, they are obviously shorter than using Fe as catalyst under the
same growth time. Figure 4.7 shows that when catalyst thickness is too thick, then carbon
cluster will be formed instead of CNTs. The cross-sectional SEM image of vertically-aligned
and randomly-oriented MWCNTs is shown as Figure 4.8 and Figure 4.9, separately. Figure.
4.10 is the patterned CNTs of SEM image.
4.5.1 SEM images of CNTs with di erent catalysts sputtering time and under-
layers
35
Figure 4.1: 1min Fe sputtering time on SiO2 underlayer
Figure 4.2: 5min Fe sputtering time on SiO2 underlayer
36
Figure 4.3: 5min Fe with Carbon 8 min sputtering time on SiO2 underlayer
Figure 4.4: 5min Fe sputtering time on SiO2 underlayer
37
Figure 4.5: 15 sec Cobalt sputtering time on SiO2 underlayer
Figure 4.6: 30 sec Cobalt sputtering time on SiO2 underlayer
38
Figure 4.7: Cobalt sputtering time above 1min on SiO2 underlayer
39
Figure 4.8: SEM image of vertically-aligned multi-wall carbon nanotubes (VA-MWCNTs).
40
Figure 4.9: Cross-sectional SEM image of randomly-oriented MWCNTs.
41
Figure 4.10: SEM image of patterned CNTs array (a) 25 m circle array with 25 m spacing,
(b) 25 m diameter circle array with 50 m spacing, (c) 50 m  50 m square array with
25 m spacing, (d) a close-up image of a single CNT bundle (from a circle array sample)
42
Table
4.4:
Various
Structural
Prop
erties
of
Catalyst-Underlying-La
yer
Com
binations
Sp
ecimen
Substrate
Underlying
Catalyst
Typ
e
Pre-treatmen
ta
Typ
es
of
CNTs
Gro
wth
Note
b
Lab
el
La
yer
(XX)
(YY)
Condition
c
Iron
f
N/A
Random
T:
700
 C;
t:20min
Table4.5
S01-XX-YY
d
Silicon
e
Oxide(0.6
 m
)
w/
dep
ositing
graphite
Vertically-Aligned
Ar:C
2H
2=75:25sccm
Cobalt
g
N/A
Random
T:
850
 C
h;
t:20min
Table4.6
w/
dep
ositing
graphite
Random
Ar:C
2H
2=75:25sccm
Iron
N/A
Random
T:
700
 C;
t:20min
Table4.7
S02-XX-YY
Silicon
N/A
i
w/
dep
ositing
graphite
Vertically-Aligned
Ar:C
2H
2=75:25sccm
Cobalt
N/A
Random
T:
850
 C;
t:20min
Table4.8
w/
dep
ositing
graphite
Random
Ar:C
2H
2=75:25sccm
Iron
N/A
Random
T:
700
 C;
t:20min
Table4.9
S03-XX-YY
Silicon
Titanium(0.6
 m
)j
w/
dep
ositing
graphite
Random
Ar:C
2H
2=75:25sccm
Cobalt
N/A
Random
T:
850
 C;
t:20min
Table4.10
w/
dep
ositing
graphite
Random
Ar:C
2H
2=75:25sccm
Iron
N/A
N/A
T:
700
 C;
t:20min
Table4.11
S04-XX-YY
Silicon
Tungsten(0.4
 m
)k
w/
dep
ositing
graphite
N/A
Ar:C
2H
2=75:25sccm
Cobalt
N/A
N/A
T:
850
 C;
t:20min
Table4.12
w/
dep
ositing
graphite
N/A
Ar:C
2H
2=75:25sccm
aSp
ecimen
with
dep
ositing
graphite
were
then
annealed
in
air
at
300
 C
for
8hrs
[63]
bF
or
detailed
information,
refer
to
the
table
cThe
grw
oth
time
of
sp
ecimens
all
are
20m
in
in
agas
mixture
of
Argon
and
Acet
ylene(75:25sccm)
in
the
pressure
70T
orr
fThe
iron
pur
ity
is
99.99%
dXX:
the
dep
osition
time
of
catalyst,
YY:
the
dep
osition
time
of
graphite
eUse
n-t
yp
e<
100
>
Si
waf
er,
resistivit
y1
to
5
 -cm
gThe
cobalt
purit
yis
99.99%
hIt
is
believ
ed
that
substrate
temp
erature
is
to
minimize
the
thermal
energy
to
activ
ate
catalysts
to
form
nanotub
es
[60]
iR
CA
cleaning
pro
cess(a
short
imme
rsion
in
a1:50
solution
of
HF
+
H2O)
to
remo
ve
nativ
eo
xide
lay
er
jThe
titanium
pur
ity
is
99.99%
kThe
tungsten
purit
yis
99.99%
43
Table
4.5:
Various
Structural
Prop
erties
of
Catalyst-Underlying-La
yer
Com
binations
Sp
ecimen
Substrate
Underlying
Catalyst
Typ
e
Pre-treatmen
t
Typ
es
of
CNTs
Gro
wth
Note
b
Lab
el
La
yer
(XX)
(YY)
Condition
a
S05-XX-YY
c
Silicon
N/A
Iron
N/A
Random
T:
700
 C;
t:20min
Table4.13
Ar:C
2H
2=75:25sccm
bF
or
detailed
information,
refer
to
the
table
aThe
grw
oth
time
of
sp
ecimens
all
are
20min
in
agas
mixture
of
Argon
and
Acet
ylene(75:25sccm)
in
the
pressure
70T
orr
cSelectiv
egrw
on
CNTs
44
Table
4.6:
MWNTs
gro
wth
on
the
catalyst-coated(F
e)
substrate
w/
SiO2
underla
yer
Sp
ecimen
Lab
el
Thic
kness
of
Catalyst
SEM
Image
Typ
e
of
CNTs
Field
Emi
ssion
Characteristics
S01-F
e-1min-00
Lo
w
de
ns
ity
gro
wth
Random
Turn
on
 eld
 
