Upload
others
View
23
Download
0
Embed Size (px)
Citation preview
THERMAL AND RADIATION EXPOSURE OF GRAPHENE-ENHANCED
GALLIUM NITRIDE ULTRAVIOLET PHOTODETECTORS FOR SPACE
EXPLORATION
A DISSERTATION
SUBMITTED TO THE DEPARTMENT OF
AERONAUTICS AND ASTRONAUTICS
AND THE COMMITTEE ON GRADUATE STUDIES
OF STANFORD UNIVERSTIY
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
Ruth A. Miller
March 2018
http://creativecommons.org/licenses/by-nc/3.0/us/
This dissertation is online at: http://purl.stanford.edu/hr475nc1100
© 2018 by Ruth Anne Miller. All Rights Reserved.
Re-distributed by Stanford University under license with the author.
This work is licensed under a Creative Commons Attribution-Noncommercial 3.0 United States License.
ii
I certify that I have read this dissertation and that, in my opinion, it is fully adequatein scope and quality as a dissertation for the degree of Doctor of Philosophy.
Debbie Senesky, Primary Adviser
I certify that I have read this dissertation and that, in my opinion, it is fully adequatein scope and quality as a dissertation for the degree of Doctor of Philosophy.
Fu-Kuo Chang
I certify that I have read this dissertation and that, in my opinion, it is fully adequatein scope and quality as a dissertation for the degree of Doctor of Philosophy.
Sigrid Close
Approved for the Stanford University Committee on Graduate Studies.
Patricia J. Gumport, Vice Provost for Graduate Education
This signature page was generated electronically upon submission of this dissertation in electronic format. An original signed hard copy of the signature page is on file inUniversity Archives.
iii
iv
Abstract
Space exploration, including gathering data and learning about Earth’s past and
possible future, sending humans to the surface of other planets, as well as searching for
life beyond Earth, requires sensors to operate reliably within the harsh environment of
space. Specifically, sensors operating within space need to withstand extreme
temperatures (from -229°C near Pluto to 460°C on the surface of Venus) and high levels
radiation (ultraviolet light, gamma rays, protons, electrons, neutrons, heavy ions, etc.)
as well as extreme pressures, high heat fluxes, and various forms of chemical and
physical corrosion. Currently, complex packaging schemes are used to keep
instrumentation at operable temperatures and shield against radiation at the cost of
added mass. By developing new sensor material platforms such as gallium nitride
(GaN), which is intrinsically radiation-hard and thermally stable, much of the extra
protective packaging can be removed. Additionally, due to its wide bandgap, GaN is
visible-blind and highly responsive to ultraviolet light making GaN an ideal material
platform for developing ultraviolet photodetectors for space-based applications.
Photodetectors that use traditional metal electrodes exhibit low quantum
efficiencies due to the metal electrodes blocking light from being absorbed by the
semiconductor. Here it is shown that graphene, a monolayer of carbon atoms, has high
v
transmission in the ultraviolet regime and creates a Schottky contact to GaN, resulting
in ultraviolet photodetectors with increased sensitivity. A microfabrication process for
manufacturing graphene-enhanced GaN MSM ultraviolet photodetectors as well as GaN
MSM photodetectors with semitransparent Ni/Au electrodes has been developed. Both
types of GaN photodetectors are characterized and compared while operating under
temperatures from room temperature up to 200°C. Thermionic, thermionic field, and
field emissions current transport models are examined and compared to the
experimental results to determine the temperature-dependent trends of GaN metal-
semiconductor-metal ultraviolet photodetectors. Additionally, the response of
graphene-enhanced and semitransparent Ni/Au GaN MSM photodetectors subjected to
2 MeV protons up to a fluence of 3.8 × 1013 cm-2 is examined through electrical
characterization and Raman spectroscopy of the graphene.
In addition to characterizing the response of GaN-based photodetectors
operating at high temperatures and when subjected to high levels of proton irradiation,
this thesis also details the implementation and testing of these sensors in flight-relevant
applications. A miniature sensor for detecting the orientation of incident ultraviolet light
was created with a focal plane consisting of a 3 × 3 array of microfabricated GaN-based
photodetectors. The photodetector array was integrated with an aperture mask to realize
a miniature sun sensor capable of determining incident light angles with a ±45° field of
view. The miniature sun sensor was integrated with a CubeSat to test its performance in
space. GaN photodetectors were also used for in situ measurements of ultraviolet shock-
layer radiation in a Titan simulant atmosphere in an Electric Arc Shock Tube in support
of atmospheric entry sensing technologies.
vi
Acknowledgments
First, I would like to acknowledge my advisor Professor Debbie Senesky. Thank
you for allowing me to join the Extreme Environment Microsystems Laboratory (XLab)
in its early days. Thank you for helping me navigate the interdisciplinary nature of my
interests and the XLab’s research goals. Thank you for pushing me to achieve more than
I ever thought I could (I didn’t think I’d be doing semiconductor device physics when I
entered grad school!) while allowing me to explore what I was interested in (Everything
and anything space and NASA related!). Thank you for being an amazing mentor and
friend on this crazy journey of grad school.
Thank you to Professor Sigrid Close and Professor Fu-Kuo Chang for your
support throughout my time at Stanford. Thank you serving on my committee and
providing your insightful comments as well and challenging questions.
I would like to thank the staff of the Stanford Nanofabrication Facility (SNF)
and Stanford Nano Shared Facilities (SNSF) for training me to use the microfabrication
equipment and helping develop my fabrication process. I would also like to express my
gratitude to Pauline Prather for her expert wirebonding and encapsulation without which
much of the harsh environment characterization would not have been possible.
vii
I would like to thank Professor Andrew Kalman for allowing me to incorporate
my miniature sun sensor on-board the QB50 CubeSat Discovery as well as teaching me
about CubeSats and developing flight software. I owe more thank yous than I can give
to David Wright and Monica Hew for working tirelessly to develop the electronics and
software to support the miniature sun sensor. I’m so grateful to have worked with both
of you. Also, many thanks to Dr. Nicolas Lee and Ana Tarano for managing the QB50
project and keeping me informed of all of the ups and downs.
I would like to thank Dr. Brett Cruden of NASA Ames research center for
teaching me about optical spectroscopy, allowing me to test my photodetectors in Ames’
Electric Arc Shock Tube (EAST) facility, and patiently answering my countless
questions. Your mentorship has been invaluable in my PhD and future career. Thank
you to Ramon Martinez for your willingness to help me conduct my experiments. Thank
you to Dr. David Hash for managing a fantastic internship program within the
Aerothermodynamics branch at Ames. Also, thanks to Mark McGlaughlin and Rick
Ryzinga for expert maintenance and operation of the EAST facility.
I would like to thank Yongqiang Wang, Joe Tesmer, and Bob Houlton of the Ion
Beam Materials Laboratory, Center for Integrated Nanotechnologies at Los Alamos
National Laboratory, who helped perform proton beam irradiation.
I would like to thank all of my XLab colleagues, past and present. I have learned
so much from all of you and I’m incredibly grateful to have been able to work with each
of you. Special thanks to Caitlin Chapin, Ateeq Suria, Minmin Hou, Ashwin Shankar,
and Karen Dowling who have been around since the beginning of the XLab. You have
all been a constant source of inspiration, knowledge, and made the XLab a wonderful
place to work. Also, the countless hours spent together in the cleanroom will not be
forgotten. Thank you to Dr. Heather Chiamori for introducing me to graphene and the
transfer process. Thank you to Dr. Hongyun So for your willingness to help as well as
reviewing and improving my papers.
viii
I am very grateful for all of the experiences I had during my time at Stanford,
inside and outside of the classroom and lab. One of the many reasons I was driven to
pursue a degree in aerospace engineering was to launch a rocket, which I was able to do
thanks to AA284. Designing, building, and launching a nitrox-paraffin hybrid rocket in
only two quarters was truly one of the most rewarding experiences. I would like to thank
everyone involved with AA284 especially my team. I will never forget the late night
testing on top of the parking structure and the bittersweet ending in the Mojave.
Participating in the young astronauts program, teaching science and engineering to local
grade school children, was another highly rewarding experience. Seeing their wide-eyed
wonderment and enthusiasm reminded me why I am here. I would also like to thank all
of my amazing friends who remind me there’s life outside of research.
Last and most important, I need to thank my family. I would not be here without
you, this degree is just as much yours as it is mine. Thank you for always encouraging
me to follow my dreams.
ix
Contents
1 Introduction 1
1.1 The Need for Space-Grade Instrumentation .................................................... 1
1.2 State-of-the-Art Sensor and Electronics .......................................................... 4
1.3 Emerging Material Platforms for Space Exploration ...................................... 5
1.4 Ultraviolet Photodetectors ............................................................................... 7
1.5 Graphene as a Space-Grade Material ............................................................. 15
1.6 Thesis Outline ................................................................................................ 16
2 GaN MSM Photodetector Operation and Modeling 18
2.1 Principal of Operation .................................................................................... 18
2.2 Semi-Analytical Model .................................................................................. 22
2.2.1 Dark Current Model ........................................................................ 22
Abstract iv
Acknowledgments vi
List of Tables xiii
List of Figures xv
x
2.2.2 Dark Current Results ....................................................................... 27
2.2.3 Photocurrent Model ......................................................................... 31
2.2.4 Photocurrent Results ....................................................................... 31
2.3 Conclusions .................................................................................................... 34
3 Fabrication and Spectral Characterization 36
3.1 Graphene-Enhanced GaN Photodetector Fabrication .................................... 36
3.1.1 Microfabrication Process ................................................................ 36
3.1.2 Graphene Transfer Process ............................................................. 38
3.2 Optical Properties and Spectral Response ..................................................... 40
3.2.1 Optical Properties of GaN Substrate ............................................... 40
3.2.2 Optical Properties of Graphene and Semitransparent Ni/Au .......... 41
3.2.3 Spectral Response Test Set-up ........................................................ 43
3.2.4 GaN Photodetector Spectral Response ............................................ 44
3.3 Conclusions .................................................................................................... 48
4 Extreme Environment Operation 50
4.1 High-Temperature Exposure .......................................................................... 50
4.1.1 Experimental Set-Up ....................................................................... 50
4.1.2 Results ............................................................................................. 51
4.2 Proton Irradiation Testing .............................................................................. 60
4.2.1 The Space Radiation Environment .................................................. 60
4.2.2 Test Set-Up ...................................................................................... 62
4.2.3 Electrical Characterization .............................................................. 63
4.2.4 Raman Spectroscopy ....................................................................... 65
4.3 Conclusions .................................................................................................... 68
5 Implementation 70
5.1 Miniature Sun Sensor ..................................................................................... 70
5.1.1 Current Sun Sensor Technology ..................................................... 70
xi
5.1.2 Principal of Operation ..................................................................... 73
5.1.3 Array Characterization .................................................................... 75
5.1.4 Miniature Sun Sensor Performance ................................................ 77
5.2 CubeSat Integration and Ground Testing ...................................................... 81
5.2.1 QB50 Mission ................................................................................. 81
5.2.2 Miniature Sun Sensor Integration ................................................... 82
5.2.3 Ground Test Results ........................................................................ 83
5.3 In Situ Shock-Layer Radiation Measurements .............................................. 86
5.3.1 Current Atmospheric Entry Sensing Technology ........................... 86
5.3.1 Shock Tube Test Set-up .................................................................. 89
5.3.3 Shock Tube Test Results ................................................................. 91
5.4 Conclusions .................................................................................................... 95
6 Conclusions and Future Work 97
6.1 Conclusions .................................................................................................... 97
6.2 Future Work ................................................................................................. 100
A MSM Photodetector Analytical Model Equations 103
B AlGaN/GaN-Based Photodetectors 108
B.1 Principal of Operation .................................................................................. 108
B.2 Fabrication and Spectral Characterization ................................................... 110
B.3 Extreme Environment Operation ................................................................. 115
B.3.1 AlGaN/GaN Phototransistor High-Temperature Testing .............. 115
B.3.2 AlGaN/GaN MSM Low-Temperature Testing ............................. 120
B.3.3 AlGaN/GaN MSM Proton Irradiation ........................................... 125
C Experiment Drawings 128
D Fabrication Run Sheet 139
xii
Bibliography 149
xiii
List of Tables
Table 1.1.1. Extreme environments experienced at planetary bodies. ........................... 3
Table 1.3.1. Semiconductor material properties comparison. ........................................ 6
Table 2.2.1. Parameter values for carrier mobility model of wurtzite GaN [94]. ........ 27
Table 2.2.2. Material properties of GaN used in MSM photodetector modeling. ........ 27
Table 4.1.1. Decay times of the graphene/GaN and semitransparent Ni/Au/GaN
photodetectors from 23°C to 200°C. ................................................................ 59
Table 4.2.1. Average pre- and 30 days post-irradiation PDCR and responsivity for
graphene and semi-transparent Ni/Au GaN-on-sapphire photodetectors. The
graphene electrodes were damaged post-irradiation during transport. ............. 65
Table 4.2.2. Proton implantation depth and material as a function of energy for the GaN-
on-sapphire substrate as calculated using the Stopping and Range of Ions in
Matter (SRIM) software. .................................................................................. 66
Table 5.1.1. Sun sensor technology comparison. ......................................................... 72
Table 5.1.2. Miniature sun sensor dimensions. ............................................................ 75
Table 5.1.3. Photodetector PDCR and responsivity at -1 V bias before attaching the
aperture mask. PD 1 was unintentionally scratched during electrical
measurements. .................................................................................................. 76
xiv
Table 5.2.1. Photodetector PDCR and responsivity at 5 V bias before attaching the
aperture mask to the QB50 miniature sun sensor. Three of the graphene
photodetectors (PD 4, PD 7 and PD 9) were not operational due to poor graphene
transfer. ............................................................................................................. 84
Table 5.2.2. “SUN ANGLE ARRAY” of the miniature sun sensor integrated with the
QB50 CubeSat with illumination angle of 0° and 5 V bias. ............................. 85
Table 5.3.1. Spacecraft to date that have or are planned to measure in-flight shock-layer
radiation during atmospheric entry. .................................................................. 88
Table 5.3.2. Shock test parameters. .............................................................................. 91
Table B.3.1. Measured thermocouple temperature versus chuck set temperature. .... 122
Table B.3.2. Average pre- and 30 days post-irradiation PDCR and responsivity for semi-
transparent Ni/Au AlGaN/GaN on GaN-on-sapphire photodetectors. ........... 127
xv
List of Figures
Figure 1.1.1. Temperature extremes and yearly radiation doses experienced at each
planet. Image adapted from discovery.com. ....................................................... 2
Figure 1.4.1. Black body radiation (Plank’s law) at 5800 K approximating the solar
irradiance spectrum outside Earth’s atmosphere with absorption
wavelengths/energies of GaN and Si indicated. ................................................. 8
Figure 1.4.2. Schematic structure of photoconductor, MSM, Schottky, p-n, p-i-n, and
avalanche semiconductor photodetectors. .......................................................... 9
Figure 1.4.3. Responsivity comparison of SiC, ZnO, GaN, and AlN photodetectors
as reported in literature with corresponding 50% and 100% quantum efficiency
values for each material indicated. ................................................................... 11
Figure 1.4.4. Maximum reported operational temperature comparison of Si, SiC,
ZnO, and III-nitride photodetectors. ................................................................. 12
Figure 2.1.1. Fundamental operating principal of semiconductor photodetectors. .. 19
Figure 2.1.2. Current transport mechanisms in metal-semiconductor junctions. ..... 20
Figure 2.1.3. MSM photodetector band diagram. .................................................... 22
Figure 2.2.1. GaN’s bandgap as a function of temperature. ..................................... 24
Figure 2.2.2. Intrinsic carrier concentration of GaN as a function of temperature. . 25
xvi
Figure 2.2.3. GaN (a) electron and (b) hole mobility as functions of temperature. . 26
Figure 2.2.4. Ni/GaN (a) electron and (b) hole barrier height as functions of
temperature. ...................................................................................................... 27
Figure 2.2.5. Semi-analytical model of GaN MSM photodetector dark current-
voltage from -100°C to 200°C for (a) thermionic emission, (b) thermionic field
emission, and (c) field emission. Note that the data shown in the left side and
right side of (a), (b), and (c) is the same. However, the figures on the right
employ a logarithmic current axis for clarity. .................................................. 29
Figure 2.2.6. Semi-analytical model of GaN MSM photodetector dark current-
voltage from -100°C to 200°C. ......................................................................... 30
Figure 2.2.7. Semi-analytical model of GaN MSM photodetector total dark current
vs. voltage for temperatures from -100°C to 200°C with (a) linear current axis
and (b) logarithmic current axis. ...................................................................... 30
Figure 2.2.8. GaN MSM photodetector dark and 365 nm illuminated current-voltage
response for temperatures from -100°C to 200°C with (a) linear current axis and
(b) logarithmic current axis. ............................................................................. 32
Figure 2.2.9. GaN MSM photodetector (a) PDCR and (b) responsivity as functions
of temperature at 200 mV bias. ........................................................................ 33
Figure 2.2.10. GaN MSM photodetector (a) PDCR and (b) responsivity as functions
of voltage at 23°C. ............................................................................................ 34
Figure 3.1.1. GaN-on-sapphire MSM photodetector microfabrication process: (a)
GaN-on-sapphire chip, (b) mesa etch for device isolation, (c) deposited SiO2
passivation layer, (d) passivation etch, (e) semitransparent Ni/Au or graphene
electrodes, and (f) deposited Ti/Au electrical contacts and interconnect.
Reprinted from [105], with the permission of AIP Publishing. ....................... 37
xvii
Figure 3.1.2. Top-view optical image of a GaN MSM photodetector with
semitransparent Ni/Au electrodes. Reprinted from [105], with the permission of
AIP Publishing. ................................................................................................. 38
Figure 3.1.3. Graphene transfer process: (a) graphene grown on copper foil via
chemical vapor deposition, (b) spin on PMMA, (c) oxygen plasma etch to
remove backside graphene, (d) etch copper foil, (e) rinse in deionized water, (f)
transfer graphene/PMMA stack to substrate, and (g) remove PMMA. ............ 39
Figure 3.1.4. (a) Top-view optical image and (b) top-view SEM of graphene-
enhanced GaN MSM photodetector. ................................................................ 40
Figure 3.2.1. Optical properties of the GaN-on-sapphire substrate: (a) absorbance and
reflectance spectra showing a cutoff wavelength around 370 nm and (b)
transmission spectrum and (αhν)2 versus hν with extrapolation of the linear
region (red dotted line) showing a bandgap of 3.34 eV. Reprinted from [105],
with the permission of AIP Publishing. ............................................................ 41
Figure 3.2.2. (a) Transmission spectra of graphene and semitransparent Ni/Au
(3 nm / 10 nm) on 1 mm thick glass and (b) corrected transmission spectra of
graphene and semitransparent Ni/Au. Reprinted from [117], with the permission
of AIP Publishing. ............................................................................................ 42
Figure 3.2.3. Cross-sectional schematic of (a) a graphene/GaN and (b) a
semitransparent/Ni/Au/GaN MSM photodetector. Reprinted from [117], with
the permission of AIP Publishing. .................................................................... 42
Figure 3.2.4. Benchtop test set-up for spectral response characterization of GaN-
based photodetectors with the applied optical power incident on the GaN-based
photodetectors shown in the inset. Reprinted from [118], with the permission of
AIP Publishing. ................................................................................................. 44
xviii
Figure 3.2.5. Current-voltage response of four semitransparent Ni/Au GaN-on-
sapphire MSM photodetectors on the same chip when unilluminated (dark) and
illuminated with wavelengths from 200 nm to 500 nm. ................................... 45
Figure 3.2.6. NPDR as a function of illumination wavelength for four
semitransparent Ni/Au GaN-on-sapphire MSM photodetectors. ..................... 47
Figure 3.2.7. Spectral responsivity of four semitransparent Ni/Au GaN-on-sapphire
MSM photodetectors. ....................................................................................... 48
Figure 4.1.1. Experimental test set-up for characterizing the elevated temperature
response of GaN-based photodetectors using a high-temperature probe
station…. .......................................................................................................... 51
Figure 4.1.2. (a) Graphene-enhanced and (b) semitransparent Ni/Au GaN-on-
sapphire MSM photodetector current-voltage curves under dark (dashed lines)
and 365 nm illuminated (solid lines) conditions from room temperature to
200°C… ............................................................................................................ 52
Figure 4.1.3. PDCR and responsivity as functions of voltage at temperatures from
23°C to 200°C for (a) and (b) a graphene/GaN photodetector and (c) and (d) a
semitransparent Ni/Au/GaN photodetector. ..................................................... 53
Figure 4.1.4. (a) PDCR and (b) responsivity as functions of temperature from room
temperature to 200°C for graphene-enhanced (shown in the inset) and
semitransparent Ni/Au GaN-on-sapphire MSM photodetectors. ..................... 55
Figure 4.1.5. Extracted barrier height as a function of temperature for (a)
graphene/GaN and (b) semitransparent Ni/Au/GaN photodetectors. ............... 56
Figure 4.1.6 GaN-based photodetectors reported in literature with responsivities
corresponding to quantum efficiencies greater than 100%. ............................. 57
Figure 4.1.7 Trapped photo-generated holes causing lowering or bending of the
Schottky barrier height resulting in an internal photo-gain. ............................. 58
xix
Figure 4.1.8. Transient current response of the (a) graphene-enhanced and (b)
semitransparent Ni/Au GaN-on-sapphire photodetectors from room temperature
to 200°C under a 365 nm UV intensity of 0.7 ± 0.1 mW/cm2. ........................ 59
Figure 4.2.1. One year trapped proton fluence as a function of energy and aluminum
shielding thickness for (a) low Earth orbit approximating the international space
station orbit and (b) Jupiter orbit approximating the science orbit of the Juno
spacecraft; calculated using the SPace ENVironment Information System
(SPENVIS) software. ....................................................................................... 61
Figure 4.2.2. (a) GaN-based photodetectors mounted inside the tandem ion
accelerator chamber at Los Alamos National Laboratory undergoing proton
irradiation and (b) test set-up for electrical characterization of proton irradiated
photodetectors. .................................................................................................. 62
Figure 4.2.3. Dark and illuminated (365 nm) scaled current-voltage response as a
function of proton irradiation fluence for GaN-on-sapphire photodetectors with
(a) graphene electrodes and (b) semitransparent Ni/Au electrodes. Reprinted
from [117], with the permission of AIP Publishing. ........................................ 64
Figure 4.2.4. (a) PDCR and (b) responsivity as a function of proton fluence for
graphene and semitransparent Ni/Au GaN-on-sapphire photodetectors.
Reprinted from [117], with the permission of AIP Publishing. ....................... 65
Figure 4.2.5. SRIM ion distribution results for (a) 200 keV and (b) 2 MeV protons
impingent on graphene on a 5 μm GaN, 200 nm AlN and 500 μm sapphire
substrate. For 200 keV protons the stopping depth and thus majority of lattice
displacement damage occurs at approximately 2 μm which is within the GaN.
Whereas for 2 MeV protons, the stopping depth is in the sapphire substrate at
52.3 μm. ............................................................................................................ 67
xx
Figure 4.2.6. Raman spectroscopy comparison of graphene samples before and after
200 keV proton irradiation up to a fluence of 3.8 × 1013 cm-2. Reprinted from
[117], with the permission of AIP Publishing. ................................................. 68
Figure 5.1.1. Cross-sectional schematic of a GaN-based miniature sun sensor
depicting the principle of operation and design parameters using an aperture
mask, spacer, and focal plane. The inset shows the top view of a GaN-on-
sapphire MSM photodetector. Reprinted from [105], with the permission of AIP
Publishing. ........................................................................................................ 74
Figure 5.1.2. Optical image of 3 × 3 GaN-based MSM photodetector array. Reprinted
from [105], with the permission of AIP Publishing. ........................................ 75
Figure 5.1.3. Scaled current-voltage response of all nine photodetectors in the array
under dark and illuminated conditions before attaching the aperture mask.
Reprinted from [105], with the permission of AIP Publishing. ....................... 77
Figure 5.1.4. Image of assembled miniature sun sensor with size comparison to a
penny. Reprinted from [105], with the permission of AIP Publishing. ............ 78
Figure 5.1.5. Scaled and normalized PDCR and responsivity comparisons of all nine
photodetectors when miniature sun sensor is illuminated (365 nm) at ((a) and
(b)) θsun = 0° and ((c) and (d)) θsun = 45°. Reprinted from [105], with the
permission of AIP Publishing. .......................................................................... 80
Figure 5.1.6. Miniature sun sensor operation at illumination angles (θsun) of (a) 0°
(illuminating PD 5) and (b) 45° (illuminating PD 2). The left-side images
indicate the illumination angle/orientation and the right-side images show the
illuminated array based on the normalized PDCR and responsivity at −1 V.
Reprinted from [105], with the permission of AIP Publishing. ....................... 81
Figure 5.2.1. QB50 CubeSat Discovery CAD model with GaN-based miniature sun
sensor and finished CubeSat. ............................................................................ 83
xxi
Figure 5.2.2. Dark and illuminated (365 nm) scaled current-voltage response of (a)
graphene and (b) semitransparent Ni/Au photodetector arrays before attaching
the aperture mask and before QB50 implementation. Three of the graphene
photodetectors (PD 4, PD 7 and PD 9) were not operational due to poor graphene
transfer.. ............................................................................................................ 84
Figure 5.2.3. Ground test set-up for integration testing of GaN sun sensor with QB50
Discovery. ......................................................................................................... 85
Figure 5.2.4. Ground test performance of miniature sun sensor integrated with QB50
at 0° illumination angle and 5 V bias; (a) graphene photodetector array and (b)
semitransparent Ni/Au photodetector array. Three of the graphene
photodetectors (PD 4, PD 7, and PD 9) were not operational due to poor
graphene transfer. ............................................................................................. 86
Figure 5.3.1. Timeline of spacecraft that have or are planned to measure in-flight
shock-layer radiation during atmospheric entry. .............................................. 88
Figure 5.3.2. Test set-up for in situ UV shock radiance measurements in NASA
Ames’ electric arc shock tube. Reprinted from [118], with the permission of AIP
Publishing. ........................................................................................................ 90
Figure 5.3.3. Transient current response (8 Hz sampling frequency) for (a)
photodetector 1 during shock test 1 and (b) photodetector 2 during shock test 2.
Unilluminated pre- and post-shock current-voltage curves (device functionality
tests) are compared in the insets. ...................................................................... 92
Figure 5.3.4. Transient current response (5 MHz sampling frequency) during shock
test for (a) photodetector 1 and (b) photodetector 2. The filtered signal displays
a distinct increase in current after the shock passes the detector giving a direct
measure of the UV shock-layer radiation. Reprinted from [118], with the
permission of AIP Publishing. .......................................................................... 92
xxii
Figure 5.3.5. Photo-generated current (Iphoto - Idark) for the GaN-based photodetectors
as a function of applied optical power calculated as a square root of power fit
through the spectral response characterization data at 365 nm. By overlaying the
photo-generated current calculated from the shock test data for (a) photodetector
1/shock test 1 and (b) photodetector 2/shock test 2, the measured optical power
is found to be about 0.30 μW for both shock tests. Reprinted from [118], with
the permission of AIP Publishing. .................................................................... 93
Figure 5.3.6. GaN photodetector chip epoxied to header mounted in shock tube port
holder (a) before shock testing, (b) post shock test 1, and (c) post shock test
2……… ............................................................................................................ 95
Figure A.1.1. MSM photodetector band diagram. .................................................. 104
Figure B.2.1. AlGaN/GaN MSM photodetector fabrication process. (a) AlGaN/GaN
on GaN-on-sapphire substrate, (b) mesa etch, (c) SiO2 passivation, (d) etch SiO2
passivation, (e) deposit semitransparent Ni/Au interdigitated electrodes, and (f)
deposit Ti/Au electrical contacts and interconnect. From [193]; reprinted by
permission of the American Institute of Aeronautics and Astronautics, Inc. . 111
Figure B.2.2. Top-view optical image of an AlGaN/GaN MSM photodetector with
semitransparent Ni/Au electrodes. .................................................................. 111
Figure B.2.3. AlGaN/GaN phototransistor fabrication process. (a) AlGaN/GaN-on-
silicon substrate, (b) mesa etch, (c) deposit Ti/Al/Pt/Au source and drain
contacts, and (d) deposit Ti/Au gate contact. ................................................. 112
Figure B.2.4. Top-view optical image of an AlGaN/GaN phototransistor with (a)
200 μm × 100 μm rectangular gate and (b) 200 μm × 100 μm spine and rib
gate…… ......................................................................................................... 113
Figure B.2.5. Absorbance and reflectance spectra of the (a) AlGaN/GaN on GaN-on-
sapphire substrate and (b) AlGaN/GaN-on-silicon substrate. ........................ 113
xxiii
Figure B.2.6. Current-voltage response of two AlGaN/GaN-on-silicon MSM
photodetectors when illuminated with wavelengths from 200 nm to
500 nm……. ................................................................................................... 114
Figure B.2.7. PDCR as a function of illumination wavelength of two AlGaN/GaN-
on-silicon MSM photodetectors. .................................................................... 114
Figure B.2.8. Spectral responsivity of two AlGaN/GaN-on-silicon MSM
photodetectors. ................................................................................................ 115
Figure B.3.1. AlGaN/GaN phototransistor (200 μm × 100 μm, rectangular gate) (a)
drain current versus gate voltage and (b) drain current versus drain-to-source
bias under dark (dashed lines) and 365 nm illuminated (solid lines) conditions
at room temperature. ....................................................................................... 116
Figure B.3.2. AlGaN/GaN phototransistor (200 μm × 100 μm, spine and rib gate) (a)
drain current versus gate voltage and (b) drain current versus drain-to-source
bias under dark (dashed lines) and 365 nm illuminated (solid lines) conditions
at room temperature. ....................................................................................... 117
Figure B.3.3. AlGaN/GaN phototransistor (200 μm × 100 μm, rectangular gate) (a)
drain current versus gate voltage (VDS = 1 V) and (b) drain current versus drain-
to-source bias (VGS = 0 V) under dark (dashed lines) and 365 nm illuminated
(solid lines) conditions from room temperature to 300°C. ............................. 118
Figure B.3.4. AlGaN/GaN phototransistor (200 μm × 100 μm, spine and rib gate) (a)
drain current versus gate voltage (VDS = 1 V) and (b) drain current versus drain-
to-source bias (VGS = 0 V) under dark (dashed lines) and 365 nm illuminated
(solid lines) conditions from room temperature to 300°C. ............................. 118
Figure B.3.5. (a) Maximum PDCR as a function of temperature for four AlGaN/GaN
phototransistors with various gate configurations and (b) the corresponding gate
voltage as a function of temperature. ............................................................. 119
xxiv
Figure B.3.6. (a) Maximum responsivity as a function of temperature for four
AlGaN/GaN phototransistors with various gate configurations and (b) the
corresponding gate voltage as a function of temperature. .............................. 119
Figure B.3.7. (a) Overall experimental test set-up for characterizing the low-
temperature response of GaN-based photodetectors, (b) temperature controlled
stage (frozen over during testing), and (c) GaN-based photodetector chip
epoxied to a PCB. From [193]; reprinted by permission of the American Institute
of Aeronautics and Astronautics, Inc. ............................................................ 121
Figure B.3.8. AlGaN/GaN on GaN-on-sapphire MSM photodetector current-voltage
curves under dark (solid lines) and 365 nm illuminated (dashed lines) conditions
from room temperature to -138°C for (a) photodetector 1 and (b) photodetector
2. The inset in (a) shows the dark-current thermal recovery of photodetector 1
after exposure to -138°C. ................................................................................ 123
Figure B.3.9. (a) PDCR and (b) responsivity as a function of temperature at -8 V bias
for photodetectors 1 and 2. ............................................................................. 123
Figure B.3.10. Transient current response of photodetector 1 at different
temperatures under a 365 nm UV intensity of 0.7 ± 0.1 mW/cm2 and a bias of -
8 V. From [193]; reprinted by permission of the American Institute of
Aeronautics and Astronautics, Inc. ................................................................. 124
Figure B.3.11. Current-voltage response for photodetector 1 at 23.3°C and while in
liquid nitrogen vapor at -195.0°C. The inset shows the condensation on the GaN
chip and PCB from the liquid nitrogen. .......................................................... 125
Figure B.3.12. Dark and illuminated (365 nm) scaled current-voltage response as a
function of proton irradiation fluence for AlGaN/GaN on GaN-on-sapphire
photodetectors with semitransparent Ni/Au electrodes. ................................. 126
xxv
Figure B.3.13. (a) PDCR and (b) responsivity as a function of proton fluence for
three semitransparent Ni/Au AlGaN/GaN on GaN-on-sapphire
photodetectors… ............................................................................................. 127
Figure C.1. Full QB50 miniature sun sensor integration circuit diagram. .............. 129
Figure C.2. Drawings of spectrometer adapter for spectral response
characterization…. .......................................................................................... 130
Figure C.3. Shock tube assembly drawings. ............................................................ 135
1
Chapter 1
Introduction
1.1 The Need for Space-Grade Instrumentation
Sensors and electronics operating in space experience temperature extremes and
are exposed to high levels of radiation. Venus, with its runaway greenhouse gas effect,
has some of the highest temperatures seen in the solar system with an average surface
temperature of 460°C [1]. To date, the longest a Venus probe has survived on the surface
of Venus is only 2 hours and 7 minutes (Soviet Venera 13 in 1982) due to thermal failure
of the electronics [1]. On the other end of the temperature spectrum, spacecraft
exploring Mars can experience temperatures as low as -120°C and spacecraft exploring
Pluto can see temperatures as low as -229°C [2, 3]. In these very low-temperature
environments, some form of heating is required to keep electronics at operable
temperatures [4]. Future missions to Europa, one of Jupiter’s icy moons, would see
average surface temperatures of -160°C [5] while being exposed to one of the harshest
radiation environments in the solar system. Due to Jupiter’s strong magnetic field and
CHAPTER 1. INTRODUCTION 2
associated radiation belts, spacecraft orbiting the planet and its moons experience a dose
of about 40 krad/day of ionizing radiation [4, 6]. This high level of radiation is very
damaging to electronic components and thus extensive protective shielding is required.
The maximum and minimum temperatures as well as average yearly dose of radiation
for each planet is shown in Fig. 1.1.1. In addition to extreme temperatures and particle
radiation, sensors and electronics operating in space also need to withstand thermal
cycling, extreme pressures, high heat fluxes, high levels of electromagnetic radiation,
as well as various forms of chemical and physical corrosion. Some of the extreme
conditions experienced at several planetary bodies are listed in Table 1.1.1.
