University of Southern California Dissertations and Theses

Growth and field emission of multi-walled carbon nanotubes

(USC Thesis Other)

Growth and field emission of multi-walled carbon nanotubes

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

GROWTH AND FIELD EMISSION OF
MULTI-WALLED CARBON NANOTUBES

by

Soo Young Kim

A Thesis Presented to the
FACULTY OF THE USC VITERBI SCHOOL OF ENGINEERING
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(ELECTRICAL ENGINEERING)

May 2009

Copyright 2009 Soo Young Kim
ii
Dedication

To my beloved parents and my sister
iii
Acknowledgments
First and foremost, I would like to thank God for all the blessings that He has given
me. He has provided me with all the wisdom, knowledge and health that I needed to
complete this thesis, and I love Him dearly.
I would like to express my deepest gratitude to Dr. Andras Kuthi, who not only
molded me into the researcher that I am today but who also taught me so much more
than just research.
I would like to thank Dr. Martin Gundersen, my advisor, for giving me the
opportunity to work on this project, and my colleagues at USC Pulsed Power Group,
who has always been supportive of my work. Without their support and advice,
none of this would have been possible.
I also would like to thank Dr. Eun Sok Kim and Dr. Chongwu Zhou for serving on my
master’s thesis committee, and special thanks to Koungmin Ryu from Dr. Zhou’s
USC Nanotechnology Research Laboratory, who helped me with nanotube growth
and SEM images.
This work was funded by Air Force Office of Scientific Research (AFOSR).
iv
Table of Contents
Dedication……………………………...………………………………………….……………….ii

Acknowledgements……………………………...………………………………….………….….iii

List of Figures………………………………………………………………….……...…...………v

Abbreviations……………………………………………………………...……………………...vii

Abstract………………………………………………………..…….........................……….…..viii

Chapter 1 Introduction………….......................................................…………...…………….…1

Chapter 2 Properties of Carbon Nanotubes and their Growth Mechanisms……….…………….4

Chapter 3 Fowler-Nordheim Electron Tunneling Equation……………………………….…..…6

Chapter 4 Carbon Nanotubes as Field Emitters………………………………………………...10

Chapter 5 Thermally Grown Multi-Walled Carbon Nanotubes…………………..….…………15
5.1 DC Powered Field Emission Measurement of Thermally Grown MWCNTs….…...18
5.2 Pulse Powered Field Emission Measurement of Thermally Grown MWCNTs….…22

Chapter 6 MWCNT Grown Via PECVD...………………………………………………….….27
6.1 PECVD for Vertically-Aligned MWCNT Growth…………………………….…....30
6.2 DC PECVD Experimental Results………………………………………….……....34

Chapter 7 Future Work…………………..….…...…………………..….…...……….…………41
7.1 Si-JFET-Controlled MWCNT Field Emitter Arrays………………………..….…...41
7.2 And Beyond..……………………………………………………………….…....…43

References………………………………………………………………………….……...……..44
v
List of Figures
Figure 1. Field emission displays using Spindt tip and carbon film………………………………2
Figure 2. Structure of CNTs…………………………………………………………………….…4
Figure 3. Amplification factor of CNTs…………………………………………………….……..5
Figure 4. Different methods of CNT growth: Arc discharge, TCVD and PECVD……....……..…7
Figure 5. Formation of Ni islands as catalyst for MWCNT growth………………………………8
Figure 6. Process flow for the vertical-JFET MNCNT array fabrication………………….…......10
Figure 7. SEM images of vertical-JFET MNCNT arrays.…………………………………….….11
Figure 8. Set-up for field emission measurement testing.…………………………………….…..11
Figure 9. Samsung CNT technology………………………………………………………….…..12
Figure 10. Gated MWCNT FE array and it field emission………………………………….……13
Figure 11. Integrated gate MWCNT cathodes……………………………………………….…...14
Figure 12. MWCNTs grown on 5nm and 10 nm Ni layer…………………………………...…...17
Figure 13. MWCNTs grown on Fe catalyst layer (10 Å) ……………………….………...….….17
Figure 14. Field emission measurement vacuum chamber……………………………...….….…19
Figure 15. DC powered field emission measurement……………………………………….……20
Figure 16. Pulsed power circuit diagram for the field emission measurement…………………...22
Figure 17. Pulsed I-V characteristics of the MWCNT cathode grown in Ni catalyst layer......…..24
Figure 18. Field emission current drawn from MWCNTs with Ni catalyst layer…………….….25
vi
Figure 19. Field emission current drawn from MWCNTs with Fe catalyst layer…….……….….26
Figure 20. MWCNTs grown with TCVD and those with PECVD…………….…………………27
Figure 21. Selective MWCNT growth using e-beam lithography technique………….……....….28
Figure 22. Field enhancement versus distance between emitters. …………….………….………28
Figure 23. Relationship between emission current and nanotube arrangement…………..………29
Figure 24. PECVD setup schematic ……………………………………………………..……….31
Figure 25. Actual PECVD setup…………………………...…………………………….……….32
Figure 26. Inside the PECVD chamber……………………………………………………..…….34
Figure 27. Plot of pressure vs. Ar gas flow rate…………………………………….……….……35
Figure 28. Ar plasma. ……………………………………...………………………………….…36
Figure 29. NH
3
/C
2
H
2
plasma during MWCNT growth. ……...……………………………….…38
Figure 30. SEM image of MWCNTs grown via DC-powered PECVD…………………...……..39
Figure 31. SEM image of amorphous carbon with magnification of (a)11K and (b) 50K.............40
Figure 32. Vertical JFET array schematic…………………………………………………....…...42
Figure 33. Die Layout with extraction and JFET gate pads……………………………..….……42
vii
Abbreviations
C
2
H
2
: Acetylene
C
2
H
4
: Ethylene
CNT: Carbon Nanotube
CVD: Chemical Vapor Deposition
DC: Direct Current
FE: Field Emission
FED: Field Emission Display
FN: Fowler-Nordheim
High-Power Microwave: HPM
JFET: Junction Field-Effect Transistor
MFC: Mass-Flo Controller
MWCNT: Multiple-Walled Carbon Nanotube
NH
3
: Ammonia
LC: Inductor-Capacitor
PECVD: Plasma-Enhanced Chemical Vapor Deposition
RC: Resistor-Capacitor
RF: Radio Frequency
SEM: Scanning Electron Microscope
TCVD: Thermal Chemical Vapor Deposition
VJFET: Vertical Junction Field-Effect Transistor
viii
Abstract
Vertically-oriented multi-walled carbon nanotubes (MWCNT) were grown via
thermal chemical vapor deposition (TCVD) technique under different conditions by
using a hydrocarbon gas (ethylene) and a reactive gas (hydrogen). In addition, a
Plasma-Enhanced CVD station has been assembled by our group in order to grow
vertically-aligned MWCNTs using another hydrocarbon gas (acetylene) and reactive
gas (ammonia). These MWCNT will be studied for application to high-power
microwave (HPM) amplification and field emission displays. A pulsed diagnostic
circuit was designed to minimize the effect of stray capacitive current and was used
during the field emission measurement to maximize the emission current while
avoiding arcing. Results revealed an emission current of up to 0.8 A/cm
2
from the
MWCNTs at an applied anode voltage of 6kV . Collected field emission current data
was analyzed by using Fowler-Nordheim equation. Careful analysis of such results
demonstrated potential for the MWCNTs to be developed into high current-emitting
cathodes.
1
Chapter 1: Introduction

