Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
GaAs nanowire optoelectronic and carbon nanotube electronic device applications
(USC Thesis Other)
GaAs nanowire optoelectronic and carbon nanotube electronic device applications
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
!
!
i!
!
GaAs Nanowire Optoelectronic and Carbon Nanotube
Electronic Device Applications
By
Sen Cong
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree of
DOCTOR OF PHILOSOPHY
(ELECTRICAL ENGINEERING)
May 2018
Copyright 2018 Sen Cong
!
!
ii!
Dedication
Dedicate to my dear parents Hongtao Cong and Yanping Deng, as well as my
lovely girlfriend Ruiwen Tang
!
!
iii!
Acknowledgement
Time flies fast. Five years and a half has passed since I first arrived at USC and began my amazing
journey to explore science as a PhD student. I am truly grateful to learn, grow and develop with so
much help from all the people who offered me help and courage along this journey. Without their
help, it is impossible for me to finish the work in my dissertation.
Firstly, I would deliver my deepest gratitude to my research advisor, Professor Chongwu Zhou,
for his kindness and guidance to help me even during the hardest time in my research. His rich
knowledge, high intelligence, and acute vision not only give me a new perspective of a particular
issue, not also help me gain the capability to discover the problems from the minute details and
the approach to solve them. Working with Professor Zhou in the ever-changing nanotechnology
field offers great opportunities to study the nanoscale details using the cutting edge technology.
Professor Zhou put tremendous effort in each single step, from design to analysis, to give me
advice and help me build my own ability to conduct it independently. Without his careful
cultivation, I cannot have the problem solving skills as I have acquired throughout the study.
Secondly, I would like to deliver my thanks to the Center for Energy Nanoscience (CEN) at USC,
a Department of Energy (DOE) Energy Frontier Research Center. The professors and colleagues
at CEN with various backgrounds and skillsets broaden my vision and provide me with different
approaches whenever a new problem arises. I would like to express my sincere gratitude to
Professor Daniel Dapkus, whose rich experience and deep knowledge in semiconductor research
often lead me to deeper understanding of the physics behind the observation. I would also like to
thank Professor Wei Wu, Professor Aiichiro Nakano, Professor Michelle Povinelli, Professor
!
!
iv!
Stephan Cronin for their helpful discussions during the CEN meetings and the collaboration for
various parts of the project.
Thirdly, I would like to thank all the current and former lab mates at USC nanolab. Dr. Maoqing
Yao, Dr Haitian Chen, and Dr. Jialu Zhang all gave me tremendous help when I first started a
project and guided me from the very beginning. They not only taught me by hand about different
steps of the experiments, but also answered my questions and encouraged me to think of and try
various methods to solve the problems. Dr. Bilu Liu, Dr. Jia Liu, Dr. Yu Cao, Dr. Gui Hui, Dr.
Xuan Cao, Pyojae Kim, Pattaramon Vuttipittayamongkol, Qingzhou Liu, Fanqi Wu, Zhen Li,
Hongyu Fu from the nanotube team gave me informative comments and suggestions throughout
my PhD study. I would also like to thank Dr. Gang Liu, Dr. Yuchi Che, Dr. Xiaoli Wang, Dr.
Luyao Zhang, Dr. Xin Fang, Dr. Noppadol Aroonyadet, Dr. Jiepeng Rong, Dr. Mingyuan Ge, Dr.
Yuqiang Ma, Dr. Liang Chen, Dr. Xin Fang, Dr. Ahmad Abbas, Anyi Zhang, Chenfei Shen,
Yihang Liu and Yilin Huang for being part of the USC nanolab family and making my PhD life
so memorable. I would like to thank other professors and researchers I have collaborated with, Dr.
Ningfeng Huang from Professor Michelle Povinelli’s group, Dr. Chunyung Chi, Dr. Yenting Lin
and Dr. Ashkan Seyedi from Professor Daniel Dapkus’ group, Dr. Shermin Arab, Bingya Hou and
Jihan Chen from Professor Stephan Cronin’s lab, Chunyang Shen from Professor Aiichiro
Nakano’s group; Dr. Liubing Huang and Dr. Mengyao Zhang from Professor Grace Lu’s group,
Dr. He Liu and Dr. Yuhan Yao from Professor Wei Wu’s group, Dr. Yufeng Wang from Professor
Eun Sok Kim’s group and Dr. Matthew Jurow from Mark Thompson’s group. They are amazing
colleagues to work with and helped me a lot. I would like to thank the financial support from King
Abdul-Aziz City for Science and Technology (KACST) via The Center of Excellence for
Nanotechnologies (CEGN).
!
!
v!
I would like to thank Professor Chongwu Zhou, Professor Daniel Dapkus, Professor Stephan
Cronin, Professor Wei Wu, Professor Armand Tanguay and Professor Aiichiro Nakano for serving
as my qualification and/or dissertation committee. I would like to thank all the staff at USC and
other institutions for offering me help and convenience during my PhD. USC cleanroom manager
Donghai Zhu helped solve the issues with the equipment quickly in the cleanroom. I also used
equipment at UCLA nanolab and UCLA ISNC, the technicians are really helpful, which is
essential for my research. I would like also to thank administrative staff members of EE-EP and
Viterbi School of Engineering, Jenny Lin, Eliza Aceves, Kim Reid, Diane Demetras, Tracy
Charles and Jennifer Gerson, whose help guaranteed me focus on research and have an enjoyable
life at USC.
The last but not least, I would like to express my thanks and gratitude to my father Hongtao Cong
and my mother Yanping Deng. I am indebted to their unconditional love and devotion throughout
the years. My father, who is a professor himself, always gives me valuable suggestions and
encouragement whenever I encounter problems. My mother, who cares about my life and help me
develop optimistic attitude toward life. I would like to thank my dear girlfriend, Ruiwen Tang, for
all the days of happiness and hardness she spent with me. With her encouragement, I always have
the faith for a better tomorrow. I sincere dedicate my dissertation to my family and all the friends
who cared about me and helped me.
!
!
vi!
Abstract
In this dissertation, I cover the study of two promising one dimensional materials, gallium arsenide
nanowires and carbon nanotubes, on their optoelectronic and electronic applications. Nanowires,
due to their small footprint and unique optical and electronic properties, are promising candidates
for next-generation high efficiency and cost effective solar cells. The relief of lattice mismatch
between different crystallographic materials from the nanowire and planar substrate enables
optimal multi-junction structure. Besides, the nanowire array can act as an anti-reflection layer,
which prevents excessive loss of the reflected light. Carbon nanotubes, on the other hand, is
another one dimensional materials which possesses transparency, high mobility, flexible and
ballistic transport properties. These desired features make carbon nanotube devices promising in
both macro-electronics and micro-electronics applications.
In my dissertation, for the GaAs nanowire part, I systematically demonstrated how my coworkers
and I progressively approached the goal of III-V nanowire-on-Si tandem solar cells. For the carbon
nanotube part, I demonstrated the integration of carbon nanotube thin film transistors (TFTs) with
polymer dispersed liquid crystal (PDLC) transparent display for the macro-electronics application.
Besides, I also studied the high on/off ratio, short channel, carbon nanotube field effect transistors
(CNTFETs) based on floating evaporative self aligned (FESA) tubes for micro-electronics
applications. The dissertation is comprised of 6 chapters. Chapters 1-3 talk about GaAs nanowires
for optoelectronic applications. Chapter 1 gives the introduction of nanowire solar cells and the
axial p-i-n junction GaAs nanowire array solar cells. This is the foundation of the III-V nanowire-
on-Si tandem solar cells. With junction depth and diameter optimizations, the efficiency can
achieve as high as 7.58%. Chapter 2 demonstrates tandem solar cells using GaAs nanowire on Si.
!
!
vii!
The observation of voltage addition and the improved efficiency of 11.4% confirm the concept of
III-V nanowire-on-Si tandem solar cells. Chapter 3 covers the study of the growth of lattice
mismatched GaAs nanowires on Si substrates. The facile five segment growth simplifies previous
seven segment growth profile. The study of twin formation has led to twin-free segments up to 80
nm, which is the highest reported so far for selective area growth (SAG). Chapters 4 and 5 talk
about carbon nanotubes. Chapter 4 gives an introduction to carbon nanotubes and the carbon
nanotube thin film transistors (CNTTFTs). One of the promising applications of CNTTFTs is the
driver circuits composed of CNTTFTs used to control flexible and transparent displays. Here, I
used the polymer dispersed liquid crystal (PDLC) flexible transparent technology, and successfully
integrated the CNTTFTs as the driving circuits for the active-matrix seven-segment PDLC display.
Chapter 5 talks about another important potential application of carbon nanotubes. Thanks to their
ballistic transport properties, carbon nanotubes have a great potential to replace Si in high speed,
low energy computing after Moore’s law. Using high semiconducting purity, highly aligned tubes
assembled by the floating evaporative self assembly (FESA) method, the CNTFETs have
demonstrated performance exceeding Si. On the other hand, problems still remain to fully realize
large scale integration. CNTFETs made from FESA-assembled tubes suffer from degradation of
on/off ratio as the channel length shrinks, especially at high drain-source voltage (V
ds
). I studied
the cause of the degradation and concluded the effect of oxygen and moisture are the main reasons
for the degradation. When measured in argon, minimizing the effects of oxygen and moisture, the
40-nm CNTFET can demonstrate on/off ratio as high as 1.29 × 10
4
at V
ds
= -0.5 V, a two order of
magnitude improvement compared to measurements in air. Besides, negative differential
resistance (NDR) phenomenon associated with gas desorption was also observed. Chapter 6
concludes the dissertation and gives a summary of the GaAs nanowire optoelectronic and carbon
!
!
viii!
nanotube electronic device applications. In the outlook section, I talked about the progress of the
GaAs nanowire solar cell and carbon nanotube electronics, and the future opportunities in these
fields.
! !
!
!
ix!
!
Table of contents
!
!
Dedication………………………………………………………………………………………....ii
Acknowledgements………………………………………………………………………………iii
List of tables………………………………………………………………………………............xi
List of figures……………………………………………………………………………….........xii
1.! GaAs Nanowire Array Solar Cells with Axial p-i-n Junctions………………………………..1
1.1!GaAs Nanowire Solar Cells……………………………………………………………1
1.2!Axial Junction versus Radial Junction…………………………………………………2
1.3!Device Structure and Optical Absorption Simulation………………………………….8
1.4!Effects of Junction Depth and Nanowire Diameter…………………………………...11
1.5!Summary……………………………………………………………………………..23
2.! Tandem Solar Cells using GaAs Nanowires on Si: Design, Fabrication, and Observation of
voltage addition………………………………………………………………………………26
2.1 Advantages of GaAs-Nanowire-on-Si Tandem Solar Cells………………………….26
2.2 The Components of Tandem Solar Cells using GaAs Nanowire on Si……………....27
2.3 The Integration of Tandem Solar Cells using GaAs Nanowire on Si and the Observation
of Voltage Addition………………………………………………………………………39
2.4 The Features of External Quantum Efficiency Measurement………………………..42
2.5 The Integration of the product of EQE and AM 1.5G photon density……………….44
2.6 Summary……………………………………………………………………………..45
3.! Facile Five-Step Heteroepitaxial Growth of GaAs Nanowires on Si Substrates and the Twin
Formation Mechanism……………………………………………………………………….50
3.1 The Advantages of III-V Semiconductor Nanowires on Si………………………….50
3.2 The History of Semiconductor Nanowires Grown on Si and the Undesired Twin
Formation inside Nanowires……………………………………………………………..51
3.3 Facile Five-Step Heteroepitaxial Growth…………………………………………….52
3.4 Twin Formation Growth Model……………………………………………………....58
3.5 Summary……………………………………………………………………………..67
!
!
x!
4.! Carbon Nanotube Macroelectronics for Active Matrix Polymer-Dispersed Liquid Crystal
Displays………………………………………………………………………………………73
4.1 The Transparent Displays and Their Driving Transistors…………………………….73
4.2 Monolithically Integrated Active Matrix PDLC Displays with CNT-TFT Based Control
Circuits…………………………………………………………………………………...74
4.3 Summary……………………………………………………………………………..87
5.! Improving Current On/Off Ratio and the Observation of Negative Differential Resistance for
40-nm-Channel Carbon Nanotube Array Transistors………………………………………...91
5.1 Promising Carbon Nanotube Array Transistors and Their Challenges………………91
5.2 Floating Evaporative Self-Assembly (FESA) Technique……………………………92
5.3 The Improvement of On/Off Ratio of 40-nm Carbon Nanotube Array Transistors
Measured in Argon……………………………………………………………………….94
5.4 Negative Differential Resistance (NDR) Effects……………………………………102
5.5 Initial Screening of Top Gated Aligned Carbon Nanotube Transistor Performance..106
5.6 Summary……………………………………………………………………………108
6.! Conclusion and Outlook…………………………………………………………………….111
6.1 Conclusion…………………………………………………………………………..111
6.2 Outlook……………………………………………………………………………...112
Bibliography……………………………………………………………………………………116
! !
!
!
xi!
List of tables
Table 1.1 .………………………………………………………………...……………………..17
Characteristics of devices with different diameter and junction depth.
!
!
xii!
List of figures
Figure 1.1 .……………………………………………..…………………………………………..2
Multi-junction solar cells. (a) Multi-junction solar cells consist of materials with different band-
gaps. Materials with larger band-gaps stack on top so different parts of the solar spectrum can be
preferentially absorbed at different depths from the surface. (b) A solar cell made from III-V
nanowire p-n junctions grown on Si substrate. (c) A three-junction solar cell made from
heterogeneous III-V nanowires grown on Si solar cells.
Figure 1.2!……………………………………………………………….………………………...4
Comparison between nanowire solar cells with axial junctions and radial junctions. (a) Schematic
diagram of axial (upper) and radial (lower) junctions in nanowires. (b) J
sc
, (c) V
oc
and (d) efficiency
of nanowire solar cells with axial junctions (black) and radial junctions (red) versus doping
concentration in n-type base. Surface recombination velocity is 3×10
5
cm/s. See text for details of
junction geometry and doping concentration. (e) 3D mapping of hole concentration in the dark and
in thermal equilibrium for the axial junction (left) and the radial junction (right). Band diagrams
under AM 1.5 short circuit conditions along the center of each nanowire (A-A’ and B-B’) are
shown on the left together with quasi Fermi levels of electrons (blue dashed line) and holes (red
dashed line). (f) Band diagram along C-C’ line in the radial junction in (e) together with quasi
Fermi levels of electrons and holes.
Figure 1.3 ……………………………………………………………………………….………...6
Efficiency as a function of n-type base doping for three radial junction structures: 30 nm p-shell/50
nm intrinsic/90 nm n-core (black), 50 nm p-shell/50 nm intrinsic/50 nm n-core (red), and 100 nm
p-shell /50 nm n-core (green).
!
!
xiii!
Figure 1.4 ………………………………………………………………………………………..10
Device structure and optical absorption simulation. (a) Schematic of a solar cell made from GaAs
nanowires with axial junctions. Carrier flow direction in solar cell operation under sunlight is
shown in the zoomed-in graph of the junction. (b) Maximum achievable J
sc
(mA/cm
2
) versus pitch
a and diameter/pitch ratio d/a. Nanowires are 3 μm tall and grown on a GaAs substrate. (c)
Maximum achievable J
sc
versus nanowire diameter for a fixed pitch of 600nm and height of 3 μm.
(d) Absorption spectra of nanowire arrays with 600 nm pitch and 100 nm (black), 250 nm (red)
and 350 nm (blue) diameter. (e) Carrier generation rate profile under AM 1.5G solar spectrum for
nanowires embedded in BCB and capped by ITO for diameters of 100 nm, 250 nm and 350 nm.
Figure 1.5 ………………………………………………………………………………………...12
Solar cell fabrication process and SEM images. (a) Fabrication steps of GaAs nanowire array solar
cells with axial junction: (i) electron beam lithography to form hole array in silicon nitride mask,
(ii) SAG of p-i-n GaAs nanowire using MOCVD, (iii) BCB infiltration, (iv) RIE to expose
nanowire tips, and (v) ITO deposition. (b) 30° tilted SEM image of as grown vertical GaAs
nanowire array on GaAs (111)B substrate. (c) SEM image after nanowires are embedded in BCB
and etched by RIE to expose short tips. (d) SEM image after coating of ITO film by sputtering. A
conformal dome-like cap is formed on the tips of nanowires.
Figure 1.6 ……………………………………………………………………………...…..…….14
Comparison between performance of nanowires with 100 nm diameter and 250 nm diameter. (a)
30° tilted SEM images of nanowires with diameter of 100 nm (sample A) and 250 nm (sample B).
(b) J-V curves of typical devices made from 100 nm and 250 nm thick nanowires in dark and under
AM 1.5 solar spectrum. (c) V
oc
(black) and FF (blue) of six devices from each batch. (d) J/V-V
curve of the device with 100 nm nanowires shown in (b) which shows linear region marked by the
blue line. Insets are schematics of a fully depleted thin wire and of a partially depleted thick wire
that still has conductive channel in the center.
!
!
xiv!
Figure 1.7 ……………………………………………………………………………………….46
Junction depth dependency and J-V under varied light intensity. (a) J-V curves of devices with
different junction depths between 100 nm and 600 nm under AM 1.5G solar spectrum. For the
device with 100 nm junction depth, diameter is increased to 320 nm. (b) Dark and AM 1.5G J-V
curves of the device with 100 nm junction depth and 320 nm diameter shown in (a) plotted in semi-
logarithmic scale. Ideality factor is found to be 2.34. (c) J-V curves of the device with 150 nm
junction depth and 220 nm diameter shown in (a) under different illuminating power between 50
μW and 10 mW from an 850 nm laser. V
oc
of 0.716 V is observed under the highest light intensity.
Insets shows the curve of V
oc
v.s. ln(I
sc
) with extracted ideality factor to be 1.72. (d) Current versus
pump power for the device shown in (c) at -2 V and 0 V bias.
Figure 1.8 ………………………………………………………………………………………...21
Spectrum response and Cathodoluminescence mapping. (a) Carrier generation rate profile under
monochromatic light of different wavelengths between 350 nm and 800 nm. All the light intensities
are 100 mW/cm
2
. Nanowires are 2.5 μm tall and 350 nm in diameter with 600 nm pitch. Nanowires
are surrounded by BCB and covered by ITO on top. (b) External quantum efficiency (left y-axis)
versus wavelength of the four devices shown in Fig. 5a. Dash-dot curve is the photon intensity of
AM 1.5G solar spectrum (right y-axis). J
sc
of each device are calculated by integrating the product
of EQE and photon intensity over the range of 350 nm to 880 nm and are indicated in the plot. (c)
SEM images and normalized cathodoluminescence mapping at wavelength of 865 nm for
nanowires with 400 nm deep junction (left) and 100 nm deep junction (right). Red curves are
relative CL intensity along the center of each nanowire.
Figure 2.1 ………………………………………………………………………………………..28
Schematic of GaAs nanowire-on-Si tandem solar cell. Top GaAs nanowire solar cell has p
+
emitter,
undoped segment, n-type base and n
+
root to form good connecting junction with p
+
Si. Bottom Si
solar cell has p
+
emitter, n-type base and n
+
back surface field. Nanowires are embedded in
transparent insulating polymer BCB. Top contact is ITO and back contact is Al.
!
!
xv!
Figure 2.2 ………………………………………………………………………………………..30
(a-c) Simulated absorption spectra of the Si bottom cell (red area), the GaAs nanowire top cell
(blue area), and the total structures (solid black line) for nanowires with nanowire heights (a) h =
500 nm, (b) h = 900 nm, and (c) h = 2000 nm. (d-f) Simulated corresponding J-V curves, which
show nearly matched currents between GaAs and Si for (e) the optimized height with h = 900 nm,
and mismatched currents for (d) shorter nanowires with h = 500 nm and (f) taller nanowires with
h = 2000 nm. (g) The simulated J
sc
in the top GaAs nanowire cell and the bottom silicon cell as
functions of the nanowire height. The short-circuit current of the tandem cell is limited by the
smaller of the two. (h) The limiting efficiency of the tandem cell as a function of the nanowire
height.
Figure 2.3 …………………………………………………………………………………...…...31
SEM images of GaAs nanowires. (a-c) SEM images of GaAs nanowires grown on Si (111)
substrate using SAG. (a) Top view, scale bar 5 μm. (b) 30° tilted view, scale bar 1 μm. (c)
Magnified top view, scale bar 500 nm.
Figure 2.4 ………………………………………………………………………………………..34
(a) Schematic of k-conservation transition in lightly doped and intrinsic semiconductor. (b)
Schematic of k-nonconservation transition in heavily doped semiconductor due to broken
translational symmetry.
Figure 2.5 ………………………………………………………………………………………..35
(a) Theoretical PL spectrum of the intrinsic GaAs nanowire using the direct band-to-band
transition mechanism (blue) and experimentally measured spectrum of undoped GaAs nanowire
array (red) at room temperature. (b) The PL spectrum of a single undoped GaAs nanowire at 4 K
and its Lorentzian peak fitting. Two major peaks at 828 nm and 834 nm are beyond the band edge
and identified as related to electron-to-carbon acceptor (e, A
0
) and carbon donor-to-carbon
acceptor (D
0
, A
0
) transitions.
Figure 2.6 ………………………………………………………………….…………………….37
Normalized room temperature PL spectra of as grown GaAs nanowire arrays on Si. Disilane flow
rate is varied between 0 and 1 sccm during the growth. Dotted lines are fitted curves. Curves are
intentionally offset for clear view.
!
!
xvi!
Figure 2.7 ………………………………………………………………………………………...37
(a) Fermi level E
f
(black) and band tail depth σ
E
(red) extracted from photoluminescence study. (b)
Carrier density calculated based on the extracted values shown in (a).
Figure 2.8 ……………………………………………………………………………………..….39
n
+
-GaAs/p
+
-Si heterojunction characterization. I-V curves of three doping levels, 0.2, 0.5 and 1
sccm, were measured. The former two shows barrier while the latter does not. Inset shows the
schematic of the device in measurement.
Figure 2.9 ………………………………………………………………………………………...42
J-V curves and EQE (a) Open circuit voltage addition of the GaAs nanowire-on-Si tandem solar
cell. J-V curves of the GaAs nanowire-on-Si tandem solar cell (black), the stand-alone GaAs
nanowire cell (blue) and the stand-alone Si cell (red). The arrows indicate the open circuit voltages
for each cell, showing an addition of V
oc
of the tandem cell (0.956 V) from separate stand-alone
GaAs nanowire (0.518 V) and stand-alone Si (0.547 V) cells. (b) Experimentally measured EQE
of the tandem cell.
Figure 2.10 ……………………………………………………………………………………...44
The simulated absorption spectra. (a) Simulated absorption spectra of GaAs nanowire subcell (blue
circles), Si subcell (red triangles), and the tandem cell (solid green), compared to the experimental
EQE (black squares) of tandem cell (b) Simulated absorption spectra of GaAs nanowire subcell
with different nanowire diameters. d = 280 nm (orange), d = 300 nm (green) and d = 320 nm
(purple).
Figure 3.1………………………………………………………………………………………...53
(a) Temperature profile of GaAs nanowire growth on Si (111). Red and blue blocks indicate the
time AsH
3
and TMG are supplied, respectively. (b) 30° tilted and (c) top view SEM images of
uniform GaAs nanowire arrays grown on Si. (d) Schematic diagram of crystal lattice at GaAs/Si
interface, viewed from [1-10] orientation. (e) HRTEM image taken at the GaAs/Si interface.
Arrows indicate sites with misfit dislocation.
!
!
xvii!
Figure 3.2………………………………………………………………………………………...56
30° tilted SEM images at the mask openings on (a) GaAs (111)B and (b) Si (111) substrates after
2 minutes growth at 790 ºC. (c) Top view SEM images of nuclei after 2 minutes growth under
different temperatures and AsH
3
partial pressures, scale bar 200 nm. TMG partial pressure was
kept constant at 7.56×10
-7
atm. (d) Schematic of relevant low-index planes viewed from <111>B
direction. (e) Projection of relevant low-index planes on (1-10) plane in GaAs zincblende lattice.
Figure 3.3 ………………………………………………………………………………………..61
(a) 30º tilted SEM images of nanowires grown at 850 ºC, the initial part was grown at 760 ºC to
help nucleate. White circles indicate nanowires with completely pinched-off tetrahedron tip, while
red circles indicate nanowires with triangular-shaped thin mesa surrounded by three tilted {-1-10}
facets (b) SEM of images of nanowire tips with different morphologies, scale bar 200 nm. Images
are organized in a sequence according to the twin formation model in (c). (c) Schematic of twin
formation process during nanowire growth. (d) TEM image of a region consists of even number
of twins. The crystals at two sides of transitional region share the same atomic registry. (e) SEM
image of a nanowire tip. After even number of twins, the wire pinches off. (f) TEM image of a
region consists of odd number of twins. The crystal at two sides of transitional region can be
considered as rotated by 180 º compared to each other. (g) SEM image of a nanowire tip with odd
number of twins.
Figure 3.4 …………………………………………………………………………………..……63
(a)-(c) TEM images of nanowire grown at different temperatures: (a) 760 ºC, (b) 790 ºC and (c)
850 ºC. The initial segment in (c) was grown at 760 ºC in order to nucleate. (d)-(f) TEM images
of nanowires grown at 850 ºC with different diameters: (d) 95 nm, (e) 180 nm and (f) 280 nm. (g)
Histogram showing length distribution of twin free segments of nanowires grown at 850 ºC with
different diameters.
Figure 3.5 ………………………………………………………………………………………..66
Probability distribution of the twin-free segment length in GaAs nanowires of the diameter 95, 180
and 280 nm at temperature (a) 850 °C and (b) 760 °C.
!
!
xviii!
Figure 4.1 ………………………………………………………………………………….…….75
(a) Schematic of the proposed seven-segment CNT-TFT driven PDLC display. (b) The cross-
sectional view of the CNT-TFT driven PDLC display, showing the structural details. (c) The side
view of the schematic of individual back gated CNT-TFTs. (d) SEM images of the CNT network.
Scale bar 2 µm. Inset, scale bar 500 nm.
Figure 4.2..……………………………………………………………………………………….77
(a) Transparent state and (b) Opaque state of the PDLC pixel. When electric field is applied to the
electrodes, liquid crystals align with the electric field permitting the incoming light to pass through
in (a). When no electric field is applied, liquid crystals are randomly aligned and incoming light is
scattered, giving opaque state in (b).
Figure 4.3…………………………………………………………………………………..…….79
PDLC transmission properties. (a) The raw and normalized transmittance of the PDLC cell at 565
nm wavelength and 400 Hz frequency versus AC voltage amplitude. (b) Transmittance of
transparent (9 V) and opaque (0 V) states versus wavelength. (c) The transmittance of PDLC versus
AC voltage frequency at 9 V amplitude and 565 nm wavelength (d) Transparent (left) and opaque
(right) states in front of printed text. In the transparent state, one can clearly see the printed text
through the PDLC cell, while in the opaque state, the text can hardly be seen. Scale bars 1 cm.
Figure 4.4 ………………………………………………………………………………….……..81
Transfer characteristics of a common back gate CNT-TFT transistor with 90 nm SiO
2
gate
dielectric and L = 20 µm, W = 200 µm (a) before and (b) after parylene passivation.
Figure 4.5.…………………………………………………………………………………….….82
Transfer characteristics of a typical individual back gate CNT-TFT on glass substrate with L = 20
µm, W = 200 µm (a) before and (b) after parylene passivation. The behaviors before and after
parylene passivation are similar.
!
!
xix!
Figure 4.6…………………………………………………………………………………......….83
(a) Transfer (I
D
-V
G
) characteristics (red) and transconductance g
m
-V
GS
characteristics (blue) of a
typical CNT-TFT (L = 20 μm, W = 2400 μm) with V
DS
= -1 V. (b) Output (I
DS
-V
DS
) characteristics
of the same device with V
G
varying from -10 to 10V in 4 V steps.
Figure 4.7………………………………………………………………………………….….….84
(a)
Schematic of CNT-TFT driven PDLC test cell. The voltage across the PDLC test cell (V
PDLC
)
is indicated. (b) The waveforms of V
PDLC
under various V
SCAN
= -10 V, -5 V, 0 V, 5 V and 10 V
with the data electrode fixed at 400 Hz sine wave and the amplitude V
DATA
= 9 V. (c) Optical
photographs showing the background blue color without PDLC test cell and the color through
PDLC test cell with the driving CNT-TFT under different V
SCAN
. The CNT-TFT can fully turn on
and turn off the transistor. Scale bar 1 mm.
Figure 4.8………………………………………………………………………….……………..86
(a) The optical image of the CNT-TFT back panel before integration. The data and scan signals
given to each electrode are also indicated. Inset: the SEM image of the CNT-TFT. The dotted
region indicates where the CNT film exists. The electrode which will be connected to PDLC pixel
with the voltage V
PDLC
in the later integrated circuit is also shown. (b) The optical image of the
seven segment display after PDLC integration. The CNT-TFT back panel, the glass with ITO top
electrode, and the gap where PDLC is injected are also indicated. (c) Histogram of the mobilities
of each CNT-TFT driving the seven segment display. (d) Histogram of the on/off of each CNT-
TFT driving the seven segment display. (e-g) Optical images of the seven segment display against
(e) blue, (f) green, and (g) red backgrounds, displaying digits “1”, “2”, and “3”, respectively.
Figure 5.1…………………………………………………….………………………………..…94
Carbon nanotube field effect transistors on FESA aligned tubes (a) Optical image of a 4-inch wafer
with FESA aligned tubes inside the blue boundary. The direction of alignment is indicated by the
red arrows. (b-d) The SEM images of the FESA aligned tubes of a larger area (b), aligned region
(c) and transient region (d).
!
!
xx!
Figure 5.2………………………………………………………………………………………...95
(a) Schematic diagram of the short-channel, back-gated CNTFETs. (b-c) The SEM images of the
short channel devices channel area of L
ch
= 100 nm (b) and L
ch
= 40 nm (c).
Figure 5.3…………………………………………………………………….…………………..96
Electrical characteristics of the short-channel CNTFETs. (a,c) I
ds
– V
gs
curves in both linear (green
and black) and log (red and blue) scales at V
ds
= -0.1 V (black and red) and V
ds
= -0.5 V (green and
blue), (a) L
ch
= 100 nm and (c) L
ch
= 65 nm. (b,d) I
ds
– V
ds
curves with V
gs
= -8 V to 0 V in steps of
2 V, (b) L
ch
= 100 nm and (d) L
ch
= 65 nm. (e,f) On state current density versus channel length
under different V
ds
, (e) V
ds
= -0.1 V and (f) V
ds
= -0.5 V.
Figure 5.4………………………………………………………………………………………...98
The electrical characteristics for the 40 nm CNTFETs in air and in argon. (a) I
ds
– V
gs
curves
measured in air for a typical 40-nm device in both linear (green and black) and log (red and blue)
scales at V
ds
= -0.1 V (black and red) and V
ds
= -0.5 V (green and blue). (b) I
ds
– V
gs
curves measured
in air for the 40-nm device with V
gs
= -6 V to 6 V in steps of 2 V and I
ds
= 0 V to -0.5 V. (c) The
same measurements as in (A) for the same device in argon. (d) The same measurements as in (b)
for the same device in argon. (e) The on/off ratio versus V
ds
in air (blue) and in argon (red). (f) The
on/off ratio of the device measured in argon compared with the highest on/off ratio in previous
publication measured in air.
Figure 5.5……………………………………………………………………………………….100
Bonded Carbon Nanotube Device on a Chip Carrier.
Figure 5.6……………………………………………………………………………………….103
The negative differential resistance phenomenon of the device. (a) I
ds
– V
gs
curves measured in air
for the 40-nm device with V
gs
= -6 V to 6 V in steps of 2 V and I
ds
= 0 V to -0.8 V. (b) The
continuous double sweep I
ds
– V
ds
curves from V
ds
= 0 V to 0.8 V under V
gs
= -6 V in air for 6 times.
(c, d) I
ds
– V
gs
curves at V
ds
= -0.1 V (c) and V
ds
= -0.5 V (d) before and after (measured immediately)
the measurements in (b).
!
!
xxi!
Figure 5.7……………………………………………………………………………………….105
The desorption and adsorption of oxygen and moisture. (a) The double sweep I
ds
– V
ds
curves from
V
ds
= -0.8 V to 0.8 V under V
gs
= -6 V in air for the L
ch
= 40 nm device. The forward sweep exhibits
negative differential resistance behaviour. (b) The same measurement in air as in (a) except that
the sweeping direction is from V
ds
= 0.8 V to -0.8 V. (c) The double sweep I
ds
– V
ds
curves from V
ds
= -0.8 V to 0.8 V under V
gs
= -6 V in argon for the L
ch
= 40 nm device. (d) The same double sweep
in argon as in (c) except that the sweeping direction is from V
ds
= 0.8 V to -0.8 V.
Figure 5.8……………………………………………………………………………………….107
Initial screening of the top gated aligned carbon nanotube performance. (a) The SEM image of the
channel area of CNTFET with W = 4 µm and L
ch
= 850 nm. (b) I
ds
-V
gs
curve of as made back gated
device. Inset: structure of back gated device. (c,d,e) I
ds
-V
gs
curve of the device measured after (c)
ALD deposition, (d) 30 minutes 400 degrees annealing in argon, and (e) top gate Ti/Au deposition.
Inset: structure of back gated device (c,d) after ALD deposition and (e) top gate Ti/Au deposition.
(f) The I
ds
-V
ds
curves of the top gated device with V
gs
= 0 to -5 V with -1 V step.
! 1!
Chapter 1 GaAs Nanowire Solar Cells with Axial p-i-n Junctions
1.1 GaAs Nanowire Solar Cells
In recent years, semiconductor nanowires, due to their 1D structure, combined with superior
electronic and optical properties, have been a topic of intense research and development for next
generation photovoltaics.
1-21
One approach to surpass the Shockley-Queisser efficiency limit is to
use multi-junction solar cells containing several p-n junctions in series.
22-26
Each junction is
designed to absorb a specific wavelength range of sun light, reducing thermalization losses, and
thereby increasing efficiency (Fig. 1.1a). Traditional, monolithically integrated multi-junction
solar cells consist of sequentially stacked thin films. The lattice constants of the materials used are
matched to allow high quality epitaxial growth. GaAs and Germanium are currently the most
widely used substrates since there are lattice-matched materials with larger bandgaps suitable for
forming a current matched set of junctions. These substrates contribute a large portion of the total
material cost and prevent large scale terrestrial implementation. The small footprint of nanowires
allows them to accommodate lattice-mismatch induced strain through elastic relaxation at the
edges.
