Embed Size (px)
Citation preview
저 시-비 리-동 조건 경허락 2.0 한민
는 아래 조건 르는 경 에 한하여 게
l 저 물 복제, 포, 전송, 전시, 공연 송할 수 습니다.
l 차적 저 물 성할 수 습니다.
다 과 같 조건 라야 합니다:
l 하는, 저 물 나 포 경 , 저 물에 적 허락조건 확하게 나타내어야 합니다.
l 저 터 허가를 러한 조건들 적 지 않습니다.
저 에 른 리는 내 에 하여 향 지 않습니다.
것 허락규약(Legal Code) 해하 쉽게 약한 것 니다.
Disclaimer
저 시. 하는 원저 를 시하여야 합니다.
비 리. 하는 저 물 리 적 할 수 없습니다.
동 조건 경허락. 하가 저 물 개 , 형 또는 가공했 경에는, 저 물과 동 한 허락조건하에서만 포할 수 습니다.
공학 사 학 논문
Fabrication of Rectifying Optical Rectenna Array
for
Sensing and Light Energy Harvesting
에너지 검출 한
학 수신기 현
2013 년 2 월
울대학 대학원
재료공학부
강 민
Fabrication of Rectifying Optical Rectenna Array
for
Sensing and Light Energy Harvesting
A DISSERTATION SUMITTED TO
DEPARTMENT OF MATERIALS SCIENCE AND ENGINEERING
SEOUL NATIONAL UNIVERSITY
FOR THE DEGREE OF
MASTER OF SCIENCE
Min Yi Kang
February 2013
i
Abstract
Since the advent of microwave power transmission in the 1960s, rectifying
optical antennas, also referred to as ‘Rectenna’, has been pursued as a means to convert
light into direct current (DC) output for sensing and possibly even light energy
harvesting applications.
Rectenna is a combination of antenna and rectifier and it has been tried to use in
visible-near infrared (near IR) spectrum nowadays. The extension from microwave to
higher frequencies (few hundreds of THz) has remained a challenge, because of
complicated fabrication for smaller feature sizes. In this dissertation, it is introduced that
the initial successes in fabricating rectenna arrays using a special projection e-beam
approach and conventional e-beam lithography. A key element in the designed function
of rectenna device is the built-in rectification with the MIM junction arrays, to convert
the alternating current (AC) in the nano antenna into DC output. For this built in MIM
structure, two different metals with a large work function difference are used: aluminum
and gold. Aluminum oxide of 10nm thickness was used as the insulating layer, formed
by atomic layer deposition (ALD). As it is mentioned previously, the nanowires
patterning process was done by the special projection e-beam lithography method –
atomic image projection e-beam lithography (AIPEL) developed over the years in our
group (nanofabrication laboratory, NFL), at Seoul National University. The
conventional e-beam lithography was also tried to pattern the nanowires. It was
followed by the metal electrodes and metal pads, which were made by conventional e-
beam lithography and photolithography, respectively. Each nanowire serves as an
element antenna of the device, which varies in length from 300nm to 480nm. The target
of frequency for operation is within near IR region. The aim is to complement the
prevailing pn diode and solar cell technologies by providing a much broader spectral
coverage. Since the visible-IR spectrum would reach the surface of Earth more than
other waves, our rectenna device is expected to be more efficient.
A series of measurement was carried out in order to characterize the device
structure and the current voltage dependence, as well as its optical response. The
representative I-V characteristic shows a dramatic increase in current response, by ~3
orders of magnitude, to incident broad-band light (white light from a microscope). With
experimental results, the spectral response to incident light in the near IR range is also
discussed as well as spectral resonance.
Keywords: Rectifying antenna, Nano antenna, Optical rectification, AIPEL, Line
patterning, MIM diode
Student Number: 2011-22859
ii
Table of contents
Abstract ..................................................................................................... i
List of Tables .......................................................................................... iv
List of Figures .......................................................................................... v
CHAPTER 1. Introduction .................................................................... 1
1.1 The energy issue .............................................................................. 2
1.2 Rectenna .......................................................................................... 5
1.2.1 Applications ............................................................................ 10
1.3 Historical background ................................................................... 15
1.4 Literature review ........................................................................... 16
1.4.1 International research trends .................................................... 16
1.4.2 National research trends .......................................................... 19
1.5 References ..................................................................................... 21
CHAPTER 2. Built in MIM structure ................................................. 23
2.1 Introduction ................................................................................... 24
2.2 Basic principles of MIM diode ...................................................... 24
2.2.1 Symmetry and asymmetry ....................................................... 25
2.2.2 Conduction Mechanism ........................................................... 28
2.2.3 J-V theory ................................................................................ 30
2.3 Design for built in MIM structure .................................................. 34
2.3.1 Design rule of antenna ............................................................. 34
2.3.2 Requirements of MIM diode for rectenna array ....................... 39
2.4 References ..................................................................................... 39
iii
CHAPTER 3. Fabrication of built in MIM structure ........................ 40
3.1 Introduction ................................................................................... 41
3.2 Fabrication of Au nanowires .......................................................... 44
3.2.1 AIPEL ..................................................................................... 44
3.2.2 Conventional e-beam lithography ............................................ 47
3.3 Fabrication of rectifying metal electrodes ...................................... 47
3.4 MIM built in structure ................................................................... 50
3.5 References ..................................................................................... 52
CHAPTER 4. Characterization ........................................................... 53
4.1 Introduction ................................................................................... 54
4.2 Experimental details ...................................................................... 54
4.3 Results and Discussions ................................................................. 56
4.3.1 Unilluminated and illuminated MIM structure ......................... 56
4.3.2 Antenna length effect .............................................................. 59
4.4 References ..................................................................................... 62
CHAPTER 5. Summary and Conclusions .......................................... 63
Abstract (in Korean) ............................................................................... 65
Acknowledgement (in Korean) ............................................................... 66
iv
List of Tables
CHAPTER 1
Table 1-1.
International research groups
Table 1-2.
National research groups
CHAPTER 2
Table 2-1.
Antenna length and antenna width for total 60 cells
v
List of Figures
CHAPTER 1
Figure 1-1.
Carbon dioxide emission trends [1]
Figure 1-2.
Best research-cell efficiency [4]
Figure 1-3.
The electromagnetic spectrum [7]
Figure 1-4.
Rectenna circuit: dipole antenna, low-pass filter, rectify diode, output DC pass filter and
resistive load [8]
Figure 1-5.
The electrical circuit diagram for a conventional rectenna
Figure 1-6.
Fuel free airplane: the stationary high altitude replay platform (SHARP) [11]
Figure 1-7.
Microwave charging for an electric vehicle by Hashimoto Lab., Japan
Figure 1-8.
Solar power satellite (SPS); SPS is beamed to a rectenna on Earth. The power beam is
converted into electrical current, which is fed into the power grid [5]
Figure 1-9.
(Left) number of patents and (right) number of journals
vi
CHAPTER 2
Figure 2-1.
(a) TEM image of a symmetric MIM (ZrCuAlNi-Al2O3-ZrCuAlNi) diode. (b)
Equilibrium energy band diagram for a symmetric MIM diode. (c) Measured and
simulated current density-field curves for the MIM diode shown in (a). (d) TEM image
of an asymmetric MIM (ZrCuAlNi-Al2O3-Al) diode. (e) Equilibrium energy band
diagram for an asymmetric MIM diode. (f) Measured and simulated current density-
field curves for the MIM diode shown in (d). [2]
Figure 2-2.
Band diagram of MIM diode when (a) Fermi level is aligned (b) negative bias is applied
(c) positive bias is applied
Figure 2-3.
F-N plots applied on Al-Al2O3 side
Figure 2-4.
F-N plots applied on Al-Al2O3 side
Figure 2-5.
A schematic band diagram of MIM diode with rectangular potential barrier, electrode 1
and electrode 2 for V< /e [4]
Figure 2-6.
A single antenna element represented by a metal rod. Incident light with wavelength
polarizes the ends and gives rise to a standing surface charge wave. [6]
vii
CHAPTER 3
Figure 3-1.
Built in MIM structure (left) 3-D image (right) 2-D cross section
Figure 3-2.
Fabrication of built in MIM structure
Figure 3-3.
(Left) a picture of AIPEL equipment (right) high resolution image of β-Si3N4, which is
used as a mask material for the AIPEL system [2]
Figure 3-4.