Multi-walled carbon nanotubes are promising materials to be applied in
nanotechnology, electronics, and optics due to their novel properties. Some of their
fascinating properties are their tensile strength of 63GPa, high elastic modulus of 1
TPa, low density of 1.4 g/cm
3
, and specific strength of up to 48,462 kN·m/kg, which
are the best among all of the known materials [1-3]. Theoretically, carbon
nanotubes have electrical conductivity of more than one-thousand times greater than
highly conductive metals such as silver and copper [1]. In addition, CNTs exhibit
high electrical conductivity at room temperature, and they have high aspect ratio
(diameter/length) due to their whisker-like shape.
Several techniques are known to effectively grow carbon nanotubes. These
techniques include growth through chemical vapor deposition (CVD), arc discharge,
and laser ablation. Among these, thermal CVD is the only process which does not
require electric field for growth and solely uses chemical reactions through thermal
heating for MWCNT growth. Therefore, arcing is not a problem during thermal
CVD growth while it is so for others.
The electron beams emitted from MWCNT cathodes can improve the
beam-wave energy conversion efficiency in high power microwave (HPM) devices.
2
Studies have shown that voltage and current waveforms of the carbon-fiber cathode
were found to be more stable and smooth than of molybdenum cathodes, while CNT
turn-on times were shorter and their durations were longer, indicating that the
carbon-fiber cathode was capable of producing high-quality electron beams [4].
Hence, it was concluded that the microwave power generated from carbon fibers was
greater than that for molybdenum, a commonly used material for HPM amplification.
Currently, Spindt cathodes with etched silicon and molybdenum tips are being
used for commercial field-emission displays (FEDs), and they emit about 50 mA/cm
2
.
Our goal is to emit 1 A/cm
2
with MWCNTs. If MWCNT arrays can be
manufactured for FEDs, then the intensity of the images displayed can be enhanced
while increasing the number of pixels due to their small size when compared to
arrays made out of silicon and molybdenum.

Figure 1. Field emission displays using Spindt tip and carbon film [5].
3
In order to further study the field emitting properties of MWCNTs, MWCNTs
were grown by thermal CVD technique and were tested with custom-made field
emission measurement station. Furthermore, a custom-made plasma-enhanced
chemical vapor deposition system was assembled to generate vertically aligned
MWCNTs with uniform length and diameter size.

4
Chapter 2: Properties of Carbon Nanotubes
and their Growth Mechanisms

Carbon nanotubes are very stable emitters even at high temperatures and can
carry a current density of 10
9
A/cm
2
[6]. In addition, they do not suffer from
electromigration because they are covalently bonded [6]. As shown in Fig. 2(d),
the n=2 (quantum number) has four electrons, and three of these electrons are
bonded to its three nearest neighbors by sp
2
bonding, in a manner similar to graphene
while the fourth electron is a π orbital perpendicular to the cylindrical surface.

Figure 2 (a) Creation of a (n,0) zigzag nanotube. a
1
and a
2
are the lattice vectors of
graphene. SWNTs are equivalent to cutting a strip in the graphene sheet (blue) and
rolling them up such that each carbon atom is bonded to its three nearest neighbors;
(b) Creation of a (n,n) armchair nanotube; (c) A(n,m) chiral nanotube; (d) The
bonding structure of a nanotube. [7]

5
The most obvious difference of carbon nanotubes over metals is that the
resistance of CNTs decreases with temperature and thus limits heat generation
(Q=I
2
R). Approximately two-thirds of single-walled CNT are semiconducting
while multi-walled CNTs are all metallic. In case of multi-walled carbon nanotubes,
they can be heated up to 2000K by their field-emitted current and yet remain stable.
Therefore, MWCNTs are ideal to be used as field emission sources [8].
Field enhancement factor is another attractive property of CNTs used as field
emissions sources. Field enhancement factor is determined by the geometric aspect
ratio (length over width) of CNTs, and if the aspect ratio for a certain MWCNTs is
500, then it means that the nanotubes can enhance their effective field to as high as
500 times the local field. Since MWCNTs can be grown with diameters smaller
than 100 nm, their aspect ratios can be significantly greater than other materials.

Figure 3 (a) Amplification factor β is determined by height h over radius r; (b)
Effective field improved by the amplification factor by β. [9]

a) b)
6
Chapter 3: Fowler-Nordheim Electron
Tunneling Equation

Fowler-Nordheim (FN) electron tunneling equation describes the field emission
current density. Tunneling current density, J, through a potential barrier between a
metal surface and vacuum is defined by de Jonge et al. [10] as:






− = ) (
3
2 8
exp
) ( 8
2 / 3
2
2 3
y v
heF
m
y t h
F e
J
φ π
φ π

Where

The equation is then simplified by using several constants and is converted into
FN linear fit:

Note that the electric field, F, is the amplification factor times the local field, the
applied voltage divided by the electrode gap (between the cathode and the anode).
The current density is expected to increase exponentially with greater aspect ratio
and local field and with smaller electrode gap while it would exponentially decrease
with greater work function (barrier height). A linear fit of the measured current
versus voltage data as in the modified FN equation above illustrate the validity of
φ πε φ
F
c
F e
y
3
0
3
4
1
= =
7
field emission data. Since MWCNTs have great aspect ratio and relatively small
barrier height of 4.9 eV [11], they can be used as field emission sources to obtain
high field emission current density.
Three common techniques illustrated in Figure 4 are used to grow MWCNTs;
these are arc discharge, thermal chemical vapor deposition (TCVD), and
plasma-enhanced chemical vapor deposition (PECVD).