27-31
This capability offers great freedom in choosing growth substrates, which are otherwise
impossible in the planar thin film case. For instance, dislocation-free InP or GaAs nanowires can
be grown on a silicon substrates,
32-34
significantly reducing substrate cost and enabling use of the
existing silicon industry infrastructure (Fig. 1.1b). Meanwhile, the freedom to use multiple,
stacked, non-latttice-matched materials within each nanowire allows the construction of multi-
junction cells with optimal band gap combinations (Fig. 1.1c). Furthermore, both simulation and
experiments have shown that nanowires are inherently excellent light absorbers due to light
scattering and the effect of resonant modes, even without additional anti-reflection coatings.
11, 35-42
! 2!
Figure 1.1 Multi-junction solar cells. (a) Multi-junction solar cells consist of materials with different band-
gaps. Materials with larger band-gaps stack on top so different parts of the solar spectrum can be
preferentially absorbed at different depths from the surface. (b) A solar cell made from III-V nanowire p-n
junctions grown on Si substrate. (c) A three-junction solar cell made from heterogeneous III-V nanowires
grown on Si solar cells.
Among various nanowires, those made from III-V compound semiconductor materials are
considered as some of the most promising platforms for photovoltaics due to high absorption,
direct band gap, superior carrier mobility, and well-developed synthesis techniques. Significant
progress with InP nanowire solar cells has recently been reported especially with axial p-i-n
junctions.
18, 21, 43
In contrast, GaAs is another important material with suitable bandgap and has the
advantage that gallium is more abundant than indium. However, progress in the development of
GaAs nanowire solar cells has been relatively slow and their potential has not been fully exploited.
1.2 Axial Junction verses Radial Junction
In terms of electrical properties, the design of the p-n junction governs the overall performance of
a solar cell. We thus carefully examined the operational conditions of nanowire solar cells with
! 3!
radial and axial junctions. Fig. 1.2a shows the schematics of these two junctions in the nanowires.
A single axial junction (top, junction along the length of the nanowire) is of considerable technical
importance, as it is an important building block for multi-junction solar cells with three or more
junctions as shown in Fig. 1.1c. Nanowires with junctions stacked in the axial direction provide
an intuitive analogue to the traditional thin film multi-junction solar cells where the incident light
can pass from materials with larger bandgap to those with lower bandgap sequentially. Radial
junctions (bottom, junction along the diameter of the nanowire) have their own advantages that
short collection lengths for excited carriers in a direction normal to the light absorption could
facilitate the efficient collection of photogenerated carriers with low minority-carrier diffusion
lengths.
44
Previous work on GaAs nanowire solar cells focused on radial p-n or p-i-n junctions. In
contrast, solar cells based on axial p-n or p-i-n junctions have not been fully studied, although it is
intriguing to compare their performance to that of radial junction devices and study the underlying
science. Our recent theoretical study indicates that axial junctions are more sensitive to the
presence of surface states, but axial junctions provide higher V
oc
and more flexibility in the design
of junction structure such as the thickness of base and emitter layers than radial junctions.
36
! 4!
!
Figure 1.2 Comparison between nanowire solar cells with axial junctions and radial junctions. (a)
Schematic diagram of axial (upper) and radial (lower) junctions in nanowires. (b) J
sc
, (c) V
oc
and (d)
efficiency of nanowire solar cells with axial junctions (black) and radial junctions (red) versus doping
concentration in n-type base. Surface recombination velocity is 3×10
5
cm/s. See text for details of junction
geometry and doping concentration. (e) 3D mapping of hole concentration in the dark and in thermal
equilibrium for the axial junction (left) and the radial junction (right). Band diagrams under AM 1.5G short
circuit conditions along the center of each nanowire (A-A’ and B-B’) are shown on the left together with
quasi Fermi levels of electrons (blue dashed line) and holes (red dashed line). (f) Band diagram along C-C’
line in the radial junction in (e) together with quasi Fermi levels of electrons and holes.
!
! 5!
To compare the two structures, we numerically solve the current density-voltage (J-V) response
using the finite-difference time-domain method
36
of the nanowire solar cells with 600 nm pitch,
250 nm diameter and 2.5 μm height. We use Synopsys Sentaurus to solve the drift-diffusion
equations for carrier transport within the nanowires. The nanowires are subject to AM 1.5G solar
irradiation. The position-dependent carrier generation rate is determined from the optical
absorption simulations above and is shown in Fig. 1.2e. For the axial junction, the p-type segment
is on top, with a thickness of 100 nm, and for the radial junction, the p-type region is on the outside,
with a thickness of 30 nm. Both intrinsic regions are 50 nm thick. The thicknesses of the p-type
regions were chosen based on our previous study.
36
Thicker shell lead to even worse performance
as shown in Fig. 1.3. We assume that the doping dependent mobilities for electrons and holes are
the same as in the bulk, with a Shockley-Read-Hall (SRH) recombination lifetime of 1 ns. Both
donor-like and acceptor-like surface state densities are fixed to be 1.5×10
12
cm
-2
, corresponding to
a surface recombination velocity (SRV) of 30,000 cm/s. We fix the doping concentration of the p-
type emitters to be 10
18
cm
-3
for the purpose of forming a low resistance ohmic contact, and vary
the n-type base doping in both the axial and radial junctions. 10 nm thick minority carrier reflectors
with doping concentrations of 10
19
cm
-3
are placed right below the top contact and above the bottom
contact to reduce recombination loss. In the real devices, heavily doped n+ substrate (2-3×10
18
cm
-
3
) serves as bottom minority reflector and the p-doping near the tip is increased to serve as top
minority reflector.
! 6!
Figure 1.3 Efficiency as a function of n-type base doping for three radial junction structures: 30 nm p-
shell/50 nm intrinsic/90 nm n-core (black), 50 nm p-shell/50 nm intrinsic/50 nm n-core (red), and 100 nm
p-shell /50 nm n-core (green).
In Fig. 1.2b, 1.2c and 1.2d we show the J
sc
, V
oc
and PCE as functions of n-type base doping
concentration. For the axial junction, increasing the doping concentration reduces J
sc
. Nanowires
usually exhibit shorter minority diffusion length than bulk, so the built-in electric field in the
junction depletion region contributes significantly to the total collection of carriers. With
increasing doping in the n-type base of axial junction, both the junction depletion region width,
which defines the carrier drift zone, and the mobility which determines carrier diffusion, are
reduced. The open circuit voltage however increases with increasing doping concentration due to
the larger built-in potential. The overall efficiency does not change much with the increasing
doping concentration. For the radial junction, the short circuit current is very high when the base
doping is higher than 10
17
cm
-3
. However, it drops sharply when the doping is smaller than 10
17
cm
-3
. The open circuit voltage and the efficiency of the radial junction have similar trends.
! 7!
We can gain insight into the difference between the J-V characteristics of axial and radial junctions
in the low doping region from Fig. 1.2e and 1.2f. Fig. 1.2e shows the band diagrams across the
center of the wires (A-A’ for the axial junction and B-B’ for the radial junction) under the AM
1.5G solar spectrum and short circuit condition and the corresponding hole concentration mapping
in the dark under thermal equilibrium conditions with a base doping of 10
16
cm
-3
. In the axial
junction, the n-type region is fully depleted by surface states under thermal equilibrium, as one can
tell from the fact that the hole concentration is close to the intrinsic carrier concentration of GaAs
(2.1×10
6
cm
-3
, 300K). However, under illumination, the nanowire behaves as a normal p-i-n
junction solar cell (see band diagram) because the surface states are largely filled by light-
generated carriers. There is a gradual band bending along the axial direction near the junction at
the heights between 2 µm and 2.4 µm (intrinsic segment is between 2.35 and 2.4 μm). This causes
an electric field pointing upwards, which helps the extraction of the light-generated carriers near
the tip of the nanowire. In a radial junction, when the core region of the nanowire is lightly doped,
the p-type shell will cause an inversion of the core carrier type to also be p-type. The structure is
hence equivalent to a p-type nanowire with a very thin n-type emitter at the bottom as can be seen
in the band diagram along B-B’ in Fig. 1.2e. This can further be confirmed by checking the cross-
sectional (C-C’) band diagrams in Fig. 1.2f, which shows the band is almost flat. The minority
carriers thus need to diffuse through the entire length of nanowire to reach the bottom and get
extracted, under which circumstance an extremely low J
sc
is expected.
In principle the heavily doped radial junction device can outperform the axial junction due to its
excellent carrier collection efficiency and its high tolerance to surface effects. However, high
doping in the base is usually undesirable because it reduces mobility and diffusion length. State-
of-the-art thin film GaAs solar cells normally use a base doping concentration of the order of
! 8!
magnitude of 10
17
cm
-3
. The possibility of using lower doping in the axial junction provides us with
more optimization space and a more robust design. Moreover, as mentioned earlier, the axial
junction is an indispensable step toward a multi-junction nanowire solar cell device with three or
more junctions. We move forward with systematic experimental study of the GaAs axial junction
nanowire solar cells.
1.3 Device Structure and Optical Absorption Simulation
Fig. 1.4a is a schematic of a solar cell made from a vertically aligned GaAs nanowire array grown
on a GaAs substrate. Nanowires are embedded in transparent insulating polymer BCB, for
mechanical support and covered by a transparent conductive indium tin oxide (ITO) front contact
to let sunlight pass through. Incident light generates electrons and holes which flow toward the n-
type and p-type region, respectively. The absorption properties of periodic nanowire arrays
strongly depend on the array structure. In order to determine the optimal structure for absorption,
we first carry out full-vectorial electromagnetic simulations to calculate the absorption of nanowire
arrays with different diameters and pitch. We calculate the maximum achievable short circuit
current from the optical absorption by assuming unity external quantum efficiency. Fig. 1.4b
shows J
sc
as a function of pitch, a, and diameter-to-pitch ratio, d/a, for a nanowire height of 3μm.
Calculations were performed using the AM 1.5G solar spectrum assuming an infinitely thick GaAs
substrate. Two local maxima are observed on this map. One is at a = 300 nm and d/a = 0.55, and
the other one is around a = 650 nm and d/a = 0.60. We focus on the maximum point associated
with pitch of 600 nm due to the fact that, without sacrificing too much absorption (<1 mA/cm
2
), a
larger pitch eases fabrication requirements. Furthermore, we will show later in the paper that the
small diameter of the other maximum point is unfavorable for the electrical properties. In Fig. 1.4c,
we plot J
sc
versus nanowire diameter for fixed pitch of 600 nm and height of 3 μm. We can achieve
! 9!
J
sc
as high as 28 mA/cm
2
for a nanowire array with diameter of 350 nm, which is close to the
theoretical limit of a single junction GaAs solar cell. Comparing the absorption spectra of arrays
with diameters of 100 nm, 250 nm and 350 nm (600 nm pitch, 3 μm height) shown in Fig. 1.4d,
the 350 nm one enhances the absorption by filling in the low absorption dips between 400 nm and
550 nm and between 700 nm and 900 nm that appear in the spectra of the 100 nm and 250 nm
diameter wire arrays. We also plot the spatial carrier generation rate distribution within the realistic
nanowire array structure (with the BCB polymer and ITO caps) for nanowires with height of 2.5
μm and diameters of 100, 250 and 350 nm, respectively (Fig. 1.4e). The maximum achievable
short circuit currents are also indicated in each profile.
! 10!
Figure 1.4 Device structure and optical absorption simulation. (a) Schematic of a solar cell made from
GaAs nanowires with axial junctions. Carrier flow direction in solar cell operation under sunlight is shown
in the zoomed-in graph of the junction. (b) Maximum achievable J
sc
(mA/cm
2
) versus pitch a and
diameter/pitch ratio d/a. Nanowires are 3 μm tall and grown on a GaAs substrate. (c) Maximum achievable
J
sc
versus nanowire diameter for a fixed pitch of 600nm and height of 3 μm. (d) Absorption spectra of
nanowire arrays with 600 nm pitch and 100 nm (black), 250 nm (red) and 350 nm (blue) diameter. (e)
Carrier generation rate profile under AM 1.5G solar spectrum for nanowires embedded in BCB and capped
by ITO for diameters of 100 nm, 250 nm and 350 nm.
!
! 11!
1.4 Effects of Junction Depth and Nanowire Diameter
To experimentally test the performance predicted by simulation we fabricated and measured solar
cells with 1 mm × 1 mm area GaAs nanowire arrays. The schematic diagram of device structure
has been shown in Fig. 1.4a and the fabrication process is shown in Fig. 1.5a. The n-type,
intentionally undoped and p-type segments are grown sequentially on a n+ (111)B substrate using
SAG in MOCVD. The growth detail has been mentioned earlier in a separate paper.
38
This growth
method does not require metal catalyst as used in vapor-liquid-solid (VLS) method which is
believed to incorporate along the growth and cause deep level traps.
45
The vapor phase epitaxial
nature of SAG can also avoid reservoir effects often encountered in VLS and achieve an abrupt
junction interface. The background doping for the i-region is not known at this point, but is
believed to be much lower than the p and n region, and may be in the range of 10
14
-10
15
cm
-3
based
on literature.
46
Nanowires are separated by 600 nm from center to center and are about 2.5 μm tall
and 320 nm in diameter, unless otherwise stated, in order to achieve nearly optimized absorption
of AM 1.5G solar irradiation. Although the nanowires only cover less than 20% of the total area,
they could potentially absorb close to 90% of the incident sun light. It’s been widely studied that
resonant modes would allow light absorption in nanowires to significantly exceed ray optics limit
given just a fraction of the material consumed in a bulk device.
11, 21
Fig. 1.5b is scanning electron
microscopy (SEM) image taken at 30° tilted angle. Nanowires distribute uniformly in the electron
beam lithography (EBL) defined template and exhibit six-fold symmetric cross section consisting
of sidewalls parallel to {110} family planes which indicate high crystal quality.
! 12!
Figure 1.5 Solar cell fabrication process and SEM images. (a) Fabrication steps of GaAs nanowire array
solar cells with axial junction: (i) electron beam lithography to form hole array in silicon nitride mask, (ii)
SAG of p-i-n GaAs nanowire using MOCVD, (iii) BCB infiltration, (iv) RIE to expose nanowire tips, and
(v) ITO deposition. (b) 30° tilted SEM image of as grown vertical GaAs nanowire array on GaAs (111)B
substrate. (c) SEM image after nanowires are embedded in BCB and etched by RIE to expose short tips.
(d) SEM image after coating of ITO film by sputtering. A conformal dome-like cap is formed on the tips
of nanowires.
After growth, nanowire arrays are planarized and etched to only expose the tip of p-type emitter.
Eventually ITO is deposited as transparent front contact (Fig. 1.5a) while AuGe is alloyed to the
backside of the substrate to form ohmic contact. The sheet resistances of planar ITO film is about
10 Ohm/ measured by four-probe method. The highly conductive ITO film allows us to achieve
low series resistance without forming additional metal fingers on top of ITO as will be shown later.
Sputtering provides a conformal layer of ITO on the non-planar top surface and forms a dome
! 13!
shape cladding over the nanowire tip which helps to concentrate the incident light near the junction
region as has been seen in Fig. 1.2e and also pointed out by Mariani et al.
14
Fig. 1.5c shows the
30° tilted SEM image of nanowire array after BCB infiltration and reactive ion etching (RIE).
About 100 nm tip is exposed for contacting. Fig. 1.5d shows ITO film conformally wrapping the
nanowire tips to ensure good conductivity at the front electrode. Basically the high uniformity is
inherited from EBL-made pattern by the subsequent process and demonstrates the superior
capability of SAG in controlling the morphology and location of nanowires.
Because of high density of surface states, carrier transport in GaAs nanowires is known to be
significantly affected by surface-to-volume ratio. For nanowires with surface Fermi level pinned
at mid gap, Chang et al.
47
pointed out majority carriers would be nearly depleted if wire diameter
is less than 100 nm, resulting in a less conductive channel which is also observed in our simulation.
Furthermore, a considerable portion of minority carriers would be captured by surface states and
subsequently annihilate through recombination for a thin wire. Thus one expects significant loss
of J
sc
when nanowire diameter gets extremely small. Fig. 1.6a shows SEM images of the 30° tilted
nanowire arrays used for two sets of devices. Both samples are around 1.5 μm tall with 300 nm
deep junction. Sample A consists of nanowires about 100 nm in diameter while the wires in sample
B are close to 250 nm in diameter. J-V characteristics are measured under dark condition as well
as under AM 1.5G solar spectrum at 1 sun illumination intensity (100 mW/cm
2
) using a solar
simulator (Photo Emission Tech). Distinct J-V curves can be seen in Fig. 1.6b for representative
devices made from these two batches. Remarkably improved performance is obtained for
nanowires of 250 nm than nanowires of 100 nm diameter due to both larger J
sc
and higher V
oc
.
Another important parameter reflecting how closely a device resembles an ideal diode is fill factor
(FF) defined by the ratio of maximum power output to the product of I
sc
and V
oc
. Fig. 1.6c shows
! 14!
the V
oc
and FF of all the samples from batch A and B and each batch contains six individual
devices. From this figure, the FF of sample B is much higher than sample A on average following
the same trend as V
oc
. In larger diameter nanowires, fewer carriers will recombine at the surface
leading to higher J
sc
. On the other hand, if the entire n-type base is depleted as a result of a small
diameter, then the built-in potential across the p-i-n junction will definitely be smaller than an ideal
junction because the Fermi-level in the n-type region is deeper into the bandgap than it would be
in an ideal case. Moreover, plots of current density divided by voltage (J/V) versus V for sample
A shows, after the diode turns on, J/V is proportional to V indicating strong quadratic dependence
on voltage for dark current as shown in Fig. 1.6d. This phenomenon can be attributed to space
charge limited (SCL) transport in a fully depleted crystal which was recently modeled and
discussed for gallium nitride nanowires
48
and indium arsenide nanowires
49
.
Figure 1.6 Comparison between performance of nanowires with 100 nm diameter and 250 nm diameter.
(a) 30° tilted SEM images of nanowires with diameter of 100 nm (sample A) and 250 nm (sample B). (b)
J-V curves of typical devices made from 100 nm and 250 nm thick nanowires in dark and under AM 1.5G
solar spectrum. (c) V
oc
(black) and FF (blue) of six devices from each batch. (d) J/V-V curve of the device
with 100 nm nanowires shown in (b) which shows linear region marked by the blue line. Insets are
! 15!
schematics of a fully depleted thin wire and of a partially depleted thick wire that still has conductive
channel in the center.
In addition to nanowire diameter we studied another important design parameter which is the
length of the p-emitter region, also mentioned as junction depth (D
jc
) interchangeably in this paper.
As has been shown in Fig. 1.2e, upon illumination of solar spectrum the carrier generation hot spot
is located very close to the nanowire tip due to the high absorption coefficient of GaAs over most
of the solar spectrum as well as the concentrating effect from the ITO cap. On the other hand, to
achieve low resistance ohmic contact between ITO and GaAs, the p-doping near the top is usually
sufficiently high leading to relatively shorter minority carrier lifetime. Thus, we want to minimize
carrier loss in the p-region while keeping a good contact to ITO. Experimentally we varied the
length of p-segment between 600 nm and 100 nm by adjusting the growth time. As shown in Fig.
1.7a, a steady increase in short circuit current can be observed when we decrease the junction
depth. Typical devices show J
sc
of 1.07, 8.52 and 11.58mA/cm
2
for devices with 220 nm diameter
and nominally 600 nm (sample C), 300 nm (sample D) and 150 nm (sample E) deep junctions
respectively. Shallower junctions also feature higher V
oc
in general. The batch with 600 nm deep
junction (sample C) exhibits relatively low V
oc
around 350 mV and, in contrast, devices in the
batch with 150 nm deep junction (sample E) have V
oc
approaching 650 mV. In batch F we further
reduce junction depth to 100 nm and also increase diameter to around 320 nm. We observe a
tremendous increase of J
sc
to above 20 mA/cm
2
with the highest being 23.28 mA/cm
2
. We believe
larger diameter is more important for this significant increase in J
sc
because devices with 320 nm
diameter and 300 nm junction depth show similar level of J
sc
. Fig. 1.7b shows the dark and 1 Sun
J-V curve of the best device from batch F plotted as a semi-logarithmic plot. The dark
characteristic shows a good rectifying behavior with an on-off ratio of 1.89×10
5
at ±1 V. The dark
! 16!
current at -1 V is only about 100 nA for a 1 mm
2
area also indicating a good junction. The small
leakage current could be attributed to the small total junction area possessed by axial junction
geometry. Vapor phase growth of the junction should, in principle, produce a sharp junction with
low leakage current. The ideality factor extracted from the intermediate forward bias regime is
around 2.34. We believe, the greater-than-unity ideality factor is due to the existence of
recombination current through surface states and space charge region in the undoped part. Further
optimization of the length of the undoped segment is ongoing. This particular device shows J
sc
of
21.08 mA/cm
2
and V
oc
of 0.565 V. With a fill factor of 0.6365 the overall PCE is 7.58%. In previous
publication,
36
we have simulated and experimentally studied the role of substrate on photon
absorption. Even though the substrate may absorb a small portion of the photons, because carrier
generation is relatively far away from the p-i-n junction, the contribution of substrate to J
sc
would
be negligible. We thus expect the power conversion capability mainly comes from the nanowire
array. The slightly lower V
oc
here in sample F compared to sample D and E is due to the logarithmic
dependence of V
oc
on I
sc
for an ideal diode which will be shown later, thus the increase of V
oc
with
shallower junction depth is not as pronounced as J
sc
, and sometimes can be masked by the device-
to-device variation as manifested by the red, blue and pink curves in Fig. 1.7a. All the devices
discussed so far are summarized in Table 1.1.
! 17!
Figure 1.7 Junction depth dependency and J-V under varied light intensity. (a) J-V curves of devices with
different junction depths between 100 nm and 600 nm under AM 1.5G solar spectrum. For the device with
100 nm junction depth, diameter is increased to 320 nm. (b) Dark and AM 1.5G J-V curves of the device
with 100 nm junction depth and 320 nm diameter shown in (a) plotted in semi-logarithmic scale. Ideality
factor is found to be 2.34. (c) J-V curves of the device with 150 nm junction depth and 220 nm diameter
shown in (a) under different illuminating power between 50 μW and 10 mW from an 850 nm laser. V
oc
of
0.716 V is observed under the highest light intensity. Insets shows the curve of V
oc
v.s. ln(I
sc
) with extracted
ideality factor to be 1.72. (d) Current versus pump power for the device shown in (c) at -2 V and 0 V bias.
Table 1.1 Characteristics of devices with different diameter and junction depth
Device #
Diameter
(nm)
Junction
depth
(nm)
J
sc
(mA/cm
2
) V
oc
(V) Efficiency (%)
A 100 300 4.20 0.412 0.68
B 250 300 7.21 0.642 3.03
C 220 600 1.07 0.410 0.21
D 220 300 8.72 0.600 3.30
E 220 150 11.58 0.592 4.62
F 320 100 21.08 0.565 7.58
G 320 300 21.21 0.511 6.56
! 18!
When a non-ideal p-n junction with a series resistance is considered with respect to the forward
bias voltages, the current across a p-n diode is given by
( )
0
exp 1
s
qV IR
II
nkT
!" − $%
= −
&' ()
&'
*+ ,-
In order to accurately determine the series resistance we can differentiate both side of the current
equation and get (Cheung’s method
50
):
( ) ln
s
dV nkT
IR
dI q
=+
A plot of dV/d(lnI) versus I will be linear and gives Rs as slope and nkT/q as the y-axis intercept.
The extracted values of ideality factor and series resistance are 2.2 and 41 Ω, respectively.
Finally, we measured device response under varied illumination power. Fig. 1.7c shows the J-V
curves of device with 300 nm junction depth and 220 nm diameter (red curve in Fig. 1.7b) in dark
and at various illumination powers from an 850 nm laser plotted in semi-logarithmic coordinates.
The short circuit current increases steadily when power is increased from 50 μW to 5 mW and
starts to saturate at higher powers which indicates reduced quantum efficiency. This behavior is
due to increased voltage drop across series resistance under high illumination as well as carrier
screening effect which reduces internal electrical field when carrier concentration is sufficiently
high. Similar trend was observed in photo detector devices made from similar nanowires recently
in our group.
51
By applying a larger reverse bias we can rebuild the strength of electrical field so
we can see incremental change of current scales with illumination power at -2 V applied bias (Fig.
1.7d) while the current at short circuit fails to track under high illumination intensity. The open
circuit voltage increases steadily with increasing pump power. V
oc
for the highest illumination
! 19!
reaches 0.72 V. Neglecting the small series resistance the diode I-V behavior under illumination
could be described as
0
0
exp 1
exp 1
L
sc
qV
II I
nkT
qV
II
nkT
!" !"
= −−
$% $%
&' &'
!" !"
≈−−
$% $%
&' &'
So we can extract V
oc
as a function of short circuit current:
( )
0
0ln 1
sc
oc
I nkT
VVI
qI
!"
== = +
#$
%&
Inset of Fig. 1.7c is the plot of V
oc
versus ln(I
sc
) and the slope is nkT/q, from which we extracted
the ideality factor to be 1.72.
From the results shown in Fig. 1.7a, it is apparent that a shallow junction is essential for
unpassivated GaAs nanowire solar cells to capture minority carriers generated close to the tip
which would otherwise recombine quickly in heavily doped p-emitter region. To better understand
the underlying physical mechanism, we simulated the carrier generation profile under
monochromatic light of different wavelengths and measured the spectrum response of the four
devices shown in Fig. 1.7a. The integration of product of EQE and AM 1.5G photon density
resulted in J
sc
similar to the value measured experimentally in Fig. 1.7a. From Fig. 1.8a we can see
most of the shorter wavelength light (400 nm) is absorbed near the wire surface while a bigger
portion of longer wavelength light (800 nm) is absorbed deeper into the bulk of the nanowire.
Carriers generated by shorter wavelength thus are more likely to recombine at the surface and
annihilate. That explains why the EQE of shorter wavelength is always lower than the EQE of
longer length for all four devices shown in Fig. 1.8b. The surface-to-volume ratio decreases with
increasing diameter so a smaller portion of the total generated carriers are distributed in the vicinity
! 20!
of surface for thicker nanowires which leads to higher EQE in 320 nm thick nanowire than in 220
nm thick nanowire with similar junction depth over a broad range of wavelength. To interpret the
junction depth dependency of J
sc
we conduct cathodoluminescence (CL) measurement (Horiba)
built in Hitachi S-4800 SEM. To excite luminescence from a single nanowire the sample was
cleaved so that electron beam can be focused and scanned normal to the sidewall surface.
Acceleration voltage was kept at 5 kV to ensure that no electrons penetrate the wire to be examined
and cause luminescence from other wires behind. SEM micrographs and normalized CL intensity
maps at 865 nm from two single nanowires, one with 400 nm long p-region (left) and the other
100 nm long p-region (right), are plotted in Fig. 1.8c and the line scans from the center of the two
wires are also superimposed. The top 300 nm of the wire with a 400 nm deep junction does not
emit efficiently. The intensity gradually increases with distance from the top of the nanowire and
eventually saturates when the electron beam moves towards the bottom of the wire. This trend
qualitatively agrees with the aforementioned model that non-radiative recombination processes are
dominant in the p-doped region due to the high impurity concentration introduced to achieve good
ohmic contact. On the other hand the luminescent intensity is much more uniform for the wire with
only 100 nm long p-region. The luminescence intensity gradient along the wire is due to the fact
that electron beam with 5kV acceleration has 200-300 nm interaction volume upon impinging at
the surface although the junction itself is considered more abrupt given the vapor phase epitaxy
employed. In addition, those holes generated close to the junction have a higher probability of
being swept by the junction before they radiatively recombine, while those holes generated far
from the junction are more likely to recombine in the n-layer. The short radiative recombination
life time and high nonradiative recombination rate observed in the p-emitter may well explain the
! 21!
increasing EQE with decreasing junction depth seen in Fig. 1.8b as less carriers would be lost in
the p-emitter when its length is reduced.
Figure 1.8 Spectrum response and Cathodoluminescence mapping. (a) Carrier generation rate profile under
monochromatic light of different wavelengths between 350 nm and 800 nm. All the light intensities are 100
mW/cm
2
. Nanowires are 2.5 μm tall and 350 nm in diameter with 600 nm pitch. Nanowires are surrounded
by BCB and covered by ITO on top. (b) External quantum efficiency (left y-axis) versus wavelength of the
four devices shown in Fig. 5a. Dash-dot curve is the photon intensity of AM 1.5G solar spectrum (right y-
axis). J
sc
of each device are calculated by integrating the product of EQE and photon intensity over the
range of 350 nm to 880 nm and are indicated in the plot. (c) SEM images and normalized
cathodoluminescence mapping at wavelength of 865 nm for nanowires with 400 nm deep junction (left)
and 100 nm deep junction (right). Red curves are relative CL intensity along the center of each nanowire.
Last but not least, the V
oc
measured from our devices is still considerably lower than the values
reported for GaAs planar p-n junctions, which can be attributed to several factors. In nanowire
solar cells because all the nanowires are isolated from each other, we need to form a front contact
that can access all of them. This has been shown to be detrimental to V
oc
in planar solar cells where
! 22!
one tries to passivate most of the emitter surface and to only make contact on area as small as
possible.
52
Intrinsic surface states together with surface defects introduced by processes such as
etching and sputtering can lead to significant recombination at the nanowire/ITO interface leading
to lower V
oc
. The presence of band bending and doping variation in the radial direction will cause
inhomogeneous barrier at the junction. Since the lower barrier part dominates the current
conduction, V
oc
is also largely determined by the lowest barrier height across the intersection. In
principle, a window layer with larger band gap and low surface states density could passivate the
GaAs nanowire surface. This part of work is still on going and we are studying the passivation
effect of shells consist of AlGaAs, GaAsP and InGaP. One issue that needs to be paid extra care
to is the presence of passivation layer should not introduce a significant shunting path between p-
and n-region which requires low unintentional doping in the shell and low interface trap density.
Furthermore, a rigorous assessment of the doping concentration within the nanowire is necessary
to determine the build-in potential which sets the upper limit of V
oc
.
1.5 Summary
In summary, we carried out optical simulation to predict the optimized wire array geometry for
maximum light absorption. We then compared the advantages and limitations of both axial
junction and radial junction design. We experimentally demonstrated solar cells realized by arrays
of GaAs nanowires with axial p-i-n junction which are grown by versatile selective area growth
method using mass production compatible MOCVD technique. Nanowire array with low filling
ratio turns out to be highly absorptive. Systematic study on effects of diameter reveals that large
diameter nanowires are favorable because of the high surface recombination velocity on the bare
GaAs nanowire surface. Junction depth also plays a significant role in carrier collection efficiency,
and shallow junction depth can lead to a short circuit current density as high as 23.28 mA/cm
2
in
! 23!
the absence of any surface passivation treatment. By reducing junction depth to around 100 nm
and keeping diameter at 320 nm we are able to achieve efficiencies as high as 7.58%. Under
concentrated 850 nm light a V
oc
as high as 0.716 V has been obtained. The results demonstrate that
GaAs nanowires are good candidates for high-efficiency and low-cost solar energy conversion and
open up great opportunities for the next generation photovoltaics based on multi-junction devices
composed of lattice mismatched material systems.
Chapter 1 References
1. Tian, B. Z.; Zheng, X. L.; Kempa, T. J.; Fang, Y.; Yu, N. F.; Yu, G. H.; Huang, J. L.; Lieber, C.
M. Nature 2007, 449, 885-889.
2. Garnett, E. C.; Yang, P. D. J. Am. Chem. Soc. 2008, 130, 9224-9225.
3. Tang, J. Y.; Huo, Z. Y.; Brittman, S.; Gao, H. W.; Yang, P. D. Nat. Nanotechnol 2011, 6, 568-572.
4. Kempa, T. J.; Cahoon, J. F.; Kim, S. K.; Day, R. W.; Bell, D. C.; Park, H. G.; Lieber, C. M. Proc.
Natl. Acad. of Sci. U.S.A. 2012, 109, 1407-1412.
5. Czaban, J. A.; Thompson, D. A.; LaPierre, R. R. Nano Lett. 2009, 9, 148-154.
6. Goto, H.; Nosaki, K.; Tomioka, K.; Hara, S.; Hiruma, K.; Motohisa, J.; Fukui, T. Appl. Phys.
Express 2009, 2, 35004.
7. Mariani, G.; Wong, P. S.; Katzenmeyer, A. M.; Leonard, F.; Shapiro, J.; Huffaker, D. L. Nano Lett.
2011, 11, 2490-2494.
8. Wei, W.; Bao, X. Y.; Soci, C.; Ding, Y.; Wang, Z. L.; Wang, D. Nano Lett. 2009, 9, 2926-2934.
9. Shin, J. C.; Kim, K. H.; Yu, K. J.; Hu, H. F.; Yin, L. J.; Ning, C. Z.; Rogers, J. A.; Zuo, J. M.; Li,
X. L. Nano Lett. 2011, 11, 4831-4838.
10. Tomioka, K.; Tanaka, T.; Hara, S.; Hiruma, K.; Fukui, T. IEEE J. Sel. Top. Quant. 2011, 17, 1112-
1129.
11. Hu, S.; Chi, C. Y.; Fountaine, K. T.; Yao, M. Q.; Atwater, H. A.; Dapkus, P. D.; Lewis, N. S.;
Zhou, C. W. Energy Environ. Sci. 2013, 6, 1879-1890.
12. Zhu, J.; Hsu, C. M.; Yu, Z. F.; Fan, S. H.; Cui, Y. Nano Lett. 2010, 10, 1979-1984.
13. Putnam, M. C.; Boettcher, S. W.; Kelzenberg, M. D.; Turner-Evans, D. B.; Spurgeon, J. M.;
Warren, E. L.; Briggs, R. M.; Lewis, N. S.; Atwater, H. A. Energy Environ. Sci. 2010, 3, 1037-1041.
14. Mariani, G.; Scofield, A. C.; Hung, C. H.; Huffaker, D. L. Nat. Commun. 2013, 4, 1497.
! 24!
15. Mariani, G.; Zhou, Z. L.; Scofield, A.; Huffaker, D. L. Nano Lett. 2013, 13, 1632-1637.
16. Holm, J. V.; Jorgensen, H. I.; Krogstrup, P.; Nygard, J.; Liu, H. Y.; Aagesen, M. Nat. Commun.
2013, 4, 1498.
17. Krogstrup, P.; Jorgensen, H. I.; Heiss, M.; Demichel, O.; Holm, J. V.; Aagesen, M.; Nygard, J.;
Morral, A. F. I. Nat. Photonics 2013, 7, 306-310.
18. Cui, Y. C.; Wang, J.; Plissard, S. R.; Cavalli, A.; Vu, T. T. T.; van Veldhoven, R. P. J.; Gao, L.;
Trainor, M.; Verheijen, M. A.; Haverkort, J. E. M.; Bakkers, E. P. A. M. Nano Lett. 2013, 13, 4113-4117.