When mission designers plan a space mission, they may not need to consider
every harsh environmental condition. For example, high temperature and high pressure
operation is applicable to a Venus mission but for a Europa mission the ability to survive
in a low-temperature, radiation-rich environment is the primary concern. However, for
every space mission, reliable operation of sensors and electronics is important
throughout all phases of the mission including during ground testing (wind tunnels, arc
jets, shock tubes, vibration, thermal vacuum cycling, etc.), launch, in flight, during
atmospheric entry, and while at other planetary bodies (orbiters, rovers, landers, etc.).
Figure 1.1.1. Temperature extremes and yearly radiation doses experienced at each planet.
Image adapted from discovery.com.
CHAPTER 1. INTRODUCTION 3
Table 1.1.1. Extreme environments experienced at planetary bodies.
Average
Temp.
(°C)
High
Temp.
(°C)
Low
Temp.
(°C)
Radiation Dose
(krad/year)
Pressure
(bar) Corrosion
Mercury 167 [7]
425 to 426
[7, 8]
-193 to -
180 [7,
8]
100 to 200 [7] --- ---
Venus 460 to 470
[7, 8]
482 to 500
[4] 0 [4] 20 [7] 92 [7] H2SO4 [7]
Earth 17 [7] 70 [7] -89 [7]
LEO: 0.1 to 1
[7, 8]
MEO: 2 to 2,000
[7, 8]
GEO: 10 to 30 [7,
8]
--- ---
Moon 31 [4] 120 to 197
[4]
-233 to -
180 [4] --- --- Dust [7]
Mars -65 to -63
[4, 7]
20 to 27
[4, 7]
-143 to -
100 [4,
7]
5 to 10 [7, 8] 0.007 [7] Dust [7]
Jupiter -116 to -
110 [7, 9]
230 (Upper
Atm.) [4] to
1000’s
(Interior) [7]
-143 [7]
Orbit: 14,000
[7, 8]
Surface: 3,000 [7]
22 [7] ---
Europa -145 [8] --- -180 [4]
Orbit: 14,600 [4]
Surface: 7,300 [4]
--- ---
Saturn -147 to -
140 [7, 9]
1000’s
(Interior) [7] -140 [7] 30 [7] --- ---
Titan -179 [10] ---
-185 to -
178 [4,
10]
--- 1.5 [7] CH4 [7]
Uranus -197 [7] 1000’s
(Interior) [7] -224 [7] 10 [7] --- ---
Neptune -201 [7]
1000’s
(Interior) [7]
-218 [7] 10 [7] --- ---
CHAPTER 1. INTRODUCTION 4
Common spacecraft instrumentation includes temperature sensors, pressure sensors,
heat flux (convective and radiative) gauges, radiation (particle and electromagnetic)
detectors, and spectrometers (mass, absorption, emission, etc.). Data gathered by these
instruments are used to validate and improve models (aerodynamic,
aerothermodynamic, radiation, materials, mechanical, etc.) which enables improved
spacecraft design, furthers our scientific knowledge, and expands our understanding of
Earth and the solar system. If there was a one-size-fits-all material platform and
corresponding sensor package capable of withstanding the extreme temperatures, high
levels of radiation, and chemical attack experienced in space, the total footprint of
instrumentation would be reduced due to fewer packaging/shielding requirements.
Additionally, development and testing costs would be lower and more data could be
gathered.
1.2 State-of-the-Art Sensor and Electronics
Most sensors and electronics today are silicon (Si)-based due to well-established
manufacturing processes, easy circuit integration, and low cost. However, most
commercial-off-the-shelf (COTS) Si-based components are only rated to temperatures
up to 125°C [11] which is a limitation of the Si material platform. Si has a relatively
high intrinsic carrier concentration at room temperature (1010 cm-3) which increases
exponentially with temperature to reach 1015 cm-3 at 300°C [1]. Additionally, the
lightest doping concentrations of Si devices range from 1014 to 1017 cm-3 [12].
Therefore, when operating at high temperatures, the intrinsic carrier concentration will
overwhelm the doping concentration and the device will no longer function as intended.
Silicon-on-insulator (SOI) devices, which have a very thin active Si layer, have shown
operation up to 300°C as well as a higher tolerance for radiation than their bulk Si
counterparts [11]. However, many applications (especially space-based applications)
require much higher operating temperatures than these COTS components allow.
CHAPTER 1. INTRODUCTION 5
Another challenge electronics face when operating in the space environment is
radiation (electromagnetic and particle) which causes ionization damage (free electron-
hole pairs) and displacement damage (atoms displaced from their lattice sites) effects
within semiconductors [13]. Since Si has a narrow bandgap (1.12 eV) and a relatively
low atomic displacement energy (13 eV) [14], the energy required to displace an atom
from its lattice, Si-based electronics suffer from both ionization and displacement
damage effects and complete device failure occurs at low total doses. For low-Earth
orbit (LEO) CubeSats or other very short duration missions, COTS components work
well. However, for long duration and interplanetary type missions where electronics
failure is unacceptable, different solutions are needed. To avoid the cost and time
intensive nature of developing new technologies, the traditional solutions used by
spacecraft designers include incorporation of thick metal shielding to protect against
radiation (for example, the electronics on the Juno mission to Jupiter were housed inside
a titanium box [15, 16]), cooling/heating systems to keep electronics at operational
temperatures, and redundancy [4].
1.3 Emerging Material Platforms for Space Exploration
To extend the operating environments of electronics, wide bandgap materials
such as silicon carbide (SiC), zinc oxide (ZnO), and III-nitride materials (gallium nitride
(GaN), aluminum nitride (AlN), etc.) are being explored as material platforms for harsh
environment sensing applications. Table 1.3.1 lists relevant material properties of SiC,
ZnO, GaN, and AlN compared to Si. The wide bandgap and high melting point of these
materials indicates they will be able to operate at higher temperatures than Si. Similarly,
the high atomic displacement energies of the wide bandgap materials, compared to Si,
indicate they will be able to withstand higher doses of radiation before device failure.
CHAPTER 1. INTRODUCTION 6
Table 1.3.1. Semiconductor material properties comparison.
Si 4H-SiC ZnO GaN AlN
EG (eV) 1.12 [17] 3.2 [17] 3.35 [17] 3.39 [17] 6.2 [17]
Excitation
Wavelength (nm)
1108
(IR)
388
(UV)
370
(UV)
365
(UV)
200
(UV)
Melting Point (°C) 1410 [17] 2550 [17] 2248 [18] 2250 [19] 3000 [20]
Max Demonstrated
Operational
Temperature (°C)
300 (SOI)
[21, 11] 800 [22] ---
600 (AlGaN/GaN)
1000 (InAlN/GaN)
[23, 24]
---
Atomic Displacement
Energy (eV) 13 [14]
Si-35
C-22 [25]
Zn-18.5
O-41.4 [26]
Ga-20.5
N-10.8 [14, 27]
Al-42
N-38 [28]
III-nitride materials, their ternary alloys, and their associated heterostructures
are chemically, thermally, and mechanically stable [29]. Due to their robustness as well
as their excellent optical and electrical properties, III-nitride materials, similar to SiC
and ZnO, have found applications in power electronics, high-temperature electronics,
as well as sensing. AlGaN/GaN high electron mobility transistors (HEMTs) have been
shown operable at 600°C in air [23] and InAlN/GaN HEMTs have been shown operable
up to 1000°C in vacuum [24]. The higher operational temperatures of III-nitride-based
devices compared to Si, can be partially attributed to their much lower intrinsic carrier
concentrations (the intrinsic carrier concentration of GaN is about 105 cm-3 at 300°C)
[12]. In addition to high-temperature operation, III-nitride-based devices have been
shown to be robust to high levels of radiation [14, 30, 27, 31].
Due to the strong covalent bonds between silicon and carbon, SiC is very stable
at high temperatures [32]. Recently, SiC integrated circuits (ICs) have been shown
operable for over 1000 hours at 500°C in Earth-atmosphere and for over 500 hours in a
Venus simulate surface environment (460°C, 9.4 MPa, CO2, and SO2) [1]. Additionally,
SiC pressure sensors have demonstrated short term operation at 800°C and extended
(over 350 hours) operation at 600°C [22, 33]. SiC is also highly chemically stable [34]
(i.e. highly resistant to corrosion and biocompatible) making it an ideal material for use
CHAPTER 1. INTRODUCTION 7
as a chemical resistant coating in environments such as Venus with its sulfuric acid
clouds. However, from a sensor and electronic device microfabrication standpoint, the
high chemical stability of SiC (and GaN) means it is very difficult to etch.
Opposite to SiC (and GaN), ZnO is easily etched via wet chemical etching
making ZnO-based electronics easy to fabricate, but not well suited for use in corrosive
environments. However, ZnO-based devices have exhibited exceptionally high
tolerance to radiation; it is more resistant to radiation damage than Si, SiC, or GaN [35].
The theoretical atomic displacement energies for the Zn and O atoms have been
estimated to be 18.5 eV and 41.4 eV, respectively [26]. Experimentally, ZnO thin films
have shown minimal damage, in terms of electrical resistance and photoluminescence
intensity, when subject to 8 MeV protons with fluences as high as 5 × 1014 cm-2 [36].
Additionally, ZnO thin film transistors irradiated to 100 Mrad (gamma ray) showed only
a slight negative threshold voltage shift after irradiation which was completely
recovered with a one minute, 200°C anneal [37].
1.4 Ultraviolet Photodetectors
Due to their wide bandgap, SiC, ZnO, GaN, and AlN are sensitive to ultraviolet
(UV) light. On the electromagnetic spectrum, the UV region spans wavelengths from
400 nm to 10 nm and is typically divided into three regions: UV-A (400 - 320 nm), UV-
B (320 - 280 nm), and UV-C (280 - 10 nm) [17, 35, 38]. Outside the Earth’s
atmosphere, UV light makes up 9% of solar radiation. The approximate solar radiation
spectrum outside Earth’s atmosphere, calculated as black body radiation at 5800 K using
Plank’s law, is shown in Fig. 1.4.1 with the absorption wavelengths and energies of
GaN and Si indicated. The Earth’s ozone layer makes the planet habitable by humans
by almost completely absorbing UV-C and significantly attenuating UV-B, which
causes cataracts, burns, and skin cancer [17]. To assess the habitability of Mars for
future manned missions, the Mars Science Laboratory (MSL) was equipped with the
CHAPTER 1. INTRODUCTION 8
rover environmental monitoring station (REMS) which takes measurements of the UV
radiation reaching the surface of Mars [39]. In addition to understanding the UV
environment of other planets, there is a need to understand the UV radiative heating
during atmospheric entry in order to design effective and efficient spacecraft thermal
protection systems [40, 41]. For these applications, UV photodetectors which are highly
sensitive, visible-blind, and robust to harsh environmental conditions are needed.
Figure 1.4.1. Black body radiation (Plank’s law) at 5800 K approximating the solar
irradiance spectrum outside Earth’s atmosphere with absorption wavelengths/energies of
GaN and Si indicated.
There are different types of semiconductor photodetectors including:
photoconductor, metal-semiconductor-metal (MSM), Schottky, p-n, p-i-n, and
avalanche. The structure for each of these photodetectors is depicted in Fig. 1.4.2.
Regardless of the type, the fundamental operating principal of all semiconductor
photodetectors is the same. Photons with energy greater or equal to the bandgap of the
semiconductor are absorbed by the semiconductor and electron-hole pairs are generated.
Ideal photodetectors will generate one electron-hole pair per absorbed photon. The
photon excitation wavelength corresponding to the bandgap of the material is listed in
CHAPTER 1. INTRODUCTION 9
Table 1.3.1 for Si, SiC, ZnO, GaN, and AlN. These photo-generated carriers will be
separated by the electric field (electrons move to the conduction band and holes to the
valance band) and a photocurrent can be measured.
Figure 1.4.2. Schematic structure of photoconductor, MSM, Schottky, p-n, p-i-n, and
avalanche semiconductor photodetectors.
There are several key parameters used to characterize photodetector
performance, including: photocurrent-to-dark current ratio (PDCR), responsivity,
quantum efficiency, rise time, and decay time. PDCR, defined as
PDCR = 𝐼𝑝ℎ𝑜𝑡𝑜 − 𝐼𝑑𝑎𝑟𝑘
𝐼𝑑𝑎𝑟𝑘 (1.1)
where Iphoto is the photocurrent and Idark is the dark current, is a measure of photodetector
sensitivity [42, 43]. A second equally important sensitivity factor is responsivity (R)
which is defined as
R = 𝐼𝑝ℎ𝑜𝑡𝑜 − 𝐼𝑑𝑎𝑟𝑘
𝑃𝑜𝑝𝑡 (1.2)
CHAPTER 1. INTRODUCTION 10
where Popt is the applied optical power [17, 44]. Additionally, the external quantum
efficiency, η, (a measure of the number of charge carriers generated per incident
photon), which is directly proportional to the responsivity, is defined as
𝜂 = 𝑅ℎ𝑐
𝑞𝜆 (1.3)
where h is Plank’s constant, c is the speed of light, q is electron charge, and λ is
wavelength [44]. Lastly, the rise time (time in which photocurrent increases from 10%
to 90%) and decay time (time in which photocurrent drops from 90% to 10%) are
important parameters to consider especially in applications where a fast response time
is required.
Si and other narrow bandgap semiconductors are not able to directly measure
UV light due to their bandgaps corresponding to near infrared or visible wavelengths.
Since UV light is higher energy than the bandgap of Si and other narrow bandgap
materials, part of the energy will be lost to heating, resulting in low quantum efficiency.
To use these materials for UV photodetection, filtering devices that absorb UV light and
reemit lower energy light need to be incorporated [38]. Additionally, Si-based
photodetectors are limited to operational temperatures below 125°C due to the thermal
generation of charge carriers dominating the photo-generated response [42, 45]. Wide
bandgap materials should be used for UV detection, rather than Si, due to their ability
to directly measure UV light and intrinsically withstand harsh environmental conditions.
A comparison of SiC, ZnO, GaN, and AlN photodetector responsivities reported
in literature is shown in Fig. 1.4.3. Additionally, the 50% and 100% quantum efficiency
values for each of the four materials are indicated where the wavelength used for the
calculation was taken to be the absorption edge corresponding to the bandgap of the
material; these excitation values are as listed in Table 1.3.1. From Fig. 1.4.3, it is easy
to see that these wide bandgap materials have the potential for high performance (> 50%
quantum efficiency) but further development is required. There is not one type of
CHAPTER 1. INTRODUCTION 11
photodetector nor one material platform that appears to outperform the rest, in terms of
responsivity. Therefore, other factors such as dark current, response time, and
wavelength region of interest need to be considered. In general, photoconductors and
avalanche photodetectors have shown high internal gain, while Schottky photodetectors
have displayed very fast response times [44]. Furthermore, SiC photodetectors are
responsive to near-visible UV and AlN photodetectors are responsive to deep UV while
the absorption edge of GaN-based photodetectors is tunable for a wide range of
wavelengths (~200 nm to ~650 nm) [17].
Figure 1.4.3. Responsivity comparison of SiC, ZnO, GaN, and AlN photodetectors as
reported in literature with corresponding 50% and 100% quantum efficiency values for each
material indicated.
For space applications, the effect of extreme conditions on UV photodetector
performance is important to consider. As discussed above, Si-based photodetectors are
only able to operate up to 125°C, while wide bandgap photodetectors have shown
operational temperatures upwards of 200°C. A comparison of the maximum operational
temperatures reported in literature for Si, SiC, ZnO, and III-nitride photodetectors is
CHAPTER 1. INTRODUCTION 12
shown in Fig. 1.4.4. The benefits and limitations of each of these wide bandgap
materials for extreme environment UV photodetection will now be discussed.
Figure 1.4.4. Maximum reported operational temperature comparison of Si, SiC, ZnO, and
III-nitride photodetectors.
SiC is the most developed of the wide bandgap materials especially with regard
to operation in high ambient temperatures [12]. Thus it is not surprising to see from Fig.
1.4.4 that SiC-based photodetectors have demonstrated the highest operational
temperatures. SiC MSM photodetectors have demonstrated PDCRs as high as 1.3 × 105
at room temperature and up to 0.62 at 450°C [42]. However, at these high operational
temperatures, the wavelengths SiC photodetectors are responsive to has been shown to
shift significantly towards longer wavelengths due to the narrowing of the bandgap [46].
In addition to high-temperature operation, SiC UV detectors have been shown to retain
their operational characteristics after being exposed to 1 MeV neutrons with a fluence
of 8 × 1014 cm-2 [47], 2 MeV protons with a fluence of 1012 cm-2 [48], as well as under
167 MeV Xe heavy ions at a fluence of 6 × 109 cm-2 [49]. SiC and III-nitride materials
have theoretically and experimentally shown similar ability to operate at high
CHAPTER 1. INTRODUCTION 13
temperatures and when exposed to high levels of radiation. However, with further
development, III-nitride-based photodetectors may outperform SiC photodetectors due
to their direct and tunable bandgap as well as their ability to form heterostructures [50,
17].
GaN photodetectors have displayed exceptionally high responsivities
(> 103 A/W [51, 52]) which have been attributed to trapping of photo-generated holes
by surface states at the metal-semiconductor interface resulting in a lowering of the
Schottky barrier height and increased current [53, 54]. From spectral responsivity data,
the cut-off wavelength of AlGaN photoconductors has been shown to shift to shorter
wavelengths as the Al concentration increases from 0% (365 nm cut-off) to 35%
(325 nm cut-off) [17]. However, these devices show only a factor of 10 UV/visible
rejection due to material quality issues which remain to be surmounted in III-nitride
materials [12]. This limitation is readily evident in the very long fall times (on the order
of hours to days) for GaN and other III-nitride-based photodetectors which has been
attributed to impurities, dislocations, and deep-level defects trapping photo-generated
carriers for long times after the illumination source has been removed. This
phenomenon has been termed persistent photoconductivity [55, 56, 57, 58, 59] and
makes GaN photodetectors an unattractive solution for applications requiring fast
response times. Spectral responsivity measurements taken using a chopper with a lock-
in amplifier, in which all phenomena occurring at rates slower than the chopper
frequency are eliminated, results in UV/visible contrasts greater than three orders of
magnitude [17, 57, 60, 61]. Thus, GaN-based optoelectronic devices display great
potential but material quality and persistent photoconductivity issues need to be
overcome.
In terms of high-temperature characterization, GaN and AlGaN/GaN MSM
photodetectors have been shown operable up to 325°C [55] and InGaN Schottky
photodetectors have shown high responsivity (5.6 A/W) and high UV/visible rejection
(> 105) at 250°C [62]. Additionally, AlN MSM photodetectors have demonstrated
working temperatures up to 300°C [43]. At the other end of the temperature scale, there
CHAPTER 1. INTRODUCTION 14
have been very few studies characterizing the low-temperature operation of GaN
photodetectors which is an important environment to consider for space applications.
GaN p-i-n photodetectors have shown only a slight shift in the peak of the spectral
response curve from 362 nm at 7°C to 356 nm at -118°C but a 55% reduction in peak
magnitude due to a narrowing of depletion region at lower temperatures [63].
While III-nitride materials have high atomic displacement energies and III-
nitride-based devices have been shown to be robust to high levels of radiation [14, 30,
27, 31], there has been little work in understanding how radiation affects UV
photodetector performance. GaN MSM photodetectors irradiated to 750 krad (gamma
ray) show more than 60% decrease in PDCR over pre-irradiation values due to trapped
charges in a SiO2 passivation layer creating large leakage currents [64]. AlN MSM
photodetectors have demonstrated tolerance to 2 MeV proton irradiation up to a fluence
of 1013 cm-2, but at 1014 cm-2 PDCR drops to zero due to proton-induced displacement
damage. Clearly further work needs to be done to understand the operation of III-nitride
(as well as SiC and ZnO) photodetectors in harsh radiation (proton, neutron, gamma,
etc.) environments.
Compared to SiC and GaN, there have been relatively few reports on the
capability of ZnO photodetectors and even fewer reports on their operation in extreme
environments. ZnO photodetectors have exhibited a sharp cut-off at 365 nm with
UV/visible contrast of a few orders of magnitude [17] as well as operation up to 200°C
[65]. However, challenges in obtaining reliable p-type ZnO have limited the
development of p-n and p-i-n photodetectors [35, 66, 67]. Perhaps more attractive than
planar photodetectors on bulk ZnO, are ZnO-based nanostructures. ZnO nanostructures
enhance UV photodetector performance due to their large surface-to-volume ratio [35]
and ability to act as an antireflective coating [68, 69, 70] resulting in increased light
absorption. Compared to ZnO thin film photodetectors, ZnO nanorod arrays have shown
a much higher on/off ratio as well as faster response and decay times [71, 70]. Since
ZnO and GaN have low lattice mismatch (~1.8%) [35] and closely aligned bandgaps,
UV photodetectors with enhanced sensitivity can be created by integrating the two
CHAPTER 1. INTRODUCTION 15
materials. Indeed, ZnO nanorod arrays on GaN showed improved performance, in terms
of PDCR, at 300°C compared to bare GaN photodetectors [68].
While all of these wide bandgap materials have their benefits and further work
should be conducted to enable harsh environment UV photodetection by all materials,
this thesis will focus on the development, characterization, and applications of GaN-
based UV photodetectors within harsh space-like environments.
1.5 Graphene as a Space-Grade Material
Graphene is a two-dimensional material consisting of a monolayer of sp2-
bonded carbon atoms. Graphene is electrically and mechanically one of the strongest
materials and forms a Schottky contact to GaN [72, 73]. Graphene/GaN Schottky diodes
have been shown to retain rectifying properties up to 277°C and recover those properties
upon cooling from as high as 377°C [73]. In addition to thermal stability, graphene has
displayed resistance to radiation damage. Graphene has a high atomic displacement
energy of 18-22 eV [74] and has been shown to act as an effective barrier layer to
particle and high energy electromagnetic radiation [75, 76, 77, 78, 79]. By encapsulating
single layer molybdenum disulfide (MoS2) with graphene, damage from electron
irradiation was eliminated thus enabling high resolution, defect free transmission
electron microscopy (TEM) imaging [75]. Additionally, graphene was shown to protect
silver nanowire networks from high intensity (millions of W/cm2) UV laser irradiation
[76]. When exposed to low dose gamma rays, the carrier density of graphene increases
with little defect generation [77]. Due to the two-dimensional nature of graphene, the
probability of ion interaction is reduced and therefore high energy particle radiation
primarily creates defects in the substrate [77], underlying the importance of coupling
graphene with a radiation-hard substrate. These experimental results show that graphene
is an excellent material choice for use as a thermally stable and radiation-hard electrode
for GaN-based electronics.
CHAPTER 1. INTRODUCTION 16
Graphene is also well known as an optically transparent material across a broad
wavelength range while maintaining high electrical conductivity. Graphene alone is not
an excellent photodetector, but when coupled with a semiconductor material (such as
GaN) and used as a transparent electrode, highly responsive optoelectronics devices can
be created [80, 81, 72, 82, 83]. Graphene-based photodetectors exhibit fast response
times due to the high mobility of graphene [81]. Graphene/GaN photodetectors have
shown PDCRs as high as 56 at 10 V bias when illuminated with a 325 nm laser
(100 μW/μm2) [72] and Graphene/AlGaN/GaN Schottky photodetectors have
demonstrated photo-to-dark contrast ratios of more than three orders of magnitude [82].
To date, the response of Graphene/GaN UV photodetectors operating in extreme
temperature and radiation-rich environments has not been thoroughly investigated, if at
all. Since graphene has demonstrated excellent electrical and optical properties as well
as robustness to harsh environmental operating conditions, graphene-enhanced
photodetectors should be developed and characterized for space-based applications
which is the subject of study for this thesis.
1.6 Thesis Outline
Chapter 2 examines the principal of operation of GaN-based MSM
photodetectors and the development of a semi-analytical model based on thermionic
emissions is presented.
Chapter 3 describes the GaN photodetector microfabrication process including
the graphene transfer process. The detailed microfabrication run sheet is included in
Appendix D. Additionally, the optical properties of the GaN substrate are presented
along with a comparison of the transmission of semitransparent Ni/Au and graphene,
the two electrode materials examined in this thesis. The spectral response of the GaN
MSM photodetectors is also presented.
CHAPTER 1. INTRODUCTION 17
Chapter 4 delves into the characterization of GaN-based photodetectors when
operating in extreme environments including high temperatures and under proton
irradiation exposure. The performance of semitransparent Ni/Au electrodes is compared
to graphene electrodes for both extreme operating conditions.
Chapter 5 details the implementation and testing of GaN-based photodetectors
in practical aerospace applications. A miniature sun sensor was created using an array
of GaN-based photodetectors. The miniature sun sensor demonstrated operation under
0° and 45° illumination and has been integrated with a CubeSat enabling future in-space
evaluation. The GaN-based photodetectors were also implemented in a shock tube and
in situ measurements of UV shock-layer radiation were made.
Chapter 6 provides the conclusions and proposed future work to further the
development of GaN-based UV photodetectors for space applications.
18
Chapter 2
GaN MSM Photodetector Operation and Modeling
2.1 Principal of Operation
A photodetector is a device that converts light into a measurable electrical signal.
The fundamental operating principal of all semiconductor photodetectors is the same;
photons with energy greater or equal to the bandgap of the semiconductor are absorbed
by the semiconductor generating electron-hole pairs as shown in Fig. 2.1.1. When an
electric field is applied across the semiconductor, the photo-generated electron-hole
pairs separate and a photocurrent can be measured. Since the bandgap of GaN is 3.4 eV,
GaN-based photodetectors will be responsive to incident light with wavelengths of
365 nm or less. When there is no light or light with energy less than the bandgap of the
semiconductor shining on the photodetector, there will be a small leakage current
through the semiconductor film generated mostly due to thermal energy; this is termed
dark current. The magnitude of this dark current is determined by the metal-
semiconductor contact.
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 19
Figure 2.1.1. Fundamental operating principal of semiconductor photodetectors.
When a metal and a semiconductor are brought into contact, a potential barrier
known as the Schottky barrier (φB) is formed. When charge carriers have sufficient
energy to overcome the Schottky barrier height, a current will be generated. This form
of current transport is known as thermionic emissions and is the dominate current
transport mechanism for metal-semiconductor devices with low doping (N < 1017 cm-3)
or operating at high temperatures [44, 84]. Thermally excited carriers that don’t have
quite enough energy to overcome the barrier height see a thinner barrier and are able to
tunnel through it. This form of current transport is termed thermionic field emissions
and is the dominate form of current transport for devices with intermediate doping (1017
< N < 1018 cm-3) operating at intermediate temperatures [44, 35, 62, 85]. For highly
doped devices direct tunneling is possible which is known as field emissions. Field
emissions becomes important for devices operating at very low temperatures [44] and
is also the basis for creating Ohmic contacts which have a linear current-voltage
response [44]. A band diagram depicting thermionic emissions, thermionic field
emissions, and field emissions is shown in Fig. 2.1.2.
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 20
Figure 2.1.2. Current transport mechanisms in metal-semiconductor junctions.
In the ideal case, the Schottky barrier height is the difference in the metal work
function and the electron affinity of the semiconductor [44]. In practice, the Schottky
barrier height is dependent on the fabrication process and can be modified by Fermi
level pinning, image force lowering, etc. [44]. However, experimental results for metal-
GaN junctions have shown that the barrier height is dependent on the metal work
function [86] and thus a metal with a large work function should be chosen to form a
Schottky contact. The electron affinity of GaN is 4.1 eV [86] and the work functions of
commonly used metal contacts to GaN are 5.65 eV for Pt, 5.1 eV for Au, 5.15 eV for
Ni, and 4.33 eV for Ti [87]. Rectifying behavior has been observed for Pt, Au, and Ni
contacts to GaN with reported barrier heights of 1.1 eV for Pt, 0.91 – 1.15 eV for Au,
and 0.66 – 0.99 eV for Ni [86]. Slightly rectifying behavior has been observed for Ti
contacts to GaN with a barrier height of 0.6 eV. However, Ti has been shown to alloy
with GaN forming TiN giving rise to nitrogen vacancy related donor states resulting in
the formation of a tunneling junction [88]. Thus, Pt, Ni, or Au should be used as a
Schottky contact to GaN rather than Ti.
For temperature tolerant photodetectors, the effect of temperature on the
Schottky barrier height (and thus the electrode material) needs careful consideration.
According to theory, the Schottky barrier height should decrease at higher temperatures
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 21
due to the higher kinetic energy of charge carriers at higher temperatures [89]. However,
experimental results have found the Schottky barrier height to increase with increasing
temperature [90, 91, 92]. This discrepancy has been attributed to the Schottky barrier
height inhomogeneity in which the Schottky barrier height is theorized to be inconsistent
across the metal-semiconductor junction [90, 91, 92]. At low temperatures, the current
will flow through the lower barrier regions and as temperature increases the charge
carriers will be able to overcome the higher barrier regions. The reported thermal limit
for Pt/GaN, Au/GaN, and Ni/GaN contacts are 400°C, 575°C and 600°C, respectively
[86, 93]. Thus, temperature tolerant photodetectors with low leakage current can be
formed using Pt, Au, or Ni electrodes on GaN.
There are several types of semiconductor photodetectors as shown in Fig. 1.4.2.
The specific type of semiconductor photodetector studied here is the metal-
semiconductor-metal (MSM) photodetector. MSM photodetectors operate by using
interdigitated Schottky electrodes (or comb fingers) on a semiconductor material [44,
84, 85]. A band diagram depicting the operation of an MSM photodetector is shown in
Fig. 2.1.3. Due to the metal-semiconductor contact, the dominate dark current transport
mechanism for MSM photodetectors is thermionic, thermionic field, and/or field
emissions. For traditional MSM photodetectors utilizing opaque metal electrodes,
photons can only be absorbed by the semiconductor region between the electrodes. Thus
the electrode width should be minimized in order to maximize the photodetector
efficiency. Another approach to maximize photodetector efficiency is use a semi-
transparent or transparent electrode which allows photon absorption underneath the
electrode. GaN-based MSM photodetectors with transparent graphene electrodes and
semitransparent Ni/Au electrodes are the subject of study in this thesis. The spectral
transmission of these two materials is presented in Chapter 3 and a comparison of the
response of GaN-based photodetectors using these two electrodes is detailed in Chapter
4. The remainder of Chapter 2 is dedicated to developing a semi-analytical model for
the temperature-dependent response of GaN-based MSM photodetectors based on
thermionic, thermionic field, and field emissions theories.
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 22
Figure 2.1.3. MSM photodetector band diagram.
2.2 Semi-Analytical Model
2.2.1 Dark Current Model
To simulate the temperature-dependent dark current-voltage response of GaN-
based MSM photodetectors, the current transport mechanisms of thermionic emissions,
thermionic field emissions, and field emissions will be considered. When a voltage is
applied to an MSM photodetector, one of the electrodes will be reverse biased and the
other forward biased. For this discussion, electrode 1 is reverse biased and electrode 2
is forward biased. Therefore, the total dark current for an MSM photodetector is the
electron current from electrode 1 added to the hole current from electrode 2. The three
bias voltage operational regions for MSM photodetectors are: (1) voltages less than
reach-through (voltage at which the sum of the depletion widths equals the electrode
spacing), (2) voltages greater than reach-through but less than flat band (voltage at
which the electric field at the forward biased electrode is zero), and (3) voltages greater
than flat band but less than breakdown. The reach-through voltage can be express as
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 23
𝑉𝑅𝑇 ≅𝑞𝑁𝑉𝐷𝐿2
2𝜀𝑠− √4𝑉𝐷
2 (2.1)
where q is elementary charge (1.602 × 10-19 C), N, is the net dopant concentration, VD
is the built-in potential, L is the electrode spacing, and εs is permittivity of the
semiconductor (relative permittivity multiplied by 8.854 × 10-12 F/m). It is easy to see
that with high net dopant concentration, the reach-through voltage will be large. For the
GaN-based MSM photodetectors fabricated (see Chapter 3) and characterized (see
Chapters 4 and 5) in this work, VRT > 1000 V, therefore only operational region (1),
voltages less than reach-through, will be considered.
For thermionic emission theory, there are three main assumptions: (1) the barrier
height is much larger than kT/q, where k is Boltzmann’s constant (1.381 × 10-23 J/K)
and T is absolute temperature, (2) the device is in thermal equilibrium, and (3) two
current fluxes can be superimposed (one from metal to semiconductor and the other
from semiconductor to metal) [44]. Additionally, a symmetric MSM photodetector will
be assumed, thus the current-voltage response will be symmetric and only the response
to positive bias voltages will be simulated. The detailed analytical equations for the
thermionic, thermionic field, and field emissions models implemented here are listed in
Appendix A. The total dark current can be expressed as the sum of the thermionic (ITE),
thermionic field (ITFE), and field emissions (IFE) currents, that is
𝐼𝑑𝑎𝑟𝑘 = 𝐼𝑇𝐸 + 𝐼𝑇𝐹𝐸 + 𝐼𝐹𝐸 . (2.2)
The four main temperature-dependent parameters incorporated in this
simulation are bandgap, intrinsic carrier concentration, mobility, and barrier height. The
temperature dependence of GaN’s bandgap (EG) is shown in Fig. 2.2.1 and can be
expressed as [94]
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 24
𝐸𝐺 = 𝐸𝐺(𝑇 = 0 𝐾) − 7.7 × 10−4 × 𝑇2
𝑇 + 600 . (2.3)
Figure 2.2.1. GaN’s bandgap as a function of temperature.
The temperature-dependent intrinsic carrier concentration (ni) is shown in Fig. 2.2.2 and
can be expressed as [94]
𝑛𝑖 = (𝑁𝑐𝑁𝑣)1/2 exp (−
𝐸𝐺
2𝑘𝑇) (2.4)
where Nc is the effective density of states in the conduction band, and Nv is the effective
density of states in the valence band. For wurtzite GaN, Nc and Nv are approximately
𝑁𝑐 ≅ (4.3 × 1014) 𝑇3/2
𝑁𝑣 ≅ (8.9 × 1015) 𝑇3/2.
(2.5)
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 25
Figure 2.2.2. Intrinsic carrier concentration of GaN as a function of temperature.