Figure 4. CNTs grown by: (a) Arc discharge method [12], (b) TCVD system [13],
and (c) PECVD system [14].
(a)
(b)
(c)
8
CNTs were first grown by Ijima who used arc discharge method [15]. Graphite
rod was the cathode material, and the CNTs were grown on the surface of the
cathode when voltage was applied onto the electrodes (cathode and anode). CNTs
grown through arc discharge had limitations such as random orientation of nanotubes,
weak adhesion onto the surface, and poor growth control. Eventually, TCVD was
developed to improve the quality of CNTs. Under the TCVD technique, MWCNTs
is grown by chemical reaction between hydrocarbon gas (acetylene C
2
H
2
or ethylene
C
2
H
4
) and hydrogen (H) gas at high temperatures. After the dissociation of atoms
from the chemical reaction, single or multiple carbon radicals from the free surface
would diffuse into the metal catalyst layer and develop into CNTs while other single
and compound radicals would react with each other and with amorphous carbon and
eventually pumped out of the system. Catalyst layer is a thin metal layer that
nucleates into small individual islands after the surface treatment through annealing.
These islands then become growth sites for CNTs. As shown in Fig. 5, the width of
CNTs depends on the size of the nucleation sites and therefore plays a key role in
controlling the diameter of the nanotubes.
9

Figure 5. Formation of Ni islands after annealing Ni thin films of varying thickness
at 700°C [16].

Although vertically oriented CNTs can be grown by using TCVD technique, its
operating temperature is typically well above 600°C and therefore cannot directly
integrate CNTs into semiconductor devices. The devices would be damaged if kept
at long time at this temperature. In addition, CNTs grown by TCVD are vertically
oriented but the individual CNTs are spaghetti-like in shape as shown in Figure 4(b).
In order to improve the field emission from the CNTs, PECVD technique has
recently been developed, and it has successfully grown vertically aligned individual
MWCNTs as shown in Figure 4(c). Reaction between acetylene (C
2
H
2
) and
ammonia (NH
3
) gases in the presence of plasma ions form into free radicals while
dissociated carbon radicals diffuse into the catalyst layer to form into CNTs. But
unlike TCVD, PECVD technique can be operated at low temperature, even at room
temperature [17], and the nanotubes can be vertically aligned due to the applied
electrical field that is applied between the electrodes during growth.
10
Chapter 4: Carbon Nanotubes as Field Emitters

Our current goal is to generate 1A/cm
2
of uniform field emission current from
the MWCNT arrays, and in order to achieve such current uniformity, vertical
junction-effect transistors (VJFETs) were designed by Dr. Qiong Shui from USC
Pulsed power Group to integrate MWCNT onto the transistor and to be used as
current controlled arrays (Fig. 6).

Figure 6. Process flow for the vertical-JFET MNCNT array fabrication.

Dr. Shui was also able to fabricate the VJFET and grow MWCNTs on Si posts as
shown in Fig. 7 by using the PECVD system from the NASA Ames Research center,
with the assistance from Dr. Alan Cassell. The nanotubes had lengths varying from
11
700 to 900 nm with diameters of 50 to 90 nm at the center while they had lengths
varying from 50 to 600 nm with diameters of 15 to 40 nm around the edges. But
only few arrays had MWCNTs grown onto them.

Figure 7. SEM images of vertical-JFET MNCNT arrays.

These vertical-JFET MNCNT arrays were tested for field emission with a setup
shown in Fig. 8. Due to the non-uniformity of size and quality the MWCNTs, field
emission current from each site was only few nanoamperes. Therefore, an
improvement in JFET fabrication and MWCNT growth is needed.

Figure 8. Set-up for field emission measurement testing.

12
Currently, large-scaled research projects are being conducted in Europe (France
and England), Korea (Samsung, Fig. 9), and the United States (NASA Ames
Research Laboratory and Naval Research Laboratory) to generate 1A/cm
2
of uniform
field emission from the MWCNT arrays, but it is yet to be achieved.

Figure 9. Samsung CNT technology- (a) Picture representing the set-up of a field
emission display; (b) SEM image of nanotube bundles projecting from the metal
electrode; (c) A sealed CNT field emission display (4.5 in.) emitting light in three
different colors. [7]

Using self-aligned process, MWCNT micro-cathodes has been fabricated by
several groups to achieve high field emission from carbon nanotubes. To date, Teo,
et al. [19] has designed several MWCNTs (approximately ten) on each emitter
window of 0.8μm in diameter and it generates 3mA/cm
2
of field emission current at
48V of gate voltage (Fig. 10).
13

Figure 10 (a) Gated MWCNT FE array; (b) Plot for peak emission current vs. gate
voltage. Turn-on voltage is 15V while its peak current density is 3mA/cm
2
at 48V .
[19]

Another design from Gangloff, et al. [20] has single MWCNT grown on each
emitter window of 0.5 μm as shown in Fig. 11, and it generates maximum field
emission current of 5 μA from an individual CNT at 46 V . Considering that there is
a space of 10 μm between each emitter sites, it should ideally generate about 5 A/cm
2
.
But due to the non-uniformity of the nanotubes structures, it emits well below 1
A/cm
2
of current. Therefore, new technique needs to be developed in order to
enhance the quality of MWCNT and its emission controlling device. VJFET is a
great candidate to be used as a controlling device due to its ability in current control,
and PECVD growth technique is a great method due to its ability to control the
orientation of the nanotubes growth.

14

Figure 11. Integrated gate MWCNT cathodes - (a) Tip view of integrated gate CNT
cathode; (b) Cross section of SEM view of the integrated gate CNT cathode. The
isotropic etching of the gate and the insulator prevent short circuits between gate and
emitter; (c) Field emission measurements: turn-on voltage of 25V and a maximum
current of 5μA at 46V . A Fowler-Nordheim (FN) linear fit illustrates validity of the
field emission data. [20]