19. Nakai, E.; Yoshimura, M.; Tomioka, K.; Fukui, T. Jpn. J. Appl. Phys. 2013, 52, 055002.
20. Gutsche, C.; Lysov, A.; Braam, D.; Regolin, I.; Keller, G.; Li, Z. A.; Geller, M.; Spasova, M.; Prost,
W.; Tegude, F. J. Adv. Funct. Mater. 2012, 22, 929-936.
21. Wallentin, J.; Anttu, N.; Asoli, D.; Huffman, M.; Aberg, I.; Magnusson, M. H.; Siefer, G.; Fuss-
Kailuweit, P.; Dimroth, F.; Witzigmann, B.; Xu, H. Q.; Samuelson, L.; Deppert, K.; Borgstrom, M. T.
Science 2013, 339, 1057-1060.
22. King, R. R.; Law, D. C.; Edmondson, K. M.; Fetzer, C. M.; Kinsey, G. S.; Yoon, H.; Sherif, R. A.;
Karam, N. H. Appl. Phys. Lett. 2007, 90, 183516.
23. Geisz, J. F.; Kurtz, S.; Wanlass, M. W.; Ward, J. S.; Duda, A.; Friedman, D. J.; Olson, J. M.;
McMahon, W. E.; Moriarty, T. E.; Kiehl, J. T. Appl. Phys. Lett. 2007, 91, 023502.
24. Geisz, J. F.; Friedman, D. J.; Ward, J. S.; Duda, A.; Olavarria, W. J.; Moriarty, T. E.; Kiehl, J. T.;
Romero, M. J.; Norman, A. G.; Jones, K. M. Appl. Phys. Lett. 2008, 93, 123505.
25. Guter, W.; Schone, J.; Philipps, S. P.; Steiner, M.; Siefer, G.; Wekkeli, A.; Welser, E.; Oliva, E.;
Bett, A. W.; Dimroth, F. Appl. Phys. Lett. 2009, 94, 223504.
26. Derkacs, D.; Jones-Albertus, R.; Suarez, F.; Fidaner, O. J. Photon. Energy 2012, 2, 021805-8.
27. Gudiksen, M. S.; Lauhon, L. J.; Wang, J.; Smith, D. C.; Lieber, C. M. Nature 2002, 415, 617-620.
28. Glas, F. Phys. Rev. B 2006, 74, 121302.
29. Sburlan, S.; Dapkus, P. D.; Nakano, A. Appl. Phys. Lett. 2012, 100, 163108.
30. Chuang, L. C.; Moewe, M.; Chase, C.; Kobayashi, N. P.; Chang-Hasnain, C.; Crankshaw, S. Appl.
Phys. Lett. 2007, 90, 043115.
31. Ertekin, E.; Greaney, P. A.; Chrzan, D. C.; Sands, T. D. J. Appl. Phys. 2005, 97, 114325.
32. Tomioka, K.; Kobayashi, Y.; Motohisa, J.; Hara, S.; Fukui, T. Nanotechnology 2009, 20, 145302.
33. Martensson, T.; Svensson, C. P. T.; Wacaser, B. A.; Larsson, M. W.; Seifert, W.; Deppert, K.;
Gustafsson, A.; Wallenberg, L. R.; Samuelson, L. Nano Lett. 2004, 4, 1987-1990.
34. Moewe, M.; Chuang, L. C.; Crankshaw, S.; Chase, C.; Chang-Hasnain, C. Appl. Phys. Lett. 2008,
93, 023116.
35. Huang, N. F.; Lin, C. X.; Povinelli, M. L. J. Opt. 2012, 14, 024004.
! 25!
36. Huang, N. F.; Lin, C. X.; Povinelli, M. L. J. Appl. Phys. 2012, 112, 064321.
37. Hu, Y.; LaPierre, R. R.; Li, M.; Chen, K.; He, J. J. J. Appl. Phys. 2012, 112, 143116.
38. Madaria, A. R.; Yao, M. Q.; Chi, C. Y.; Huang, N. F.; Lin, C. X.; Li, R. J.; Povinelli, M. L.; Dapkus,
P. D.; Zhou, C. W. Nano Lett. 2012, 12, 2839-2845.
39. Lin, C. X.; Povinelli, M. L. Opt. Express 2009, 17, 19371-19381.
40. Lin, C. X.; Povinelli, M. L. Opt. Express 2011, 19, A1148-A1154.
41. Lin, C. X.; Huang, N. F.; Povinelli, M. L. Opt. Express 2012, 20, A125-A132.
42. Kelzenberg, M. D.; Boettcher, S. W.; Petykiewicz, J. A.; Turner-Evans, D. B.; Putnam, M. C.;
Warren, E. L.; Spurgeon, J. M.; Briggs, R. M.; Lewis, N. S.; Atwater, H. A. Nat. Mater. 2010, 9, 239-244.
43. Heurlin, M.; Wickert, P.; Falt, S.; Borgstrom, M. T.; Deppert, K.; Samuelson, L.; Magnusson, M.
H. Nano Lett. 2011, 11, 2028-2031.
44. Kayes, B. M.; Atwater, H. A.; Lewis, N. S. J. Appl. Phys. 2005, 97, 114302.
45. Breuer, S.; Pfuller, C.; Flissikowski, T.; Brandt, O.; Grahn, H. T.; Geelhaar, L.; Riechert, H. Nano
Lett. 2011, 11, 1276-1279.
46. Kushibe, M.; Eguchi, K.; Funamizu, M.; Ohba, Y. Appl. Phys. Lett. 1990, 56, 1248-1250.
47. Chang, C. C.; Chi, C. Y.; Yao, M. Q.; Huang, N. F.; Chen, C. C.; Theiss, J.; Bushmaker, A. W.;
LaLumondiere, S.; Yeh, T. W.; Povinelli, M. L.; Zhou, C. W.; Dapkus, P. D.; Cronin, S. B. Nano Lett.
2012, 12, 4484-4489.
48. Talin, A. A.; Leonard, F.; Swartzentruber, B. S.; Wang, X.; Hersee, S. D. Phys. Rev. Lett. 2008,
101, 076802.
49. Katzenmeyer, A. M.; Leonard, F.; Talin, A. A.; Toimil-Molares, M. E.; Cederberg, J. G.; Huang,
J. Y.; Lensch-Falk, J. L. IEEE Trans. Nanotechnol. 2011, 10, 92-95.
50. Cheung, S. K.; Cheung, N. W. Appl. Phys. Lett. 1986, 49, 85-87.
51. Seyedi, M. A.; Yao, M.; O'Brien, J.; Wang, S. Y.; Dapkus, P. D. Appl. Phys. Lett. 2013, 103,
251109.
52. Zhao, J. H.; Wang, A. H.; Altermatt, P. P.; Wenham, S. R.; Green, M. A. Sol. Energ. Mat. Sol.
Cells 1996, 41-42, 87-99.
!
! 26!
Chapter 2 Tandem Solar Cells using GaAs Nanowires on Si: Design,
Fabrication, and Observation of Voltage Addition
2.1 Advantages of GaAs-Nanowire-on-Si Tandem Solar Cells
As introduced in chapter 1, GaAs nanowires have superior 1D geometric structure as well as
electronic and optical properties for next generation photovoltaics. Multijunction solar cells aiming
at surpassing the Shockley-Queisser limit
1
of conventional solar cells are facing challenges in
terms of monolithic fabrication. Lattice matching imposes a significant constraint on material
choices and adds complication to the synthesis process. In lattice-mismatched metamorphic
devices, the complexity in the growth of buffer layers or in the wafer peeling-off process is by no
means easy. Furthermore, expensive Ge and GaAs substrates significantly increase the
manufacturing cost so that the efficiency advantage is outweighed by the low cost of Si solar cells
for many applications. The advancements in the understanding of nanowire material properties,
especially GaAs nanowires, as well as various device architectures provide alternative
approaches.
2-9
Furthermore, enhanced interaction between light and nanowires leads to efficient
absorption,
5, 10-16
and lattice-mismatch-induced strain can be relaxed through the nanowire
sidewall
17-21
. Theoretical analysis indicates an optimal tandem cell with 1.1/1.7 eV bandgap
combination can achieve efficiency higher than 40%,
22
making nanowire-on-Si heterostructures
promising candidates to compete with today’s three- and four-junction solar cells. Here, in chapter
2, we fabricated the first nanowire-on-Si tandem solar cells and observed open circuit voltage (V
oc
)
addition of separate GaAs nanowire and Si solar cells up to 0.956 V and efficiency (η) of 11.4%.
! 27!
2.2 The Components of Tandem Solar Cells using GaAs Nanowire on Si
A schematic structure of the GaAs-nanowire-on-Si tandem solar cell is depicted in Fig. 2.1. The
bottom Si cells start from n-type float-zone Si substrates. p
+
emitter and n
+
back surface field are
made by boron and phosphorous implantation followed by rapid thermal annealing. A thin segment
of n
+
doped GaAs nanowires is grown to form good connecting junction with Si. The n-type,
intentionally undoped, and p-type segments are grown sequentially for the top GaAs-nanowire
cell. Nanowire arrays are embedded in a transparent insulating polymer benzocyclobutene (BCB),
which is etched to expose the nanowire tips. The top contact is made by a radio-frequency-
sputtered indium-tin-oxide (ITO) transparent conductive layer and the back contact is e-beam
evaporated Al. For this concept of GaAs-nanowire-on-Si tandem cell to work, we need to work
out the following technology components: current matching between subcells, characterization
and optimization of doping in GaAs nanowires, ohmic connecting junction across n
+
-GaAs/p
+
-Si
hetero-interface, and the integration of subcells to make a functioning tandem cell. Below, we will
study these components in detail.
! 28!
Figure 2.1 Schematic of GaAs nanowire-on-Si tandem solar cell. Top GaAs nanowire solar cell has p
+
emitter, undoped segment, n-type base and n
+
root to form good connecting junction with p
+
Si. Bottom Si
solar cell has p
+
emitter, n-type base and n
+
back surface field. Nanowires are embedded in transparent
insulating polymer BCB. Top contact is ITO and back contact is Al.
The tandem cell should have the top and bottom cell connected in series by a low-resistance
connecting junction. The smaller short-circuit current between the two subcells hence limits the
output current of the tandem cell.
23
We decided the optimal structure for the tandem cell by full-
vector electromagnetic simulations
24
and considered a square array of GaAs nanowires with 300
nm diameter and 600 nm pitch on a semi-infinite silicon substrate. Our previous simulation work
on single-junction GaAs nanowire solar cells on GaAs substrates examines a wide range of
nanowire dimensions such as diameter and pitch, and shows that this set of structural parameters
(diameter of 300 nm and pitch of 600 nm) are around the optimal parameters with high optical
absorption and high short-circuit current density (J
sc
).
9
Fig. 2.2a-c shows the simulated absorption
spectra of the silicon bottom cells (red area) and the GaAs nanowire top cells (blue area) with
different nanowire heights (500 nm for Fig. 2.2a, 900 nm for Fig. 2.2b, and 2000 nm for Fig. 2.2c).
The absorptions of the total structures are also plotted as solid black lines. Although all three
! 29!
structures have similar and high absorptions (>80%), the portion that the GaAs nanowires absorb
becomes larger as the height of the nanowires increases from 500 nm to 2000 nm, which strongly
affects the efficiency of the tandem cells. In Fig. 2.2d-f, we plot the simulated current density (J)
vs. voltage (V) curves
1, 11
for the three cases shown in Fig. 2.2a-c. The absorption was weighted by
AM 1.5G solar spectrum
17
and the solar cells were assumed to have perfect carrier collection and
no nonradiative recombination. A perfect conducting junction was assumed to connect the top and
the bottom cells. For 500 nm nanowire height, the GaAs nanowire top cell can only absorb a small
amount of light, and a significant portion of light with photon energy above the band gap energy
of GaAs (λ < 867 nm) is transmitted through the nanowire structure and is absorbed in the silicon
bottom cell. The J
sc
of the silicon bottom cell (~23 mA/cm
2
) is much higher than that of GaAs
nanowire top cell (~15 mA/cm
2
), which limits the total current and thus the efficiency of the
tandem cell, as indicated by the black curve shown in Fig. 2.2d. On the other hand, Fig. 2.2e
illustrates a nearly optimal case for this system. When the height of GaAs nanowire array is 900
nm, the short-circuit currents of top and bottom cells are similar, leading to the maximum overall
J
sc
of ~ 20 mA/cm
2
. The efficiency also increases from 25.4% to 32.4% under this current-matching
condition. If the height of the nanowire array further increases to 2000 nm (Fig. 2.2f), the silicon
bottom cell absorbs insufficient light and limits the overall current and efficiency. To further show
the importance of current matching, we plot the short-circuit current densities of top and bottom
cells (Fig. 2.2g) and the limiting efficiency (Fig. 2.2h) as functions of the GaAs nanowire array
height. The limiting efficiency peaks at 900 nm nanowire height where the short-circuit current
densities of the top and bottom cells match. We applied the optimized nanowire height of 900 nm
in the experimental work described below. We note that it is also possible to achieve matched
! 30!
currents in the tandem cell by selecting a reasonable nanowire height and then optimizing the
nanowire diameter and pitch.
Figure 2.2 (a-c) Simulated absorption spectra of the Si bottom cell (red area), the GaAs nanowire top cell
(blue area), and the total structures (solid black line) for nanowires with nanowire heights (a) h = 500 nm,
(b) h = 900 nm, and (c) h = 2000 nm. (d-f) Simulated corresponding J-V curves, which show nearly matched
currents between GaAs and Si for (e) the optimized height with h = 900 nm, and mismatched currents for
(d) shorter nanowires with h = 500 nm and (f) taller nanowires with h = 2000 nm. (g) The simulated J
sc
in
the top GaAs nanowire cell and the bottom silicon cell as functions of the nanowire height. The short-circuit
current of the tandem cell is limited by the smaller of the two. (h) The limiting efficiency of the tandem cell
as a function of the nanowire height.
400 600 800 1000
0.0
0.2
0.4
0.6
0.8
1.0
Absorption
Wavelength8(nm)
400 600 800 1000
0.0
0.2
0.4
0.6
0.8
1.0
8silicon
8GaAs
8total
Absorption
Wavelength8(nm)
400 600 800 1000
0.0
0.2
0.4
0.6
0.8
1.0
Absorption
Wavelength8(nm)
0.0 0.5 1.0 1.5 2.0
?25
?20
?15
?10
?5
0
8silicon
8GaAs
8total
Current8density8(mA/cm
2
)
Voltage8(V)
0.0 0.5 1.0 1.5 2.0
?25
?20
?15
?10
?5
0
Current8density8(mA/cm
2
)
Voltage8(V)
0.0 0.5 1.0 1.5 2.0
?25
?20
?15
?10
?5
0
Current8density8(mA/cm
2
)
Voltage8(V)
a b c
d e
f
0 500 1000 1500 2000
0
5
10
15
20
25
30
8GaAs8top8cell8(a=6008nm8d=3008nm)
8Silicon8bottom8cell
J
sc
8(mA/cm
2
)
GaAs8nanowire8height8(nm)
0 500 1000 1500 2000
5
10
15
20
25
30
35
Limiting8efficiency8(%)
GaAs8nanowire8height8(nm)
h8=85008nm h8=89008nm h8=820008nm
η=825.4%
η=832.4%
η=825.6%
g
h
! 31!
GaAs nanowires used in this study were grown by selective area growth (SAG) technique using
metalorganic chemical vapor deposition (MOCVD). Briefly, trimethylgallium (TMG) and arsine
(AsH
3
) were used as the precursors for Ga and As with partial pressures of 7.56×10
-7
atm and
2.14×10
-4
atm, respectively. Annealing in hydrogen ambient at 925 ºC for 5 minutes was essential
for high-yield vertical nanowire growth. Disilane (diluted, 100 ppm in hydrogen) was used as the
precursor for Si, which is the n-type dopant for GaAs nanowires. Fig. 2.3a-c shows the scanning
electron microscopy (SEM) images of GaAs nanowires grown on Si (111) substrates. Fig. 2.3a
presents the uniformity of growth from the top-view. The yield of vertical nanowires is 100%. Fig.
2.3b is a 30° tilted view showing the uniform wires have hexagonal cross sections enclosed by {1-
10} facets, while Fig. 2.3c is a magnified top-view image.
Figure 2.3 SEM images of GaAs nanowires. (a-c) SEM images of GaAs nanowires grown on Si (111)
substrate using SAG. (a) Top view, scale bar 5 µm. (b) 30° tilted view, scale bar 1 µm. (c) Magnified top
view, scale bar 500 nm.
A tandem cell requires a low-resistance connecting junction (usually a tunnel junction) sandwiched
by the two subcells, which can conduct current through a reverse-biased p-n junction. Typical
connecting junctions require doping concentration higher than 10
19
cm
-3
.
25, 26
Well-controlled
degenerate doping in nanowires, however, presents a big challenge due to the lack of a precise
doping characterization technique and the presence of surface states which reduce the effective
! 32!
carrier concentration.
27
Recent progress on Si/InAs nanowire heterojunction demonstrated very
high current density,
28, 29
indicating heavily doped III-V nanowire-on-Si heterojunction could be a
viable approach to serve as the connecting junction. Previous studies
relied on field effect transistor
measurements to assess the carrier density, but the precision was limited by the estimation of gate
capacitance. Atom probe tomography is another emerging technique to map dopant distribution,
which involves sophisticated data reconstruction.
30
The technological issues in doping
characterization have hindered the progress of nanowire-on-Si tandem solar cells.
It has been well studied that heavy impurity incorporation results in notable changes in the
semiconductor band structure. The random distribution of impurities disturbs the original periodic
energy potential. Because the translational symmetry is broken, indirect k-nonconserving
transitions become possible. Another important effect is the blue shift of interband transition
energy due to band filling in materials with degenerate electron distribution, which is named
Burstein-Moss effect.
31, 32
These effects on the band structure are directly reflected in the
spontaneous emission spectra.
Depending on the impurity concentration, two optical transition mechanisms exist:
i.! When translational symmetry is preserved, k-conservation direct band-to-band transition
dominates (Figure 2.4a). The spontaneous emission intensity can be described as:
(1)
( ) IE
( )
1
*
12
2
**
1
*
2
**
() 1 exp
11 exp
e
gF g
h
g
eh
h
gF g
e
eh
eh
EE E E
m
IE E E E
mm kT kT
EE E E
m
M
mm kT kT
−
−
"# $% −−
∝× − × + ×− () *+
*+
+
()
,- ./
01
"# $% −−
−22
×− + ×− × () *+ 34
*+
+
()
22 ,- ./
56
! 33!
where the first square term refers to the simple photon density of states in the bulk, the second
square root term refers to the density of states of the electron. The next term in the brackets
refers to the probability of a state in the conduction band being occupied by an electron based
on Fermi-Dirac distribution, while the term in braces is the probability that a corresponding
vertical state in the valence band is unoccupied. The last term is the transition matrix element
determining the oscillator strength of the transition.
ii.! When translational symmetry is broken, indirect band-to-acceptor transition without k-
conservation dominates (Fig. 2.4b). Most photo-created minority carriers thermalize
completely before radiative recombination and are energetically situated at the extremum of
the band tail. Thus the PL spectra directly reflects the carrier population in the conduction
band:
(2),
where ρ
c
is the density of states in conduction band. In sufficiently heavily doped
semiconductors, the band edge fluctuates spatially. By assuming a Gaussian distribution of the
edge fluctuation and the local density of states (DOS) still following a parabolic relationship,
Kane gave the analytic form of distorted density of states
33
:
(3),
where E
c
is the integral variable to take into account all the possible band edge positions and
σ
E
is the root mean square of the energy fluctuation of the band edge.
( )
1
2
2
1 exp
f
ceh
EE
IE E M
kT
ρ
−
#−$ %&
∝× × + ×
)* +,
-. /0
( )
2 32
*
22
12 1 1
exp
22 2
E
c
cc c
E E
E m
EEE dE ρ
πσ πσ
−∞
&'
() ()
*+ = −−
,- ,-
*+ ./ ./
01
∫
! 34!
Figure 2.4 (a) Schematic of k-conservation transition in lightly doped and intrinsic semiconductor. (b)
Schematic of k-nonconservation transition in heavily doped semiconductor due to broken translational
symmetry.
To avoid the donor-acceptor transition which is dominant in low-temperature PL, our study
focused on room temperature PL. As grown nanowire arrays were excited by a 532 nm laser with
relatively low power intensity (~1 W cm
-2
) to avoid optical heating. Room temperature PL spectra
were collected using a 100x objective lens with 0.8 numerical aperture, an 1800 mm
-1
grating, and
a Si charge-couple device (CCD) detector. Fig. 2.5a shows the theoretical PL spectrum of the
intrinsic GaAs nanowire using the direct band-to-band transition mechanism (blue) and
experimentally measured spectrum of undoped GaAs nanowire array (red) at room temperature.
Although no intentional doping was introduced in the undoped sample, the band-to-band transition
model spectrum deviates from the experimentally measured spectrum. While the high energy side
matches well, a significant tail below the band edge is observed in the experimental spectrum.
Since no disilane was supplied, we speculate the band tail is induced by unintentionally
incorporated carbon from methyl radicals in TMG during the pyrolysis process
34, 35
and is enhanced
on As rich surfaces.
36
Fig. 2.5b is the PL spectrum measured from a single undoped nanowire at 4
! 35!
K. Peak fitting with Lorentzian line shapes was conducted to deconvolve individual peaks. Two
dominant peaks are located at 828 and 834 nm, well beyond the band edge at 816 nm. These two
peaks have been previously reported to be related to electron-to-carbon acceptor (e, A
0
) and carbon
donor-to-carbon acceptor (D
0
, A
0
) transitions.
37
Figure 2.5 (a) Theoretical PL spectrum of the intrinsic GaAs nanowire using the direct band-to-band
transition mechanism (blue) and experimentally measured spectrum of undoped GaAs nanowire array (red)
at room temperature. (b) The PL spectrum of a single undoped GaAs nanowire at 4 K and its Lorentzian
peak fitting. Two major peaks at 828 nm and 834 nm are beyond the band edge and identified as related to
electron-to-carbon acceptor (e, A
0
) and carbon donor-to-carbon acceptor (D
0
, A
0
) transitions.
Given the presence of tail states, we interpreted all the measured PL spectra through the proposed
indirect band-to-acceptor transition model. We used equation (2) and (3) to fit the experimental
data using Fermi-level E
f
and mean square root of band edge fluctuation σ
E
as fitting parameters.
E
g
is fixed at the nominal band gap of 1.424 eV for GaAs at room temperature. The best fittings
are shown in Fig. 2.6 with the black dash-dot fitted curves superimposed on the solid measured
curves. All the curves are normalized between 0 and 1, and are offset to better illustrate the line
shape evolution with increasing disilane flow rate. The standard deviation for each fitted curve is
! 36!
on the order of 10
-4
. E
f
and σ
E
of the best fittings for each doping level are plotted in Fig. 2.3h,
based on which the carrier density is calculated as:
! (4),!
where ρ
c
is defined in equation (3). The calculated carrier density is plotted in Figure 3i. The carrier
concentration increases approximately linearly until it reaches about 2×10
18
cm
-3
. It then saturates
with further increases of the disilane flow rates and eventually decreases after it reaches 2.25×10
18
cm
-3
. Previous studies of Si doped bulk GaAs also found that the carrier density increases linearly
up to 3×10
18
cm
-3
with increasing disilane-to-TMG ratio and the Si atoms are fully ionized as
donors.
38
Further increasing the disilane flow rates leads to linearly increased atomic Si
concentrations in GaAs while the carrier density tends to saturate. Such a deviation is caused by
the self-compensation mechanism of Si given its amphoteric nature.
39, 40
Domke et al. atomically
resolved various compensation mechanisms in real-space by scanning tunneling microscopy
40
.
With increasing Si concentrations, the Si
Ga
donors are consecutively electrically deactivated by
Si
As
acceptors, Si clusters and Si
Ga
-Ga vacancy complexes.
( )
1
1 exp
f
c
EE
n E dE
kT
ρ
−
∞
−∞
$−% &'
= × +
)* +,
-. /0
∫
! 37!
Figure 2.6 Normalized room temperature PL spectra of as grown GaAs nanowire arrays on Si. Disilane
flow rate is varied between 0 and 1 sccm during the growth. Dotted lines are fitted curves. Curves are
intentionally offset for clear view.
Figure 2.7 (a) Fermi level E
f
(black) and band tail depth σ
E
(red) extracted from photoluminescence study.
(b) Carrier density calculated based on the extracted values shown in (a).
Based on the doping information in the GaAs nanowires, we then characterized the transport
properties of the proposed n
+
-GaAs nanowire/p
+
-Si heterojunction. The device schematic is shown
! 38!
in the inset of Fig. 2.8. The Si wafer was implanted with boron ion followed by thermal activation.
The doping concentration in the surface layer was around 1×10
21
cm
-3
. After BCB infiltration, the
nanowire arrays were planarized and etched back to expose the tips. Ohmic top contact was formed
by depositing AuGe/Ni/Au and rapid thermal annealing. Nanowires doped with 0.2, 0.5 and 1
sccm disilane were studied. From the current (I) vs. voltage (V) curves in Fig. 2.4, the 0.2 and 0.5
sccm samples show symmetrical shapes with barriers while the 1 sccm sample shows ohmic
behavior. We have seen earlier in Fig. 3i that the carrier density for the sample doped with 1 sccm
disilane flow rate is lower than that of the sample doped with a 0.5 sccm disilane flow rate.
However, the former shows higher current and lower energy barrier when forming a heterojunction
with p
+
Si. We believe that the deep acceptor states in GaAs with higher Si concentration
effectively reduced the valence band discontinuity between GaAs and Si, lowering the barrier for
hole current. More deep level states can also facilitate trap-assisted tunneling process at the hetero-
interface. We note that the heterojunction of n
+
-GaAs/p
+
-Si prepared using SAG MOCVD is rather
new and has not been fully studied, and whether the low-resistance n
+
-GaAs/p
+
-Si connecting
junction we have made is a tunnel junction deserves further study. The extracted resistance from
Fig. 2.8 for the sample doped with 1 sccm disilane flow rate is 50 Ω for a device area of 1 mm ×
1 mm, meaning the voltage drop across the heterojunction would be only around 0.01 V for an
estimated 20 mA/cm
2
J
sc
for the tandem cell. Thus this heterojunction is a good candidate as
interconnection for the proposed tandem solar cell architecture.
! 39!
Figure 2.8 n
+
-GaAs/p
+
-Si heterojunction characterization. I-V curves of three doping levels, 0.2, 0.5 and 1
sccm, were measured. The former two shows barrier while the latter does not. Inset shows the schematic of
the device in measurement.
2.3 The Integration of Tandem Solar Cells using GaAs Nanowire on Si and the Observation
of Voltage Addition
In order to gain more insight into the performance of the tandem cells, we fabricated and
characterized separate Si solar cells and GaAs nanowire solar cells. When characterizing stand-
alone Si solar cells, we deposited Ti/Pd/Ag as the front contact and Al as the back contact. The Si
solar cell fabrication uses float-zone n-type silicon wafers of 3 Ω cm resistivity. Emitter was
formed by boron ion implantation. The ion energy was 30 keV and the dose was 1.5×10
16
cm
-2
.
Back surface field was formed by phosphorous implantation. The ion energy was 40 keV and the
dose was 8×10
15
cm
-2
. After cleaning, ions were activated in forming gas at 1050 °C for 1 minute
using rapid thermal annealing. 1 μm thick Al was deposited on backside and annealed at 400 °C
to form good ohmic back contact. Around 70 nm plasma enhanced chemical vapor deposition
I"V
p+&Si
GaAs
AuGe
BCB
"1.5 "1.0 "0.5 0.0 0.5 1.0 1.5
"0.03
"0.02
"0.01
0.00
0.01
0.02
0.03
0.04
Current&(A )
Voltage&(V)
&disilane&1sccm
&0.5sccm
&0.2sccm
&
&
&
&
! 40!
grown silicon nitride acted as antireflection layer for boron-doped emitter. Top metal contact
pattern was defined by photolithography and buffered oxide etching removed silicon nitride in
metal contact region. Ti/Pd/Ag were sequentially deposited by e-beam evaporation for front metal
contact. After lift-off, the wafers were diced into 1 cm × 1 cm pieces. The J-V curves of Si solar
cells were measured both in dark and under AM 1.5G solar spectrum, which give J
sc
of 26.28
mA/cm
2
, V
oc
of 0.547 V, fill factor (FF) of 0.658 and overall efficiency of 9.41%. Shunt resistance
and series resistance are 8.61 kΩ and 3.24 Ω for a 1 cm
2
cell. Ruling out the effect of shunt and
series resistance, the FF is 0.77 and η is 11.03%. The performance is well below that of state-of-
the-art commercial Si solar cells due to the lack of optimization in many aspects. To just name a
few, no surface texturing was formed to reduce reflection loss, surface passivation layers were not
introduced to lower the surface recombination losses at both sides, the doping of emitter layer was
high in order to achieve good heterojunction, resulting in short minority lifetime,
41
and the shunt
resistance was finite due to the small dimensions of the lab cells. To characterize the top GaAs
nanowire cell, we applied the same techniques as in chapter 1
15
to fabricate stand-alone GaAs
nanowire cells except that the nanowire heights were modified to be 900 nm. The cells show J
sc
of
21.50 mA/cm
2
, V
oc
of 0.518 V, FF of 0.581 and η of 6.44%. This short-circuit current density is
similar to that of the simulated GaAs subcell in optimal tandem cell structure with 900 nm
nanowire height, which verifies that the experimental results are consistent with our simulation.
The final accomplishment was to integrate all essential components and fabricate functioning
nanowire-on-Si tandem cells. Following the same steps for Si cell fabrication up to implanted ion
activation, we grew a thin segment of GaAs nanowires with 1 sccm disilane on Si bottom cell in
order to form low-resistance connecting junction mentioned above. During the growth of GaAs
nanowires for the top cell, we controlled the heights of the nanowires to 900 nm to keep them the
! 41!
same as in our stand-alone GaAs nanowire cell. After nanowire growth, transparent insulating
polymer BCB (Cyclotene, Dow) was drop-casted on the as grown samples and left infiltrating for
about 12 minutes for complete penetration to the roots of nanowires. The polymer was then spun
at 6000 rpm. The polymer was cured at 250
o
C in vacuum furnace and etched by CF
4
/O
2
RIE to
expose the tip consisting p-type emitter. Subsequently front electrodes were formed by depositing
a 300 nm thick ITO using radio frequency sputtering at 300 W power and 5 mT chamber pressure.
1 μm thick Al was deposited on backside and annealed at 400
o
C to form back contact. The tandem
solar cell performance was measured under AM 1.5G solar spectrum. Samples were diced into the
area with nanowires grown on top (1 mm
2
) for fair evaluation. The most efficient tandem cell
shows J
sc
of 20.64 mA/cm
2
, V
oc
of 0.956 V, FF of 0.578 and η of 11.4%. The J-V curves of the
stand-alone Si cell, the stand-alone GaAs nanowire cell, as well as the integrated tandem cell are
shown in Fig. 2.9a. The open circuit voltages of the stand-alone Si cell (red), the stand-alone GaAs
nanowire cell (blue), and the integrated tandem cell (black) are 0.547 V, 0.518 V and 0.956 V
respectively, as indicated by arrows and directly compared in Fig. 2.9a. The open circuit voltage
of the tandem cell is approximately equal to the summation of the open circuit voltages of the Si
and the GaAs nanowire cells. The slight drop in the tandem cell’s V
oc
as compared to the
summation could be due to the enhanced surface recombination at the hetero-interface for the Si
solar cell and nonradiative recombination at heavily doped nanowire root for the GaAs nanowire
solar cell. In addition, the J
sc
of the tandem cell (20.64 mA/cm
2
) is very close to the J
sc
in the
simulated current-matching case (19.4 mA/cm
2
) in Fig. 2.2e. Based on these evidences, we believe
the GaAs nanowire subcell and the Si subcell in our tandem cell are connected in series with nearly
matched currents.
! 42!
Figure 2.9 J-V curves and EQE (a) Open circuit voltage addition of the GaAs nanowire-on-Si tandem solar
cell. J-V curves of the GaAs nanowire-on-Si tandem solar cell (black), the stand-alone GaAs nanowire cell
(blue) and the stand-alone Si cell (red). The arrows indicate the open circuit voltages for each cell, showing
an addition of V
oc
of the tandem cell (0.956 V) from separate stand-alone GaAs nanowire (0.518 V) and
stand-alone Si (0.547 V) cells. (b) Experimentally measured EQE of the tandem cell.
2.4 The Features of External Quantum Efficiency Measurement
We also carried out external quantum efficiency (EQE) measurement on the tandem cell. The
measured EQE vs. wavelength plot is shown in Fig. 2.9b. To study the features of the EQE curve,
the simulated absorption spectra of the GaAs nanowire top cell (blue circles) and the Si bottom
cell (red triangles) are plotted in Fig. 2.10a. Here, we adopt nanowires with 300 nm diameter and
600 nm pitch in the array. The absorption spectrum of the GaAs nanowire top cell has a peak at
the wavelength of 650 nm, which is due to the GaAs absorption coefficient properties and the
resonant modes at this nanowire array geometry. As the tandem cell current is limited by the
smaller of the two currents in the subcells, the simulated EQE curve of the tandem cell (solid green
curve in Fig. 2.10a) is the smaller of the GaAs and Si absorption spectra. When the wavelength is
larger than 650 nm, as the wavelength increases, the absorption in the GaAs subcell decreases, as
!0.2 0.0 0.2 0.4 0.6 0.8 1.0
!30
!25
!20
!15
!10
!5
0
5
10
+
+
V
oc
$=$0.956$V
V
oc
$=$0.547$V
Current+Density+(mA/cm
2
)
Voltage+(V)
$Tandem$Cell
$GaAs$Standalone
$Si$Standalone
V
oc
$=$0.518$V
a b
0
0.1
0.2
0.3
0.4
0.5
480 580 680 780 880
EQE
Wavelength$(nm)
Experimental$tandem$cell$EQE
! 43!
a result, more photons can reach the Si bottom cell, and the absorption in Si subcell increases.
Their intersection at 740 nm gives a peak in the tandem cell output current, which corresponds to
the current matching condition. Correspondingly, there is a valley in the simulated Si bottom cell
absorption at 650 nm, which is due to the strong absorption of the GaAs nanowire top cell, and is
also reflected in the simulated EQE curve. The simulated EQE curve (solid green) and the
measured EQE curve (black squares) in Fig. 2.10a are qualitatively consistent. For example, both
curves show sharp drops for wavelength greater than 867 nm, as photons with energy below the
GaAs bandgap would not get absorbed by GaAs. In addition, both curves show two peaks with a
valley in between, although the specific peak positions have some difference, probably due to
factors such as variations in the nanowire geometry and the presence of the ITO top contacts,
which are hard to be fully studied in the simulation. Nevertheless, we further studied the effect of
the nanowire diameter variation on the absorption spectra. Measurements of the nanowire diameter
in the nanowire samples reveal that the standard deviation of the diameter variation is 6.7%, i.e.,
20 nm standard deviation for 300 nm average diameter. Simulations of the absorption of tandem
cells with different diameters (Fig. 2.10b, orange curve for d = 280 nm, green curve for d = 300
nm and purple curve for d = 320 nm) show that the valley and peak positions shift right as the
nanowire diameter increases. The combined effect is that the valley in the tandem cell absorption
near 650 nm would become much shallower when there is variation in the nanowire diameter,
which is consistent with the rather shallow valley in the measured EQE curve.