SEM images taken after the fabrication of Au nanowires by AIPEL (left) a whole
exposure area with dose 800 C/cm2, at magnification of 1.6K (right) central part of the
whole exposure area with dose 800 C/cm2 at magnification of 5K
Figure 3-5.
SEM images taken after the fabrication of Au nanowires and Au electrode by
conventional e-beam lithography with dose 2500 C/cm2 (left) at magnification of 10K
(right) at magnification of 25K
Figure 3-6.
SEM images of the final state of the device showing (left) a rectangular shape at
magnification of 15K (right) a trapezoidal shape at magnification of 20K
viii
CHAPTER 4
Figure 4-1.
Equipment used for this analysis: (top) Agilent 4156C (middle) KEITHLEY 4200-SCS
(bottom) IPCE
Figure 4-2.
J-V curve for 380nm of antenna length and 40nm of antenna width
Figure 4-3.
I-V curve for 380nm of antenna length and 40nm of antenna width
Figure 4-4.
I-V curves for illuminated device with 80nm of antenna width (left) 480nm of antenna
length (right) 300nm of antenna length
Figure 4-5.
A graph showing current signal of a cell (380nm of antenna length and 40nm of antenna
width) in visible-near IR spectrum
1
CHAPTER 1.
Introduction
2
1.1 The energy issue
Over the last few years, demands for fossil fuels are becoming more and more
concentrated in some countries and the threat of climate change is pushing us towards
reducing greenhouse emissions. The following Figure 1-1 shows the emission of carbon
dioxide (CO2), which has been the main cause of global warming.
Even though global emission of CO2 dropped by one percent in 2009 due to economic
recessions, it is likely to increase in 2010. [2] No surprisingly, trends in global CO2
emissions continued upwards afterwards. According to the annual report from Joint
Research Centre (JRC) and the Netherlands Environmental Assessment Agency (PBL),
the global emissions of carbon dioxide reached an all-time high of almost 34 billion
tonnes in 2011. [3]
Energy demand ideally needs to be satisfied. Therefore, new energy era has
been highlighted. In other words, since energy demand continues to rise every year, the
research on renewable energy has been growing. One of such source of energy is solar
radiation, which is unlimited in supply, is clean and renewable. Traditional p-n junction
solar cells are the most mature of the solar cell energy harvesting technologies, however,
we have seen its efficiency limitation. This is because heat energy loss is huge. Basically,
the crystalline silicon (Si) solar cell uses basic physics of energy absorption and carrier
generation in the material, showing electric properties. A photon needs only have greater
energy than the bandgap in order to excite an electron from the valence band into the
conduction band. Those higher energy photons will be absorbed by the solar cell, but
the difference in energy between these photons and the silicon bandgap (1.12eV) is
converted into heat energy mostly via lattice vibration. It occurs rather than converting
into usable energy, such as electric energy. Refer to Figure 1-2, for a single Si solar cell,
its maximum efficiency is around 33.7% and a present-day record 40.7% of efficiency
was achieve with a triple-junction version of the cell.
Rather than generating single electron-hole pair as in the photovoltaic, using the
incoming electromagnetic (EM) field itself from the Sun is another approach that we
could use to harvest new energy. And the device, which convert EM wave into direct
current (DC) electricity, is called as ‘rectenna’.
3
Figure 1-1 Carbon dioxide emission trends [1]
4
Figure 1-2 Best research-cell efficiency [4]
5
1.2 Rectenna
William C. Brown, who was an American electrical engineer, brought an idea about
microwave power transmission in the 1960s. [5] Following his works, a rectifying
antenna (also called as a ‘Rectenna’), which is a special type of antenna that is used to
directly convert microwave energy into DC electricity [6], has been highlighted as a
good example of new energy conversion concept. It has been researched in several
groups over the world in the suitable frequency range.
Microwave power transmission is the use of microwave to transmit power through
the atmosphere without electrical connection. Refer to Figure 1-3, the frequency range
of microwave is between 1GHz and 30GHz and its wavelength is between 30cm and
1cm. For example, the wavelength of a microwave oven is around 12.2 cm.
A rectenna is simply described as a combination of antenna and rectifier. Antenna is
a conductor device that absorbs electromagnetic (EM) wave from the Sun and transmits
as alternating (AC) current signal through it and rectifier is a device that converts AC
electricity to direct current (DC) electricity. In order to make a rectenna device works, a
support of electrical circuit can be added so that the device can transmit and convert
wave energy into DC electricity. In the past, a conventional rectenna used to be
designed with a diode placed between dipole antennas as shown below;
As shown in Figure 1-4, a half wave dipole is attached to a low-pass filter, which
inserted between the antenna and rectifying circuit. A diode placed across the
transmission line is the rectifier. DC pass filter consists of a large capacitor and a
resistor, which measures the output DC, is placed at the end of circuit. Hence, once the
EM wave is received by the individual antenna, it passes through low pass filter, the
high frequency rectifying diode and resistive load. Pass filter let low frequency signals
pass but diminishes signals with higher than the cutoff frequency. A diode rectifies the
current induced in the antenna by the wave. i.e. make the current goes one way.
6
Figure 1-3 The electromagnetic spectrum [7]
7
Figure 1-4 Rectenna circuit: dipole antenna, low-pass filter, rectify diode, output
DC pass filter and resistive load [8]
8
In order to explain in a different way, a simple electrical circuit is attached as
Figure 1-5; in this case, it will be explained by introducing a half wave rectifier circuit.
Note that bridge rectifier has higher output voltage.
At the first stage, AC signals go into the circuit and the waveforms look like a simple
sine graph. The waveform of AC input voltage is rectified by a diode. During the
positive half wave of AC input, the diode of the rectifier is forward biased and so it
conducts. This charges the capacitor quickly in order to peak a value of supply voltage,
which is also shown by a maximum point of the wave. (e.g. point b and d in the last
graph of Figure 1-5) After being fully charged, the capacitor holds the charge until AC
input supply to the rectifier goes negative. During negative half cycle, the diode gets
reverse biased and stop conduction. These steps are repeated and at the end, we get a
straight line of DC signal by a regulator and it can be measured by a resistor.
Rectenna has been highly efficient converting microwave energy to direct
current electricity if the conversion efficiency of heat energy (~40%) is considered.
Rectenna was involved in microwave power transmission since 1960s and its
conversion efficiency of a rectenna in the microwave regime was upto 90%,
theoretically.
9
Figure 1-5 The electrical circuit diagram for a conventional rectenna
10
1.2.1 Applications
Rectenna can be used in various applications such as solar power satellite (SPS),
wireless networks (e.g. wireless charging for electric devices) and so on. [9] As it is
mentioned previously, a conventional rectenna was used for the transmission of
microwave energy since W. C. Brown brought its idea in 1960s. Furthermore, it is well
known that micro power transmission has been applied in aircrafts since 1960s. [10]
Raytheon Company showed a proof of a concept demonstration; it transmitted
microwave energy to a helicopter so that it could fly at an altitude of 15km. [10] The
helicopter carried a rectenna array incorporating thousands of diodes to convert the
microwave beam into useful electrical power.
Following its work, Figure 1-6 shows the microwave powered aircraft, which was
conducted by Canada Research Community Centre under the project name of SHARP
(Stationary High Altitude Relay Platform). [11] It was also an unmanned plane and
could fly in a circle at 150 meters above transmitting antenna. [11] A circle plane shown
in the figure is a rectenna array, which is placed between wings and a tale. Thus, fuel
free aircrafts can be also a good example of clean energy applications using rectenna.
11
Figure 1-6 Fuel free airplane: the stationary high altitude replay platform (SHARP)
[11]
12
Microwave technology can be used in wireless charging for electrical devices such as
mobile phones and cameras. Furthermore, it can also charge the electrical vehicles
(Figure 1-7). It is good news since the batteries for electric vehicles are quite heavy and
still have a problem of degradation. Therefore, the electrical vehicles supported by
rectenna arrays will shift away from oil powered cars in a long period.
Solar Power Satellite (SPS) relates with a space based solar power (SBSP) system. SPS
has introduced in the late of 1960s and has been mainly investigated by National
Aeronautics and Space Administration (NASA). [10] The efficiency of rectenna
developed by NASA in 1974 and 1975 was around 54%. [10] Refer to Figure 1-8, a
rectenna on Earth receives the solar power and generates electricity continuously in
order to feed the power grid. In addition, the facilities are focused on the system
construction in order to provide electricity from clean energy nowadays. Therefore, the
space solar power (SSP) can spark the industry as well as economy.