(c)
15
Chapter 5: Thermally Grown Multi-Walled
Carbon Nanotubes

MWCNTs were grown on top of the silicon substrate with 10Å of nickel catalyst
layer and 30Å of chromium adhesion layer by using the TCVD system from Dr.
Chongwu Zhou’s USC Nanotechnology Research Laboratory. TCVD system was
composed of a furnace, quartz tube and gas injection system for ethylene, hydrogen
and argon gases as shown previously in Fig. 4(b). During sample preparation,
nickel and chrome metal layers were directly deposited on top of the Si substrate
using e-beam evaporator. The Ni/Cr deposited Si substrate sample was inserted
into the quartz tube and was heated to 550°C while argon gas was injected in the
tube during heating in order to prevent oxidation of the metal layer. One thing to
note at this point is that the melting temperature of Ni layer decreases from its
original melting point of 1453°C to around 800°C due to its atomic level thickness
that makes Ni atoms more reactive than usual when exposed to elevated temperature.
When the furnace temperature was at 550°C, the substrate sample was pretreated
with hydrogen gas for 10 minutes. During pretreatment, Ni particles aggregated
and formed into islands in the order of tens of nanometers and became nucleation
sites for MWCNT growth as shown previously in Fig. 5. After pretreatment, the
16
substrate was further heated to 800°C with Ar gas, and once the desired peak
temperature of 800°C was reached, 350 sccm of ethylene gas and 700 sccm of
hydrogen gas were injected into the quartz tube. Activated by the high temperature,
ethylene gas and hydrogen gas reacted and decomposed into carbon, hydrocarbon,
and hydrogen radicals. While other radicals reacted with one another and remained
as gases, carbon radicals diffused into Ni nucleation sites extruded off the catalyst as
the growing nanotubes via sp
2
bonding.
As the result, vertically-oriented MWCNTs were grown on top of the Fe catalyst
layer, and detailed SEM imaging illustrated that vertically-oriented individual
nanotubes were spaghetti-like in shape and was entangled with one another as
illustrated in Fig. 12. These nanotubes had about 50 nm in diameter and 4 μm in
length for an aspect ratio of 80. In order to achieve vertical orientation of MWCNT
growth using TCVD technique, the nanotubes should be as dense as possible (no
more than 100nm apart from each other) so that the nanotubes would push each other
up during growth and grow vertically. While several other recipes were used to
grow MWCNTs by changing pretreatment temperature, growth temperature, catalyst
layer thickness, and reaction gas flow rate during the experiment, the recipe
described above generated best result in terms of uniformity in nanotubes thickness,
density and length.
17

Figure 12. MWCNTs grown on top of (a) 5nm and (b) 10 nm Ni catalyst layer.

Similar growth procedures were used to grow the MWCNTs on Si substrate with
10 Å of iron (Fe) catalyst layer but without the adhesion layer. The substrate was
heated to 550°C with Ar gas and pretreated with hydrogen gas at 550°C for ten
minutes. After the substrate was further heated to 820°C, ethylene and hydrogen
gases were injected into the quartz tube to react and form into MWCNTs.

Figure 13. MWCNTs grown on Fe catalyst layer (10 Å). Lengths of the nanotubes
are approximately 200 μm and their diameters vary from 20 nm to 100nm.
18
5.1. DC Powered Field Emission Measurement of
Thermally Grown MWCNTs
Field emission current of thermally grown MWCNTs was measured by using a
custom-built field emission measurement system as shown in Fig. 14. Inside the
chamber, 316 stainless steel anode and cathode were placed about 1cm apart from
each other, and the linear motion manipulator attached onto the cathode freely
adjusted the gap at will. Linear motion manipulator was insulated from the cathode
by a viton o-ring and the emitted current was collected through enamel-insulated
copper wire attached onto the cathode. 100μm thick stainless steel mask that has an
exposed area with a diameter of 1cm (corresponding MWCNT exposure area of
0.785cm
2
) was used to screen the MWCNT Si substrate, and it prevented arcing on
the sharp edges of the substrate during high voltage measurements. Vacuum system
used for the field emission experiment was composed of three stage pumps: dry
membrane roughing pump (down to 0.5mTorr), turbomolecular pump (down to
6×10
-6
Torr) and ionization pump (down to 7×10
-7
Torr). All the field emission
measurements were conducted at 700 nTorr.
19

Figure 14. Field emission measurement vacuum chamber - (a) Schematic of the
chamber with anode and cathode; (b) Actual picture of the chamber; (c) Detailed
schematic of the cathode. Note that the cathode is floating.

For the first experiment, dc-power supply was used to measure the MWCNT
cathode grown on top of Ni catalyst, and the gap between the cathode and the anode
was adjusted by using the linear motion manipulator. Our setup was also used with
pulsed voltage. In pulsed mode we measured two different currents: capacitive
current from the electrodes and the field emission current. The capacitive current
was subtracted from the total current measurement in order to distinguish the field
emission current from the total current. After several sets of experiments, the
20
nanotubes generated maximum field emission current density of 0.38mA/cm
2
at an
applied dc voltage of 2.7kV with an electrode gap of 0.88mm.
The measurements were analyzed using Fowler-Nordheim equation (Ch. 3),
which described the relationship between the applied voltage and the field emission
current. And a linear fit, represented as y=mx+b, was used with Ln[I/V
2
] as the y
term, 1/V as the x term, and [-E
0
·d·Ln(I
0
/d
2
·E
0
2
)] as the slope, m, where d = electrode
gap, I
0
= 6.3 A (local current) and E
0
= 19.1kV (local field). The FN linear fit of the
data demonstrated the validity of the measured current as the field emission current
as plotted in Fig. 15. During the measurement, the nanotubes started to heat and
often became red-hot. Undesired arcing also occurred during the measurement
between the electrodes due to the unparalleled surface between the electrodes and led
to uneven field emission depending on the location of the nanotubes, which
eventually led to arcing. But it was also found that arcing and red-hot heating of
the nanotubes burned off the weak nanotubes and impurities such as dusts.
Therefore, field emission current eventually increased and arcing reduced since
arcing reduced unwanted nanotubes and impurities while stable nanotubes continued
to emit without disturbance.
21

Figure 15. DC powered field emission measurement.

For the next dc-powered measurement, the gap between the cathode and the
anode was decreased from 0.88mm to 0.22mm, and an improvement in field
emission current was expected since smaller gap would increase the electric field.
Field emission current of 0.13mA/cm
2
was measured at an applied voltage of 750V
as shown in Fig. 15 while the previous experiment with 0.88mm electrode gap
required more than 2.3kV of applied voltage to achieve such current level. But
further increase in voltage generated severe arcing and thus prohibited measurement
beyond these points.

22
5.2. Pulse Powered Field Emission Measurement of
Thermally Grown MWCNTs
Our next step was to measure the field emission current from the nanotubes
using pulsed power. The advantage of using pulsed power over dc power for the
field emission measurement is that field emission current values can be analyzed at
different applied voltages within one single shot of pulse. In addition, short pulse
width greatly reduces the chances for the electrodes to arc and thus higher voltage
can be applied without arcing. For the setup, pulse powered circuit was as shown in
Fig. 16. The circuit was powered by a dc power supply and a pulse generator
connected with a silicon-controlled rectifier (SCR) switch. As the pulse triggered
the SCR, the voltage from the DC power supply was delivered to the LC circuit.
The transformer with the ratio of one-to-ten amplified the oscillating voltage and
delivered the voltage to the RC circuit that consisted of a 500kΩ current limiting
resistor and a capacitor composed of MWCNT cathode and stainless steel anode. A
voltage divider was used in order to protect the oscilloscope and it decreased the
actual voltage by one-thousandth. Dual back-to-back parallel diodes limited the
voltage at the scope input in case of high voltage arcing. A 50Ω resistor at I
out
was
used to measure the total current from the MWCNT capacitor/field-emitter
combination.
23

Figure 16. Pulsed power circuit diagram for the field emission measurement

Pulsed power measurements for the MWCNT cathode with the Ni catalyst layer
(shown in Fig. 12(b)) generated two different peaks of current, one from the
capacitive current I
C
, and the other from the field emission current I
FE
(Fig. 17). It
was found that the peak capacitive current I
C
was 9mA at a peak anode voltage V
out

of 6kV and the pulse width was 24μsec.
24

Figure 17. Pulsed I-V characteristics of the MWCNTs grown on Ni catalyst layer.