! 44!
Figure 2.10 The simulated absorption spectra. (a) Simulated absorption spectra of GaAs nanowire subcell
(blue circles), Si subcell (red triangles), and the tandem cell (solid green), compared to the experimental
EQE (black squares) of tandem cell (b) Simulated absorption spectra of GaAs nanowire subcell with
different nanowire diameters. d = 280 nm (orange), d = 300 nm (green) and d = 320 nm (purple).
2.5 The Integration of the product of EQE and AM 1.5G photon density
The AM 1.5G solar spectrum is a portfolio of different wavelengths of weighted intensity. For a
single junction cell, as there is no current limiting effect of different junctions, the integration of
the product of EQE and AM 1.5G photon density should give the same J
sc
as the measured J
sc
,
when the solar cell is illuminated by white light with AM 1.5G spectrum. For a tandem cell, the
current limiting effect can make a difference. In our case, when measuring the EQE at a single
short wavelength, the GaAs nanowire top cell absorption is strong, leaving very few photons
a
b
0
0.1
0.2
0.3
0.4
0.5
480 580 680 780 880
Absorption
Wavelength;(nm)
d;=;280;nm
d;=;300;nm
d;=;320;nm
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
480 580 680 780 880
Absorption
Wavelength;(nm)
Simulated;GaAs;subcell
Simulated;Si;subcell
Simulated;Tandem;Cell
Experimental;EQE
! 45!
reaching the Si bottom cell for generating photocurrent, so the Si subcell is the limiting junction.
The current limiting effect leaves the photocurrent flowing across the tandem cell very small, and
thus the EQE at this short wavelength is very small. On the contrary, when measuring J
sc
under
AM 1.5G spectrum, despite that very few photons reach the Si subcell at the same short
wavelength, photons at other wavelengths can still reach the Si subcell. As a result, the current
generated by the Si limiting junction increases, more current flows across the tandem cell and more
photons absorbed by GaAs subcell at this short wavelength can contribute to the measured tandem
cell J
sc
. Similarly, at long wavelengths, especially at wavelengths with photon energy smaller than
the GaAs bandgap, the EQE is very small or even zero. On the other hand, under AM 1.5G
spectrum, the photocurrent generated by Si subcell at these long wavelengths can still contribute
to measured J
sc
as there is reasonable photocurrent in the GaAs subcell generated at other
wavelengths. So for a tandem cell, the integration of the product of EQE and AM 1.5G photon
density results in J
sc
much smaller than the measured J
sc
under AM 1.5G spectrum. Based on Fig.
9b, the integration of the product of EQE and AM 1.5 G photon density results in a value of 7.58
mA/cm
2
, which is indeed smaller than the measured J
sc
of 20.64 mA/cm
2
.
2.6 Summary
Although the overall efficiency is still low compared to today’s GaAs or Si solar cells, our
demonstration of tandem cell behavior through nanowires on a mismatched substrate serves as a
starting point and opens up routes for future development and optimization. For example, surface
passivation of both Si and GaAs nanowire subcells would allow their individual V
oc
to approach
that of state-of-the-art Si and GaAs solar cells. In our study, a silicon nitride layer was prepared
using plasma enhanced chemical vapor deposition to serve as both the growth mask and the front
surface passivation layer of the Si solar cell; however, the silicon nitride deposition condition needs
! 46!
to be further tuned and optimized to minimize the dangling bonds on the Si surface. In addition,
the back surface of the Si cell was not passivated in our study, which would be another source of
recombination loss, and we believe that the passivation of the back Si surface can lead to improved
efficiency. Even more important is the surface treatment of GaAs nanowire, since GaAs is known
for high density of surface states and the surface-to-volume ratio is very large for nanowire
structures. Well-engineered shells with wide band gaps, e.g., AlGaAs and InGaP, could be
promising candidates.
42, 43
Another significant constraint is the short diffusion length in the emitter
of the Si solar cell due to the ultra-high doping. In case one can confine the high doping only to
the area underneath the nanowires, both V
oc
and J
sc
of the Si cells can be further improved.
To conclude, the work presents the concept of nanowire-on-Si heterostructure involving
monolithic integration of lattice-mismatched materials. Our work starts from the design of optimal
structure to achieve current matching between subcells, the epitaxial growth of the top GaAs
nanowire cell on the bottom planar Si cell, and the formation of a low-resistance connecting
junction between the n
+
-GaAs/p
+
-Si hetero-interface. Our simulation indicates that GaAs nanowire
length affects current matching and thus is critical to high efficiency. We also demonstrated
uniform GaAs nanowire growth on lattice-mismatched Si substrate with 100% yield and developed
an optical and non-destructive method to estimate the doping concentration using
photoluminescence (PL). By varying the n-type dopant precursor flow rate, we were able to
achieve low-resistance ohmic connecting junction between GaAs nanowire and Si subcells. The
fabricated nanowire-on-Si tandem cells take advantage of the optimized parameters for each
component, and it is the first demonstration of tandem solar cells using top GaAs nanowire array
cells grown on lattice-mismatched bottom Si cells, which not only shows current matching
between individual subcells but also achieves low-resistance ohmic behavior across the
! 47!
heterojunction. The tandem cells observe voltage addition of GaAs nanowire and Si cells showing
V
oc
up to 0.956 V and efficiency of 11.4%. The presented structure opens up great opportunities
for future nanowire-based, low-cost, and high-efficiency multijunction solar cells.
Chapter 2 References
1. Shockley, W.; Queisser, H. J. J. Appl. Phys. 1961, 32, 510-519.
2. Tian, B. Z.; Zheng, X. L.; Kempa, T. J.; Fang, Y.; Yu, N. F.; Yu, G. H.; Huang, J. L.; Lieber, C.
M. Nature 2007, 449, 885-889.
3. Goto, H.; Nosaki, K.; Tomioka, K.; Hara, S.; Hiruma, K.; Motohisa, J.; Fukui, T. Appl. Phys.
Express 2009, 2, 035004.
4. Tomioka, K.; Tanaka, T.; Hara, S.; Hiruma, K.; Fukui, T. IEEE J. Sel. Top. Quant. 2011, 17, 1112-
1129.
5. Hu, S.; Chi, C. Y.; Fountaine, K. T.; Yao, M. Q.; Atwater, H. A.; Dapkus, P. D.; Lewis, N. S.;
Zhou, C. W. Energy Environ. Sci. 2013, 6, 1879-1890.
6. Mariani, G.; Scofield, A. C.; Hung, C. H.; Huffaker, D. L. Nat. Commun. 2013, 4, 1497.
7. Cui, Y. C.; Wang, J.; Plissard, S. R.; Cavalli, A.; Vu, T. T. T.; van Veldhoven, R. P. J.; Gao, L.;
Trainor, M.; Verheijen, M. A.; Haverkort, J. E. M.; Bakkers, E. P. A. M. Nano Lett. 2013, 13, 4113-4117.
8. Wallentin, J.; Anttu, N.; Asoli, D.; Huffman, M.; Aberg, I.; Magnusson, M. H.; Siefer, G.; Fuss-
Kailuweit, P.; Dimroth, F.; Witzigmann, B.; Xu, H. Q.; Samuelson, L.; Deppert, K.; Borgstrom, M. T.
Science 2013, 339, 1057-1060.
9. Yao, M. Q.; Huang, N. F.; Cong, S.; Chi, C. Y.; Seyedi, M. A.; Lin, Y. T.; Cao, Y.; Povinelli, M.
L.; Dapkus, P. D.; Zhou, C. W. Nano Lett. 2014, 14, 3293-3303.
10. Huang, N. F.; Lin, C. X.; Povinelli, M. L. J. Opt. 2012, 14, 024004.
11. Huang, N. F.; Lin, C. X.; Povinelli, M. L. J. Appl.Phys. 2012, 112, 064321.
12. Wen, L.; Zhao, Z. F.; Li, X. H.; Shen, Y. F.; Guo, H. M.; Wang, Y. Q. Appl. Phys. Lett. 2011, 99,
143116.
13. Madaria, A. R.; Yao, M. Q.; Chi, C. Y.; Huang, N. F.; Lin, C. X.; Li, R. J.; Povinelli, M. L.; Dapkus,
P. D.; Zhou, C. W. Nano Lett. 2012, 12, 2839-2845.
14. Lin, C. X.; Povinelli, M. L. Opt. Express 2009, 17, 19371-19381.
15. Kelzenberg, M. D.; Boettcher, S. W.; Petykiewicz, J. A.; Turner-Evans, D. B.; Putnam, M. C.;
Warren, E. L.; Spurgeon, J. M.; Briggs, R. M.; Lewis, N. S.; Atwater, H. A. Nat. Mater. 2010, 9, 239-244.
16. LaPierre, R. R. J. Appl. Phys. 2011, 110, 014310.
! 48!
17. Gudiksen, M. S.; Lauhon, L. J.; Wang, J.; Smith, D. C.; Lieber, C. M. Nature 2002, 415, 617-620.
18. Glas, F. Phys. Rev. B 2006, 74, 121302.
19. Sburlan, S.; Dapkus, P. D.; Nakano, A. Appl. Phys. Lett. 2012, 100, 163108.
20. Chuang, L. C.; Moewe, M.; Chase, C.; Kobayashi, N. P.; Chang-Hasnain, C.; Crankshaw, S. Appl.
Phys. Lett. 2007, 90, 043115.
21. Ertekin, E.; Greaney, P. A.; Chrzan, D. C.; Sands, T. D. J. Appl. Phys. 2005, 97, 114325.
22. Tobias, I.; Luque, A. Prog. Photovoltaics 2002, 10, 323-329.
23. Hu, Y.; Li, M.; He, J. J.; LaPierre, R. R. Nanotechnology 2013, 24, 065402.
24. Li, M.; Hu, X. H.; Ye, Z.; Ho, K. M.; Cao, J. R.; Miyawaki, M. Opt. Lett. 2006, 31, 3498-3500.
25. Olson, J. M.; Kurtz, S. R.; Kibbler, A. E.; Faine, P. Appl. Phys. Lett. 1990, 56, 623-625.
26. Bertness, K. A.; Kurtz, S. R.; Friedman, D. J.; Kibbler, A. E.; Kramer, C.; Olson, J. M. Appl. Phys.
Lett. 1994, 65, 989-991.
27. Casadei, A.; Krogstrup, P.; Heiss, M.; Rohr, J. A.; Colombo, C.; Ruelle, T.; Upadhyay, S.;
Sorensen, C. B.; Nygard, J.; Morral, A. F. I. Appl. Phys. Lett. 2013, 102, 013117.
28. Bessire, C. D.; Bjork, M. T.; Schmid, H.; Schenk, A.; Reuter, K. B.; Riel, H. Nano Lett. 2011, 11,
4195-4199.
29. Yang, T.; Hertenberger, S.; Morkotter, S.; Abstreiter, G.; Koblmuller, G. Appl. Phys. Lett. 2012,
101, 233102.
30. Perea, D. E.; Hemesath, E. R.; Schwalbach, E. J.; Lensch-Falk, J. L.; Voorhees, P. W.; Lauhon, L.
J. Nat. Nanotechnol. 2009, 4, 315-319.
31. Burstein, E. Phys. Rev. 1954, 93, 632-633.
32. Moss, T. S. P. Phys. Soc. Lond. B 1954, 67, 775-782.
33. Kane, E. O. Phys. Rev. 1963, 131, 79-88.
34. Kushibe, M.; Eguchi, K.; Funamizu, M.; Ohba, Y. Appl. Phys. Lett. 1990, 56, 1248-1250.
35. Kuech, T. F.; Redwing, J. M. J. Cryst. Growth 1994, 145, 382-389.
36. Kuech, T. F.; Veuhoff, E. J. Cryst. Growth 1984, 68, 148-156.
37. Bose, S. S.; Lee, B.; Kim, M. H.; Stillman, G. E.; Wang, W. I. J. Appl. Phys. 1988, 63, 743-748.
38. Shimazu, M.; Kamon, K.; Kimura, K.; Mashita, M.; Mihara, M.; Ishii, M. J. Cryst. Growth 1987,
83, 327-333.
39. Druminski, M.; Wolf, H.-D.; Zschauer, K.-H.; Wittmaack, K. J. Cryst. Growth 1982, 57, 318-324.
40. Domke, C.; Ebert, P.; Heinrich, M.; Urban, K. Phys. Rev. B 1996, 54, 10288-10291.
41. Green, M. A. Prog. Photovoltaics 2009, 17, 183-189.
! 49!
42. Chang, C. C.; Chi, C. Y.; Yao, M. Q.; Huang, N. F.; Chen, C. C.; Theiss, J.; Bushmaker, A. W.;
LaLumondiere, S.; Yeh, T. W.; Povinelli, M. L.; Zhou, C. W.; Dapkus, P. D.; Cronin, S. B. Nano Lett.
2012, 12, 4484-4489.
43. Huang, N. F.; Povinelli, M. L. IEEE J. Photovolt. 2014, 4, 1511-1517.
!
! 50!
Chapter 3 Facile Five-Step Heteroepitaxial Growth of GaAs
Nanowires on Si Substrates and the Twin Formation Mechanism
3.1 The Advantages of of Semiconductor Nanowires on Si
While III-V compound semiconductor nanowires have attracted a great deal of research interest
due to their unique electronic and optical properties,
1-8
Si is still the dominant material in today’s
semiconductor industry because of its relatively low cost and mature processing technology. It
thus has been a natural desire to integrate III-V semiconductors with Si to enable novel functional
devices that can take advantage of the benefits offered by both materials. One noticeable example
is the development of tandem solar cells and future multijunction solar cells using III-V materials
grown on low-cost and sturdy Si substrates. Furthermore, the technology could also substantially
influence the integrated circuit industry by the monolithic integration of III-V materials and Si
circuitry.
To achieve such integration in a thin film fashion represents a significant technological challenge
caused by several problems intrinsic to the interface of the two dissimilar materials. Mismatches
in the lattice constants and thermal expansion coefficients can lead to misfit dislocations or even
non-epitaxial growth.
9, 10
The polar nature of the bonds in III-V materials makes the growth on
nonpolar Si difficult and causes anti-phase boundaries (APBs).
11-14
The native oxide layer
commonly observed on the Si surface is another major factor hindering the epitaxial growth.
15-17
The use of nanowire geometry for GaAs on Si provides benefits in this integration owing to the
small lateral dimension. Strain due to lattice mismatch can be effectively relieved in the lateral
directions
18-21
and the probability of forming APBs is remarkably reduced. High-quality single-
crystal III-V nanowires grown on Si substrate have been successfully demonstrated in recent
! 51!
years.
15-17, 22-46
A wide range of applications have also been explored including high speed
transistors,
47
tunnel diodes,
48, 49
light emitting diodes (LED),
50-52
room temperature lasers,
53
and
solar cells.
54
3.2 The History of Semiconductor Nanowires Grown on Si and the Undesired Twin
Formation inside Nanowires
The vapor-liquid-solid (VLS) growth method, which was first developed by Wagner and others in
1960s and 1970s,
55
has become the most prevalent approach to grow III-V nanowires on Si
substrates. Au is widely used as the catalyst due to the ease of preparation and low reactivity.
Vertical nanowire arrays have been successfully obtained through Au-catalyzed VLS in molecular
beam epitaxy (MBE)
33, 43, 46
and metalorganic chemical vapor deposition (MOCVD).
15, 30, 32, 34, 35, 45
On the other hand, the use of Au during the growth
56-58
leads to the speculation that deep levels
acting as recombination centers may be created.
59
Group III atom self-catalyzed VLS growth has
thus attracted attention since no foreign metal intermixing is involved in this method.
23-27, 29, 36-39
However, the metal droplet places constraints on the growth of axial heterostructure with different
group III elements. The droplet usually remains after growth due to the properties of VLS growth
and is undesired in some applications such as efficient light extraction in LEDs and good ohmic
contact formation in solar cells. Significant progress has been made with non-VLS methods,
including selective area growth (SAG),
16, 17
and oxide-assisted self-induced growth.
40, 41
In
particular, pioneering work was done by Tomioka et al.
16, 17
using SAG, where the nanowires were
precisely located by a pre-defined array of openings in a dielectric mask and no catalyst was
required. In Tomioka’s work, they used a seven-step approach for the epitaxial growth of GaAs
nanowires on Si, including: (1) ramp-up of the sample temperature, (2) a thermal annealing at
925 °C to remove the native oxide, (3) ramp-down of the sample temperature to 400 °C, (4) surface
! 52!
treatment with only AsH
3
flow at 400 °C, (5) ramp-up of the sample temperature again to growth
temperature, (6) the GaAs growth at 750 °C, and (7) cooling down the sample temperature.
The high uniformity of nanowire morphology allows the fabrication of versatile devices in a
controllable manner. Photodetectors,
60
LEDs,
61
and solar cells
62, 63
have been achieved using
nanowires. Nevertheless, rotational twins and polytypism are common defects observed in both
VLS and non-VLS grown nanowires. These defects are believed to cause altered band structure in
nanowires compared to their bulk counterpart. Carrier scattering, local quantum confinement and
non-intrinsic carrier transition processes are incurred due to the band discontinuity between
zincblende (ZB) and wurtzite (WZ) regions.
64-68
Understanding of the driving force for twin
formation and control of its presence has become a topic of intense research. Substantial progress
in eliminating twins and polytypism has been achieved for the VLS mode with a general agreement
that the thermodynamics at the triple-phase interface line is the dominant factor controlling the
defect formation.
23, 25, 69-77
The understanding of twin formation mechanism in SAG mode is not
fully understood yet, and was interpreted from several different perspectives including faceted
growth
78, 79
and thermodynamics of nucleation.
67, 80
3.3 Facile Five-Step Heteroepitaxial Growth
We successfully used facile five-step approach for heteroepitaxial growth of GaAs nanowires on
Si (111) substrates using SAG. The growth steps are summarized in Fig. 3a. Briefly, after substrate
preparation, the Si (111) substrates are immediately loaded into the MOCVD reactor. Step 1 is to
ramp up the temperature from room temperature to 925 °C for hydrogen annealing. Step 2 is to
keep the temperature stable at 925 °C and anneal the substrates in hydrogen ambient for 5 minutes
to remove native oxides. Step 3 is to ramp down the temperature directly to the desired growth
temperature between 700 and 790 °C, and for the case shown in Fig. 3.1a, the temperature is
! 53!
760 °C. Step 4 is the step when the nanowires grow. Precursors including trimethylgallium (TMG)
and arsine (AsH
3
) are introduced simultaneously at the beginning of the step and the temperature
is kept at the growth temperature. During this step, the nanowires would grow in the vertical GaAs
<111>B direction. After the desired growth time (typically ~1 hour), the final step is to close the
TMG supply and cool down the system while AsH
3
precursor continues to flow until the
temperature is below 300 °C to prevent nanowire decomposition. Fig. 3.1b shows a scanning
electron microscopy (SEM) image taken at a 30° tilted angle of nanowires grown at 760 °C. All
the wires are located uniformly in the pre-defined template and exhibit 6-fold symmetric cross
section consisting of {1-10} sidewall facets. Fig. 3.1c shows a top view SEM image of the
nanowire array. The yield of vertical nanowires is 100%. Compared to the seven-step growth
reported by Tomioka et al.,
16, 17
our five-step growth method could also give 100% yield and
similarly high uniformity of nanowire morphology, while our direct heteroepitaxial growth
without AsH
3
treatment could significantly simplify the growth temperature profile.
Figure 3.1 (a) Temperature profile of GaAs nanowire growth on Si (111). Red and blue blocks indicate the
time AsH
3
and TMG are supplied, respectively. (b) 30° tilted and (c) top view SEM images of uniform
GaAs nanowire arrays grown on Si. (d) Schematic diagram of crystal lattice at GaAs/Si interface, viewed
AsH
3
TMG
Temperature/℃ 925
760
300
RT
Time
GaAs
Si
Ga
As
Si
3.47Å
5/μm 10/nm
a b
c d
e
1μm
! 54!
from [1-10] orientation. (e) HRTEM image taken at the GaAs/Si interface. Arrows indicate sites with misfit
dislocation.
The lattice constants of Si and GaAs are different by 4.1%. GaAs nanowires with small enough
footprint can, however, be grown on Si coherently without generating misfit dislocations (MFDs)
at the interface. Such a MFD-free interface was observed by Tomioka et al. when they grew GaAs
on Si with actual opening diameter of 19 nm.
16
For larger nanowires, the increased strain energy
would lead to MFDs at the interface once the nanowires are beyond a critical size. Previous
calculations based on continuum elasticity have estimated the critical height and diameter of NW
for MFD generation and shown considerable stress relief through MFDs.
20, 81
The nanowires
discussed in this paper fall into this category. As shown in Fig. 3.1d, the distance between two
adjacent As atoms in the [11-2] direction is 3.47 Å and the lattice mismatch is 4.1%, so the period
of MFD is 3.47 Å /0.041 = 8.46 nm. To examine the crystal structure of our nanowires and the
heteroepitaxial interface, we cut a 70 nm thin slice parallel to the {1-10} crystal planes from the
center of a nanowire using focused ion beam (FIB, JEOL). High resolution TEM images were
taken with [1-10] zone axis. High density of twins were observed immediately after the growth
started for nanowires grown at 760 °C in Fig. 3.1e. The size of our nanowires is above the critical
diameter and height for MFD-free growth, so periodic MFDs are observed at the GaAs/Si interface
together with periodic contrast variation on the GaAs side due to strain distribution. The average
period of the MFDs is 8.45 nm, which agrees with the calculated value based on the analysis above.
Our measured period also agrees with the value calculated by Yuan et al. combining molecular-
dynamics and quantum-mechanics simulations.
81
To initiate coherent epitaxial growth of GaAs nanowires on Si, previous studies indicate a
nucleation or surface treatment step is often required prior to the nanowire growth. For the Au-
! 55!
catalyzed VLS growth, Kang et al. found vertical nanowires can only be obtained with high-
temperature nucleation followed by low-temperature nanowire growth.
35
In the case of SAG,
Tomioka et al. pointed out that the Si surface needs to be soaked in As ambient first to form Si
(111):As surface reconstruction, which is an analogue of GaAs or InAs (111)B surface. To prevent
As evaporation from this surface, low temperature surface treatment in AsH
3
was performed prior
to ramping up temperature for the nanowire growth.
16, 17
However, we used direct heteroepitaxial
growth without AsH
3
treatment, which simplified the temperature profile of the growth procedure
to just five steps (Fig. 3.1a). On the contrary, the surface treatment step at lower temperatures only
decreased the vertical nanowire yield. Furthermore, high-yield vertical nanowires could be grown
even on a Si surface exposed to dry etching, showing the robustness of our method. In our study,
the sizes of the mask openings for samples prepared using electron beam lithography (EBL) and
photolithography were around 100 nm and 200 nm, respectively. One possible mechanism that
allows us to grow without low temperature surface treatment could be the relatively large size of
mask openings in our study compared to those reported in other papers.
16, 17, 35
Because the initial
growth adopts Volmer-Weber (island) mode, as will be shown later, abundant sources captured in
large mask openings ensure adequate local supply of adatoms on the nuclei and prevent them from
decomposition.
To better understand the growth behavior at the initial stage, we grew nanowire material using the
five-step growth method for only 2 minutes (defined by the TMG supply time) on both GaAs
(111)B and Si (111) substrates and examined their nuclei morphologies under SEM. The SEM
images show the nuclei formed in the mask openings after growth at 790 °C and 2.14×10
-5
atm
AsH
3
partial pressure (equivalent V/III ratio = 283) on GaAs (111)B substrate (Fig. 3.2a) and on
Si (111) substrate (Fig. 3.2b). The material deposited on the GaAs substrate fills the entire opening
! 56!
and maintains a flat surface. In contrast, the nuclei on the Si substrate appear to be islands with
crystalline facets. The classic models to depict different film growth mechanisms are
82
: layer-by-
layer growth (Frank – van-der-Merve, or FM, mechanism), island growth (Volmer – Weber, or
VW, mechanism), and the Stranski – Krastanov (SK) (an initially continuous film that becomes
islanded but with a thin continuous ‘wetting’ layer left). The interplay among surface energies of
the substrate (γ
s
), the film (γ
f
), and their interface (γ
i
) decides the actual growth mechanism.
79
If
the substrate surface energy is smaller than the sum of the two other surface energies, the island
growth (VW) occurs because the exposed substrate surface is energetically favorable:
γ
s
< γ
f
+ γ
i
,
whereas the reverse inequality:
γ
s
> γ
f
+ γ
i
is associated with layer-by-layer growth (FM).
Figure 3.2 30° tilted SEM images at the mask openings on (a) GaAs (111)B and (b) Si (111) substrates
after 2 minutes growth at 790 ºC. (c) Top view SEM images of nuclei after 2 minutes growth under different
200#nm
670 730 790
2.14e.5 2.14e.4 1.43e.3
Temperature#(#℃)
AsH
3
partial#pressure#(atm)
(111)B
(0.10)
(.1.10) (0.1.1)
(.1.10)
(.100) (00.1)
(111)A
(.101) (10.1)
(.1.12)
(1.21)
(2.1.1)
(.211)
(.12.1)
(11.2)
(111)B
(%1%12)
a
b
c
d e
200#nm
! 57!
temperatures and AsH
3
partial pressures, scale bar 200 nm. TMG partial pressure was kept constant at
7.56×10
-7
atm. (d) Schematic of relevant low-index planes viewed from <111>B direction. (e) Projection
of relevant low-index planes on (1-10) plane in GaAs zincblende lattice.
Our result agrees with early studies that GaAs-on-Si growth occurs in VW mode.
83-85
As-
passivated Si (e.g. Si (111):As surface mentioned earlier) was found to be highly inert (low γ
s
) due
to the existence of As lone-pair states and was even shown to be almost unaffected by exposure to
oxygen or air.
86
Previous studies have also shown that the growth temperature (T
g
) and the AsH
3
partial pressure (P
AsH3
) can affect the shape of grown GaAs crystals during SAG MOCVD.
78, 87
So
we varied T
g
between 670 and 790 °C, and P
AsH3
between 2.14×10
-5
and 1.43×10
-3
atm while
keeping the TMG partial pressure constant at 7.56×10
-7
atm. Fig. 3.2c shows the top view SEM
images of typical nuclei under each growth condition. All the conditions result in VW growth and
a hexagonal cross section emerges after a short time. However, the growth rates along different
orientations show strong dependence on both T
g
and P
AsH3
. In general, lateral growth is suppressed
under lower P
AsH3
or higher T
g
while the vertical growth is enhanced under these conditions. We
believe this is related to the different T
g
and P
AsH3
dependence of the growth rate for different crystal
planes. Fig. 3.2d shows the relevant low-index planes viewed from the <111>B direction and Fig.
3.2e shows their orientations in relation to the zincblende GaAs crystal lattice viewed from the [1-
10] direction. In crystal growth, facets are expressed because of their slow growth rate. At higher
T
g
and lower P
AsH3
(those within the blue background in Fig. 3.2c), the growth in <111>B direction
is faster than in the <1-10> directions so the adatoms migrate to the top of nuclei and contribute to
the vertical growth. The lateral growth is instead suppressed and clear {1-10} sidewall facets can
be seen due to their slow growth. The tilted {-1-10} facets have the same surface atomic
configurations as those of {1-10} sidewalls, but due to adatom accumulation at the top, the growth
in the tilted <-1-10> directions is still faster than that in the vertical <1-10> directions. The tilted
! 58!
{-1-10} facets are not expressed and top surface is usually flat. However, in the extreme case of
790 °C T
g
and 2.14×10
-5
P
AsH3
, the nuclei tend to be pyramids enclosed by tilted {-1-10} facets
meaning the growth rates in those directions are also significantly reduced. In contrast, when T
g
decreases and P
AsH3
increases, the growth rate in <111>B direction decreases and those in the <1-
10> directions increase. The nuclei under these conditions (those within the pink background in
Fig. 3.2c) extend further laterally and show much smaller height than those grown under higher T
g
and lower P
AsH3
. The nuclei grown at 670 °C T
g
and 2.14×10
-3
P
AsH3
are not as tall as the silicon
nitride mask while those grown at 790 °C T
g
and 2.14×10
-5
P
AsH3
are around 100 nm tall. Under low
T
g
and high P
AsH3
conditions, due to the increased growth rates in the <1-10> directions, the
hexagonal cross sections are less obvious and multiple nuclei are observed. Variation of the growth
rates in the <111>B direction is believed to be dominated by the surface reconstruction that is
related to As surface enrichment and is a strong function of T
g
and P
AsH3
. Ultra-high vacuum studies
by various groups found that at low T
g
or high P
AsH3
, a (2×2) reconstruction takes place which
features a chemically stable As-trimer structure leading to reduced growth rate in <111>B
direction, and at high T
g
or low P
AsH3
( 19× 19) reconstruction appears.
88-93
Nishida et al.
reported the same phase transition on GaAs (111)B surface during MOCVD growth.
94
3.4 Twin Formation Growth Model
The growth condition not only changes the nucleation behavior, but also affects the crystal
structure in terms of twin defects. Ikejiri et al. proposed a hypothetical growth model of GaAs
shape evolution in early growth stage during SAG MOCVD.
95
In our work, we have verified
Ikejiri’s growth model and can explain the twin formation mechanism by direct observation of
nanowire morphologies corresponding to each evolution stage. Using our five-step growth method,
the highest nanowire growth temperature was 800 °C. Above this temperature, few GaAs
! 59!
nanowires were observed to grow from the Si
3
N
4
mask opening, and we believe the reason is that
at temperatures above 800 °C, the desorption of the reaction species from the Si
3
N
4
and Si surface
is so rapid that little deposition of GaAs could occur. However, we were able to grow nanowires
at 850 °C by starting the growth at 760 °C for 5 minutes and then quickly ramping up the
temperature to 850 °C, as the growth at 760 °C would lead to GaAs nanowires in the openings,
and then even at 850 °C, the reaction species can wet the GaAs nanowires better than the Si
3
N
4
and Si surface, thus leading to growth even at 850 °C. Yet, as shown in Fig. 3.3a, nanowires
become much less uniform in height probably due to the longer Ga diffusion length at higher
temperature and the increased desorption rate may also influence the non-uniformities observed
during growth at temperatures above 800 °C. Uneven nanowire tops appear with tilted {-1-10}
facets and other low-index facets to various degrees. Some nanowires have a completely pinched-
off tetrahedron tip (enclosed in white circles in Fig 3.3a), while some others have a triangular-
shaped thin mesa surrounded by three tilted {-1-10} facets (enclosed in red circles in Fig. 3.3a).
Based on the various nanowire tip morphologies (Fig. 3.3b), we demonstrate the various growth
fronts and the mechanism driving the twin formation. At a temperature as high as 850 °C, the
growth rate in <-1-10> directions is extremely low. When a nanowire grows in the vertical <111>B
direction, the three tilted {-1-10} facets extend toward the center from three corners at the tip.
Hence, the top (111)B facet is intersected by these three facets and appears to be a triangle (Fig.
3.3c, stage i). This triangle keeps shrinking while the {-1-10} facets further extend (Fig. 3.3c, stage
ii). When the triangle reaches a certain critical dimension, the probability of forming a rotational
twin, upon the deposition of the next layer, would be dramatically increased (Fig 3.3c, stage iii).
The critical dimension of the triangle is related to the change of the Gibbs free energy for the
growth of the next bilayer of the triangular area. In smaller triangles (i.e., larger peripheral/area
! 60!
ratios), the smaller peripheral energy density of the rotated stacking, compared to that of the normal
zinc-blende stacking, makes twin formation energetically more favorable. Mathematical details of
the nucleation-growth model will be provided later in this manuscript. If no twin forms, the tip
will pinch off and the nanowire growth terminates. Otherwise the triangular (111)B facet will stop
shrinking but expand laterally as a mesa instead after a twin is introduced (Fig. 3.3c, stage iv).
78,
79
When a twin is formed at the interface of the nucleus and the substrate, the crystal lattice of the
nucleus can be considered to be rotated by 60° (or 180° given the three-fold symmetry) azimuthally
in relation to that of the substrate. The tilted <-1-10> directions of the substrate are thus no longer
the same slow-grow directions for the nucleus and lateral growth rate in those directions become
considerable. The twinned mesa will fill up all the space above the original {-1-10} facets, making
the nanowire top flat (Fig. 3.3c, stage v to vii). Sometimes another slow-grow facet, (111)A facet,
will appear temporarily before the mesa forms a complete flat hexagonal top (Fig. 3.3c, stage vii).
When the growth proceeds, three {-1-10} facets of the newly grown crystal above the twin will
start to emerge at the three corners different from those where the original {-1-10} facets stem. At
this point, the (111)B facet starts to shrink again and the growth will repeat the whole process
mentioned above. Through this repeating growth sequence, quasiperiodic twins are introduced
during the nanowire growth. We also notice that around the critical size of (111)B triangle, the
twinned mesa is metastable. After the first twin forms, multiple consecutive twins are also highly
likely to take place. If even numbers of the twins form before the mesa eventually stabilizes, the
new stable crystal shares the same lattice orientation as the original crystal underneath the first
twin (TEM image in Fig. 3.3d). Thus, three {-1-10} facets emerge from the edges of the triangle
mesa and would finally pinch off the tip (Fig. 3.3e). Similar to the effect of a single twin, odd
! 61!
number of twins will rotate the lattice by 60° (TEM image in Fig. 3.3f) and allow the stabilized
mesa to expand laterally (Fig. 3.3g).
Figure 3.3 (a) 30º tilted SEM images of nanowires grown at 850 ºC, the initial part was grown at 760 ºC to
help nucleate. White circles indicate nanowires with completely pinched-off tetrahedron tip, while red
circles indicate nanowires with triangular-shaped thin mesa surrounded by three tilted {-1-10} facets (b)
SEM of images of nanowire tips with different morphologies, scale bar 200 nm. Images are organized in a
sequence according to the twin formation model in (c). (c) Schematic of twin formation process during
nanowire growth. (d) TEM image of a region consists of even number of twins. The crystals at two sides
of transitional region share the same atomic registry. (e) SEM image of a nanowire tip. After even number
of twins, the wire pinches off. (f) TEM image of a region consists of odd number of twins. The crystal at
two sides of transitional region can be considered as rotated by 180 º compared to each other.!(g) SEM
image of a nanowire tip with odd number of twins.!