1:86
V= m
Saturation
curen
t 
5:4
 
10
 3
A=cm
2
S01-F
e-3min-00
t 
7nm
Lo
w
de
ns
ity
gro
wth
Random
Turn
on
 eld
 
1:57
V= m
Saturation
curen
t 
5:2
 
10
 3
A=cm
2
S01-F
e-5min-00
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:71
V= m
Saturation
curen
t 
4:1
 
10
 3
A=cm
2
S01-F
e-8min-00
t 
7nm
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:71
V= m
Saturation
curen
t 
4:8
 
10
 3
A=cm
2
S01-F
e-10min-00
t 
10nm
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:57
V= m
Saturation
curen
t 
4:9
 
10
 3
A=cm
2
S01-F
e-15min-00
t 
15nm
No
gro
wth
a
N/A
N/A
S01-F
e-5min-C-1min
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:71
V= m
Saturation
curen
t 
4:1
 
10
 3
S01-F
e-5min-C-3min
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:71
V= m
Saturation
curen
t 
5:3
 
10
 3
A=cm
2
S01-F
e-5min-C-5min
Fe
catalyst:
t 
7nm
a
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:71
V= m
Saturation
curen
t 
4:78
 
10
 3
A=cm
2
S01-F
e-5min-C-8min
Extreme
densit
ygro
wth
Vertically
Aligned
Turn
on
 eld
 
2:14
V= m
Saturation
curen
t 
1:78
 
10
 3
A=cm
2
S01-F
e-5min-C-10min
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:57
v= m
Saturation
curen
t 
4:58
 
10
 3
A=cm
2
S01-F
e-5min-C-15min
No
gro
wth
N/A
N/A
aF
orm
the
carb
on
cluster
aCho
osing
Fe-5min
as
the
base,and
then
pretreated
by
dep
ositing
graphite(C)
an
dannealing
8hrs.
45
Figure
4.11:
Field
emission
curv
efor
gro
wing
CNTs
on
SiO
2
using
Fe
as
catalyst
w/o
pre-treatmen
t
46
Figure
4.12:
Fo
wler
-N
ordeim
curv
eof
CNT
gro
wn
on
SiO
2
using
Fe
as
catalyst
w/o
pre-treatmen
t
47
Figure
4.13:
Field
emission
curv
efor
gro
wing
CNTs
on
SiO
2
using
Fe
as
catalyst
with
pre-treatmen
t
48
Figure
4.14:
Fo
wler-Nordeim
curv
eof
CNT
gro
wn
on
SiO
2
using
Fe
as
catalyst
with
pre-treatmen
t
49
Table
4.7:
MWNTs
gro
wth
on
the
catalyst-coated(Co)
substrate
w/
SiO
2underla
yer
Sp
ecimen
Lab
el
Thic
kness
of
Catalyst
SEM
Image
Typ
e
of
CNTs
Field
Emiss
ion
Characteristics
S01-Co-15sec-00
Lo
w
de
ns
ity
gro
wth
Random
Turn
on
 eld
 