A semi-empirical analytical approximation for the carrier mobility of wurtzite GaN
considering temperature and concentration dependencies is shown in Fig. 2.2.3 and
given by [95]
𝜇𝑖(𝑁, 𝑇) = 𝜇𝑚𝑎𝑥,𝑖(𝑇0)𝐵𝑖(𝑁) (
𝑇𝑇0
)𝛽𝑖
1 + 𝐵𝑖(𝑁) (𝑇𝑇0
)𝛼𝑖+𝛽𝑖
(2.6)
where
𝐵𝑖(𝑁) =
[ 𝜇𝑚𝑖𝑛,𝑖 + 𝜇𝑚𝑎𝑥,𝑖 (
𝑁𝑔,𝑖
𝑁 )𝛾𝑖
𝜇𝑚𝑎𝑥,𝑖 − 𝜇𝑚𝑖𝑛,𝑖
]
||
𝑇=𝑇0
(2.7)
and the parameters μmax,i, μmin,i, Ng,i, γi, αi, and βi, as determined from a best fit to
experimental results, are listed in Table 2.2.1. For a metal electrode on n-type wurtzite
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 26
GaN, the temperature-dependent electron barrier height (φn) is estimated using a simple
Schottky model [96, 97, 89] as
𝜑𝑛 = 𝜑𝑒𝑙𝑒𝑐𝑡𝑟𝑜𝑑𝑒(300 𝐾) + 𝜁𝑒𝑙𝑒𝑐𝑡𝑟𝑜𝑑𝑒(𝑇 − 300) − 𝜒𝐺𝑎𝑁 (2.8)
where φelectrode is the work function of the electrode, ζelectrode is the electrode temperature
coefficient, and χGaN is the electron affinity of GaN. Note that the temperature
dependence of the electron affinity is assumed negligible compared to the variation of
the electrode work function. The hole barrier height can then be calculated as the
difference between the semiconductor bandgap and the electron barrier height [44], that
is
𝜑𝑝 = 𝐸𝐺 − 𝜑𝑛. (2.9)
Using a Ni electrode as an example where φNi = 5.15 eV and ζelectrode = 5.6 ×10-4 eV/K
[89], the electron and hole barrier heights as functions of temperature are shown in Fig.
2.2.4. The GaN material properties used in MSM photodetector modeling are listed in
Table 2.2.2.
Figure 2.2.3. GaN (a) electron and (b) hole mobility as functions of temperature.
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 27
Table 2.2.1. Parameter values for carrier mobility model of wurtzite GaN [94].
Type of Carrier μmax,i μmin,i Ng,i γi αi βi
Electrons 1000 55 2 × 1017 1.0 2.0 0.7
Holes 170 3 3 × 1017 2.0 5.0 ---
Figure 2.2.4. Ni/GaN (a) electron and (b) hole barrier height as functions of temperature.
Table 2.2.2. Material properties of GaN used in MSM photodetector modeling.
Property Variable Value Unit Reference
Bandgap (at 0 K) 𝐸𝐺(𝑇 = 0 𝐾) 3.47 eV [94]
Permittivity of Semiconductor 𝜀𝑠 10.4 F/m [44]
Electron Affinity 𝜒𝐺𝑎𝑁 4.1 eV [86]
Effective Mass of Electrons 𝑚𝑛 0.27 kg [44]
Effective Mass of Holes 𝑚𝑝 0.8 kg [44]
Net Dopant Concentration 𝑁 5 × 1017 cm-3 ---
Lattice Constant (at 300 K) a 0.3189 Nm [44]
2.2.2 Dark Current Results
The semi-analytical model results for the dark current-voltage response over
temperatures ranging from -100°C to 200°C are shown in Fig. 2.2.5a for thermionic
emissions, Fig. 2.2.5b for thermionic field emissions, and Fig. 2.2.5c for field emissions.
To easily visualize the temperature dependence of the current-voltage response, the
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 28
figures on the right side in Fig. 2.2.5 employ a logarithmic current axis. The current-
voltage responses from the three current transport models are plotted together in Fig.
2.2.6. From Fig. 2.2.5 and Fig. 2.2.6, it can be seen that field emissions is the dominate
form of current transport for temperatures below -100°C while thermionic field
emissions governs the response from -100°C up to 23°C. From 23°C to 200°C, the
current-voltage response is a combination of thermionic emissions and thermionic field
emissions. Lastly, thermionic emission is the dominate form of current transport for
temperatures higher than 200°C. The total dark current-voltage response (see Eq. 2.2)
of the modeled GaN MSM photodetector is shown in Fig. 2.2.7. From the above results
it can be concluded that thermionic field emissions is a good first approximation for the
dark current transport process for GaN MSM photodetectors operating at temperatures
between -100°C to 200°C. A similar result showing that thermionic field emissions was
the dominate current transport mechanism in InGaN Schottky diodes for temperatures
up to 250°C was found by Sang et al. [62]. A qualitative comparison of the semi-
analytical model results with experimental results is presented in Chapter 4.
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 29
Figure 2.2.5. Semi-analytical model of GaN MSM photodetector dark current-voltage
from -100°C to 200°C for (a) thermionic emission, (b) thermionic field emission, and (c) field
emission. Note that the data shown in the left side and right side of (a), (b), and (c) is the
same. However, the figures on the right employ a logarithmic current axis for clarity.
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 30
Figure 2.2.6. Semi-analytical model of GaN MSM photodetector dark current-voltage
from -100°C to 200°C.
Figure 2.2.7. Semi-analytical model of GaN MSM photodetector total dark current vs.
voltage for temperatures from -100°C to 200°C with (a) linear current axis and (b) logarithmic
current axis.
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 31
2.2.3 Photocurrent Model
The photo-generated current for GaN-based photodetectors has been shown to
be proportional to the square root of the applied optical power [61]. Thus, a simple
empirical model for the photo-generated current of GaN-based photodetectors can be
expressed as [61]
𝐼𝑜𝑝𝑡 = (𝛼𝑉 + 𝛽)√𝑃𝑜𝑝𝑡 (2.10)
where α and β are constants found by fitting experimental data, V is the applied bias
voltage, and Popt is the applied optical power. The total measured photocurrent is the
sum of the dark current and photo-generated current, that is
𝐼𝑝ℎ𝑜𝑡𝑜 = 𝐼𝑑𝑎𝑟𝑘 + 𝐼𝑜𝑝𝑡. (2.11)
2.2.4 Photocurrent Results
The simulated dark and 365 nm illuminated current-voltage response for a GaN
MSM photodetector operating at temperatures between -100°C to 200°C is shown in
Fig. 2.2.8 where α was taken as 0.079 and β was taken as 0.0 as suggested by Zhang et
al [61]. Since the photo-generated current model used in this simulation (Eq. 2.10) is
independent of temperature, it is not surprising that the total measured photocurrent is
of the same order of magnitude at each temperature. To easily visualize the effect of
temperature on photodetector performance, PDCR (see Eq. 1.1) and responsivity (see
Eq. 1.2) were computed as functions of temperature as shown in Fig. 2.2.9. In Fig.
2.2.9a, PDCR decreases with increasing temperature indicating thermal generation of
charge carriers (leakage current) dominates the photo-generation at elevated
temperatures which is a commonly reported result in literature for wide bandgap
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 32
photodetectors [62, 98, 42, 43]. The constant responsivity vs. temperature shown in Fig.
2.2.9b is due to the fact that photo-generated current model implemented does not have
a temperature-dependent component. Experimental results reported in literature for the
temperature-dependent responsivity of GaN-based photodetectors varies widely
showing responsivity to be approximately constant from 23°C to 200°C [98], constant
from 23°C to 150°C and then increasing [62], as well as increasing from 70°C to 130°C
then constant until 225°C and finally decreasing [55]. The inconsistencies for reported
responsivity vs. temperature can be attributed to traps at the metal-semiconductor
interface [62] as well as defects in the GaN [55] trapping photo-generated carriers at
lower temperatures and subsequently releasing the trapped carriers at higher
temperatures.
Figure 2.2.8. GaN MSM photodetector dark and 365 nm illuminated current-voltage
response for temperatures from -100°C to 200°C with (a) linear current axis and (b)
logarithmic current axis.
The PDCR and responsivity presented in Fig. 2.2.9 were taken at 200 mV bias.
To see the effect of bias voltage on PDCR and responsivity, both figures of merit were
plotted as functions of voltage at room temperature as shown in Fig. 2.2.10. From Fig.
2.2.10a, it can be seen that there is an optimal bias voltage to obtain the largest PDCR
whereas Fig. 2.2.10b shows responsivity continually increases with increasing voltage.
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 33
The responsivity increases linearly with voltage due to the linear relationship between
photo-generated current and voltage in the implemented model (Eq. 2.10). Physically,
the photo-generated carriers are collected more efficiently when operating under a
higher electric field resulting in enhanced responsivity at higher applied bias voltage. A
linear relationship between responsivity and voltage is consistent with experimental
results reported in the literature [61, 99, 100, 101, 102]. At higher bias voltages,
saturated carrier mobility and the sweep-out effect (all photo-generated carriers are
collected before they can recombine) may cause the responsivity-voltage relationship to
plateau [50, 101]. The linear relationship between photo-generated current and voltage
coupled with the non-linear relationship between dark current and voltage results in the
PDCR peaking at a relatively low voltage. Reported experimental results for PDCR as
a function of bias voltage are consistent with the analytical model results presented here
[103, 104, 62]. Lastly, the simulation results presented in Fig. 2.2.8, Fig. 2.2.9, and Fig.
2.2.10 are compared to experimental results for GaN-based MSM UV photodetectors in
Chapter 4.
Figure 2.2.9. GaN MSM photodetector (a) PDCR and (b) responsivity as functions of
temperature at 200 mV bias.
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 34
Figure 2.2.10. GaN MSM photodetector (a) PDCR and (b) responsivity as functions of
voltage at 23°C.
2.3 Conclusions
A semi-analytical model was developed to simulate the temperature-dependent
response of gallium nitride metal-semiconductor-metal ultraviolet photodetectors. The
current transport mechanisms of thermionic emissions (charge carriers have sufficient
energy to overcome the Schottky barrier height), thermionic field emissions (charge
carriers that do not have enough energy to overcome the Schottky barrier height but see
a thinner barrier and are able to tunnel through it), and field emissions (charge carriers
are able to directly tunnel though the Schottky barrier) were considered. For GaN MSM
photodetectors operating at temperatures between -100°C and 200°C, thermionic field
emissions was found to be the dominate form of current transport. The dark current was
seen to increase with increasing temperature due to increased thermal generation of
charge carriers. The effect of temperature on the photo-generated current was analyzed
in terms of the key device parameters of PDCR and responsivity. With increasing
temperature PDCR was found to decrease due to the thermal generation of charge
carriers dominating the photo-generation at elevated temperatures while responsivity
CHAPTER 2. GaN MSM PHOTODETECTOR OPERATION AND MODELING 35
remained constant. The analytical model results presented here are fairly consistent with
experimental results for GaN MSM photodetectors to be discussed in detail in
Chapter 4.
36
Chapter 3
Fabrication and Spectral Characterization
3.1 Graphene-Enhanced GaN Photodetector Fabrication
3.1.1 Microfabrication Process
A cross-sectional view of the photodetector microfabrication process is shown
in Fig. 3.1.1. The GaN-on-sapphire substrate stack, from top to bottom, was composed
of 5 μm doped n-type GaN (< 5 Ω-cm resistivity), 200 nm AlN, and 500 μm sapphire
(Fig. 3.1.1a). The fabrication process began with a mesa etch of the GaN film via
reactive ion etching to isolate each device from their neighbors and mitigate
photodetector cross-talk (i.e., interaction between photodetectors) (Fig. 3.1.1b). Next,
50 nm of SiO2 was deposited via atomic layer deposition (Fig. 3.1.1c). The oxide layer
was then selectively etched to create the active area for each photodetector and allow
the interdigitated electrodes to be in direct contact with the GaN (Fig. 3.1.1d). Half of
the microfabricated photodetectors had semi-transparent Ni/Au (3 nm / 10 nm)
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 37
electrodes formed using a standard metal lift-off procedure and the other half had
graphene electrodes formed using a poly(methyl-methacrylate) (PMMA)-mediated
transfer method (Fig. 3.1.1e). Lastly, 20 nm Ti / 200 nm Au films were deposited for
electrical contacts and interconnect (Fig. 3.1.1f). The photodetector active area is
250 μm × 250 μm with an electrode length of 240 μm, width of 5 μm, and spacing of
5 μm. Standard 1:1 contact photolithography processing was used for each step. A top-
view optical image of a typical microfabricated GaN-on-sapphire MSM photodetector
with semitransparent Ni/Au electrodes is shown in Fig. 3.1.2.
Figure 3.1.1. GaN-on-sapphire MSM photodetector microfabrication process: (a) GaN-
on-sapphire chip, (b) mesa etch for device isolation, (c) deposited SiO2 passivation layer, (d)
passivation etch, (e) semitransparent Ni/Au or graphene electrodes, and (f) deposited Ti/Au
electrical contacts and interconnect. Reprinted from [105], with the permission of AIP
Publishing.
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 38
Figure 3.1.2. Top-view optical image of a GaN MSM photodetector with semitransparent
Ni/Au electrodes. Reprinted from [105], with the permission of AIP Publishing.
3.1.2 Graphene Transfer Process
There are several methods for fabricating graphene including mechanical
cleavage, epitaxial growth, reduction of graphene oxide, and chemical vapor deposition
(CVD) [80, 106]. Mechanical cleavage methods such as the Scotch tape method provide
high quality and easily transferable graphene sheets, but are limited to micrometer sizes
[80, 107]. Graphene grown using epitaxial growth methods such as thermal
decomposition of SiC, where SiC is annealed at high temperature causing the silicon
atoms to evaporate leaving behind carbon atoms at the surface, cannot be transferred to
arbitrary substrates limiting its usefulness [80, 106]. Graphene formed via reduction of
graphene oxide contains a substantial amount of residual functionalities and thus the
resulting graphene has poor electrical properties [80]. The CVD method of graphene
growth involves decomposing gaseous carbon sources on transition metals [80, 106].
High-quality and large-area graphene sheets have been demonstrated using the CVD
growth and subsequent transfer method [80, 108, 109, 110, 111, 112, 81].
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 39
Figure 3.1.3. Graphene transfer process: (a) graphene grown on copper foil via chemical
vapor deposition, (b) spin on PMMA, (c) oxygen plasma etch to remove backside graphene,
(d) etch copper foil, (e) rinse in deionized water, (f) transfer graphene/PMMA stack to
substrate, and (g) remove PMMA.
The graphene transfer process used here began with graphene grown on copper
via CVD (Fig. 3.1.3a). Note that the CVD process results in graphene growth on both
sides of the copper. To enable transfer of the graphene, a layer of PMMA was spun onto
one side of the graphene on copper sheet (Fig. 3.1.3b). Next, the
PMMA/graphene/copper stack is flipped upside down and an oxygen plasma etch was
used to remove the backside graphene (Fig. 3.1.3c). The PMMA/graphene/copper stack
is flipped over again and floated in a bath of ferric chloride (FeCl3) to remove the copper
(Fig. 3.1.3d). Once the copper is etched away the PMMA/graphene stack is transferred
to a water bath to rinse off the FeCl3 (Fig. 3.1.3e). Lastly, the PMMA/graphene stack is
transferred to the GaN-on-sapphire substrate (Fig. 3.1.3f) and the PMMA is removed
using solvents (Fig. 3.1.3g). Standard photolithography and an oxygen plasma etch was
used to form the interdigitated electrodes. The detailed photodetector fabrication run
sheet including graphene transfer process is listed in Appendix D. A top-view optical
image of a typical graphene/GaN MSM photodetector is shown in Fig. 3.1.4a where the
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 40
2D graphene electrodes are not visible. The graphene electrodes can be seen in a
scanning electron microscope (SEM) image as shown in Fig. 3.1.4b.
Figure 3.1.4. (a) Top-view optical image and (b) top-view SEM of graphene-enhanced
GaN MSM photodetector.
3.2 Optical Properties and Spectral Response
3.2.1 Optical Properties of GaN Substrate
The optical properties of the GaN-on-sapphire substrate were measured using a
UV-Vis-NIR spectrophotometer (Cary 6000i, Agilent Technologies) and are shown in
Fig. 3.2.1. The absorbance and reflectance spectra in Fig. 3.2.1a show the intrinsic
visible blindness of GaN with a sharp cutoff wavelength near 370 nm. Additionally,
Fabry-Pérot oscillations (constructive interference of incident light with reflected light
in multilayer materials) in the visible light region confirm the good uniformity and
interface quality of the GaN-on-sapphire substrate [99, 113, 93]. Fig. 3.2.1b shows the
transmission spectrum and (αhν)2 versus incident photon energy (hν) where α is the
absorption coefficient, h is Plank’s constant, and ν is the photon frequency. By
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 41
extrapolating the linear region of the (αhν)2 curve (red dotted line) [99, 42, 114], the
bandgap of the GaN substrate is seen to be approximately 3.34 eV which, as expected,
is comparable to the wide bandgap of wurtzite crystallized GaN (~3.4 eV) [115, 116].
Figure 3.2.1. Optical properties of the GaN-on-sapphire substrate: (a) absorbance and
reflectance spectra showing a cutoff wavelength around 370 nm and (b) transmission
spectrum and (αhν)2 versus hν with extrapolation of the linear region (red dotted line)
showing a bandgap of 3.34 eV. Reprinted from [105], with the permission of AIP Publishing.
3.2.2 Optical Properties of Graphene and Semitransparent Ni/Au
A comparison of the transmittance spectra (measured using an Agilent Cary
6000i UV-Vis-NIR spectrophotometer) of semitransparent Ni/Au (3 nm / 10 nm) and
graphene on a 1 mm thick glass substrate is shown in Fig. 3.2.2a and the corrected
transmittance spectra for graphene and semitransparent Ni/Au accounting for the glass
substrate is shown in Fig. 3.2.2b. For wavelengths longer than 300 nm, the transmittance
of semitransparent Ni/Au is between 30 to 45% and only 20 to 30% for wavelengths
less than 300 nm. Since the glass substrate is strongly absorbing for wavelength less
than 300 nm, the signal is very low which is why the corrected data appears noisy at the
low wavelengths. The relatively low transmittance of the Ni/Au agrees with previously
reported values for 100 Å thick gold electrodes of 30 to 50% in the UV regime [17].
The transmittance of graphene is greater than 90% for wavelengths longer than 300 nm
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 42
and greater than 80% for wavelengths shorter than 300 nm. In particular, at 365 nm, the
transmittance of graphene is 94% compared to only 35% for semitransparent Ni/Au.
Therefore, photodetectors with graphene electrodes will allow more of the incident UV
light to be absorbed by the active GaN region as shown in Fig. 3.2.2 and thus will result
in more sensitive photodetectors.
Figure 3.2.2. (a) Transmission spectra of graphene and semitransparent Ni/Au
(3 nm / 10 nm) on 1 mm thick glass and (b) corrected transmission spectra of graphene and
semitransparent Ni/Au. Reprinted from [117], with the permission of AIP Publishing.
Figure 3.2.3. Cross-sectional schematic of (a) a graphene/GaN and (b) a
semitransparent/Ni/Au/GaN MSM photodetector. Reprinted from [117], with the permission
of AIP Publishing.
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 43
3.2.3 Spectral Response Test Set-up
To characterize the spectral response of the GaN-on-sapphire photodetectors, a
laser driven light source (EQ-1500, Energetiq) with nearly constant radiance across the
UV spectra and spectrometer (McPherson 218) with diffraction grating
(246.16 grooves/mm and 226 nm blaze) set-up was utilized as shown in Fig. 3.2.4. To
begin, the spectrometer count to power relationship was determined using a NIST-
traceable calibrated reference photodetector (918D-UV-OD3R, Newport) with a power
meter (841-P-USB, Newport). The inlet and exit slit widths of the spectrometer were
set to 2 mm and 0.75 mm, respectively, to maximize the applied optical power with a
reasonable wavelength resolution (10% of peak at ±20 nm). Next, the calibrated
photodetector was used to measure the power of the laser driven light source from
500 nm to 200 nm in increments of 10 nm. Above 500 nm, the laser driven light source
had low intensity and since wavelengths less than 200 nm (known as vacuum-
ultraviolet, VUV) are absorbed by oxygen, experimentation under vacuum is required
which was not implemented in this study. In order to reduce second order diffraction
affects, a long pass filter was used for wavelengths greater than 340 nm. Finally,
current-voltage measurements were taken for the GaN-based photodetectors under
illumination from 500 nm to 200 nm in 10 nm increments. The inset of Fig. 3.2.4 shows
the power incident on the GaN photodetectors as a function of wavelength as measured
at the imaging plane of the spectrometer by the reference detector with the difference in
detector size between the reference (0.75 mm × 1.13 cm) and GaN (250 μm × 250 μm)
photodetectors accounted for. Two separate measurements were taken above and below
340 nm with the filter in place. Differences in source alignment cause the applied power
to differ between the two measurements. All electrical measurements were taken using
a semiconductor parameter analyzer (B1500A, Agilent Technologies). Drawings of
spectrometer adapter that enabled spectral response characterization of the GaN-based
photodetectors are shown in Appendix C.
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 44
Figure 3.2.4. Benchtop test set-up for spectral response characterization of GaN-based
photodetectors with the applied optical power incident on the GaN-based photodetectors
shown in the inset. Reprinted from [118], with the permission of AIP Publishing.
3.2.4 GaN Photodetector Spectral Response
The current-voltage responses for four separate semitransparent Ni/Au GaN-on-
sapphire MSM photodetectors are shown in Fig. 3.2.5 under dark (unilluminated) and
illuminated (500 nm, 365 nm, and 200 nm wavelengths) conditions. All four
photodetectors were fabricated on the same GaN-on-sapphire chip. From the figures, it
is easily seen that the photodetectors are most responsive to 365 nm illumination which
directly corresponds to the bandgap of GaN-on-sapphire substrate (3.34 eV).
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 45
Figure 3.2.5. Current-voltage response of four semitransparent Ni/Au GaN-on-sapphire
MSM photodetectors on the same chip when unilluminated (dark) and illuminated with
wavelengths from 200 nm to 500 nm.
As described in Chapter 1, photodetector performance is typically quantified
using two figures of merit, PDCR (defined in Eq. 1.1) and responsivity (defined in Eq.
1.2). Since the photo-generated current (Iphoto - Idark) for GaN-based photodetectors is
proportional to the square root of the applied optical power [61], a more objective metric
for spectral characterization would be the normalized photocurrent-to-dark current ratio
(NPDR) [119, 108] defined as
NPDR = (𝐼𝑝ℎ𝑜𝑡𝑜 − 𝐼𝑑𝑎𝑟𝑘
𝐼𝑑𝑎𝑟𝑘)
√𝑃𝑜𝑝𝑡
. (3.1)
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 46
The NPDR as a function of illumination wavelength for the four semitransparent
Ni/Au/GaN photodetectors are shown in Fig. 3.2.6. The NPDR peaks around 365 nm
and quickly drops off in the visible wavelength region. The spectral responsivity (R)
can then be defined as,
R = 𝐼𝑝ℎ𝑜𝑡𝑜 − 𝐼𝑑𝑎𝑟𝑘
√𝑃𝑜𝑝𝑡
(3.2)
where, once again the square root dependence of photo-generated current with applied
optical power is taken into account. The spectral responsivities for the four characterized
GaN-on-sapphire photodetectors are shown in Fig. 3.2.7. The applied optical power
used for these calculations was that measured using the calibrated reference
photodetector as shown in Fig. 3.2.4.
The spectral responsivity of the GaN-based photodetectors peaks at 365 nm and
exhibits a 365 nm to visible rejection of about one order of magnitude while the
responsivity at 200 nm is about one-third the responsivity at 365 nm. The reduced
responsivity at shorter illumination wavelengths is unexpected from simply analyzing
the absorbance of the GaN-on-sapphire substrate. From Fig. 3.2.1, the absorbance of the
GaN-on-sapphire substrate is approximately constant for wavelengths shorter than
365 nm. The strong peak in the spectral responsivity has been previously observed in
GaN-based photodetectors and attributed to excitonic effects where excitons (bound
electron-hole pairs) with energy slightly less than GaN’s bandgap act as traps for photo-
generated carriers [120, 55, 121, 122, 44].
These results are consistent with other DC spectral responsivity measurements
of GaN-based photodetectors reported in literature [17, 60]. The relatively low
UV/visible rejection ratio can be attributed to persistent photoconductivity effects due
to impurities, dislocations, and defects trapping photo-generated carriers for long times
after the illumination source has been removed [55, 98, 17, 60, 57]. Most often, spectral
responsivity measurements are reported using a chopper with a lock-in amplifier, in
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 47
which all phenomena occurring at rates slower than the chopper frequency are
eliminated. Using this technique results in UV/visible contrasts greater than three orders
of magnitude. However, the chopping frequency affects the absolute value of the
responsivity, its variation with incident optical power, as well as its variation with the
excitation wavelength [17, 61, 60, 57]. For implementation in any other real-world
application without a pulsed illumination source, it is most important to understand the
DC characteristics of the device to accurately quantify the amount UV radiation present.
Figure 3.2.6. NPDR as a function of illumination wavelength for four semitransparent
Ni/Au GaN-on-sapphire MSM photodetectors.
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 48
Figure 3.2.7. Spectral responsivity of four semitransparent Ni/Au GaN-on-sapphire MSM
photodetectors.
3.3 Conclusions
The absorbance and reflectance spectra of the GaN-on-sapphire substrate
demonstrated the intrinsic visible blindness of GaN with a sharp cutoff wavelength near
370 nm. Additionally, the bandgap of the GaN film was shown to be 3.34 eV
comparable to the reported value of 3.4 eV for the wide bandgap of wurtzite crystallized
GaN. GaN-on-sapphire UV photodetectors with semitransparent Ni/Au (3 nm / 10 nm)
interdigitated electrodes were fabricated and shown to be highly responsive to UV
radiation, especially at 365 nm, demonstrating an order of magnitude contrast from
CHAPTER 3. FABRICATION AND SPECTRAL CHARACTERIZATION 49
visible radiation. Semitransparent Ni/Au (3 nm / 10 nm) films demonstrated relatively
low transmittance between 30 - 40% in the UV regime while graphene was shown to
have a very high transmittance (> 80%) in the UV regime. In order to capitalize on this
high transmittance, GaN-on-sapphire UV photodetectors with graphene interdigitated
electrodes were fabricated. In the following chapters, the semitransparent Ni/Au/GaN
photodetectors and graphene/GaN photodetectors will be characterized and compared
while operating in high temperature and proton irradiation environments.
50
Chapter 4
Extreme Environment Operation
4.1 High-Temperature Exposure
4.1.1 Experimental Set-Up
The test set-up for characterizing the GaN-based photodetectors in air at elevated
temperatures using a high-temperature probe station (Signatone S-1060) is shown in
Fig. 4.1.1. Measurements began at room temperature (23°C) with dark current-voltage
measurements followed by photocurrent-voltage measurements. Testing proceeded in
50°C steps up to 200°C with dark current-voltage and photocurrent-voltage measured
at each temperature step. In addition to current-voltage measurements, transient current
measurements were taken at room temperature, 100°C, 150°C, and 200°C. Transient
current measurements were taken by measuring dark current for five minutes followed
by five minutes illuminated (365 nm) and then five minutes in the dark. The GaN-based
photodetector chip was held at each temperature for at least 10 minutes before
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 51
measurements were taken. A semiconductor parameter analyzer (B1500A, Agilent
Technologies) was used for all electrical characterization and a 6 W, 365 nm UV lamp
(UVP, EL series UV lamp) was used for UV illumination. A UV light meter was used
to measure the incident UV intensity, which for all illuminated electrical measurements
was 0.7 mW/cm2. For this intensity, the optical power applied on the photodetectors is
approximately 438 nW.
Figure 4.1.1. Experimental test set-up for characterizing the elevated temperature response
of GaN-based photodetectors using a high-temperature probe station.
4.1.2 Results
One GaN-on-sapphire MSM photodetector with graphene electrodes and two
GaN-on-sapphire MSM photodetectors with semitransparent Ni/Au electrodes were
characterized during this testing. Dark (dashed lines) and 365 nm illuminated (solid
lines) current-voltage curves from room temperature up to 200°C are shown in Fig.
4.1.2a for a representative graphene/GaN photodetector and Fig. 4.1.2b for a
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 52
representative semitransparent Ni/Au/GaN photodetector. For both types of electrodes,
dark current increases as temperature increases as expected due to increased thermal
generation of carriers [44].
Figure 4.1.2. (a) Graphene-enhanced and (b) semitransparent Ni/Au GaN-on-sapphire
MSM photodetector current-voltage curves under dark (dashed lines) and 365 nm illuminated
(solid lines) conditions from room temperature to 200°C.
To quantitatively compare the performance of the graphene/GaN and
semitransparent Ni/Au/GaN photodetectors, PDCR (as defined in Eq. 1.1) and
responsivity (as defined in Eq. 1.2) as functions of voltage for temperatures from 23°C
to 200°C were calculated as shown in Fig. 4.1.3. Both types of photodetectors exhibit a
peak PDCR (30.8 for the graphene/GaN photodetector and 5.6 for the semitransparent
Ni/Au/GaN photodetectors at room temperature) at relatively low bias voltages which
is in good agreement with the analytical model developed and discussed in Chapter 2.
The responsivity for both types of photodetectors increases with increasing voltage
which is consistent with the analytical model results from Chapter 2. However, the room
temperature responsivity of the semitransparent Ni/Au/GaN photodetectors becomes
nearly constant for voltages greater than 2 V which can be attributed to the sweep-out
phenomenon where all photo-generated carriers are collected before they can recombine
resulting in responsivity being saturated above a certain bias voltage [123, 101, 50]. As
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 53
temperature increases, the voltage at which sweep-out occurs increases which can be
attributed to the increased recombination rate of photo-excited carriers at high
temperatures [56]. Since graphene has a higher UV transmittance than semitransparent
Ni/Au (see Fig. 3.2.2), the graphene/GaN device is able to absorb more incident photons
thus sweep-out would occur at a higher bias voltage.
Figure 4.1.3. PDCR and responsivity as functions of voltage at temperatures from 23°C to
200°C for (a) and (b) a graphene/GaN photodetector and (c) and (d) a semitransparent
Ni/Au/GaN photodetector.
To easily visualize the temperature-dependent trends, PDCR and responsivity as
functions of temperature are shown in Fig. 4.1.4. At room temperature, the
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 54
graphene/GaN photodetectors exhibit PDCRs four times greater than the
semitransparent Ni/Au/GaN photodetectors. The larger PDCR for the graphene/GaN
photodetectors is due to the higher transmission of the graphene electrodes allowing the
GaN to absorb more of the incident photons as discussed in Chapter 3, as well as the
lower dark current as seen in Fig. 4.1.2. For both electrode materials, PDCR decreases
as temperature increases up to 200°C which is in good agreement with the analytical
model results described in Chapter 2. For temperatures beyond 200°C, PDCR drops to
zero indicating the sensors no longer operate as photodetectors. The semitransparent
Ni/Au/GaN photodetectors showed 32 times greater room temperature responsivity
compared to the graphene/GaN photodetectors which can be attributed to the much
higher leakage (dark) current of the semitransparent Ni/Au/GaN devices, as will be
discussed further below. For both types of photodetectors, responsivity first increases
with temperature, but then decreases as temperature is further increased, reaching zero
above 200°C. This result is similar to that seen by De Vittorio et al. with the initial
increase attributed to thermal ionization of photo-generated carriers trapped in defect
sites resulting in increased responsivity at higher temperatures. The subsequent
reduction in responsivity at even higher temperatures is due to increased lattice
scattering and a reduced bandgap [55, 65]. From Fig. 2.2.1, the bandgap of GaN at 23°C
is 3.4 eV which corresponds to an absorption edge of 365 nm. The bandgap of GaN at
200°C is only 3.3 eV resulting in a shifted absorption edge of 376 nm. Assuming the
spectral responsivity curves shown in Fig. 3.2.7 shift by 11 nm toward longer
wavelengths, the responsivity of the GaN-based photodetectors will be reduced by
approximately 20% when illuminated by 365 nm light.
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 55
Figure 4.1.4. (a) PDCR and (b) responsivity as functions of temperature from room
temperature to 200°C for graphene-enhanced (shown in the inset) and semitransparent Ni/Au
GaN-on-sapphire MSM photodetectors.
The magnitude of the leakage current is much lower for the graphene/GaN
photodetectors (nanoamps at 23°C) compared to the semitransparent Ni/Au/GaN
photodetectors (milliamps at 23°C). To understand this, the barrier heights for both
types of photodetectors were extracted from the forward current-voltage characteristics
by fitting the current-voltage curves to the thermionic field emissions model described
in Chapter 2. At room temperature, the barrier height for the graphene/GaN
photodetectors was found to be 0.80 eV while that for the semitransparent Ni/Au/GaN
photodetectors was only 0.46 eV. Thus, the lower dark current in the graphene/GaN
photodetectors compared to the semitransparent Ni/Au/GaN photodetectors is due to the
superior contact of the graphene to the GaN creating a higher Schottky barrier height.
The extracted barrier heights from room temperature up to 200°C are shown in Fig.
4.1.5. As temperature increases, the barrier height increases which has been reported in
previous experimental results and attributed to the Schottky barrier height
inhomogeneity [90, 91, 92] as discussed in Chapter 2. Reported values for the barrier
height of graphene on GaN range from 0.49 eV to 0.74 eV [73, 72] and those for Ni on
GaN are between 0.66 - 0.99 eV [86]. The relatively low barrier height for Ni/Au on
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 56
GaN reported here could be due to the Ti/Au electrical contacts alloying with the GaN
forming TiN and generating leakage pathways as discussed in Chapter 2.
Figure 4.1.5. Extracted barrier height as a function of temperature for (a) graphene/GaN
and (b) semitransparent Ni/Au/GaN photodetectors.
The room temperature quantum efficiency (defined in Eq. 1.3) for the
graphene/GaN photodetector is 1.1 × 103 % while the semitransparent Ni/Au is
1.9 × 106 %. As shown in Fig. 4.1.6, quantum efficiencies greater than 100% are often
reported for GaN-based photodetectors and attributed to a photo-gain mechanism [124,
125, 51, 52, 126, 102]. The photo-gain is caused by photo-generated holes trapped at
negatively-charged surface states lowering or bending the Schottky barrier with respect
to the dark Schottky barrier as shown in Fig. 4.1.7. This change in the Schottky barrier
results in an enhanced leakage current. Since the Schottky barrier height of the
semitransparent Ni/Au/GaN photodetector was found to be lower than the
graphene/GaN photodetector, the additional lowering of the barrier height by trapped
photo-generated holes has a greater effect on the semitransparent Ni/Au/GaN
photodetectors resulting in a much larger responsivity. As the operating temperature is
increased, thermal energy releases the trapped carriers. The lowering or bending of the
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 57
Schottky barrier is lessened thus the extra leakage current and the enhancement of the
photocurrent is reduced. At 200°C, the quantum efficiencies are 1.0 × 103 % and
1.9 × 105 % for the graphene/GaN and semitransparent Ni/Au/GaN photodetectors,
respectively. Additionally, Zhang et al. reported that the internal photo-gain is much
higher when operating in DC mode [61], as opposed to using a chopper and lock-in
amplifier in which all phenomena occurring at rates less than the chopper frequency are
eliminated. This observation further supports the notion that trapped photo-generated
carriers, which effect the device response on long time scales, are creating the observed
internal photo-gain. Trapped photo-generated carriers also give rise to persistent
photoconductivity as mentioned in Chapter 1 and discussed further below.