According to the growth model, twin density or length of a twin-free segment is determined by
how soon a flat hexagonal (111)B top surface shrinks to the triangle of critical dimension. In other
words, the critical triangle size and the nanowire diameter together determine the length of a single
twin-free segment. Fig. 3.4a to 3.4c are TEM images of nanowires grown at different temperatures
and the different crystal orientations between two segments separated by a twin is manifested by
! 62!
the brightness contrast. Distinct twin densities indicate strong temperature dependence of the
growth characteristics. For nanowire grown at 760 °C (Fig. 3.4a), twins form every few
monolayers, while twin-free segments can extend longer than 10 nm in nanowires grown at 790 °C
(Fig. 3.4b). The difference is even more striking in Fig. 3.4c where we have the initial part grown
at 760 °C and later part grown at 850 °C. An abrupt interface exists, across which the twin density
changes dramatically. A strong correlation between the growth temperature and the twin density
is demonstrated. The twins in the 760 °C part are so dense that in many cases they form every
monolayer which makes the crystal to be effectively in the wurtzite structure. As will be discussed
in detail later, higher temperature features a smaller critical dimension, while in the case of lower
temperature, the probability of forming a twin is considerable even when the (111)B facet is pretty
large.
! 63!
Figure 3.4 (a)-(c) TEM images of nanowire grown at different temperatures: (a) 760 ºC, (b) 790 ºC and (c)
850 ºC. The initial segment in (c) was grown at 760 ºC in order to nucleate. (d)-(f) TEM images of nanowires
grown at 850 ºC with different diameters: (d) 95 nm, (e) 180 nm and (f) 280 nm. (g) Histogram showing
length distribution of twin free segments of nanowires grown at 850 ºC with different diameters.
Because the probability of forming a twin increases when the tip is around the critical dimension,
larger diameter would also favor fewer twins as it takes longer time for the tip to reach that critical
size. Our observation of twin-free segments on nanowires with different diameters grown at the
same temperature is consistent with the observation by Yoshida et al..
78
Yoshida’s work indicates
that the thickness of the twin-free segments decreases as the nanowire diameter decreases. They
used nanowires with 200 nm and 380 nm diameters and showed that the twin occurrence
probability per monolayer for the 200 nm diameter nanowires is higher than that for 380 nm
diameter nanowires. The largest thickness of twin-free segments is ~42 monolayers, corresponding
! 64!
to ~14 nm in length. In our work, nanowires were all grown at 850 °C with 95, 180 and 280 nm
diameters controlled by the size of the opening in the silicon nitride mask. Fig. 3.4d to 3.4f are the
TEM pictures of these nanowires with different diameters (95 nm for Fig. 3.4d, 180 nm for Fig.
3.4e and 280 nm for Fig. 3.4f). Although they were all grown at the same temperature, the length
distributions of twin-free segments are substantially different. As shown by the statistics in Fig.
3.4g, the length of twin-free segment becomes larger with increasing diameter. The average twin-
free lengths for nanowires of 95, 180 and 280 nm diameters are 17.9, 39.1 and 53.1 nm,
respectively. Gaussian peak fittings are superimposed on the histogram to show that the peaks shift
toward larger twin free length with increasing diameters. The nature of the twin peaks for 180 nm
nanowire is not fully understood at this point. The longest twin-free segment comes from our 280
nm sample, with a twin-free length of 80 nm. This twin-free length of 80 nm in GaAs nanowires
using SAG is a significant improvement compared to previous work.
78
We attribute our
improvement in the twin-free length to the unusually high growth temperature (850 °C) we can
achieve.
Based on the growth model mentioned earlier, the maximum length of a twin-free segment occurs
when the tip is completely pinched off. According to the geometry of the tetrahedron enclosed by
three {-1-10} facets, the maximum length would be 33.6, 63.6 and 99 nm for nanowires of 95, 180
and 280 nm diameters. All the lengths we measured are below the theoretical limit with only one
exception for the 95 nm nanowire as can be seen in Fig. 3.4g (one data point is between 40 and 50
nm). This discrepancy could be due to the fact that the TEM sample might not have been cut right
through the center of the nanowire so the actual diameter is larger than what we observed. We also
assumed zero growth rate in <-1-10> directions, while in the real situation, a finite growth rate
cannot be ruled out.
! 65!
To interpret the effects of growth temperature and diameter, we have calculated the twin-free
segment length probability distribution P(h) using a simple nucleation-growth model.
67
We
considered a hexagonal GaAs nanowire with diameter D, in which a new twin-free segment starts
to grow at height h = 0 to form a tetrahedral island on the (111)B top surface. At height h, the next
GaAs bilayer of area A(h) can grow with either the zinc-blende (zb) or fault (f) stacking.
67
The
change of the Gibbs free energy for the growth of the next bilayer of area A is given by
(λ = zb or f), (1)
where and are the areal and peripheral energy densities of the bilayer and is the
chemical potential of the vapor, which depends on the temperature T in addition to the vapor
pressures of the reactant gases.
67
takes a maximum value, ) (
*
0 *
A G G
λ λ
= , at a critical area
A
*
. We note that decreases with increasing temperature, so the critical area A
*
also
decreases with increasing temperature, leading to increased twin-free segments. The Gibbs free
energy used to obtain the twin statistics is then
!
"
#
>
≤
=
) ) ( (
) ) ( ( )) ( (
) (
*
*
*
0
A h A G
A h A h A G
h G
λ
λ
λ
(λ = zb, or λ = f when
*
) ( A h A > ). (2)
In the case of fault bilayers (where the twin boundaries are, λ = f and
*
) ( A h A ≤ ), we also need to
take into account the twin-twin interaction,
) ( ) ( ) ( ) (
int
f
0
f
h A h h G h G ε + = , (3)
where is the twin-twin interaction energy density.
96
The probability to find a twin at height
h is calculated as
, (4)
G
0
λ
(A)=[ε
area
λ
−µ
gas
(T)]A+ε
edge
λ
A
ε
area
λ
ε
edge
λ
µ
gas
(T)
G
0
λ
(A)
µ
gas
(T)
ε
int
(h)
p(h)=
exp[−G
f
(h)/k
B
T]
exp[−G
f
(h)/k
B
T]+exp[−G
zb
(h)/k
B
T]
! 66!
where k
B
is the Boltzmann constant,
) (
zb
h G
and ) (
f
h G are calculated using (2) and (3),
respectively, as the twins occur at
*
) ( A h A ≤ . The probability for the first occurrence of a stacking
fault at the n-th bilayer is then
, (5)
where Δ is the distance between consecutive bilayers.
Fig. 3.5 shows the calculated P(h) for three diameters, D = 95, 180 and 280 nm at temperatures
T = 850 °C and T = 760 °C. At T = 850 °C, the calculated distribution in Fig. 3.5a exhibits a sharp
peak for every diameter, where the peak position is an increasing function of D. Accordingly, the
average twin-free segment length is 19, 37 and 53 nm for D = 95, 180 and 280 nm, in reasonable
agreement with the experimental observations. In contrast, at a lower T = 760 °C, when the twin-
free segment is expected to be much shorter, the distribution is nearly independent of D and decays
rapidly to zero within the first few bilayers, as can be seen in Fig. 3.5b.
Figure 3.5 Probability distribution of the twin-free segment length in GaAs nanowires of the diameter 95,
180 and 280 nm at temperature (a) 850 °C and (b) 760 °C.
P(nΔ)= [1− p(iΔ)]p(nΔ)
i=1
n−1
∏
! 67!
3.5 Summary!
We have reported facile five-step heteroepitaxial growth of GaAs nanowires on Si (111) substrates
using SAG MOCVD. Highly uniform nanowire arrays could be obtained without low temperature
surface treatment in arsine before nanowire growth. The effects of growth temperature and arsine
partial pressure were carefully studied during the initial nucleation process for GaAs nanowires on
Si. We are able to grow nanowires at unusually high temperature (850 °C) and obtain a variety of
tip morphologies to demonstrate the twin formation mechanism. Twin-free length up to 80 nm in
our GaAs nanowires growth at 850 °C with 280 nm diameter is a significant improvement
compared to previous work. The temperature dependence as well as diameter dependence were
well interpreted by a mechanism of faceted growth and nucleation thermodynamics. The results
deepen our understanding of twin defects frequently observed in nanowires and could open up
great opportunities of heterostructure devices requiring epitaxial III-V material on Si substrate.
Chapter 3 References
1. Huang, Y.; Duan, X.; Cui, Y.; Lauhon, L. J.; Kim, K.-H.; Lieber, C. M. Science 2001, 294, 1313-
1317.
2. Gudiksen, M. S.; Lauhon, L. J.; Wang, J.; Smith, D. C.; Lieber, C. M. Nature 2002, 415, 617-620.
3. Lauhon, L. J.; Gudiksen, M. S.; Wang, D.; Lieber, C. M. Nature 2002, 420, 57-61.
4. Johnson, J. C.; Choi, H.-J.; Knutsen, K. P.; Schaller, R. D.; Yang, P.; Saykally, R. J. Nat. Mater.
2002, 1, 106-110.
5. Parkinson, P.; Lloyd-Hughes, J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Johnston, M. B.; Herz, L. M.
Nano Lett. 2007, 7, 2162-2165.
6. Tang, T.; Han, S.; Jin, W.; Liu, X. L.; Li, C.; Zhang, D. H.; Zhou, C. W.; Chen, B.; Han, J.;
Meyyapan, M. J. Mater. Res. 2004, 19, 423-426.
7. Lei, B.; Li, C.; Zhang, D. H.; Zhou, Q. F.; Shung, K. K.; Zhou, C. W. Appl. Phys. Lett. 2004, 84,
4553-4555.
8. Li, C.; Lei, B.; Zhang, D. H.; Liu, X. L.; Han, S.; Tang, T.; Rouhanizadeh, M.; Hsiai, T.; Zhou, C.
W. Appl. Phys. Lett. 2003, 83, 4014-4016.
! 68!
9. Fang, S. F.; Adomi, K.; Iyer, S.; Morkoç, H.; Zabel, H.; Choi, C.; Otsuka, N. J. Appl. Phys. 1990,
68, R31-R58.
10. Ishiwara, H.; Hoshino, T.; Katahama, H. Materials Chemistry and Physics 1995, 40, 225-229.
11. Cohen, D.; Carter, C. B. Journal of Microscopy 2002, 208, 84-99.
12. Komninou, P.; Stoemenos, J.; Dimitrakopulos, G. P.; Karakostas, T. J. Appl. Phys. 1994, 75, 143-
152.
13. Yu, B. B.; Pchelyakov, O. P. Physics-Uspekhi 2008, 51, 437.
14. Kawanami, H. Sol. Energ. Mat. Sol. Cells 2001, 66, 479-486.
15. Mårtensson, T.; Svensson, C. P. T.; Wacaser, B. A.; Larsson, M. W.; Seifert, W.; Deppert, K.;
Gustafsson, A.; Wallenberg, L. R.; Samuelson, L. Nano Lett. 2004, 4, 1987-1990.
16. Tomioka, K.; Kobayashi, Y.; Motohisa, J.; Hara, S.; Fukui, T. Nanotechnology 2009, 20, 145302.
17. Tomioka, K.; Motohisa, J.; Hara, S.; Fukui, T. Nano Lett. 2008, 8, 3475-3480.
18. Glas, F. Phys. Rev. B, PRB 2006, 74, 121302.
19. Chuang, L. C.; Moewe, M.; Chase, C.; Kobayashi, N. P.; Chang-Hasnain, C.; Crankshaw, S. Appl.
Phys. Lett. 2007, 90, 043115.
20. Sburlan, S.; Dapkus, P. D.; Nakano, A. Appl. Phys. Lett. 2012, 100, 163108.
21. Cirlin, G. E.; Dubrovskii, V. G.; Soshnikov, I. P.; Sibirev, N. V.; Samsonenko, Y. B.; Bouravleuv,
A. D.; Harmand, J. C.; Glas, F. Phys. Status Solidi RRL 2009, 3, 112-114.
22. Hertenberger, S.; Funk, S.; Vizbaras, K.; Yadav, A.; Rudolph, D.; Becker, J.; Bolte, S.; Döblinger,
M.; Bichler, M.; Scarpa, G.; Lugli, P.; Zardo, I.; Finley, J. J.; Amann, M.-C.; Abstreiter, G.; Koblmüller,
G. Appl. Phys. Lett. 2012, 101, 043116.
23. Krogstrup, P.; Popovitz-Biro, R.; Johnson, E.; Madsen, M. H.; Nygård, J.; Shtrikman, H. Nano
Lett. 2010, 10, 4475-4482.
24. Munshi, A. M.; Dheeraj, D. L.; Todorovic, J.; van Helvoort, A. T. J.; Weman, H.; Fimland, B.-O.
J. Cryst. Growth 2013, 372, 163-169.
25. Cirlin, G. E.; Dubrovskii, V. G.; Samsonenko, Y. B.; Bouravleuv, A. D.; Durose, K.; Proskuryakov,
Y. Y.; Mendes, B.; Bowen, L.; Kaliteevski, M. A.; Abram, R. A.; Zeze, D. Phys. Rev. B 2010, 82, 035302.
26. Sandra, J. G.; Jonathan, P. B.; Ray, R. L. Semiconductor Science and Technology 2013, 28, 105025.
27. Jabeen, F.; Grillo, V.; Rubini, S.; Martelli, F. Nanotechnology 2008, 19, 275711.
28. Mandl, B.; Stangl, J.; Mårtensson, T.; Mikkelsen, A.; Eriksson, J.; Karlsson, L. S.; Bauer, G.;
Samuelson, L.; Seifert, W. Nano Lett. 2006, 6, 1817-1821.
29. Paek, J. H.; Nishiwaki, T.; Yamaguchi, M.; Sawaki, N. physica status solidi (c) 2009, 6, 1436-
1440.
! 69!
30. Bao, X.-Y.; Soci, C.; Susac, D.; Bratvold, J.; Aplin, D. P. R.; Wei, W.; Chen, C.-Y.; Dayeh, S. A.;
Kavanagh, K. L.; Wang, D. Nano Lett. 2008, 8, 3755-3760.
31. Wei, W.; Bao, X.-Y.; Soci, C.; Ding, Y.; Wang, Z.-L.; Wang, D. Nano Lett. 2009, 9, 2926-2934.
32. Roest, A. L.; Verheijen, M. A.; Wunnicke, O.; Serafin, S.; Wondergem, H.; Bakkers, E. P. A. M.
Nanotechnology 2006, 17, S271.
33. Ihn, S.-G.; Song, J.-I.; Kim, Y.-H.; Lee, J. Y. Appl. Phys. Lett. 2006, 89, 053106.
34. Huang, H.; Ren, X.; Ye, X.; Guo, J.; Wang, Q.; Yang, Y.; Cai, S.; Huang, Y. Nano Lett. 2009, 10,
64-68.
35. Kang, J. H.; Gao, Q.; Joyce, H. J.; Tan, H. H.; Jagadish, C.; Kim, Y.; Choi, D. Y.; Guo, Y.; Xu, H.;
Zou, J.; Fickenscher, M. A.; Smith, L. M.; Jackson, H. E.; Yarrison-Rice, J. M. Nanotechnology 2010, 21,
035604.
36. Plissard, S.; Dick, K. A.; Wallart, X.; Caroff, P. Appl. Phys. Lett. 2010, 96, 121901.
37. Munshi, A. M.; Dheeraj, D. L.; Fauske, V. T.; Kim, D. C.; Huh, J.; Reinertsen, J. F.; Ahtapodov,
L.; Lee, K. D.; Heidari, B.; van Helvoort, A. T. J.; Fimland, B. O.; Weman, H. Nano Lett. 2014, 14, 960-
966.
38. Plissard, S.; Dick, K., A.; Larrieu, G.; Godey, S.; Addad, A.; Wallart, X.; Caroff, P.
Nanotechnology 2010, 21, 385602.
39. Kwoen, J.; Watanabe, K.; Iwamoto, S.; Arakawa, Y. J. Cryst. Growth 2013, 378, 562-565.
40. Koblmüller, G.; Hertenberger, S.; Vizbaras, K.; Bichler, M.; Bao, F.; Zhang, J. P.; Abstreiter, G.
Nanotechnology 2010, 21, 365602.
41. Rudolph, D.; Hertenberger, S.; Bolte, S.; Paosangthong, W.; Spirkoska, D. e.; Döblinger, M.;
Bichler, M.; Finley, J. J.; Abstreiter, G.; Koblmüller, G. Nano Lett. 2011, 11, 3848-3854.
42. Tateno, K.; Hibino, H.; Gotoh, H.; Nakano, H. Appl. Phys. Lett. 2006, 89, 033114.
43. Ihn, S.-G.; Song, J.-I.; Kim, T.-W.; Leem, D.-S.; Lee, T.; Lee, S.-G.; Koh, E. K.; Song, K. Nano
Lett. 2007, 7, 39-44.
44. Mårtensson, T.; Wagner, J. B.; Hilner, E.; Mikkelsen, A.; Thelander, C.; Stangl, J.; Ohlsson, B. J.;
Gustafsson, A.; Lundgren, E.; Samuelson, L.; Seifert, W. Advanced Materials 2007, 19, 1801-1806.
45. Mohseni, P. K.; Maunders, C.; Botton, G. A.; LaPierre, R. R. Nanotechnology 2007, 18, 445304.
46. Paek, J. H.; Nishiwaki, T.; Yamaguchi, M.; Sawaki, N. physica status solidi (c) 2008, 5, 2740-
2742.
47. Tomioka, K.; Yoshimura, M.; Fukui, T. Nature 2012, 488, 189-192.
48. Tomioka, K.; Fukui, T. Appl. Phys. Lett. 2011, 98, 083114.
49. Yang, T.; Hertenberger, S.; Morkotter, S.; Abstreiter, G.; Koblmuller, G. Appl. Phys. Lett. 2012,
101, 233102.
! 70!
50. Svensson, C. P. T.; Mårtensson, T.; Trägårdh, J.; Larsson, C.; Rask, M.; Hessman, D.; Samuelson,
L.; Ohlsson, J. Nanotechnology 2008, 19, 305201.
51. Guo, W.; Zhang, M.; Banerjee, A.; Bhattacharya, P. Nano Lett. 2010, 10, 3355-3359.
52. Tomioka, K.; Motohisa, J.; Hara, S.; Hiruma, K.; Fukui, T. Nano Lett. 2010, 10, 1639-1644.
53. Mayer, B.; Rudolph, D.; Schnell, J.; Morkötter, S.; Winnerl, J.; Treu, J.; Müller, K.; Bracher, G.;
Abstreiter, G.; Koblmüller, G.; Finley, J. J. Nat. Commun. 2013, 4, 2931.
54. Heurlin, M.; Wickert, P.; Fält, S.; Borgström, M. T.; Deppert, K.; Samuelson, L.; Magnusson, M.
H. Nano Lett. 2011, 11, 2028-2031.
55. Wagner, R. S.; Ellis, W. C. Appl. Phys. Lett. 1964, 4, 89-90.
56. Bar-Sadan, M.; Barthel, J.; Shtrikman, H.; Houben, L. Nano Lett. 2012, 12, 2352-2356.
57. Perea, D. E.; Allen, J. E.; May, S. J.; Wessels, B. W.; Seidman, D. N.; Lauhon, L. J. Nano Lett.
2005, 6, 181-185.
58. Allen, J. E.; Hemesath, E. R.; Perea, D. E.; Lensch-Falk, J. L.; LiZ.Y; Yin, F.; Gass, M. H.; Wang,
P.; Bleloch, A. L.; Palmer, R. E.; Lauhon, L. J. Nat. Nanotechnol. 2008, 3, 168-173.
59. Breuer, S.; Pfüller, C.; Flissikowski, T.; Brandt, O.; Grahn, H. T.; Geelhaar, L.; Riechert, H. Nano
Lett. 2011, 11, 1276-1279.
60. Seyedi, M. A.; Yao, M.; O'Brien, J.; Wang, S. Y.; Dapkus, P. D. Appl. Phys. Lett. 2013, 103,
251109.
61. Fukui, T.; Yoshimura, M.; Nakai, E.; Tomioka, K. AMBIO 2012, 41, 119-124.
62. Mariani, G.; Scofield, A. C.; Hung, C.-H.; Huffaker, D. L. Nat. Commun. 2013, 4, 1497.
63. Yao, M.; Huang, N.; Cong, S.; Chi, C.-Y.; Seyedi, M. A.; Lin, Y.-T.; Cao, Y.; Povinelli, M. L.;
Dapkus, P. D.; Zhou, C. Nano Lett. 2014, 14, 3293-3303.
64. Parkinson, P.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Zhang, X.; Zou, J.; Jagadish, C.; Herz, L. M.;
Johnston, M. B. Nano Lett. 2009, 9, 3349-3353.
65. Spirkoska, D.; Arbiol, J.; Gustafsson, A.; Conesa-Boj, S.; Glas, F.; Zardo, I.; Heigoldt, M.; Gass,
M. H.; Bleloch, A. L.; Estrade, S.; Kaniber, M.; Rossler, J.; Peiro, F.; Morante, J. R.; Abstreiter, G.;
Samuelson, L.; Fontcuberta i Morral, A. Phys. Rev. B 2009, 80, 245325.
66. Thelander, C.; Caroff, P.; Plissard, S. b.; Dey, A. W.; Dick, K. A. Nano Lett. 2011, 11, 2424-2429.
67. Yuan, Z.; Nomura, K.-i.; Nakano, A. Appl. Phys. Lett. 2012, 100, 163103.
68. Ikejiri, K.; Kitauchi, Y.; Tomioka, K.; Motohisa, J.; Fukui, T. Nano Lett. 2011, 11, 4314-4318.
69. Algra, R. E.; Verheijen, M. A.; Borgstrom, M. T.; Feiner, L.-F.; Immink, G.; van Enckevort, W. J.
P.; Vlieg, E.; Bakkers, E. P. A. M. Nature 2008, 456, 369-372.
70. Algra, R. E.; Verheijen, M. A.; Feiner, L.-F.; Immink, G. G. W.; Enckevort, W. J. P. v.; Vlieg, E.;
Bakkers, E. P. A. M. Nano Lett. 2011, 11, 1259-1264.
! 71!
71. Caroff, P.; Dick, K. A.; Johansson, J.; Messing, M. E.; Deppert, K.; Samuelson, L. Nat.
Nanotechnol. 2009, 4, 50-55.
72. Dick, K. A.; Bolinsson, J.; Messing, M. E.; Lehmann, S.; Johansson, J.; Caroff, P. Journal of
Vacuum Science & Technology B 2011, 29, 04D103.
73. Johansson, J.; Karlsson, L. S.; Patrik T. Svensson, C.; Martensson, T.; Wacaser, B. A.; Deppert,
K.; Samuelson, L.; Seifert, W. Nat. Mater. 2006, 5, 574-580.
74. Johansson, J.; Dick, K. A.; Caroff, P.; Messing, M. E.; Bolinsson, J.; Deppert, K.; Samuelson, L.
Journal of Physical Chemistry C 2010, 114, 3837-3842.
75. Joyce, H. J.; Wong-Leung, J.; Gao, Q.; Tan, H. H.; Jagadish, C. Nano Lett. 2010, 10, 908-915.
76. Glas, F.; Harmand, J.-C.; Patriarche, G. Phys. Rev. Lett. 2007, 99, 146101.
77. Dubrovskii, V. G.; Sibirev, N. V.; Harmand, J. C.; Glas, F. Phys. Rev. B 2008, 78, 235301.
78. Yoshida, H.; Ikejiri, K.; Sato, T.; Hara, S.; Hiruma, K.; Motohisa, J.; Fukui, T. J. Cryst. Growth
2009, 312, 52-57.
79. Chi, C.-Y.; Chang, C.-C.; Hu, S.; Yeh, T.-W.; Cronin, S. B.; Dapkus, P. D. Nano Lett. 2013, 13,
2506-2515.
80. Shapiro, J. N.; Lin, A.; Ratsch, C.; Huffaker, D. L. Nanotechnology 2013, 24, 475601.
81. Yuan, Z.; Shimamura, K.; Shimojo, F.; Nakano, A. J. Appl. Phys. 2013, 114, 074316.
82. Tournié, E.; Ploog, K. H. J. Cryst. Growth 1994, 135, 97-112.
83. Zinke#Allmang, M.; Feldman, L. C.; Nakahara, S. Appl. Phys. Lett. 1988, 52, 144-146.
84. Biegelsen, D. K.; Ponce, F. A.; Smith, A. J.; Tramontana, J. C. J. Appl. Phys. 1987, 61, 1856-1859.
85. Hull, R.; Fischer#Colbrie, A.; Rosner, S. J.; Koch, S. M.; Harris, J. S. Appl. Phys. Lett. 1987, 51,
1723-1725.
86. Uhrberg, R. I. G.; Bringans, R. D.; Olmstead, M. A.; Bachrach, R. Z.; Northrup, J. E. Phys. Rev. B
1987, 35, 3945-3951.
87. Ando, S.; Kobayashi, N.; Ando, H. J. Cryst. Growth 1994, 145, 302-307.
88. Biegelsen, D. K.; Bringans, R. D.; Northrup, J. E.; Swartz, L. E. Phys. Rev. Lett. 1990, 65, 452-
455.
89. Thornton, J. M. C.; Woolf, D. A.; Weightman, P. Applied Surface Science 1998, 123–124, 115-
119.
90. Rajkumar, K. C.; Chen, P.; Madhukar, A. J. Appl. Phys. 1991, 69, 2219-2223.
91. Woolf, D. A.; Westwood, D. I.; Williams, R. H. Appl. Phys. Lett. 1993, 62, 1370-1372.
92. Avery, A. R.; Tok, E. S.; Jones, T. S. Surface Science 1997, 376, L397-L402.
93. Chen, P.; Rajkumar, K. C.; Madhukar, A. Appl. Phys. Lett. 1991, 58, 1771-1773.
94. Nishida, T.; Uwai, K.; Kobayashi, Y.; Kobayashi, N. Jpn. J. Appl. Phys. 1995, 34, 6326.
! 72!
95. Ikejiri, K.; Sato, T.; Yoshida, H.; Hiruma, K.; Motohisa, J.; Hara, S.; Fukui, T. Nanotechnology
2008, 19, 265604.
96. Yuan, Z.; Nakano, A. Nano Lett. 2013, 13, 4925-4930.
! 73!
Chapter 4 Carbon Nanotube Macroelectronics for Active Matrix
Polymer-Dispersed Liquid Crystal Displays
4.1 The Transparent Displays and Their Driving Transistors
Transparent display is an emerging technology which offers texts and images in front of the
background while enabling people to see the scenery in the back of the panel.
1-3
It enhances the
observers’ view of world and gives people experience for better life. Furthermore, lightweight,
energy saving, and other advantages make transparent display a hot area of research.
2, 4-7
Polymer-
dispersed liquid crystal (PDLC) and organic light emitting diodes (OLEDs) are the two major
candidates for transparent display applications. In particular, with no polarizers, PDLC is the most
promising technology due to its simple structure, good chemically stability and non-emissive
property.
2, 6, 8
Besides, its mechanical flexibility makes flexible transparent display possible. Rigid
and flexible PDLC transparent displays have been demonstrated driven by thin film transistors
(TFTs) with amorphous Si (a-Si)
2, 6
and organic thin film
9-11
channel materials, respectively.
However, there are still remaining issues with these transistor channel materials. Amorphous
silicon is not intrinsically flexible or transparent. Although organic materials are flexible, they are
easy to degrade upon exposure to oxygen and moisture, and thus sophisticated passivation
techniques are necessary. In addition, the low mobilities (~ 1 cm
2
V
-1
S
-1
) of both a-Si
12
and organic
materials
13
are not preferred for large area display. It is thus desirable to figure out transistor
channel materials which are not only intrinsically flexible and transparent, but also possess good
air stability and high mobility for flexible transparent display applications. Carbon nanotubes
(CNTs), as promising channel materials, can meet all the aforementioned requirements.
14
Flexible
high-performance carbon nanotube integrated circuits have already been demonstrated.
15, 16
With
! 74!
the progress in carbon nanotube growth,
17-20
purification,
21-23
characterization,
24
and fabrication
techniques,
25-29
carbon nanotube thin film transistors (CNT-TFTs) are becoming more and more
promising for transparent displays. Active matrix organic light emitting diodes (AMOLEDs)
driven by CNT-TFTs have been reported by various groups including our own.
30-32
On the other
hand, PDLC display driven by CNT-TFTs has not been reported yet.
4.2 Monolithically Integrated Active Matrix PDLC Displays with CNT-TFT Based Control
Circuits
The schematic of a CNT-TFT-driven seven-segment PDLC display is depicted in Fig. 4.1a. Each
PDLC display pixel has a CNT driver transistor. In each transistor, the source is connected to the
data signal, which is the alternating current (AC) voltage required to drive the PDLC display. The
gate of the driver transistor is connected to the scan signal, which turns the PDLC pixel on and off.
The structural details of the PDLC display are depicted in a cross-sectional image in Fig. 4.1b.
The side view of the schematic of individual back gated CNT-TFTs is shown in Fig. 4.1c, and the
scanning electron microscope (SEM) images show the uniform CNT network in the channel area
in Fig. 4.1d. The CNT-TFTs start from a blank glass substrate. The back gate is e-beam evaporated
Ti/Au. Atomic layer deposition (ALD) Al
2
O
3
acts as the high κ gate dielectric. Another thin layer
of SiO
2
on top of Al
2
O
3
can improve carbon nanotube adhesion, as reported previously.
25, 31
The
channel material is the high semiconducting ratio random CNT network. Ti/Pd source (S) and
drain (D) complete the transistor structure. We notice that a passivation layer of parylene is
necessary to avoid performance degradation of the transistor in the integrated structure.
33
! 75!
Figure 4.1. (a) Schematic of the proposed seven-segment CNT-TFT driven PDLC display. (b) The cross-
sectional view of the CNT-TFT driven PDLC display, showing the structural details. (c) The side view of
the schematic of individual back gated CNT-TFTs. (d) SEM images of the CNT network. Scale bar 2 µm.
Inset, scale bar 500 nm.
For PDLC pixels, the PDLCs are sandwiched between two transparent indium tin oxide (ITO)
electrodes. The two electrodes face each other and their overlapping area defines the active region.
To connect the PDLC pixel to the driving transistor, the pixel’s bottom ITO electrode overlaps
with the drain of the transistor. Depending on the ON or OFF state of the PDLC pixel, the active
region allows or blocks the passing through of the incoming light, exhibiting transparent and
opaque states, respectively.
PDLCs are composite materials which take advantages of both the solid polymers and fluid liquid
crystals. Solid-like PDLC films are formed by embedding micro-sized liquid crystals into polymer
! 76!
matrices. No polarizers are required for PDLC displays.
34
Instead, when an electric field is applied,
the liquid crystal droplets are aligned in the same orientation, where the refractive indices of the
liquid crystal droplets and the polymer matrix are the same. The incoming light can pass through
PDLC film and the transparent indium tin oxide (ITO) electrodes in a straight line, giving the
display a transparent, ON state (Fig. 4.2a), and people call it transparent state. On the other hand,
when no electric field is applied, the liquid crystal droplets are randomly aligned and there are
significant differences in refractive indices between the liquid crystal droplets and the polymer
matrix. The incoming light is randomly scattered and the display is in an opaque, OFF state (Fig.
4.2b), and people call it opaque or scattering state. In some situations, even though the transmission
for the randomly scattered light is not low enough to make the display totally opaque, people still
call it opaque state to distinguish from the transparent state. Without the need for polarizers,
PDLCs enjoy a wide range of advantages, such as reduced weight, high transparency, low power
consumption and good flexibility. For the CNT-TFT driven PDLC concept to work, we need to
study and characterize its components individually, i.e., the PDLC transmission properties and the
CNT driver transistor behavior. Below, we will study these components in detail.
! 77!
Figure 4.2 (a) Transparent state and (b) Opaque state of the PDLC pixel. When electric field is applied
to the electrodes, liquid crystals align with the electric field permitting the incoming light to pass through
in (a). When no electric field is applied, liquid crystals are randomly aligned and incoming light is scattered,
giving opaque state in (b).
Norland optical adhesive 65 (NOA65) has been widely used as the fast photocurable pre-
polymer for PDLC. Here, we choose NOA65 (Norland Co.) and E7 (Merck) as the pre-polymer
and nematic liquid crystal for our PDLC, respectively. In order to study the transmission properties
of PDLC, we fabricated 1.5 cm× 1.5 cm PDLC test cell without the driving transistor. The NOA65
and E7 were mixed in weight ratio of 3:7 and then injected between two ITO-coated glasses 7 µm
apart with the separation defined by transparent spacers (Magsphere Co.). The structure was
subsequently cured by UV irradiation (365 nm wavelength) at 6.0 mJ/(cm
2
⋅s) for 18 minutes. After
fabrication, the spectra of PDLC transmittance were measured by UV/Vis spectroscopy (Cary
5000, Varian). Fig. 4.3a shows the raw (blue) and normalized (red) transmittance of the PDLC cell
versus the applied AC voltage amplitude (A) at 565 nm wavelength and 400 Hz frequency.
Previous study shows that the ON state PDLC transmittance is the highest at this frequency.
35
The
raw transmittance data is the transmittance of whole glass-ITO-PDLC-ITO-glass test cell while
! 78!
the normalized one only accounts for the PDLC transmittance by dividing the raw transmittance
by the bare glass-ITO-ITO-glass structure transmittance. As we can see, with zero electric field,
the normalized transmittance is 2.0%. With increasing AC amplitudes, the transmittance increases
correspondingly and begins to saturate around 9 V. At 9 V, the normalized transmittance is 80%,
meaning the contrast ratio is about 40. The performance of our PDLC cell is similar to previous
low operating voltage PDLC study
2
. According to the analysis, we choose A = 9 V as our data
voltage in this study. Furthermore, the transmittance curves versus wavelength for the transparent
state (pink, A = 9 V) and the opaque (black, A = 0 V) are plotted in Fig. 4.3b. The difference in
transmittance between the transparent and opaque states are clearly demonstrated in the visible
spectrum. We also studied the effect of signal frequency on PDLC transmission. The transmittance
of PDLC versus AC voltage frequency at A = 9 V and 565 nm wavelength is plotted in Fig. 4.3c.
At 400 Hz, the PDLC transmittance is the highest, which also agrees with literature
35
and we
choose this frequency as the AC driving frequency in this paper. Fig. 4.3d shows the PDLC test
cell with printed text in the background in the transparent (left) and opaque, (right) states. As for
the transparent state, one can easily see the printed text through the PDLC cell, while for the
opaque state, the light is scattered by the PDLC and the printed text can hardly be seen. The 2.0%
transmittance in the opaque state causes slight color variations in the center and the edge of the
PDLC cell, and the opaque state transmittance is similar to previous low operating voltage PDLC
study.