13
Figure 1-7 Microwave charging for an electric vehicle by Hashimoto Lab., Japan
14
Figure 1-8 Solar power satellite (SPS); SPS is beamed to a rectenna on Earth. The
power beam is converted into electrical current, which is fed into the power grid [5]
15
1.3 Historical background
In order to understand microwave power transmission for a rectenna, power
transmission by radio waves should be introduced. In 1886, Heinrich Hertz developed
an antenna receiver which revealed power transmission of radio waves. [12] Following
his work, Nikola Tesla sought to apply the principle to the wireless transmission of
electrical power from one point to another. [13, 14] In the late 1930s, microwave power
transmission is introduced by two developments: 1) Klystron tube by O. Heil [10] and 2)
Microwave cavity magnetron during World War II. [15] History of free space
microwave power transmission begun in the late 1950s by Goubau Schwering and W. C.
Brown. Goubau Schwering demonstrated a beam waveguide consisting of lenses or
reflecting mirrors can make microwave power be transmitted and its efficiency reaches
100 percent. [16, 17] W. C. Brown proposed an amplitron, which was developed into a
super power tube producing hundreds of kilowatts of continuous power. [18] In 1960,
Prof. George and Prof. Sabbaugh from Purdue University set an early stage of the
investigation of the semiconductor diodes as rectifiers. [18] Three years later, closed
thermionic diode rectifier was firstly demonstrated by W. C. Brown. In addition, the
demonstration of solid state diodes as rectifier, which made the feasibility to convert the
microwave power into dc power, was shown by W. C. Brown. [10] It came out the idea
of a rectenna. Especially, Roscoe George of Purdue University developed dense arrays
within a waveguide attached to a receiving horn. It had low collection efficiency. [10]
Thus, the individual full wave rectifiers out of the waveguide attached to half wave
dipoles were proposed as a solution [19], which is a basic of current rectenna research
model. Hence, the first rectenna made by W. C. Brown had 40% and 50% of collection
efficiency for 4W and 7W of dc power, respectively. [10] Later, Si schottky barrier
diode was developed by Hewlett Packard Associates, which was more efficient and had
more power handling capability. [10] In 1968, a rectenna was applied into solar power
satellite. [10] In 1971, Wes Mathei of Microstate used GaAs schottky barrier diode,
which had greater efficiency and a power handling capability. [10] Since the middle of
1970s, the property of rectenna has been found; according to MSFC Sponsored Program,
power density at the center of the rectenna is greater than the edge of it. Overall
efficiency was around 48±2%. [10] In terms of solar power satellite (SPS), from 1974 to
1975, NASA presented the upgrade version of rectenna, which has 54±1% of efficiency.
[10] Another SPS study by NASA resulted in electrical and mechanical improvement to
rectenna from 1976 to 1977, supported by Lewis Research Center (LeRC). [10] Hence,
starting from the first demonstration of rectenna by W. C. Brown in 1960s, the research
of rectenna has been continued around the world.
16
1.4 Literature review
There are around 150 published papers about rectenna since 1980s. The
following Figure 1-9 shows that the trend of rectenna is upwards. Especially, in Figure
1-9 (right), it is found that the number of published journals is doubled in the 21st
century.
In this section, it is divided into two categories: international research trends
and national research trends.
1.4.1 International research trends
The following Table 1-1 and Table 1-2 show that which researches have been
done by groups / year. Each paper was found on Web of Science website [20] and
especially, papers found in Table 1-1 were selected due to higher citation numbers
(i.e.≥ 25).
International research groups on the above Table 1-1 have been mostly focused on a
frequency range which rectennas rely on. Also, material selections for diode and design
of rectenna array have been investigated in several groups in order to improve the
conversion efficiency. Furthermore, D. Y. Goswami et al. mentioned a new concept of
nano sized antenna in 2004, which is solar energy conversion. [25]
17
Figure 1-9 (Left) number of patents and (right) number of journals
18
Table 1-1 International research groups
Author Year Title Contents Reference
W. C. Brown (Raytheon Company, USA) and E. Eugene Eves (Raytheon Company, USA)
1992 Beamed microwave transmission and its application to space
General principle of beamed microwave power transmission system and discussion about its application to the space program. 54% of efficiency has achieved.
[21]
T. W. Yoo (Texas A&M University, USA) and K. Chang (Texas A&M University, USA)
1992 Theoretical and experimental development of 10 and 35GHz rectennas
A theoretical analysis to predict the performance of the rectenna. A 35GHz rectenna with 39% conversion efficiency.
[22]
James O. McSpadden et al. (Boeing Information, Space, and Defence Systems Group, USA)
1998 Design and experiments of a high conversion efficiency 5.8GHz rectenna
High efficiency rectenna at 5.8GHz with a silicon Schottky barrier mixer diode.
[8]
Y. H. Suh et al. (Texas A&M University, USA)
2000 Circularly polarized truncated corner square patch microstrip rectenna for wireless power transmission
The microstrip rectenna at 5.8GHz with GaAs Schottky barrier diode has 60% of efficiency.
[23]
Y. H. Suh (Texas A&M University, USA) and K. Chang (Texas A&M University, USA)
2002 A high-efficiency dual-frequency rectenna for 2.45-and 5.8-GHz wireless power transmission
A dual frequency printed dipole rectenna at 2.45 and 5.8GHz with GaAs Schottky barrier diode has 84.4 and 82.7% of efficiency, respectively.
[24]
D.Y. Goswami et al (University of Florida, USA)
2004 New and emerging developments in solar energy
Nanoscale antenna as future energy source conversion device and its challenges
[25]
J. A. Hagerty et al. (Boulder company, USA)
2004 Recycling ambient microwave energy with broad-band rectenna arrays
A 64 element dual circularly polarized spiral rectenna array over a frequency range of 2-18GHz. Increased in rectenna efficiency for multitone input waves.
[26]
J. P. Curty et al. (Swiss Federal Institute of Technology, EPFL)
2005 A model for mu-power rectifier analysis and design
A rectenna designed in a finite ground coplanar waveguide (FG-CPW) circuit has 68.5% of efficiency.
[27]
19
1.4.2 National research trends
Besides, Korean research groups are mainly focused on printing rectenna
elements on flexible materials.
It seems that the two groups are focused on the design of the micro sized rectenna. For
instance, J. Kim’s group has been improved a flexible rectenna using by cellulose
electro active paper (EAPap). Sang H Choi et al. has been developed an on/off power
switch for saving the energy. This year, H. Park et al. also published on Nanotechnology
about printing technologies.
20
Table 1-2 National research groups
Author Year Title Contents Reference J. Kim et al. (Inha Uni., South Korea)
2006 Micro transfer printing on cellulose electro-active paper
A micro patterning tech, gold electrode patterning on cellulose electro active paper (EAPap) for a microwave dipole rectenna
[28]
J. Kim et al. (Inha Uni., South Korea and NASA Langley Research Center, USA)
Performance characterization of flexible dipole rectennas for smart actuator use
A flexible rectenna over a freq. range of 9-12GHz (20-50%, depending on input power)
[29]
Microwave power transmission using a flexible rectenna for microwave-powered aerial vehicles
Two flexible rectenna over a freq. range of 9-12GHz for a microwave powered aerial vehicle (MPAV)
[30]
* Sang H Choi et al. (NASA Langley Research Center, USA) and T. Itoh (University of California, USA)
2008 A 60 GHz retrodirective array system with efficient power management for wireless multimedia sensor server applications
Efficient control of power; extremely low power consumption for a sleeping mode or, working as a retrodirective array at >0.013mW/cm of received power density
[31]
S. Y. Yang et al. (J. Kim’s group, Inha University)
2010 Wirelessly driven electro-active paper actuator made with cellulose-polypyrrole-ionic liquid and dipole rectenna
Wirelessly driven electro-polypyrrole-ionic liquid (CPIL) electrode-active paper actuators are reported.
[32]
H. Park et al. (Sunchon National University, Korea, PARU Co. Ltd, Korea and GTFAM regional innovation center, Korea)
2012 Fully roll-to-roll gravure printed rectenna on plastic foils for wireless power transmission at 13.56MHz
Six printing units of R2R gravure have been successfully employed to print a rectenna as a wireless power transmission device at 13.56 MHz very cheaply.
[33]
*Work with Chung-Ang University in South Korea and Norfolk state University in USA.