Figure 4.1.6 GaN-based photodetectors reported in literature with responsivities
corresponding to quantum efficiencies greater than 100%.
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 58
Figure 4.1.7 Trapped photo-generated holes causing lowering or bending of the Schottky
barrier height resulting in an internal photo-gain.
The transient current response curves at 200 mV bias for graphene/GaN and
semitransparent Ni/Au/GaN photodetectors at temperatures between 23°C and 200°C
are shown in Fig. 4.1.8a and Fig. 4.1.8b, respectively. It is readily apparent from Fig.
4.1.8 that increasing temperatures result in a lower net photo-generated current (dark
current subtracted from illuminated current) which was the same result found above by
examining the current-voltage responses. From Fig. 4.1.8, it can also be seen that
increasing temperature results in faster photocurrent decay. Table 4.1.1 lists the 90% to
10% decay times for the graphene/GaN and semitransparent Ni/Au GaN photodetectors.
At room temperature the decay time for the graphene/GaN photodetector is 5900 s and
that for the semitransparent Ni/Au/GaN photodetector is 1250 s. Long photocurrent
decay times are characteristic of GaN photodetectors and have been attributed to
impurities, dislocations, and deep-level defects trapping photo-generated carriers for
long times after the illumination source has been removed [58, 59, 55, 56, 57]. This
phenomenon has been termed persistent photoconductivity. The photocurrent decay has
been modeled as a stretched exponential of the form
𝐼(𝑡) = 𝐼0exp [− (𝑡
𝜏)𝛽
] (4.1)
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 59
where I0 is the initial photocurrent, τ is the decay time constant, and β is the decay
exponent. At room temperature, reported values for β are between 0.2 and 0.4 [57, 59].
Values for β extracted from the experimental data range from 0.16 to 0.14 for the
graphene/GaN photodetectors and from 0.26 to 0.15 for the semitransparent Ni/Au/GaN
photodetectors from room temperature to 200°C. It has been shown that by increasing
the temperature of GaN-based photodetectors, photocurrent decay times are reduced and
thus persistent photoconductivity is suppressed [56, 57, 55, 98, 127]. Thus, at elevated
temperatures, thermal energy is able to release photo-generated carriers trapped at
defects sites.
Figure 4.1.8. Transient current response of the (a) graphene-enhanced and (b)
semitransparent Ni/Au GaN-on-sapphire photodetectors from room temperature to 200°C
under a 365 nm UV intensity of 0.7 ± 0.1 mW/cm2.
Table 4.1.1. Decay times of the graphene/GaN and semitransparent Ni/Au/GaN photodetectors
from 23°C to 200°C.
Temperature (°C)
23 100 150 200
Graphene/GaN 5900 s 2900 s --- ---
Semitransparent Ni/Au/GaN 1250 s 35 s 13 s 0.2 s
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 60
The rise time (10% to 90% rise time) of GaN-based photodetectors has been
shown to be on the order of milliseconds at room temperature [35]. While it is not
possible to accurately quantify the rise time of the GaN-based photodetectors due to the
slow turn-on time of the UV lamp, it is easy to visually see from Fig. 4.1.8 that rise time
decreases as temperature increases. Since charge carriers generated far from the
electrodes may recombine before they are collected [17], the faster rise time at higher
temperatures suggests that the recapture rate of photo-excited carriers becomes large
compared to the excitation rate as temperature increases [56].
4.2 Proton Irradiation Testing
4.2.1 The Space Radiation Environment
Electronic devices operating in space experience high levels of radiation in the
form of electromagnetic waves (ultraviolet light, gamma rays, etc.) and particles
(electrons, protons, neutrons, heavy ions, etc.), as discussed in Chapter 1. These high
levels of radiation can introduce atomic defects in the material lattice which alter or
even destroy electronic performance. Thick aluminum shielding (~10 mm) is often used
to protect spacecraft instrumentation from this damaging radiation [4]. Predictions for
the integral trapped proton fluence as a function of energy and aluminum shielding
thickness for one year in low-Earth orbit (LEO) (orbit approximating the international
space station orbit, perigee: 400.2 km, apogee: 409.5 km, inclination: 51.64°) computed
using the SPace ENVironment Information System (SPENVIS), a software package that
combines several models to determine the space environment and its effect on materials,
are shown in Fig. 4.2.1a. In particular, SPENVIS predicts a 2 MeV integral trapped
proton fluence of 4.04 × 109 cm-2 for unshielded electronics which is reduced to a
fluence of 6.76 × 107 cm-2 (98.3% reduction) with 10 mm of aluminum shielding. One
of the most severe radiation environments in the solar system is encountered near Jupiter
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 61
due to its strong magnetic field [4, 6]. SPENVIS predictions for the integral trapped
proton fluence as a function of energy and aluminum shielding thickness for one year
in orbit around Jupiter (orbit approximating the science orbit of the Juno spacecraft,
perijove: 4,200 km, apojove: 7,900 km, inclination: 90°) are shown in Fig. 4.2.1b.
SPENVIS predicts a 2 MeV integral trapped proton fluence of 3.35 × 1011 cm-2 for
unshielded electronics in this Jovian orbit which is an 83 times larger fluence compared
to one year in LEO. With 10 mm aluminum shielding the 2 MeV integral trapped proton
fluence is reduced to 1.60 × 1010 cm-2. However, for both orbits, the aluminum shielding
does little to protect instrumentation against higher energy protons as can be easily seen
in Fig. 4.2.1. Additionally, the aluminum shielding adds considerable mass and thus
cost to spacecraft.
The high atomic displacement energy of GaN makes it an intrinsically radiation-
hard material. Thus, GaN is an intriguing material platform for developing radiation-
tolerant sensors that do not require protective shielding. In the next sections, the
response of GaN UV photodetectors subjected to 2 MeV proton irradiation will be
discussed.
Figure 4.2.1. One year trapped proton fluence as a function of energy and aluminum
shielding thickness for (a) low Earth orbit approximating the international space station orbit
and (b) Jupiter orbit approximating the science orbit of the Juno spacecraft; calculated using
the SPace ENVironment Information System (SPENVIS) software.
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 62
4.2.2 Test Set-Up
Proton irradiation of the GaN-based photodetectors was conducted using the
NEC Pelletron Tandem Accelerator (9SDH-2) at Ion Beam Materials Laboratory
(IBML) at Los Alamos National Laboratory. The proton beam was scattered using a
very thin gold foil (~50 nm) with the photodetectors located at a forward angle of 45°
to the beam direction as shown in Fig. 4.2.2a. The proton flux was concurrently
monitored with a silicon surface barrier detector located at a backward angle of 167°.
After exposure to irradiation, the photodetectors were removed from the proton
irradiation chamber for electrical characterization at estimated average fluences of
8.9 × 1011, 1.3 × 1013, 2.7 × 1013, and 3.8 × 1013 cm-2 on the photodetectors. Current-
voltage measurements were taken using a semiconductor parameter analyzer (Agilent
Technologies B1500A) under dark and illuminated (365 nm, 1 mW/cm2) conditions as
shown in Fig. 4.2.2b. Two graphene/GaN and three semitransparent Ni/Au/GaN
photodetectors were characterized during this testing.
Figure 4.2.2. (a) GaN-based photodetectors mounted inside the tandem ion accelerator
chamber at Los Alamos National Laboratory undergoing proton irradiation and (b) test set-
up for electrical characterization of proton irradiated photodetectors.
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 63
4.2.3 Electrical Characterization
Current-voltage response curves (dark and 365 nm illuminated) at each
irradiation fluence are shown in Fig. 4.2.3a for a representative graphene/GaN
photodetector and Fig. 4.2.3b for a semitransparent Ni/Au/GaN photodetector. From
Fig. 4.2.3, it can be visually seen that 2 MeV proton irradiation up to a fluence of
3.8 × 1013 cm-2 does not greatly affect device performance. To quantitatively show this,
PDCR and responsivity as functions of proton fluence for both graphene/GaN and
semitransparent Ni/Au/GaN photodetectors are shown in Fig. 4.2.4. For both electrode
materials, the results in Fig. 4.2.4 show that proton irradiation up to a fluence of
3.8 × 1013 cm-2 has very little effect on the devices response to UV light. A similar result
was also shown by Khanna et al. using photoluminescence spectroscopy to study defect
formation in bare GaN films due to proton irradiation [128]. Additionally, the
graphene/GaN devices exhibit much larger PDCR (three times larger over all irradiation
fluences tested) and responsivity (five times larger over all irradiation fluences tested)
than the semitransparent Ni/Au/GaN devices which can be attributed to the higher
transmittance of the graphene electrodes as discussed in Chapter 3. The very high
responsivities exhibited by both devices has been previously reported in literature and
attributed to trapping of photo-generated holes by surface states at the metal-
semiconductor interface resulting in a lowering of the Schottky barrier height and
increased leakage current [53, 54], as discussed in the high-temperature test results and
shown in Fig. 4.1.7. For graphene/GaN photodetectors, this photoconductive gain is
further enhanced by the very short space distance at the graphene/GaN interface [72].
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 64
Figure 4.2.3. Dark and illuminated (365 nm) scaled current-voltage response as a function
of proton irradiation fluence for GaN-on-sapphire photodetectors with (a) graphene
electrodes and (b) semitransparent Ni/Au electrodes. Reprinted from [117], with the
permission of AIP Publishing.
Since pre- and 30 days post-irradiation current-voltage measurements were
taken using the high-temperature probe station (see Fig. 4.1.1) rather than the set-up
shown in Fig. 4.2.2b, a direct comparison of the pre- and post-irradiation data to the
measurements made during irradiation is not possible. However, an overall comparison
is still informative. The average pre- and 30 days post-irradiation PDCR and
responsivity for the graphene/GaN and semitransparent Ni/Au/GaN photodetectors are
listed in Table 4.2.1. The graphene electrodes were damaged post-irradiation during
transport, therefore no measurements of those devices could be made. The pre- and post-
irradiation PDCR and responsivity of the semitransparent Ni/Au/GaN photodetectors
are consistent with the measurements made during irradiation showing a slight increase
in PDCR and a slight decrease in responsivity post-irradiation.
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 65
Figure 4.2.4. (a) PDCR and (b) responsivity as a function of proton fluence for graphene
and semitransparent Ni/Au GaN-on-sapphire photodetectors. Reprinted from [117], with the
permission of AIP Publishing.
Table 4.2.1. Average pre- and 30 days post-irradiation PDCR and responsivity for graphene
and semi-transparent Ni/Au GaN-on-sapphire photodetectors. The graphene electrodes were
damaged post-irradiation during transport.
Electrode Material PDCR Responsivity (A/W)
Pre-Irradiation Graphene 0.69 3388
Semitransparent Ni/Au 0.06 351
Post-Irradiation Graphene --- ---
Semitransparent Ni/Au 0.09 310
4.2.4 Raman Spectroscopy
Predictions for the proton implantation depth as a function of energy computed
using the Stopping and Range of Ions in Matter (SRIM) software, a Monte Carlo
simulation which estimates the transport of ions in matter, are listed in Table 4.2.2. For
2 MeV protons impingent on graphene on GaN-on-sapphire, SRIM predicts a stopping
depth of 52 μm which is within the sapphire substrate, thus the majority of the
displacement damage occurs in the sapphire substrate, well below the active region of
the photodetector. To intentionally increase displacement damage within the active GaN
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 66
layer, 200 keV protons were used to irradiate graphene specimens based on SRIM
calculations which estimate that 200 keV protons stop at approximately 2 μm depth,
which is within the GaN. This irradiation may have a greater effect on device
performance since defects in the active semiconductor have been shown to influence the
graphene/semiconductor interface [77]. Additionally, previous reports have indicated
that defect formation in graphene increases as energy decreases [79]. SRIM ion
distributions for 200 keV and 2 MeV protons impingent on graphene on GaN-on-
sapphire are shown in Fig. 4.2.5.
Table 4.2.2. Proton implantation depth and material as a function of energy for the GaN-on-
sapphire substrate as calculated using the Stopping and Range of Ions in Matter (SRIM)
software.
Energy (MeV) Implantation Depth (μm) Implantation Material
0.1 1.0 GaN
0.2 2.0 GaN
0.5 6.2 Sapphire
1 17.5 Sapphire
2 52.4 Sapphire
5 242 Sapphire
7 432 Sapphire
Raman spectroscopy (HORIBA Scientific LabRAM HR Evolution
spectrometer, laser wavelength of 532 nm) was used to investigate the damage done to
graphene by 200 keV protons at a fluence of 3.8 × 1013 cm-2. Figure 4.2.6 shows the
Raman spectra of a graphene on a SiO2/Si sample before irradiation and 30 days after
irradiation (the spectra are offset for clarity). Before irradiation, the intensity of the 2D-
peak (2686 cm-1) compared to the G-peak (1596 cm-1) (I2D/IG) was 2.4 and the intensity
of the D-peak (1347 cm-1) compared to the G-peak (ID/IG) was 0.53. The presence of
two main peaks (G-peak and 2D-peak) and a negligible D-peak, which is the most
common method to detect defect formation in graphene, indicate a pristine graphene
sample [79, 129, 130]. Additionally, the full-width-at-half-maximum (FWHM) of the
2D-peak was 35.3 cm-1 which indicates monolayer graphene [131]. After irradiation,
I2D/IG decreased to 1.8 and ID/IG increased slightly to 0.56. The increased ID/IG ratio as
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 67
well as the appearance of the D'-peak (1621 cm-1) brought about by proton irradiation
indicate increased disorder in the graphene corresponding to vacancy related defects
[79, 129, 130]. Another indication of the damaging effect of proton irradiation is the
decrease in the 2D peak intensity. These changes in the Raman spectrum are quite small
considering the low energy (200 keV) and high fluence (3.8 × 1013 cm-2) proton
irradiation the graphene was exposed to. These results further confirm the radiation-
hardness of graphene and its potential to enhance other material platforms.
Figure 4.2.5. SRIM ion distribution results for (a) 200 keV and (b) 2 MeV protons
impingent on graphene on a 5 μm GaN, 200 nm AlN and 500 μm sapphire substrate. For
200 keV protons the stopping depth and thus majority of lattice displacement damage occurs
at approximately 2 μm which is within the GaN. Whereas for 2 MeV protons, the stopping
depth is in the sapphire substrate at 52.3 μm.
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 68
Figure 4.2.6. Raman spectroscopy comparison of graphene samples before and after
200 keV proton irradiation up to a fluence of 3.8 × 1013 cm-2. Reprinted from [117], with the
permission of AIP Publishing.
4.3 Conclusions
GaN UV photodetectors with graphene and semitransparent Ni/Au electrodes
were electrically characterized and compared when subjected to temperatures ranging
from 23°C to 200°C. The graphene/GaN photodetectors exhibited a higher PDCR
compared to the semitransparent Ni/Au/GaN photodetectors over all temperatures tested
which can be attributed to graphene’s very high transmittance (> 80%) in the UV regime
compared to semitransparent Ni/Au (> 30%). For both types of photodetectors, PDCR
was seen to decrease with increasing temperature eventually reaching zero above
200°C. The decreasing PDCR with increasing temperatures is due to the thermal
generation of charge carriers overwhelming the photo-generation. The responsivity of
both types of photodetectors was seen to decrease at high temperatures due to increased
lattice scattering as well as a reduced bandgap. From transient current response data, it
was shown that both graphene/GaN and semitransparent Ni/Au/GaN photodetectors
CHAPTER 4. EXTREME ENVIRONMENT OPERATION 69
demonstrated faster rise and decay times at elevated temperatures. The faster rise times
at high temperatures were attributed to a higher recapture rate of photo-excited carriers
while the faster decay times were attributed to thermal energy releasing trapped photo-
generated carriers. Additionally, these temperature-dependent trends seen in the
experimental results match the predictions of the analytical model developed in Chapter
2.
The graphene/GaN and semitransparent Ni/Au/GaN photodetectors were also
characterized when subjected to 2 MeV proton irradiation up to a fluence of
3.8 × 1013 cm-2. Electrical measurements made throughout irradiation testing showed no
significant degradation in device performance in terms of PDCR and responsivity.
Additionally, Raman spectroscopy of 200 keV irradiated graphene showed only slight
increase of disorder in the form of an increased ID/IG ratio, the appearance of the D'-
peak, and a decrease in the 2D peak intensity. These results support the use of graphene
and GaN as temperature tolerant and radiation-hard material platforms for UV
photodetection with high sensitivity and responsivity.
70
Chapter 5
Implementation
5.1 Miniature Sun Sensor
5.1.1 Current Sun Sensor Technology
There are several sensor technologies used for spacecraft attitude determination
which include sun sensors, star trackers, gyroscopes, magnetometers, and the global
positioning system (GPS) [132, 133]. Sun sensors indicate the orientation of the
spacecraft relative to the sun while star trackers use multiple vector measurements to
different stars, resulting in a more accurate measurement at the cost of more mass and
higher power consumption [133]. Gyroscopes give relative angular measurements with
high accuracy and a fast response time, but must be coupled with an absolute reference
such as a star tracker. Magnetometers determine spacecraft attitude relative to the local
magnetic field, thus for Earth satellites can only be used below about 6,000 km. Lastly,
GPS can be used for attitude determination by Earth satellites in communication in GPS
CHAPTER 5. IMPLEMENTATION 71
satellites [132]. Of all of these technologies, sun sensors are the most commonly used.
Sun sensors have been used for aerospace applications since the early 1950’s and were
used before that for guidance of ground-based telescopes among other applications
[134]. Sun sensors have low mass (< 1 kg) and low power (< 0.2 W) requirements
compared to the other sensor technologies while maintaining decent accuracy (0.01°)
and a wide field of view (± 130°) [133]. The reliable operation of these sensors within
the harsh environment of space is critical for the success of space missions.
The two-axis sun sensors on-board the Voyager 1 and Voyager 2 spacecraft
(launched in 1977) used recrystallized photoresistive cadmium sulfide (CdS) detectors
for their sensing element and had a field of view of ± 20° [135, 136, 137]. These sun
sensors were relatively large and had additional packaging for radiation attenuation. Sun
sensor technology has advanced greatly since 1977; for example, the sun sensor on-
board the Curiosity rover (launched 2011) used a Si complementary metal-oxide-
semiconductor (CMOS) active pixel sensor (APS), had a field of view of ± 60°, and had
a mass of only 35 grams [138, 139, 140, 141]. Other two-axis sun sensors have been
reported including APS cameras [139, 142, 143], charge coupled device (CCD) cameras
[144], Si-based photodiodes [145, 146], and solar cells [147]. Additionally, three-axis
sun sensors utilizing a hemisphere [148] and the Doppler shift of the Sun’s photosphere
[132] have been reported.
Table 5.1.1 compares the sun sensors discussed above, from the CdS-based
detectors used on the early space missions through the Si-based APS and CCD detectors
being developed today. All of the sun sensors discussed are responsive to infrared and/or
visible light and most are based on Si technology. A GaN-based sun sensor will respond
to UV light which enhances sun sensing technology by expanding the operational
regime. Since GaN is a temperature tolerant and radiation-hard material platform, a
GaN-based sun sensor will be able to intrinsically withstand the harsh space
environment thus reducing extra protective packaging requirements. Therefore, for UV
photodetection and reliable space operation, a miniature sun sensor utilizing an array of
CHAPTER 5. IMPLEMENTATION 72
GaN-based photodetectors could address many technological challenges in one device
and will be discussed in the following sections.
Table 5.1.1. Sun sensor technology comparison.
Year Sensor
Element Field of View Accuracy Notes Reference
1971 Photoresistor ± 2.5° × ± 20° --- Cadmium sulfide
Flown on Mariner spacecraft [149]
1977 Photoresistor ± 20° ---
Cadmium sulfide
Flown on Voyager 1 and 2
[136, 137,
135]
2005 CMOS APS ~ ± 50° × ± 50° 0.1° Used on JAXA's small
satellites [150]
2009 CMOS APS 94° × 81° 1 arcmin
Uses a multi-aperture mask
to increase precision
[143]
2009 Solar Cell ---
0.1° with
12 bit A/D
converter
Low-cost coarse sun sensor
using already on-board solar
cells for Pico satellites
[147]
2011 CMOS APS ± 45° 0.01 pixels
Uses a multi-aperture mask
to increase precision
[151]
2011 CMOS APS ± 60° 1 arcmin Flown on Mars Curiosity
rover
[139, 138,
140, 141]
2012 Photodiode
Array ± 60° 20°
Uses several sensors to
extend the field of view while
maintaining high-precision
[152]
2014
TSL230R
Light-to-
Frequency
Converter
--- 5° Azimuth
1° Zenith
Uses commercial light
intensity sensors in a
hemispherical arrangement
[148]
2016 CCD ± 65° × ± 65° 0.1°
Three-axis sun sensor
Uses a V-shaped mask and
highly integrated control
circuit
[144]
2016
Magneto-
Optical
Filters and
CCD
25° with 200°/s
slew rate 0.1°
Images the Doppler shift of
the sun's photosphere to
measure the solar rotation
axis
[132]
CHAPTER 5. IMPLEMENTATION 73
5.1.2 Principal of Operation
A sun sensor composed of a focal plane, spacer, and aperture mask operates by
allowing light to shine through small openings in the opaque aperture mask. As the
sensor moves relative to the sun, the illumination pattern on the focal plane will change.
The position of the sensor relative to the sun can be determined by tracking the variation
in the illumination pattern. The principle of operation of a sun sensor is much like a sun
dial. A cross-sectional schematic of the sun sensor’s operating principle and design
parameters are shown in Fig. 5.1.1. For this design the optical focal plane consists of a
3 × 3 array of GaN-on-sapphire MSM photodetectors (as shown in Fig. 5.1.2) and the
spacer was made of glass (Pyrex, Corning Inc.). It should be noted that the upper left
photodetector in the array will be referred to as PD 1 with numbering continuing left to
right with the bottom right photodetector called PD 9. It should also be noted that due
to the scalability of microfabrication, larger array sizes could be utilized to increase field
of view and resolution. For reliable operation, several parameters had to be taken into
consideration in the design including the length (Lapp) and thickness (t) of the aperture,
length of the photodetector (LPD), spacing between the photodetectors (Lspace), and
height of the spacer (h), all labeled in Fig. 5.1.1. In brief, LPD, Lspace, and h determine
the resolution of the sun sensor (i.e. the closer together and smaller the photodetectors
are, the more precisely the incident light angle will be known), as well as the size of the
sun sensor (i.e., the taller the spacer, the farther apart the photodetectors need to be to
measure the same angle). Therefore a tall spacer with a larger array of closely spaced
photodetectors (achievable with today’s micro- and nano-fabrication techniques) would
give the maximum field of view with highest resolution.
The sun sensor was designed such that when the angle of the incident light (θsun)
is 45°, one of the edge photodetectors at the angle of θPD after refraction would be
illuminated. To accomplish this, Snell’s law of refraction was considered,
CHAPTER 5. IMPLEMENTATION 74
𝑛𝑎𝑖𝑟 sin(𝜃𝑠𝑢𝑛) = 𝑛𝑝𝑦𝑟𝑒𝑥 sin(𝜃𝑃𝐷) (5.1)
where the index of refraction of air (nair) is 1.00 and that of Pyrex (npyrex) was taken as
1.4734. The parameters of the final sun sensor are listed in Table 5.1.2. The ratio of the
wavelength of the incoming light to the size of the aperture determines how much a
wave spreads out behind an opening regardless of the shape of the opening. For 365 nm
wavelength light and the aperture size of 350 μm, this ratio calculates to be 0.001 and it
is easily seen that diffraction effects are negligible for this design. Since the spreading
due to diffraction is negligible to a first order approximation, the ray light model can be
employed and the light can be considered as traveling in straight lines [153].
Figure 5.1.1. Cross-sectional schematic of a GaN-based miniature sun sensor depicting the
principle of operation and design parameters using an aperture mask, spacer, and focal plane.
The inset shows the top view of a GaN-on-sapphire MSM photodetector. Reprinted from
[105], with the permission of AIP Publishing.
CHAPTER 5. IMPLEMENTATION 75
Figure 5.1.2. Optical image of 3 × 3 GaN-based MSM photodetector array. Reprinted
from [105], with the permission of AIP Publishing.
Table 5.1.2. Miniature sun sensor dimensions.
Parameter Designed Value
Lapp 350 μm
Lspace 300 μm
LPD 250 μm
t 110 nm
h 1.1 mm
5.1.3 Array Characterization
The electrical performance, in terms of dark and illuminated current versus
voltage, of the GaN MSM photodetectors was determined using a semiconductor
parameter analyzer (B1500A, Agilent Technologies) and a 6 W, 365 nm UV lamp
(UVP, EL series UV lamp). A UV light meter was used to measure the incident UV
intensity, which for all illuminated electrical measurements was 1.0 mW/cm2. For this
intensity, the optical power applied on the photodetector is approximately 625 nW.
The magnitude of the dark current and the response to UV illumination without
the aperture mask is about the same for each photodetector in the array with an
approximate array average PDCR (defined in Eq. 1.1) of 2.4 and responsivity (defined
in Eq. 1.2) of 3200 A/W at -1 V bias. PDCR values for GaN-based photodetectors
reported in literature range from less than one up to tens of thousands [98, 154, 155,
CHAPTER 5. IMPLEMENTATION 76
156]. The large PDCR values are obtained using high intensity UV sources (e.g., up to
1020 mW/cm2 using He-Cd laser [42]). Since the photo-generated current is highly
dependent on the incident light intensity, it is important to note that an increased PDCR
will be achieved with a higher intensity UV source. Reported responsivities range from
0.1 A/W well into the thousands [52, 51] with the large responsivity values attributed to
large internal gains. Since responsivity decreases with increasing optical power and
PDCR increases, these figure of merit are highly coupled and both need careful
consideration. From this, it can be concluded that the photodetector performance
reported here is on par with previously reported results for GaN-based photodetectors.
The PDCR and responsivity for each photodetector at -1 V bias is listed in Table
5.1.3. To show the nominal response of the devices, these values were calculated from
the raw pre-aperture mask current-voltage data. It is seen that PD 1 has a different
response than the rest of the array with a much lower PDCR and much lower
responsivity. This is mainly because PD 1 was unintentionally scratched by probe tips
during electrical measurements which gave it an altered response. However, this single
anomalous photodetector does not affect the sun sensor performance as a whole, as will
be described below.
Table 5.1.3. Photodetector PDCR and responsivity at -1 V bias before attaching the aperture
mask. PD 1 was unintentionally scratched during electrical measurements.
PDCR Responsivity [A/W]
PD 1 0.17 1785.7
PD 2 2.27 3952.3
PD 3 2.05 3409.4
PD 4 2.38 4102.9
PD 5 2.40 3647.0
PD 6 2.40 3696.9
PD 7 1.90 3600.6
PD 8 2.72 3042.4
PD 9 2.02 2182.0
The performance of each photodetector is slightly different, thus for accuracy
these slight variations in response from device to device need to be considered. To easily
CHAPTER 5. IMPLEMENTATION 77
visualize and compare the dark and illuminated current-voltage characteristics for all
nine photodetectors in the array, a scaling was performed so that the dark current versus
voltage curve is the same for each photodetector while maintaining each individual
photodetector’s PDCR. The result of this scaling for data taken before the aperture mask
was attached is shown in Fig. 5.1.3. From the figure it is seen that the nominal response
of all devices in the array is about the same except for PD 1 as discussed above.
Figure 5.1.3. Scaled current-voltage response of all nine photodetectors in the array under
dark and illuminated conditions before attaching the aperture mask. Reprinted from [105],
with the permission of AIP Publishing.
5.1.4 Miniature Sun Sensor Performance
With the photodetector array constituting the focal plane, a Pyrex spacer with
opaque gold aperture mask was attached with space-grade non-outgassing epoxy (EPO-
TEK 301-2). The aperture mask was fabricated using a standard metal lift-off procedure
with 10 nm Ti / 100 nm Au on a Pyrex wafer. As stated in Table 5.1.2, the size of the
aperture opening is 350 μm × 350 μm. For comparison purposes, two 3 × 3 arrays of
photodetectors were fabricated on a single 1 cm × 1 cm GaN-on-sapphire chip, resulting
CHAPTER 5. IMPLEMENTATION 78
in two sun sensors per chip. Optical images of the fabricated MSM photodetectors are
shown in Fig. 5.1.2 and the final miniature sun sensor is shown in Fig. 5.1.4.
To determine the “sun” (UV source) angle under simulated conditions (the UV
lamp was rotated and clamped at the specific test angle), a simple algorithm that
compares the PDCR and responsivity for each of the nine photodetectors was developed
as outlined below. In practice, a processing circuit could be used to implement this
simple algorithm.
Figure 5.1.4. Image of assembled miniature sun sensor with size comparison to a penny.
Reprinted from [105], with the permission of AIP Publishing.
1. Test all nine photodetectors in the 3 × 3 array before the aperture mask
is attached (dark and illuminated at θsun = 0°)
Calculate PDCR and responsivity for each device – call these the
“NOMINAL” values
2. Attach aperture mask
3. Test all devices in the array (dark and illuminated at θsun = 0°)
Calculate PDCR and responsivity for each device – call these the
“0° AP MASK” values
CHAPTER 5. IMPLEMENTATION 79
4. Normalize the “0° AP MASK” values by the “NOMIAL” values – call
these the “NORMALIZED ARRAYS”
5. Scale the “NORMALIZED ARRAYS” values to be between 0 and 1 –
call these the “SCALED and NORMALIZED ARRAYS”
6. Calculate the “SUN ANGLE ARRAY” by equally weighting the
PCDR and responsivity “SCALED and NORMALIZED ARRAYS”
values
7. Repeat steps 3 to 6 for θsun = 45° with normalizing the “45° AP
MASK” values by the “0° AP MASK” values
The results of this algorithm for two test cases are shown in Fig. 5.1.5 and Fig.
5.1.6. Fig. 5.1.5 shows the PDCR and responsivity “SCALED and NORMALIZED
ARRAYS” for illumination angles of θsun = 0° and 45° for a range of bias voltages.
From Fig. 5.1.5a and Fig. 5.1.5b, it is easily seen that PD 5 was the illuminated
photodetector (at θsun = 0°) and from Fig. 5.1.5c and Fig. 5.1.5d, it is PD 2 that was
illuminated (θsun = 45°). In Fig. 5.1.5c and Fig. 5.1.5d, it is seen that PD 3 has a slightly
higher scaled and normalized PDCR and responsivity than the other unilluminated
photodetectors. This anomaly was attributed to a misalignment during testing allowing
the UV light to leak around the edge of the aperture.
In Fig. 5.1.6, the “SUN ANGLE ARRAY” values at -1 V for the two test cases
shown in Fig. 5.1.5 have been overlaid on an optical image of the photodetector array
to provide visualization of the photodetector illumination pattern. Using the “SUN
ANGLE ARRAY” values calculated from the algorithm described above and shown in
Fig. 5.1.6, the minimum illuminated photodetector to dark photodetector (PDil/PDdark)
ratio is 1.6 with an average ratio of 2.4 for the 0° illumination case. For the 45°
illumination case, the minimum ratio is 1.4 with an average of 2.1. The overall sun
sensor performance is efficient and reliable considering the low 1.0 mW/cm2 intensity
(limitation of the UV lamp utilized). Since photo-generated current increases with
increasing intensity, it is obvious that the more UV light present, will lead to improved
CHAPTER 5. IMPLEMENTATION 80
performance (increased sensitivity) of the sun sensor. For comparison, the irradiance of
the sun at the earth’s mean distance from the sun outside the earth’s atmosphere is
137 mW/cm2 of which about 9% is UV [157]. Therefore, a sun sensor just outside the
earth’s atmosphere would experience a UV intensity of ~12 mW/cm2 which is 12 times
greater than that of the UV lamp used in this study. In addition to intensity, the
performance could be improved by further increasing the illuminated photodetector to
dark photodetector ratio by refining the aperture mask design to reduce the light leakage
and reflection around the edges and corners.
Figure 5.1.5. Scaled and normalized PDCR and responsivity comparisons of all nine
photodetectors when miniature sun sensor is illuminated (365 nm) at ((a) and (b)) θsun = 0°
and ((c) and (d)) θsun = 45°. Reprinted from [105], with the permission of AIP Publishing.
CHAPTER 5. IMPLEMENTATION 81
Figure 5.1.6. Miniature sun sensor operation at illumination angles (θsun) of (a) 0°
(illuminating PD 5) and (b) 45° (illuminating PD 2). The left-side images indicate the
illumination angle/orientation and the right-side images show the illuminated array based on
the normalized PDCR and responsivity at −1 V. Reprinted from [105], with the permission
of AIP Publishing.
5.2 CubeSat Integration and Ground Testing
5.2.1 QB50 Mission
The QB50 mission consists of a network of 50 satellites built by university teams
all over the world to collect in situ scientific data in Earth’s lower thermosphere. The
main payload on-board the QB50 CubeSat developed at Stanford (known as Discovery)
is an ion-neutral mass spectrometer (INMS). The INMS uses mass spectroscopy to
determine the composition of low mass ionized and neutral particles and has been
optimized for O, O2, NO, and N2, the major constituents in Earth’s lower thermosphere.
CHAPTER 5. IMPLEMENTATION 82
The GaN-based miniature sun sensor described in Chapter 5.1 was integrated on-board
CubeSat Discovery as a secondary payload. The integration and ground test results will
be described in the following sections. QB50 CubeSat Discovery is expected to be
launched off the international space station (ISS) in the near future.