2
On the other hand, the high contrast between transparent and opaque states proves that our
PDLCs are suitable for CNT-TFT driven PDLC display study.
! 79!
Figure 4.3 PDLC transmission properties. (a) The raw and normalized transmittance of the PDLC cell at
565 nm wavelength and 400 Hz frequency versus AC voltage amplitude. (b) Transmittance of transparent
(9 V) and opaque (0 V) states versus wavelength. (c) The transmittance of PDLC versus AC voltage
frequency at 9 V amplitude and 565 nm wavelength (d) Transparent (left) and opaque (right) states in front
of printed text. In the transparent state, one can clearly see the printed text through the PDLC cell, while in
the opaque state, the text can hardly be seen. Scale bars 1 cm.
In order to gain more insight into the final active matrix PDLC display, we fabricated and
characterized the electrical performance of the CNT driving transistors. Individual back gated
devices with various dimensions, including the same dimensions (L = 20 µm, W = 2,400 µm) as
in the final structure were fabricated. High semiconducting purity (>99.9% semiconducting single
wall carbon nanotubes, Nanointegris) CNT network was used as the channel. Detailed fabrication
steps are as follows:
! 80!
1.! The blank glass slide was first cleaned by oxygen plasma at RF power 100 W and O
2
pressure 150 mT for 5 mins.
2.! The back gates were patterned by photolithography. 5 nm Ti and 35 nm Au were
sequentially deposited followed by liftoff.
3.! 90 nm of ALD Al
2
O
3
and 10 nm e-beam evaporated SiO
2
were deposited onto the whole
surface.
4.! The sample went through another photolithography to define the exposed area of the
bottom gate contact pads. The sample was dipped into 7:1 buffered oxide etch for 90 s to
remove the dielectric above the pads.
5.! High semiconducting purity (>99.9% semiconducting single wall carbon nanotubes,
Nanointegris) CNT solution was drop casted onto the sample for 2 mins to form
semiconducting CNT network. After incubation, excess CNT solution was washed away
by toluene and the sample was baked at 200 degrees for 1 hour.
6.! The source and drain electrodes were defined by another round of photolithography
followed by 1 nm Ti/50 nm Pd deposition and subsequent liftoff.
7.! Unwanted CNTs outside the channel area were etched away by oxygen plasma at 100 W
and 150 mT for 80 s and concludes the fabricaiton
The passivation layer is critical as direct contact of PDLCs with the transistors in the integrated
structure can degrade transistor performance. Note that we used parylene to passivate the
transistors and this passivation layer does not affect the transistor performance significantly, which
! 81!
agrees with previous studies.
33
The transfer characteristics (I
DS
− V
GS
) before (Fig. 4.4a) and after
(Fig. 4.4b) parylene passivation of a common back gate CNT-TFT transistor on Si substrate with
90 nm SiO
2
gate dielectric and L = 20 µm, W = 200 µm are shown in Fig. 4.4. The V
DS
is kept
constant at −1 V. The on/off ratio only decreases slightly from 4.51×10
6
to 2.14×10
5
, the mobility
even increases from 27.64 cm
2
/(V⋅S) to 35.72 cm
2
/(V⋅S). The performance does not degrade
noticeably. The transfer characteristics (I
DS
− V
GS
) before (Fig. 4.5a) and after (Fig. 4.5b) parylene
passivation of an individual back gate CNT-TFT transistor on glass substrate with ALD Al
2
O
3
gate
dielectric and L = 20 µm, W = 2400 µm under the same measurement condition are shown in Fig.
4.5. Similarly, the device shows good on/off ratio as well as mobility after parylene passivation.
The on/off ratio only decreases slightly from 4.11×10
5
to 3.09×10
5
, the mobility increases from
7.68 cm
2
/(V⋅S) to 9.59 cm
2
/(V⋅S). The results show that the parylene passivation technique is good
enough for final integration with PDLC display.
Figure 4.4 Transfer characteristics of a common back gate CNT-TFT transistor with 90 nm SiO
2
gate
dielectric and L = 20 µm, W = 200 µm (a) before and (b) after parylene passivation.
! 82!
Figure 4.5 Transfer characteristics of a typical individual back gate CNT-TFT on glass substrate with L =
20 µm, W = 200 µm (a) before and (b) after parylene passivation. The behaviors before and after parylene
passivation are similar.
Furthermore, the transfer characteristics (I
DS
− V
GS
) of the same individual back gate CNT-TFT on
glass substrate after parylene passivation is replotted in the red curve in Fig. 4.6a and the
corresponding transconductance (g
m
− V
GS
) is also plotted (blue) in the same graph. Fig. 4.6b shows
the output (I
DS
− V
DS
) characteristics of the same transistor with V
GS
varying from -10 to 10 V with
4 V steps. The output curve shows the good saturation behavior and Ohmic contacts of the
transistor. The analysis shows the good field-effect transistor behaviors of our CNT-TFTs. We
also note that CNT-TFTs demonstrate mobility, typically ~10 cm
2
/(V⋅S), higher than both a-Si and
organic transistors.
12,13
! 83!
Figure 4.6 (a) Transfer (I
D
-V
G
) characteristics (red) and transconductance g
m
-V
GS
characteristics (blue) of
a typical CNT-TFT (L = 20 µm, W = 2400 µm) with V
DS
= -1 V. (b) Output (I
DS
-V
DS
) characteristics of the
same device with V
G
varying from -10 to 10V in 4 V steps.
To further understand the behavior of the CNT-TFT controlled PDLC display, we connected the
PDLC test cell to the CNT driving transistor externally via wire bonding. The active area of the
PDLC test cell was modified to 2 mm × 2 mm to match the PDLC pixel area in the final seven
segment display. The schematic of this testing circuit is shown in Fig. 4.7a, and the voltage across
the PDLC test cell (V
PDLC
) is also indicated. From previous typical direct current (DC)
characteristic of the CNT-TFT, we see that the ON state resistance of typical driving transistor is
around 10 kΩ. Assuming the dielectric constant of the PDLC is 20, the corresponding capacitance
is calculated to be ~ 0.1 nF. At 400 Hz driving frequency, the impedance of the PDLC is 4.0 MΩ,
which is ~ 400 times larger than the impedance of the driving transistor, indicating the latter has
very little effect on the PDLC performance. On the other hand, the OFF state impedance of the
driving transistor is at least 200 MΩ, which is 50 times larger than the impedance of the PDLC.
To confirm our estimations, we measured the waveforms of V
PDLC
with the data electrode fixed at
400 Hz sine wave with the amplitude V
DATA
= 9 V. The waveforms of V
PDLC
under various V
SCAN
! 84!
= -10 V, -5 V, 0 V, 5 V and 10 V are plotted in Fig. 4.7b. The positive peaks for the corresponding
waveforms are 98.3%, 76.0%, 63.1%, 14.7% and 1.13% of V
DATA
, respectively, whereas the
negative peaks are 96.7%, 54.3%, 35.9% 3.49% and 0.33% of V
DATA
, respectively. At V
SCAN
=10
V, the maximum peak of V
PDLC
is 0.10 V, and the pixel remains in the opaque state, while at V
SCAN
= -10 V, the maximum peak value becomes 8.86 V, and the pixel is in the transparent state. The
obvious change in the transmission of the PDLC cell can be visually seen in response with the
change in V
SCAN
against blue background in Fig. 4.7c. The optical photographs represent the blue
background without the presence of PDLC test cell as well the PDLC test cell in front of the blue
background under various V
SCAN
voltages of -10, -5, 0, 5, and 10 V, respectively. The contrast
demonstrates that the PDLC test cell can be fully turned on and turned off at V
SCAN
voltages of -10
and 10 V and these voltages are chosen as the ON and OFF voltages of the scan lines of the PDLC.
Figure 4.7 (a)
Schematic of CNT-TFT driven PDLC test cell. The voltage across the PDLC test cell (V
PDLC
)
is indicated. (b) The waveforms of V
PDLC
under various V
SCAN
= -10 V, -5 V, 0 V, 5 V and 10 V with the
data electrode fixed at 400 Hz sine wave and the amplitude V
DATA
= 9 V. (c) Optical
photographs showing
the background blue color without PDLC test cell and the color through PDLC test cell with the driving
! 85!
CNT-TFT under different V
SCAN
. The CNT-TFT can fully turn on and turn off the transistor. Scale bar 1
mm.
The final accomplishment was to integrate the essential components and demonstrate active matrix
seven segment PDLC display. First, seven segment back panel CNT-TFT control circuit was
fabricated following the same procedures as the individual back gated devices described
previously. 200 nm radio-frequency (RF) sputtered ITO connects to the drain of CNT-TFTs and
defines the active pixel area of the PDLC. After depositing parylene passivation layer throughout
the whole structure, the parylene above the ITO areas and the probing pads were
photolithographically patterned and subsequently removed by oxygen plasma. The PDLC mixture
was prepared and assembled with the back panel following the same recipe as the PDLC test cell.
The optical image of the CNT-TFT back panel before integration is shown in Fig. 4.8a. The inset
shows the zoomed-in SEM image of the details of a CNT-TFT. The red dotted region indicates
where the CNT film exists. Figure 4b is the optical image of the final structure after integration.
The details of the final structure are what we proposed earlier in Fig. 4.1b. The relative positions
of the ITO on glass top top electrode, the CNT-TFT back panel, and the gap in between where
PDLC was injected are also indicated in Fig. 4.8b. The vertical alignment marker in the left part
of the back panel was patterned together with the source and drain of the transistor but does not
play any role in the circuit. The top ITO electrode sticks out to the left of the back panel, where it
connects to the ground in the circuit. In our back panel, the CNT-TFTs driving the seven-segment
PDLC display demonstrate an average mobility of 13.23 cm
2
/(V⋅s). The histogram of the mobilities
of these seven CNT-TFTs is plotted in Fig. 4.8c. The mobility is similar to previous publication.
25
In PDLC display, the current passing through the driving transistor charges and discharges the
capacitance of the PDLC pixel, so the on/off ratio is another important transistor parameter. Fig.
! 86!
4.8d shows the distribution of the on/off ratio of the CNT-TFTs after passivation in the seven-
segment display. As we can see, all the transistors demonstrate on/off ratio larger than 5× 10
4
with
an average of 1.28 × 10
6
,
and the champion device has an on/off ratio of 4.3× 10
6
. The high on/off
ratio and the uniformity in device performance guarantee the control circuits can fully turn off the
PDLC pixel.
Fig. 4.8 (a)
The optical image of the CNT-TFT back panel before integration. The data and scan signals
given to each electrode are also indicated. Inset: the SEM image of the CNT-TFT. The dotted region
! 87!
indicates where the CNT film exists. The electrode which will be connected to PDLC pixel with the voltage
V
PDLC
in the later integrated circuit is also shown. (b) The optical image of the seven segment display after
PDLC integration. The CNT-TFT back panel, the glass with ITO top electrode, and the gap where PDLC
is injected are also indicated. (c) Histogram of the mobilities of each CNT-TFT driving the seven segment
display.
(d) Histogram of the on/off of each CNT-TFT driving the seven segment display. (e-g)
Optical
images of the seven segment display against (e) blue, (f) green, and (g) red backgrounds, displaying digits
“1”, “2”, and “3”, respectively.
In our PDLC displays, both the scan lines and the data lines are individually controlled for each
seven segment PDLC pixel, so that each pixel can be individually addressed using the CNT-TFTs
at different data line voltages (0 to 9 V) with different scan line voltages (-10 to 10 V). Fig. 4.8e,
f and g show the optical images of the seven-segment PDLC display with numerical digits “1”,
“2” and “3” against blue, green and red backgrounds, respectively. While the areas outside the
seven PDLC pixels remain opaque, only the seven pixels can act as the active areas and switch
between transparent and opaque states depending on the states of their driving CNT-TFTs. One
can observe different background colors through the pixels when they are transparent. The PDLC
display can show different and easily recognizable numerical digits against different backgrounds.
Our CNT-TFT driven PDLC displays can serve as a starting point for future improvements and
developments of flexible and high-performance transparent displays. For example, the rigid
substrates and the metal pads can be replaced by bendable materials to enable flexible features.
ITO or other transparent conducting materials can be used as the electrodes of the CNT-TFTs to
make the structure fully transparent, etc. Our work demonstrates the great potential of
semiconductor-enriched CNT-TFTs in flexible transparent display applications.
4.3 Summary
The work presents the great potential of using CNT-TFTs for transparent display electronics. We
! 88!
successfully monolithically integrated active matrix PDLC seven-segment display with CNT-TFT
based control circuits. First, the CNT-TFTs were fabricated using polymer sorted carbon
nanotubes and show good electrical performance with average device mobility of 13.23 cm
2
/(V⋅s)
and on/off ratio of 1.29 ×10
6
. In addition, the PDLC transmission properties were well studied.
More importantly, we have successfully monolithically integrated the CNT-TFTs and the PDLC
pixels to demonstrate fully functioning seven-segment displays. The demonstration takes
advantages of high mobility, high on/off CNT-TFTs to fully control PDLC pixels in active matrix
display. We envision that the transparency and flexibility of the carbon nanotubes can add more
features to future transparent displays. Our achievements may open a door to future flexible carbon
nanotube based transparent displays.
Chapter 4 References
1. Gorrn, P.; Sander, M.; Meyer, J.; Kroger, M.; Becker, E.; Johannes, H. H.; Kowalsky, W.; Riedl,
T. Adv. Mater. 2006, 18, 738-741.
2. Su, C. W.; Chen, M. Y. J. Disp. Technol. 2014, 10, 683-687.
3. Facchetti, A.; Marks, T. J., Transparent Electronics: From Synthesis to Applications. John Wiley
and Sons, Ltd: Chichester, UK, 2010.
4. Zhang, J. L.; Wang, C.; Zhou, C. W. ACS Nano 2012, 6, 7412-7419.
5. Ju, S.; Li, J. F.; Liu, J.; Chen, P. C.; Ha, Y. G.; Ishikawa, F.; Chang, H.; Zhou, C. W.; Facchetti,
A.; Janes, D. B.; Marks, T. J. Nano Lett. 2008, 8, 997-1004.
6. Su, C. W.; Liao, C. C.; Chen, M. Y. J. Disp. Technol. 2016, 12, 31-34.
7. Lim, T.; Kim, H.; Meyyappan, M.; Ju, S. ACS Nano 2012, 6, 4912-4920.
8. Chung, S. H.; Noh, H. Y. Opt. Express 2015, 23, 32149-32157.
9. Mach, P.; Rodriguez, S. J.; Nortrup, R.; Wiltzius, P.; Rogers, J. A. Appl. Phys. Lett. 2001, 78, 3592-
3594.
10. Sheraw, C. D.; Zhou, L.; Huang, J. R.; Gundlach, D. J.; Jackson, T. N.; Kane, M. G.; Hill, I. G.;
Hammond, M. S.; Campi, J.; Greening, B. K.; Francl, J.; West, J. Appl. Phys. Lett. 2002, 80, 1088-1090.
! 89!
11. Lee, J.; Kim, D. H.; Kim, J. Y.; Yoo, B.; Chung, J. W.; Park, J. I.; Lee, B. L.; Jung, J. Y.; Park, J.
S.; Koo, B.; Im, S.; Kim, J. W.; Song, B.; Jung, M. H.; Jang, J. E.; Jin, Y. W.; Lee, S. Y. Adv. Mater. 2013,
25, 5886-5892.
12. Han, L.; Mandlik, P.; Wagner, S. IEEE Electron Device Lett. 2009, 30, 502-504.
13. Kumar, B.; Kaushik, B. K.; Negi, Y. S. J. Mater. Sci. Mater. Electron. 2014, 25, 1-30.
14. Javey, A.; Kong, J., Carbon Nanotube Electronics. Springer: New York, USA, 2009.
15. Sun, D. M.; Timmermans, M. Y.; Tian, Y.; Nasibulin, A. G.; Kauppinen, E. I.; Kishimoto, S.;
Mizutani, T.; Ohno, Y. Nat. Nanotechnol. 2011, 6, 156-161.
16. Cao, Q.; Kim, H. S.; Pimparkar, N.; Kulkarni, J. P.; Wang, C. J.; Shim, M.; Roy, K.; Alam, M. A.;
Rogers, J. A. Nature 2008, 454, 495-500.
17. Bhaviripudi, S.; Mile, E.; Steiner, S. A.; Zare, A. T.; Dresselhaus, M. S.; Belcher, A. M.; Kong, J.
J. Am. Chem. Soc. 2007, 129, 1516-1517.
18. Liu, B. L.; Ren, W. C.; Gao, L. B.; Li, S. S.; Pei, S. F.; Liu, C.; Jiang, C. B.; Cheng, H. M. J. Am.
Chem. Soc. 2009, 131, 2082-2083.
19. Liu, J.; Wang, C.; Tu, X. M.; Liu, B. L.; Chen, L.; Zheng, M.; Zhou, C. W. Nat. Commun. 2012, 3,
1199.
20. Zhang, F.; Hou, P. X.; Liu, C.; Wang, B. W.; Jiang, H.; Chen, M. L.; Sun, D. M.; Li, J. C.; Cong,
H. T.; Kauppinen, E. I.; Cheng, H. M. Nat. Commun. 2016, 7, 11160.
21. Arnold, M. S.; Green, A. A.; Hulvat, J. F.; Stupp, S. I.; Hersam, M. C. Nat. Nanotechnol. 2006, 1,
60-65.
22. Mistry, K. S.; Larsen, B. A.; Blackburn, J. L. ACS Nano 2013, 7, 2231-2239.
23. Gui, H.; Chen, H. T.; Khripin, C. Y.; Liu, B. L.; Fagan, J. A.; Zhou, C. W.; Zheng, M. Nanoscale
2016, 8, 3467-3473.
24. Cai, L.; Zhang, S. M.; Miao, J. S.; Wei, Q. Q.; Wang, C. Nanoscale Res. Lett. 2015, 10, 291.
25. Chen, H. T.; Cao, Y.; Zhang, J. L.; Zhou, C. W. Nat. Commun. 2014, 5, 4097.
26. Brady, G. J.; Joo, Y.; Wu, M. Y.; Shea, M. J.; Gopalan, P.; Arnold, M. S. ACS Nano 2014, 8, 11614-
11621.
27. Zhao, Y. D.; Li, Q. Q.; Xiao, X. Y.; Li, G. H.; Jin, Y. H.; Jiang, K. L.; Wang, J. P.; Fan, S. S. ACS
Nano 2016, 10, 2193-2202.
28. Wei, H.; Shulaker, M.; Wong, H. S. P.; Mitra, S., Monolithic Three-Dimensional Integration of
Carbon Nanotube FET Complementary Logic Circuits. In 2013 IEEE International Electron Devices
Meeting (IEDM), Washington, DC, USA, 2013.
29. Liu, X. L.; Han, S.; Zhou, C. W. Nano Lett. 2006, 6, 34-39.
! 90!
30. Wang, C.; Hwang, D.; Yu, Z. B.; Takei, K.; Park, J.; Chen, T.; Ma, B. W.; Javey, A. Nat. Mater.
2013, 12, 899-904.
31. Zhang, J. L.; Fu, Y.; Wang, C.; Chen, P. C.; Liu, Z. W.; Wei, W.; Wu, C.; Thompson, M. E.; Zhou,
C. W. Nano Lett. 2011, 11, 4852-4858.
32. Zou, J. P.; Zhang, K.; Li, J. Q.; Zhao, Y. B.; Wang, Y. L.; Pillai, S. K. R.; Demir, H. V.; Sun, X.
W.; Chan-Park, M. B.; Zhang, Q. Sci. Rep. 2015, 5, 11755.
33. Selvarasah, S.; Li, X. H.; Busnaina, A.; Dokmeci, M. R. Appl. Phys. Lett. 2010, 97, 153120.
34. Cox, S. J.; Reshetnyak, V. Y.; Sluckin, T. J. Mol. Cryst. Liq. Crys. A 1998, 320, 301-319.
35. Yang, K. J.; Kang, J. K.; Choi, B. D. Jpn. J. Appl. Phys. 2014, 53, 08NF03.
! 91!
Chapter 5 Improving Current On/Off Ratio and the Observation of
Negative Differential Resistance for 40-nm-Channel Carbon
Nanotube Array Transistors
5.1 Promising Carbon Nanotube Array Transistors and Their Challenges
In recent years, intense research and studies have been conducted on the new generation of
materials for energy-efficient and high-speed nanoelectronics after Si transistors approach their
physical and theoretical limits.
1, 2
Carbon nanotube, due to its one-dimensional nature and excellent
electronic properties, is a promising channel material which can enable high-speed and low-power
electronics in sub-5 nm logic technology nodes.
3
Significant accomplishments have been made on
both performance and scaling for carbon nanotube field effect transistors (CNTFETs).
4-6
Quasi-
ballistic carbon nanotube array transistors with current density exceeding Si and GaAs
7
as well as
5-nm gate length complementary CNTFETs
3
have been achieved. Recently, Cao et al
demonstrated carbon nanotube transistors scaled to a 40-nanometer footprint which can deliver
higher current than that of the best-competing silicon devices.
8
On the other hand, many scientific
challenges still need to be solved for reliable industrial-scale applications, such as the driving
capabilities of the driver transistor. The on/off ratio is the critical factor deciding whether the
current driver transistor output can effectively control the input of the next stage. It is desirable
that the on/off ratio larger than 10
4
for operation drain-source voltage (V
ds
) on the order of the
supply voltage. For short channel devices with channel length (L
ch
) less than 50 nm, single CNT
transistors can provide high on/off ratio of 10
5
at V
ds
= 0.4 V.
6, 9
To achieve high current density,
CNT array transistors can take advantage of simultaneous contributions of the carrying currents
passed by multiple-nanotube channels. Under this circumstance, the on/off ratio usually degrades
! 92!
due to the variations in off current, threshold voltage etc.
9
Heinze et al observed the trend of higher
off state current for higher V
ds
, and pointed out that under the Schottky barrier model, ultra-scaled
CNTFETs will exhibit unacceptable off state leakage under useful operating voltages, even for
devices with small-diameter tubes of ~1 nm.
10
Experimentally, even with the thin high-κ dielectric
thickness, which can provide good gating effects, the on/off ratios for sub-40-nm-channel CNT
array transistors have not exceeded 10
3
at V
ds
= 0.5 V.
8
Improving the low on/off ratio is critical
before the realization of industrial-scale applications of CNT array transistors.
5.2 Floating Evaporative Self-Assembly (FESA) Technique
Among the nanotube alignment techniques, the floating evaporative self-assembly (FESA)
technique can align high density and high semiconducting purity carbon nanotubes directly to the
SiO
2
surfaces of the substrates.
7, 11, 12
Without the need of additional transfer, carbon nanotube array
transistors made by the FESA technique can not only enjoy the benefits of the high aligned
nanotube density, but also fewer defects in the channel area. Furthermore, the FESA carbon
nanotube array transistors have been demonstrated to exceed the performance of their Si and GaAs
counterparts at L
ch
= 100 nm.
7
It is thus necessary to investigate and understand the performance
of FESA CNTFETs with sub-100-nm channel length, and the ways to mitigate any adverse effects.
The high-density, well-aligned semiconducting nanotubes used in this study were deposited by
FESA technique. Dispersions of semiconducting SWCNTs for aligned CNT deposition were
prepared following previous publication.
7
Briefly, arc-discharge CNTs were sorted for
semiconducting enrichment using PFO-BPy in toluene. Pellets of the polymer-sorted CNTs were
formed via centrifugation, the toluene was decanted away from the CNT pellet, and the pellet was
dispersed into chloroform. Semiconducting-enriched CNTs in chloroform were used as the ink to
deposit aligned arrays of CNTs onto a SiO
2
(15 nm, thermal growth) surface of a Si wafer using
! 93!
an adapted FESA method.
7, 11-13
In the present study, FESA deposition yielded bands of aligned
CNTs that immediately neighbor one another. Between the aligned bands a narrow interface region
exists, where disorder in the CNT alignment is present. The average width of the aligned bands
and the transition regions were 30 µm and 1 µm, respectively. We do not observe gaps between
the bands. The deposition of the high-density, well-aligned semiconducting nanotubes can be
achieved uniformly over ~3 cm × 5 cm on a 4-inch wafer (Fig. 5.1a). The blue rectangle indicates
the region where the aligned nanotubes exist and the red arrow indicates the alignment direction.
The scanning electron microscope (SEM) images show the uniform aligned CNTs reside in stripes
of width ~30 µm (Fig. 5.1b). Fig. 5.1c shows a zoomed-in image of the aligned region. The tubes
are well-aligned with a density of 40-50 tubes/µm. Between the stripes are the transient regions
between each pulse of CNT droplet during the deposition process. The transient regions have
misaligned tubes, forming random carbon nanotube networks of a few micrometers wide, and can
be as narrow as ~1 µm (Fig. 5.1d).
! 94!
Figure 5.1 Carbon nanotube field effect transistors on FESA aligned tubes (a) Optical image of a 4-inch
wafer with FESA aligned tubes inside the blue boundary. The direction of alignment is indicated by the red
arrows. (b-d) The SEM images of the FESA aligned tubes of a larger area (b), aligned region (c) and
transient region (d).
5.3 The Improvement of On/Off Ratio of 40-nm Carbon Nanotube Array Transistors
Measured in Argon
The schematic of the back-gated device structure based on the FESA CNTs is depicted in Fig. 5.2a.
The active channel areas are selectively chosen to land in the aligned regions. Heavily p-type doped
Si is used as the common back gate. The gate dielectric is a thermally grown 15 nm thick SiO
2
layer. High semiconducting ratio, densely aligned carbon nanotubes are deposited onto the SiO
2
surface via the FESA method. Pure palladium (30 nm) forms the source and drain electrodes for
the transistors. The fabrication consists of the following steps: (1) The alignment markers were
patterned by electron beam lithography (EBL). 1 nm Ti and 50 nm Au were sequentially deposited
! 95!
followed by liftoff. (2) A second layer of EBL was performed to define the channel areas of
CNTFETs. Unwanted CNTs outside the channel area were etched away by oxygen plasma at 100
W and 150 mTorr for 20 s. The channel width is defined to be 3.5 µm by the unwanted etching.
(3) After striping off the PMMA, the carbon nanotubes were annealed in high vacuum for 1 hour
to remove polymer residues. (4) The third layer of EBL is used to define the source and drain
separated by the short channel. To achieve the precision, a single layer of PMMA 950k was used,
the spinning rate is 1600 rpm and the dosage is controlled at 116 µC/cm
2
. (5) After developing the
pattern for 30 s in MIBK:IPA = 1: 3 solution, 30 nm Pd was deposited followed by subsequent
liftoff in acetone.
Figure 5.2 (a) Schematic diagram of the short-channel, back-gated CNTFETs. (b-c) The SEM images of
the short channel devices channel area of L
ch
= 100 nm (b) and L
ch
= 40 nm (c).
Fig. 5.2b shows the SEM image of the channel area of short channel device with L
ch
= 100 nm,
individual CNTs can be clearly identified inside the channel area. However, as the channel length
decreases, it is getting more and more difficult to image individual tubes inside the short channel
! 96!
area. At L
ch
= 40 nm, the details of the channel area cannot be clearly captured, but the channel
length can be determined from Fig. 5.2c.
Brady et al. observed a decrease in on/off ratio for FESA aligned CNT array transistors when the
channel length is less than 100 nm at V
ds
= -0.1 V. They hypothesized that the misalignment and
the overlapping of tubes render a loss of controllability of the gate. We observed similar trend in
the decrease of on/off ratio at L
ch
= 65 nm compared to L
ch
= 100 nm. In addition, at higher V
ds
= -
0.5 V, the degradation is more significant, as shown in Fig. 5.3.
Figure 5.3 Electrical characteristics of the short-channel CNTFETs. (a,c) I
ds
– V
gs
curves in both linear
(green and black) and log (red and blue) scales at V
ds
= -0.1 V (black and red) and V
ds
= -0.5 V (green and
! 97!
blue), (a) L
ch
= 100 nm and (c) L
ch
= 65 nm. (b,d) I
ds
– V
ds
curves with V
gs
= -8 V to 0 V in steps of 2 V, (b)
L
ch
= 100 nm and (d) L
ch
= 65 nm. (e,f) On state current density versus channel length under different V
ds
,
(e) V
ds
= -0.1 V and (f) V
ds
= -0.5 V.
!
We further fabricated CNTFETs with channel length of 40 nm. We studied further on other
possible effects on the on/off ratio for these 40 nm devices. We noticed that at shorter channel
lengths, the devices are more vulnerable to dielectric breakdown. Making the devices more robust
without the loss of generality and understanding the physics, we reduced the gate voltage sweeping
range to V
gs
= -6 V to 6 V. Fig. 5.4a shows the typical I
ds
-V
gs
curve for a 40 nm device for V
ds
= -
0.1 V and -0.5 V. The red the blue curves are in log scale while the green and black curves are in
linear scale. As we can see that the on/off degrades to only 49.1 at V
ds
= -0.1 V and 29.8 at V
ds
= -
0.5 V. Besides, unlike longer channel length devices where there is a clear cutoff voltage, I
ds
for
the short channel devices monotonically decrease as V
gs
ramps up, without the trend for n-branch
conduction. This implies the holes are still the majority carriers injecting into carbon nanotubes
from Pd electrodes either via tunneling or thermal activation. The decrease in on/off ratio is due to
the increase in off current. Fig. 5.4b shows the I
ds
-V
ds
curve for the device and one can see that
there is an obvious off current at V
ds
= -0.5 V.
! 98!
Figure 5.4 The electrical characteristics for the 40 nm CNTFETs in air and in argon. (a) I
ds
– V
gs
curves
measured in air for a typical 40-nm device in both linear (green and black) and log (red and blue) scales at
V
ds
= -0.1 V (black and red) and V
ds
= -0.5 V (green and blue). (b) I
ds
– V
gs
curves measured in air for the
40-nm device with V
gs
= -6 V to 6 V in steps of 2 V and I
ds
= 0 V to -0.5 V. (c) The same measurements as
in (A) for the same device in argon. (d) The same measurements as in (b) for the same device in argon. (e)
The on/off ratio versus V
ds
in air (blue) and in argon (red). (f) The on/off ratio of the device measured in
argon compared with the highest on/off ratio in previous publication measured in air.
The origins of the prominent decrease in on/off ratio can be a combination of various reasons. One
possible reason can be the short channel effects, which includes possible tunneling and barrier
lowering effects.
14
Another possible reason is the existence of metallic tube directly bridging
! 99!
source and drain.
15
Other possibilities include the reduced gate controllability over the channel as
well as the tube-tube interaction due to the densely aligned nature.
9
Furthermore, oxygen and
water in ambient conditions can affect transistor performance, and their effects can be magnified
at short channels.
16-19
As a result, it is helpful to understand the origin of the low on/off ratio by
minimizing the effects of active gas molecules like oxygen and studying the device performance
in inert atmosphere like argon.
The devices were wire-bonded to a chip carrier and then placed into an enclosed chamber of 10
cm× 5 cm× 5 cm. The devices after wire-bonding are shown in Fig. 5.5. All the measurements
were carried out after the chamber was purged with pure argon at 500 sccm for 10 minutes. Fig.
5.4c shows the I
ds
-V
gs
curves for a typical device in argon at V
ds
= -0.1 V and -0.5 V. The devices
show reduced leakage current for more than two orders of magnitude in argon compared with the
case in air while only suffers minimal decrease in on-state current of ~15%. The decreased leakage
current significantly improves the on/off ratio. At V
ds
= -0.1 V, the on/off ratio is 1.00×10
5
, and at
V
ds
= -0.5 V, the on/off ratio is 1.29×10
4
. This on/off ratio is the highest observed value for densely
aligned semiconducting CNTFETs at V
ds
= 0.5 V of sub-50 nm channel length. The reduction of
the off current can also be observed in I
ds
-V
ds
curve shown in Fig. 5.4d. In air, it is very hard to
completely turn the device off, even at high V
gs
= 6 V, while in argon, we can see that the device
shows negligible I
ds
for V
gs
>= 0 V. The comparison of on/off ratio under various V
ds
from -0.1 V
to -0.8 V in air (blue) and argon (red) at fixed sweeping range of V
gs
from -6 V to 6 V is plotted in
Fig. 5.4e. As V
ds
increases, both the on/off ratios in air and argon decrease, and the on/off ratios in
argon remain consistently around two orders of magnitude higher than those in air. The
improvement in on/off ratio for 40 nm FESA CNTFETs measured in argon (star) in reference to
the best devices in previous publication when measured in air (square) can be clearly seen in Fig.
! 100!
5.4f. As we can see, the on/off ratio has improved more than 10 times measured in air at the same
channel length of 40 nm.
Figure 5.5 Bonded Carbon Nanotube Device on a Chip Carrier
!
Despite the combined effect of various mechanisms can be complicated, the observation of high
on/off ratio in argon excludes several possible reasons for the major cause of the decreased on/off
ratio in air at L
ch
= 40 nm. First, the high on/off ratio in argon proves the high semiconducting
purity of the CNTs. As only one metallic tube in the channel can increase the off current
substantially, all the tubes in the active channel are semiconducting. Second, the imperfect tube
morphology is not the fundamental reason for our FESA platform causing the reduced gate
controllability at L
ch
= 40 nm in air. Our FESA platform covers ~5% of the total surface area,
bilayer carbon nanotubes with the cross tube junctions are not frequent but can sometimes happen
in the active channel area. Instead, the interactions between molecules in air and the device
electrodes or the nanotubes themselves can be the main reason responsible for the degraded on/off
ratio. Studies show that oxygen can effectively be a p-type dopant for long channel CNTFETs, by
creating a lowest unoccupied molecular orbit (LUMO) carrier trap.
16
For short channel devices,
oxygen could more likely affect the device performance via the modification of the electrode-tube
interactions.
17
! 101!
For CNTFETs, the accepted model for device operation is the Schottky barrier model.
14, 20
High
work function metals are usually used as the CNTFET source and drains for better Fermi level
alignment for CNTFET devices. When the gate voltage is low, the CNTFET is in on state, and the
resulting high work function can help hole injection via both the thermionic and tunneling
processes from the metal to CNTs. During the off state, the high gate voltage suppresses both the
thermionic and tunneling processes for the hole injection by modifying the Schottky barrier. At
the same time, the Schottky barrier for the electrons remains high, giving a high on/off current
ratio. For the 40-nm channel length devices, the significant degradation in the on/off ratio is mainly
due to the increase in the off state current. The short channel devices are very sensitive to the
surrounding gas environment and exposure to oxygen could increase the Pd electrode work
function. Besides, despite the short channel, oxygen can still possibly modify the band structure
of the carbon nanotubes and behaves like a p-type dopant. The combined effects of oxygen in air
could cause barrier lowering for thermal activation processes. Furthermore, it is also possible to
cause the thinning of the barrier width, making hole tunneling process easier. Even for the off state,
a noticeable amount of holes can still either thermally jump over or tunneling through the Schottky
barrier, leaving a large level of leakage current.