2:72
V= m
t 
7nm
Saturation
curen
t 
4:5
 
10
 3
A=cm
2
S01-Co-30sec-00
Lo
w
de
ns
ity
gro
wth
Random
Turn
on
 eld
 
5:57
V= m
Saturation
curen
t 
2:09
 
10
 3
A=cm
2
S01-Co-45sec-00
t 
15nm
Lo
w
de
ns
ity
gro
wth
Random
Turn
on
 eld
 
1:72
V= m
Saturation
curen
t 
3:8
 
10
 3
A=cm
2
S01-Co-1min-00
t 
25nm
Lo
w
de
ns
ity
gro
wth
Random
Turn
on
 eld
 
2:72
V= m
Saturation
curen
t 
3:7
 
10
 3
A=cm
2
S01-Co-Ab
ov
e1min-00
t>
25nm
No
gro
wth
N/A
N/A
S01-Co-45sec-C-3min
No
gro
wth
N/A
N/A
S01-Co-45sec-C-4min
No
gro
wth
N/A
N/A
S01-Co-45sec-C-5min
Co
catalyst:
t 
25nm
No
gro
wth
N/A
N/A
S01-Co-45sec-C-7min
No
gro
wth
N/A
N/A
S01-Co-45sec-C-10min
No
gro
wth
N/A
N/A
50
Figure
4.15:
Field
emission
curv
efor
gro
wing
CNTs
on
SiO
2
using
Co
as
catalyst
w/o
pre-tre
atmen
t
51
Figure
4.16:
Fo
wle
r-Nordeim
curv
eof
CNT
gro
wn
on
SiO
2
using
Co
as
catalyst
w/o
pre-tr
eatmen
t
52
Table
4.8:
MWNTs
gro
wth
on
the
catalyst-coated(F
e)
substrate
w/o
SiO2
underla
yer
Sp
ecimen
Lab
el
Thic
kness
of
Catalyst
SEM
Image
Typ
e
of
CNTs
Field
Emi
ssion
Characteristics
S02-F
e-1min-00
Lo
w
de
ns
ity
gro
wth
Random
Turn
on
 eld
 
1:21
V= m
Saturation
curen
t 
4:33
 
10
 3
A=cm
2
S02-F
e-3min-00
t 
7nm
Lo
w
de
ns
ity
gro
wth
Random
Turn
on
 eld
 
1:22
V= m
Saturation
curen
t 
6:47
 
10
 3
A=cm
2
S02-F
e-5min-00
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:78
V= m
Saturation
curen
t 
5:45
 
10
 3
A=cm
2
S02-F
e-8min-00
t 
7nm
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:43
V= m
Saturation
curen
t 
6:4
 
10
 3
A=cm
2
S02-F
e-10min-00
t 
10nm
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:14
V= m
Saturation
curen
t 
6:6
 
10
 3
A=cm
2
S02-F
e-15min-00
t 
15nm
No
gro
wth
a
N/A
N/A
S02-F
e-5min-C-1min
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:14
V= m
Saturation
curen
t 
5:74
 
10
 3
A=cm
2
S02-F
e-5min-C-3min
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:37
V= m
Saturation
curen
t 
5:1
 
10
 3
A=cm
2
S02-F
e-5min-C-5min
Fe
catalyst:
t 
7nm
a
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:36
V= m
Saturation
curen
t 
6:4
 
10
 3
A=cm
2
S02-F
e-5min-C-8min
Extreme
densit
ygro
wth
Vertically
Aligned
Turn
on
 eld
 