21
1.5 References
[1] Jos G.J.Oliver, Greet janssesns-Maenhout, Jeroen A.H.W.Peters and Julian Wilson
[Online], 2011, Long-term trend in global CO2 emissions. 2011 report,
http://edgar.jrc.ec.europa.eu/news_docs/C02%20Mondiaal_%20webdef_19sept.pdf
[2] Conservation Magazine, [Online], 2012, Brief Dip in Emissions Then Record Rise,
http://www.conservationmagazine.org/2011/11/economic-meltdown/
[3] Jos G.J.Oliver, Greet janssesns-Maenhout and Jeroen A.H.W.Peters, [Online], 2012,
Trends in global CO2 emissions; 2012 Report,
http://edgar.jrc.ec.europa.eu/CO2REPORT2012.pdf
[4] SolarContact, [Online], 2012, Solar Cell Efficiency,
http://www.solarcontact.com/solar-panels/solar-cells/efficiency
[5] IEEE Global History Network, [Online], 2012, W. C. Brown: Biography,
http://www.ieeeghn.org/wiki/index.php/William_C._Brown
[6] Wikipedia, [Online], 2010, Rectenna and Nantenna,
http://en.wikipedia.org/wiki/Rectenna and http://en.wikipedia.org/wiki/Nantenna
[7] Princeton University, [Online], 2010, Electromagnetic spectrum,
http://web.princeton.edu/sites/ehs/laserguide/index.htm
[8] J. O. McSpadden et al., IEEE Trans. Microw. Theory Tech., 46, 2053-2060, (1998)
[9] Space Studies Institute, [Online], 2011, http://ssi.org/space-art/ssi-sample-slides/
[10] William C. Brown, IEEE Trans. Microwave Theory Tech.,32, 1230-1242, (1984)
[11] How Wireless Power Works, Tracy V. Wilson, [Online], 2012,
http://electronics.howstuffworks.com/wireless-power.htm/printable
[12] H. Hertz, Dictionary of Scientific Biography, vol. VI. New York: Scribner, pp. 340-
349
[13] J. J. O’Nell, Prodigal Genius – the Life of Nikola Tesla, New York, Washburn,
1944
[14] M. Cheney, Tesla, Man Out of Time. Englewood Cliffs, NJ: Prentice-Hall, 1981
[15] H. Boot and J. Randall, IEEE Trans. Electron Devices, ED-23, (1976)
[16] G. Goubau and F. Schwering, IRE Trans. Antennas Propagat., AP-9, 248-256,
(1961)
22
[17] J. Degenford, W. Sirkis, and W. Steier, IEEE Trans. Microwave Theory Tech., 445-
453 (1964)
[18] W. C. Brown, Proc. IRE., 45, 1209-1222, (1957)
[19] W. C. Brown et al., U.S. Patent 3434678, March 25, 1969
[20] ISI Web of Knowledge search tool, Web of Science, [Online],
http://apps.isiknowledge.com
[21] W. C. Brown and E. E. Eves, IEEE Trans. Microw. Theory Tech., 40, 1239-1250,
(1992)
[22] T. W. Yoo and K. Chang, IEEE Trans. Microw. Theory Tech., 40, 1259-1266,
(1992)
[23] Y. H. Suh et al., Electron. Lett., 36, 600-602, (2000).
[24] Y. H. Suh and K. Chang, IEEE Trans. Microw. Theory Tech., 50, 1784-1789,
(2002).
[25] D.Y.Goswami et al., Solar Energy, 76, 33-43 (2004)
[26] J. A. Hagerty et al., IEEE Trans. Microw. Theory Tech., 52, 1014-1024, (2004)
[27] J. P. Curty et al., IEEE Trans. Circuits Syst. I-Regul. Pap., 52, 2771-2779, (2005)
[28] J. Kim et al., Smart Mater. Struct. 15, 889-892 (2006)
[29] J. Kim et al., Smart Mater. Struct., 15, 809-815 (2006)
[30] J. Kim et al., Smart Mater. Struct., 15, 1243-1248 (2006)
[31] S. Lim and T. Itoh, IET Microw. Antennas Propag., 2, 615-621 (2008)
[32] S. Y. Yang et al., Smart Mater. Struct., 19, (2010)
[33] H. Park et al., Nanotechnology, 23, 344006 (2012)
23
CHAPTER 2.
Built in MIM structure
24
2.1 Introduction
A rectenna is a combination of antenna and rectifier. Once wave energy (i.e.
electromagnetic wave from Sun) comes, antenna captures it and converts into
alternating (AC) current signal. In other words, when the wave energy comes to the
antenna, which is a conductor, electrons inside the conductor vibrates, thus, the
oscillation of electrons occurs. The AC current signal transmits through the antenna then
needs to be rectified. And the role of diodes is the rectification.
Nowadays, nanotechnology has been developed surprisingly so that small
features even in nanometer (nm) scales are available. In the past, rectenna used to
convert microwave energy into a usable output, however, it is now able to convert near
infrared (near IR) and/or visible spectrum due to the improvement of nanotechnology.
Therefore, it is possible to aim a high frequency range, such as few hundreds of
terahertz (THz).
If schottky diode has been used as rectenna element in microwave energy, a
different type of diode is required in higher frequency regime. Rectification in such high
frequency is a big challenge and Metal-Insulator-Metal (MIM) diodes are suitable due
to its high speed (around femtoseconds, 10-15 of a second) and capability of rectification.
[1]
If conventional rectenna has two elements, antenna and rectifier, separately
(they are usually connected in an electrical circuit), our device combined the two
element together in one device. i.e. built in MIM structure. In this chapter, the basic
principle of MIM diode is explained followed by a brief discussion about conduction
mechanism in diodes. Then design rule of antenna and MIM diode will be introduced
for a unique design, MIM built in structure. Lastly, requirements of MIM diode in
rectenna array will be discussed at energy harvesting point of view.
2.2 Basic principles of MIM diode
Once oscillating current occurs through antenna, element that rectifies it is
needed. Diode does the work in rectenna device. In microwave regime, schottky diode
was frequently used, however, at higher frequency (few hundreds of terahertz), a diode
that has high speed is required. Metal-Insulator-Metal (MIM) diode became suitable due
to its femtosecond of speed. Hence, MIM diode can do the work in near infrared (near
IR) and/or visible range.
MIM diode stands for metal-insulator-metal diode; incorporates an insulator
that is a few nanometers (nm) thick between two metals. i.e. sandwich structure In the
past, Cat Whisker diodes were frequently used; a tungsten wire connected to oxidized
25
metal sheet. Due to the development of nanotechnology, small size (in nm scales) of
MIM diode became available.
2.2.1 Symmetry and asymmetry
There are two types of MIM diode: symmetry and asymmetry. When two
metals are the same (refer to Figure 2-1 (b)), work function of those two metals are the
same. It means that electron tunneling probability from M1 to M2 or M2 to M1 is the
same. Hence, we have symmetrical J-V curve as shown in Figure 2-1 (c). If two
different metals with different work function are used in MIM diode, the graph shifts to
one direction due to different tunneling probability on both sides. Note that simulation
curve in Figure 2-1 is based on Fowler-Nordheim (F-N) equation, which will be
explained within this Chapter 2.
The role of diode in rectenna device is to make current signal in one way in
order to produce usable output direct current (DC). Therefore, we decided to use two
different metals in our device in order to induce more current due to high asymmetry.
Gold (Au) and aluminum (Al) are used in this study; work function of Au is 5.3eV and
that of Al is 3.9eV. Furthermore, barrier height of Au-aluminum oxide (Al2O3) is 4.1eV
and that of Al-Al2O3 is 2.1eV. Refer to Figure 2-2 (b), the Fermi level shifts upwards
when negative bias is applied on Al side. On the other hand, the Fermi level shifts down
when positive bias is applied.
26
Figure 2-1 (a) TEM image of a symmetric MIM (ZrCuAlNi-Al2O3-ZrCuAlNi)
diode. (b) Equilibrium energy band diagram for a symmetric MIM diode. (c)
Measured and simulated current density-field curves for the MIM diode shown in
(a). (d) TEM image of an asymmetric MIM (ZrCuAlNi-Al2O3-Al) diode. (e)
Equilibrium energy band diagram for an asymmetric MIM diode. (f) Measured
and simulated current density-field curves for the MIM diode shown in (d). [2]
27
Figure 2-2 Band diagram of MIM diode when (a) Fermi level is aligned (b)
negative bias is applied (c) positive bias is applied
28
2.2.2 Conduction Mechanism
On our MIM diode, three conduction mechanisms can be mainly discussed:
Schottky emission (field assisted thermionic emission), Fowler-Nordheim (F-N)
tunneling and Direct tunneling.