5.2.2 Miniature Sun Sensor Integration
The GaN-based miniature sun sensor was epoxied and wirebonded to the
CubeSat’s sun-facing printed circuit board (PCB) near the CubeSat’s attitude
determination and control system (ADCS). In order to protect the very delicate
wirebonds from breaking, they were encapsulated in an epoxy (EPO-TEK 301-2) with
low-outgassing properties, high optical transmission (> 94% for wavelengths
> 320 nm), and a wide operating temperature range (-55°C to 200°C). The 1 cm × 1 cm
GaN-chip has two 3 × 3 arrays of photodetectors, one array of graphene photodetectors
and one array of semitransparent Ni/Au photodetectors, resulting in two miniature sun
sensors per chip. A computer-aided design (CAD) model of the CubeSat and miniature
sun sensor as well as a photo of the completed CubeSat are shown in Fig. 5.2.1.
To enable independent measurement of all 18 photodetectors, a circuit was
designed which powers only one column of photodetectors at once and measures the
voltage across one row. By switching which column is powered and which row is
measured, all 18 photodetectors can be read in quick succession and with post-
processing of the data, the angle of the incident light (i.e. the sun) can be determined.
The power supplied can be chosen ranging from 0 V to 5 V. The raw data returned by
the circuit is in bits which can be converted to volts by dividing by 4096 (the circuit
uses a 12 bit analog-to-digital converter) and multiplying by the bias voltage. This
voltage can be converted to current by dividing by the 40x amplifier gain. The full
circuit diagram for the miniature sun sensor integrated with the QB50 CubeSat is shown
in Appendix C. The CubeSat measures all 18 photodetectors of the miniature sun sensor
CHAPTER 5. IMPLEMENTATION 83
twice per orbit and stores it in a data packet that includes a timestamp, GPS location,
ADCS sun sensor data, and temperature measured by a thermocouple on the CubeSat’s
sun-facing PBC.
Figure 5.2.1. QB50 CubeSat Discovery CAD model with GaN-based miniature sun sensor
and finished CubeSat.
5.2.3 Ground Test Results
The GaN-based photodetector arrays were characterized before integrating them
with the CubeSat. The scaled current-voltage responses of the (a) graphene and (b)
semitransparent Ni/Au photodetector arrays under dark and illuminated (365 nm)
conditions before attaching the aperture mask are shown in Fig. 5.2.2. Three of the
graphene photodetectors (PD 4, PD 7, and PD 9) were not operational due to poor
CHAPTER 5. IMPLEMENTATION 84
graphene transfer. The corresponding PDCR and responsivity at 5 V bias are listed in
Table 5.2.1.
Figure 5.2.2. Dark and illuminated (365 nm) scaled current-voltage response of (a)
graphene and (b) semitransparent Ni/Au photodetector arrays before attaching the aperture
mask and before QB50 implementation. Three of the graphene photodetectors (PD 4, PD 7
and PD 9) were not operational due to poor graphene transfer.
Table 5.2.1. Photodetector PDCR and responsivity at 5 V bias before attaching the aperture
mask to the QB50 miniature sun sensor. Three of the graphene photodetectors (PD 4, PD 7
and PD 9) were not operational due to poor graphene transfer.
Graphene Semitransparent Ni/Au
PDCR Responsivity [A/W] PDCR Responsivity [A/W]
PD 1 0.15 3317.3 0.13 2302.5
PD 2 0.21 3605.3 0.15 2100.6
PD 3 0.12 2018.8 0.05 846.5
PD 4 X X 0.13 2579.2
PD 5 0.06 1099.7 0.06 911.3
PD 6 0.10 2195.2 0.09 1719.0
PD 7 X X 0.12 2352.0
PD 8 0.08 1569.5 0.09 1900.8
PD 9 X X 0.25 3465.7
The set-up for testing and debugging the miniature sun sensor circuit with the
QB50 CubeSat Discovery hardware is shown in Fig. 5.2.3. The miniature sun sensor
was tested at 5 V bias with the 365 nm UV lamp positioned at 0°. The simple algorithm
developed in Chapter 5.1 was used to determine the “SUN ANGLE ARRAY” with the
quantitative results listed in Table 5.2.2 and photodetector array visualization shown in
CHAPTER 5. IMPLEMENTATION 85
Fig. 5.2.4. The minimum PDil/PDdark ratio is 1.1 for the graphene photodetector array
and 1.2 for the semitransparent Ni/Au photodetector array. The average PDil/PDdark ratio
is 3.2 and 3.7 for the graphene and semitransparent Ni/Au photodetector arrays
respectively. Slight misalignment of the UV lamp could be contributing to the variation
of PDil/PDdark across the array. The performance of the miniature sun sensor is quite
good considering the low 1.0 mW/cm2 intensity of the UV lamp and is expected to be
much better with the higher intensity of the sun outside the Earth’s atmosphere.
Figure 5.2.3. Ground test set-up for integration testing of GaN sun sensor with QB50
Discovery.
Table 5.2.2. “SUN ANGLE ARRAY” of the miniature sun sensor integrated with the QB50
CubeSat with illumination angle of 0° and 5 V bias.
Graphene Semitransparent Ni/Au
PD 1 0.17 0.21
PD 2 0.18 0.29
PD 3 0.42 0.82
PD 4 X 0.42
PD 5 1 1
PD 6 0.88 0.49
PD 7 X 0.35
PD 8 0.87 0.40
PD 9 X 0.10
CHAPTER 5. IMPLEMENTATION 86
Figure 5.2.4. Ground test performance of miniature sun sensor integrated with QB50 at 0°
illumination angle and 5 V bias; (a) graphene photodetector array and (b) semitransparent
Ni/Au photodetector array. Three of the graphene photodetectors (PD 4, PD 7, and PD 9)
were not operational due to poor graphene transfer.
5.3 In Situ Shock-Layer Radiation Measurements
5.3.1 Current Atmospheric Entry Sensing Technology
Planetary exploration is constrained by the capabilities of atmospheric entry
technologies, including thermal protection systems (TPS). Spacecraft experience high
heat fluxes during atmospheric entry and thus require a TPS to protect the payload [4].
In-flight temperature and pressure measurements on the heatshields of the Flight
Investigation Reentry Environment II (FIRE II) vehicle, the Planetary Atmosphere
Experiments Test (PAET) vehicle, Apollo, Space Shuttle, and Mars Science Laboratory
(MSL), among other spacecraft, have enabled aerothermal model refinement and
improved TPS design [158, 159]. However, advancement of atmospheric entry sensing
technologies have been limited due to the high heat fluxes and complexity of
instrumenting the TPS. Current state-of-the-art seen on the Mars Entry Descent and
Landing Instrumentation (MEDLI) sensor suite on the MSL heatshield consists of
thermocouples embedded in a plug of TPS material (Phenolic-Impregnated Carbon
CHAPTER 5. IMPLEMENTATION 87
Ablator, PICA) and pressure transducers connected to pressure ports in the TPS [160,
161, 162].
Prediction of the entry heat load and subsequent sizing of the TPS is often
focused on convective heating. However, for large vehicles and high entry velocities,
shock-layer radiation can be a major contributor [163, 40, 164, 165]. As of now, few in-
flight shock-layer radiation measurements have been made. A timeline of spacecraft
that have or are planned to measure in-flight shock-layer radiation during atmospheric
entry is shown in Fig. 5.3.1 and listed in Table 5.3.1. To measure shock-layer radiative
heating a radiometer, consisting of a sensing element (often a thermopile) behind a
window, is embedded in the TPS. In the late 1960’s to early 1970’s, FIRE II, Apollo
4/6, and PAET, used on-board radiometers to measure radiation during Earth reentry
[166, 167, 158, 168]. Due to the window and sensor materials used by the radiometers
on-board Apollo and PAET, measurements were limited to visible to near infrared
radiation, neglecting UV radiation, which can account for half of the radiative heating
[166, 41]. Furthermore, shock-layer radiative heating in the VUV regime (< 200 nm),
which has never been measured in-flight, may contribute up to seven times the non-
VUV radiative flux on the backshell of Earth reentry vehicles [41]. More recently, the
sapphire window on the radiometer on-board the Orion Exploration Flight Test-1 (EFT-
1) vehicle enabled UV measurements down to 220 nm [41]. The first measurement of
shock-layer radiative heating during Mars entry was made by a broadband radiometer
on the heatshield of the European Space Agency’s ExoMars Schiaparelli lander with
data analysis forthcoming [169]. The next in-flight measurement of radiative heating is
slated to be NASA’s Mars Entry Descent and Landing Instrumentation II (MEDLI II)
sensor suite on the Mars 2020 heat shield [170].
Clearly there are gaps in our knowledge of shock-layer radiative heating seen in
the discrepancies between the in-flight data, ground test experiments, and simulations.
There are technological hurdles to overcome in order to advance atmospheric entry
sensing technology including developing high-temperature capable sensors, refining
CHAPTER 5. IMPLEMENTATION 88
sensor packaging for embedding in TPS so as to not degrade the integrity of the TPS,
and developing radiometers capable of UV radiation measurements.
Figure 5.3.1. Timeline of spacecraft that have or are planned to measure in-flight shock-
layer radiation during atmospheric entry.
Table 5.3.1. Spacecraft to date that have or are planned to measure in-flight shock-layer
radiation during atmospheric entry.
Vehicle Year Planet Sensing Element Window Material Location
FIRE II 1965 Earth --- Fused Quartz Fore and Aftbody
Apollo 4/6 1967/8 Earth Thermopile Quartz Fore and Aftbody
PAET 1971 Earth PIN Photodiode Fused Silica Forebody
EFT-1 2014 Earth Thermopile Sapphire Rod Forebody
ExoMars 2016 Mars --- --- Aftbody
MEDLI 2 2020 Mars Heat Flux Sensor Sapphire Aftbody
In this study, the feasibility of implementing GaN-based photodetectors to detect
shock-layer UV radiative heating through in situ measurements in a shock tube is
explored. GaN-based photodetectors offer many benefits over traditional solutions
including thermal and chemical stability enabling GaN-based devices to be placed
closer to the surface of the TPS allowing for more accurate measurements to be obtained
as well as the ability to measure the entire UV spectrum, as discussed in Chapter 1.
GaN-on-sapphire metal-semiconductor-metal (MSM) photodetectors were installed in
CHAPTER 5. IMPLEMENTATION 89
the Electric Arc Shock Tube (EAST) facility at NASA Ames Research Center and
characterized in a flight-relevant Titan entry environment (98% N2, 2% CH4). Titan,
Saturn’s largest moon, is the subject of many future exploration studies due to the
Cassini-Huygens mission revealing Titan’s Earth-like meteorology and geology as well
as its subsurface ocean [171]. Accurate characterization of the UV shock-layer radiation
is especially important for a Titan aeroshell as the peak radiative heat flux is predicted
to be four times greater than the peak convective heat flux [40, 164, 165]. Titan’s
atmosphere consists primarily of nitrogen with small amounts methane and argon. The
shock wave formed by the entering spacecraft dissociates the methane which then reacts
with the atmospheric nitrogen forming CN molecules. These CN radicals are strong
radiators especially in the UV and are responsible for the majority of the radiative
heating [40, 164, 165].
5.3.1 Shock Tube Test Set-up
The EAST facility at NASA Ames Research Center is a ground-based facility
that produces flight relevant radiation. The facility uses four separate spectrometers to
observe and characterize radiation from the shock wave from VUV (120 nm) to near
infrared (1700 nm) [163, 40, 172]. A schematic of the EAST facility is shown in Fig.
5.3.2 as well as the test set-up for characterizing the performance of the GaN-based
photodetectors in the shock tube. The 5 mm × 5 mm GaN-on-sapphire photodetector
chip was epoxied (Devcon, 5 Minute) and wirebonded to a transistor outline (TO)
header (Spectrum Semiconductor Materials, Inc., TO-8). To protect the wirebonds
during shock testing they were encapsulated in epoxy while leaving the photodetectors
exposed to the shock wave. The packaged photodetectors were soldered to wire leads in
the vacuum-tight shock port holder (see drawings in Appendix C) and installed in the
test section of the shock tube just downstream of the spectrometers used by the facility.
Two GaN-on-sapphire MSM photodetectors (referred to as photodetector 1 and
CHAPTER 5. IMPLEMENTATION 90
photodetector 2) were used to measure the UV radiance during two separate shock tests.
A source meter (2400 Series, Keithley) was used to take current-voltage measurements
pre- and post-shock as well as apply a voltage and measure the transient current response
during the shock test. Due to limitations of the equipment, the sampling rate during the
shock test was only about 8 Hz. Since the Titan entry conditions and test parameters
dictated that the shock speeds should be around 5-6 km/s, a separate high frequency
(5 MHz) data acquisition system (DAQ) was also used to acquire data during the shock
test. The DAQ was triggered to record 4 ms of data when the burst disk ruptured
whereas the source meter collected data for several minutes before and after the shock
event. The high frequency measurements were taken as voltage measurements (Vmeas)
across a resistor from which the photodetector current was calculated. For all tests, the
applied voltage (Vapp) was 200 mV.
Figure 5.3.2. Test set-up for in situ UV shock radiance measurements in NASA Ames’
electric arc shock tube. Reprinted from [118], with the permission of AIP Publishing.
CHAPTER 5. IMPLEMENTATION 91
5.3.3 Shock Tube Test Results
Table 5.3.2 lists the parameters for the two shock tests conducted, with
photodetector 1 active during the first shock test and photodetector 2 active during the
second. The first shock test melted the front edge of the TO header and broke the
wirebonds to photodetector 1, rendering the low frequency transient current
measurements useless as shown in Fig. 5.3.3a. The transient current response of
photodetector 2 during the second test, measured with the source meter (8 Hz sampling
frequency), is shown in Fig. 5.3.3b. The shock event is clearly visible in this data and
the slow current decay after the shock has past is consistent with the persistent
photoconductivity phenomena described in Chapter 1. The unilluminated (dark)
current-voltage response for photodetector 2 is unchanged from pre- to post-shock as
shown in the inset of Fig. 5.3.3b. A change in the dark current would indicate material
damage (displacements, dislocations, etc.) and/or electrical degradation (Schottky
barrier height). Since the unilluminated current-voltage response is unchanged, this
indicates that there is little degradation in the device performance after being subjected
to the shock wave.
Table 5.3.2. Shock test parameters.
Test 1 Test 2
Active Photodetector Photodetector 1 Photodetector 2
Test Gas 98% N2, 2% CH4 98% N2, 2% CH4
Driver Gas 93.2% He, 6.8% Ar 93.2% He, 6.8% Ar
Test Section Pressure (torr) 0.1 0.1
Shock Speed (km/s) 5.92 5.32
Equilibrium Temperature of Shocked Gas (K) 5589.53 5324.28
The 5 MHz sampling frequency transient current response during the shock test
is shown in Fig. 5.3.4a for photodetector 1 and Fig. 5.3.4b for photodetector 2. A moving
average filter with a window size of 100 samples was applied to the collected data to
reduce the noise. The filtered signal displays a distinct increase in current after the shock
passes the detector giving a direct measure of the UV shock-layer radiation. In
CHAPTER 5. IMPLEMENTATION 92
Fig. 5.3.4, the time when the shock and contact surface pass the detector is easily
identifiable due the large spikes in current. These spikes can be attributed to the exposed
contacts collecting free electrons in the partially ionized gases or forming a ground loop
between the measurement circuit and the facility.
Figure 5.3.3. Transient current response (8 Hz sampling frequency) for (a) photodetector
1 during shock test 1 and (b) photodetector 2 during shock test 2. Unilluminated pre- and
post-shock current-voltage curves (device functionality tests) are compared in the insets.
Figure 5.3.4. Transient current response (5 MHz sampling frequency) during shock test
for (a) photodetector 1 and (b) photodetector 2. The filtered signal displays a distinct increase
in current after the shock passes the detector giving a direct measure of the UV shock-layer
radiation. Reprinted from [118], with the permission of AIP Publishing.
For GaN-based photodetectors, the photo-generated current is proportional to
the square root of the applied optical power [61]. Applying this relation to the
CHAPTER 5. IMPLEMENTATION 93
measurements from the spectral response characterization at 365 nm (termed “Spectral
Response Data” in Fig. 5.3.5) results in the fits (termed “P1/2 Fit” in Fig. 5.3.5) shown
in Fig. 5.3.5. The photo-generated current from the shock tests can be calculated from
Fig. 5.3.4 as the average current before the shock subtracted from the average of the
maximum 100 filtered samples between the shock and contact surface. Overlaying this
photo-generated current from the shock tests (termed “Shock Test 1” and “Shock Test
2” in Fig. 5.3.5) on top of the square root of power fits in Fig. 5.3.5 results in a measured
optical power of about 0.30 μW for both shock tests. From the EAST facility
spectrometers, the average UV (190 nm to 370 nm) intensity during the two shock tests
were measured to be about 49.5 μJ/cm2 and 45.0 μJ/cm2. Multiplying these values by
the active area of the respective GaN-based photodetector and the characteristic time
gives 0.31 μW and 0.28 μW which are in excellent agreement with the 0.30 μW result
from the square root of power fit to the data from GaN-based photodetectors.
Figure 5.3.5. Photo-generated current (Iphoto - Idark) for the GaN-based photodetectors as a
function of applied optical power calculated as a square root of power fit through the spectral
response characterization data at 365 nm. By overlaying the photo-generated current
calculated from the shock test data for (a) photodetector 1/shock test 1 and (b) photodetector
2/shock test 2, the measured optical power is found to be about 0.30 μW for both shock tests.
Reprinted from [118], with the permission of AIP Publishing.
Using NASA’s chemical equilibrium with applications (CEA) software [173],
the temperatures behind the shocks were calculated to be, 5589.53 K and 5324.28 K for
CHAPTER 5. IMPLEMENTATION 94
tests 1 and 2, as listed in Table 5.3.2. These very high temperatures persist for less than
1 ms and thus the photodetectors are not heated substantially. Therefore it is reasonable
to neglect temperature effects in the proceeding analysis. However, the descent through
a planetary bodies’ atmosphere could be on the order of several minutes. To protect the
photodetectors from these very high temperatures over this relatively long duration, they
could be implemented behind a window or on the backside of the aeroshell, similar to
the implementation of previous heat shield sensors. However, due to the temperature
tolerance of GaN, these sensors can be placed closer to the surface of the TPS than
previously possible thus increasing the understanding of the UV radiative heat load.
Figure 5.3.6a shows the GaN photodetector chip epoxied to the TO header and
mounted in the shock tube port holder before shock testing. The damaged TO header
after the first shock test is shown in Fig. 5.3.6b. The upstream side of the TO header
experienced significant melting (which broke the wirebonds to photodetector 1) while
the downstream side was relatively unaffected. Since the packaged GaN chip does not
have the same curvature as the shock tube (i.e. it is flat), it sits just inside the diameter
of the shock tube and thus it is likely that an oblique shock formed over the leading edge
of the TO header. Thus, the damage to the TO header is attributed to the formation of
the oblique shock. The TO header incurred slightly more damage after the second shock
test, as shown in Fig. 5.3.6c. Since the GaN-based photodetectors were able to measure
the shock-layer radiation and showed no measurable signs of electrical degradation
post-shock testing, packaging appears to be the limiting factor and requires further
development.
CHAPTER 5. IMPLEMENTATION 95
Figure 5.3.6. GaN photodetector chip epoxied to header mounted in shock tube port holder
(a) before shock testing, (b) post shock test 1, and (c) post shock test 2.
5.4 Conclusions
GaN-based UV photodetectors were implemented and tested in space-relevant
applications including a miniature sun sensor for spacecraft orientation. A GaN-based
miniature sun sensor capable of operation in the harsh environment of space was
developed using scalable microfabrication techniques, which can be leveraged to create
larger device arrays for high-resolution sun angle detection. Photodetectors with an
average PDCR of 2.4 and responsivity of 3200 A/W at -1 V bias were microfabricated
and integrated with an opaque aperture mask to create the UV-sensitive sun sensor. The
successful operation of this instrument at 0° and 45° incident light angles with an
average illuminated photodetector to dark photodetector selectivity of 2.1 was
demonstrated using a simple sun angle determination algorithm. This algorithm was
able to correctly identify the illuminated photodetector for ten test cases with results
similar to those shown here. The repeatability of the measurements and results show the
feasibility of a space-environment capable UV sun sensor using an array of GaN-based
photodetectors. The GaN-based miniature sun sensor was integrated with the QB50
CubeSat Discovery which will be launched in the near future and will enable in-flight
evaluation of the GaN-based photodetectors.
CHAPTER 5. IMPLEMENTATION 96
In addition to sun sensing applications, GaN-based UV photodetectors were also
tested in a shock tube for atmospheric entry UV radiation sensing. Atmospheric entry
conditions for many planetary bodies have been predicted to have high heat fluxes
dominated by shock-layer radiation rather than thermal convection. In order to design
thermal protection systems to enable exploration missions to these places, this heat load
needs to be well understood. Since GaN-based photodetectors are highly responsive to
UV radiation and are temperature tolerant, it is worth exploring their ability to detect
shock-layer radiation in the UV regime. GaN-on-sapphire UV photodetectors were
successfully used to detect UV radiation behind a shock in a simulated Titan entry
environment in an electric arc shock tube. The calculated optical power from the
transient current response measurements made by the GaN photodetectors was
0.30 μW, which is in agreement with the UV radiance measured by EAST facility
spectrometers. Furthermore, the electrical response of the photodetectors was not
measurably degraded by the shock as well as there was no visible damage to the
photodetectors post-shock. These results demonstrate the feasibility of small-scale
GaN-based sensors operating in the harsh atmospheric entry environment.
97
Chapter 6
Conclusions and Future Work
6.1 Conclusions
Wide bandgap semiconductors, such as GaN, are intriguing material platforms
for developing space-based sensors and electronics. Compared to Si, GaN has a much
higher melting point and atomic displacement energy which allows for operation in high
temperature and radiation-rich environments. GaN has a wide and direct bandgap of
3.4 eV which enables ultraviolet photodetection. Photon absorption (and thus
sensitivity) can be increased in GaN-based photodetectors by using transparent
electrodes such as graphene. Graphene has a high atomic displacement energy and has
been shown to create a thermally stable Schottky contact to GaN. The properties of both
GaN and graphene are well suited for harsh environment operation and the
graphene/GaN material platform, specifically, GaN UV photodetectors with graphene
electrodes, should be characterized as a space-exploration enabling technology.
However, there are still many challenges to overcome before GaN-based sensors are
CHAPTER 6. CONCLUSIONS AND FUTURE WORK 98
ready for space-flight, including but not limited to, addressing material quality issues,
advancement of microfabrication processing, and development of extreme environment
packaging schemes.
Chapter 4 of this thesis delved into the characterization of the graphene/GaN and
semitransparent Ni/Au/GaN photodetectors when operating in extreme environments
including high temperatures and under proton irradiation exposure. The graphene/GaN
photodetectors exhibited a higher PDCR compared to the semitransparent Ni/Au/GaN
photodetectors over all temperatures tested (-100°C to 200°C) which can be attributed
to graphene’s very high transmittance (> 80%) in the UV regime compared to
semitransparent Ni/Au (> 30%) as detailed in Chapter 3. For both types of
photodetectors, PDCR was seen to decrease with increasing temperature eventually
reaching zero above 200°C. The decreasing PDCR with increasing temperatures is due
to the thermal generation of charge carriers overwhelming the photo-generation at
elevated temperatures. The responsivity of both types of photodetectors was seen to
decrease at high temperatures due to increased lattice scattering as well as a reduced
bandgap. From transient current response data, it was shown that both graphene/GaN
and semitransparent Ni/Au/GaN photodetectors demonstrated faster rise and decay
times at elevated temperatures. The faster rise times at high temperatures were attributed
to a higher recapture rate of photo-excited carriers while the faster decay times were
attributed to thermal energy releasing trapped photo-generated carriers. Additionally,
these temperature-dependent trends seen in the experimental results match the
predictions of the semi-analytical model based on thermionic emissions theory
developed in Chapter 2. The graphene/GaN and semitransparent Ni/Au/GaN
photodetectors were also characterized when subjected to 2 MeV proton irradiation up
to a fluence of 3.8 × 1013 cm-2. Electrical measurements made throughout irradiation
testing showed no significant degradation in device performance in terms of PDCR and
responsivity. Additionally, Raman spectroscopy of 200 keV irradiated graphene showed
only slight increase of disorder in the form of an increased ID/IG ratio, the appearance of
the D'-peak, and a decrease in the 2D peak intensity. These results support the use of
CHAPTER 6. CONCLUSIONS AND FUTURE WORK 99
graphene and GaN as temperature tolerant and radiation-hard material platforms for UV
photodetection with high sensitivity and responsivity.
Chapter 5 detailed the implementation and testing of GaN-based photodetectors
in practical aerospace applications. An ultraviolet miniature sun sensor was created
using an array of GaN-based photodetectors coupled with an opaque aperture mask. The
miniature sun sensor demonstrated successful operation under 0° and 45° UV
illumination with an average illuminated photodetector to dark photodetector selectivity
of 2.1. The miniature sun sensor has been integrated with a CubeSat enabling future in-
space evaluation of the graphene/GaN material platform. The GaN-based
photodetectors were also implemented in an electric arc shock tube and in situ
measurements of UV shock-layer radiation in a Titan simulant environment were made.
The calculated optical power from the transient current response measurements made
by the GaN photodetectors was 0.30 μW, which is in agreement with the UV radiance
measured by EAST facility spectrometers. Furthermore, the electrical response of the
photodetectors was not measurably degraded by the shock as well as there was no visible
damage to the photodetectors post-shock. These results demonstrate the feasibility of
small-scale GaN-based sensors for future use in space sensing applications, specifically,
for spacecraft orientation and detection of radiative heating during atmospheric entry.
In summary, this thesis described the design and microfabrication of GaN UV
photodetectors with transparent graphene electrodes and the experimental test results
supporting the development of these devices for space-based sensing applications. The
graphene/GaN material platform overcomes the limitations of current Si-based sensor
technology by extending the operational temperature range and radiation exposure
limits. Sensors and electronics using the graphene/GaN material platform will require
less protective packaging which will increase the size of scientific payloads that can be
flown.
CHAPTER 6. CONCLUSIONS AND FUTURE WORK 100
6.2 Future Work
Future work should be conducted to further understand the operation of and
improve the performance of GaN-based ultraviolet photodetectors operating in harsh
environments. In terms of photodetector device design, it is important to consider ways
to reduce the dark (leakage) current without sacrificing performance. A photodetector
architecture that incorporates an insulator such as a metal-insulator-semiconductor
(MIS) photodetector is an intriguing option since the insulator layer will increase the
Schottky barrier height thus reducing the dark current [62]. However, for space-based
applications, the insulator material needs to be chosen with care and needs to withstand
high levels of radiation and extreme temperatures. For example SiO2 thin films have
been shown to trap charges after exposure to gamma radiation resulting in large leakage
currents [64, 174]. Thin films made of HfO2, Al2O3, and h-BN have displayed
robustness to gamma radiation and may be good options [174], however difficulty in
wet etching HfO2 may prove to be limiting in terms of fabrication [175, 176]. In addition
to adding an insulating layer, further research needs to be conducted to develop reliable
high-temperature capable Schottky contacts to GaN [177, 178, 179, 180]. A
metallization stack that can operate at high temperatures while maintaining a high
barrier height would be highly desirable and would allow GaN-based photodetectors to
operate up to 600°C or above.
Since AlGaN/GaN transistors have demonstrated operation at 600°C in air [23],
photodetectors utilizing the AlGaN/GaN material platform may further increase the
temperature limit of GaN-based photodetectors. AlGaN/GaN photodetectors have
shown fast room temperature response times [181, 182], high UV sensitivity [181, 183,
182], and the ability to tune the absorption edge through the Al doping concentration
[17]. Additionally, self-heating of suspended AlGaN/GaN photodetectors has shown to
be an effective way to reduce the highly undesirable phenomenon of persistent
photoconductivity [127]. Future work should be conducted to realize the potential of
AlGaN/GaN photodetectors for harsh environment sensing. Further discussion and
CHAPTER 6. CONCLUSIONS AND FUTURE WORK 101
preliminary experimental results for AlGaN/GaN MSM photodetectors are provided in
Appendix B.
Further testing is needed to fully understand the response of GaN-based
photodetectors operating in extreme environments (i.e. high temperature, low
temperature, radiation-rich, shock, corrosive, etc.). For example, it is important to
understand what happens to the microstructures of GaN, graphene, etc. when exposed
to extreme temperatures for long time periods and how that affects photodetector
sensitivity. Additionally, while operating in space, sensors and electronics can
experience extreme thermal cycling which may cause premature device degradation.
Also worth exploring is how long term exposure to UV light changes device
performance. It was noted above that the MIS photodetector architecture results in lower
leakage current than MSM photodetectors, however extended exposure to UV light
tends to damage oxide layers [17, 184]. Consequently, the MIS structure may help
reduce leakage current and with correct selection of oxide, can withstand particle
radiation, but further work needs to be conducted with regards to material degradation
in the presence of high intensity UV light. In terms of proton radiation exposure,
irradiating to higher fluences would be very informative to determine when
photodetector sensitivity degrades. Also, it is important to understand how GaN sensors
and electronics respond to electrons, neutrons, and heavy ions for successful operation
in space.
Future work to further the implementation work discussed in Chapter 5 includes
analysis of the data sent back from the miniature sun sensor integrated with the QB50
CubeSat. This data analysis will not only demonstrate the performance of the miniature
sun sensor when subjected to the sun’s high intensity UV light outside the Earth’s
atmosphere, but will also inform on how well the materials (GaN, graphene, etc.)
themselves withstand the space environment. To further improve the performance of the
miniature sun sensor, the GaN-based array should be scaled to include more closely
spaced photodetectors (achievable with today’s micro- and nano-fabrication techniques)
and a taller spacer should be implemented to achieve the maximum field of view with
CHAPTER 6. CONCLUSIONS AND FUTURE WORK 102
the highest resolution. Another suggestion to improve the miniature sun sensor is to
consider alternate packaging methods. Wirebonding the GaN photodetector chip to a
PCB and encapsulating the wirebonds with epoxy worked well for proof of concept and
preliminary prototyping. However, as the number of photodetectors in the array is
increased the number of wirebonds will become overwhelming. Thus this is not a robust
solution. In lieu of GaN integrated circuit technology, which is currently being
researched and may be realized on a large scale in the near future [185, 186], it is
suggested to implement flip chip bonding [187, 188] and eliminate the excessive and
fragile wirebonds. In this packaging scheme, the sapphire substrate of the GaN-on-
sapphire wafer is the spacer in the miniature sun sensor and the photodetector array is
considered to be back-illuminated [99, 61].
The last suggestion for future work is to create a gratingless UV spectrometer
using an array GaN-based photodetectors. As discussed in Chapter 1, the absorption
edge of GaN-based photodetectors can be tuned through doping. Thus by selectively
doping the GaN film, an array of UV photodetectors that are sensitive to different
wavelengths can be created. One of the many applications for a gratingless UV
spectrometer is in detecting the UV spectra of shock-layer radiation during atmospheric
entry as discussed in Chapter 5.
103
Appendix A
MSM Photodetector Analytical Model Equations
To simulate the temperature-dependent dark current-voltage response of GaN-
based MSM photodetectors, the current transport mechanisms of thermionic emissions,
thermionic field emissions, and field emissions were considered in Chapter 2. The
thermionic, thermionic field, and field emission equations implemented in the analytical
model developed in Chapter 2 are detailed here.
When a voltage is applied to an MSM photodetector, one of the electrodes will
be reverse biased and the other forward biased. For this discussion, electrode 1 is reverse
biased and electrode 2 is forward biased, as shown in Fig. A.1.1. Therefore, the total
dark current for an MSM photodetector is the electron current from electrode 1 added
to the hole current from electrode 2. The three bias voltage operational regions for MSM
photodetectors: (1) voltages less than reach-through (voltage at which the sum of the
depletion widths equals the electrode spacing), (2) voltages greater than reach-through
but less than flat band (voltage at which the electric field at the forward biased electrode
APPENDIX A. MSM PHOTODETECTOR ANALYTICAL MODEL EQUATIONS 104
is zero), and (3) voltages greater than flat band but less than breakdown. The reach-
through voltage can be express as
𝑉𝑅𝑇 ≅𝑞𝑁𝑉𝐷𝐿2
2𝜀𝑠− √4𝑉𝐷
2 (A.1)
where q is elementary charge (1.602 × 10-19 C), N, is the net dopant concentration, VD
is the built-in potential, L is the electrode spacing, and εs is permittivity of the
semiconductor (relative permittivity multiplied by 8.854 × 10-12 F/m). It is easy to see
that with high net dopant concentration, the reach-through voltage will be large. For the
GaN-based MSM photodetectors fabricated (see Chapter 3) and characterized (see
Chapters 4 and 5) in this work, VRT > 1000 V, therefore only operational region (1),
voltages less than reach-through, will be considered.
Figure A.1.1. MSM photodetector band diagram.
APPENDIX A. MSM PHOTODETECTOR ANALYTICAL MODEL EQUATIONS 105
For thermionic emission theory, there are three main assumptions: (1) the barrier
height is much larger than kT/q, where k is Boltzmann’s constant (1.381 × 10-23 J/K)
and T is absolute temperature, (2) the device is in thermal equilibrium, and (3) two
current fluxes can be superimposed (one from metal to semiconductor and the other
from semiconductor to metal) [44]. Additionally, a symmetric MSM photodetector will
be assumed thus the current-voltage response will be symmetric and only the response
to positive bias voltages will be simulated. For symmetric MSM photodetectors
operating at voltages less than the reach-through voltage, the current density according
to thermionic emissions is [84]
𝐽𝑇𝐸 = 𝐽𝑛𝑠𝑒𝛽∆𝜑𝑛1(1 − 𝑒−𝛽𝑉1)
+ [𝑞𝐷𝑝𝑝𝑛0 tanh[(𝑥1 − 𝑥2)/𝐿𝑝]
𝐿𝑝(1 − 𝑒−𝛽𝑉1)
+𝐽𝑝𝑠𝑒
−𝛽𝑉𝐷
cosh[(𝑥2 − 𝑥1)/𝐿𝑝](𝑒𝛽𝑉2 − 1)]
(A.2)
where 𝛽 = 𝑞/𝑘𝑇, V1 and V2 are the applied voltage at electrodes 1 and 2, x1 and x2 are
the locations of the depletion widths for electrodes 1 and 2, Dp is the hole diffusion
constant, Pn0 is the equilibrium hole density, and Lp is the diffusion length. The term
∆𝜑𝑛1 (= √𝑞𝐸𝑚1
4𝜋𝜀𝑠) is known as the image-force lowering or Schottky barrier lowering
and accounts for the reduction in the Schottky barrier height in the presence of an
electric field. The electron and hole saturation current densities are
𝐽𝑛𝑠 ≡ 𝐴𝑛∗ 𝑇2𝑒−𝑞𝜑𝑛1/𝑘𝑇
𝐽𝑝𝑠 ≡ 𝐴𝑝∗ 𝑇2𝑒−𝑞𝜑𝑝2/𝑘𝑇
(A.3)
APPENDIX A. MSM PHOTODETECTOR ANALYTICAL MODEL EQUATIONS 106
where φn1 is the electron barrier height at electrode 1, and φp2 is the hole barrier height
at electrode 2. The effective Richardson constants for electrons and holes are [44]
𝐴𝑛∗ =
4𝜋𝑞𝑚𝑛𝑘2
ℎ3
𝐴𝑝∗ =
4𝜋𝑞𝑚𝑝𝑘2
ℎ3
(A.4)
where, mn is the effective mass of electrons, mp is the effective mass of holes, and h is
Planck’s constant (6.626 × 10-34 Js).