2:14
V= m
Saturation
curen
t 
1:41
 
10
 4
A=cm
2
S02-F
e-5min-C-10min
High
densit
ygro
wth
Random
Turn
on
 eld
 
1:86
v= m
Saturation
curen
t 
2:08
 
10
 3
A=cm
2
S02-F
e-5min-C-15min
No
gro
wth
N/A
N/A
aF
orm
the
carb
on
cluster
aCho
osing
Fe-5min
as
the
base,and
then
pretreated
by
dep
ositing
graphite(C)
an
dannealing
8hrs.
53
Figure
4.17:
Field
emission
curv
efor
gro
wing
CNTs
on
plain
silicon
using
Fe
as
catalyst
w/o
pre-treatm
en
t
54
Figure
4.18:
Fo
wler
-Nordeim
curv
eof
CNT
gro
wn
on
plain
silicon
using
Fe
as
catalyst
w/o
pr
e-treatmen
t
55
Figure
4.19:
Field
emission
curv
efor
gro
wing
CNTs
on
plain
silicon
using
Fe
as
catalyst
with
pre-treatme
nt
56
Figure
4.20:
Fo
wler-Nordeim
curv
eof
CNT
gro
wn
on
plain
silicon
using
Fe
as
catalyst
with
pre-treatmen
t
57
Table
4.9:
MWNTs
gro
wth
on
the
catalyst-coated(Co)
substrate
w/o
SiO2
underla
yer
Sp
ecimen
Lab
el
Thic
kness
of
Catalyst
SEM
Image
Typ
e
of
CNTs
Field
Emissi
on
Characteristics
S02-Co-15sec-00
No
gro
wth
N/A
N/A
t 
7nm
S02-Co-30sec-00
No
gro
wth
N/A
N/A
S02-Co-45sec-00
t 
15nm
No
gro
wth
N/A
N/A
S02-Co-1min-00
t 
25nm
No
gro
wth
N/A
N/A
S02-Co-45sec-C-3min
No
gro
wth
N/A
N/A
S02-Co-45sec-C-4min
No
gro
wth
N/A
N/A
S02-Co-45sec-C-5min
Co
catalyst:
t 
25nm
No
gro
wth
N/A
N/A
S02-Co-45sec-C-7min
No
gro
wth
N/A
N/A
S02-Co-45sec-C-10min
No
gro
wth
N/A
N/A
58
Table
4.10:
MWNTs
gro
wth
on
the
catalyst(F
e)-underlying
Ti
lay
er-coated
substrate
Sp
ecimen
Lab
el
Thic
kness
of
Catalyst
SEM
Image
Typ
e
of
CNTs
Field
Emiss
ion
Characteristics
S03-F
e-5min-00
Very
low
densit
ygro
wth
Random
Turn
on
 eld
 
2:9
V= m
Saturation
curen
t 
2:52
 
10
 3
mA=cm
2
S03-F
e-8min-00
Fe
catalyst:
t 
7nm
Very
low
densit
ygro
wth
Random
Turn
on
 eld
 
3:2
V= m
Ti
lay
er
 
0:6
 m
Saturation
curen
t 
1:54
 
10
 3
mA=cm
2
S03-F
e-10min-00
Very
low
densit
ygro
wth
Random
Turn
on
 eld
 
2:3
V= m
Saturation
curen
t 
2:64
 
10
 3
mA=cm
2
S03-F
e-5min-C-5min
Very
low
densit
ygro
wth
Random
Turn
on
 eld
 
2:14
V= m
Saturation
curen
t 
2:04
 
10
 3
mA=cm
2
S03-F
e-5min-C-8min
Fe
catalyst:
t 
7nm
Very
low
densit
ygro
wth
Random
Turn
on
 eld
 