In case of thermionic emission, heat-induced flow of charge carriers occurs
from a surface or over a potential energy barrier. [3] And the related Equation 2-1 is
shown below;
J = ∗ − ( − /4
Equation 2- 1
where the effective Richardson constant, ∗ = ∗
ℏ , T = absolute temperature in
K, q = magnitude of electronic charge, = barrier height, E = electric field and =
insulator dynamic permittivity.
Except thermionic emission, there are two main tunneling mechanisms for
MIM diode: F-N tunneling and direct tunneling. If the applied bias is largely on, the
oxide region tends to be triangular shape. In other words, oxide thickness gets
effectively smaller so that more electron tunneling probability could be made. The
following Equation 2-2 shows F-N relationship;
J = A −
Equation 2- 2
where A =
ℏ
∗ and B =
( ∗ ) /
ℏ
/ .
In our case, applied bias voltage should be over 2V for having F-N phenomena. It was
confirmed theoretically (by drawing a band diagram) and experimentally (Figure 2-3
and Figure 2-4). Refer to Figure 2-3 and Figure 2-4, the barrier height of each
electrode-oxide layer is found to be extremely small. i.e. F-N tunneling does not match
in this case.
Therefore, direct tunneling is expected for our device. Current density – Voltage
(J-V) theory could be a good explanation for direct tunneling. A brief derivation would
be introduced in the next section, 2.2.3 J-V theory.
29
Figure 2-3 F-N plots applied on Al-Al2O3 side
Figure 2-4 F-N plots applied on Al-Al2O3 side
30
2.2.3 J-V theory
In this section, standard practice to use the analytical approximation of MIM
diode will be shown and J-V curve is derived by John G. Simmons [4] and it also uses
the WKB approximation [5].
Current density – voltage theory (J-V) is derived from Schrödinger equation for
low temperature so that thermal current could be neglected. The well-known time
independent equation would be shown as following;
+2
ℏ [ − ( )] ( ) = 0
Equation 2- 3
where E-V(x) term indicates kinetic energy of a particle and ℏ is Planck’s constant.
Tunneling probability can be expressed by WBK approximation and current expression
for MIM diode can be extracted from electron tunneling probability, T;
T~exp −2 | ( )|
Equation 2- 4
where absolute value of the wave vector of carrier in the barrier,
| ( )| =
ℏ( − ). So, the probability that an electron can penetrate a potential
barrier of height;
D(Ex) = exp −4
ℏ [2 ( ( ) − ( )] /
Equation 2- 5
where = the energy component of the incident electron in the x direction.
31
Figure 2-5 A schematic band diagram of MIM diode with rectangular potential
barrier, electrode 1 and electrode 2 for V< /e [4]
32
Refer to Figure 2-5; the number N1 of electrons tunneling through the barrier from
electrode 1 to electrode 2 and the number of N2 of electrons tunneling through the
barrier from electrode 2 to 1. The tunneling probability, D(Ex) is the same in either
direction and it is assumed that a positive potential is applied on electrode 2 so that the
Fermi-Dirac function is written as f(E+eV) as following;
1 = (
) ( ) =1
( ) ( )
=4
ℏ ( )
( )
N2 =4
ℏ ( ) ( + )
Equation 2- 6
where = the maximum energy of the electron in the electrode, ( ) = the
number of electrons per unit volume with velocity between and + and
= /2 since it is expressed in polar coordinates.
Net flow of electrons, N=N1-N2;
=4
ℏ ( ) ( )
− ( ) ( + )
=4
ℏ ( ) [ ( ) − ( + )]
Equation 2- 7
where f(E) = the Fermi-Dirac distribution function.
Let 1 and 2 be;
1 =
ℏ ∫ ( )
and 2 =
ℏ ∫ ( + )
Equation 2- 8
Substitute 1 and 2 into net flow of electron, N;
J = D( )ζd
Equation 2- 9
33
Then current density, J, can be expressed as;
J=
ℏ ∫ exp[− ( + − )
/
] + ∫ ( − )exp[− ( + −
) / ]
Equation 2- 10
where = mean barrier height and = Fermi-level.
After its expansion and some integral works;
= exp − − ( + )exp[− ( + )]
Equation 2- 11
where =
ℏ( )2 , = 1 −
∫ [ ( ) − ]2
(if ~1 , it is a good
approximation), s = thickness of insulating film and s1 and s2 = limits of barrier at
Fermi level (Δs = s2 − s1).
Since the electrical measurement was carried out in a range, -1<V<1, it is defined as
intermediate-voltage range (V <
) so that Δs = sandφ = −
where =
height of rectangular barrier. Hence, J-V expression is finally concluded as;
=
2 2 −
2 exp
−4
ℏ(2 )
−
2
− +
2
−4
ℏ(2 )
+
2
Equation 2- 12
34
2.3 Design for built in MIM structure
In this part, 2.3.1 design rule of antenna and 2.3.2 requirements of MIM
diode for rectenna array will be introduced.
2.3.1 Design rule of antenna
Typical antennas are based on the half wave dipole antennas and they have a
resonant frequency. However, in nanometer scales, the rule of resonant frequency (e.g.
half wave) is meant to be changed. There are various methods of defining antenna
length but Novonty’s method, which was published in Physical review letters (2007), is
standard. [6] It is beyond human work to understand the theory, however, it is the well-
known one of defining nano sized antenna length. Thus, we decided our antenna length
based on Novonty’s method. According to his method, antenna length is half of
effective wavelength. In his assumption, antenna is made of linear segments with radius
R<<λ and metal can be described by a free electron gas as shown in Figure 2-6.
35
Figure 2-6 A single antenna element represented by a metal rod. Incident light with
wavelength polarizes the ends and gives rise to a standing surface charge wave. [6]
36
A single antenna is a cylindrical shape and the following equation was assumed as;
L = 2
= 1 + 2
Equation 2- 13
where = effective wavelength, = plasma wavelength and n1 and n2 are
coefficients with dimensions of length that depend on antenna geometry and static
dielectric properties.
And the effective wavelength is expressed as;
= − 4
Equation 2- 14
where free space wavenumber, = 2 / and γ = propagation constant of the
surface charge wave.
In Equation 2-14, 4R is approximation value considering that electromagnetic (EM)
wave may react with both ends of the rod. Thus, 4R is subtracted in the equation since
we want to find out the antenna length. Furthermore, in order to calculate propagation
constant gamma (γ), electrical field mode simulation is used. And the following
equation is shown analytically;
( ) ( )
( )−
( )( )
( )( )
= 0
Equation 2- 15
where J = cylindrical Bessel functions (to find a solution for wave propagation in a
cylindrical waveguide) and H = cylindrical Hankel functions (which are Bessel
function’s 3rd kind). K1 and K2 are defined by the propagation constant as =
( −
) / and = ( −
) / , respectively. ( )= dielectric function
and = dielectric constant for the rod which is embedded in a medium.
37
Based on the above equations, we used Mathematica program to calculate
antenna length for near infrared (near IR) spectrum. The following Table 2-1 shows the
results.
Our targeting is within near infrared (near IR) region; between 925nm to 1192nm of
wavelength. The detailed fabrication process of antenna will be explained in Chapter 3
of this dissertation.
38
Table 2-1 Antenna length and antenna width for total 60 cells
L= antenna length (nm) and W=antenna width (nm) L 424~
1375 394~ 1100
424~ 1375
394~ 1100
424~ 1375
394~ 1100
424~ 1375
394~ 1100
424~ 1375
394~ 1100
W 100 100 80 80 60 60 40 40 20 20 L 360~
840 330~ 570
360~ 840
330~ 570
360~ 840
330~ 570
360~ 840
330~ 570
360~ 840
330~ 570
W 100 100 80 80 60 60 40 40 20 20 L 480 460 440 420 400 380 360 340 320 300 W 100 100 100 100 100 100 100 100 100 100 L 480 460 440 420 400 380 360 340 320 300 W 80 80 80 80 80 80 80 80 80 80 L 480 460 440 420 400 380 360 340 320 300 W 60 60 60 60 60 60 60 60 60 60 L 480 460 440 420 400 380 360 340 320 300 W 40 40 40 40 40 40 40 40 40 40
39
2.3.2 Requirements of MIM diode for rectenna array
MIM diode incorporates a thin insulating layer that is a few nanometers thick
between two metals. MIM diode in rectenna requires two factors: high asymmetry and
fast responsivity. Firstly, alternating current (AC) signal, obtained from antenna, is
needed to pass in one direction. At this point of view, high asymmetry is required. In our
device, we used two different metals for MIM diode in order to get high asymmetry
factor. Secondly, we targeted few hundreds of THz, the device needs high speed. This is
why we chose MIM diode over other diodes.