Since the applied voltage was known, it was used to derive the capacitive current,
and subtracting the capacitive current from the total current resulted in emission
current curve with respect to time (Fig. 18(a)). According to the calculation, the
highest emission current obtained was 12 mA/cm
2
at 10 μsec, and it surpassed the
peak capacitive current by 2mA. A linear fit was generated from the data to
observe the quality of the measured field-emission data, similar to the DC
measurements, and the linear fit demonstrated validity of our measured field
emission current (Fig. 18(b)).
25

Figure 18 (a) Behavior of emission current with respect to time for MWCNTs with
Ni catalyst layer; (b) Corresponding Linear plot using Fowler-Nordheim equation.

As expected, higher voltage and a smaller electrode gap (1mm) between the
anode and the cathode generated higher emission current, but further reducing the
gap between the electrodes while maintaining the anode voltage generated heavy
arcing. Similarly, anode voltages higher than 6kV generated undesired arcing
between the nanotubes and the anode. A possible solution to this problem is to set
the electrodes perfectly parallel to each other. Since arcing is more likely to occur
with smaller electrode gaps, such adjustment would equalize the distance between
the cathode and the anode at any location and thus would not generate selective
arcing at specific locations where the electrode gaps are smaller than others.
Pulsed power measurements for the MWCNT cathode with the Fe catalyst layer
revealed an emission current of 0.8 A/cm2, which was approximately 70 times
greater than that from the MWCNTs with the Ni catalyst layer. Greater emission
(a)
(b)
26
current was expected from these nanotubes since their aspect ratio (β = height /
diameter = 4000) was much greater than the aspect ratio (β = 80) from the nanotubes
with the Ni catalyst later.
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
0 5 10 15 20 25 30 35 40
time (μsec)
J (A /cm
2
)
Ln(I/V
2
) = -3.9299(1/kV) - 16.417
R
2
= 0.8356
-21
-20.5
-20
-19.5
-19
-18.5
-18
-17.5
0.4 0.5 0.6 0.7 0.8 0.9 1
1/kV
L n (I/V
2
)

Figure 19 (a) Behavior of emission current with respect to time for MWCNTs with
Fe catalyst layer; (b) Corresponding Linear plot using Fowler-Nordheim equation.

Field emission current can be increased by improving the flaws described above,
but the main flaw that needs to be adjusted is the geometrical shape of the nanotubes.
Since thermally-grown MWCNTs are spaghetti-like in shape and have different
lengths, their field emission properties are unstable. But if we were to grow
MWCNTs that are vertically aligned and are similar in length, then their field
emission properties would greatly improve. Therefore, a PECVD system that
grows vertically-aligned MWCNTs was designed with desire to achieve our field
emission goal of 1A/cm
2
.

27
Chapter 6: MWCNTs Grown Via PECVD

The greatest advantage of using PECVD over TCVD for MWCNT growth is
that the effective amplification factor of the nanotubes is much greater when grown
under PECVD. Close-packed CNTs are not ideal because close packing of the
tubes screens the applied field, and hence reduces the field enhancement for low
aspect ratio as illustrated in Fig. 20(a). But on the other hand, vertically-aligned
CNTs can control screening of the applied field and thus can obtain high field
enhancement (Fig. 20(b)).

Figure 20 (a) MWCNTs grown with TCVD; (b) MWCNTs grown with PECVD. [5]

In addition, MWCNTs can be grown selectively via PECVD (Fig. 21), meaning
that MWCNTs can be grown at any location as desired using e-beam writing
lithography technique. Depending on the size of the catalyst, either single or
28
multiple nanotubes can be grown. From Fig. 21, it can be observed that catalysts
with diameters of 100 nm – 300 nm generated single nanotubes while catalysts with
greater diameters generated multiple nanotubes.

Figure 21. Selective MWCNT growth using e-beam lithography technique [19].

Nillson et al. [19] ran a simulation to demonstrate that closely spaced tips screen
field enhancement and that nanotube spacing must be equal to or greater than twice
their height in order to obtain maximum field enhancement (Fig. 22 (b)).

Figure 22. Field enhancement versus distance between emitters [5, 19].
29
Relationship between field emission current and spacing between the nanotubes
is further explained by Meletchko et al. [13] in Fig. 23. Their experiments
demonstrate that densely packed nanotubes generate the smallest emission current at
0.3 μA and sparsely arranged nanotubes generate moderate current at 1 μA while the
nanotube arrays, are separated approximately twice the height of the nanotubes,
generate the greatest current at 5.5 μA. Although densely packed nanotubes are
much longer than sparsely arranged nanotubes, sparsely arranged nanotubes generate
more than twice the current of the densely packed nanotubes. Similarly, nanotubes
in array formation are still shorter than densely-packed nanotubes, but the spacing
between the nanotubes not only compensates for the shorter length but also generates
much greater emission current due to field enhancement. Therefore, MWCNT
growth via PECVD is the ideal method to achieve optimal field emission.

Figure 23. Relationship between emission current and nanotube arrangement - (a)
densely packed MWCNTs; (b) sparsely arranged MWCNTs; (c) MWCNT arrays
where nanotubes are separated from each other by twice their heights; (d) plot
illustrating that MWCNT arrays are superior over others. [13]
30
6.1. PECVD for Vertically-Aligned MWCNT Growth
Plasma-enhanced chemical vapor deposition is a deposition technique in which
plasma is created from the reactant gases by using a designated power source such as
direct current power or radio frequency power. Once the plasma is ignited, the ions
in the plasma are in excited states and thus easily react with each other and with the
Ni nucleation sites from the Si substrate without the need for an elevated temperature
as in the conventional (thermal) CVD.
For our MWCNT growth experiment, acetylene (C
2
H
2
) gas is used as the carbon
source while ammonia (NH
3
) gas is used as the catalyst and dilution gas. Vertically
aligned carbon nanotubes with controllable diameters and lengths can be obtained
through this process due to the local electric field applied from the cathode to the
anode that controls the growth orientation of the nanotubes and due to the catalyst
diameter size ions that controls the diameter of the nanotubes. According to the
MWCNT growth research conducted at NASA Ames Research center [14], the
catalyst material (nickel, Ni) undergoes a surface pretreatment for 10 minutes as the
tungsten filament elevates the temperature of the environment at a current of 20 A
and a corresponding power of 150 - 200 W in NH
3
atmosphere at a pressure of 4 Torr.
Filament provides direct radiation to the substrate and indirect chemical energy
transfer in H/NH
3
dissociation and recombination process. During the surface
31
pretreatment, NH
3
dissociates into N
3+
, H
+
, NH
2+
, and NH
2
+
radicals, and they
recombine into N
2
and H
2
as well. After the pretreatment process, NH
3
and C
2
H
2