2:71
V= m
Saturation
curen
t 
1:43
 
10
 3
mA=cm
2
S03-F
e-5min-C-10min
Very
low
densit
ygro
wth
Random
Turn
on
 eld
 
3:86
V= m
Saturation
curen
t 
1:55
 
10
 3
mA=cm
2
59
Figure
4.21:
Field
emission
curv
efor
gro
wing
CNTs
on
Ti
underlying
lay
er
using
Fe
as
catalyst
w/o
pr
e-treatmen
t
60
Figure
4.22:
Fo
wler
-Nordeim
curv
eof
CNT
gro
wn
on
plain
silicon
using
Fe
as
catalyst
w/o
pr
e-treatmen
t
61
Figure
4.23:
Field
emission
curv
efor
gro
wing
CNTs
on
Ti
underlying
lay
er
us
ing
Fe
as
catalyst
with
pre
-treatmen
t
62
Figure
4.24:
Fo
wler-Nordeim
curv
eof
CNT
gro
wn
on
Ti
underlying
lay
er
using
Fe
as
catalyst
with
pre-treatmen
t
63
Table
4.11:
MWNTs
gro
wth
on
the
catalyst(Co)-underly
ing
Ti
lay
er-coated
substrate
Sp
ecimen
Lab
el
Thic
kness
of
Catalyst
SEM
Image
Typ
e
of
CNTs
Field
Emi
ssion
Characteristics
S03-Co-30sec-00
t 
7nm
No
gro
wth
N/A
N/A
S03-Co-45sec-00
t 
15nm
No
gro
wth
N/A
N/A
S03-Co-1min-00
t 
25nm
No
gro
wth
N/A
N/A
S03-Co-1min-C-5min
No
gro
wth
N/A
N/A
S03-Co-1min-C-8min
t 
25nm
No
gro
wth
N/A
N/A
S03-Co-1min-C-10min
No
gro
wth
N/A
N/A
64
Table
4.12:
MWNTs
gro
wth
on
the
catalyst(F
e)-underlying
W
lay
er-coated
substrate
Sp
ecimen
Lab
el
Thic
kness
of
Catalyst
SEM
Image
Typ
e
of
CNTs
Field
Emiss
ion
Characteristics
S04-F
e-5min-00
t 
7nm
No
gro
wth
N/A
N/A
S04-F
e-8min-00
t 
7nm
No
gro
wth
N/A
N/A
S04-F
e-10min-00
t 
10nm
No
gro
wth
N/A
N/A
S04-F
e-5min-C-5min
No
gro
wth
N/A
N/A
S04-F
e-5min-C-8min
Fe
catalyst:
t 
7nm
No
gro
wth
N/A
N/A
S04-F
e-5min-C-10min
No
gro
wth
N/A
N/A
65
Table
4.13:
MWNTs
gro
wth
on
the
catalyst(Co)-under
lying
W
lay
er-coated
substrate
Sp
ecimen
Lab
el
Thic
kness
of
Catalyst
SEM
Image
Typ
e
of
CNTs
Field
Emi
ssion
Characteristics
S04-Co-30sec-00
t 
7nm
No
gro
wth
N/A
N/A
S04-Co-45sec-00
t 
15nm
No
gro
wth
N/A
N/A
S04-Co-1min-00
t 
25nm
No
gro
wth
N/A
N/A
S04-Co-1min-C-5min
No
gro
wth
N/A
N/A
S04-Co-1min-C-8min
Co
catalyst:
t 
25nm
No
gro
wth
N/A
N/A
S04-Co-1min-C-10min
No
gro
wth
N/A
N/A
66
Table
4.14:
Selectiv
egro
wth
of
MWNTs
on
the
catalyst(F
e)-coated
substrate
w/o
SiO2
Sample
Lab
el
Size
Shap
e
Arra
y
Spa
cing
SEM
Image
Typ
e
of
CNTs
Field
Em
ission
Characteristics
25
 m
in
diameter
Circle
25
 m
High
densit
y
Random
Turn
on
 eld
 