Hence, under design rule of antenna and MIM diode, we fabricated built in
MIM structure for our rectenna device. In the next Chapter 3, the built in MIM
structure will be introduced with detailed fabrication process.
2.4 References
[1] S. Krishnan, Ph.D. Dissertation, University of South Florida (2008)
[2] E. William Cowell III et al., Adv. Mater., 23, 74-78, (2011)
[3] S. M. Sze, Physics of Semiconductor Devices, 1981, John Wiley & Sons, Inc., New
Jersey
[4] J. G. Simmons, J. Appl. Phys., 34, 1793 (1963)
[5] D. Bohm, Quantum Theory, 1951, Prentice-Hall, Inc., New Jersey
[6] L. Novonty, PRL, 98, 266802 (2007)
40
CHAPTER 3.
Fabrication of
built in MIM structure
41
3.1 Introduction
This work focuses on the rectenna working in near infrared (near IR) and/or
visible light, which requires hundreds of terahertz (THz) in frequency. It calls for the
most precise patterning, yet also large areal coverage.
As it is explained in the previous Chapter 2, MIM diodes which consist of
sandwich structure- an insulating layer between two metals- is suitable for wave energy
with high frequency. Two metals with different work functions were used in MIM
structure in order to make alternating current (AC) signals pass in one way. In this case,
aluminum (Al) and gold (Au) were selected; work function of Au is around 5.3eV and
that of Al is around 3.9eV. Aluminum oxide (Al2O3) is used as an insulating layer.
Unlike conventional rectenna, two elements, antenna and diode, are combined
in one device as shown below. i.e. built in MIM structure – a rectifying antenna coupled
diode (Figure 3-1).
The following Figure 3-2 shows the whole fabrication process. The fabrication
of the rectenna device consists two main parts: 3.2 Fabrication of Au nanowires and
3.3 Fabrication of rectifying metal electrodes.
42
Figure 3-1 Built in MIM structure (left) 3-D image (right) 2-D cross section
43
Figure 3-2 Fabrication of built in MIM structure
44
3.2 Fabrication of Au nanowires
Two methods were used in order to p Au nanowires; AIPEL and conventional e-
beam lithography. For each method, different types of photoresist (PR) are used so that
the process would be slightly different compared to Figure 3-2. Note that positive PR is
used in the above Figure 3-2.
3.2.1 AIPEL
AIPEL stands for atomic image projection e-beam lithography. It is the
modified 200kV transmission electron microscopy (TEM) with a field emission gun
(JEM-2010F, JEOL Ltd.). AIPEL is one of nanolithography processes using a high-
resolution image of crystal material by phase-contrast generation in a transmission
electron microscopy. [3] The Ångstrom-scale lattice image of a crystalline material is
magnified 50 to 300 times within the electron microscope and projected onto an
electron-beam-resist-coated silicon substrate. [1,2] Via this method, the successful
fabrication of periodic arrays of metal nanowires were successfully patterned for the
first trial by using beta-silicon nitride (β-Si3N4) as mask materials. This method
promises the potential for high-throughput lithography for various interesting periodic
nanostructure with substantially smaller feature sizes. [1] With this method, gold (metal)
nanowires were successfully patterned.
45
Figure 3-3 (Left) a picture of AIPEL equipment (right) high resolution image of β-
Si3N4, which is used as a mask material for the AIPEL system [2]
46
Figure 3-4 SEM images taken after the fabrication of Au nanowires by AIPEL (left)
a whole exposure area with dose 800 C/cm2, at magnification of 1.6K (right)
central part of the whole exposure area with dose 800 C/cm2 at magnification of
5K
47
In order to obtain Au nanowires as shown Figure 3-4, the following steps
should be done; a sample is prepared as silicon dioxide (SiO2) with a thickness of
100nm is deposited on a 4 inch p type silicon (Si) wafer. After the preparation, titanium
(Ti) with a thickness of 3nm and gold (Au) with a thickness of 30nm were also
deposited on the same wafer. For the deposition of two metals, e-beam evaporator
(MAESTECH ZZS550-2/D) was used and it was done by in-situ deposition. In this case,
Ti is meant to serve as an adhesion layer. For the adhesion between hydrogen
silsequioxane (HSQ) and Au layer, amorphous silicon layer (a-Si) with a thickness of
5nm is also deposited by e-gun evaporator. Patterning by e-beam lithography has been
known as one of methods to produce nanowires. For this process, HSQ, which is a
negative e-beam resist material, is spin-coated. It is known that it can resolve 10nm
scale nanostructures. [4] It is also necessary to develop after the exposure to e-beam
irradiation so that the exposed area of HSQ would remain. The unexposed area of HSQ
would be removed by tetramethylammonium hydroxide (TMAH). After the develop
process, physical etching is done with Cl2 gas using plasma dry etcher (Oxford RIE 80
Plus). The results of nanowires can be observed by scanning electron microscope (JEOL
JEM-7401F SEM), and confirm that the designed Au nanowires with a length of ~60nm
were fabricated from end to end of the exposed area, with ~30-40μm of diameter.
3.2.2 Conventional e-beam lithography
To define precise nanowire length, conventional e-beam lithography was also
tried and successfully done. And the sample that conventional e-beam lithography used
was also measured and the results will be shown and discussed in the next Chapter 4.
Compared to 3.2.1 AIPEL, the experimental steps are slightly different due to different
type of e-beam resist materials. In this case (also refer to Figure 3-2), a positive type e-
beam resist material, poly methyl methacrylate (PMMA) was spin-coated on silicon
dioxide (SiO2) wafer then e-beam was exposed by conventional e-beam lithography.
After the develop process, titanium (Ti) and gold (Au) were deposited by e-gun
evaporator. Since a positive type resist material was used, lift-off process was required.
For lift-off procedure, the sample was dipped into acetone solution.
3.3 Fabrication of rectifying metal electrodes
In this part, it is a combination of photolithography and e-beam lithography
methods. Gold (Au) nanowires and Au electrode was patterned together by e-beam
lithography as shown in Figure 3-5. In Figure 3-5, horizontal rods indicate Au
nanowires and a vertical rod shows Au electrode.
48
Figure 3-5 SEM images taken after the fabrication of Au nanowires and Au
electrode by conventional e-beam lithography with dose 2500 C/cm2 (left) at
magnification of 10K (right) at magnification of 25K
49
Before patterning the metal bottom pads, AZ5214, a negative photoresist (PR),
was spin-coated on the sample. Photolithography was used for patterning of top metal
pads. Note that optimized exposure time is found to be 7 seconds. The deposition of
titanium (Ti) and gold (Au) was carried out by e-beam evaporator. Once again, Ti is
used for adhesion. Thus, very thin (less than 5nm) Ti layer is needed and 3nm of Ti was
used in this case. After lift-off process, the sample can be checked by scanning electron
microscope (SEM). Since the device structure is MIM (Figure 3-1), dielectric layer is
needed. An insulating layer of the device, aluminum oxide (Al2O3) with a thickness of
10nm is deposited on the sample by atomic layer deposition (ALD, Lucida D100) at
220oC. After ALD process, the conventional e-beam lithography method is proceeded in
order to pattern aluminum (Al) electrode. Again, poly methyl methacrylate (PMMA)
was done before e-beam lithography session. Current optimized dose is 2500μC/cm2
and 1nA of current is used. The deposition of Al was done by e-beam evaporator and the
lift-off process was done in order to keep Al remain where the electrode should be.
Lastly, metal top pads are deposited by e-gun evaporator followed by the lift-off process.
The final state of sample can be checked by scanning electron microscope as shown in
Figure 3-6.
50
3.4 MIM built in structure
Refer to Figure 3-6, the whole fabrication of rectifying optical antenna array
was successfully fabricated. Once again, horizontal rods indicate gold (Au) nanowires
and two vertical rods show aluminum (Al) electrode and Au electrode. Between two
electrodes, a thin aluminum oxide (Al2O3) layer exists, forming a sandwich structure. i.e.