Apart from oxygen, moisture is another factor causing the increased off current in air. The
hydroxyl/water groups on the surface form electron-traps.
18, 19
The electrons in the traps in the
vicinity of the CNTs will provide the negative bias biasing the CNTs toward the on state. As the
channel length is reduced, the drain-source electric field increases and more of the traps are filled.
Also, having a high-density of arrayed CNTs will enhance ability for the traps to fill. As a result,
the off state will be biased to a lesser extent, and lead to a leakage current increase.
! 102!
In argon, the effects of the oxygen and moisture are minimized, leading to an exponential decrease
in the off current. We also note the relatively small change in the on state current between the
measurements in air and argon, which could also be a combination of the effects of oxygen and
moisture. During the on state, when oxygen is present, the barrier height is low and the barrier
width is thin, and holes can easily inject into the tubes from the electrode. Without the presence of
oxygen, the on state current should drop significantly if oxygen is the only major effect. On the
other hand, the presence of moisture can have an opposite effect. When moisture is present, due to
the applied negative gate voltage, positive charge traps are formed. These charge traps act as a
positive electric potential at SWCNT surface and reduce the effective negative gate voltage, so the
on-state current will decease. As both oxygen and moisture effects are minimized, we do not
observe a significant change in on state current. However, the on state current does observe a
noticeable drop of 15%, and we conclude that the effects caused by oxygen are of more importance
than those caused by moisture for the on state current.
5.4 Negative Differential Resistance (NDR) Effects
Another interesting observation is that when measuring I
ds
-V
ds
curves at a larger V
ds
range of 0 V
to -0.8 V, for the first sweep, negative differential resistance (NDR) effects appear. (Fig. 5.6a).
Previously, negative differential resistance effects were only observed in metallic CNTs. It is
predicted that for semiconducting tubes, when the current level is high, it is possibly to observe
NDR. For our case, we only observed NDR in the first forward sweep, and the subsequent sweeps
and the backward sweep does not exhibit such phenomenon. We believe that this is associated
with oxygen desorption process when the currents heat up the device.
To verify the oxygen/moisture desorption processes, we measured I
ds
-V
ds
curve under constant V
gs
= -6 V for 6 consecutive sweeps. As shown in Fig. 5.6b, the negative differential resistance effects
! 103!
disappear after the first sweep and hysteresis becomes smaller, indicating the adsorbed oxygen and
moisture moves away from the surface. After such repeated measurements, we immediately
measured the I
ds
-V
gs
curves. With fewer adsorbed oxygen and water molecules, the off current
becomes lower and the on/off ratio increases. As one can see in Fig. 5.6c, the on/off ratio increases
from 49.1 to 359 for V
ds
= -0.1 V and from 29.8 to 326 for V
ds
= -0.5 V. The improvements do not
match the high on/off ratio observed in argon, but still an order of magnitude higher than the
original I
ds
-V
gs
sweeps. These observations confirm the effects of oxygen and moisture in
determining the CNTFET on/off ratio.
Figure 5.6 The negative differential resistance phenomenon of the device. (a) I
ds
– V
gs
curves measured in
air for the 40-nm device with V
gs
= -6 V to 6 V in steps of 2 V and I
ds
= 0 V to -0.8 V. (b) The continuous
double sweep I
ds
– V
ds
curves from V
ds
= 0 V to 0.8 V under V
gs
= -6 V in air for 6 times. (c, d) I
ds
– V
gs
curves at V
ds
= -0.1 V (c) and V
ds
= -0.5 V (d) before and after (measured immediately) the measurements
in (b).
! 104!
Furthermore, the outgassing and restoring of the oxygen molecules can be a quick process, even
in the same sweep of the measurement.
17
For the I
gs
-V
gs
curve in air, the current is large initially,
the generated heat will cause oxygen adsorbed on Pd surface suddenly “detached” from the surface.
This sudden drop in oxygen level will also cause the decrease in work function, which partially
responsible for the small difference in the on state currents for air and argon. As the current
becomes smaller, the substrate becomes cooler and the oxygen will re-attach to the Pd surface.
With the re-adsorption of oxygen, the work function of Pd will increase, contributing to the large
off state current. The evidence of the re-adsorption of oxygen can be also observed in the I
ds
-V
ds
curve in Fig. 5.7a. V
ds
sweeps from -0.8 V to 0.8 V, and then backwards at V
gs
= -6 V. When V
ds
is
negative (segment 1), the CNTFET is turned on and under high current, and the channel area
becomes hot, causing oxygen molecules to desorb from the Pd surface. During the sweeping, after
the current becomes small under small V
ds
, the surface becomes cooler, and the oxygen molecules
quickly re-land on the Pd electrode. When V
ds
enters the positive regime (segment 2), the oxygen
will desorb again, causing the current to saturate and even drop when V
ds
> 0.5 V, which is the
negative differential resistance phenomenon. Amer et al also reported kink behavior in current-
voltage characteristics associated with gas desorption process.
21
For the backward sweep, as
oxygen is degassed, the current does not increase but monotonically decrease as V
ds
decreases
(segment 3). Furthermore, the oxygen repopulation is much weaker in strength and cannot restore
to the original level, we did not observe a drop in current at V
ds
> 0.5 V (segment 4) as in the
forward sweep (segment 2), where there are more oxygen molecules. However, the current indeed
begins to saturate around V
ds
= -0.6 V. A similar desorption-adsorption-desorption process can be
observed when V
ds
double sweeps from V
ds
= 0.8 V to V
ds
= -0.8 V (Fig. 5.7b).
! 105!
On the contrary, when the devices are measured in argon, the adsorbed oxygen molecules are
purged away. As expected, no obvious NDR effects are present in the I
ds
-V
ds
curves (Fig. 5.7c and
5.7d). Instead, inflection points, which are characteristics of large Schottky barrier can be
identified from the curves.
22
In addition, in the same double sweep for the measurement in air in
Fig. 5.7a, the forward sweep sees a uniform slope between V
ds
= -0.4 V and 0.4 V. During the
backward sweep, the slope of the curve is slightly smaller around V
ds
= 0 V compared with those
around V
ds
= -0.3 V and 0.3 V, which further confirms a slight increase in the Schottky barrier
caused by the desorption of oxygen in the backward sweep.
Figure 5.7 The desorption and adsorption of oxygen and moisture. (a) The double sweep I
ds
– V
ds
curves
from V
ds
= -0.8 V to 0.8 V under V
gs
= -6 V in air for the L
ch
= 40 nm device. The forward sweep exhibits
negative differential resistance behaviour. (b) The same measurement in air as in (a) except that the
sweeping direction is from V
ds
= 0.8 V to -0.8 V. (c) The double sweep I
ds
– V
ds
curves from V
ds
= -0.8 V
to 0.8 V under V
gs
= -6 V in argon for the L
ch
= 40 nm device. (d) The same double sweep in argon as in
(c) except that the sweeping direction is from V
ds
= 0.8 V to -0.8 V.
! 106!
5.5 Initial Screening of Top Gated Aligned Carbon Nanotube Transistor Performance
After obtaining good performance bottom gate CNTFETs with high on current and on/off ratio,
the next milestone is to make these CNTFETs as the building blocks to fabricate high performance
circuits. The aligned tubes assembled using the FESA technique rely on the tension between the
substrate and the water surfaces. At this time, good alignment can be only achieved with flat
substrate surfaces. With this restriction, in order to build circuits involving connections between
transistors as well as controlling each individual transistors, the layout requires top gate structures
for the CNTFETs. This motivates the investigation of the performance of top-gated CNTFETs.
We started by the fabrication of back gated CNTFETs with 100 nm SiO
2
on highly conductive Si.
The initial screening used transistors of channel width of 4 µm and channel length of 850 nm. The
SEM image of the channel area is shown in Fig. 5.8a. We measured the I
ds
-V
ds
curve of the back-
gated device, as shown in Fig 5.8b. The inset shows the schematic of the device. The device shows
on state current density of 65.75 µA/µm, and on/off ratio of 4.7×10
5
. As we need dielectric layer
and the atomic layer deposition (ALD) technique usually gives good dielectric quality, we choose
ALD Al
2
O
3
as the dielectric material. Here, we create the pattern of ALD Al
2
O
3
by using the lift
off method. We write the pattern using electron beam lithography similar to other layers. In order
to prevent the hard bake of the PMMA resist and the subsequent difficulty of lift off, the ALD
deposition temperature is set to 90 ℃ . The dielectric thickness is 12 nm. The lift off in acetone
for the ALD Al
2
O
3
layer typically takes longer than lift off for other layers, and requires overnight
soaking.
! 107!
Figure 5.8 Initial screening of the top gated aligned carbon nanotube performance. (a) The SEM image of
the channel area of CNTFET with W = 4 µm and L
ch
= 850 nm. (b) I
ds
-V
gs
curve of as made back gated
device. Inset: structure of back gated device. (c,d,e) I
ds
-V
gs
curve of the device measured after (c) ALD
deposition, (d) 30 minutes 400 degrees annealing in argon, and (e) top gate Ti/Au deposition. Inset: structure
of back gated device (c,d) after ALD deposition and (e) top gate Ti/Au deposition. (f) The I
ds
-V
ds
curves of
the top gated device with V
gs
= 0 to -5 V with -1 V step.
After ALD deposition, the device shows ambipolar behavior, as one can see a significant n-branch
in Fig. 5.8c. The increase in n-branch conductance as well as decrease in p-branch conductance
! 108!
can be explained by the elimination of oxygen vacancies, and the pumping away of moisture and
oxygen molecules in air during the ALD process. Similar observations have been reported in
previous publications.
23, 24
After annealing in argon at 400 degrees for 30 minutes, the unipolar p-
type conductance has been restored (Fig. 5.8d), but the conductance level is lower than the as made
devices (Fig. 5.8a). After another layer of e-beam lithography and the Ti/Au top gate deposition,
the top gated structure is shown in the inset of Fig. 5.8e. As the dielectric thickness is only 12 nm,
the gate voltage sweeping range reduces to -3 V to 3 V, and the I
ds
-V
gs
curve is shown in Fig. 5.8e.
The device exhibits high on/off ratio of 5.8×10
4
, and the on current is 33.9 µm/µA. The I
ds
-V
ds
curves of the top gated device with V
gs
= 0 to -5 V with -1 V step are plotted in Fig. 5.8f. Fine
tuning of the fabrication parameters needs to be studied to achieve better performance, but the
initial screening of top gated CNTFETs is very promising.
5.6 Summary
The work presents the significant improvement in on/off ratio of short channel densely aligned
FESA CNTFETs in argon atmosphere. We attribute the on/off ratio improvements to the lack of
oxygen and the subsequent decrease in palladium electrode work function. Removal of moisture
also help improve the on/off ratio. We also studied the adsorption and desorption processes of
oxygen while sweeping measurement voltages and the observation of negative differential
resistance. These findings help the understanding of the device physics and will benefit future high
on/off ratio, high on-state current CNTFETs. The initial screening of top-gated structure also
shows good on current and on/off ratio. Our achievements may open a door to future large-scale,
high-performance and energy-efficient carbon nanotube based computing devices.
! 109!
Chapter 5 References
1. Shulaker, M. M.; Hills, G.; Patil, N.; Wei, H.; Chen, H. Y.; PhilipWong, H. S.; Mitra, S. Nature
2013, 501, 526-530.
2. Peng, L. M.; Zhang, Z. Y.; Wang, S. Mater. Today 2014, 17, 433-442.
3. Qiu, C. G.; Zhang, Z. Y.; Xiao, M. M.; Yang, Y. J.; Zhong, D. L.; Peng, L. M. Science 2017, 355,
271-276.
4. Franklin, A. D.; Chen, Z. H. Nat. Nanotechnol. 2010, 5, 858-862.
5. Javey, A.; Guo, J.; Wang, Q.; Lundstrom, M.; Dai, H. J. Nature 2003, 424, 654-657.
6. Franklin, A. D.; Luisier, M.; Han, S. J.; Tulevski, G.; Breslin, C. M.; Gignac, L.; Lundstrom, M.
S.; Haensch, W. Nano Lett. 2012, 12, 758-762.
7. Brady, G. J.; Way, A. J.; Safron, N. S.; Evensen, H. T.; Gopalan, P.; Arnold, M. S. Sci. Adv.
2016, 2, e1601240.
8. Cao, Q.; Tersoff, J.; Farmer, D. B.; Zhu, Y.; Han, S. J. Science 2017, 356, 1369-1372.
9. Gerald J. Brady, K. R. J., Michael S. Arnold. Journal of Applied Physics 2017, 122, 124506.
10. Heinze, S.; Tersoff, J.; Avouris, P. Appl. Phys. Lett. 2003, 83, 5038-5040.
11. Joo, Y.; Brady, G. J.; Arnold, M. S.; Gopalan, P. Langmuir 2014, 30, 3460-3466.
12. Brady, G. J.; Joo, Y.; Wu, M. Y.; Shea, M. J.; Gopalan, P.; Arnold, M. S. ACS Nano 2014, 8,
11614-11621.
13. Brady, G. J.; Joo, Y.; Roy, S. S.; Gopalan, P.; Arnold, M. S. Appl. Phys. Lett. 2014, 104.
14. Franklin, A. D.; Farmer, D. B.; Haensch, W. ACS Nano 2014, 8, 7333-7339.
15. Che, Y. C.; Wang, C.; Liu, J.; Liu, B. L.; Lin, X.; Parker, J.; Beasley, C.; Wong, H. S. P.; Zhou,
C. W. ACS Nano 2012, 6, 7454-7462.
16. Kang, D.; Park, N.; Ko, J. H.; Bae, E.; Park, W. Nanotechnology 2005, 16, 1048-1052.
17. Derycke, V.; Martel, R.; Appenzeller, J.; Avouris, P. Appl. Phys. Lett. 2002, 80, 2773-2775.
18. Mudimela, P. R.; Grigoras, K.; Anoshkin, I. V.; Varpula, A.; Ermolov, V.; Anisimov, A. S.;
Nasibulin, A. G.; Novikov, S.; Kauppinen, E. I. J. Sensors 2012, 2012, 496546.
19. Gao, F.; Chen, D.; Tuller, H. L.; Thompson, C. V.; Palacios, T. Journal of Applied Physics 2014,
115, 124506.
20. Heinze, S.; Tersoff, J.; Martel, R.; Derycke, V.; Appenzeller, J.; Avouris, P. Phys. Rev. Lett.
2002, 89, 106801.
21. Amer, M.; Bushmaker, A.; Cronin, S. Nano Res. 2012, 5, 172-180.
22. McClain, D.; Thomas, N.; Youkey, S.; Schaller, R.; Jiao, J.; O'Brien, K. P. Carbon 2009, 47,
1493-1500.
23. Zhang, J. L.; Wang, C.; Fu, Y.; He, Y. C.; Zhou, C. W. ACS Nano 2011, 5, 3284-3292.
! 110!
24. Moriyama, N.; Ohno, Y.; Suzuki, K.; Kishimoto, S.; Mizutani, T. Appl. Phys. Express 2010, 3,
105102.
! 111!
Chapter 6 Conclusion and Outlook
6.1 Conclusion
In this dissertation we have studied the GaAs nanowire optoelectronic and carbon nanotube
electronic device applications in detail.
For the GaAs nanowire growth, characterization and application in solar cells. Starting from
material synthesis, we systematically developed and evaluated each component involved in a
GaAs-on-Si tandem solar cell and successfully demonstrated a pilot device with the proposed
concept and architecture. In specific, we first achieved growth of highly uniform GaAs nanowire
array using non-catalytic selective-area-growth that allows the precise control of nanowire size
and location. This growth technique then was applied to demonstrate GaAs nanowire
homojunction solar cell devices with a champion efficiency of 7.58%, highest among all the GaAs
nanowire array solar cells so far. The result encourages us to take one step forward to make a
nanowire-on-Si tandem solar cell. In order to achieve this structure, we first solved the issue of
growing nanowires on Si (111) substrate. This is a nontrivial advancement from growth on GaAs
(111)B substrate given the dissimilar crystal registry of GaAs and Si as well as complicated surface
chemistry. Nevertheless, we successfully obtained epitaxially grown GaAs nanowires on Si with
100% vertical yield, and even more importantly, the growth does not require complicated
nucleation or growth of buffer layer. Heavily doped n-type GaAs nanowires were then grown on
p+ Si substrate using this technique. The doping concentration was assessed innovatively through
a non-destructive method using photoluminescence measurement based on Burstein-Moss effect.
Ohmic transport is observed across this heavily doped heterojunction which paves the way toward
the tandem solar cells. An 11.4% efficiency tandem solar cell monolithically integrating GaAs
nanowire top cell and Si planar bottom cell was eventually delivered by optimizing the length of
! 112!
nanowire. This is the first time the concept of epitaxial III-V on Si solar cell is materialized and
sets a milestone in the progress of multijunction solar cells. It opens up opportunity for optimized
bandgap combination and allows low cost by using inexpensive Si substrate.
For the carbon nanotube electronic applications, we focused on two parts: macroelectronics and
microelectronics. For the macroelectronics part, we studied the carbon nanotube thin film
transistor applications by monolithically integrating it with active matrix polymer dispersed liquid
crystal displays. In order to achieve the integration, we first assembled PDLC test cells and studied
their transmission properties in response to different electrical inputs. Later, we fabricated
individual back gated carbon nanotube thin film transistors with little performance degradation
after parylene passivation. The CNT-TFTs demonstrate good average mobility of 13.23 cm
2
/(V⋅s)
and on/off ratio of 1.29 ×10
6
. The excellent CNT-TFTs electrical properties guarantee good
controllability of the PDLC pixels. Eventually, we monolithically integrated the CNT-TFT back
panel and the PDLC pixels, demonstrating fully functioning seven-segment active matrix
transparent displays. For the aligned carbon nanotube nanoelectronics applications, we used FESA
technique to align high-semiconducting-ratio CNTs directly onto Si/SiO
2
wafer over a 4 inch wafer
with wide aligned stripes of ~30 µm. We fabricated back-gated CNTFETs and showed the on/off
decreases as the channel length shrinks, the decrease is even more noticeable for large V
ds
. We
purged the 40 nm channel length device in argon and observed the on/off ratio increase more than
two orders of magnitude consistently for a variety of drain source voltages without significant
sacrifice in the on-state current. The improvements in on/off ratio measured in argon mainly come
from the reduction in the off state current. We observed negative differential resistance
phenomenon in the I
ds
-V
ds
curve in air and after repeated I
ds
-V
ds
measurements, the on/off ratio of
the device also increase, which are indications of partial desorption of oxygen and moisture caused
! 113!
by heating during measurements. We attribute the reasons to the removal of oxygen and moisture
in air, in which oxygen modifies the Schottky barrier of the transistor, and the hydroxyl/water
groups on the substrate surface form electron traps, biasing the CNTs towards the on state.
6.2 Outlook
For GaAs nanowire solar cells, there is still plenty room to improve the device performance. With
regard to the GaAs-on-Si tandem solar cells, the GaAs nanowire top cell needs to be significantly
improved. The open circuit voltage obtained so far is way below that of a state-of-art GaAs single
junction solar cell. Doping in the n-type segment can be further increased. As presented in chapter
2, the doping concentration corresponding to 0.1 sccm disilane is around mid of 10
17
cm
-3
. Further
increasing the doping can increase the build-in potential across the p-n junction and also alleviate
the surface depletion effect. GaAs is known for its high density of surface states so surface
passivation is important. Successfully surface passivation using large band gap material has been
successfully applied to radial junction devices. Passivation in axial junction is more complicated
as one need to avoid creating shunting path through the passivation shell. We have attempted to
use AlGaAs as passivation. However, in spite of increased J
sc
, V
oc
is always lower in a passivated
device compared with that of non-passivated device. Careful band alignment design is thus
required. High density of twin defects could be another reason causing degraded performance. The
discontinuity in band structure is known to scatter carriers and to reduce minority diffusion length.
Several other aspects could be optimized to improve the performance of the tandem cell other than
the GaAs nanowire top cell. The Si planar bottom cell used in chapter 2 is still far from the state-
of-art Si solar cell in terms of efficiency. V
oc
is relatively low, but can be improved through careful
configuration of junction properties and surface passivation. The doping in the emitter is now
around 10
21
cm
-3
, and might be too high and lead to significant nonradiative recombination.
! 114!
Junction depth is around 100 nm and should be further optimized to achieve low series resistance.
Front surface is simply passivated by PECVD silicon nitride and back surface is not passivated.
Both sides need better passivation strategies to reduce surface recombination and to increase Voc.
The transport mechanism across the n+-GaAs/p+-Si heterojunction needs to be better understood
to interpret the observed ohmic behavior. As discussed in chapter 2, deep acceptor-assisted
tunneling might dominate the carrier transport. To fully enable the potential of III-V-on-Si two
junction solar cell, we should also explore nanowires with 1.7 eV bandgap as the top cell because
this combination constructs the optimal bandgap combination as discussed in chapter 2. AlGaAs,
InGaP and GaAsP could be candidates, to just name a few. The friendlier surface chemistry of
these materials compared with that of GaAs makes the performance of the optimized tandem cell
even more promising.
For carbon nanotube macroelectronics applications, more complicated active matrix driving
schemes can be used for the polymer dispersed liquid crystal display. 4 transistors and 1 capacitor
(4T1C) is a common driving scheme to improve the uniformity of the current charging and
discharging the pixels. Fully transparent display requires the carbon nanotube transistor electrode
to be transparent as well. Usually, the source and drain electrode can affect transistor performance
significantly, high performance transparent carbon nanotube transistor is another area to work on.
The transparent display properties also have room to improve, such as low driving voltage and
high transparency. Besides polymer dispersed liquid crystals, other forms of transparent display
also have the potential to integrate with flexible and transparent carbon nanotube transistors.
In terms of carbon nanotube microelectronics applications, short channel transistor uniformity and
driving capabilities are the areas to work on. The FESA aligned tube platform can improve the
alignment density, reduce the tube-tube junction as well as increase the aligned areas. Top gated
! 115!
field effect transistors with lower operation voltages are desired as well. The load matching
between the driving transistor and the next stage has to be optimized. Circuitries with increasing
complexities is the next step. Especially, high performance carbon nanotube computers with
machine learning capabilities are the directions to work on to realize carbon nanotube
nanoelectronics exceeding Si.
!
! 116!
Biliography
Tian, B. Z.; Zheng, X. L.; Kempa, T. J.; Fang, Y.; Yu, N. F.; Yu, G. H.; Huang, J. L.; Lieber, C.
M. Nature 2007, 449, 885-889.
Garnett, E. C.; Yang, P. D. J. Am. Chem. Soc. 2008, 130, 9224-9225.
Tang, J. Y.; Huo, Z. Y.; Brittman, S.; Gao, H. W.; Yang, P. D. Nat. Nanotechnol 2011, 6, 568-
572.
Kempa, T. J.; Cahoon, J. F.; Kim, S. K.; Day, R. W.; Bell, D. C.; Park, H. G.; Lieber, C. M. Proc.
Natl. Acad. of Sci. U.S.A. 2012, 109, 1407-1412.
Czaban, J. A.; Thompson, D. A.; LaPierre, R. R. Nano Lett. 2009, 9, 148-154.
Goto, H.; Nosaki, K.; Tomioka, K.; Hara, S.; Hiruma, K.; Motohisa, J.; Fukui, T. Appl. Phys.
Express 2009, 2, 035004.
Mariani, G.; Wong, P. S.; Katzenmeyer, A. M.; Leonard, F.; Shapiro, J.; Huffaker, D. L. Nano
Lett. 2011, 11, 2490-2494.
Wei, W.; Bao, X. Y.; Soci, C.; Ding, Y.; Wang, Z. L.; Wang, D. Nano Lett. 2009, 9, 2926-2934.
Shin, J. C.; Kim, K. H.; Yu, K. J.; Hu, H. F.; Yin, L. J.; Ning, C. Z.; Rogers, J. A.; Zuo, J. M.; Li,
X. L. Nano Lett. 2011, 11, 4831-4838.
Tomioka, K.; Tanaka, T.; Hara, S.; Hiruma, K.; Fukui, T. IEEE J. Sel. Top. Quant. 2011, 17, 1112-
1129.
Hu, S.; Chi, C. Y.; Fountaine, K. T.; Yao, M. Q.; Atwater, H. A.; Dapkus, P. D.; Lewis, N. S.;
Zhou, C. W. Energy Environ. Sci. 2013, 6, 1879-1890.
Zhu, J.; Hsu, C. M.; Yu, Z. F.; Fan, S. H.; Cui, Y. Nano Lett. 2010, 10, 1979-1984.
Putnam, M. C.; Boettcher, S. W.; Kelzenberg, M. D.; Turner-Evans, D. B.; Spurgeon, J. M.;
Warren, E. L.; Briggs, R. M.; Lewis, N. S.; Atwater, H. A. Energy Environ. Sci. 2010, 3, 1037-
1041.
Mariani, G.; Scofield, A. C.; Hung, C. H.; Huffaker, D. L. Nat. Commun. 2013, 4, 1497.
Mariani, G.; Zhou, Z. L.; Scofield, A.; Huffaker, D. L. Nano Lett. 2013, 13, 1632-1637.
Holm, J. V.; Jorgensen, H. I.; Krogstrup, P.; Nygard, J.; Liu, H. Y.; Aagesen, M. Nat. Commun.
2013, 4, 1498.
Krogstrup, P.; Jorgensen, H. I.; Heiss, M.; Demichel, O.; Holm, J. V.; Aagesen, M.; Nygard, J.;
Morral, A. F. I. Nat. Photonics 2013, 7, 306-310.
! 117!
Cui, Y. C.; Wang, J.; Plissard, S. R.; Cavalli, A.; Vu, T. T. T.; van Veldhoven, R. P. J.; Gao, L.;
Trainor, M.; Verheijen, M. A.; Haverkort, J. E. M.; Bakkers, E. P. A. M. Nano Lett. 2013, 13,
4113-4117.
Nakai, E.; Yoshimura, M.; Tomioka, K.; Fukui, T. Jpn. J. Appl. Phys. 2013, 52, 055002.
Gutsche, C.; Lysov, A.; Braam, D.; Regolin, I.; Keller, G.; Li, Z. A.; Geller, M.; Spasova, M.;
Prost, W.; Tegude, F. J. Adv. Funct. Mater. 2012, 22, 929-936.
Wallentin, J.; Anttu, N.; Asoli, D.; Huffman, M.; Aberg, I.; Magnusson, M. H.; Siefer, G.; Fuss-
Kailuweit, P.; Dimroth, F.; Witzigmann, B.; Xu, H. Q.; Samuelson, L.; Deppert, K.; Borgstrom,
M. T. Science 2013, 339, 1057-1060.
King, R. R.; Law, D. C.; Edmondson, K. M.; Fetzer, C. M.; Kinsey, G. S.; Yoon, H.; Sherif, R.
A.; Karam, N. H. Appl. Phys. Lett. 2007, 90, 183516.
Geisz, J. F.; Kurtz, S.; Wanlass, M. W.; Ward, J. S.; Duda, A.; Friedman, D. J.; Olson, J. M.;
McMahon, W. E.; Moriarty, T. E.; Kiehl, J. T. Appl. Phys. Lett. 2007, 91, 023502.
Geisz, J. F.; Friedman, D. J.; Ward, J. S.; Duda, A.; Olavarria, W. J.; Moriarty, T. E.; Kiehl, J. T.;
Romero, M. J.; Norman, A. G.; Jones, K. M. Appl. Phys. Lett. 2008, 93, 123505.
Guter, W.; Schone, J.; Philipps, S. P.; Steiner, M.; Siefer, G.; Wekkeli, A.; Welser, E.; Oliva, E.;
Bett, A. W.; Dimroth, F. Appl. Phys. Lett. 2009, 94, 223504.
Derkacs, D.; Jones-Albertus, R.; Suarez, F.; Fidaner, O. J. Photon. Energy 2012, 2, 021805-8.
Gudiksen, M. S.; Lauhon, L. J.; Wang, J.; Smith, D. C.; Lieber, C. M. Nature 2002, 415, 617-620.
Glas, F. Phys. Rev. B 2006, 74, 121302.
Sburlan, S.; Dapkus, P. D.; Nakano, A. Appl. Phys. Lett. 2012, 100, 163108.
Chuang, L. C.; Moewe, M.; Chase, C.; Kobayashi, N. P.; Chang-Hasnain, C.; Crankshaw, S. Appl.
Phys. Lett. 2007, 90, 043115.
Ertekin, E.; Greaney, P. A.; Chrzan, D. C.; Sands, T. D. J. Appl. Phys. 2005, 97, 114325.
Tomioka, K.; Kobayashi, Y.; Motohisa, J.; Hara, S.; Fukui, T. Nanotechnology 2009, 20, 145302.
Martensson, T.; Svensson, C. P. T.; Wacaser, B. A.; Larsson, M. W.; Seifert, W.; Deppert, K.;
Gustafsson, A.; Wallenberg, L. R.; Samuelson, L. Nano Lett. 2004, 4, 1987-1990.
Moewe, M.; Chuang, L. C.; Crankshaw, S.; Chase, C.; Chang-Hasnain, C. Appl. Phys. Lett. 2008,
93, 023116.
Huang, N. F.; Lin, C. X.; Povinelli, M. L. J. Opt. 2012, 14, 024004.
Huang, N. F.; Lin, C. X.; Povinelli, M. L. J. Appl. Phys. 2012, 112, 064321.
! 118!
Hu, Y.; LaPierre, R. R.; Li, M.; Chen, K.; He, J. J. J. Appl. Phys. 2012, 112, 143116.
Madaria, A. R.; Yao, M. Q.; Chi, C. Y.; Huang, N. F.; Lin, C. X.; Li, R. J.; Povinelli, M. L.;
Dapkus, P. D.; Zhou, C. W. Nano Lett. 2012, 12, 2839-2845.
Lin, C. X.; Povinelli, M. L. Opt. Express 2009, 17, 19371-19381.
Lin, C. X.; Povinelli, M. L. Opt. Express 2011, 19, A1148-A1154.
Lin, C. X.; Huang, N. F.; Povinelli, M. L. Opt. Express 2012, 20, A125-A132.
Kelzenberg, M. D.; Boettcher, S. W.; Petykiewicz, J. A.; Turner-Evans, D. B.; Putnam, M. C.;
Warren, E. L.; Spurgeon, J. M.; Briggs, R. M.; Lewis, N. S.; Atwater, H. A. Nat. Mater. 2010, 9,
239-244.
Heurlin, M.; Wickert, P.; Falt, S.; Borgstrom, M. T.; Deppert, K.; Samuelson, L.; Magnusson, M.
H. Nano Lett. 2011, 11, 2028-2031.
Kayes, B. M.; Atwater, H. A.; Lewis, N. S. J. Appl. Phys. 2005, 97, 114302.
Breuer, S.; Pfuller, C.; Flissikowski, T.; Brandt, O.; Grahn, H. T.; Geelhaar, L.; Riechert, H. Nano
Lett. 2011, 11, 1276-1279.
Kushibe, M.; Eguchi, K.; Funamizu, M.; Ohba, Y. Appl. Phys. Lett. 1990, 56, 1248-1250.
Chang, C. C.; Chi, C. Y.; Yao, M. Q.; Huang, N. F.; Chen, C. C.; Theiss, J.; Bushmaker, A. W.;
LaLumondiere, S.; Yeh, T. W.; Povinelli, M. L.; Zhou, C. W.; Dapkus, P. D.; Cronin, S. B. Nano
Lett. 2012, 12, 4484-4489.
Talin, A. A.; Leonard, F.; Swartzentruber, B. S.; Wang, X.; Hersee, S. D. Phys. Rev. Lett. 2008,
101, 076802.
Katzenmeyer, A. M.; Leonard, F.; Talin, A. A.; Toimil-Molares, M. E.; Cederberg, J. G.; Huang,
J. Y.; Lensch-Falk, J. L. IEEE Trans. Nanotechnol. 2011, 10, 92-95.
Cheung, S. K.; Cheung, N. W. Appl. Phys. Lett. 1986, 49, 85-87.
Seyedi, M. A.; Yao, M.; O'Brien, J.; Wang, S. Y.; Dapkus, P. D. Appl. Phys. Lett. 2013, 103,
251109.
Zhao, J. H.; Wang, A. H.; Altermatt, P. P.; Wenham, S. R.; Green, M. A. Sol. Energ. Mat. Sol.
Cells 1996, 41-42, 87-99.
Shockley, W.; Queisser, H. J. J. Appl. Phys. 1961, 32, 510-519.
Henry, C. H. J. Appl. Phys. 1980, 51, 4494-4500.
Bedair, S. M.; Lamorte, M. F.; Hauser, J. R. Appl. Phys. Lett. 1979, 34, 38-39.
! 119!
Virshup, G. F.; Chung, B. C.; Werthen, J. G. 20th IEEE Photovoltaic Specialists Conference, 1988,
pp 441-445.
Green, M. A.; Emery, K.; Hishikawa, Y.; Warta, W.; Dunlop, E. D. Prog. Photovoltaics 2014, 22,
701-710.
Dimroth, F.; Grave, M.; Beutel, P.; Fiedeler, U.; Karcher, C.; Tibbits, T. N. D.; Oliva, E.; Siefer,
G.; Schachtner, M.; Wekkeli, A.; Bett, A. W.; Krause, R.; Piccin, M.; Blanc, N.; Drazek, C.; Guiot,
E.; Ghyselen, B.; Salvetat, T.; Tauzin, A.; Signamarcheix, T.; Dobrich, A.; Hannappel, T.;
Schwarzburg, K. Prog. Photovoltaics 2014, 22, 277-282.
Yao, M. Q.; Huang, N. F.; Cong, S.; Chi, C. Y.; Seyedi, M. A.; Lin, Y. T.; Cao, Y.; Povinelli, M.
L.; Dapkus, P. D.; Zhou, C. W. Nano Lett. 2014, 14, 3293-3303.
LaPierre, R. R. J. Appl. Phys. 2011, 110, 014310.
Tobias, I.; Luque, A. Prog. Photovoltaics 2002, 10, 323-329.
Hu, Y.; Li, M.; He, J. J.; LaPierre, R. R. Nanotechnology 2013, 24, 065402.
Li, M.; Hu, X. H.; Ye, Z.; Ho, K. M.; Cao, J. R.; Miyawaki, M. Opt. Lett. 2006, 31, 3498-3500.
Reference Solar Spectral Irradiance: Air Mass 1.5 Spectra.
http://rredc.nrel.gov/solar/spectra/am1.5 (April 28th, 2015)
Olson, J. M.; Kurtz, S. R.; Kibbler, A. E.; Faine, P. Appl. Phys. Lett. 1990, 56, 623-625.
Bertness, K. A.; Kurtz, S. R.; Friedman, D. J.; Kibbler, A. E.; Kramer, C.; Olson, J. M. Appl. Phys.