2:26
V= m
Saturation
curen
t 
2:47
 
10
 3
A=cm
2
25
 m
in
diameter
Circle
50
 m
High
densit
y
Random
Turn
on
 eld
 
3:12
V= m
Saturation
curen
t 
0:3
 
10
 3
A=cm
2
25
 m
in
diameter
Circle
100
 m
High
densit
y
Random
Turn
on
 eld
 
2:42
V= m
S05-F
e-5min-00
Saturation
curen
t 
1 
10
 3
A=cm
2
50
 m
 
50
 m
Square
25
 m
High
densit
y
Random
Turn
on
 eld
 
2:28
V= m
Saturation
curen
t 
4 
10
 3
A=cm
2
0.5cm
 
0.5cm
Square
0.5cm
High
densit
y
Random
Turn
on
 eld
 
2:82
V= m
Saturation
curen
t 
0:4
 
10
 4
A=cm
2
67
Figure
4.25:
Field
emission
curv
efor
gro
wing
selectiv
eCNTs
on
plain
silicon
using
Fe
as
catalyst
w/o
pre-treatmen
t
68
Figure
4.26:
Fo
wl
er-Nordeim
curv
eof
selectiv
eCNT
gro
wn
on
plain
silicon
us
ing
Fe
as
catalyst
w/o
pre-
treatmen
t
69
Chapter 5
SUMMARY AND FUTURE DIRECTIONS
Since the discovery of carbon nanotubes (CNTs), many studies have been carried out on
their synthesis. With their outstanding electrical, structure, and physical properties, CNTs
are envisioned to impact furture electronic applications such as nanoelectronics, sensors,
electrodes, and nanophotonics. These applications generally require controlled growth on
patterned substrates. Therefore, there has been a strong focus on using thermal chemical
vapor deposition (CVD) technique to realized CNT structures for these applications. Field
emission is one of the most advanced and broadly studied applications of CNTs. They
can emit electrons easily due to their high aspect ratios, compared to other cold-cathode
materials.
In this work, CNTs have been selective and non-selective grown on a variety of catalyst-
coated underlaying layers (plain silicon, SiO2, Ti, W) substrate at two di erent temperatures
(700 C, 800 C) by using thermal chemical vapor technique. We investigated their  eld emis-
sion characteristics. From the experiemtal results, it clearly indicates that metal catlayst
thickness is a critical parameter for the growth of CNTs. In addition, the deposition tem-
perature and underlayers also play the important roles. Field emission results from carbon
nanotube showed clearly the Fowler-Nordheim behavior. Vertically-aligned CNTs can be
successfully synthesized at a certain growth condition. There are many still obstacle remain-
ing. The growth conditions of CNTs have not been fully determined such as gas  ow rate
(acetylene, argon), deposition temperature, pressure, catalyst thickness, underlying layer,
etc. The  eld emisision and underlayers mechanism for CNTs growth need to be deeply
investigated.
70
Bibliography
[1] S. Iijima. Helical microtubules of graphitic carbon. Nature, 354:56{58, November 1991.
[2] A. A. Talin, K. A. Dean, and J. E. Kaskie. Solid-State Electron, 45:963, 2001.
[3] J. M. Bonard, H. Kind, T. Stockli, and L. Nilsson. Solid-State Electron, 45:893, 2001.
[4] H. Murakami, M. Hirakawa, C. Tanaka, and H. Yamakawa. Appl. Phys. Lette, 76:1776,
2000.
[5] W. Zhu, C. Bower, G. P. Kochankski, and S. Jin. Solid-State Electron, 45:921, 2001.
[6] Ramesh Bokka. Master?s thesis, Auburn University, ECE Department, May 2011.
[7] Anastasios John Hart. PhD thesis, Massacusetts Institute of Technology, August 2006.
[8] T. Hayashi, Y.A. Kim, T. Matoba, M. Esaka, K. Nishimura, T. Tsukada, M. Endo, and
M.S. Dresselhaus. Nano Letters, pages 887{889, 2003.
[9] T. W. Ebbesen, H. Hiura, J. Fujita, Y. Ochiai, S. Matsui, and K. Tangiaki. Chem.
Phys. Lett., 209:83{90, 1993.
[10] A. Thess, R. Lee, P. Nikolaev, H. Dai, P. Petit, J. Robert, C. Xu, Y.H. Lee, S.G. Kim,
A.G. Rinzler, D.T. Colbert, G.E. Scuseria, D. Tomanek, J.E. Fischer, and R.E. Smalley.