MIM
Since the whole device contains nanowires (antennas) and diode together, we
called it as ‘built in MIM structure’, which is a unique design compared to the
conventional rectenna array. This is because the conventional rectenna arrays have two
components, antenna and diode, separately and they are connected in an electrical
circuit.
Our design is also divided into two; rectangular shape and trapezoidal shape;
the shape indicates the region between two electrodes in the SEM images. Figure 3-6
(left) shows a rectangular shape; two electrodes are parallel to each other. It was aimed
to confirm resonant wavelength for each antenna length since the same length of the
antenna exists in the device. Note that the antenna length equals to the length of Au
nanowires (horizontal rods in the SEM images) between two electrodes (two different
vertical rods in the SEM images). Figure 3-6 (right) shows a trapezoidal shape; two
electrodes are not perpendicular to each other compared to Figure 3-6 (left) and the
area between two electrodes forms a trapezoid. This design was aimed to confirm the
light response as whole antenna length exists together. The measurement of this device
will be shown and discussed in the following Chapter 4 of this dissertation.
51
Figure 3-6 SEM images of the final state of the device showing (left) a rectangular
shape at magnification of 15K (right) a trapezoidal shape at magnification of 20K
52
3.5 References
[1] H. S. Lee and K. B. Kim et al., Adv. Mater. (2007)
[2] [Patent] An Apparatus and a Method for Forming a pattern Using a Crystal Structure
of Material, PCT/KR02/00109 (Jan. 24 2002)
[3] B. S. Kim et al., 2nd US-Korea NanoForum, Los Angeles, USA February 17-19,
2005
[4] S. W. Nam et al., J. Electrochem. Soc., 154, H844-H847 (2007)
53
CHAPTER 4.
Characterization
54
4.1 Introduction
Rectenna was made under a unique design, built in MIM structure, and it has
been found to be an optical sensing and/or energy harvesting device. The design rules
and fabrication process has been interpreted in the previous chapters.
A series of measurements were conducted in order to characterize the device
structure and the current voltage dependence, as well as optical response of the
fabricated rectenna array devices. As it was explained in the previous Chapter 3, two
different types of device were measured; trapezoidal shape and rectangular shape
(Figure 3-6). The trapezoidal shape could confirm whether the device reacts with light
(i.e. optical sensing) and the rectangular shape shows its antenna effect. Thus, ability of
optical sensing was tested as well as antenna effect. Furthermore, asymmetry factor was
calculated to test our MIM structure.
Based on the analysis of the device, the contribution and its future will be
discussed in the next Chapter 5 of this dissertation.
4.2 Experimental details
For this analysis, three different equipments were used: Agilent 4156C,
KEITHLEY 4200-SCS and IPCE (incident photon to charge carrier efficiency).
The device characterization was done by an electrical measurement. Agilent 4156C was
used for I-V measurement. KEITHLEY 4200-SCS was used for characterization of
illuminated MIM structure (under visible spectrum). Lastly, IPCE was applied to find
out resonant antenna length over a broad wavelength of incident light (from 300nm to
1100nm). Hence, antenna effect could be tested from this technique.
55
Figure 4-1 Equipment used for this analysis: (top) Agilent 4156C (middle)
KEITHLEY 4200-SCS (bottom) IPCE
56
4.3 Results and Discussions
4.3.1 Unilluminated and illuminated MIM structure
Current density – voltage (J-V) was produced when Agilent 4156C was used.
The voltage range was from -1V to +1V.
Two different metals were chosen for MIM structure and it was intended to get
asymmetry phenomena. It could be confirmed by checking asymmetry factor. [1]
Asymmetry factor is the absolute value of the ratio of forward bias current to reverse
bias current. [1] For example, if asymmetry factor equals to 1, it indicates no
rectification. From the graph of Figure 4-2, asymmetry factor was found to be 1.6. This
is because current density values are -15.12A/cm2 at -1V and 9.45A/cm2 at 1V. In this
device, it shows rectification. In other words, the barrier introduces an asymmetry on J-
V curve favoring negative bias voltage. Hence, the nonlinearity in MIM diode structure
is evident directly from the measured data.
As it was designed as an optical sensing device, the measurement was carried
out under the light. i.e. illuminated MIM structure.
Current-Voltage (I-V) curve was obtained as shown in Figure 4-3. The graph contains
two: light off and light on. Light source was just white light with 150W of power and
the equipment was the same as previous one, Agilent 4156C. Surprisingly, the curve
shows optical sensing and light energy harvesting effect; the red curve shifts upward
compared to the dark one. As the power incident on the diode increases, corresponding
to an increasing number of photons incident of a diode, the illuminated I-V curve shifts
upward. In other words, under the light, more number of photons is captured by antenna
showing higher current level. Although the current level is low (~nA), at least the device
shows its sensitivity of light.
57
Figure 4-2 J-V curve for 380nm of antenna length and 40nm of antenna width
58
Figure 4-3 I-V curve for 380nm of antenna length and 40nm of antenna width
59
4.3.2 Antenna length effect
When rectenna was originally designed, we would like to find out resonant
antenna length for each wavelength of incident light. i.e. antenna length effect. Since
KEITHLEY 4200-SCS equipment has a filter so that five different wavelengths (450,
500, 550, 600 and 650nm) can be applied.
Refer to Figure 4-4 (left), the cell with 480nm of antenna length shows the highest
current when the wavelength of incident light is 650nm. And as shown in Figure 4-4
(right), the highest current level for 300nm of antenna length is observed when the
wavelength of incident light is 600nm. Resonant wavelength of those different antenna
length are not the same, however, the current levels are very similar. It is expected that
our device does not show antenna effect in visible spectrum.
The targeting resonant wavelength is within near infrared spectrum for our
designed antenna. Theoretical value of resonant wavelength is calculated from
Novonty’s method [2] by using Mathematica. Then, IPCE became suitable for this
analysis. Note that little fluctuation shown on the graph of Figure 4-5 is expected as
noise.
As a result, it is experimentally found that 900nm is resonant wavelength for 380nm of
antenna length. The resonant wavelength is meant to be 1044nm theoretically. Those
two values do not match. Therefore, it is concluded that the device has antenna length
effect since it has highest current level at a wavelength value. However, the antenna
length effect is extremely small.
60
Figure 4-4 I-V curves for illuminated device with 80nm of antenna width (left)
480nm of antenna length (right) 300nm of antenna length
61
Figure 4-5 A graph showing current signal of a cell (380nm of antenna length and
40nm of antenna width) in visible-near IR spectrum
62
4.4 References
[1] P. Periasamy et al., Adv. Mater., 23, 3080-3085 (2011)
[2] L. Novonty, PRL, 98, 266802 (2007)
63
CHAPTER 5.
Summary and Conclusions
64
Rectenna is a combination of antenna and rectifier (also referred to as a
rectifying optical antenna). It has been pursued as a mean to directly convert light into
direct current (DC) output for sensing and possibly even light energy harvesting.
In the past, rectenna used to target microwave energy, however, it has been tried
to convert visible-near infrared spectrum. Its extension to higher frequencies has
remained a challenge, because of the correspondingly smaller feature sizes and the need
for scaling up the array size, despite the efforts by a number of research groups around
the world.
In this dissertation, a unique design of rectenna device, which is built in MIM
structure, is introduced. In Chapter 2, it includes design rules of antenna and
requirements of MIM diode in rectenna array. In Chapter 3, it shows the fabrication of
rectenna arrays using a special projection e-beam lithography and conventional e-beam
lithography. These arrays are designed to operate in near infrared/ infrared light
spectrum, which is around few hundreds of THz of frequency. In Chapter 4,
characterization of the device is shown and discussed; a series of measurement is
conducted in order to characterize the device structure and the current voltage
dependence, as well as its optical response of the fabricated rectenna array devices.
Especially, for illuminated device, there is a dramatic increase in current response, by
around 3 orders of magnitude, to incident broad-band light (white light from a
microscope). In addition, each antenna length shows different level of current signal,
however, its effect is very small. Overall, our rectenna device, which consists of built in
MIM structure, can be concluded as an optical sensing device and/or a light energy
harvesting device.