Lett. 1994, 65, 989-991.
Casadei, A.; Krogstrup, P.; Heiss, M.; Rohr, J. A.; Colombo, C.; Ruelle, T.; Upadhyay, S.;
Sorensen, C. B.; Nygard, J.; Morral, A. F. I. Appl. Phys. Lett. 2013, 102, 013117.
Bessire, C. D.; Bjork, M. T.; Schmid, H.; Schenk, A.; Reuter, K. B.; Riel, H. Nano Lett. 2011, 11,
4195-4199.
Yang, T.; Hertenberger, S.; Morkotter, S.; Abstreiter, G.; Koblmuller, G. Appl. Phys. Lett. 2012,
101, 233102.
Perea, D. E.; Hemesath, E. R.; Schwalbach, E. J.; Lensch-Falk, J. L.; Voorhees, P. W.; Lauhon, L.
J. Nat. Nanotechnol. 2009, 4, 315-319.
Burstein, E. Phys. Rev. 1954, 93, 632-633.
Moss, T. S. P. Phys. Soc. Lond. B 1954, 67, 775-782.
Kane, E. O. Phys. Rev. 1963, 131, 79-88.
Kuech, T. F.; Redwing, J. M. J. Cryst. Growth 1994, 145, 382-389.
! 120!
Kuech, T. F.; Veuhoff, E. J. Cryst. Growth 1984, 68, 148-156.
Bose, S. S.; Lee, B.; Kim, M. H.; Stillman, G. E.; Wang, W. I. J. Appl. Phys. 1988, 63, 743-748.
Shimazu, M.; Kamon, K.; Kimura, K.; Mashita, M.; Mihara, M.; Ishii, M. J. Cryst. Growth 1987,
83, 327-333.
Druminski, M.; Wolf, H.-D.; Zschauer, K.-H.; Wittmaack, K. J. Cryst. Growth 1982, 57, 318-324.
Domke, C.; Ebert, P.; Heinrich, M.; Urban, K. Phys. Rev. B 1996, 54, 10288-10291.
Green, M. A. Prog. Photovoltaics 2009, 17, 183-189.
Huang, N. F.; Povinelli, M. L. IEEE J. Photovolt. 2014, 4, 1511-1517.
Huang, Y.; Duan, X.; Cui, Y.; Lauhon, L. J.; Kim, K.-H.; Lieber, C. M. Science 2001, 294, 1313-
1317.
Lauhon, L. J.; Gudiksen, M. S.; Wang, D.; Lieber, C. M. Nature 2002, 420, 57-61.
Johnson, J. C.; Choi, H.-J.; Knutsen, K. P.; Schaller, R. D.; Yang, P.; Saykally, R. J. Nat. Mater.
2002, 1, 106-110.
Parkinson, P.; Lloyd-Hughes, J.; Gao, Q.; Tan, H. H.; Jagadish, C.; Johnston, M. B.; Herz, L. M.
Nano Lett. 2007, 7, 2162-2165.
Tang, T.; Han, S.; Jin, W.; Liu, X. L.; Li, C.; Zhang, D. H.; Zhou, C. W.; Chen, B.; Han, J.;
Meyyapan, M. J. Mater. Res. 2004, 19, 423-426.
Lei, B.; Li, C.; Zhang, D. H.; Zhou, Q. F.; Shung, K. K.; Zhou, C. W. Appl. Phys. Lett. 2004, 84,
4553-4555.
Li, C.; Lei, B.; Zhang, D. H.; Liu, X. L.; Han, S.; Tang, T.; Rouhanizadeh, M.; Hsiai, T.; Zhou, C.
W. Appl. Phys. Lett. 2003, 83, 4014-4016.
Fang, S. F.; Adomi, K.; Iyer, S.; Morkoç, H.; Zabel, H.; Choi, C.; Otsuka, N. J. Appl. Phys. 1990,
68, R31-R58.
Ishiwara, H.; Hoshino, T.; Katahama, H. Materials Chemistry and Physics 1995, 40, 225-229.
Cohen, D.; Carter, C. B. Journal of Microscopy 2002, 208, 84-99.
Komninou, P.; Stoemenos, J.; Dimitrakopulos, G. P.; Karakostas, T. J. Appl. Phys. 1994, 75, 143-
152.
Yu, B. B.; Pchelyakov, O. P. Physics-Uspekhi 2008, 51, 437.
Kawanami, H. Sol. Energ. Mat. Sol. Cells 2001, 66, 479-486.
Tomioka, K.; Motohisa, J.; Hara, S.; Fukui, T. Nano Lett. 2008, 8, 3475-3480.
! 121!
Cirlin, G. E.; Dubrovskii, V. G.; Soshnikov, I. P.; Sibirev, N. V.; Samsonenko, Y. B.; Bouravleuv,
A. D.; Harmand, J. C.; Glas, F. Phys. Status Solidi RRL 2009, 3, 112-114.
Hertenberger, S.; Funk, S.; Vizbaras, K.; Yadav, A.; Rudolph, D.; Becker, J.; Bolte, S.; Döblinger,
M.; Bichler, M.; Scarpa, G.; Lugli, P.; Zardo, I.; Finley, J. J.; Amann, M.-C.; Abstreiter, G.;
Koblmüller, G. Appl. Phys. Lett. 2012, 101, 043116.
Krogstrup, P.; Popovitz-Biro, R.; Johnson, E.; Madsen, M. H.; Nygård, J.; Shtrikman, H. Nano
Lett. 2010, 10, 4475-4482.
Munshi, A. M.; Dheeraj, D. L.; Todorovic, J.; van Helvoort, A. T. J.; Weman, H.; Fimland, B.-O.
J. Cryst. Growth 2013, 372, 163-169.
Cirlin, G. E.; Dubrovskii, V. G.; Samsonenko, Y. B.; Bouravleuv, A. D.; Durose, K.;
Proskuryakov, Y. Y.; Mendes, B.; Bowen, L.; Kaliteevski, M. A.; Abram, R. A.; Zeze, D. Phys.
Rev. B 2010, 82, 035302.
Sandra, J. G.; Jonathan, P. B.; Ray, R. L. Semiconductor Science and Technology 2013, 28,
105025.
Jabeen, F.; Grillo, V.; Rubini, S.; Martelli, F. Nanotechnology 2008, 19, 275711.
Mandl, B.; Stangl, J.; Mårtensson, T.; Mikkelsen, A.; Eriksson, J.; Karlsson, L. S.; Bauer, G.;
Samuelson, L.; Seifert, W. Nano Lett. 2006, 6, 1817-1821.
Paek, J. H.; Nishiwaki, T.; Yamaguchi, M.; Sawaki, N. physica status solidi (c) 2009, 6, 1436-
1440.
Bao, X.-Y.; Soci, C.; Susac, D.; Bratvold, J.; Aplin, D. P. R.; Wei, W.; Chen, C.-Y.; Dayeh, S. A.;
Kavanagh, K. L.; Wang, D. Nano Lett. 2008, 8, 3755-3760.
Roest, A. L.; Verheijen, M. A.; Wunnicke, O.; Serafin, S.; Wondergem, H.; Bakkers, E. P. A. M.
Nanotechnology 2006, 17, S271.
Ihn, S.-G.; Song, J.-I.; Kim, Y.-H.; Lee, J. Y. Appl. Phys. Lett. 2006, 89, 053106.
Huang, H.; Ren, X.; Ye, X.; Guo, J.; Wang, Q.; Yang, Y.; Cai, S.; Huang, Y. Nano Lett. 2009, 10,
64-68.
Kang, J. H.; Gao, Q.; Joyce, H. J.; Tan, H. H.; Jagadish, C.; Kim, Y.; Choi, D. Y.; Guo, Y.; Xu,
H.; Zou, J.; Fickenscher, M. A.; Smith, L. M.; Jackson, H. E.; Yarrison-Rice, J. M.
Nanotechnology 2010, 21, 035604.
Plissard, S.; Dick, K. A.; Wallart, X.; Caroff, P. Appl. Phys. Lett. 2010, 96, 121901.
! 122!
Munshi, A. M.; Dheeraj, D. L.; Fauske, V. T.; Kim, D. C.; Huh, J.; Reinertsen, J. F.; Ahtapodov,
L.; Lee, K. D.; Heidari, B.; van Helvoort, A. T. J.; Fimland, B. O.; Weman, H. Nano Lett. 2014,
14, 960-966.
Plissard, S.; Dick, K., A.; Larrieu, G.; Godey, S.; Addad, A.; Wallart, X.; Caroff, P.
Nanotechnology 2010, 21, 385602.
Kwoen, J.; Watanabe, K.; Iwamoto, S.; Arakawa, Y. J. Cryst. Growth 2013, 378, 562-565.
Koblmüller, G.; Hertenberger, S.; Vizbaras, K.; Bichler, M.; Bao, F.; Zhang, J. P.; Abstreiter, G.
Nanotechnology 2010, 21, 365602.
Rudolph, D.; Hertenberger, S.; Bolte, S.; Paosangthong, W.; Spirkoska, D. e.; Döblinger, M.;
Bichler, M.; Finley, J. J.; Abstreiter, G.; Koblmüller, G. Nano Lett. 2011, 11, 3848-3854.
Tateno, K.; Hibino, H.; Gotoh, H.; Nakano, H. Appl. Phys. Lett. 2006, 89, 033114.
Ihn, S.-G.; Song, J.-I.; Kim, T.-W.; Leem, D.-S.; Lee, T.; Lee, S.-G.; Koh, E. K.; Song, K. Nano
Lett. 2007, 7, 39-44.
Mårtensson, T.; Wagner, J. B.; Hilner, E.; Mikkelsen, A.; Thelander, C.; Stangl, J.; Ohlsson, B. J.;
Gustafsson, A.; Lundgren, E.; Samuelson, L.; Seifert, W. Advanced Materials 2007, 19, 1801-
1806.
Mohseni, P. K.; Maunders, C.; Botton, G. A.; LaPierre, R. R. Nanotechnology 2007, 18, 445304.
Paek, J. H.; Nishiwaki, T.; Yamaguchi, M.; Sawaki, N. physica status solidi (c) 2008, 5, 2740-
2742.
Tomioka, K.; Yoshimura, M.; Fukui, T. Nature 2012, 488, 189-192.
Tomioka, K.; Fukui, T. Appl. Phys. Lett. 2011, 98, 083114.
Svensson, C. P. T.; Mårtensson, T.; Trägårdh, J.; Larsson, C.; Rask, M.; Hessman, D.; Samuelson,
L.; Ohlsson, J. Nanotechnology 2008, 19, 305201.
Guo, W.; Zhang, M.; Banerjee, A.; Bhattacharya, P. Nano Lett. 2010, 10, 3355-3359.
Tomioka, K.; Motohisa, J.; Hara, S.; Hiruma, K.; Fukui, T. Nano Lett. 2010, 10, 1639-1644.
Mayer, B.; Rudolph, D.; Schnell, J.; Morkötter, S.; Winnerl, J.; Treu, J.; Müller, K.; Bracher, G.;
Abstreiter, G.; Koblmüller, G.; Finley, J. J. Nat. Commun. 2013, 4, 2931.
Wagner, R. S.; Ellis, W. C. Appl. Phys. Lett. 1964, 4, 89-90.
Bar-Sadan, M.; Barthel, J.; Shtrikman, H.; Houben, L. Nano Lett. 2012, 12, 2352-2356.
Perea, D. E.; Allen, J. E.; May, S. J.; Wessels, B. W.; Seidman, D. N.; Lauhon, L. J. Nano Lett.
2005, 6, 181-185.
! 123!
Allen, J. E.; Hemesath, E. R.; Perea, D. E.; Lensch-Falk, J. L.; LiZ.Y; Yin, F.; Gass, M. H.; Wang,
P.; Bleloch, A. L.; Palmer, R. E.; Lauhon, L. J. Nat. Nanotechnol. 2008, 3, 168-173.
Fukui, T.; Yoshimura, M.; Nakai, E.; Tomioka, K. AMBIO 2012, 41, 119-124.
Parkinson, P.; Joyce, H. J.; Gao, Q.; Tan, H. H.; Zhang, X.; Zou, J.; Jagadish, C.; Herz, L. M.;
Johnston, M. B. Nano Lett. 2009, 9, 3349-3353.
Spirkoska, D.; Arbiol, J.; Gustafsson, A.; Conesa-Boj, S.; Glas, F.; Zardo, I.; Heigoldt, M.; Gass,
M. H.; Bleloch, A. L.; Estrade, S.; Kaniber, M.; Rossler, J.; Peiro, F.; Morante, J. R.; Abstreiter,
G.; Samuelson, L.; Fontcuberta i Morral, A. Phys. Rev. B 2009, 80, 245325.
Thelander, C.; Caroff, P.; Plissard, S. b.; Dey, A. W.; Dick, K. A. Nano Lett. 2011, 11, 2424-2429.
Yuan, Z.; Nomura, K.-i.; Nakano, A. Appl. Phys. Lett. 2012, 100, 163103.
Ikejiri, K.; Kitauchi, Y.; Tomioka, K.; Motohisa, J.; Fukui, T. Nano Lett. 2011, 11, 4314-4318.
Algra, R. E.; Verheijen, M. A.; Borgstrom, M. T.; Feiner, L.-F.; Immink, G.; van Enckevort, W.
J. P.; Vlieg, E.; Bakkers, E. P. A. M. Nature 2008, 456, 369-372.
Algra, R. E.; Verheijen, M. A.; Feiner, L.-F.; Immink, G. G. W.; Enckevort, W. J. P. v.; Vlieg, E.;
Bakkers, E. P. A. M. Nano Lett. 2011, 11, 1259-1264.
Caroff, P.; Dick, K. A.; Johansson, J.; Messing, M. E.; Deppert, K.; Samuelson, L. Nat.
Nanotechnol. 2009, 4, 50-55.
Dick, K. A.; Bolinsson, J.; Messing, M. E.; Lehmann, S.; Johansson, J.; Caroff, P. Journal of
Vacuum Science & Technology B 2011, 29, 04D103.
Johansson, J.; Karlsson, L. S.; Patrik T. Svensson, C.; Martensson, T.; Wacaser, B. A.; Deppert,
K.; Samuelson, L.; Seifert, W. Nat. Mater. 2006, 5, 574-580.
Johansson, J.; Dick, K. A.; Caroff, P.; Messing, M. E.; Bolinsson, J.; Deppert, K.; Samuelson, L.
Journal of Physical Chemistry C 2010, 114, 3837-3842.
Joyce, H. J.; Wong-Leung, J.; Gao, Q.; Tan, H. H.; Jagadish, C. Nano Lett. 2010, 10, 908-915.
Glas, F.; Harmand, J.-C.; Patriarche, G. Phys. Rev. Lett. 2007, 99, 146101.
Dubrovskii, V. G.; Sibirev, N. V.; Harmand, J. C.; Glas, F. Phys. Rev. B 2008, 78, 235301.
Yoshida, H.; Ikejiri, K.; Sato, T.; Hara, S.; Hiruma, K.; Motohisa, J.; Fukui, T. J. Cryst. Growth
2009, 312, 52-57.
Chi, C.-Y.; Chang, C.-C.; Hu, S.; Yeh, T.-W.; Cronin, S. B.; Dapkus, P. D. Nano Lett. 2013, 13,
2506-2515.
Shapiro, J. N.; Lin, A.; Ratsch, C.; Huffaker, D. L. Nanotechnology 2013, 24, 475601.
! 124!
Yuan, Z.; Shimamura, K.; Shimojo, F.; Nakano, A. J. Appl. Phys. 2013, 114, 074316.
Tournié, E.; Ploog, K. H. J. Cryst. Growth 1994, 135, 97-112.
Zinke+Allmang, M.; Feldman, L. C.; Nakahara, S. Appl. Phys. Lett. 1988, 52, 144-146.
Biegelsen, D. K.; Ponce, F. A.; Smith, A. J.; Tramontana, J. C. J. Appl. Phys. 1987, 61, 1856-1859.
Hull, R.; Fischer+Colbrie, A.; Rosner, S. J.; Koch, S. M.; Harris, J. S. Appl. Phys. Lett. 1987, 51,
1723-1725.
Uhrberg, R. I. G.; Bringans, R. D.; Olmstead, M. A.; Bachrach, R. Z.; Northrup, J. E. Phys. Rev.
B 1987, 35, 3945-3951.
Ando, S.; Kobayashi, N.; Ando, H. J. Cryst. Growth 1994, 145, 302-307.
Biegelsen, D. K.; Bringans, R. D.; Northrup, J. E.; Swartz, L. E. Phys. Rev. Lett. 1990, 65, 452-
455.
Thornton, J. M. C.; Woolf, D. A.; Weightman, P. Applied Surface Science 1998, 123–124, 115-
119.
Rajkumar, K. C.; Chen, P.; Madhukar, A. J. Appl. Phys. 1991, 69, 2219-2223.
Woolf, D. A.; Westwood, D. I.; Williams, R. H. Appl. Phys. Lett. 1993, 62, 1370-1372.
Avery, A. R.; Tok, E. S.; Jones, T. S. Surface Science 1997, 376, L397-L402.
Chen, P.; Rajkumar, K. C.; Madhukar, A. Appl. Phys. Lett. 1991, 58, 1771-1773.
Nishida, T.; Uwai, K.; Kobayashi, Y.; Kobayashi, N. Jpn. J. Appl. Phys. 1995, 34, 6326.
Ikejiri, K.; Sato, T.; Yoshida, H.; Hiruma, K.; Motohisa, J.; Hara, S.; Fukui, T. Nanotechnology
2008, 19, 265604.
Yuan, Z.; Nakano, A. Nano Lett. 2013, 13, 4925-4930.
Gorrn, P.; Sander, M.; Meyer, J.; Kroger, M.; Becker, E.; Johannes, H. H.; Kowalsky, W.; Riedl,
T. Adv. Mater. 2006, 18, 738-741.
Su, C. W.; Chen, M. Y. J. Disp. Technol. 2014, 10, 683-687.
Facchetti, A.; Marks, T. J., Transparent Electronics: From Synthesis to Applications. John Wiley
and Sons, Ltd: Chichester, UK, 2010.
Zhang, J. L.; Wang, C.; Zhou, C. W. ACS Nano 2012, 6, 7412-7419.
Ju, S.; Li, J. F.; Liu, J.; Chen, P. C.; Ha, Y. G.; Ishikawa, F.; Chang, H.; Zhou, C. W.; Facchetti,
A.; Janes, D. B.; Marks, T. J. Nano Lett. 2008, 8, 997-1004.
Su, C. W.; Liao, C. C.; Chen, M. Y. J. Disp. Technol. 2016, 12, 31-34.
! 125!
Lim, T.; Kim, H.; Meyyappan, M.; Ju, S. ACS Nano 2012, 6, 4912-4920.
Chung, S. H.; Noh, H. Y. Opt. Express 2015, 23, 32149-32157.
Mach, P.; Rodriguez, S. J.; Nortrup, R.; Wiltzius, P.; Rogers, J. A. Appl. Phys. Lett. 2001, 78,
3592-3594.
Sheraw, C. D.; Zhou, L.; Huang, J. R.; Gundlach, D. J.; Jackson, T. N.; Kane, M. G.; Hill, I. G.;
Hammond, M. S.; Campi, J.; Greening, B. K.; Francl, J.; West, J. Appl. Phys. Lett. 2002, 80, 1088-
1090.
Lee, J.; Kim, D. H.; Kim, J. Y.; Yoo, B.; Chung, J. W.; Park, J. I.; Lee, B. L.; Jung, J. Y.; Park, J.
S.; Koo, B.; Im, S.; Kim, J. W.; Song, B.; Jung, M. H.; Jang, J. E.; Jin, Y. W.; Lee, S. Y. Adv.
Mater. 2013, 25, 5886-5892.
Han, L.; Mandlik, P.; Wagner, S. IEEE Electron Device Lett. 2009, 30, 502-504.
Kumar, B.; Kaushik, B. K.; Negi, Y. S. J. Mater. Sci. Mater. Electron. 2014, 25, 1-30.
Javey, A.; Kong, J., Carbon Nanotube Electronics. Springer: New York, USA, 2009.
Sun, D. M.; Timmermans, M. Y.; Tian, Y.; Nasibulin, A. G.; Kauppinen, E. I.; Kishimoto, S.;
Mizutani, T.; Ohno, Y. Nat. Nanotechnol. 2011, 6, 156-161.
Cao, Q.; Kim, H. S.; Pimparkar, N.; Kulkarni, J. P.; Wang, C. J.; Shim, M.; Roy, K.; Alam, M. A.;
Rogers, J. A. Nature 2008, 454, 495-500.
Bhaviripudi, S.; Mile, E.; Steiner, S. A.; Zare, A. T.; Dresselhaus, M. S.; Belcher, A. M.; Kong, J.
J. Am. Chem. Soc. 2007, 129, 1516-1517.
Liu, B. L.; Ren, W. C.; Gao, L. B.; Li, S. S.; Pei, S. F.; Liu, C.; Jiang, C. B.; Cheng, H. M. J. Am.
Chem. Soc. 2009, 131, 2082-2083.
Liu, J.; Wang, C.; Tu, X. M.; Liu, B. L.; Chen, L.; Zheng, M.; Zhou, C. W. Nat. Commun. 2012,
3, 1199.
Zhang, F.; Hou, P. X.; Liu, C.; Wang, B. W.; Jiang, H.; Chen, M. L.; Sun, D. M.; Li, J. C.; Cong,
H. T.; Kauppinen, E. I.; Cheng, H. M. Nat. Commun. 2016, 7, 11160.
Arnold, M. S.; Green, A. A.; Hulvat, J. F.; Stupp, S. I.; Hersam, M. C. Nat. Nanotechnol. 2006, 1,
60-65.
Mistry, K. S.; Larsen, B. A.; Blackburn, J. L. ACS Nano 2013, 7, 2231-2239.
Gui, H.; Chen, H. T.; Khripin, C. Y.; Liu, B. L.; Fagan, J. A.; Zhou, C. W.; Zheng, M. Nanoscale
2016, 8, 3467-3473.
Cai, L.; Zhang, S. M.; Miao, J. S.; Wei, Q. Q.; Wang, C. Nanoscale Res. Lett. 2015, 10, 291.
! 126!
Chen, H. T.; Cao, Y.; Zhang, J. L.; Zhou, C. W. Nat. Commun. 2014, 5, 4097.
Brady, G. J.; Joo, Y.; Wu, M. Y.; Shea, M. J.; Gopalan, P.; Arnold, M. S. ACS Nano 2014, 8,
11614-11621.
Zhao, Y. D.; Li, Q. Q.; Xiao, X. Y.; Li, G. H.; Jin, Y. H.; Jiang, K. L.; Wang, J. P.; Fan, S. S. ACS
Nano 2016, 10, 2193-2202.
Wei, H.; Shulaker, M.; Wong, H. S. P.; Mitra, S., Monolithic Three-Dimensional Integration of
Carbon Nanotube FET Complementary Logic Circuits. In 2013 IEEE International Electron
Devices Meeting (IEDM), Washington, DC, USA, 2013.
Liu, X. L.; Han, S.; Zhou, C. W. Nano Lett. 2006, 6, 34-39.
Wang, C.; Hwang, D.; Yu, Z. B.; Takei, K.; Park, J.; Chen, T.; Ma, B. W.; Javey, A. Nat. Mater.
2013, 12, 899-904.
Zhang, J. L.; Fu, Y.; Wang, C.; Chen, P. C.; Liu, Z. W.; Wei, W.; Wu, C.; Thompson, M. E.; Zhou,
C. W. Nano Lett. 2011, 11, 4852-4858.
Zou, J. P.; Zhang, K.; Li, J. Q.; Zhao, Y. B.; Wang, Y. L.; Pillai, S. K. R.; Demir, H. V.; Sun, X.
W.; Chan-Park, M. B.; Zhang, Q. Sci. Rep. 2015, 5, 11755.
Selvarasah, S.; Li, X. H.; Busnaina, A.; Dokmeci, M. R. Appl. Phys. Lett. 2010, 97, 153120.
Cox, S. J.; Reshetnyak, V. Y.; Sluckin, T. J. Mol. Cryst. Liq. Crys. A 1998, 320, 301-319.
Yang, K. J.; Kang, J. K.; Choi, B. D. Jpn. J. Appl. Phys. 2014, 53, 08NF03.
Shulaker, M. M.; Hills, G.; Patil, N.; Wei, H.; Chen, H. Y.; PhilipWong, H. S.; Mitra, S. Nature
2013, 501, 526-530.
Peng, L. M.; Zhang, Z. Y.; Wang, S. Mater. Today 2014, 17, 433-442.
Qiu, C. G.; Zhang, Z. Y.; Xiao, M. M.; Yang, Y. J.; Zhong, D. L.; Peng, L. M. Science 2017, 355,
271-276.
Franklin, A. D.; Chen, Z. H. Nat. Nanotechnol. 2010, 5, 858-862.
Javey, A.; Guo, J.; Wang, Q.; Lundstrom, M.; Dai, H. J. Nature 2003, 424, 654-657.
Franklin, A. D.; Luisier, M.; Han, S. J.; Tulevski, G.; Breslin, C. M.; Gignac, L.; Lundstrom, M.
S.; Haensch, W. Nano Lett. 2012, 12, 758-762.
Brady, G. J.; Way, A. J.; Safron, N. S.; Evensen, H. T.; Gopalan, P.; Arnold, M. S. Sci. Adv. 2016,
2, e1601240.
Cao, Q.; Tersoff, J.; Farmer, D. B.; Zhu, Y.; Han, S. J. Science 2017, 356, 1369-1372.
Gerald J. Brady, K. R. J., Michael S. Arnold. Journal of Applied Physics 2017, 122, 124506.
! 127!
Heinze, S.; Tersoff, J.; Avouris, P. Appl. Phys. Lett. 2003, 83, 5038-5040.
Joo, Y.; Brady, G. J.; Arnold, M. S.; Gopalan, P. Langmuir 2014, 30, 3460-3466.
Brady, G. J.; Joo, Y.; Wu, M. Y.; Shea, M. J.; Gopalan, P.; Arnold, M. S. ACS Nano 2014, 8,
11614-11621.
Brady, G. J.; Joo, Y.; Roy, S. S.; Gopalan, P.; Arnold, M. S. Appl. Phys. Lett. 2014, 104.
Franklin, A. D.; Farmer, D. B.; Haensch, W. ACS Nano 2014, 8, 7333-7339.
Che, Y. C.; Wang, C.; Liu, J.; Liu, B. L.; Lin, X.; Parker, J.; Beasley, C.; Wong, H. S. P.; Zhou,
C. W. ACS Nano 2012, 6, 7454-7462.
Kang, D.; Park, N.; Ko, J. H.; Bae, E.; Park, W. Nanotechnology 2005, 16, 1048-1052.
Derycke, V.; Martel, R.; Appenzeller, J.; Avouris, P. Appl. Phys. Lett. 2002, 80, 2773-2775.
Mudimela, P. R.; Grigoras, K.; Anoshkin, I. V.; Varpula, A.; Ermolov, V.; Anisimov, A. S.;
Nasibulin, A. G.; Novikov, S.; Kauppinen, E. I. J. Sensors 2012, 2012, 496546.
Gao, F.; Chen, D.; Tuller, H. L.; Thompson, C. V.; Palacios, T. Journal of Applied Physics 2014,
115, 124506.
Heinze, S.; Tersoff, J.; Martel, R.; Derycke, V.; Appenzeller, J.; Avouris, P. Phys. Rev. Lett. 2002,
89, 106801.
Amer, M.; Bushmaker, A.; Cronin, S. Nano Res. 2012, 5, 172-180.
McClain, D.; Thomas, N.; Youkey, S.; Schaller, R.; Jiao, J.; O'Brien, K. P. Carbon 2009, 47, 1493-
1500.
Zhang, J. L.; Wang, C.; Fu, Y.; He, Y. C.; Zhou, C. W. ACS Nano 2011, 5, 3284-3292.
Moriyama, N.; Ohno, Y.; Suzuki, K.; Kishimoto, S.; Mizutani, T. Appl. Phys. Express 2010, 3,
105102.
Abstract (if available)
Abstract
In this dissertation, I cover the study of two promising one dimensional materials, gallium arsenide nanowires and carbon nanotubes, on their optoelectronic and electronic applications. Nanowires, due to their small footprint and unique optical and electronic properties, are promising candidates for next-generation high efficiency and cost effective solar cells. The relief of lattice mismatch between different crystallographic materials from the nanowire and planar substrate enables optimal multi-junction structure. Besides, the nanowire array can act as an anti-reflection layer, which prevents excessive loss of the reflected light. Carbon nanotubes, on the other hand, is another one dimensional materials which possesses transparency, high mobility, flexible and ballistic transport properties. These desired features make carbon nanotube devices promising in both macro-electronics and micro-electronics applications. ❧ In my dissertation, for the GaAs nanowire part, I systematically demonstrated how my coworkers and I progressively approached the goal of III-V nanowire-on-Si tandem solar cells. For the carbon nanotube part, I demonstrated the integration of carbon nanotube thin film transistors (TFTs) with polymer dispersed liquid crystal (PDLC) transparent display for the macro-electronics application. Besides, I also studied the high on/off ratio, short channel, carbon nanotube field effect transistors (CNTFETs) based on floating evaporative self aligned (FESA) tubes for micro-electronics applications. The dissertation is comprised of 6 chapters. Chapters 1-3 talk about GaAs nanowires for optoelectronic applications. Chapter 1 gives the introduction of nanowire solar cells and the axial p-i-n junction GaAs nanowire array solar cells. This is the foundation of the III-V nanowire-on-Si tandem solar cells. With junction depth and diameter optimizations, the efficiency can achieve as high as 7.58%. Chapter 2 demonstrates tandem solar cells using GaAs nanowire on Si. The observation of voltage addition and the improved efficiency of 11.4% confirm the concept of III-V nanowire-on-Si tandem solar cells. Chapter 3 covers the study of the growth of lattice mismatched GaAs nanowires on Si substrates. The facile five segment growth simplifies previous seven segment growth profile. The study of twin formation has led to twin-free segments up to 80 nm, which is the highest reported so far for selective area growth (SAG). Chapters 4 and 5 talk about carbon nanotubes. Chapter 4 gives an introduction to carbon nanotubes and the carbon nanotube thin film transistors (CNTTFTs). One of the promising applications of CNTTFTs is the driver circuits composed of CNTTFTs used to control flexible and transparent displays. Here, I used the polymer dispersed liquid crystal (PDLC) flexible transparent technology, and successfully integrated the CNTTFTs as the driving circuits for the active-matrix seven-segment PDLC display. Chapter 5 talks about another important potential application of carbon nanotubes. Thanks to their ballistic transport properties, carbon nanotubes have a great potential to replace Si in high speed, low energy computing after Moore’s law. Using high semiconducting purity, highly aligned tubes assembled by the floating evaporative self assembly (FESA) method, the CNTFETs have demonstrated performance exceeding Si. On the other hand, problems still remain to fully realize large scale integration. CNTFETs made from FESA-assembled tubes suffer from degradation of on/off ratio as the channel length shrinks, especially at high drain-source voltage (Vds). I studied the cause of the degradation and concluded the effect of oxygen and moisture are the main reasons for the degradation. When measured in argon, minimizing the effects of oxygen and moisture, the 40-nm CNTFET can demonstrate on/off ratio as high as 1.29 × 10⁴ at Vds = −0.5 V, a two order of magnitude improvement compared to measurements in air. Besides, negative differential resistance (NDR) phenomenon associated with gas desorption was also observed. Chapter 6 concludes the dissertation and gives a summary of the GaAs nanowire optoelectronic and carbon nanotube electronic device applications. In the outlook section, I talked about the progress of the GaAs nanowire solar cell and carbon nanotube electronics, and the future opportunities in these fields.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Growth, characterization of gallium arsenide based nanowires and application in photovoltaic cells
PDF
Carbon nanotube macroelectronics
PDF
Electronic and optoelectronic devices based on quasi-metallic carbon nanotubes
PDF
Carbon nanotube nanoelectronics and macroelectronics
PDF
Multilayer grown ultrathin nanostructured GaAs solar cells towards high-efficiency, cost-competitive III-V photovoltaics
PDF
Printed and flexible carbon nanotube macroelectronics
PDF
Nanomaterials for macroelectronics and energy storage device
PDF
Printed electronics based on carbon nanotubes and two-dimensional transition metal dichalcogenides
PDF
Carbon material-based nanoelectronics
PDF
Single-wall carbon nanotubes separation and their device study
PDF
Optical, mechanical, and electrical properties of nano-structured materials
PDF
Controlled synthesis, characterization and applications of carbon nanotubes
PDF
Synthesis, assembly, and applications of single-walled carbon nanotube
PDF
A study of junction effect transistors and their roles in carbon nanotube field emission cathodes in compact pulsed power applications
PDF
One-dimensional nanomaterials for electronic and sensing applications
PDF
Optoelectronic, thermoelectric, and photocatalytic properties of low dimensional materials
PDF
Zero-dimensional and one-dimensional nanostructured materials for application in photovoltaic cells
PDF
Raman spectroscopy and electrical transport in suspended carbon nanotube field effect transistors under applied bias and gate voltages
PDF
Light management in nanostructures: nanowire solar cells and optical epitaxial growth
PDF
Optoelectronic properties and device physics of individual suspended carbon nanotubes
Asset Metadata
Creator
Cong, Sen
(author)
Core Title
GaAs nanowire optoelectronic and carbon nanotube electronic device applications
School
Viterbi School of Engineering
Degree
Doctor of Philosophy
Degree Program
Electrical Engineering
Publication Date
02/14/2018
Defense Date
12/07/2017
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
aligned tubes,axial junction,carbon nanotube,field effect transistor,floating evaporative self assembly,gallium arsenide,high density,micro-electronics,multi-junction,nano-electronics,nanowire,negative differential resistance,OAI-PMH Harvest,on/off ratio,optoelectronic,solar cell,thin film transistor,transparent display
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Zhou, Chongwu (
committee chair
), Cronin, Stephen Burke (
committee member
), Dapkus, Daniel Paul (
committee member
), Nakano, Aiichiro (
committee member
), Tanguay, Armand Rene, Jr. (
committee member
), Wu, Wei (
committee member
)
Creator Email
sencong@usc.edu,ss_cong@yahoo.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c40-475630
Unique identifier
UC11266467
Identifier
etd-CongSen-6039.pdf (filename),usctheses-c40-475630 (legacy record id)
Legacy Identifier
etd-CongSen-6039.pdf
Dmrecord
475630
Document Type
Dissertation
Rights
Cong, Sen
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
aligned tubes
axial junction
carbon nanotube
field effect transistor
floating evaporative self assembly
gallium arsenide
high density
micro-electronics
multi-junction
nano-electronics
nanowire
negative differential resistance
on/off ratio
optoelectronic
solar cell
thin film transistor
transparent display