Science, 273:483, 1996.
[11] C.J. Lee and J. Park. Appl. Phys. Lett., 77:3397, 2000.
[12] Haitao Zhao. PhD thesis, Auburn University, August 2012.
71
[13] OTTO ZHOU, HIDEO SHIMODA, BO GAO, SOOJIN OH, LES FLEMING, and
GUOZHEN YUE. Materials Science of Carbon Nanotubes: Fabrication, Integration,
and Properties of Macroscopic Structures of Carbon Nanotubes. Acc. Chem. Res., pages
1045{1053, 2002.
[14] Bhupesh Chandra. PhD thesis, Columbia Uniersity, 2009.
[15] Anton Anisimov. PhD thesis, Aalto University, 2010.
[16] Mauricio Terrones. Science and technology of the twenty- rst century: Synthesis, Prop-
erties and Applications of Carbon Nanotubes. Annu. Rev. Mater.Res., pages 419{501,
2003.
[17] M. S. Dresselhaus, G. Dresselhaus, and Ph. Avouris. Carbon nanotubes, Topics in
Applied Physics. 2001.
[18] M.S. Dresselhaus, G. Dressehaus, and P.C. Eklund. Science of Fullenrenes and Carbon
Nanotubes. 1996.
[19] Chao Liu. Cold electron emitters using chemically vapor deposited carbon nanotubes.
Master?s thesis, Auburn University, ECE Department, 2000.
[20] R. Saito, G. Dresselhaus, and M.S. Dresselhaus. Appl. Phys., 73:494, 1993.
[21] M.S. Dresselhaus, G. Dressehaus, R. Saito, and A. Jorio. Raman spectroscopy of Carbon
Nanotubes. Physics Reports, 2004.
[22] Ryuichi Kuzuo, Masami Terauchi, and Michiyoshi Tanaka. Electron Energy-Loss Spec-
tra of Carbon Nanotubes. Japanese Journal of Applied Physics, 1992.
[23] Mayumi Kosaka, Thomas W. Ebbesen, Hidefumi Hiura, and Katsumi Tanigaki. Electron
spin resonance of carbon nanotubes. Chemical Physcis Letters, 1994.
72
[24] L. P. Biro, TS. Lazarescu, Ph. Lambin, P. A. Thiry, A. Fonseca, J. B. Nagy, and A. A.
Lucas. Scanning tunneling microscope investigation of carbon nanotubes produced by
catalytic decomposition of acetylene. Physical Review B, 1997.
[25] T.W. Ebbesen, H. Lezec, H. Hiura, J.W. Bennett, H.F. Ghaemi, and T. Thio. Nature,
54:382, 1996.
[26] A. Thess, R. Lee, P. Nikolaev, H. Dai, P. Petit, J. Robert, C. Xu, Y.H. Lee, S.G. Kim,
A.G. Rinzler, D.T. Colbert, G.E. Scuseria, D. Tomanek, J.E. Fischer, and R.E. Smalley.
Science, 273:483, 1994.
[27] O. Lourie abd H.D. Wagner. Mater. Res., 13:9, 1998.
[28] M.F. Yu, B.S. Files, S. Arepalli, and R.S. Ruo . Phys. Rev. Lett., 84:24, 2000.
[29] C.A. Cooper, R.J. Young, and M. Halsall. Investigation into the deformation of carbon
nanotubes and their composites through the use of Raman spectroscopy. Composites:
Part A, 2001.
[30] M.M.J. Treacy, T.W. Ebbesen, and J.M. Gibson. Exceptionally high Young?s modulus
observed for individual carbon nanotubes. Nature, 381:678{690, 1996.
[31] A. Krishnan, E. Dujardin, T.W. Ebbesen, P.N. Yianilos, and M.M.J. Treacy. Youngs
modulus of single-walled nanotubes. Phys. Rev. B, 58:14013{14019, 1998.
[32] F. J. Garca-Vidal, J. M. Pitarke, and J. B. Pendry. E ective medium theory of the
optical properties of aligned carbon nanotubes. Phys. Rev. Lett., 1997.
[33] Zu-Po Yang, Lijie Ci, James A. Bur, Shawn-Yu Lin, and Pulickel M. Ajayan. Experi-
mental Observation of an Extremely Dark Material Made by a Low-Density Nanotube
Array. Phys. Rev. B, 8:446{451, 2008.
[34] T. W. Ebbesen and P.M. Ajayan. Nature, 220:358, 1992.
73
[35] Yu-Chun Chen. PhD thesis, Auburn University, ECE Department, 2003.
[36] B.I. Yakobson and R.E. Smalley. American Scientist, 85:324, 1997.
[37] M.L. Terranova1, V. Sessa1, and M. Rossi. Chemical Vapor Deposition, 12:315{325,
2006.
[38] M. Daenen, R.D. de Fouw, B. Hamers, P.G.A. Janssen, K. Schouteden, and M.A.J.
Veld. The Wondrous World of Carbon Nanotubes. 2003.
[39] M. Su, B. Zheng, and J. Liu. Chem. Phys. Lett., 322:321, 2000.
[40] H. Lee, Y.S. Kang, P.S. Lee, and J. Y. Lee. J. Alloys Compd., 330:569, 2005.
[41] P. Mauron, C. Emmenegger, A. Zuttel, C. Nutzenadel, P. Sudan, and L. Schlapbach.
Carbon, 40:1339, 2002.
[42] J. B. Park, G. S. Choi, Y. S. Cho, S. Y. Hong, D. Kim, S. Y. Choi, J. H. Lee, and K. I.
Cho. J. Crystal Grwoth, 244:211, 2002.
[43] M. Kumar and Y. Ando. J. of Nanoscience and Nanotechnology, 10:3739{3758, 2010.
[44] C. Adessi and M. Devel. Phys. Rev. B, 20, 2000.
[45] J.-M. Bonard, J.-P. Salvetat, T. Stockli, L. Forro, and A. Chatelain. Appl. Phys. A.,
69, 1999.
[46] Y. Satio and S. Uemura. Carbon, 2000.
[47] Yi-Chun Chen, Hsiu-Fung Cheng, Yun-Shuo Hsieh, and You-Ming Tsau. J. Appl.
Physic., 94, 2003.
[48] Goki Edal, Giovanni Fanchini1, and Manish Chhowalla1. Nature Nanotechnology, 2008.
[49] J.H. Hafner, C.-L. Cheung, A.T. Woolley, and C.M. Lieber. Progress in Biophysics and
Molecular Biology, 77, 2001.
74
[50] Walt A. de Heer, Jean-Marc Bonard, Kai Fauth, Andr6 Chiitelain, Laszlo Forro, and
Daniel Ugarte. Advanced Materials, 2004.
[51] Bishun N Khare, Patrick Wilhite, and M Meyyappan. Nanotechnology, 2004.
[52] Johannes Svensson, Niklas Lindahl, Hoyeol Yun, Miri Seo, Daniel Midtvedt, Yury
Tarakanov, Niclas Lindvall, Oleg Nerushev, Jari Kinaret, SangWook Lee, and Eleanor
E. B. Campbell. Nano Letters, 11:3569{3575, 2001.
[53] Chin-Lung Cheng, Chien-Wei Liu, Bau-Tong Dai, and Ming-Yen Lee. Physical and
Electrical Characteristics of Carbon Nanotube Network Field-E ect Transistors Syn-
thesized by Alcohol Catalytic Chemical Vapor Deposition. J. of nanomaterials, 2011,
2011.
[54] Y. Tzeng, C. Liu, and Z. Chen. Low temperature CVD carbon nanotubes on glass
plates for  at panel display applications. In Materials Research Society Spring Meeting,
2000.
[55] Z.P. Huang, D.Z. Wang, J.G. Wen, M. Sennett, H. Gibson, and Z.F. Ren. E ect of
nickel, iron and cobalt on growth of aligned carbon nanotubes. Appl. Phys. A, 2002.
[56] E. Flahaut, A. Govindaraj, A. Peigney, Ch. Laurent, A. Rousset, and C.N.R. Rao.
Synthesis of single-walled carbon nanotubes using binary Fe, Co, Ni alloy nanoparticles
prepared in situ by the reduction of oxide solid solutions. Chemical Physics Letters,
1999.
[57] Cheol Jin Lee, Jeunghee Park, and Jeong A. Yu. Catalyst e ect on carbon nanotubes
synthesized by thermal chemical vapor deposition. Chemical Physics Letters, 2002.
75
[58] Yoshikazu Homma, Yoshiro Kobayashi, Toshio Ogino, Daisuke Takagi, Roichi Ito,
Yung Joon Jung, and Pulickel M. Ajayan. Role of Transition Metal Catalysts in Single-
Walled Carbon Nanotube Growth in Chemical Vapor Deposition. J. Phys. Chem. B,
2003.
[59] M. Chhowalla, K. B. K. Teo, C. Ducati, N. L. Rupesinghe, G. A. J. Amaratunga,
A. C. Ferrari, D. Roy, J. Robertson, and W. I. Milne. Growth process conditions of
vertically aligned carbon nanotubes using plasma enhanced chemical vapor deposition.
J. of Applied Physics, 2001.
[60] Chao Hsun Lin, Hui Lin Chang, Ming Her Tsai, and Cheng Tzu Kuo. Growth mecha-
nism and properties of the large area well-aligned carbon nano-structures deposited by
microwave plasma electron cyclotron resonance chemical vapor deposition. Diamond
and Related Materials, pages 922{926, November 2002.
[61] R.H. Fowler and L.W. Nordheim. page 173, 1928.
[62] X. Lu, Q. Yang, C. Xiao, and A. Hirose. Nonlinear Fowler-Nordheim plots of the  eld
electron emission from graphitic nanocones: In uence of non-uniform  eld enhancement
factor. PhD thesis, 2006.
[63] C. Liu, Y.C., and Y. Tzeng. Diamond and Related Materials, 13:1274{1280, 2004.
76