For symmetric MSM photodetectors, the current density according to thermionic
field emissions theory is [44]
𝐽𝑇𝐹𝐸 =𝐴𝑛
∗ 𝑇
𝑘√𝜋𝐸00𝑞 [𝑉1 +
𝜑𝑛1
𝑐𝑜𝑠ℎ2(𝐸00/𝑘𝑇)] 𝑒𝑥𝑝 (
−𝑞𝜑𝑛1
𝐸0) (𝑒𝑥𝑝 (
𝑞𝑉1
𝜀′) − 1)
+𝐴𝑝
∗ 𝑇√𝜋𝐸00𝑞(𝜑𝑝2 − 𝜑𝑛 − 𝑉2)
𝑘𝑐𝑜𝑠ℎ(𝐸00/𝑘𝑇)𝑒𝑥𝑝 (
−𝑞𝜑𝑛
𝑘𝑇
−𝑞(𝜑𝑝2 − 𝜑𝑛)
𝐸0) (𝑒𝑥𝑝 (
𝑞𝑉2
𝐸0) − 1)
(A.5)
where φn (= (𝐸𝑐 − 𝐸𝐹)/𝑞) is the fermi potential from conduction band edge in n-type
semiconductors. E0, E00, and ε’ are defined as follows where m* is mn or mp as
appropriate:
𝐸00 ≡𝑞ħ
2√
𝑁
𝑚∗𝜀𝑠, (A.6)
APPENDIX A. MSM PHOTODETECTOR ANALYTICAL MODEL EQUATIONS 107
𝐸0 ≡ 𝐸00 coth (𝐸00
𝑘𝑇) , (A.7)
𝜀′ =𝐸00
(𝐸00/𝑘𝑇) − tanh(𝐸00/𝑘𝑇). (A.8)
For symmetric MSM photodetectors, the current density according to field
emissions theory is [44]
𝐽𝐹𝐸 = 𝐴𝑛∗ (
𝐸00
𝑘)
2
(𝜑𝑛1 + 𝑉1
𝜑𝑛1) exp(−
2𝑞𝜑𝑛13/2
3𝐸00√𝜑𝑛1 + 𝑉1
)
+𝐴𝑝
∗ 𝑇𝜋 exp(−𝑞(𝜑𝑝2 − 𝑉2)/𝐸00)
𝑐1𝑘 𝑠𝑖𝑛(𝜋𝑐1𝑘𝑇)
(A.9)
where
𝑐1 ≡1
2𝐸00log [
4(𝜑𝑝2 − 𝑉2)
−𝜑𝑛]. (A.10)
The total dark current can be expressed as the sum of the current densities
multiplied by the photodetector active area (A) that is
𝐼𝑑𝑎𝑟𝑘 = (𝐽𝑇𝐸 + 𝐽𝑇𝐹𝐸 + 𝐽𝐹𝐸)𝐴. (A.11)
108
Appendix B
AlGaN/GaN-Based Photodetectors
B.1 Principal of Operation
When AlGaN is grown on top of GaN, a highly conductive sheet of electrons is
formed known as the two-dimensional electron gas (2DEG). The 2DEG forms due to
spontaneous and piezoelectric polarization of the AlGaN and GaN films [189, 190].
Since GaN is a piezoelectric material (applied stress generates a voltage and vice versa),
the piezoelectric polarization is induced due to the lattice mismatch between the AlGaN
and GaN which puts the AlGaN in tension. The 2DEG formed by the AlGaN/GaN
heterojunction is often used to create high electron mobility transistors (HEMTs) and
has been used as the sensing element for various sensors (surface acoustic wave
resonators, pressure or strain sensors, photodetectors, etc.) [189, 182, 191, 183]. Here,
AlGaN/GaN-based UV photodetectors will be examined.
For AlGaN/GaN MSM photodetectors, it has been shown that majority carriers
dominate the response and their charge transport is through the 2DEG [182]. Photons
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 109
absorbed in the AlGaN, 2DEG, or GaN will create electron-hole pairs which will first
travel to the 2DEG and then to the electrical contacts due to the strong vertical electric
field. This is in contrast to GaN-based MSM photodetectors where photo-generated
charge carriers travel laterally between the interdigitated electrodes. Since the AlGaN
layer is very thin (on the order of 10’s of nm) the response time of AlGaN/GaN-based
photodetectors is much faster than their MSM GaN-based counterparts which are
response time limited by the spacing between electrodes (on the order of a few μm).
For AlGaN/GaN HEMT photodetectors (termed phototransistors), electron
transport is through the 2DEG, similar to AlGaN/GaN MSM photodetectors, while hole
transport, which can greatly affect the device response, is dependent on where the
electron-hole pair was generated [183]. Holes generated in the AlGaN will be swept to
the surface due to the strong piezoelectric polarization field and will combine with
trapped surface electrons. This produces a positive gate bias effect and increases the
2DEG concentration. Holes generated in the GaN will drift toward the substrate due to
the built-in electric field creating a positive back-gate which further increases the 2DEG
concentration. The increased 2DEG concentration from absorbed UV photons should
result in UV phototransistors with high responsivities.
Devices fabricated on the AlGaN/GaN material platform have demonstrated
operation in temperatures as low as -257°C [192] and as high as 600°C [23].
Additionally, AlGaN/GaN devices have shown the ability to operate when exposed to
high levels of radiation [14]. Therefore, by using the AlGaN/GaN material platform, a
temperature tolerant, radiation hard, highly responsive, and fast response time UV
photodetector can be created. In the following sections, the fabrication and spectral
characterization as well as the high temperature, low temperature, and proton irradiation
response of AlGaN/GaN MSM photodetectors and AlGaN/GaN phototransistors will
be explored.
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 110
B.2 Fabrication and Spectral Characterization
Fabrication of the AlGaN/GaN MSM photodetectors began by growing 3 nm
GaN, 30 nm AlGaN, and 1.5 µm of GaN via metal organic chemical vapor deposition
(MOVCD) on a GaN-on-sapphire template (5 µm GaN, 200 nm AlN, and 500 µm
sapphire). MSM photodetectors were also fabricated on an AlGaN/GaN-on-silicon
(1 nm GaN, 30 nm AlGaN, 1.5 µm GaN, 1.5 µm buffer, and 625 µm Si) substrate. The
AlGaN/GaN substrates were then mesa etched via reactive ion etching for device
isolation. Next, 50 nm of SiO2 passivation was deposited via atomic layer deposition
(ALD) and subsequently etched to create the active area for each photodetector. Semi-
transparent electrodes with 30 to 50% UV transmittance were formed using a standard
metal lift-off procedure with 3 nm Ni / 10 nm Au. Lastly, electrical contact metal was
deposited (20 nm Ti / 200 nm Au). The photodetector active area is 250 μm × 250 μm
with an electrode length of 240 μm, width of 5 μm, and spacing of 5 μm. The complete
fabrication process for the AlGaN/GaN on GaN-on-sapphire MSM photodetectors is
shown in Fig. B.2.1 and a top-view optical image of a typical AlGaN/GaN MSM
photodetector is depicted in Fig. B.2.2.
Fabrication of the AlGaN/GaN phototransistors began with a mesa etch via
reactive ion etching of an AlGaN/GaN-on-silicon (1 nm GaN, 30 nm AlGaN, 1.5 µm
GaN, 1.5 µm buffer, and 625 µm Si) substrate. Ohmic source and drain contacts were
formed using a standard metal lift-off procedure with 20 nm Ti / 100 nm Al / 40 nm Pt
/ 80 nm Au and annealed in N2 at 850°C for 35 seconds. Gate electrodes were formed
using a standard metal lift-off procedure with 20 nm Ti / 80 nm Au. In an attempt to
increase the external quantum efficiency, photodetectors utilizing spine and rib gates
were fabricated in addition to rectangular gates. For both gate geometries, two sizes
were fabricated with photodetector active areas of 200 μm × 100 μm and
200 μm × 50 μm. The complete fabrication process for the AlGaN/GaN
phototransistors is shown in Fig. B.2.3 and top-view optical images of typical
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 111
AlGaN/GaN rectangular gate and spine and rib gate phototransistor are depicted in Fig.
B.2.4a and Fig. B.2.4b, respectively.
Figure B.2.1. AlGaN/GaN MSM photodetector fabrication process. (a) AlGaN/GaN on
GaN-on-sapphire substrate, (b) mesa etch, (c) SiO2 passivation, (d) etch SiO2 passivation, (e)
deposit semitransparent Ni/Au interdigitated electrodes, and (f) deposit Ti/Au electrical
contacts and interconnect. From [193]; reprinted by permission of the American Institute of
Aeronautics and Astronautics, Inc.
Figure B.2.2. Top-view optical image of an AlGaN/GaN MSM photodetector with
semitransparent Ni/Au electrodes.
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 112
The absorbance and reflectance spectra of the AlGaN/GaN on GaN-on-sapphire
substrate shown in Fig. B.2.5a. show a sharp cutoff near 370 nm demonstrating the
intrinsic visible-blindness of GaN. The absorbance and reflectance spectra of the
AlGaN/GaN-on-silicon substrate shown in Fig. B.2.5b demonstrates a similar result.
The absorbance spectrum for both substrates shows a strong peak at 365 nm due to
photon absorption in the GaN (3.4 eV) and a plateau between 310 nm to 260 nm which
can be attributed to absorption in the AlGaN (~4 eV) and buffer layers. The oscillations
in the absorbance and reflectance data above 400 nm can be attributed to Fabry-Pérot
interference from light reflected off the interfaces of the different layers [99, 194, 195].
The relatively high absorbance in the visible region for the AlGaN/GaN-on-silicon
substrate is attributed to the silicon rather than the photo-sensitive AlGaN and GaN
layers.
Figure B.2.3. AlGaN/GaN phototransistor fabrication process. (a) AlGaN/GaN-on-silicon
substrate, (b) mesa etch, (c) deposit Ti/Al/Pt/Au source and drain contacts, and (d) deposit
Ti/Au gate contact.
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 113
Figure B.2.4. Top-view optical image of an AlGaN/GaN phototransistor with (a) 200 μm
× 100 μm rectangular gate and (b) 200 μm × 100 μm spine and rib gate.
Figure B.2.5. Absorbance and reflectance spectra of the (a) AlGaN/GaN on GaN-on-
sapphire substrate and (b) AlGaN/GaN-on-silicon substrate.
Similar to the spectral response characterization of the GaN-on-sapphire
photodetectors described in Chapter 3, AlGaN/GaN-on-silicon MSM photodetectors
were characterized using a laser driven light source (EQ-1500, Energetiq) with nearly
constant radiance across the UV spectra and spectrometer (McPherson 218) with
diffraction grating (246.16 grooves/mm and 226 nm blaze). For wavelengths longer
than 340 nm, a long pass filter was implemented to reduce second order diffraction
affects. The current-voltage response of two AlGaN/GaN-on-silicon MSM
photodetectors illuminated with wavelengths from 200 nm to 500 nm is shown in Fig.
B.2.6. PDCR (as defined in Eq. 1.1) and responsivity (as defined in Eq. 1.2) as functions
of the illumination wavelength are shown in Fig. B.2.7 and Fig. B.2.8 respectively. The
jumps in the spectral PDCR data at 340 nm can be attributed to misalignments after the
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 114
long pass filter was implemented resulting in a slightly higher light intensity. The
increase in PDCR for wavelengths shorter than 400 nm and the relatively constant
PDCR from 200 nm to 340 nm matches the results from the absorbance and reflectance
data in Fig. B.2.5b. However, the responsivity continually increases for wavelengths
less than 400 nm which is not as expected and could be the result of a number of things
including: absorption in the AlN-based buffer layers generating electron-hole pairs,
persistent photoconductivity, or misalignments in the test set-up. Further testing is
needed to confirm and fully understand this result.
Figure B.2.6. Current-voltage response of two AlGaN/GaN-on-silicon MSM
photodetectors when illuminated with wavelengths from 200 nm to 500 nm.
Figure B.2.7. PDCR as a function of illumination wavelength of two AlGaN/GaN-on-
silicon MSM photodetectors.
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 115
Figure B.2.8. Spectral responsivity of two AlGaN/GaN-on-silicon MSM photodetectors.
B.3 Extreme Environment Operation
B.3.1 AlGaN/GaN Phototransistor High-Temperature Testing
Since traditional rectangular metal gate electrodes transmit less than 30% of
incident UV light [17], a spine and rib gate was designed and the performance compared
to that of the traditional rectangular gate. The experimental set-up for comparing these
two gate geometries when exposed to high temperatures is described in Chapter 4 and
shown in Fig. 4.1.1. The drain current (ID) versus gate-to-source bias (VGS) and drain
current versus drain-to-source bias (VDS) under dark and 365 nm illuminated conditions
at room temperature for an AlGaN/GaN phototransistor with a 200 μm × 100 μm
rectangular gate are shown in Fig. B.3.1a and Fig. B.3.1b respectively. The same curves
are shown in Fig. B.3.2a and Fig. B.3.2b for a 200 μm × 100 μm spine and rib gate. For
both devices, increasing drain-to-source bias or increasing gate-to-source bias results in
an increased drain current. When the rectangular gated device is illuminated with
365 nm light, the drain current increases slightly and the threshold voltage (Vth) shifts
to more negative drain-to-source bias. For the spine and rib gated device, 365 nm
illumination results in a dramatic shift of the threshold voltage to more negative drain-
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 116
to-source bias. The maximum PDCR of the rectangular gated device was 4.6 at a gate-
to-source bias of -11.6 V while the spine and rib device exhibited a maximum PDCR of
13.5 at a gate-to-source bias of -12.4 V. The maximum responsivity of the rectangular
gated device was 0.8 A/W at a gate-to-source bias of -11.3 V and the responsivity of the
spine and rib device was 2.3 A/W at -12.8 V gate-to-source bias. The almost three times
increase in PDCR and responsivity for the spine and rib gate compared to the rectangular
gate can be partially attributed to the increased photon absorption by the AlGaN and
GaN. Additionally, the electric field is altered by using the spine and rib gate geometry
which slight changes the characteristics of device response. This can be easily seen by
comparing the slopes of the ID-VGS curves for the two different gate geometries.
Figure B.3.1. AlGaN/GaN phototransistor (200 μm × 100 μm, rectangular gate) (a) drain
current versus gate voltage and (b) drain current versus drain-to-source bias under dark
(dashed lines) and 365 nm illuminated (solid lines) conditions at room temperature.
The drain current versus gate bias (VDS = 1 V) and drain current versus drain-
to-source bias (VGS = 0 V) under dark (dashed lines) and 365 nm illuminated (solid
lines) conditions from room temperature to 300°C are shown in Fig. B.3.3a and Fig.
B.3.3b respectively for a 200 μm × 100 μm rectangular gate phototransistor. The same
curves are shown in Fig. B.4.4a and Fig. B.4.4b for a 200 μm × 100 μm spine and rib
gate phototransistor. For both devices, increasing temperature results in a decreased
drain current and a decreased on/off current ratio which can be attributed to the reduced
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 117
mobility of the 2DEG. Under dark conditions, the threshold voltage shifts only slightly
as temperature increases, while under 365 nm illuminated conditions the threshold
voltage dramatically shifts to less negative values as temperature increases.
The maximum PDCR as a function of temperature for four AlGaN/GaN
phototransistors with various gate configurations (200 μm × 100 μm rectangular gate,
200 μm × 50 μm rectangular gate, 200 μm × 100 μm spine and rib gate, 200 μm × 50 μm
spine and rib gate) is shown in Fig. B.3.5a. The gate-to-source bias corresponding to the
maximum PDCR as a function temperature is shown in Fig. B.3.5b. Similarly, the
maximum responsivity as a function of temperature for four AlGaN/GaN
phototransistors with various gate configurations and the corresponding gate-to-source
bias as a function of temperature are shown in Fig. B.3.6. For all four gate
configurations, it is seen that the maximum PDCR is relatively constant up to 200°C but
significantly diminishes at higher temperatures and reaches zero at 350°C. On the
contrary, the maximum responsivity continually decreases as temperature increases
reaching zero at 350°C. The gate-to-source bias where the maximum PDCR and
maximum responsivity occur generally decreases as temperature increases which is in
accordance with the positive shifting of the threshold voltage with temperature.
Figure B.3.2. AlGaN/GaN phototransistor (200 μm × 100 μm, spine and rib gate) (a) drain
current versus gate voltage and (b) drain current versus drain-to-source bias under dark
(dashed lines) and 365 nm illuminated (solid lines) conditions at room temperature.
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 118
Figure B.3.3. AlGaN/GaN phototransistor (200 μm × 100 μm, rectangular gate) (a) drain
current versus gate voltage (VDS = 1 V) and (b) drain current versus drain-to-source bias (VGS
= 0 V) under dark (dashed lines) and 365 nm illuminated (solid lines) conditions from room
temperature to 300°C.
Figure B.3.4. AlGaN/GaN phototransistor (200 μm × 100 μm, spine and rib gate) (a) drain
current versus gate voltage (VDS = 1 V) and (b) drain current versus drain-to-source bias (VGS
= 0 V) under dark (dashed lines) and 365 nm illuminated (solid lines) conditions from room
temperature to 300°C.
These results demonstrate that AlGaN/GaN HEMTs are highly responsive to
incident UV light up to temperatures of 300°C. The incident UV light increases the drain
current and causes the threshold voltage to shift to more negative drain-to-source bias.
Therefore, the AlGaN/GaN HEMT architecture can be used to create highly responsive
and high operating temperature capable UV phototransistors. However, careful
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 119
consideration needs to be made when selecting packaging materials in order to shield
AlGaN/GaN HEMTs from unwanted UV exposure which will drastically alter device
performance.
Figure B.3.5. (a) Maximum PDCR as a function of temperature for four AlGaN/GaN
phototransistors with various gate configurations and (b) the corresponding gate voltage as a
function of temperature.
Figure B.3.6. (a) Maximum responsivity as a function of temperature for four AlGaN/GaN
phototransistors with various gate configurations and (b) the corresponding gate voltage as a
function of temperature.
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 120
B.3.2 AlGaN/GaN MSM Low-Temperature Testing
Cryogenic tests on GaN-based HEMTs have shown improved performance (in
the form of slightly higher transconductance and electron mobility, among other
parameters) at low temperatures compared to room temperature [192, 196]. Since higher
mobility leads to the collection of more photo-generated carriers, AlGaN/GaN-based
MSM photodetectors should also demonstrate good performance at low temperatures.
However, the experimental response of AlGaN/GaN MSM photodetectors under
cryogenic conditions has not been thoroughly investigated, if at all. Here, two
AlGaN/GaN on GaN-on-sapphire MSM photodetectors (called photodetector 1 and
photodetector 2) are characterized when subject to temperatures as low as -138.0°C
through in situ monitoring of the dark-current and photo-generated current.
The test set-up for characterizing the GaN-based photodetectors at near-
cryogenic temperatures using a liquid nitrogen cooled pressure and temperature
controlled chamber (Linkam Scientific THMS600-PS) is shown in Fig. B.3.7. A close-
up view of the stage (iced over during testing due to condensation of water vapor in air)
is shown in Fig. B.3.7b. The GaN photodetector chip was epoxied onto a PCB as shown
in Fig. B.3.7c with a k-type thermocouple epoxied on top of the chip to measure the
photodetector temperature which differed from the chuck set temperature. The PCB was
secured in place on the temperature-controlled chuck inside the chamber with the
photodetectors visible through the small window on the cover. Once the chamber was
closed, an argon purge was used to remove any water from the air so that it did not
condense on the photodetectors during cooling. Dark current-voltage measurements
were then taken with the window covered to block any ambient light from affecting the
measurement. The UV lamp was then turned on and a photocurrent measurement taken.
After these initial reference measurements, the temperature of the chuck was decreased
until the thermocouple read 0°C. The chip was allowed to soak at this low temperature
for about 30 minutes and then dark and illuminated measurements were retaken. This
process was continued down to -138°C. In addition to current-voltage measurements,
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 121
transient current measurements were taken at room temperature, -50°C, and -98°C.
Transient current measurements were taken by measuring dark current for five minutes
followed by five minutes illuminated (365 nm) and finally five minutes in the dark. A
semiconductor parameter analyzer (B1500A, Agilent Technologies) was used for all
electrical characterization and a 6 W, 365 nm UV lamp (UVP, EL series UV lamp) was
used for UV illumination. A UV light meter (Sper Scientific UVA/UVB Light Meter)
was used to measure the incident UV intensity, which for all illuminated electrical
measurements was 0.7 mW/cm2. For this intensity, the optical power applied on the
photodetectors is approximately 438 nW.
The chuck set temperature and corresponding measured thermocouple
temperature (located on the GaN photodetector chip) are listed in Table B.3.1. This
difference in temperature is likely due to the fact that the photodetectors are not in direct
thermal contact with the chuck. Additionally, the argon purge is contributing to
convection across the surface. Thus, all results presented are with respect to the
thermocouple temperature.
Figure B.3.7. (a) Overall experimental test set-up for characterizing the low-temperature
response of GaN-based photodetectors, (b) temperature controlled stage (frozen over during
testing), and (c) GaN-based photodetector chip epoxied to a PCB. From [193]; reprinted by
permission of the American Institute of Aeronautics and Astronautics, Inc.
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 122
Table B.3.1. Measured thermocouple temperature versus chuck set temperature.
Thermocouple Temperature
(± 2.5°C)
Chuck Temperature
(± 0.15°C)
23.3 25.0
-0.2 -7.0
-49.9 -70.0
-98.0 -142.0
-138.0 -188.0
The current-voltage curves for two AlGaN/GaN on GaN-on-sapphire MSM
photodetectors under dark (solid lines) and 365 nm illuminated (dashed lines)
conditions from room temperature to -138°C are shown in Fig. B.3.8. As the
temperature decreases, the dark-current is reduced which is consistent with the idea that
thermal excitation of electrons is a dominate factor in the creation of dark-current. Thus
by lowering the temperature the dark-current is suppressed. However, this effect is not
permanent as can be seen in the dark-current thermal recovery of photodetector 1 after
exposure to -138°C shown in the inset in Fig. B.3.8a. Following the same trend as the
dark-current, the photocurrent also decreases with decreasing temperature; this is
expected since the photocurrent is generated by both thermal excitation and photo-
excitation of charge carriers. The PDCR and responsivity as functions of temperature at
-8 V bias for photodetectors 1 and 2 are shown in Fig. B.3.9a and Fig. B.3.9b,
respectively. From these figures, it can be said that PDCR and responsivity are not
significantly affected by temperature down to -138.0°C.
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 123
Figure B.3.8. AlGaN/GaN on GaN-on-sapphire MSM photodetector current-voltage
curves under dark (solid lines) and 365 nm illuminated (dashed lines) conditions from room
temperature to -138°C for (a) photodetector 1 and (b) photodetector 2. The inset in (a) shows
the dark-current thermal recovery of photodetector 1 after exposure to -138°C.
Figure B.3.9. (a) PDCR and (b) responsivity as a function of temperature at -8 V bias for
photodetectors 1 and 2.
The transient current response of photodetector 1 at -0.2°C, -49.9°C and -98.0°C
is shown in Fig. B.3.10. For these measurements, the device was biased at -8 V and
dark-current was measured for 5 minutes, then the UV light was switched on and
photocurrent (365 nm, 0.7 ± 0.1 mW/cm2) measured for five minutes, and finally the
UV light was turned off and dark-current measured for five minutes. Once again, the
reduction of dark-current and photocurrent with decreasing temperatures is readily
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 124
apparent. Also apparent is the increase in the decay time (time between 90% and 10%
steady-state photocurrent) with decreasing temperatures. For photodetector 1, at -0.2°C
the decay time was 132 seconds and for temperatures lower than -49.9°C the decay time
was greater than the five minute measurement time. The decay time is much longer than
the rise time (time between 10% and 90% steady-state photocurrent), which has been
seen in GaN-based and other III-V photodetectors and could be due to impurities,
dislocations, and deep-level defects creating a persistent photoconductivity [55, 56, 57,
58, 181, 59]. Defects act as traps for photo-generated carriers at low temperatures. At
higher temperatures, thermal energy is able to release these carriers. Therefore, the
increase in decay time experienced at low temperatures can be attributed to the electron-
hole recombination rate slowing down (decreasing recapture rate of photo-excited
electrons) at decreased temperatures [55, 56, 57].
Figure B.3.10. Transient current response of photodetector 1 at different temperatures under
a 365 nm UV intensity of 0.7 ± 0.1 mW/cm2 and a bias of -8 V. From [193]; reprinted by
permission of the American Institute of Aeronautics and Astronautics, Inc.
When the chuck set temperature was as low as it could be (-190°C), the on-chip
thermocouple read -138°C. In order to test the photodetectors at lower temperatures, the
PCB with AlGaN/GaN photodetector chip was held in liquid nitrogen vapor. Figure
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 125
B.3.11 shows the current-voltage response for photodetector 1 at 23.3°C and while in
liquid nitrogen vapor when the on-chip thermocouple read -195.0°C. This figure clearly
shows that AlGaN/GaN MSM photodetectors are still operational down to -195.0°C.
The PDCR and responsivity at -8 V bias were 0.06 and 173 A/W. The relatively poor
performance of the photodetector at this cryogenic temperature can be attributed to the
liquid nitrogen vapor blocking the UV light. Additionally, the inset of Fig. B.3.11 shows
the condensation on the GaN chip and PCB from the liquid nitrogen.
Figure B.3.11. Current-voltage response for photodetector 1 at 23.3°C and while in liquid
nitrogen vapor at -195.0°C. The inset shows the condensation on the GaN chip and PCB
from the liquid nitrogen.
B.3.3 AlGaN/GaN MSM Proton Irradiation
The experimental set-up for proton irradiation and electrical characterization of
GaN-based photodetectors is described in Chapter 4 and shown in Fig. 4.4.1. The dark
and illuminated (365 nm) scaled current-voltage response as a function of 2 MeV proton
irradiation fluence for an AlGaN/GaN on GaN-on-sapphire MSM photodetector with
semitransparent Ni/Au electrodes is shown in Fig. B.3.12. In an attempt to clearly show
the effect proton irradiation has on the photo-generated current, the current-voltage
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 126
curves were scaled such that the dark current is the same at each irradiation fluence.
From Fig. B.3.12 it is evident that there is little to no degradation, in terms of UV
response, of the AlGaN/GaN photodetectors when subjected to 2 MeV protons with
fluences up to about 3.8 × 1013 cm-2. The PDCR and responsivity as functions of proton
fluence for three semitransparent Ni/Au AlGaN/GaN on GaN-on-sapphire
photodetectors are shown in Fig. B.3.13. From examining how these figures of merit
change with increasing irradiation fluence, it is again apparent that the UV response
experiences minimal degradation. The average pre-irradiation PDCR and responsivity
were 1.13 and 4.88 A/W and the 30 days post-irradiation PDCR and responsivity were
0.75 and 8.59 A/W as listed in Table B.3.2.
Figure B.3.12. Dark and illuminated (365 nm) scaled current-voltage response as a function
of proton irradiation fluence for AlGaN/GaN on GaN-on-sapphire photodetectors with
semitransparent Ni/Au electrodes.
Compared to the proton irradiated GaN-on-sapphire photodetectors described in
Chapter 4, it is seen that the current is much lower for the Ni/Au AlGaN/GaN
photodetectors (about 1 mA at 5 V) compared to the Ni/Au GaN-on-sapphire
photodetectors (about 10 mA at 5 V). This intuitively makes sense since the Schottky
barrier height is higher for Ni on AlGaN (1.11-1.36 eV) than it is for Ni on GaN (0.84-
1.00 eV) [197]. The PDCR is larger for AlGaN/GaN-based photodetectors (about 1)
APPENDIX B. AlGaN/GaN-BASED PHOTODETECTORS 127
than GaN-based photodetectors (about 0.1). Conversely, the responsivity is much larger
for GaN-based photodetectors (about 300 A/W) than for AlGaN/GaN-based
photodetectors (about 13 A/W). The photo-generated current (dark current subtracted
from the photocurrent) is higher for the GaN-based photodetectors which once again
can be attributed to the lower Schottky barrier height. While the GaN-based
photodetectors have higher photo-generated current, they also have higher dark current
which results in lower PDCR and higher responsivity when compared to the
AlGaN/GaN photodetectors. Additionally, the high responsivities exhibited by both
devices has been previously reported in literature and attributed to trapping of photo-
generated holes by surface states at the metal-semiconductor interface resulting in a
further lowering of the Schottky barrier height and increased leakage current [53, 54].
Figure B.3.13. (a) PDCR and (b) responsivity as a function of proton fluence for three
semitransparent Ni/Au AlGaN/GaN on GaN-on-sapphire photodetectors.
Table B.3.2. Average pre- and 30 days post-irradiation PDCR and responsivity for semi-
transparent Ni/Au AlGaN/GaN on GaN-on-sapphire photodetectors.
Electrode Material PDCR Responsivity (A/W)
Pre-Irradiation Ni/Au 1.13 4.88
Post-Irradiation Ni/Au 0.75 8.59
128
Appendix C
Experiment Drawings
APPENDIX C. EXPERIMENT DRAWINGS 129
Fig
ure C
.1.
Full Q
B50 m
iniatu
re sun sen
sor in
tegratio
n circu
it diag
ram.
APPENDIX C. EXPERIMENT DRAWINGS 130
Fig
ure C
.2.
Draw
ings o
f spectro
meter ad
apter fo
r spectral resp
onse ch
aracterization
.
APPENDIX C. EXPERIMENT DRAWINGS 131
Figure C.2. Drawings of spectrometer adapter for spectral response characterization.
APPENDIX C. EXPERIMENT DRAWINGS 132
Fig
ure C
.2.
Draw
ings o
f spectro
meter ad
apter fo
r spectral resp
onse ch
aracterization
.
APPENDIX C. EXPERIMENT DRAWINGS 133
Fig
ure C
.2.
Draw
ings o
f spectro
meter ad
apter fo
r spectral resp
on
se characterizatio
n.
APPENDIX C. EXPERIMENT DRAWINGS 134
Fig
ure C
.2.
Draw
ings o
f spectro
meter ad
apter fo
r spectral resp
on
se characterizatio
n.
APPENDIX C. EXPERIMENT DRAWINGS 135
Fig
ure C
.3.
Shock
tube assem
bly
draw
ings.
APPENDIX C. EXPERIMENT DRAWINGS 136
Fig
ure C
.3.
Shock
tube assem
bly
draw
ings.
APPENDIX C. EXPERIMENT DRAWINGS 137
Fig
ure C
.3.
Shock
tube assem
bly
draw
ings.
APPENDIX C. EXPERIMENT DRAWINGS 138
Fig
ure C
.3.
Shock
tube assem
bly
draw
ings.
139
Appendix D
Fabrication Run Sheet
Steps preformed in the Stanford Nano Shared Facility (SNSF) denoted, all other steps
preformed in Stanford Nanofabrication Facility (SNF).
1. Clean tweezers
a. Acetone, 10 mins
b. Methanol, 10 mins
c. IPA, 10 mins
2. Clean wafer(s) (including silicon dummy wafers)
a. 9:1 Piranha (H2SO4 (sulfuric acid): H2O2 (hydrogen peroxide)),
wbflexcorr2, 30 mins, 120°C
b. Water rinse, N2 dry
***************************** MESA ******************************
********************************************************************
APPENDIX D. FABRICATION RUN SHEET 140
LITHO – MESA ETCH
3. Dehydrate
a. 5 mins, 115°C hotplate
4. YES oven: dehydrate and prime with HMDS
5. Headway manual resist spinner
a. Shipley 3612, 60 sec, 5000 rpm
b. 1 min, 90°C hotplate
6. KarlSuss aligner
a. Clean mask: acetone, methanol, IPA, N2 dry
b. Hard contact, 40 um gap, expose 1.1 sec (Check exposure time on dummy
wafer first!)
c. 1 min, 115°C hotplate
7. Develop
a. MF-26A, 55 sec (Check development time on dummy wafer first!)
b. Water rinse and N2 dry
8. Inspect under microscope
9. Hard bake
a. 5 mins, 115°C hotplate
MESA ETCH
10. Oxford III-V etcher, xlab_GaN_slow_etch (Check etch rate on dummy wafer
first!)