gases are injected simultaneously into the growth chamber with a flow rate of 80 and
22.5 cubic centimeter per second or sccm, respectively, to generate 380 W plasma
with a corresponding voltage and current of 525 V and 730 mA, respectively, for
optimal growth (Fig. 16).
Using the NASA Ames Research Center PECVD system as the reference guide,
PECVD system was custom-designed and assembled by our group (Fig. 24, 25).

Figure 24. Schematic of the PECVD setup.
32

Figure 25. Actual PECVD setup.

Two four-way crosses from MDC were connected and used as the PECVD
chamber while rotary vane vacuum pump controlled the pressure level for the system.
Steel wool trap was connected between the vacuum pump and the chamber in order
to prevent oil backstreaming from the pump since oil backstreaming from the
vacuum pump can severely contaminate the vacuum chamber with oil vapors.
Throttle valve was also attached between the vacuum pump and the chamber in order
to regulate the pressure level of the chamber when exposed to gas injection and a
capacitance manometer with a pressure measurement range between 1m Torr and
100 Torr was attached on top of the four-way cross for pressure monitoring.
33
Multi-port 6-inch flange was installed on top of the four-way cross, and the
multi-port flange contained linear motion manipulator, two gas tubes, and three DC
power supply feedthroughs. Linear motion manipulator held the anode and one of
the nickel-wired feedthroughs powered the cathode. Two feedthroughs were used
to power the tungsten filament used for pretreatment of nickel catalyst layer under
ammonia environment as described on the previous paragraph. One of the gas
tubing supplied controlled flow rate of gases into the chamber via mass-flo
controllers (MFCs) while the other gas tubing was used to purge the system with
argon gas before and after the injection of NH
3
and C
2
H
2
gases. Cathode and anode
were made of 316 stainless steel and had a polished 2.5 inch diameter surface within
a separation distance of 2.5 inches while the linear motion manipulator that held the
anode could decrease the separation distance to as low as 0.5 inches. The anode,
perforated with thirty-six perforated small holes, functioned as a showerhead for the
gases to flow through the anode and to effectively reach the cathode for plasma
generation. On top of the stainless steel cathode, a graphite spacer, a tantalum foil,
and a 1cm x 1cm substrate was placed respectively, and a custom-made aluminum
cover on the top of the multi-port flange was added to provide safety cover and
grounding for the chamber.
34

Figure 26. Inside the PECVD chamber

6.2. DC PECVD Experimental Results
First, the vacuum chamber was tested to investigate the dependency of pressure
level with respect to the argon gas injection in order to observe the reliability of the
vacuum pump and Mass-Flo controller. As illustrated on Fig. 27(a), Argon gas
injection varied from 0 to 100 sccm and the chamber pressure varied from 1 mTorr to
11 Torr. It was demonstrated that the chamber pressure level did not have a linear
relationship with the gas flow rate. This is because the pressure was being limited
by the pumping speed of the roughing pump. A closer observation was made by
varying the Ar flow rate from 0 to 10 sccm (Fig. 27(b)) and it was revealed that the
35
pressure remained at 1 mTorr until the flow rate was greater than 2 sccm. After 2
sccm, the pressure increased somewhat linearly until the flow rate reached to 10
sccm. Such observation further proved non-linear relationship of pressure level and
gas flow rate and demonstrated that throttle valve should be used to control the
pressure level rather than controlling the flow rate of the gases.

Figure 27. Plot of pressure vs. Ar gas flow rate: (a) from 0 to 100 sccm and (b) from
0 to 10 sccm.

Next step was to generate Ar plasma and observe its stability at different
pressure levels. The pressure was set to 4 Torr and the chamber was only filled
with Ar gas. Plasma started to appear at 30 W with corresponding voltage and
current of 600 V and 50 mA, and its intensity increased with increase in DC power
level. But once the power level reached to 70 W (900 V and 100 mA), plasma
generation was interrupted by arcing between the electrodes. In order to generate
plasma with stronger intensity, the pressure level was raised tom 6 Torr by using Ar
36
gas. As expected, higher pressure level provided higher stability because the higher
ion density generated required lower voltages. Plasma was generated without any
arcing above 70 W, and it reached up to 200 W with its corresponding voltage and
current of 1000 V and 200 mA until arcing occurred. At this power level, the
tantalum foil turned red-hot as desired. Red-hot tantalum foil signified that the
plasma heated the tantalum foil to a degree that is between 500 and 790 °C. It is
important that the tantalum foil is heated to a high temperature because it will
function as an external heater to heat the silicon substrate for nickel layer
pretreatment and nanotube growth. There is a need to precisely control the
temperature of the tantalum foil since the device can be destroyed if the tantalum foil
reaches to a temperature that is higher than the melting temperature of the Ni catalyst
layer. But for now, we will continue adjusting the temperature of the tantalum foil
by adjusting the intensity of the plasma rather than using an external heater.

Figure 28. Ar plasma at 70 Watts.
37
After successful Ar plasma generation, NH
3
/C
2
H
2
plasma was generated for
MWCNT growth. NH
3
and C
2
H
2
gases were injected simultaneously into the
growth chamber with a flow rate of 80 sccm and 22.5 sccm, respectively, and the
chamber pressure was adjusted to 4 Torr. But when the power was raised to 50 W
with corresponding voltage and current of 1000 V and 50 mA, the electrodes started
to arc. Such result revealed that NH
3
/C
2
H
2
plasma was more vulnerable to arcing
when compared to Ar plasma since Ar plasma started to arc at 70 W with
corresponding voltage and current of 600 V and 50 mA. This is because higher
voltage (1000 V instead of 600 V) was applied to draw the same amount of current
(50 mA), and such high voltage would make the electrodes more susceptible to
arcing. And since plasma was generated at under 70 W, its intensity was too weak
for MWCNTs to be grown. It is important that the plasma generates high enough
intensity to heat the substrate and activate the gas particles at the same time, but
apparently, 70 W was not high enough for such phenomena to occur.
On the next experiment, the pressure level was fixed at 6 Torr while gas flow
rates remained constant from previous experiment at 80 sccm and 22.5 sccm for NH
3