Hence, the extension to higher frequencies is still a challenge due to the
complexity of fabrication in nano scales. The improvement on built in MIM structure is
necessary for a greater efficiency by achieving high asymmetry and fast responsivity. In
addition, the illuminated I-V curve could be formulized by the substitution of incident
photon term into the generalized equation for dark (unilluminated) curve. There are
quite a few analytical simulations out there, however, we may need the comparison of
the experimental data of built in MIM structure for rectenna devices.
65
문
1960 microwave power transmission 시 래, 흔히
‘Rectenna’라고 는 rectifying optical antenna가 에
전 주는 역 감 나 에
는 술에 다.
Rectenna 는 antenna rectifier 치 , 근 들 가시
과 근 적 에 시 고 다. 전에는 크 에
치가 주 수 (수 THz)에 나 스 샘 만들
복 제 과정 여 큰 전 다. 본 연 에 는,
projection e-beam approach 여 처 rectenna
치 만드는 첫 공 것과 conventional e-beam lithography
제 시 다. 본 연 에 개 rectenna 치는 built
in MIM structure , 에 antenna rectifier 가 께 는
, antenna 내 에 들 전 rectifier, , MIM diode
낸다. 조에는 asymmetry 여 work
function 다 개 aluminum과 gold가 쓰 , insulating
layer 는 ALD (atomic layer deposition) 착 10nm aluminum oxide
가 쓰 다. 여 antenna역 는 nanowire는
projection e-beam approach AIPEL (atomic image projection e-beam
lithography) 여 patterning 고 또 conventional e-beam
lithography 제 다. 그 , metal electrodes metal pad는 각각
conventional e-beam lithography photolithography 만들 다.
Rectenna array 에 nanowire는 300nm에 480nm 고 는
수 THz 근 적 역 겨냥 다. 본 연 적 존에 연
가 많 고 는 pn diode solar cell과 여 다 접근
시 rectenna 치 제 과, 나 가 , 에 많 달 는
-근적 시 , 나 본다.
제 에 공 rectenna device 가 고 측정 본 결과, 랍게
, 쬐 주 전 신 가 3 order 정 는 보 다.
, 제 는 에 는 sensitivity 보여주 다. 또 , 실험
각 antenna length에 resonant wavelength
만, 그에 그 antenna length effect는 적다고 고 다.
66
Acknowledgement (in Korean)
감
무 여 날, 생 시 NFL에 들 그
제 같 2 시간 나 제가 감 적 니 실감
나 습니다. 처 에는 느 게 흘러가 만 시계 늘
느 저 게 졸 시 습니다. 흐드러 게
고 눈 수 곤 채 졸 생각
니 무거워 고 에 많 그 것 같습니다.
생 동 무 것 저에게 많 주신 여러 들
께 감 저 조 나 전 드 고 니다.
, 저 수님 신 수님께 감 말
드 니다. 저는 처 수님과 뷰 날 수가 습니
다. 원 생 조 저 다정히 맞 주시고
제 주 그 당시 나 뻤는 겠습니다.
생 동 에 연 뿐만 니라 제 생에 큰 그 그 보
는 키워주 습니다. 수님께 가 쳐 주 든 것들
제 에 쳐 보 니다. 수님 조 생각
가 겠습니다. 또 , 저 문 심 맡 주신 주
수님과 남 태 수님께 감 말 드 니다. 심 제게
주신 조 들 졸 문 쓸 많 습니다. I
also would like to thank Prof. Jimmy Xu for his great support. I do
appreciate for his research guidance. Without his help, this
dissertation could not be achieved. Thank you, professor.
생 제가 가 커다란 물 NFL 연
만난 들 니다. 저 주고 졸 좋
주신 들에게 감 말 드 고 싶습니다. ,
실험실 큰 니 신 미 니, 저 연 처 끝 조
주 습니다. 찌 보 저는 니께 큰 실험실 생
것 같습니다. 감 드 니다. 또 룸에 카드 같
많 보고 싶 것 니다. 그 고 연 . 철
는 뻐 주 나 감 겠습니다. 게
주시고 생에 값 조 주 큰
가 많 습니다. 다시 감 드 니다. 그 고 니.
67
시 107 에 내 가 니 는
맞 주시고 거 나눠주 습니다. 곧 태 날 주 건강
게 라 드 겠습니다. 제 졸 는 보 제는
득 . 날 주 께 눈물 거
가 그제 같 시간 게 났습니다.
가신 좋 보 럽습니다. 그 고 저 첫 수
동복 . 스 시 에 전 챙겨주 나 감동 는
겠습니다. 샛 니 께 복 시고 건강 시 랍니다. 겉
무심 보 만 가 신 . 졸 문에 주
말 니다. 스누 많
그 것 같습니다. 그 고 가 많 민 .
다 겠 만, 그 동 감 습니다. 니
졸 것 같습니다. 층에 내 가 포스 신
승 . 에는 시고 좋 만나 좋
겠습니다. 난 많 걸 . 가 같
좋겠습니다. 에 종종 티격 태격 나날들
많 것 같습니다. 티 승 .
냐고 늦 전에 나가라고 는 느 제가 졸 여
감 가 습니다. 술과 친 라고 들 는 제는 나
시니 조 좋겠습니다. 가 실 .
내 다 고 보게 것 같습니다. 감
드 고 시험 좋 결과 좋겠습니다. 실험실
에 많 친 니. 저 님 결 식
보 보 좋 습니다. 그 전에 종종 찮 신 걸
는 건강 좋겠습니다. 그 고 졸 에 만남
는 주 . 저 여 제는 생 습니다.
고맙고 동무 고 다 그 시절 그 연 쭉
갔 좋겠습니다. 그 고 랑 는 나 동 들. 동 들
저는 간에 그만 니다. 니 러 정
랑 시는 주 . 힘들 퇴근 에 02 스
에 주신 그 날 니다. 좋 식?
좋겠습니다. 제 다 들 주신 스
. 택 는 복 보 보 좋습니다.
전 는 심 시고 휴가 나 실 종종
좋겠습니다. 그 고 것 는 경 . 매 드
68
러 미 주 힘 가 많 습니다. 여 친
랑 , 나 포 연 에 공 여 끼
랍니다. 또 에 같 만 가
들에게 감 보냅니다. 점
많 정 . 침 힘든 제가 에 열심히 시는
보 칠 가 많 습니다. 찌 보 나 만
런스 맞 실 것 라 생각 니다. lol 신, . 는
근 들 친 것 같 많 쉽습니다. 여 습 말고
는 는 연 뿐만 니라 든 에 신 것 같습니다.
그 고 들 난에 연 . 고 는 연 에 고
민 많 만, 연 니 곧 보 것 라 생각 니다. 실험실 생
동 적 만 시 힘든 다 제든
연락 만나 좋겠습니다. 개 게 시는 천
. 본 들 많 쉽습니다.
다 에 에 다 들 주시 보겠습니다. 그 고
같 87라 식 . 스스럼 매 저 는
그 드 돌 겠습니다. 또 단 과 많 친 져 좋
결과 좋겠습니다. 저 컴퓨 에 종종 신
생들, 민수, , , , 단 . 찌 보 저 같
점에 는 시 는 그 시고 실험실에 거
많 좋겠습니다. 주제는 정 만 원 시는
에 게 연 시 랍니다. Kumar, I wish you all the best in
your future. It was great to see you at NFL. And please say hello to
your son for me.
제 생 담고 는 엔 친 들
수 습니다. / 연 끈
정 . 제가 에 주 만나 고 는 제는
많 보 쉽습니다. 저 에 나가 열심히 는
습 보 좋습니다. 그 고 제 눈만 든 것 수 는
Scotch 친 들, , , 수민, 택 , 낙헌. 다 께 에
날 다 니다. 곳에 원 나
고 겠습니다. 졸 에 연 닿 에 좋
IC 들, 정, 수 , 문균, . 각 치에
다들 무 집니다. 종종 수 타는 저 고맙고 2013
69
에 것 라 생각 니다. 그 고 랫동
같 만 제 만난 원 니. 저 는 그 주
고맙고 니 저는 든 것 공 절 헤 것
같습니다. 니 만 미 에 는 동갑내 친 같습니다. 감
니다.
저 주는 랑 는 님, 원
많 만 무 랑과 께 끝 믿 주 감 드 니
다. 그 고 곧 5월 전역 나뿐 동생, 승 에게 고
전 니다. 조 만 견 열 것 니다. 그 고 내
니 1주 라 누 보다 느낄
랑 다고 전 고 싶습니다.
2013 1월 17 , 강민