11. Clean
a. 9:1 Piranha (H2SO4 (sulfuric acid): H2O2 (hydrogen peroxide)),
wbflexcorr2, 30 mins, 120°C
b. Water rinse, N2 dry
***************************** OHMIC *****************************
********************************************************************
APPENDIX D. FABRICATION RUN SHEET 141
LITHO - OHMIC
12. Dehydrate
a. 5 mins, 115°C hotplate
13. YES oven: dehydrate and prime with HMDS
14. Headway manual resist spinner
a. LOL 2000, 60 sec, 3000 rpm
b. 5 mins, 160°C hotplate
c. Shipley 3612, 60 sec, 5000 rpm
d. 1 min, 90°C hotplate
15. KarlSuss aligner
a. Clean mask: acetone, methanol, IPA, N2 dry
b. Hard contact, 40 um gap, expose 2.2 sec (check exposure time on dummy
wafer first)
c. 1 min, 115°C hotplate
16. Develop
a. MF-26A, 20 sec (check development time on dummy wafer first)
b. Water rinse and N2 dry
17. Inspect under microscope
18. Hard bake
a. 5 mins, 115°C hotplate
19. Drytek 2, descum
20. Weak ammonium hydroxide (1:20, ammonium hydroxide: water), 10 sec
OHMIC METAL
21. KJL e-beam evaporator - SNSF
a. 20 nm Ti / 100 nm Al / 40 nm Pt / 80 nm Au
22. Liftoff
a. 1165 overnight
b. Acetone, 5 mins
APPENDIX D. FABRICATION RUN SHEET 142
c. Methanol, 1 min
d. IPA, 1 min
e. N2 dry and inspect under microscope
f. Clean with PRS 1000, 20 mins, 40°C
g. Acetone, 5 mins
h. Methanol, 1 min
i. IPA, 1 min
j. N2 dry
23. Dehydrate
a. 5 mins, 115°C hotplate
24. Rapid thermal anneal
a. awin_r, P_850_N2, 850°C, 35 sec
***************************** OXIDE *****************************
********************************************************************
OXIDE DEPOSITION
25. Atomic layer deposition (ALD) SiO2, 50 nm
LITHO – OXIDE ETCH
26. Dehydrate
a. 5 mins, 115°C hotplate
27. YES oven: dehydrate and prime with HMDS
28. Headway manual resist spinner
a. Shipley 3612, 60 sec, 5000 rpm
b. 1 min, 90°C hotplate
29. KarlSuss aligner
a. Clean mask: acetone, methanol, IPA, N2 dry
b. Hard contact, 40 um gap, expose 1.1 sec
c. 1 min, 115°C hotplate
APPENDIX D. FABRICATION RUN SHEET 143
30. Develop
a. MF-26A, 55 sec
b. Water rinse and N2 dry
31. Inspect under microscope
OXIDE ETCH
32. 50:1 hydrofluoric acid (HF) dip, 1 min 30 sec
33. Remove photoresist
a. Acetone, 5 mins
b. Methanol, 1 min
c. IPA, 1 min
d. N2 dry
******************** SEMI-TRANSPARENT Ni/Au ********************
*********************************************************************
LITHO – SEMI-TRANSPARENT Ni/Au
34. Dehydrate
a. 5 mins, 115°C hotplate
35. YES oven: dehydrate and prime with HMDS
36. Headway manual resist spinner
a. LOL 2000, 60 sec, 3000 rpm
b. 5 mins, 160°C hotplate
c. Shipley 3612, 60 sec, 5000 rpm
d. 1 min, 90°C hotplate
37. KarlSuss aligner
a. Clean mask: acetone, methanol, IPA, N2 dry
b. Hard contact, 40 um gap, expose 2.2 sec
c. 1 min, 115°C hotplate
38. Develop
APPENDIX D. FABRICATION RUN SHEET 144
a. MF-26A for 20 sec
b. Water rinse and N2 dry
39. Inspect under microscope
40. Hard bake
a. 5 mins, 115°C hotplate
41. Drytek 2, descum
SEMI-TRANSPARENT Ni/Au
42. Innotec e-beam evaporator
a. 3 nm Ni
b. 10 nm Au
43. Liftoff
a. 1165 overnight
b. Acetone, 5 mins
c. Methanol, 1 min
d. IPA, 1 min
e. N2 dry and inspect under microscope
f. Clean with PRS 1000, 20 mins, 40°C
g. Acetone, 5 mins
h. Methanol, 1 min
i. IPA, 1 min
j. N2 dry
**************************** BONDPAD ****************************
*********************************************************************
LITHO - BONDPAD
44. Dehydrate
a. 5 mins, 115°C hotplate
45. YES oven: dehydrate and prime with HMDS
APPENDIX D. FABRICATION RUN SHEET 145
46. Headway manual resist spinner
a. LOL 2000, 30 sec, 3000 rpm
b. 5 mins, 160°C hotplate
c. Shipley 3612, 60 sec, 5000 rpm
d. 1 min, 90°C hotplate
47. Karlsuss aligner
a. Clean mask: acetone, methanol, IPA, N2 dry
a. Hard contact, 40 um gap, expose 2.2 sec
b. 1 min, 115°C hotplate
48. Develop
a. MF-26A for 20 sec
b. Water rinse and N2 dry
49. Inspect under microscope
50. Hard bake
a. 5 mins, 115°C hotplate
51. DryTek2, descum
BONDPAD METAL
52. KJL e-beam evaporator - SNSF
a. 20 nm Ti
b. 200 nm Au
53. Liftoff
a. 1165 overnight
b. Acetone, 5 mins
c. Methanol, 1 min
d. IPA, 1 min
e. N2 Dry and inspect under microscope
f. Clean with PRS 1000, 20 mins, 40°C
g. Acetone, 5 mins
APPENDIX D. FABRICATION RUN SHEET 146
h. Methanol, 1 min
i. IPA, 1 min
j. N2 dry
****************************** DICE ******************************
********************************************************************
LITHO FOR DICING
54. Dehydrate
a. 5 mins, 115°C hotplate
55. YES oven: dehydrate and prime with HMDS
56. Headway manual resist spinner
a. Shipley 3612, 60 sec, 5000 rpm
57. Hard bake
a. 5 mins, 115°C hotplate
58. Dice
59. Remove photoresist
a. Acetone, 5 mins
b. Methanol, 1 min
c. IPA, 1 min
d. N2 Dry and inspect under microscope
*************************** GRAPHENE ****************************
*********************************************************************
GRAPHENE TRANSFER - SNSF
60. Start with graphene grown on copper through chemical vapor deposition (CVD)
61. Cut a piece of graphene slightly larger than sample size, using weigh paper as a
support while cutting
62. Flatten piece between two microscope slides
63. Attach to microscope slide using a tiny drop of water
APPENDIX D. FABRICATION RUN SHEET 147
64. Spin PMMA (495,000g/mol, 4% dissolved in anisole) onto top side of sample at 0
rpm for 10 sec and 1000 rpm for 45 sec
65. Let PMMA rest for 12 hours or bake at 100°C for 1 min
66. Flip piece over, etch away backside graphene, 30 sec, 3% oxygen, 300 W. Use a
glass slide to weigh down the edges so the pieces don’t get blown away.
67. Flip back over. Place in 1 M FeCl3 solution. Let sit until copper is etched away.
68. Scoop PMMA/graphene stack into DI water. Let sit for 10 mins then scoop into
2nd DI water. Let sit for 10 mins then scoop into 3rd DI water. Scoop
PMMA/graphene stack onto actual substrate. Absorb excess water with a clean
room wipe and let sit to dry
69. 15 mins, 150°C hotplate
70. Remove PMMA
a. Acetone, 5 mins
b. IPA, 5 mins
c. N2 dry
71. 10 mins, 200°C hotplate
LITHO – GRAPHENE ETCH
72. Headway manual resist spinner
a. Shipley 3612, 60 sec, 5000 rpm
b. 1 min, 90°C hotplate
73. KarlSuss aligner
a. Clean mask: acetone, methanol, IPA, N2 dry
a. Hard contact, 40 um gap, expose 1.1 sec
b. 1 min, 115°C hotplate
74. Develop
a. MF-26A, 55 sec
b. Water rinse and N2 dry
APPENDIX D. FABRICATION RUN SHEET 148
GRAPHENE ETCH - SNSF
75. Etch graphene, 60 sec, 3% oxygen, 300 W
76. Remove photoresist
a. Acetone, 5 mins
b. Methanol, 1 min
c. IPA, 1 min
d. N2 dry
149
Bibliography
[1] P. G. Neudeck, R. D. Meredith, L. Chen, D. J. Spry, L. M. Nakley and G. W.
Hunter, "Prolonged silicon carbide integrated circuit operation in Venus surface
atmospheric conditions," AIP Advances, vol. 6, p. 125119, 2016.
[2] R. L. Patterson, A. Hannoud, J. E. Dickman, S. Gerber, E. Overton and M.
Elbuluk, "Electrical devices and circuits for low temperature space
applications," NASA TM-212600, 2003.
[3] K. Lodders and B. Fegley Jr., The planetary scientist's companion, New York:
Oxford University Press, 1998.
[4] E. Kolawa and et al, "Extreme environment technologies for future space science
missions," NASA JPL D-32832, 2007.
[5] E. Pettinelli, B. Cosciotti, F. Di Paolo, S. E. Lauro, E. Mattei, R. Orosei and G.
Vannaroni, "Dielectric properties of Jovian satellite ice analogs for subsurface
radar exploration: A review," Review of Geophysics, vol. 53, pp. 593-641, 2015.
[6] T. S. Balint, E. A. Kolawa, J. A. Cutts and C. E. Peterson, "Extreme environment
technologies for NASA's robotic planetary exploration," Acta Astronautica, vol.
63, pp. 285-298, 2008.
BIBLIOGRAPHY 150
[7] J. D. Cressler and H. A. Mantooth, Extreme environment electronics, Boca
Raton: Taylor & Francis Group, 2013.
[8] M. P. Petkov, "The effects of space environments on electronic components,"
Jet Propulsion Laboratory, 2003.
[9] R. W. Hobbs and S. L. Knapp, "Planetary temperatures at 9.55 mm
wavelengths," Icarus, vol. 14, pp. 204-209, 1971.
[10] V. Cottini, C. Nixon, D. E. Jennings, R. de Kok, N. A. Teanby, P. G. J. Irwin
and F. M. Flasar, "Spatial and temporal variations in Titan's surface temperatures
from Cassini CIRS observations," Planetary and Space Science, vol. 60, pp. 62-
71, 2012.
[11] K. Weide-Zaage and M. Chrzanowska-Jeske, Semiconductor devices in harsh
conditions, Boca Raton, Florida: CRC Press, Taylor & Francis Group, 2017.
[12] P. G. Neudeck, R. S. Okojie and L.-Y. Chen, "High-temperature electronics-a
role for wide bandgap semiconductors?," Proceedings of the IEEE, vol. 90, pp.
1065-1076, 2002.
[13] C. Claeys and E. Simoen, Radiation effects in advanced semiconductor materials
and devices, New York: Springer, 2002.
[14] S. J. Pearton, F. Ren, E. Patrick, M. E. Law and A. Y. Polyakov, "Review-
Ionizing radiation damage effects on GaN devices," ECS Journal of Solid State
Science and Technology, vol. 5, pp. Q35-Q60, 2016.
[15] R. Nybakken, "The Juno mission to Jupiter - A pre-launch update," in IEEE
Aerospace Conference, Big Sky, MT, 2011.
[16] J. E. P. Connerney, M. Benn, J. B. Bjarno, T. Denver, J. Espley, J. L. Jorgensen,
P. S. Jorgensen, P. Lawton, A. Malinnikova, J. M. Merayo, S. Murphy, J. Odom,
R. Oliversen, R. Schnurr, D. Sheppard and E. J. Smith, "The Juno magnetic field
investigation," Space Science Reviews, vol. 213, pp. 39-138, 2017.
BIBLIOGRAPHY 151
[17] E. Monroy, F. Omnes and F. Calle, "Wide-bandgap semiconductor ultraviolet
photodetectors," Semiconductor Science and Technology, vol. 18, pp. R33-R51,
2003.
[18] G. Guisbiers and S. Pereira, "Theoretical investigation of size and shape effects
on the melting temperature of ZnO nanostructure," Nanotechnology, vol. 18, p.
435710, 2007.
[19] K. Harafuji, T. Tsuchiya and K. Kawamura, "Melting point of wurtzite-type
GaN crystal," Physica Status Solidi C, vol. 0, pp. 2420-2423, 2003.
[20] Y. A. Goldberg, "Semiconductor near-ultraviolet photoelectronics,"
Semiconductor Science and Technology, vol. 14, pp. R41-R60, 1999.
[21] H. Kappert, N. Kordas, S. Dreiner and U. K. R. Paschen, "High temperature SOI
CMOS technology and circuit realization for applications up to 300C," in IEEE
International Symposium on Circuits and Systems (ISCAS), Lisbon, Portugal,
2015.
[22] R. S. Okojie, D. Lukco, V. Nguyen and E. Savrun, "4H-SiC piezoresistive
pressure sensors at 800C with observed sensitivity recovery," IEEE Electron
Device Letters, vol. 36, pp. 174-176, 2015.
[23] A. J. Suria, A. S. Yalamarthy, H. So and D. G. Senesky, "DC characteristics of
ALD-grown Al2O3/AlGaN/GaN MIS-HEMTs and HEMTs at 600C in air,"
Semiconductor Science and Technology, vol. 31, p. 115017, 2016.
[24] D. Maier, M. Alomari, N. Grandjean, J.-F. Carlin, M.-A. Diforte-Poisson, C.
Dua, S. Delage and E. Kohn, "InAlN/GaN HEMTs for operation in the 1000C
regime: a first experiment," IEEE Electron Device Letters, vol. 33, pp. 985-987,
2012.
BIBLIOGRAPHY 152
[25] F. Nava, G. Bertuccio, A. Cavallini and E. Vittone, "Silicon carbide and its use
as a radiation detector material," Measurement Science and Technology, vol. 19,
p. 102001, 2008.
[26] J. Zhan, Y. Bando, J. Hu and D. Golberg, "Nanofabrication on ZnO nanowires,"
Applied Physics Letters, vol. 89, p. 243111, 2006.
[27] A. Ionascut-Nedelcescu, C. Carlone, A. Houdayer, H. J. von Bardeleben, J.-L.
Cantin and S. Raymond, "Radiation hardness of gallium nitride," IEEE
Transactions on Nuclear Science, vol. 49, pp. 2733-2738, 2002.
[28] E. Wendler, "In-situ RBS channeling studies of ion implanted semiconductors
and insulators," AIP Conference Proceedings, vol. 1336, p. 621, 2011.
[29] H. Morkoc, "General properties of nitrides," in Handbook of nitride
semiconductors and devices: materials properties, physics, and growth, volume
1, Wiley-VCH Verlag GmbH & Co. KGaA, 2008, pp. 1-129.
[30] K.-a. Son, A. Liao, G. Lung, M. Gallegos, T. Hatake, R. D. Harris, L. Z. Scheick
and W. D. Smythe, "GaN-based high temperature and radiation-hard electronics
for harsh environments," Nanoscience and Nanotechnology Letters, vol. 2, pp.
89-95, 2010.
[31] S. J. Pearton, Y. H. Hwang and F. Ren, "Radiation effects in AlGaN/GaN and
InAlN/GaN high electron mobility transistors," ECS Transactions, vol. 66, pp.
3-13, 2015.
[32] D. G. Senesky, B. Jamshidi, K. B. Cheng and A. P. Pisano, "Harsh environment
silicon carbide sensors for health and performance monitoring of aerospace
systems: a review," IEEE Sensors Journal, vol. 9, pp. 1472-1478, 2009.
[33] R. S. Okojie, V. Nguyen, E. Savrun and D. Lukco, "Improved reliability of SiC
pressure sensors for long term high temperature applications," in 16th
BIBLIOGRAPHY 153
International Solid-State Sensors, Actuators and Microsystems Conference
(Transducers), Beijing, China, 2011.
[34] R. Maboudian, C. Carraro, D. G. Senesky and C. S. Roper, "Advances in silicon
carbide science and technology at the micro- and nanoscales," Journal of
Vacuum Science & Technology A: Vacuum, Surfaces, and Films, vol. 31, p.
050805, 2013.
[35] Z. Alaie, S. Mohammad Nejad and M. H. Yousefi, "Recent advances in
ultraviolet photodetectors," Materials Science in Semiconductor Processing,
vol. 29, pp. 16-55, 2015.
[36] K. Koike, T. Aoki, R. Fujimoto, S. Sasa, M. Yano, S.-i. Gonda, R. Ishigami and
K. Kume, "Radiation hardness of single-crystalline zinc oxide films," Physica
Status Solidi C, vol. 9, pp. 1577-1579, 2012.
[37] J. I. Ramirez, Y. V. Li, H. Basantani, K. Leedy, B. Bayraktaroglu, G. H. Jessen
and T. N. Jackson, "Radiation-hard ZnO thin film transistors," IEEE
Transactions on Nuclear Science, vol. 62, pp. 1399-1404, 2015.
[38] D. Decoster and J. Harari, "Ultraviolet photodetectors," in Optoelectronic
Sensors, Hoboken, New Jersey, John Wiley & Sons, Inc., 2009, pp. 181-222.
[39] J. Gomez-Elvira and e. al., "REMS: The environmental sensor suite for the mars
science laboratory rover," Space Science Reviews, vol. 170, pp. 583-640, 2012.
[40] D. Bose, M. J. Wright, D. W. Bogdanoff, G. A. Raiche and G. A. Allen Jr.,
"Modeling and experimental assessment of CN radiation behind a strong shock
wave," Journal of Thermophysics and Heat Transfer, vol. 20, pp. 220-230, 2006.
[41] C. O. Johnston and A. M. Brandis, "Features of afterbody radiative heating for
earth entry," Journal of Spacecraft and Rockets, vol. 52, pp. 105-119, 2015.
BIBLIOGRAPHY 154
[42] W.-C. Lien, D.-S. Tsai, D.-H. Lien, D. G. Senesky, J.-H. He and A. P. Pisano,
"4H-SiC metal-semiconductor-metal ultraviolet photodetectors in operation of
450C," IEEE Electron Device Letters, vol. 33, pp. 1586-1588, 2012.
[43] D.-S. Tsai, W.-C. Lien, D.-H. Lien, K.-M. Chen, M.-L. Tsai, D. G. Senesky, Y.-
C. Yu, A. P. Pisano and J.-H. He, "Solar-blind photodetectors for harsh
electronics," Scientific Reports, vol. 3, p. 2628, 2013.
[44] S. M. Sze and K. K. Ng, Physics of semiconductor devices, Hoboken: John
Wiley & Sons, Inc., 2007.
[45] A. Vijayakumar, R. M. Todi and K. B. Sundaram, "Amorphous-SiCBN-based
metal-semiconductor-metal photodetector for high-temperature applications,"
IEEE Electron Device Letters, vol. 28, pp. 713-715, 2007.
[46] S. Hou, H. P-E., C.-M. Zetterling and M. Ostling, "550C 4H-SiC p-i-n
photodiode array with two-layer metallization," IEEE Electron Device Letters,
vol. 37, pp. 1594-1596, 2016.
[47] F. Nava, G. Bertuccio, A. Cavallini and E. Vittone, "Silicon carbide and its use
as a radiation detector material," Measurement Science and Technology, vol. 19,
p. 102001, 2008.
[48] J. Hu, X. Xin, J. H. Zhao, B. L. VanMil, R. Myers-Ward, C. R. Eddy Jr. and D.
K. Gaskill, "Proton irradiation of ultraviolet 4H-SiC single photon avalanche
diodes," IEEE Transactions on Nuclear Science, vol. 58, pp. 3343-3347, 2011.
[49] E. V. Kalinina, A. A. Lebedev, E. Bogdanova, B. Berenquier, L. Ottaviani, G.
N. Violina and V. A. Skuratov, "Irradiation of 4H-SiC UV detectors with heavy
ions," Physics of Semiconductor Devices, vol. 49, pp. 540-546, 2015.
[50] M. Razeghi and A. Rogalski, "Semiconductor ultraviolet detectors," Journal of
Applied Physics, vol. 79, pp. 7433-7473, 1996.
BIBLIOGRAPHY 155
[51] E. Munoz, E. Monroy, J. A. Garrido, I. Izpura, F. J. Sanchez, M. A. Sanchez-
Garcia, E. Calleja, B. Beaumont and P. Gibart, "Photoconductor gain
mechanisms in GaN ultraviolet detectors," Applied Physics Letters, vol. 71, pp.
870-872, 1997.
[52] F. Binet, J. Y. Duboz, E. Rosencher, F. Scholz and V. Harle, "Mechanisms of
recombination in GaN photodetectors," Applied Physics Letters, vol. 69, pp.
1202-1204, 1996.
[53] F. Xie, H. Lu, D. Chen, R. Zhang and Y. Zheng, "GaN MSM photodetectors
fabricated on bulk GaN with low dark-current and high UV/visible rejection
ratio," Physica Status Solidi C, vol. 8, pp. 2473-2475, 2011.
[54] O. Katz, V. Garber, B. Meyler, G. Bahir and J. Salzman, "Gain mechanism in
GaN Schottky ultraviolet detectors," Applied Physics Letters, vol. 79, pp. 1417-
1419, 2001.
[55] M. De Vittorio, B. Poti, M. T. Todaro, M. C. Frassanito, A. Pomarico, A.
Passaseo, M. Lomascolo and R. Cingolani, "High temperature characterization
of GaN-based photodetectors," Sensors and Actuators A, vol. 113, pp. 329-333,
2004.
[56] M. T. Hirsch, J. A. Wolk and W. H. E. E. Walukiewicz, "Persistent
photoconductivity in n-type GaN," Applied Physics Letters, vol. 71, pp. 1098-
1100, 1997.
[57] C. H. Qiu and J. I. Pankove, "Deep levels and persistent photoconductivity in
GaN thin films," Applied Physics Letters, vol. 70, pp. 1983-1985, 1997.
[58] B. Poti, A. Passaseo, M. Lomascolo, R. Cingolani and M. De Vittorio,
"Persistent photocurrent spectroscopy of GaN metal-semiconductor-metal
photodetectors on long time scales," Applied Physics Letters, vol. 85, pp. 6083-
6085, 2004.
BIBLIOGRAPHY 156
[59] E. Arslan, S. Butun, S. B. Lisesivdin, M. Kasap, S. Ozcelik and E. Ozbay, "The
persistent photoconductivity effect in AlGaN/GaN heterostructures grown on
sapphire and SiC substrates," Journal of Applied Physics, vol. 103, p. 103701,
2008.
[60] E. Monroy, F. Calle, E. Munoz, F. Omnes, B. Beaumont and P. Gibart, "Visible-
blindness in photoconductive and photovoltaic AlGaN ultraviolet detectors,"
Journal of Electronic Materials, vol. 28, pp. 240-245, 1999.
[61] S. K. Zhang, W. B. Wang, I. Shtau, F. Yun, L. He, H. Morkoc, X. Zhou, M.
Tamargo and R. R. Alfano, "Backilluminated GaN/AlGaN heterojunction
ultraviolet photodetector with high internal gain," Applied Physics Letters, vol.
81, pp. 4862-4864, 2002.
[62] L. Sang, M. Liao, Y. Koide and M. Sumiya, "High-temperature ultraviolet
detection based on InGaN Schottky photodiodes," Applied Physics Letters, vol.
99, p. 031115, 2011.
[63] F. Liu, J. Xu, P. Wang, C. Li, D. Qian, N. Wang and X. Li, "Characteristics of
GaN-based p-i-n photodetector at low temperature and performances of
MCT/GaN-based infrared/ultraviolet dual-colour detector," Journal of Physics:
Conference Series, vol. 276, p. 012178, 2011.
[64] H. C. Chiamori, R. Miller, A. Suria, N. Broad and D. G. Senesky, "Irradiation
effects of graphene-enhanced gallium nitride (GaN) metal-semiconductor-metal
(MSM) ultraviolet photodetectors," in SPIE Proceedings: Micro/Nano Sensing
Technology for Harsh Environments, Baltimore, Maryland, 2015.
[65] G. Li, J. Zhang and X. Hou, "Temperature dependence of performance of ZnO-
based metal-semiconductor-metal ultraviolet photodetectors," Sensors and
Actuators A: Physical, vol. 209, pp. 149-153, 2014.
BIBLIOGRAPHY 157
[66] V. Avrutin, D. J. Silversmith and H. Morkoc, "Doping asymmetry problem in
ZnO: current status and outlook," Proceedings of the IEEE, vol. 98, pp. 1269-
1280, 2010.
[67] S. B. Zhang, S.-H. Wei and A. Zunger, "Intrinsic n-type versus p-type doping
asymmetry and the defect physics of ZnO," Physical Review B, vol. 63, p.
075205, 2001.
[68] H. So and D. G. Senesky, "ZnO nanorod arrays and direct wire bonding on GaN
surface for rapid fabrication of antireflective, high-temperature ultraviolet
sensors," Applied Surface Science, vol. 387, pp. 280-284, 2016.
[69] Y.-C. Chao, C.-Y. Chen, C.-A. Lin, Y.-A. Dai and J.-H. He, "Antireflection
effect of ZnO nanorod arrays," Journal of Materials Chemistry, vol. 20, pp.
8134-8138, 2010.
[70] D.-S. Tsai, C.-A. Lin, W.-C. Lien, H.-C. Chang, Y.-L. Wang and J.-H. He,
"Ultra-high-responsivity broadband detection of Si metal-semiconductor-metal
Schottky photodetectors improved by ZnO nanorod arrays," ACS Nano, vol. 5,
pp. 7748-7753, 2011.
[71] F. Yi, Y. Huang, X. Yan, Z. Bai, Z. Wang and X. Z. Q. Z. Y. Chen, "High
ultraviolet photoresponse from ZnO nanorod arrays," in International Photonics
and Optoelectronics Meetings, Nanophotonics, Nanoelectronics, and
Nanosensor, Wuhan, China, 2013.
[72] F. Lin, S.-W. Chen, J. Meng, G. Tse, X.-W. Fu, F.-J. Xu, B. Shen, Z.-M. Liao
and D.-P. Yu, "Graphene/GaN diodes for ultraviolet and visible photodetectors,"
Applied Physics Letters, vol. 105, p. 073103, 2014.
[73] S. Tongay, M. Lemaitre, T. Schumann, K. Berke, B. R. Appleton, B. Gila and
A. F. Hebard, "Graphene/GaN Schottky diodes: Stability at elevated
temperatures," Applied Physics Letters, vol. 99, p. 102102, 2011.
BIBLIOGRAPHY 158
[74] F. Banhart, J. Kotakoski and A. V. Krasheninnikov, "Structural defects in
graphene," ACS Nano, vol. 5, pp. 26-41, 2011.
[75] R. Zan, Q. M. Ramasse, R. Jalil, T. Georgiou, U. Bangert and K. S. Novoselov,
"Control of radiation damage in MoS2 by graphene encapsulation," ACS Nano,
vol. 7, pp. 10167-10174, 2013.
[76] S. R. Das, Q. Nian, M. Saei, S. Jin, D. Back, P. Kumar, D. B. Janes, M. A. Alam
and G. J. Cheng, "Single-layer graphene as a barrier layer for intense UV laser-
induced damages for silver nanowire network," ACS Nano, vol. 9, pp. 11121-
11133, 2015.
[77] R. C. Walker II, T. Shi, E. C. Silva, I. Jovanovic and J. A. Robinson, "Radiation
effects on two-dimensional materials," Physica Status Solidi A, vol. 213, pp.
3065-3077, 2016.
[78] C. Lee, J. Kim, S. Kim, Y. J. Chang, K. S. Kim, B. Hong and E. J. Choi, "Strong
hole-doping and robust resistance-decrease in proton-irradiated graphene,"
Scientific Reports, vol. 6, p. 21311, 2016.
[79] S. Lee, J. Seo, J. Hong, S. H. Park, J.-H. Lee, B.-W. Min and T. Lee, "Proton
irradiation energy dependence of defect formation in graphene," Applied Surface
Science, vol. 344, pp. 52-56, 2015.
[80] Y. Xu and J. Liu, "Graphene as transparent electrodes: fabrication and new
emerging applications," Small, vol. 12, pp. 1400-1419, 2016.
[81] F. Bonaccorso, Z. Sun, T. Hasan and A. C. Ferrari, "Graphene photonics and
optoelectronics," Nature Photonics, vol. 4, pp. 611-622, 2010.
[82] M. Kumar, H. Jeong, K. Polat, A. D. Okyay and D. Lee, "Fabrication and
characterization of graphene/AlGaN/GaN ultraviolet Schottky photodetector,"
Journal of Physics D: Applied Physics, vol. 49, p. 275105, 2016.
BIBLIOGRAPHY 159
[83] Y. Lin and X. Guo, "Ultrahigh photogain nanoscale hybrid photodetectors,"
Small, vol. 11, pp. 2856-2861, 2015.
[84] S. M. Sze, D. J. Coleman and A. Loya, "Current transport in metal-
semiconductor-metal (MSM) structures," Solid-State Electronics, vol. 14, pp.
1209-1218, 1971.
[85] J. C. Carrano, T. Li, P. A. Grudowshi, C. J. Eiting, R. D. Dupuis and J. C.
Campbell, "Current transport mechanisms in GaN-based metal-semiconductor-
metal photodetectors," Applied Physics Letters, vol. 72, pp. 542-544, 1998.
[86] S. J. Pearton, J. C. Zolper, R. J. Shul and F. Ren, "GaN: Processing, defects, and
devices," Journal of Applied Physics, vol. 86, pp. 1-78, 1999.
[87] H. B. Michaelson, "Relation between an atomic electronegativity scale and the
work function," IBM Journal of Research and Development, vol. 22, pp. 72-80,
1978.
[88] Y.-J. Lin, Y.-M. Chen, T.-J. Cheng and Q. Ker, "Schottky barrier height and
nitrogen-vacancy-related defects in Ti alloyed Ohmic contacts to n-GaN,"
Journal of Applied Physics, vol. 95, pp. 571-575, 2004.
[89] Y.-H. Wang, Y. C. Liang, G. S. Samudra, T.-F. Chang, C.-F. Huang, L. Yuan
and G.-Q. Lo, "Analytical modelling of high temperature characteristics of the
DC response for Schottky-gate AlGaN/GaN HEMT devices," Semiconductor
Science and Technology, vol. 28, p. 125010, 2013.
[90] Y. Zhou, D. Wang, C. Ahyi, C.-C. Tin, J. Williams, M. Park, N. M. Williams,
A. Hanser and E. A. Preble, "Temperature-dependent electrical characteristics
of bulk GaN Schottky rectifier," Journal of Applied Physics, vol. 101, p. 024506,
2007.
[91] W. Lim, J.-H. Jeong, J.-H. Lee, S.-B. Hur, J.-K. Ryu, K.-S. Kim, T.-H. Kim, S.
Y. Song, J.-I. Yang and S. J. Pearton, "Temperature dependence of current-
BIBLIOGRAPHY 160
voltage characteristics of Ni-AlGaN/GaN Schottky diodes," Applied Physics
Letters, vol. 97, p. 242103, 2010.
[92] J. Osvald, J. Kuzmik, G. Konstantinidis, P. Lobotka and A. Gerogakilas,
"Temperature dependence of GaN Schottky diodes I-V characteristics,"
Microelectronic Engineering, vol. 81, pp. 181-187, 2005.
[93] E. T. Yu and M. O. Manasreh, III-V nitride semiconductors: applications and
devices, New York: Taylor & Francis Books, Inc., 2003.
[94] M. E. Levinshtein, S. L. Rumyantsev and M. S. Shur, Properties of advanced
semiconductor materials: GaN, AlN, InN, BN, SiGe, New York: John Wiley &
Sons, Inc., 2001.
[95] T. T. Mnatsakanov, M. E. Levinshtein, L. I. Pomortseva, S. N. Yurkov, G. S.
Simin and M. A. Khan, "Carrier mobility model for GaN," Solid-State
Electronics, vol. 47, pp. 111-115, 2013.
[96] V. M. Bermudez, "Simple interpretation of metal/wurtzite-GaN barrier heights,"
Journal of Applied Physics, vol. 86, pp. 1170-1171, 1999.
[97] A. S. Yalamarthy and D. G. Senesky, "Strain- and temperature-induced effects
in AlGaN/GaN high electrode mobility transistors," Semiconductor Science and
Technology, vol. 31, p. 035024, 2016.
[98] H. So, J. Lim and D. G. Senesky, "Continuous V-grooved AlGaN/GaN surfaces
for high-temperature ultraviolet photodetectors," IEEE Sensors Journal, vol. 16,
pp. 3633-3639, 2016.
[99] G. Wang, F. Xie, H. Lu, D. Chen, R. Zhang, Y. Zheng, L. Li and J. Zhou,
"Performance comparison of front- and back-illuminated AlGaN-based metal-
semiconductor-metal solar-blind ultraviolet photodetectors," Journal of Vacuum
Science and Technology B, vol. 31, p. 011202, 2013.
BIBLIOGRAPHY 161
[100] D. Walker, X. Zhang, A. Saxler, P. Kung, J. Xu and M. Razeghi, "Al x Ga 1-x
N ultraviolet photodetectors grown on sapphire by metal-organic chemical-
vapor deposition," Applied Physics Letters, vol. 70, pp. 949-951, 1997.
[101] K. W. Liu, J. G. Ma, J. Y. Zhang, Y. M. Lu, D. Y. Jiang, B. H. Li, D. X. Zhao,
Z. Z. Zhang, B. Yao and D. Z. Shen, "Ultraviolet photoconductive detector with
high visible rejection and fast photoresponse based on ZnO thin film," Solid-
State Electronics, vol. 51, pp. 757-761, 2007.
[102] J. C. Carrano, P. A. Grudowski, C. J. Eiting, R. D. Dupuis and J. C. and
Campbell, "Very low dark current metal-semiconductor-metal ultraviolet
photodetectors fabricated on single-crystal GaN epitaxial layers," Applied
Physics Letters, vol. 70, pp. 1992-1994, 1997.
[103] H. So and D. G. Senesky, "Effect of frost formation on operation of GaN
ultraviolet photodetectors at low temperatures," IEEE Sensors Journal, vol. 17,
pp. 4752-4756, 2017.
[104] Z. Chen, O. Baklenov, E. T. Kim, I. Mukhametzhanov, J. Tie, A. Madhukar, Z.
Ye and J. C. Campbell, "Normal incidence InAs/AlxGa1-xAs quantum dot
infrared photodetectors with undoped active region," Journal of Applied
Physics, vol. 89, pp. 4558-4563, 2001.
[105] R. A. Miller, H. So, H. C. Chiamori, A. J. Suria, C. A. Chapin and D. G. Senesky,
"A microfabricated sun sensor using GaN-on-sapphire ultraviolet photodetector
arrays," Review of Scientific Instruments, vol. 87, p. 095003, 2016.
[106] N. Mishra, J. Boeckl, N. Motta and F. Iacopi, "Graphene growth on silicon
carbide: A review," Physica Status Solidi A, vol. 213, pp. 2277-2289, 2016.
[107] K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V.
Dubonos, I. V. Grigorieva and A. A. Firsov, "Electric field effect in atomically
thin carbon films," Science, vol. 306, pp. 666-669, 2004.
BIBLIOGRAPHY 162
[108] Y. An, A. Behnam, E. Pop and A. Ural, "Metal-semiconductor-metal
photodetectors based on graphene/p-type silicon Schottky junctions," Applied
Physics Letters, vol. 102, p. 013110, 2013.
[109] X. Li, Y. Zhu, W. Cai, M. Borysiak, B. Han, D. Chen, R. D. Piner, L. Colombo
and R. S. Ruoff, "Transfer of large-area graphene films for high-performance
transparent conductive electrodes," Nano Letters, vol. 9, pp. 4359-4363, 2009.
[110] X. Liang, B. A. Sperling, I. Calizo, G. Cheng, C. A. Hacker, Q. Zhang, Y. Obeng,
K. Yan, H. Peng, Q. Li, X. Zhu, H. Yuan, A. R. H. Walker, Z. Liu, L.-m. Peng
and C. A. Richter, "Toward clean and crackless transfer of graphene," ACS
Nano, vol. 5, pp. 9144-9153, 2011.