and C
2
H
2
gases, respectively. At first, the chamber was injected with NH
3
for
pretreatment of Ni catalyst layer. Supplied power was fixed at 80 W with
corresponding voltage and current of 1000 V and 80mA. At this point, there was
38
no arcing although the applied voltage was identical to the one from our previous
experiment at 4 Torr. After surface pretreatment for 10 minutes, power supply was
turned off and C
2
H
2
gas was introduced into the chamber.
Once the pressure stabilized at 6 Torr, the power supply was turned on and it
reached to 130 W with its corresponding voltage and current of 1300 V and 100 mA.
At this point, plasma was at high intensity (Fig. 29). There was some arcing, but
arcing appeared once or twice for every 30 seconds, and therefore, the experiment
was carried on, expecting MWCNT growth. After 15 minutes of growth, plasma
was turned off, and the sample was observed by SEM.

Figure 29. NH
3
/C
2
H
2
plasma during MWCNT growth at 80 Watts.

As revealed in Fig. 30, multi-walled carbon nanotubes were grown, but their
diameter were too thin to be grown vertically. In order for the nanotubes to grow
vertically, they need to have a minimum diameter of 50 nm [21], but the SEM
imaging suggested that the diameter of the nanotubes were about 10 nm.
39

Figure 30. SEM image of MWCNTs grown via DC-powered PECVD

SEM imaging on the edges of the sample as shown in Fig. 31 revealed that
amorphous carbon was dominant over carbon nanotubes on certain areas. Such
phenomenon could have occurred due to the arcing that destroyed crystalline
bonding among carbon atoms or due to the need for a higher plasma intensity to
elevate the temperature and energy for MWCNT growth. Therefore, a possible
solution is to increase the pressure level in order to increase the power level supplied
onto the anode. Another reason for this could be due to lack of hydrogen radicals.
Free hydrogen radicals have the ability to eliminate amorphous carbons by bonding
with them and converting amorphous carbons into gases, but it is possible that there
were not enough hydrogen radicals to eliminate them. Therefore, a possible
solution might be to increase the flow rate of ammonia while maintaining the flow
40
rate of acetylene consistent. Amorphous carbon can be the reason for the growth of
the nanotubes with such thin diameter since carbon atoms that are supposed to
contribute to nanotube growth are being used up to grow amorphous carbon. If a
solution is found to eliminate amorphous carbon growth, then the diameter of the
nanotubes should be thicker and thus should be able to grow vertically as desired.
Further experiments will take place in attempt to grow vertically aligned MWCNTs
in the future with dc-powered PECVD.

Figure 31. SEM images of amorphous carbon grown via PECVD with magnification
view of (a)11K and (b) 50K.

41
Chapter 7: Future Work

7.1. Si-JFET-Controlled MWCNT Field Emitter Arrays
Once MWCNTs can be successfully grown by using our PECVD system, they
will be grown on top of Si-JFETs to be used as controlled field emitter arrays. Dr.
Andras Kuthi has modified Dr. Shui’s design of Si-JFET arrays by eliminating extra
gate contact and Si posts while and adding SiO
2
trenches for insulation and contact
pads for extraction gates and JFET gates as shown in Fig. 32. Elimination of a gate
contact will provide more space between the gates while Si posts will eliminate
several processing steps. Si posts were eliminated because it was learned that they
do not contribute to the field emission of CNTs. On the other hand, SiO
2
trenches
were added to isolate each element of the array from another. Also, contact pads
were added as shown in Fig. 33 to ease device testing. Now with our new design,
100x100 arrays will have one extraction gate and JFET gate pads that are connected
to individual gates and thus will simplify contact with external sources.

42

Figure 32. Vertical JFET array schematic.

Figure 33. Die Layout with extraction and JFET gate pads.

Once the device is fabricated with MWCNTs grown on the emitter site (on top
of the Ni layer), the arrays can be tested for field emission. Our thermally-grown
MWCNT field emission measurements generated up to 0.8A/cm
2
, and with the
Ni/Cr
43
vertically-aligned MWCNT arrays grown by our PECVD system in the near future,
we can expect their field emission current to be greater than 1A/cm
2
.

7.2. And Beyond
Other future work includes successful vertically-aligned MWCNT growth via
DC and RF PECVD. The addition of RF power to the substrate allows the
reduction of DC voltage for a given power and will aid in eliminating arcing related
damage. To date, a prototype RF circuit has been developed by Jason Sanders and
Taehyun Kim from USC Pulsed Power Group, and it has successfully generated Ar
and NH
3
/C
2
H
2
plasmas. And once the complete circuit is built and tested, we will
attempt to grow MWCNTs using RF PECVD.
Furthermore, the use of e-beam writing lithography technique is under
investigation to grow MWCNT selectively (as demonstrated in Fig. 21) on the
VJFET arrays and thus maximize field emission from the nanotubes. In addition,
consistent MWCNT growth with uniform length will need to be investigated as well
by changing catalyst materials, reactant gases and growth recipes in order to
optimize field emission current of the nanotubes.

44
References

Anantram, M. P. and Leonard F. “Physics of Carbon Nanotube Electronic Devices.”
Rep. Prog. Phys. 69(3): 507–561 (2006). [7]

Bonard, J.M. et al. “Carbon nanotube films as electron field emitters.” Carbon
40:1715-1728 (2002).

Boskovic, B. O. et al. “Large-Area Synthesis of Carbon Nanofibers at Room
Temperature.” Nature Mat. 1:165-167 (2002). [17]

Chhowalla, M. et al. “Growth Process Conditions of Vertically Aligned Carbon
Nanotubes using Plasma Enhanced Chemical Vapor Deposition.” J. Appl. Phys.
90:5308 (2001). [15]

Collins, P. and Avouris, P. “Nanotubes for Electronics.” Scientific American, pp.
68-69, December 2000. [1]

Cruden, B. et al. “Reactor Design Considerations in the hot Filament/Direct Current
Plasma Synthesis of Carbon Nanofibers.” J. Appl. Phys. 94:6, 15 September 2003.
[16, 18]

de Jonge, N. et al. “Characterization of the Field Emission Properties of Individual
Thin Carbon Nanotubes.” Appl. Phys. Lett. 84: 2004, 30 Aug. 2004. [10]

Forro, L. et al. “Electronic and Mechanical Properties of Carbon Nanotubes.”
Departement de Physique, EPF-Lausanne, CH-1015 Lausanne, Switzerland.
[http://ipn2.epfl.ch/CHBU/papers/ourpapers/Forro_NT99.pdf] [2]

Hofmann, S. et al. “Low-Temperature Growth of Carbon Nanotubes by
Plasma-Enhanced Chemical Vapor Deposition.” APL 83:135 (2003).