[111] A. V. Babichev, S. A. Rykov, M. Tchernycheva, A. N. Smirnov, V. Y. Davydov,
Y. A. Kumzerov and V. Y. Butko, "Influence of substrate microstructure on the
transport properties of CVD-graphene," ACS Applied Materials and Interfaces,
vol. 8, pp. 240-246, 2016.
[112] G. B. Barin, Y. Song, I. d. F. Gimenez, A. G. S. Filho, L. S. Barreto and J. Kong,
"Optimized graphene transfer: Influence of polymethylmethacrylate (PMMA)
layer concentration and baking time on graphene final performance," Carbon,
vol. 84, pp. 82-90, 2015.
[113] J. H. Simmons and K. S. Potter, Optical materials, San Diego: Academic Press,
2000.
[114] A. A. Ogwu, E. Bouquerel, O. Ademosu, S. Moh, E. Crossan and F. Placido,
"The influence of rf power and oxygen flow rate during deposition on the optical
transmittance of copper oxide thin films prepared by reactive magnetron
sputtering," Journal of Physics D: Applied Physics, vol. 38, pp. 266-271, 2005.
BIBLIOGRAPHY 163
[115] H. Morkoc, S. Strite, G. B. Gao, M. E. Lin, B. Sverdlov and M. Burns, "Large-
band-gap SiC, III-V nitride, and II-VI ZnSe-based semiconductor device
technologies," Journal of Applied Physics, vol. 76, pp. 1363-1398, 1994.
[116] S. Strite and H. Morkoc, "GaN, AlN, and InN: A review," Journal of Vacuum
Science and Technology B, vol. 10, pp. 1237-1266, 1992.
[117] R. A. Miller, H. So, H. C. Chiamori, K. M. Dowling, Y. Wang and D. G.
Senesky, "Graphene-enhanced gallium nitride ultraviolet photodetectors under
2 MeV proton irradiation," Applied Physics Letters, vol. 111, p. 241902, 2017.
[118] R. A. Miller, B. A. Cruden, R. Martinez and D. G. Senesky, "In situ ultraviolet
shock radiance measurements using GaN-on-sapphire photodetectors," Review
of Scientific Instruments, vol. 88, p. 115004, 2017.
[119] C. O. Chui, A. K. Okyay and K. C. Saraswat, "Effective dark current suppression
with asymmetric MSM photodetectors in group IV semiconductors," IEEE
Photonics Technology Letters, vol. 15, pp. 1585-1587, 2003.
[120] L. B. Flannery, I. Harrison, D. E. Lacklison, R. I. Dykeman, T. S. Cheng and C.
T. Foxon, "Fabrication and characterisation of p-type GaN metal-
semiconductor-metal ultraviolet photoconductors grown by MBE," Materials
Science and Engineering B, vol. 50, pp. 307-310, 1997.
[121] E. Munoz, E. Monroy, J. L. Pau, F. Calle, F. Omnes and P. Gibart, "III nitrides
and UV detection," Journal of Physics: Condensed Matter, vol. 13, pp. 7115-
7137, 2001.
[122] F. Binet, J. Y. Duboz, E. Rosencher, F. Scholz and V. Harle, "Electric field
effects on excitons in gallium nitride," Physics Review B, vol. 54, pp. 8116-8121,
1996.
[123] B. Shen, K. Yang, L. Zang, Z.-z. Chen, Y.-g. Zhou, P. Chen, R. Zhang, Z.-c.
Huang, H.-s. Zhou and Y.-d. Zheng, "Study of photocurrent properties of GaN
BIBLIOGRAPHY 164
ultraviolet photoconductor grown on 6H-SiC substrate," Japanese Journal of
Applied Physics, vol. 38, pp. 767-769, 1999.
[124] K. S. Stevens, M. Kinniburgh and R. Beresford, "Photoconductive ultraviolet
sensor using Mg-doped GaN on Si(111)," Applied Physics Letters, vol. 66, pp.
3518-3520, 1995.
[125] B. W. Lim, Q. C. Chen, J. Y. Yang and M. Asif Khan, "High responsivity
intrinsic photoconductors based on AlxGa1-xN," Applied Physics Letters, vol.
68, pp. 3761-3762, 1996.
[126] M. Asif Khan, J. N. Kuznia, D. T. Olson, J. M. Van Hove, M. Blasingame and
L. F. Reitz, "High-responsivity photoconductive ultraviolet sensors based on
insulating single-crystal GaN epilayers," Applied Physics Letters, vol. 1992, pp.
2917-2919, 1992.
[127] M. Hou, H. So, A. J. Suria, A. S. Yalamarthy and D. G. Senesky, "Suppression
of persistent photoconductivity in AlGaN/GaN ultraviolet photodetectors using
in-situ heating," IEEE Electron Device Letters, vol. 38, pp. 56-59, 2017.
[128] S. M. Khanna, J. Webb, H. Tang, A. J. Houdayer and C. Carlone, "2 MeV proton
radiation damage studies of gallium nitride films through low temperature
photoluminescence spectroscopy measurements," IEEE Transactions on
Nuclear Science, vol. 47, pp. 2322-2328, 2000.
[129] S. Mathew, T. K. Chan, D. Zhan, K. Gopinadhan, A. R. Barman, M. B. H.
Breese, S. Dhar, Z. X. Shen, T. Venkatesan and J. T. L. Thong, "Mega-electron-
volt proton irradiation on supported and suspended graphene: A Raman
spectroscopic layer dependent study," Journal of Applied Physics, vol. 110, p.
084309, 2011.
BIBLIOGRAPHY 165
[130] G. Yang, B.-J. Kim, K. Kim, J. W. Han and J. Kim, "Energy and dose
dependence of proton-irradiation damage in graphene," RSC Advances, vol. 5,
pp. 31861-31865, 2015.
[131] Y. Hao, Y. Wang, L. Wang, Z. Ni, Z. Wang, R. Wang, C. K. Koo, Z. Shen and
J. T. L. Thong, "Probing layer number and stacking order of few-layer graphene
by Raman spectroscopy," Small, vol. 6, pp. 195-200, 2010.
[132] C. C. Liebe, N. Murphy, L. Dorsky and N. Udomkesmalee, "Three-axis sun
sensor for attitude determination," IEEE Aerospace and Electronics Systems
Magazine, vol. 31, pp. 6-11, 2016.
[133] W. J. Larson and J. R. Wertz, Space mission analysis and design, 3rd Ed., El
Segundo, California: Microcosm Press and Kluwer Academic Publishers, 1999.
[134] "Spacecraft sun sensors," National Aeronautics and Space Administration, SP-
8047, 1970.
[135] T. C. Clarke and E. L. Divita, "Evaluation of radiation interference in the
voyager sun sensor's cadmium sulfide detector," IEEE Transactions on Nuclear
Science, vol. 25, pp. 1324-1328, 1978.
[136] M. H. Jahanshahi, "Simultaneous calibrations of voyager celestial and inertial
attitude control systems in flight," IEEE Transactions on Aerospace and
Electronic Systems, vol. 18, pp. 21-28, 1982.
[137] R. L. Heacock, "The voyager spacecraft," Institution of Mechanical Engineers,
vol. 194, pp. 211-224, 1980.
[138] C. C. Liebe and S. Mobasser, "MEMS based sun sensor," in IEEE Aerospace
Conference, Big Sky, Montana, 2001.
[139] C. C. Liebe, "Solar compass chip," IEEE Sensors Journal, vol. 4, pp. 779-786,
2004.
BIBLIOGRAPHY 166
[140] S. Mobasser and C. C. Liebe, "Micro sun sensor for spacecraft attitude control,"
in IFAC Automatic Control in Aerospace, Saint-Petersburg, Russia, 2004.
[141] S. Mobasser, C. C. Liebe, J. Naegle and L. Choonsup, "Flight qualified micro
sun sensor for mars applications," in 2nd International Conference on Recent
Advances in Space Technologies, Istanbul, Turkey, 2005.
[142] N. Xie and A. J. P. Theuwissen, "Low-power high-accuracy micro-digital sun
sensor by means of a CMOS image sensor," Journal of Electronic Imaging, vol.
22, p. 033030, 2013.
[143] G. Rufino and M. Grassi, "Multi-aperture CMOS sun sensor for microsatellite
attitude determination," Sensors, vol. 9, pp. 4503-4524, 2009.
[144] Q.-y. Fan, J.-w. Peng and X.-y. Gao, "Micro digital sun sensor with linear
detector," Review of Scientific Instruments, vol. 87, p. 075003, 2016.
[145] M. A. Post, J. Li and R. Lee, "A low-cost photodiode sun sensor for cubesat and
planetary microrover," International Journal of Aerospace Engineering, vol.
2013, p. 549080, 2013.
[146] P. Ortega, G. Lopez-Rodriguez, J. Ricart, M. Dominguez, L. M. Castaner, J. M.
Quero, C. L. Tarrida, J. Garcia, M. Reina, A. Gras and M. Angulo, "A
miniaturized two axis sun sensor for attitude control of nano-satellites," IEEE
Sensors Journal, vol. 10, pp. 1623-1632, 2010.
[147] M. Zahran and M. Aly, "A solar cell based coarse sun sensor for a small LEO
satellite attitude determination," Journal of Power Electronics, vol. 9, pp. 631-
642, 2009.
[148] J. Barnes, C. Liu and K. Ariyur, "A hemispherical sun sensor for orientation and
geolocation," IEEE Sensors Journal, vol. 14, pp. 4423-4433, 2014.
BIBLIOGRAPHY 167
[149] L. Schumacher, "Mariner Mars 1971 sun sensor model development and
simulation, TM 33-610," Jet Propulsion Laboratory, National Aeronautics and
Space Administration, Pasadena, California, 1973.
[150] Y. Yoshihara, H. Hashimoto, T. Yamamoto, H. Saito, E. Hirokawa and M. Mita,
"Micro sun sensor with CMOS imager for small satellite attitude control," in
19th Annual AIAA/USU Conference on Small Satellites, Logan, Utah, 2005.
[151] Z. You, J. Sun, F. Xing and G.-F. Zhang, "A novel multi-aperture based sun
sensor based on a fast multi-point MEANSHIFT (FMMS) algorithm," Sensors,
vol. 11, pp. 2857-2874, 2011.
[152] F. J. Delgado, J. M. Quero, J. Garcia, C. L. Tarrida, P. R. Ortega and S. Bermejo,
"Accurate and wide-field-of-view MEMS-based sun sensor for industrial
applications," IEEE Transactions on Industrial Electronics, vol. 59, pp. 4871-
4880, 2012.
[153] R. D. Knight, Physics for scientists and engineers 2nd ed., San Francisco:
Pearson Addison-Wesley, 2008.
[154] H. Hung, S. J. Chang, Y. C. Lin, H. Kuan and R. M. Lin, "AlGaN MSM
photodetectors with recess-etched LT-AlGaN cap layers," IET Optoelectronics,
vol. 1, pp. 147-149, 2007.
[155] C. L. Yu, C. H. Chen, S. J. Chang and P. C. Chang, "GaN metal-semiconductor-
metal ultraviolet photodetectors with Ir/Pt contact electrodes," Journal of the
Electrochemical Society, vol. 153, pp. J71-J72, 2007.
[156] C. K. Wang, S. J. Chang, Y. K. Su, C. S. Chang, Y. Z. Chiou, C. H. Kuo, T. K.
Lin, T. K. Ko and J. J. Tang, "GaN MSM photodetectors with TiW transparent
electrodes," Materials Science and Engineering B, vol. 112, pp. 25-29, 2004.
[157] B. L. Diffey, "Sources and measurement of ultraviolet radiation," Methods, vol.
28, pp. 4-13, 2002.
BIBLIOGRAPHY 168
[158] B. A. Woollard, R. D. Braun and D. Bose, "Aerothermodynamic and thermal
protection system instrumentation reference guide," in IEEE Aerospace
Conference, Big Sky, Montana, 2016.
[159] E. R. Martinez, C. T. Weber, T. Oishi, J. Santos and J. Mach, "Development of
a sheathed miniature aerothermal reentry thermocouple for thermal protection
system materials," in 42nd AIAA Thermophysics Conference, Honolulu, Hawaii,
2011.
[160] M. J. Gazarik, M. J. Wright, A. Little, F. M. Cheatwood, J. A. Herath, M. M.
Munk, F. J. Novak and E. R. Martinez, "Overview of the MEDLI project," in
IEEE Aerospace Conference, Big Sky, Montana, 2008.
[161] D. Bose, J. A. Santos, E. Rodriguez, T. White, M. Olson and M. Mahzari, "Mars
science laboratory heat shield instrumentation and arc jet characterization," in
44th AIAA Thermophysics Conference, San Diego, California, 2013.
[162] A. Little, D. Bose, C. Karlgaard, M. Munk, C. Kuhl, M. Schoenenberger, C.
Antill, R. Verhappen, P. Kutty and T. White, "The mars science laboratory
(MSL) entry, descent and landing instrumentation (MEDLI): hardware
performance and data reconstruction," in 36th AAS Guidance and Control
Conference, Breckenridge, Colorado, 2013.
[163] B. A. Cruden, D. Prabhu and R. Martinez, "Absolute radiation measurements in
venus and mars entry conditions," Journal of Spacecraft and Rockets, vol. 49,
pp. 1069-1079, 2012.
[164] J. Olejniczak, M. Wright, D. Prabhu, N. Takashima, B. Hollis, E. V. Zoby and
K. Sutton, "An analysis of the radiative heating environment for aerocapture at
titan," in 39th AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit,
Huntsville, Alabama, 2003.
BIBLIOGRAPHY 169
[165] A. M. Brandis and B. A. Cruden, "Titan atmospheric entry radiative heating," in
47th AIAA Thermophysics Conference, AIAA Aviation Forum, Washington,
D.C., 2016.
[166] R. C. Ried Jr., W. C. Rochelle and J. D. Milhoan, "Radiative heating to the
Apollo command module: engineering prediction and flight measurement,"
NASA TM-X-58091, Houston, Texas, 1972.
[167] A. Seiff, D. E. Reese, S. C. Sommer, D. B. Kirk, E. E. Whiting and H. B.
Niemann, "PAET, an entry probe experiment in the earth's atmosphere," Icarus,
vol. 18, pp. 525-563, 1973.
[168] D. L. Cauchon, "Radiative heating results from the FIRE II flight experiment at
a reentry velocity of 11.4 kilometers per second," NASA TM-X-1402, Hampton,
Virginia, 1967.
[169] A. Gulhan, T. Thiele, F. Siebe, T. Schleutker, R. Kronen, J. Annaloro, P. Omaly
and P. J. Hebert, "Achievements of the COMARS+ instrumentation package
during the entry flight phase of the exomars schiaparelli lander," in 14th
International Planetary Probe Workshop, The Hague, The Netherlands, 2017.
[170] H. H. Hwang, D. Bose, T. R. White, H. S. Wright, M. Schoenenberger, C. A.
Kuhl, D. Trombetta, J. A. Santos, T. Oishi, C. D. Karlgaard, M. Mahzari and S.
P. Pennington, "Mars 2020 entry, descent and landing instrumentation 2
(MEDLI2)," in 46th AIAA Thermophysics Conference, Washington, D. C., 2016.
[171] I. Muller-Wodarg, C. A. Griffith, E. Lellouch and T. E. Cravens, Titan: Interior,
surface, atmosphere, and space environment, New York: Cambridge University
Press, 2014.
[172] B. A. Cruden, R. Martinez, J. H. Grinstead and J. Olejniczak, "Simultaneous
vacuum ultraviolet through near IR absolute radiation measurement with
BIBLIOGRAPHY 170
spatiotemporal resolution in an electric arc shock tube," in 41st AIAA
Thermophysics Conference, San Antonio, Texas, 2009.
[173] S. Gordon and B. J. McBride, "Computer program for calculation of complex
chemical equilibrium compositions and applications," National Aeronautics and
Space Administration RP 1311, 1994.
[174] A. J. Suria, H. C. Chiamori, A. Shankar and D. G. Senesky, "Capacitance-
voltage characteristics of gamma irradiated Al2O3, HfO2, and SiO2 thin films
grown by plasma-enhanced atomic layer deposition," in Proceeding of SPIE,
Sensors for Extreme Harsh Environments II, Baltimore, Maryland, 2015.
[175] V. Lowalekar and S. Raghavan, "Etching of zirconium oxide, hafnium oxide,
and hafnium silicates in dilute hydrofluoric acid solutions," Journal of Materials
Research, vol. 19, pp. 1149-1156, 2004.
[176] D. Shamiryan, M. Baklanov, M. Claes, W. Boullart and V. Paraschiv, "Selective
removal of high-k gate dielectrics," Chemical Engineering Communications,
vol. 196, pp. 1475-1535, 2009.
[177] A. J. Suria, A. S. Yalamarthy, T. A. Heuser, A. Bruefach, C. A. Chapin, H. So
and D. G. Senesky, "Thickness engineering of atomic layer deposited Al2O3
films to suppress interfacial reaction and diffusion of Ni/Au gate metal in
AlGaN/GaN HEMTs up to 600C in air," Applied Physics Letters, vol. 110, p.
253505, 2017.
[178] N. Miura, T. Nanjo, M. Suita, T. Oishi, Y. Abe, T. Ozeki, H. Ishikawa, T. Egawa
and T. Jimbo, "Thermal annealing effects on Ni/Au based Schottky contacts on
n-GaN and AlGaN/GaN with insertion of high work function metal," Solid-State
Electronics, vol. 48, pp. 689-695, 2004.
BIBLIOGRAPHY 171
[179] K. J. Duxstad, E. E. Haller, K. M. Yu, M. T. Hirsch, W. R. Imler, D. A.
Steigerwald, F. A. Ponce and L. T. Romano, "Thermal stability of Pt, Pd, and Ni
on GaN," in Materials Research Society Conference Proceedings, 1996.
[180] J. D. Guo, F. M. Pan, M. S. Feng, R. J. Guo, P. F. Chou and C. Y. Chang,
"Schottky contact and the thermal stability of Ni on n-type GaN," Journal of
Applied Physics, vol. 80, pp. 1623-1627, 1996.
[181] W. Yang, T. Nohava, S. Krishnankutty, R. Torreano, S. McPherson and H.
Marsh, "High gain GaN/AlGaN heterojunction phototransistor," Applied
Physics Letters, vol. 73, pp. 978-980, 1998.
[182] Y. Huang, D. J. Chen, H. Lu, H. B. Shi, P. Han, R. Zhang and Y. D. Zheng,
"Photocurrent characteristics of two-dimensional-electron-gas-based
AlGaN/GaN metal-semiconductor-metal photodetectors," Applied Physics
Letters, vol. 96, p. 243503, 2010.
[183] Z. H. Zaidi and P. A. Houston, "Highly sensitive UV detection mechanism in
AlGaN/GaN HEMTs," IEEE Transactions on Electron Devices, vol. 60, pp.
2776-2781, 2013.
[184] V. V. Afanas'ev and A. Stesmans, "Leakage currents induced in ultrathin oxides
on (100)Si by deep-UV photons," Materials Science and Engineering B, vol. 71,
pp. 56-61, 2000.
[185] R. Gaska, M. Gaevski, R. Jain, J. Deng, M. Islam, G. Simin and M. Shur, "Novel
AlInN/GaN integrated circuits operating up to 500C," Solid-State Electronics,
vol. 113, pp. 22-27, 2015.
[186] H. Wang, A. M. H. Kwan, Q. Jiang and K. J. Chen, "A GaN pulse width
modulation integrated circuit for GaN power converters," IEEE Transactions on
Electon Devices, vol. 62, pp. 1143-1149, 2015.
BIBLIOGRAPHY 172
[187] T. K. Ko, S. C. Shei, S. J. Chang, Y. K. Su, Y. Z. Chiou, Y. C. Lin, C. S. Chang,
C. S. Chen, C. K. Wang, J. K. Sheu and W. C. Lai, "Flip-chip p(GaN)-i(GaN)-
n(AlGaN) narrowband UV-A photosensors," IEEE Sensors Journal, vol. 6, pp.
964-969, 2006.
[188] H. Jia, Y. Xu, Y. Kang and X. Li, "Effects of visible light on a readout integrated
circuit for GaN-based UV focal plane array," Semiconductor Science and
Technology, vol. 21, pp. 112-115, 2006.
[189] O. Ambacher, B. Foutz, J. Smart, J. R. Shealy, N. G. Weimann, K. Chu, M.
Murphy, A. J. Sierakowski, W. J. Schaff and L. F. Eastman, "Two dimensional
electron gases induced by spontaneous and piezoelectric polarization in undoped
and doped AlGaN/GaN heterostructures," Journal of Applied Physics, vol. 87,
p. 334, 2000.
[190] M. Stutzmann, G. Steinhoff, M. Eickhoff, O. Ambacher, C. E. Nebel, J.
Schalwig, R. Neuberger and G. Muller, "GaN-based heterostructures for sensor
applications," Diamond and Related Materials, vol. 11, pp. 886-891, 2002.
[191] T. Narita, A. Wakejima and T. Egawa, "Ultraviolet photodetectors using
transparent gate AlGaN/GaN high electron mobility transistor on silicon
substrate," Japanese Journal of Applied Physics, vol. 52, p. 01AG06, 2013.
[192] A. Endoh, I. Watanabe, Y. Yamashita, T. Mimura and T. Matsui, "Effect of
temperature on cryogenic characteristics of AlGaN/GaN MIS-HEMTs," Physica
Status Solidi C, vol. 6, pp. S964-D967, 2009.
[193] R. A. Miller, C. A. Chapin, K. M. Dowling, R. Chen, A. J. Suria and D. G.
Senesky, "Low-temperature operation of gallium nitride based ultraviolet
photodetectors," in AIAA SPACE Forum, Long Beach, California, 2016.
[194] J. H. Simmons and K. S. Potter, in Optical Materials, San Diego, Academic
Press, 2000, pp. 165-168.
BIBLIOGRAPHY 173
[195] E. T. Yu and M. O. Manasreh, in III-V nitride semiconductors: applications and
devices, New York, Taylor & Francis Books, Inc., 2003, pp. 95-100.
[196] M. Shur, B. Gelmont and M. Asif Khan, "Electron mobility in two-dimensional
electron gas in AlGaN/GaN heterostructures and in bulk GaN," Journal of
Electronic Materials, vol. 25, pp. 777-785, 1996.
[197] D. Qiao, L. S. Yu, S. S. Lau, J. M. Redwing, J. Y. Lin and H. X. Jiang,
"Dependence of Ni/AlGaN Schottky barrier height on Al mole fraction," Journal
of Applied Physics, vol. 87, pp. 801-804, 2000.
[198] S. Chang, M. Chang and Y. Yang, "Enhanced responsivity of GaN metal-
semiconductor-metal (MSM) photodetectors on GaN substrate," IEEE
Photonics Journal, vol. 9, p. 6801707, 2017.
[199] F. Xie, H. Lu, D. Chen, X. Ji, F. Yan, R. Zhang, Y. Zheng, L. Li and J. Zhou,
"Ultra-low dark current AlGaN-based solar-blind metal-semiconductor-metal
photodetectors for high-temperature applications," IEEE Sensors Journal, vol.
12, pp. 2086-2090, 2012.
[200] W.-C. Lien, D.-S. Tsai, S.-H. Chiu, D. G. Senesky, R. Maboudian, A. P. Pisano
and J.-H. He, "Low-temperature, ion beam-assisted SiC thin films with
antireflective ZnO nanorod arrays for high temperature photodetection," IEEE
Electron Device Letters, vol. 32, pp. 1564-1566, 2011.
[201] W. T. Hsieh, Y. K. Fang, K. H. Wu, W. J. Lee, J. J. Ho and C. W. Ho, "Using
porous silicon as semi-insulating substrate for b-SiC high temperature optical-
sensing devices," IEEE Transactions on Electron Devices, vol. 48, pp. 801-803,
2001.
[202] L. W. Sang, M. Y. Liao, Y. Koide and M. Sumiya, "InGaN photodiodes using
CaF2 insulator for high-temperature UV detection," Physica Status Solidi C, vol.
9, pp. 953-956, 2012.
BIBLIOGRAPHY 174
[203] M. Mazzillo, A. Sciuto and S. Marchese, "Impact of the epilayer doping on the
performance of thin metal film Ni2Si/4H-SiC schottky photodiodes," Journal of
Instrumentation, vol. 9, p. 12001, 2014.
[204] Y. Xu, D. Zhou, H. Lu, D. Chen, F. Ren, R. Zhang and Y. Zheng, "High-
temperature and reliability performance of 4H-SiC schottky-barrier photodiodes
for UV detection," Journal of Vacuum Science and Technology B, vol. 33, p.
040602, 2015.
[205] T. V. Blank, Y. A. Goldberg, E. V. Kalinina, O. V. Konstantinov, A. O.
Konstantinov and A. Hallen, "Temperature dependence of the photoelectric
conversion quantum efficiency of 4H-SiC schottky UV photodetectors,"
Semiconductor Science and Technology, vol. 20, pp. 710-715, 2005.
[206] N. Watanabe, T. Kimoto and J. Suda, "4H-SiC pn photodiodes with temperature-
independent photoresponse up to 300Cq," Applied Physics Express, vol. 5, p.
094101, 2012.
[207] Y. Ueda, S. Akita, Y. Nomura, Y. Nakayama and H. Naito, "Study of high
temperature photocurrent properties of 6H-SiC UV sensor," Thin Solid Films,
vol. 517, pp. 1471-1473, 2008.
[208] X. Bai, X. Guo, D. C. Mcintosh, H.-D. Liu and J. C. Campbell, "High detection
sensitivity of ultraviolet 4H-SiC avalanche photodiodes," IEEE Journal of
Quantum Electronics, vol. 43, pp. 1159-1162, 2007.
[209] S. P. Chang, S. J. Chang, Y. Z. Chiou, C. Y. Lu, T. K. Lin, Y. C. Lin, C. F. Kuo
and H. M. Chang, "ZnO photoconductive sensor epitaxially grown on sapphire
substrates," Sensors and Actuators A, vol. 140, pp. 60-64, 2007.
[210] S. P. Chang, R. W. Chuang, S. J. Chang, C. Y. Lu, Y. Z. Chiou and S. F. Hsieh,
"Surface HCl treatment in ZnO photoconductive sensors," Thin Solid Films, vol.
517, pp. 5050-5053, 2009.
BIBLIOGRAPHY 175
[211] S. J. Young, L. W. Ji, S. J. Chang and Y. K. Su, "ZnO metal-semiconductor-
metal ultraviolet sensor with various contact electrodes," Journal of Crystal
Growth, vol. 293, pp. 43-47, 2006.
[212] S. J. Young, L. W. Ji, T. H. Fang, S. J. Chang, Y. K. Su and X. L. Du, "ZnO
ultraviolet photodiodes with Pd contact electrodes," Acta Materialia, vol. 55, pp.
329-333, 2007.
[213] S. Nikishin, B. Borisov, M. Pandikunta, R. Dahal, J. K. Lin, H. X. Jiang, H.
Harris and M. Holtz, "High quality AlN for deep UV photodetectors," Applied
Physics Letters, vol. 95, p. 054101, 2009.
[214] R. W. Chuang, S. P. Chang, S. J. Chang, Y. Z. Chiou, C. Y. Lu, T. K. Lin, Y. C.
Lin, C. F. Kuo and H. M. Chang, "Gallium nitride metal-semiconductor-metal
photodetectors prepared on silicon substrates," Journal of Applied Physics, vol.
102, p. 073110, 2007.
[215] Y.-K. Su, F.-S. Juang and M.-H. Chen, "GaN metal-semiconductor-metal
visible-blind photodetectors with transparent indium-tin-oxide contact
electrodes," Japanese Journal of Applied Physics, vol. 42, pp. 2257-2259, 2003.
[216] L. Ravikiran, K. Radhakrishnan, N. Dharmarasu, M. Agrawal, Z. Wang, A.
Bruno, C. Soci, T. Lihuang and K. S. Ang, "GaN Schottky metal-semiconductor-
metal UV photodetectors on Si(111) grown by ammonia-MBE," IEEE Sensors
Journal, vol. 17, pp. 72-77, 2017.
[217] F.-R. Juang, Y.-K. Fang, Y.-T. Chiang, T.-C. Wei and B.-W. Lin, "The poly-SiC
MSM photodiode on ceramic SiC substrate for low-cost ultraviolet light sensing
applications," IEEE Sensors Journal, vol. 11, pp. 3446-3450, 2011.
[218] W. Yang, F. Zhang, Z. Liu and Z. Wu, "Effects of annealing on the performance
of 4H-SiC metal-semiconductor-metal ultraviolet photodetectors," Materials
Science in Semiconductor Processing, vol. 11, pp. 59-62, 2008.
BIBLIOGRAPHY 176
[219] F. Zhang, W. Yang, H. Huang, X. Chen, Z. Wu, H. Qi, J. Yao, Z. Fan and J.
Shao, "High-performance 4H-SiC based metal-semiconductor-metal ultraviolet
photodetectors with Al2O3/SiO2 films," Applied Physics Letters, vol. 92, p.
251102, 2008.
[220] L. Ostlung, Q. Wang, R. Esteve, S. Almqvist, D. Rihtnesberg, S. Reshanov, A.
Z. Z. Zhang, J.-K. Lim, M. Bakowski, A. Schoner and W. Kaplan, "4H- and 6H-
SiC UV photodetectors," Physica Status Solidi C, vol. 9, pp. 1680-1682, 2012.
[221] W.-C. Lien, A. P. Pisano, D.-S. Tsai, J.-H. He and D. G. Senesky, "Extreme
temperature 4H-SiC metal-semiconductor-metal ultraviolet photodetectors," in
European Solid-State Device Research Conference, Bordeaux, France, 2012.
[222] S. J. Young, "Ir/n-ZnO Schottky barrier ultraviolet photodiodes," IEEE Sensors
Journal, vol. 11, pp. 1129-1133, 2011.
[223] R. Dahal, J. Li, Z. Y. Fan, M. L. Nakarmi, T. M. Al Tahtamouni, J. Y. Lin and
H. Z. Jiang, "AlN MSM and Schottky photodetectors," Physica Status Solidi C,
vol. 5, pp. 2148-2151, 2008.
[224] H. Endo, M. Sugibuchi, K. Takahashi, S. Goto, S. Sugimura, K. Hane and Y.
Kashiwaba, "Schottky ultraviolet photodiode using a ZnO hydrothermally
grown single crystal substrate," Applied Physics Letters, vol. 90, p. 121906,
2007.
[225] D. G. Zhao, D. S. Jiang, J. J. Zhu, Z. S. Liu, S. M. Zhang, J. W. Liang, H. Yang,
X. Li, X. Y. Li and H. M. Gong, "Influence of defects in n-GaN layer on the
responsivity of Schottky barrier ultraviolet photodetectors," Applied Physics
Letters, vol. 90, p. 062106, 2007.
[226] D. T. Thanh, J. Jin, K. B. Ko, B. D. Ryu, M. Han, T. V. Cuong and C.-H. Hong,
"Hexagonal boron nitride pattern embedded in AlN template layer for visible-
BIBLIOGRAPHY 177
blind ultraviolet photodetectors," Optical Materials Express, vol. 7, pp. 1463-
1472, 2017.
[227] E. Monroy, E. Munoz, F. J. Sanchez, F. Calle, E. Calleja, B. Beaumont, P.
Gibart, J. A. Munoz and F. Cusso, "High-performance GaN p-n junction
photodetectors for solar ultraviolet applications," Semiconductor Science and
Technology, vol. 13, pp. 1042-1046, 1998.
[228] J. D. Hwang, D. H. Wu and S. B. Hwang, "Inserting an i-ZnO layer to increase
the performance of p-Si/n-ZnO heterojunction photodetectors," Materials
Science in Semiconductor Processing, vol. 39, pp. 132-135, 2015.
[229] J. M. Van Hove, R. Hickman, J. J. Klaassen, P. P. Chow and P. P. Ruden,
"Ultraviolet-sensitive, visible-blind GaN photodiodes fabricated by molecular
beam epitaxy," Applied Physics Letters, vol. 70, pp. 2282-2284, 1997.
[230] J. C. Lin, Y. K. Su, S. J. Chang, W. H. Lan, W. R. Chen, K. C. Huang, Y. C.
Cheng and W. J. Lin, "Low dark current GaN p-i-n photodetectors with a low-
temperature AlN interlayer," IEEE Photonics Technology Letters, vol. 20, pp.
1255-1257, 2008.
[231] X. Chen, W. Yang and Z. Wu, "Visible blind p-i-n ultraviolet photodetector
fabricated on 4H-SiC," Microelectronic Engineering, vol. 83, pp. 104-106,
2006.
[232] X. Chen, H. Zhu, J. Cai and Z. Wu, "High-performance 4H-SiC based ultraviolet
p-i-n photodetector," Journal of Applied Physics, vol. 102, p. 024505, 2007.
[233] J. Cai, X. Chen, R. Hong, W. Yang and Z. Wu, "High-performance 4H-SiC
based p-i-n photodiode and investigation of its capacitance characteristics,"
Optics Communications, vol. 333, pp. 182-186, 2014.
BIBLIOGRAPHY 178
[234] F. Sun, C.-X. Shan, S.-P. Wang, B.-H. Li, Z.-Z. Zhang, C.-L. Yang and D.-Z.
Shen, "Ultraviolet photodetectors fabricated from ZnO p-i-n homojunction
structures," Materials Chemistry and Physics, vol. 129, pp. 27-29, 2011.
[235] T. Tut, M. Gokkavas, A. Inal and E. Ozbay, "AlGaN-based avalanche
photodiode with high reproducible avalanche gain," Applied Physics Letters,
vol. 90, p. 163506, 2007.
[236] H. Zhu, X. Chen, J. Cai and Z. Wu, "4H-SiC ultraviolet avalanche
photodetectors with low breakdown voltage and high gain," Solid-State
Electronics, vol. 53, pp. 7-10, 2009.
[237] H.-D. Liu, D. McIntosh, X. Bai, H. Pan, M. Liu and J. C. Campbell, "4H-SiC
PIN recessed window avalanche photodiode," in 20th annual meeting of the
IEEE Lasers and Electro-Optics Society, Lake Buena Vista, Florida, 2007.
[238] D. Zhou, F. Liu, H. Lu, D. Chen, F. Ren, R. Zhang and Y. Zheng, "High-
temperature single photon detection performance of 4H-SiC avalanche
photodiodes," IEEE Photonics Technology Letters, vol. 26, pp. 1136-1138,
2014.