Gangloff, L. et al. “Self-Aligned, Gated Arrays of Individual Nanotube and
Nanowire Emitters.” Nano Lett. 4:9 (2004). [20]

Gjerde, K. et al. “Integrating Nanotubes into Microsystems with Electron Beam
Lithography and In Situ Catalytically Activated Growth.” phys. stat. sol. (a)
203(6):1058–1063 (2006).
45
Iijima, S. “Helical Microtubules of Graphitic Carbon.” Nature 254: 56 (1991). [15]

Iljin Nanotech, “Schematic for the Arc Discharge System for Carbon Nanotube
Growth.” [http://www.iljinnanotech.co.kr/en/material/r-4-1.htm] [12]

Jensen, K. L. “Field Emitter Arrays for Plasma and Microwave Source
Applications.” Physics of Plasmas 6(5):2241-2253 (1999).

Melechko, A. et al. “Vertically Aligned Carbon Nanofibers and Related Structures:
Controlled Synthesis and Directed Assembly.” Journ. of appl. Phys. 97:041301
(2005). [13]

Meyyappan, M. et al. “Carbon Nanotube Growth by PECVD: A Review.” Plasma
Sources Sci. Technol. 12:205-216 (2003). [21]

Milne, W. I. “Carbon Nanotubes as Electron Sources.” J. Mater. Chem. 14:1–12
(2004). [8]

Nilsson, L. et al. “Scanning Field Emission from Patterned Carbon Nanotube Films.”
APL 75:2071 (2000). [19]

Lee, S. B. et al. “Fabrication and Electrical Characteristics of Carbon
Nanotube-Based Microcathodes for Use in a Parallel Electron-Beam Lithography
System.” J. Vac. Sci. Technol. B 21 3, May/Jun 2003.

Legagneux, P. et al. “GHz Modulation of Carbon Nanotube Cathodes for Microwave
Amplifiers.” Proceedings of 2005 5th IEEE Conference on Nanotechnology,
Nagoya, Japan, July 2005.

Lie, L. et al. “Fabrication of Carbon-Fiber Cathode for High-Power Microwave
Applications.” IEEE Transactions on Plasma Science 32(4):1742-1746, Aug. 2004.
[4]

Liu, P. et al. “Measuring the Work Function of Carbon Nanotubes with Thermoionic
Method.” Nano Lett. 8(2):647-651 (2008). [11]

Physica Status Solidi, WILEY-VCH Verlag GmbH & Co. 205(1), (2008). [9]

46
Robertson, J. “Field Emission Devices.” Lecture Notes. Cambridge University, UK.
[jr@eng.cam.ac.uk] [5]

Robertson, “Growth of Nanotubes for Electronics.” J. Materials Today 10: 1-2,
Jan-Feb 2007.

Sato, H. et al. “Fabrication of Carbon Nanotube Array and Its Field Emission
Property.” J. Vac. Sci. Technol. B 22.3, May/Jun 2004.

Shiffler, D. et al. “Emission Uniformity and Emission area of Explosive Field
Emission Cathodes.” APL 79:18, 29 October 2001.

Svizhenko, A. and Anantram, M. P. “Effect of scattering and contacts on current and
electrostatics in carbon nanotubes.” Phys. Rev. B 72, 085430 (2005). [6]

Teo, S. et al, “Characterization of Plasma-Enhanced Chemical Vapor Deposition
Carbon Nanotubes by Auger Electron Spectroscopy.” J Vac. Sci. Technol., B 20, 116
(2002). [20]

Umstattd, R. J. “Gas Evolution During Operation of a Csl-Coated Carbon Fiber
Cathode in a Closed Vacuum System.” IEEE Transactions on Plasma Science 33:2,
April 2005.

Utsumi, T. “Vacuum Microelectronics: What’s New and Exciting.” IEEE Trans.
Electron Dev. 38:2276 (1991).

Yu, M. et al. “Strength and Breaking Mechanism of Multiwalled Carbon Nanotubes
Under Tensile Load.” Science 287:637-640 (2000). [3]

Abstract (if available)

Abstract Vertically-oriented multi-walled carbon nanotubes (MWCNT) were grown via thermal chemical vapor deposition (TCVD) technique under different conditions by using a hydrocarbon gas (ethylene) and a reactive gas (hydrogen). In addition, a Plasma-Enhanced CVD station has been assembled by our group in order to grow vertically-aligned MWCNTs using another hydrocarbon gas (acetylene) and reactive gas (ammonia). These MWCNT will be studied for application to high-power microwave (HPM) amplification and field emission displays. A pulsed diagnostic circuit was designed to minimize the effect of stray capacitive current and was used during the field emission measurement to maximize the emission current while avoiding arcing. Results revealed an emission current of up to 0.8 A/cm2 from the MWCNTs at an applied anode voltage of 6kV. Collected field emission current data was analyzed by using Fowler-Nordheim equation. Careful analysis of such results demonstrated potential for the MWCNTs to be developed into high current-emitting cathodes.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A study of junction effect transistors and their roles in carbon nanotube field emission cathodes in compact pulsed power applications

PDF

Synthesis, assembly, and applications of single-walled carbon nanotube

PDF

Single-wall carbon nanotubes separation and their device study

PDF

GaAs nanowire optoelectronic and carbon nanotube electronic device applications

PDF

In2O3 COVID-19 biosensors and two-dimensional materials: synthesis and applications

PDF

Design and modification of electrocatalysts for use in fuel cells and CO₂ reduction

Asset Metadata

Creator Kim, Soo Young (author)

Core Title Growth and field emission of multi-walled carbon nanotubes

School Viterbi School of Engineering

Degree Master of Science

Degree Program Electrical Engineering

Publication Date 04/13/2009

Defense Date 03/27/2009

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag carbon nanotubes,CNT,field emission,Fowler-Nordheim,multi-walled chemical vapor deposition,MWCNTS,OAI-PMH Harvest,PECVD

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Gundersen, Martin A. (committee chair), Kim, Eun Sok (committee member), Zhou, Chongwu (committee member)

Creator Email alexkim83@hotmail.com,sooyoung@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2082

Unique identifier UC1131787

Identifier etd-Kim-2843 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-206502 (legacy record id),usctheses-m2082 (legacy record id)

Legacy Identifier etd-Kim-2843.pdf

Dmrecord 206502

Document Type Thesis

Rights Kim, Soo Young

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu